Complete Manual Training 14

ROBOT MILLENNIUM v 16.
5 - USERS MANUAL UPDATE, May 2003

Web: www.robot-structures.com e-mail : sales@robobat.fr.com Techsupport : service@robot-
structures.com
1
TABLE OF CONTENTS
NOVELTIES IN THE ROBOT MILLENNIUM PROGRAM (VERSION 16.5).................................. 4
GENERAL PART ..................................................................................................................................... 4
BAR STRUCTURES ................................................................................................................................. 5
PLATES, SHELLS, PLANE STRESS / PLANE DEFORMATION STRUCTURES................................................... 6
SOLIDS.................................................................................................................................................. 6
RC STRUCTURES................................................................................................................................... 6
STRUCTURE ANALYSIS .......................................................................................................................... 7
ROBOT EXPERT ..................................................................................................................................... 7
ROBOT OPEN STANDARD ....................................................................................................................... 8
1. ROBOT MILLENNIUM INSTALLATION...................................................................................... 9
1.1 HARDWARE REQUIREMENTS....................................................................................................... 9
1.2 INSTALLATION PROGRAM........................................................................................................... 9
1.2.1. Installation on Single Computers ........................................................................................... 10
1.2.2. Network Installation of Robot Millennium.............................................................................. 15
INSTALLATION OF NETWORK PROTECTION............................................................................. 16
NETWORK SERVER VERSION INSTALLATION OF ROBOT MILLENNIUM................................ 17
NODE INSTALLATION OF THE NETWORK VERSION OF ROBOT MILLENNIUM..................... 19
Functionality of the Network Version .......................................................................................................... 22
PROGRAM ADMINISTRATOR....................................................................................................... 23
Examples of how to use the Administrator program...................................................................................... 24
NETWORK MONITOR.................................................................................................................... 25
1.2.3. Protection Settings................................................................................................................. 26
1.3 GENERATION OF INSTALLATION REPORT................................................................................... 29
2. INTRODUCTION......................................................................................................................... 30
2.1 GENERAL DESCRIPTION OF THE PROGRAM................................................................................ 30
2.2 GENERAL RULES...................................................................................................................... 34
2.2.1 The Layout System........................................................................................................... 35
2.2.2 Menu, Context Menu, Toolbars ........................................................................................ 36
2.2.3 Preferences and Job Preferences...................................................................................... 38
2.2.4 Selection and Filters ........................................................................................................ 40
2.2.5 Display of Structural Attributes........................................................................................ 43
2.2.6 Lists Used in the Program................................................................................................ 44
2.2.7 Common Features of Dialog Boxes (Graphic Cursor, Calculator) .................................... 45
2.3 SIGN CONVENTION................................................................................................................... 46
Sign Convention for Bar Elements................................................................................................... 46
Sign Convention for Planar Finite Elements .................................................................................... 47
Sign Convention for Volumetric Finite Elements.............................................................................. 48
2.4 LIST OF SHORTCUTS ................................................................................................................. 49
2.5 SNAP SETTINGS........................................................................................................................ 50
3. GENERAL RULES ON STRUCTURE MODEL DEFINITION................................................. 52
3.1 STRUCTURE TYPES................................................................................................................... 52
3.2 NODES, BARS .......................................................................................................................... 52
3.3 BAR SECTIONS AND MATERIALS............................................................................................... 55
3.4 PANELS ................................................................................................................................... 60
3.4.1 Types of Planar Finite Elements....................................................................................... 62
Delaunay's Triangulation Method ................................................................................................................ 62
Coons' Method............................................................................................................................................ 62
Parameters of FE Mesh Generation.............................................................................................................. 64
3.4.2 Emitters, FE mesh refinement, consolidation and quality.................................................. 67
3.5 PANEL THICKNESS ................................................................................................................... 69
3.6 SOLIDS (VOLUMETRIC STRUCTURES) ........................................................................................ 71
3.6.1 Description of Volumetric Finite Elements.............................................................................. 74
ROBOT MILLENNIUM v 16.5 - USERS MANUAL UPDATE, May 2003
structures.com
2
3.7 OPERATIONS ON 2D OBJECTS (EXTRUDE, REVOLVE, COMBINATION) AND ON 3D OBJECTS
(FILLET, CHAMFER)............................................................................................................................. 78
3.8 SUPPORTS................................................................................................................................ 84
3.8.1 Elastic Support Definition (Layered Soils)............................................................................... 86
3.9 LOADS..................................................................................................................................... 88
3.9.1 Load Case Combinations........................................................................................................ 96
3.9.2 Code Combinations ................................................................................................................ 97
3.9.3 Moving Loads....................................................................................................................... 100
3.9.4. Autoloader Moving Loads................................................................................................. 105
3.10 SNOW/WIND LOADS............................................................................................................... 108
3.10.1 Snow/Wind Loads in 3D...................................................................................................... 110
3.10.2 Wind Loads Masts / Towers ............................................................................................. 111
3.10.3. Snow/Wind Loads on 3D Objects ....................................................................................... 112
3.10.4. Wind Loads for the Structure with a Polygon Base (Prism)................................................. 114
3.10.5 Load Report (Loads Taken from Database) ........................................................................ 117
3.10.6 Automatic Definition of Loads Induced by Soil Pressure...................................................... 119
3.11 CABLES ................................................................................................................................. 122
Equations Governing the Problem................................................................................................. 123
Cables in Robot Millennium.......................................................................................................... 125
Assembling Load Case .................................................................................................................. 126
Load Cases after Anchorage.......................................................................................................... 127
3.12 OTHER STRUCTURAL ATTRIBUTES.......................................................................................... 128
3.13 OTHER STRUCTURE COMPONENTS .......................................................................................... 131
3.13.1 Edit Operations .................................................................................................................. 135
3.14 TYPICAL STRUCTURES............................................................................................................ 136
3.15 PHASE STRUCTURES............................................................................................................... 137
3.16 SURFACE FINITE ELEMENTS ................................................................................................... 138
3.17 COORDINATE SYSTEM............................................................................................................ 140
3.17.1 Global Systems ................................................................................................................. 140
3.17.2 Local Coordinate Systems................................................................................................. 140
3.17.2.1 Bar Elements .............................................................................................................................. 140
3.17.2.2 Finite Elements........................................................................................................................... 141
3.17.3 Global-local Systems Relations ......................................................................................... 142
3.18 SIGN CONVENTIONS ............................................................................................................... 143
3.18.1 Bar Elements ..................................................................................................................... 143
3.18.2 Finite Elements.................................................................................................................. 144
3.18.3 Volumetric Finite Elements................................................................................................. 145
3.19 SECTION DEFINITION.............................................................................................................. 145
4. STRUCTURE ANALYSIS.......................................................................................................... 152
4.1 STRUCTURE CALCULATIONS................................................................................................... 152
4.2 AVAILABLE ANALYSIS TYPES................................................................................................. 153
4.3 NEW CASE DEFINITION AND ANALYSIS TYPE CHANGE ............................................................ 166
5. RESULT ANALYSIS.................................................................................................................. 173
5.1 DIAGRAMS............................................................................................................................. 174
5.2 TABLES ................................................................................................................................. 177
5.3 MAPS ON BARS ...................................................................................................................... 180
5.4 MAPS (PANELS) ..................................................................................................................... 181
5.5 PANEL CUTS .......................................................................................................................... 185
5.6 MAPS ON SOLIDS.................................................................................................................... 188
5.7 SOLID CUTS ........................................................................................................................... 191
5.8 STRESS ANALYSIS (BARS) ...................................................................................................... 194
5.9 STRUCTURE STRESS ANALYSIS ............................................................................................... 196
5.10 GLOBAL ANALYSIS - BARS..................................................................................................... 199
5.11 DETAILED ANALYSIS.............................................................................................................. 201
5.12 INFLUENCE LINES .................................................................................................................. 205
5.13 REDUCED RESULTS FOR PANELS............................................................................................. 206
5.14 TIME HISTORY ANALYSIS / ADVANCED ANALYSES DIAGRAMS AND TABLES........................ 210
structures.com
3
6. STRUCTURE DESIGN.............................................................................................................. 215
6.1 STEEL/ALUMINUM DESIGN..................................................................................................... 215
6.1.1. Detailed Analysis - Eurocode3 ...................................................................................... 222
6.2 DESIGN OF RC STRUCTURE ELEMENTS ................................................................................... 226
6.2.1 RC Beam Design........................................................................................................... 228
6.2.2 RC Column Design ....................................................................................................... 233
6.2.3 Foundation Design ........................................................................................................ 237
6.2.4. Continuous Footing Design ................................................................................................. 240
6.2.5 Deep Beam Design......................................................................................................... 241
6.2.6 Final Drawings.............................................................................................................. 243
6.3 DESIGN OF RC BARS.............................................................................................................. 246
6.4 STEEL CONNECTION DESIGN .................................................................................................. 255
6.5 TIMBER DESIGN..................................................................................................................... 260
6.6 PLATE AND SHELL REINFORCEMENT....................................................................................... 263
7. SECTION DATABASES............................................................................................................ 273
8. PRINTOUTS............................................................................................................................... 276
8.1 CALCULATION NOTES ............................................................................................................ 276
8.2 PRINTOUT COMPOSITION........................................................................................................ 276
8.2.1 Standard Tab........................................................................................................................ 278
8.2.2 Screen Captures Tab ............................................................................................................ 279
8.2.3 Templates Tab...................................................................................................................... 279
8.2.4 Simplified Printout Tab......................................................................................................... 280
STAGE 1.................................................................................................................................................. 280
STAGE 2.................................................................................................................................................. 280
STAGE 3.................................................................................................................................................. 281
8.3 PAGE SETUP........................................................................................................................... 281
APPENDICES .................................................................................................................................... 282
APPENDIX 1................................................................................................................................... 283
APPENDIX 2................................................................................................................................... 288
APPENDIX 3................................................................................................................................... 298
APPENDIX 4................................................................................................................................... 334
APPENDIX 5................................................................................................................................... 335
APPENDIX 6................................................................................................................................... 351
structures.com
4
Novelties in the ROBOT MILLENNIUM PROGRAM
(Version 16.5)
General Part
1. Structure axes possibility of defining any structure axes in the form of:
- straight lines
- rays
- segments
2. Structure axes possibility of transforming lines and bars into any structure axes
3. Structure axes possibility of selecting objects-parts of a structure by means of axes
- new option Selection by means of Structure Axes (the Edit / Select Special menu)
- possibility of indicating the axes between which selected objects are positioned (the
Selection dialog box)
- possibility of indicating the structure axes between which selected objects are positioned
in the graphical viewer using the mouse cursor (by indicating a single axis and through
window selection)
4. Structure axes altered manner of activation of the modification dialog box (at present, the
modification mechanism is called up by the Properties option in the context menu).
5. Structure axes possibility of defining axes in the Cartesian system rotated by a
determined angle
6. Structure axes possibility of reading structure axes defined in AutoCAD (as a result,
arbitrary structure axes in the Robot program are obtained)
7. Operations on structure model in the graphical viewer enhanced operation of the
mechanism for highlighting structure parts (nodes, bars, lines, panels, solids, etc.) by
means of the mouse cursor, which include as follows:
- not highlighting the objects that are invisible (the display is switched off in the Display
dialog box)
- not highlighting the objects for which the snap mechanism of the mouse cursor is
switched off (the Snap Settings dialog box)
8. Visibility of structure attributes display of attributes activated by opening certain dialog
boxes is switched off automatically on closing these dialog boxes
9. Visibility of structure attributes possibility of defining sets of attributes to be displayed on
structure parts (the Display dialog box)
10.Definition of the local work system possibility of saving a work plane defined by 3 points
as a predefined structure view available from the level of the View History dialog box and
the dialog box for setting a view (the button in the graphical viewer)
11.A window with a structure view in OPEN GL a new cursor mode which enables changing
the structure view depending on a cursor position in 4 quarters of the graphical viewer
12.Loading of projects generated under previous versions of the program the initial view set
automatically (such a zoom selected so that the entire structure is shown in the graphical
viewer)
13.New printout language for results Croatian (037)
14.Data exchange format SAT possibility of loading a file in which units other than meters
were used
structures.com
5
Bar Structures
DEFINITION
1. Possibility of displaying groups of bars used in steel design as bar attributes in a structure
graphical view
2. New types of steel compound sections
- two angles with legs back to back (there is a possibility to choose if a section should be
set shorter leg back to back or longer leg back to back)
- C-section and I-section back to back
3. New types of user sections:
- steel sections: round filled and box II
- concrete I-section
4. Concrete sections possibility of reduction of geometrical moments in static and dynamic
analysis
5. Moving loads
- possibility of a moving load case analysis for overhead traveling cranes
- possibility of definition and analysis of moving load cases for a user-defined vehicle
6. Snow and wind loads load generation according to the American code ASCE 7-98
7. Loads - combinations:
- possibility of combining moving load cases with static load cases
- possibility of using other combinations, including user quadratic combinations, in
definition of a combination
CALCULATIONS
8. Consideration of added masses in static calculations for the needs of off-shore structures
RESULTS
9. Results for the elasto-plastic analysis
- new table of plasticity state history (presenting values of the plastic ratio in the course of
the iteration process)
- possibility of displaying the plastic ratio in tables of forces and stresses
- possibility of displaying the plastic ratio in the global analysis
10.Results in a tabular form possibility of displaying results at intermediate points - division
of calculation elements into n-points
11.The dialog boxes Info-bars and Info-nodes possibility of generating a calculation note
with entire data contained in these dialog boxes
DESIGN OF STEEL AND TIMBER MEMBERS
12.Design of multi-chord columns according to the Polish code PN-90/B-03200
13.Consideration of stiffness of a section internal flange in the analysis of a section with
stiffener according to Eurocode3
CONNECTION DESIGN
14. New connection type a column - beam (two-sided) connection according to the codes
CM66 and EC3
structures.com
6
Plates, Shells, Plane Stress / Plane Deformation Structures
DEFINITION
1. Possibility of graphic selection of panels and contours through indicating their description
with the mouse cursor
2. Snow and wind loads for cylindrical containers - load generation according to the
American code ASCE 7-98
RESULTS
3. Possibility of displaying a finite element mesh on a result map
See also some of the novelties for bar structures
Solids
DEFINITION
1. A number of volumetric finite elements considered in statistical data
See also some of the novelties for bar, plate and shell structures
RC Structures
REQUIRED MEMBER REINFORCEMENT
1. reinforcement calculation according to the Canadian code CSA A23.3-94
2. reinforcement calculation according to the Romanian code STAS 10107/0-90
3. reinforcement calculation according to the Singaporean code CP65 (1999)
4. calculation update according to the Polish code PN-B-03264: December 2002
REQUIRED REINFORCEMENT OF PLATES AND SHELLS
1. reinforcement calculation according to the Canadian code CSA A23.3-94
2. reinforcement calculation according to the Romanian code STAS 10107/0-90
RC BEAMS
3. consideration of releases defined for bars when importing beams via *.DAI files
structures.com
7
RC COLUMNS
3. possibility of considering reactions that result from adjoining beams in column calculations
RC PLATES
1. calculation of plate reinforcement according to the code SNIP 2.03.01-84 (including
calculation of punching and considering wire fabric databases characteristic of a given
code)
4. possibility of selecting a set of plates (multiselection) when importing *.DAI files from the
DALLES module
DEEP - BEAMS
FOUNDATIONS
Structure Analysis
1. Moving loads
- analysis of moving loads induced by overhead traveling cranes
- analysis of loads induced by user-defined vehicles
2. Structure seismic analysis according to the Canadian code
Robot Expert
1. New RC calculator according to the Canadian code CSA A23.3-94
2. New RC calculator according to the Romanian code STAS
3. RC calculator according to the Polish code update to PN-B-03264: December 2002
structures.com
8
Robot Open Standard
1. Support of the following bar attribute: inactive bar
2. Possibility of changing bar orientation (local coordinate system)
- possibility of setting the local systems sense with respect to the selected axis of the
global coordinate system
3. Definition of the size of FE mesh by specifying length of the element edge
4. Possibility of switching on automatic adjustment of FE mesh on touching objects
5. Possibility of dividing object edges:
- into N parts
- at a selected point
- by cutting with a selected plane
6. Recognition of the object type (if it is a solid type object)
7. Calculations of:
- object area
- solid volume
8. Saving the current state of the project
structures.com
9
1. Robot Millennium INSTALLATION
1.1 Hardware Requirements
To successfully install Robot Millennium, your computer must be equipped with the
following:
a Windows NT/2000/XP operating system (Windows 2000 or XP Professional system is
recommended)
a complete IBM PC compatible computer with at least a P5 processor,
128 MB RAM and about 500 MB of free disk space when the installation is completed (for
advanced users 512 (or more) Mb RAM and 5 GB of free disk space is recommended),
resolution at least 1024x768.
Robot Millennium is delivered along with a protection key (this is necessary during
installation and usage of the program) or with disk protection. It is advised to insert the key to
a parallel port (printer port) prior to the installation. The key should be inserted into the port
before every program activation and should remain there while the program is in use.
In order to correctly print the Robot Millennium calculation notes, it is necessary to have a
word processor supporting *.rtf (Rich Text Format) format files. MS Word version 6.0 (or
higher) or WordPad which is included in the standard Windows 95 installation are examples
of such word processors. The Robot Millennium system also has its own word processor
that supports *.rtf format files.
1.2 Installation Program
Robot Millennium is equipped with an automatic installation program. It allows the user to
perform a fast and effective installation. During the installation of the program, a few short
questions will appear on the screen that the user must answer for the installation to continue.
First, the protection key must be inserted into a parallel port of the computer (printer port).
Next, the CD with the Robot Millenniuminstallation must be inserted into the appropriate CD
drive. The system installation start window will appear on the screen. Once the Installations
option is selected, the options that allow for the Robot Millenniumsystem installation appear
on the screen.
Individual buttons located on the screen allow for:
RobotMillennium/Robot LT/Robot PL installation of Robot Millennium, Robot Light
and Robot Plate (R/C plate design)
Robot Expert installation of calculators offered by RoboBAT company: the calculator
allowing the calculation and verification of RC sections may begin, the retaining wall
calculator, the EC3 calculator for steel structures, timber structure calculator according to
EC5
Robot Load Report starting the installation of the Robot Load Report program
R/C tools: Robot Dalle, Robot Mur, ACI Beam, ACI Column, ACI Foundation
commencement of the installation of additional programs for calculations of R/C structure
elements
structures.com
10
Network tools: Network protection, Network Monitor starting the installation of
network protection for the network version of the Robot program and Monitor program
that allows one to monitor utilization of license for the network version of Robot (see the
description in chapter 1.2.2)
General tools - activation of installation of additional drivers, VRML Player, DAO as well
as Solid DXF Converter.
To install Robot Millennium on the disk, press the Robot MILLENNIUM / Robot LT/Robot
PL button.
NOTE: It is possible, if required during installation, to return to the previous dialog
box shown on the screen in order to make corrections to the data previously
entered by the user.
1.2.1. Installation on Single Computers
The consecutive stages of the Robot Millennium installation on a stand-alone unit are listed
below:
1. A dialog box for the selection of the desired installation language will appear as shown on
the drawing below.
2. The Robot Millennium License Agreement dialog box will appear; to continue the
installation once you have familiarized yourself with the contents, press the YES button
(accepting the license agreement conditions); the following dialog box presents notes
pertaining to the performance of Robot Millennium; press the NEXT button to continue.
3. A dialog box prompting basic user data (see the figure below) will appear; enter the name
(initials) of the user, the company name, and optionally the e-mail address; press the
NEXT button to continue.
structures.com
11
4. A dialog box in which the type of Robot Millennium installation may be selected will
appear. Available are: full, typical, custom; network installations are also listed in this
dialog box - server and node; one may also select in this dialog box the installation of
Robot LT and Robot PL
NOTE: The network installation (network server and nodes) is described in the next
chapter of the manual (Chapter 1.2.2).
Once the appropriate type of stand-alone installation is selected, press the NEXT button to
continue the installation.
structures.com
12
One of three installation types on a single computer may be selected in the above dialog box:
full allows to install the full version of the program: all available languages (interface,
calculation notes, help) and national codes
typical allows to install the full version of the program for one language version only
custom a selection of installation components may be performed; once this option is
selected and the NEXT button is pressed, the dialog box shown below will appear on the
screen.
It is also possible to select the network installation described in Chapter 1.2.2.
structures.com
13
NOTE: during the custom installation, at least two options must be selected: The
basic program files and, for example, for the English language version Files
for the English language version.
5. There appears a dialog box that allows one to define the target folder of program
installation on hard disk (see the figure below): the C:\Program Files\ROBOT Structural
Office\Robot MILLENNIUM folder is the default folder of the installation. The localization
may be typed directly from the keyboard or selected by means of the Browse button,
available in the lower part of the dialog box; if the defined folder is not present on hard
disk, the installation program will create the folder with the user-defined name; press the
CONTINUE button to continue.
NOTE: It is not recommended to install Robot program on a substituted disk. In such
a case, after restarting Windows system Robot program support will not be
registered; to make it available, it should be installed by means of appropriate
icon from Robot folder.
structures.com
14
6. A dialog box for entering the group name in which Robot Millennium will be installed will
appear (see the drawing below); the default group name which the user may change
appears on the screen (Robot Structural Office); to continue, press the NEXT button.
7. A dialog box confirming the entered information (program location, name, user
information) appears; press the NEXT button to continue;
structures.com
15
8. The program begins the Robot Millennium installation (copying files); the installation
Progress window appears on the screen. Short notes about the program are displayed on
the screen during the installation;
9. Once the installation is complete, the group dialog box containing the icons Robot
Millennium and the auxiliary programs is displayed;
10. The Robot Millennium program may be uninstalled by selecting the Add/Remove
programs available in the Control Panel. Selection of this option allows for modification of
previously installed Robot program version (adding or removing components) and version
reinstallation according to currently selected options (repair installation).
The Robot Millennium system has been installed to a disk. It is necessary to restart
WINDOWS NT/2000/XP to run Robot Millennium; therefore a warning to restart the
computer before activating the program appears on the screen. Robot Millennium is
activated by:
double-clicking the icon of the system located in the Windows desktop (if the user decided
to place the icon on the desktop during installation),
selecting the Robot Millennium command in the Robot Structural Office group created
during installation.
1.2.2. Network Installation of Robot Millennium
NOTE: Network installation of the Robot Millenniumprogram should be performed
by Network Administrator or, in the worst case, in his presence.
Network installation consists of three parts:
installation of network protection
installation of the server
installation of nodes.
In the first stages, the network installation (for the server installation as well as for network
nodes) is carried out in the same way as for a single computer installation (see points 1-4
described in Chapter 1.2.1). To install the network version, the Robot Millennium server and
network protection must be installed first.
Network protection may be installed on any computer connected to the network. The
installation is independent from the network installation of Robot Millennium and does not
have to be carried out on the server where Robot Millennium is to be installed. The other
users, in order to use protection during node installation, must have access rights to read
from the folder where the License Protection files are installed.
The Robot Millennium server must be installed on a disk that is visible to other users. The
main folder of the server version of Robot Millennium must be available to other users with
reading rights, whereas the USERS folder with reading and saving rights. The Robot
Millennium server should be installed by the administrator of the Robot system (the person
installing the Robot Millennium server is given such rights). Once the server is installed, the
users may install the nodal versions on their computers. During node installation, the network
address of the Robot Millennium server should be given.
NOTE: During the Robot Millennium server installation the use of short names for
the folder in which the server version will be installed is recommended
(maximum 8 characters of the folder name and 3 extension characters) see
point 4 in chapter 1.2.1.
structures.com
16
The configuration recommended for network installation of the program:
Network protection server - on any computer, yet never on an NT server (installation of
protection drivers)
Robot Millennium server - on the computer where the backup copies will be stored (it may
be an NT server, for installation of Robot Millennium server does not comprise installation
of any drivers, but, to maintain network security, this should be another server, especially
in the case of big networks).
INSTALLATION OF NETWORK PROTECTION
Installation of network protection allows one to install network protection on any computer
connected to the network. The installation is independent from the network installation of
Robot Millenniumand does not have to be carried out on the server where Robot Millennium
will be installed. Moreover, it must not be carried out on an NT server. Network protection
supports both the Robot Millennium program and all the calculators (RC calculator, Robot
Expert, etc.).
Network protection installation is started once the Network Protection button is pressed in
the menu of the Installation layout (see also chapter 1.2). Installation of network protection
should be carried out before the installation of Robot Millennium server and before the
installation of Robot Millennium program nodes.
Installation of network protection allows one to set protection for the entire network by means
of one protection key (red HASP). The key must be inserted in the printer port of the computer
where the installation of network protection has been carried out, for it is there that the drivers
of the key are installed.
Moreover, the installation of the protection key requires the License Protection to be installed
first. The License Protection is provided on a diskette delivered together with the network key.
The installer of network protection demands one to insert this diskette by displaying the
appropriate message:
The installation procedure reads License Protection from the diskette and displays its number
in the below-presented dialog box.
structures.com
17
NOTE: Each License Protection is installed in the following folder: D:\Program
Files\Common Files\RoboBAT\Prot, where D:\Program Files\Common
Files is a Windows folder. The above folder must be made accessible for
other users to read, for the user must indicate the folder of the installed
License Protection during node installation.
Once the installation is completed, one should restart the Windows system to activate the
drivers of the protection key.
NETWORK SERVER VERSION INSTALLATION OF ROBOT
MILLENNIUM
During the Robot Millennium server installation, select the following option in the Setup
Type dialog box: Network - server.
structures.com
18
After pressing the NEXT button, Robot Millennium will be installed on the server. After the
server installation of Robot Millennium, this program cannot be activated. A node installation
is necessary to do this. A node may be installed on the same computer as the server version
but the installation must be performed in a different folder.
In installation of the network version the possibility of selecting User File Server has been
implemented; this is a place where user files are saved. Hitherto, while working in the
network, user files have been saved in the place where the Robot program server (USERS
folder) is installed. At present, files are saved in the User File Server. In the dialog box below,
the place of their installation may be selected; the installation program creates the USERS
folder at the indicated location.
structures.com
19
Once the files are installed on the server, the note shown below will be displayed.
If the server version of Robot Millennium was installed on the NTFS partitions (Windows
NT), the main database of Robot Millennium program should be made available to other
users, whereas the "Users" database must be made available with saving rights.
The administrator should have the right to save to the main database of Robot Millennium
program. During the installation of a network version node, the Robot Millenniumfolder must
be made available with the rights to write.
NODE INSTALLATION OF THE NETWORK VERSION OF ROBOT
MILLENNIUM
During the installation of a node of the network version of Robot Millennium, the Network
node option should be chosen from the Setup Type dialog box (Best Performance or Efficient
Space).
As during version installation on single computers, components for installation may also be
chosen for both types of network nodes.
NOTE: It is recommended that the same language versions of Robot Millenniumbe
installed on all network nodes.
structures.com
20
During installation, the server version Robot Millennium installation folder should be
indicated (the Robot Millennium server should be installed first). It should be noted whether
the Robot Millennium version folder has been given as the networks address or whether it
was mapped (changed to a disk letter). In the second case, options for re-mapping the disk
after restarting the Windows system should be set.
In the installation of the network version, the possibility of selecting User File Server has been
implemented; this is a place where user files are saved. In the dialog box below the
destination folder for the Robot program server may be selected; the installation program
checks the right of writing in the indicated location and interrupts further installation, if the user
is not given such a right.
structures.com
21
Once the files are installed, the Protection License is installed on the node version of Robot
Millennium. The dialog box shown on the drawing below in which one should determine the
path to the Protection License that was installed during the installation of network protection
for Robot Millennium.
Nodal protection can be installed directly from the protection diskette. Then the path to the
diskette should be given (e.g. A:\).
structures.com
22
Once the installation is complete, the Windows system should be restarted.
Functionality of the Network Version
The basic features of the network version of the Robot Millenniumare presented
below:
1. Work in the local or server mode.
The user may select the place where his/her files will be kept (projects and configuration
files). This may be a local computer that is being used, or, the server computer. In the
second case, the user will be able to work on his projects from any network computer.
2. Possibility of work on a project from any computer
If the user chooses to work in the server mode, he/she will be able to work on projects
from any network computer (without needing to copy files).
3. Orientation of the program towards the user.
Every user has its own folder with configuration files and projects.
4. Advanced options of the configuration file management
The administrator may define common configuration files on the server for all network
users. The user is informed about any changes and may easily perform an actualization of
configuration files (compare the below presented description of the Administrator
program).
5. Possibility of limiting access to Robot.
If the necessity arises, the administrator may define a list of users with the right to access
Robot, a right to make changes to the default configuration, etc. (compare the below
presented description of the Administrator program).
structures.com
23
PROGRAM ADMINISTRATOR
The Administrator program (admn.exe) is used to configure the work parameters of the
network version of Robot Millennium. The program is located in the SYSTEM/EXE folder of
the installed Robot Millennium program. Its capabilities depend on the users rights. The
following three tabs will appear to a regular user, once the application is activated: Files, Work
Mode and Robot Expert. The Robot Millennium program Administrator has access to three
additional ones (Users, System and Network Monitor).
NOTE: Prior to opening the Administrator program, it should be checked that Robot
Millennium program is not open.
Below, one can find a description of particular tabs that appear in the dialog box of the
Administrator.
The Users tab (option available only for the administrator)
The Users tab allows the administrator to define a list of users and assign them the
appropriate user rights.
The dialog box contains the following options:
User name -- the name under which the user is identified in the network.
Access rights the user may run Robot.
Administrator rights the user is given full administrator rights.
Rights to work on custom (own) configuration the user may work on his/her own
configuration files otherwise the user work configuration is the network configuration.
Option Check the user list at start results in denying access to the program to all those
not listed on the user list.
System tab (option available only for the administrator)
The dialog box contains the following options:
Modification of the configuration files once this option is selected, the administrators
configuration will become the server configuration (The users will be able to update their
own configuration files with the changes made by the administrator).
Protection Code in the case of a protection code change (e.g. once a new option is
purchased), the new code should be entered on the edit field. Once Robot Millennium is
run by a user, the application will automatically update the protection code.
Creating a new user database removal of the old user database and creating a new
one.
Network Monitor tab (options available only for the administrator)
The options located in this tab allow one to define the network address of the computer where
the Network Monitor is installed. The defined address refers to all installed Robot Expert
programs (calculators).
Robot Millennium finds this address automatically.
Files tab
A message about the necessity to update configuration or system files may appear at the
start of Robot. The Files tab informs the user about the state of the files and allows for
automatic updating.
NOTE: Robot should be closed at this time.
structures.com
24
Work Mode tab
The Work Mode tab allows the user to define his/her work parameters on the net.
A local work mode means that the user configuration file will be located on the users
computer (faster access).
A server work mode means that the configuration file will be located on the server. This is
recommended if the user wishes to work on the same project on various computers.
The configuration selection allows the user to declare whether he/she wants to use the local
or server configuration (defined by the system administrator).
As long as we want to continue working on the same configuration files copying of local
configuration files and projects to the server or server files to the local disk should be
performed once the work mode is changed (appropriately from the local to server mode or
from server to local mode).
Robot Expert Tab
Robot Expert tab allows one to update the network address of the computer where the
Network Monitor is installed.
Examples of how to use the Administrator program.
A. The user cannot run Robot Millennium.
Check that the server is active
if during node installation the path to the server was mapped, check if after restarting the
mapping is still current.
check if the user has reading rights in the System folder on the server.
check if the user has saving rights for the Users folder \ user name
check if the user has the right to use Robot Millennium.
B. Messages at start of the Robot Millennium program.
in the case of the appearance of a message about the necessity to update the configuration
or system files, after activating the Administrator program it is required to switch to the Files
tab and open the appropriate option.
C. Work on the network version
If the user wishes to work on projects from any network computer, then the Network work
mode option should be activated on the Mode tab.
If the user is working in the local mode (configuration files are located on a local computer)
and wants to work on the same files on a different computer, the following should be done:
run the Administrator program
switch to the Work Mode tab
use the option for copying files on the server
change the work mode to server mode
D. The job of the network version administrator.
Modification of the server configuration (changes in the configuration visible for all users)
the administrator checks that he is working on the current configuration (File tab), if not,
then an update of configuration files is performed.
the administrator enters the required changes (new material addition, definition of a new
section, supports, addition of a new template, etc.)
he activates the Administrator program
structures.com
25
on the System tab, he selects the Configuration File Modification option.
the administrator closes the Administrator program
the user who will then run Robot Millennium will receive a warning about the need to
update the configuration files.
Modification of the protection code (e.g. once new options have been purchased)
run the Administrator program
on the System tab, enter a new protection code
the user who will later run Robot Millennium will have the protection code automatically
updated.
NETWORK MONITOR
Possibilities of the Network Monitor:
Tracking the programs with protection run in the network (Licenses tab)
Generation of reports concerning the statistics of using license (Reports tab)
Gathering information concerning certain events (Events tab)
Sending e-mails to Users.
Program Monitor may be installed on any computer in the network. For the program to
operate according to its goals, it should be launched (preferably in Autostart) and not
switched off till there are protected programs operating in the network. Once the Monitor is
installed, one should type the name of the computer where the Monitor is installed in
the "Network Monitor" tab of the "Administrator" program. Each program (Robot
Millennium or Robot Expert) sends to the Monitor information that it has been activated and
uses protection. The Licenses tab will display information concerning a given program (the
name of the computer where it has been run, user's name, number of the protection used,
name of the application, process identifier, time of activation, state of the application).
The application tree consists of roots (computer names) and the activated applications. The
applications activated on one computer use one license. During its operation, Robot
Millennium periodically (every few minutes) sends control information to the Monitor. The
time of sending the information is introduced in the Last Ping field. If the hour introduced in
this field is not up-to-date, it means that the application is not active. Pings are not sent while
a structure is being prepared for calculations.
NOTE: Robot Expert Programs do not ping and Robot Millennium cannot ping
when opening an *.str file.
Exiting an application sends appropriate message to the Monitor and the program is removed
from the tree of applications in the Licenses tab.
Diagnostics of using a license consists in watching the tree in the Licenses tab. The
Administrator may see who has been using the license and for how long. He may also see
whether the license is blocked by process jamming or not (Last Ping received a long time
ago). It is not possible to take away the license from a user by remote control, or to
unblock the license taken away by a jammed process. The Remove Application option is
used to delete from the tree the application that does not exist (it must have ended in an error
and the protection has not been returned). The Monitor should be run during the entire time
when protection is used. After exiting and re-running the Monitor, one can obtain information
concerning the programs operated in the network, with the exception of the information
structures.com
26
concerning the programs that ended their operation in an error when the Monitor was inactive
(these programs will not be included in the tree, but they will keep blocking the license).
The Administrator may generate reports concerning the use of the license by the users.
Several report templates will be provided, but the user will be able to create his own reports,
as need be. A user-defined report will appear in the table in the Reports tab and it may be
printed. The default report contains the entire database (All the data are introduced into an
*.mdb database and can be viewed through Access). Checking out the Logout Status field
means that the application has sent the message notifying that its operation has terminated
(standard behavior). If the field is not checked out, it means that the application terminated in
an error or it terminated when the Monitor was switched off.
NOTE: From time to time, the unnecessary records should be removed from the
database.
If an application (Robot Millennium or Robot Expert) has not been assigned a license, then,
a message will be sent to the Monitor and it will be displayed in the Events tab.
It is possible to sent e-mails to users. The Network Monitor program contains a dialog box
that allows one to define the addressee and the message contents. E-mails sent to a user
during his work are displayed immediately (not later than in 5 minutes after sending in the
case of work on a demo version), while the remaining ones will be displayed during the next
Robot session.
NOTE: E-mails may be sent only to the users of Robot Millennium.
For the e-mails to be received, one should check out the "Message support" field in the
License Monitor tab of the Administrator program (the tab is available only for the
Administrator).
For Robot Expert programs to be able to register, one should perform the following operations
in the "Network Monitor":
Run the "Administrator" program on the computer where the Robot Expert programs will
be used
Go to the Robot Expert tab and update the License Monitor Address (if the button is
grayed out, it means that the actualization has been performed).
During actualization, there may appear an error caused by the fact that the user is not
authorized to modify registers. In such case, the actualization should be performed by the
user with the rights of an Administrator for the computer in question (not to be confused with
the rights of the Administrator of the Robot Millennium system), or else, the actualization may
be performed by running the RegEdit program (contained in system Windows NT/2000/XP)
and setting the key value "HKEY_LOCAL_MACHINE \ SOFTWARE \ RoboBAT \ Protection \
Net \ LicenseServerName" to the current address of the License Monitor (provided over the
"Actualize" button in the Robot Expert tab of the Administrator program).
1.2.3. Protection Settings
The protection key delivered together with the Robot program can be changed by specifying
the appropriate code. Such a situation may occur when the user purchases new options or
when the temporary protection terminates and the user decides to prolong it.
The Protection - Settings dialog box is used to modify or check protection settings on the
user computer. After installing the program, the options provided in this dialog box enable
modification of protection settings.
The Protection - Settings dialog box opens after choosing the following option: Tools /
Protection settings located in the Robot Structural Office group in the START menu of the
Windows operating system.
structures.com
27
NOTE: Prior to opening the Protection - Settings dialog box the user should make
sure that the programs using RoboBAT protection keys have been closed.
Once the Protection - Settings option is selected, the dialog box shown in the drawing below
appears on the screen.
The top part of the dialog box contains the Select key list; from this list the user should select
a protection key to be modified. Protection key types are designated with letters from A to E;
the appropriate letter is shown on the selection list. For network protection, the Select key
field contains the relevant item describing a network key.
NOTE: Take note that protection keys may be connected simultaneously ONLY if
they are of different types (otherwise the keys can be damaged).
After a protection key has been selected, its number and identifier are read automatically (and
entered to the edit fields: Number and ID). For network protection, zero value is entered to the
ID field.
The Select key field enables as well, support of the total number of local keys A, B, C, D and
E; the installation report generated for such a setting corresponds to all the options available
on these keys.
The Source field contains options used to provide codes changing protection settings saved
on the protection key chosen earlier. The user should select the source from which the
information about modifications in the protection should be gotten (Internet, CD, delivered file
or entered manually), and next, press the Apply button in the bottom part of the dialog box. In
the Personalization edit field the user may enter program authorization (a character string
displayed, among others, on the Robot program vignette); the maximal name length to be
given cannot exceed 32 characters. To save the modified personalization on the protection
key, the Apply button in the bottom part of the dialog box should be pressed.
There are the following buttons included in the bottom part of the Protection - Settings
dialog box:
Local / Network pressing this button opens the Protection - Local/Network dialog box
(see the description below)
structures.com
28
Refresh - pressing this button causes rereading of the connected local keys and network
protection settings; the option is useful if there are several keys of the same type (e.g. A).
Since many keys of the same type cannot be connected simultaneously, all operations
on the keys have to be performed one by one. It can be done by replacing the key and
pressing the Refresh button without exiting from the program
Report - pressing this button starts generation of a report containing the current
protection settings for a selected key
Test - pressing this button runs a protection test. The testing procedure verifies the
protection system checking all the local keys and active network licenses (activated by
means of the Local / Network option). A result of the test taken is a text file with
information which should allow identification of possible protection problems.
If problems with the protection system occur, the user should press the Test button and
send the file with the test result to the RoboBAT technical support.
Apply - pressing this button modifies the protection
Close - pressing this button closes the Protection - Settings dialog box.
The dialog box shown in the figure below is displayed on the screen after pressing the
Local/Network button in the Protection - Settings dialog box.
The above dialog box includes options allowing configuration of local and network protection
for all the RoboBAT products.
The top part of the dialog box contains the Local / Network field concerned with all the
programs. If the Local option is chosen only, then the program will not be looking for the
network key; if the Network option is set only, the program will not be checking the local key.
If both options are selected (i.e. Local and Network), then first the key connected locally to a
computer will be checked, if there is no such one, the network key will be checked.
structures.com
29
In the field located to the right of the Network option, the user may indicate the folder
containing the ~TMP folder with a tmp.pmn file including a definition of network protection
settings; it can be done either by entering the path directly to the edit field or by indicating the
folder location after pressing the (...) button.
Pressing the Apply button in the bottom part of the dialog box saves the modified settings.
Below are the following fields located: Local / Network - RCAD and Local / Network - ESOP.
They include options similar to those described above located in the Local / Network field.
There is, however, a possibility of independent protection configuration for RCAD program
(the same settings apply to both RCAD Steel and RCAD Concrete) and ESOP program.
Thanks to that, the user may buy a local Robot program license as well as network RCAD
program license and work simultaneously with both programs on the same computer.
For example, to enable on one computer, the local Robot program protection and the network
RCAD program protection, the following operations should be carried out in the dialog box
above:
switch on the Independent settings for RCAD option
switch on the Network option located below and point the location where ~TMP folder
with the appropriate tmp.pmn file is contained
press the Apply button in the bottom part of the dialog box to save the changes made.
Similarly, the user may set the protection for the ESOP program independently of the
protection settings for the remaining programs.
After closing the Protection - Local/Network dialog box, the user returns to the main
Protection - Settings dialog box where the information about the available protection keys is
automatically updated.
1.3 Generation of Installation Report
Once program installation is completed and the program is activated, an installation report
which presents information concerning Robot program will be automatically generated. If the
report is not printed by a user then, there are several ways of installation report (protection
report) generation. To start generation of installation report, a user should choose one of the
possibilities presented below, which are provided by the program:
1. in the Robot program select Help / Installation report option from the menu; in the
Protection Settings dialog box press the Report button
2. from the START menu of the Windows system run the option: Robot / Tools / Protection
parameters; in the Protection Settings dialog box press the Report button.
NOTE: Generation of installation report may take some time; once it is completed, a
text editor will appear on the screen to present the generated report.
structures.com
30
2. INTRODUCTION
2.1 General Description of the Program
Robot Millennium is an integrated graphic program used for modeling, analyzing and
designing various types of structures. The program allows users to create structures, carry
out calculations and verify results. Unlike other finite element programs, Robot Millennium
also allows to prepare a calculation report for the designed and calculated structure.
The most important features of Robot Millennium are listed below:
complete graphic structure definition in the graphic editor (user may also load e.g. DXF
format files containing the structures geometry which have been prepared in a different
graphic program)
possibility of graphic display of the designed structure and display of various calculation
results (forces, displacements, simultaneous work in several windows on the same
screen, etc.)
possibility of calculating (designing) the structure while designing another structure
(multithreading)
possibility of carrying out static and dynamic structure analysis
possibility of assigning bar types during the structure creation instead of assigning them in
the code modules
possibility of an arbitrary printout composition (calculation notes, screen capture, printout
composition, transferring objects to other programs).
Robot Millennium is made up of several modules that are each responsible for a specific
step in the structure design (creating the structure model, calculating the structure,
designing). The modules work in the same environment.
Once Robot Millennium is activated (click on the appropriate icon on the desktop or choose
the appropriate command from the taskbar), the window shown below will appear on the
screen. The window is used to select the type of structure that will be designed; load an
existing structure or choose the structure design module.
NOTE: During first activation of Robot program an installation report is generated
that presents information concerning Robot program. Generation of
installation report may take some time; once it is completed a text editor will
appear on screen to present the generated report.
structures.com
31
Individual icons found in the above window indicate:
- 2D frame, - shell,
- 2D truss, - plane stress structure,
- grillage, - plane deformation structure,
- 3D truss, - axisymmetric structure,
- 3D frame, - volumetric structure (solid)
- plate,
NOTE: The axisymmetric structure type models a solid generated due to rotation
around the rotation axis, by means of a 2D, vertical section through a solid
(see the drawing below). It is assumed that the global Z axis is a vertical axis
of a solid; half of a section is defined on the positive side of the X axis. To
mark the position of the section vertical axis in an axisymmetric structure, in
the structure view auxiliary structural axes with X coordinates equaling 0 have
been added.
structures.com
32
These are found in the top two rows and are used to select the structure type:
the next six icons are used to design concrete elements of the structure:
R/C beam design, foundation design,
R/C column design, continuous footing design,
R/C plate design deep beam design,
the next icon is used to select the parametric structures;
the next two icons are used to perform:
steel connection design section definition (solid or thin-walled sections);
the penultimate icon allows the user to open the previously created structure file.
the last icon is used to open a new project
The above presented dialog box allows one also to select one of the following icons:
viewer of section databases
editor for creating templates of plotter drawings.
NOTE: When the cursor is positioned on an icon, a short description of its use is
displayed.
Once one of the options listed above is chosen, Robot Millennium parameters are adjusted
to the functions of the chosen code module or the selected structure type. Depending on the
function of the module, an edit viewer allowing for structure editing or a layout customized to
the module function (mostly pertains to code modules), will be displayed. The main elements
found on the screen in most of the system modules are shown below. As an example, the
initial Robot Millennium work screen is presented.
structures.com
33
The above screen may be divided into several sections:
top bar where basic information pertaining to the task is given (project name, structure
calculation data: available results, unavailable results, results during analysis, etc.);
menu and toolbars (an additional toolbar at the right side of the screen containing the
most frequently used icons) and the Robot Millennium layout selection list;
selection lists of the following: nodes, bars, load cases and modes;
a graphic viewer (graphic editor) which is used for structure modeling and visualization;
field at the bottom of the screen where the following information is presented: names of
the open viewers, cursor location coordinates, used units and several options which open
dialog boxes (Display, Snap settings) or display information about the available
resources.
Icons in the lower left part of the layout allow one to:
select the cursor mode (snap settings)
opening the Display dialog box
restore the default attributes of the structures presented on screen.
In case of volumetric structure definition three additional icons appear in the lower part of the
screen:
without shading (none)
shading
quick shading.
structures.com
34
In the bottom left part of the screen there is the icon (e.g. icon) which presents
symbolically the plane in which the view of a defined structure is shown. After pressing this
icon, the program opens the View dialog box.
There is a possibility to choose work in:
2D view (the 2D button)
2D view - deep projection (the 2D/3D button)
3D view (the 3D button).
When the 2D and 2D/3D buttons are pressed, the following become available: selection list
provided under these buttons and two buttons: and (these buttons are also located on the
icon in the bottom right corner of the screen). Pressing the () button causes moving to the
next (previous) level of defined structure axes (the level means structure axes defined for
any X, Y or Z axis). The selection list enables choosing any existing level of structure axes (it
may also be identified by the structure axis name).
After switching on the 2D/3D option, the XY, XZ and YZ buttons are accessible and enable
selection of a work plane. The selection list provided under these buttons, contains available
structure views (top, bottom, etc.; the views SW, SE, NW, NE are isometric and show the
structure as seen from the following directions: southwest, southeast, northwest and
northeast, respectively).
NOTE: The data provided in the field in the bottom part of the layout depend on the
module in which the user works at the given moment (e.g. slightly different
information is provided in the modules for designing RC structure elements).
Once one of the code module icons is chosen (reinforced concrete design, steel design, steel
connection design) the set of viewers and tables corresponding to the module functions are
activated. Additional information pertaining to Robot Millennium layouts is provided in
Chapter 2.2.1.
2.2 General Rules
To begin the description of work in Robot Millennium, several general rules should be listed:
New nodes are created automatically during bar definition. If a bar is created based on
existing nodes, new nodes will not be generated.
When a bar is deleted its nodes remain in place.
Once nodes, bars, panels and solids are determined other features such as supports,
sections, panel thickness and loads must be defined and assigned to them. It is also
possible to invert the order of properties selection in the design process.
The bar type may be assigned while defining the structure.
The UNDO option may not work with some edit operations.
structures.com
35
2.2.1 The Layout System
Robot Millennium has been equipped with a layout mechanism that simplifies the design
process. The layouts in Robot Millennium are specially designed systems of dialog boxes,
viewers and tables that are used to perform specific defined operations. Layouts available in
Robot Millennium were created to make consecutive operations leading to defining,
calculating, and designing the structure easier. In order to avoid damage to the layout system,
the dialog boxes and tables that open once a layout is activated may not be closed. The state
of the dialog boxes is saved at closing; the dialog boxes in the layout should re-open in the
same state as at closing.
Layouts are available once the selection list found in the upper part of the screen is chosen
(see the figure below).
The list, that is shown below opens after clicking in the layout selection field. It is a list of
standard layouts available in Robot Millennium. The list does not include all layouts defined
in several layout groups. Only layouts used for the bar structure model definition are
presented on the drawing below.
The layout order and arrangement was created to suggest the next design step. It is not
necessary to define the structure according to the layout order. This may be done in any order
chosen by the user; the layout system was introduced in such a way that Robot Millennium
structure definition is intuitive and efficient. All Robot Millennium operations may also be
performed without using the defined layouts. As an example of using layouts, the screen
arrangement opened after choosing the BARS layout is shown on the figure below.
structures.com
36
The screen may be divided into three main parts:
the graphic viewer where the structure is defined,
the Bars dialog box (in the case of a different layout, it will be a different dialog box) used
to define the consecutive structure bars,
a table operating like a spreadsheet where data pertaining to the defined bars is shown,
(in the case of a different layout, it will be data concerning other objects such as: nodes,
loads, supports, etc.). The table allows you to edit the entered data. It is also possible to
copy the contents of the table into other programs (i.e. MS Excel).
By choosing the consecutive layouts defined in Robot Millennium, the user may easily
define, calculate and design the structure. Once the structure is chosen, it may be necessary
to modify some bars (i.e. by changing the bar section); the recalculation of the entire structure
could be necessary. The layout system greatly simplifies and speeds up these stages of
analysis, design and modification.
2.2.2 Menu, Context Menu, Toolbars
The Robot Millennium system menu consists of two parts: a text menu and toolbars with
appropriate icons. They can be used interchangeably, according to the users needs and
preferences.
Both are displayed in the same way - as a horizontal bar at the top of the screen (additionally,
for most layouts in the Robot Millennium system, another toolbar is displayed on the right
side of the screen). Basic options available within the modules are accessible both from the
text menu and the toolbar. Though contents of the text menu and toolbars for successive
modules vary, main options are always available regardless of which module is active. The
structures.com
37
figure below illustrates both types of menus; the main menu that appears once the START
layout is selected is shown as an example.
Text menu
Toolbar
Clicking with the left mouse button on any command from the text menu results in opening a
corresponding submenu with detailed options. All commands are of the text type (activated by
the command name).
Each toolbar contains options presented as icons. The main menu contains only basic
options. Clicking with the left mouse button on some icons from the main menu results in
displaying an additional submenu with groups of related options (structure definition,
operations used for structure edition on the screen, tools). Clicking on other icons from the
main icon menu results in executing given operations (save, print, preview, copy, initial view
etc.) or opening appropriate dialog boxes (analysis types).
The text menu is organized in a hierarchical order. To select an option, place the cursor on
the option name and click with the left mouse button. If the command is already highlighted
the option can also be selected by pressing the <Enter> button. Another way to execute an
option is the hot-key system (by pressing the underlined letter). Arrows from the keyboard can
be used to select options in the text menu.
Once one of the options from the main menu is selected, the appropriate submenu with a
group of related options will be displayed. For some options from the submenu further
submenus are available.
Toolbars are organized in a similar way. Clicking with the left mouse button results either in
executing a given operation or opening a corresponding submenu with a group of related
options.
For example, in order to open the Structural Axis dialog box to define structural axes, it is
necessary to select:
the Geometry / Structural Axis option from the menu
the icon from the side toolbar of the START layout.
Both operations open the same dialog box.
NOTE: In the Robot Millenniumsystem, menus are adjusted to successive modules
(structure definition, results preview, design). The menu that is currently
presented on the screen corresponds to the active (highlighted) viewer, table
or graphic editor. In order to change the menu activate another viewer or
table by clicking on it with the left mouse button.
The menu, context menu, hot keys and toolbars defined in the Robot Millenniumsystem can
be modified (with the Tools/Adjust command available from the menu). Dialog boxes for
adjusting menus/context menus, hot keys and toolbars will be opened and can be modified
according to the users needs.
structures.com
38
While working with the graphic editor or a table, clicking the right
mouse button opens an additional context menu containing the
most widely used options.
The context menu that opens while working in the viewer of the
START layout is shown as an example (2D frame).
2.2.3 Preferences and Job Preferences
The two options, Preferences and Job Preferences, allow the user to set program
parameters in the Robot Millennium system. The Preferences dialog box presented below
is used to define basic parameters in the program. It is available from:
the menu by selecting: Tools / Preferences
by selecting the Preferences icon located in the TOOLS toolbar.
The above dialog box can be divided into several sections, namely:
the upper section - containing four icons and a field for selecting the preference file.
The name of the currently used preference file is displayed in the upper partr of the dialog
box. This field allows the user to select a previously created preference file. Pressing the
arrow at the end of this field opens a list of previously created files where the appropriate
one can be selected
structures.com
39
the left section - containing the tree from which one of the Robot Millennium system
preference options (listed below) can be selected with the mouse:
languages - selection of regional settings (definition of the country whose codes
materials and regulations - e.g. code combination regulations - will be used during the
design process, calculations and structure design) and working and printout language
general parameters (saving parameters, number of recently used structures, sound
on/off etc.)
display parameters (colors and fonts for screen components)
toolbar and menu (menu type and the type of toolbars)
printout parameters (colors and fonts for printouts, scale and symbols, line thickness)
protection parameters (protection, authorization) - for changing the system protection
and authorization modifications are made in the Protection Settings dialog box
(see chapter 1)
advanced - options that allow clearing the TEMP folder on the computer and options
enabling selection of edit operations based on the ACIS kernel to perform Boolean
operations
COM interface - presentation of the registered additional programs/modules
the right section - containing the field that is updated any time a new command from the
option tree is selected.
The Job Preferences dialog box, presented below, allows you to define general program
parameters to be used in a given task. This dialog box is available from:
the menu by selecting: Tools / Job Preferences,
the TOOLS toolbar by clicking on the "Job Preferences icon .
The structure and operation of the dialog box is similar to those used in the Preferences
dialog box.
The following options are available in the above dialog box:
number units and formats (dimensions, forces, possibility of unit edition),
materials (selection of material set, according to the country and the possibility of creating
user-defined material),
structures.com
40
section database (selection of the appropriate database with member sections),
vehicle databases (selection of the appropriate database with code-defined vehicles),
load database (selection of the appropriate database with loads used for composing the
loads acting on structure elements),
soil database (selection of the appropriate database containing the list of soils),
codes (selection of codes to be used during the structure design steel, R/C, connection
design, etc.); while selecting codes of code combinations, the user may run the editor of
code combination regulations it is done by pressing the () button located to the right
of the list for selection of code combination regulations
structure analysis parameters (selection of the static analysis method and definition of
basic parameters for dynamic and non-linear analyses; selection of analysis types,
possibility of saving results for seismic analysis combination of seismic cases)
parameters for generation of surface finite element mesh for plates and shells.
2.2.4 Selection and Filters
The selection options are commonly used while working in different modules of the program.
They allow the user to define the node and bar list (only!) which will be used in further
selection. Once the selection is made, a set of bars or nodes is created and it is active until
the next selection is made.
The selection can be made:
graphically on the screen (by selecting one of the options
from the Edit / Select Special submenu or by selecting the
Select option from the context menu)
from the Selection dialog box (using the Edit / Select option)
from the list of nodes and bars located under the menu (any
number of bars/nodes for selection can be entered in this
field) or from the table by highlighting appropriate lines.
Selection is the same in all tables and viewers. This means that
once the nodes/bars are chosen in the graphic viewer, the
selection will also be presented in the bar/node table. As an
analogy, selection of bars/nodes in the table will result in
highlighting these objects in viewers. The dialog box for
selection (called up by the Edit / Select command) is presented
beside.
The above dialog box allows for the selection of nodes/bars in the structure. It can be
performed either by using buttons located in the upper part of the dialog box (All, None,
Inversion, Previous) or from options in the Attributes tab. Selected bars or nodes will be
entered to the field located in the upper part of the Selection dialog box. In particular cases,
the dialog box can be opened in a special mode, i.e. selection of bars, only. The shape of the
bottom part of the Selection dialog box depends on what object is indicated.
To perform a selection:
indicate the object (node, bar, load case, mode), which will undergo selection
structures.com
41
enter the corresponding numbers of the selected objects in the appropriate fields, or use
the options found in the lower portion of the dialog box.
At the end of the field where the list of selected nodes is
introduced, the user finds a field for determining the
definition of selection mode. The field may be checked
out or not. If the field is not checked out, the field of node
selection containing a support will be filled with node
numbers corresponding to the nodes where the support
is applied (see the figure beside).
If the selection field is checked out, the field of node
selection containing a support will be filled with the
following text: Support = any (see the figure beside).
The difference in performance between the two buttons found in the dialog box is as follows:
once the button is pressed, selected objects will be added to the current selection
once the button is pressed, the current selection will be removed and numbers of the
selected objects will be entered.
The Selection dialog box is used to define a group of nodes or bars. Once the selection is
made (in any of the described ways), switch to the Group tab. The lower part of the dialog box
will be displayed as below:
Once the selection is made and the button is pressed, a small dialog box will be open.
The color and the name of the group is to be entered there. After pressing the Apply button,
the group of a given name and color will be added to the field presented on the figure above.
In the Selection dialog box the user may define selection of objects by means of a defined
grid of structure axes. The Selection dialog box includes the additional Geometry tab
available, shown in the drawing below.
structures.com
42
The above dialog box contains options enabling selection of objects positioned on the
selected structure axes; the user may determine positions of the beginning and end structure
axes (in all directions of the coordinate system), between which objects are to be selected.
Structure elements may also be selected by means of structure axes after choosing the
following command from the menu: Edit / Select Special / Structure Axes.
Object selection in the program should be distinguished from the filter. Selection is used to
specify which objects (nodes, bars, cases and modes) are to be displayed whereas filters are
different for each viewer and table. Filtering can be done in tables by opening the Filters
dialog box (using the View / Filters command). In case of graphic viewer, filtering is applicable
only for load cases and modes.
To perform filtering in edit viewers:
open the Selection dialog box and choose load cases or modes,
select the appropriate load case or mode from the list of cases and modes located under
the menu.
The program also provides an option (Result Filters) which is used for global selection of
results obtained for nodes, bars, etc. defined in a structure. Description of this option is
presented in the chapter 4.3.
For description of the filter mechanism in tables see chapter 5.2.
Simple, basic operations allowing one to perform selection are discussed in examples:
if an object is indicated (node, bar, panel), it is selected
indicating an object with a mouse with the CTRL or SHIFT button pressed results in
adding objects to or removing them from the groups of selected objects (as in the
Windows operating system)
indicating again an already selected object results in entering into the mode of object
position modification within the structure (in this way, one may move, for instance, a bar
to another position)
selection of objects may be performed by means of moving the mouse cursor in the
graphical viewer with the left mouse button pressed; however, there is a difference in the
selection results:
if the selection with window-cursor is
performed from the upper left corner,
only the bars (objects) will be selected
that fall entirely within the defined
cursor-window;
if the selection with window-cursor is
performed from the lower right corner,
all the bars (objects) that enter,
however partially, into the window will
be selected the figure below shows
the difference.
The program provides user with the possibility to select elements of a created structure which
will be graphically selected on a graphical screen. Graphical selection filter option allows for it.
A dialog box can be opened once the Edit/ Select special/ Graphical selection filter command
from the menu is chosen. After choosing this option the dialog box shown on the drawing
below appears on the screen.
structures.com
43
The above dialog box contains options which enable selection of structure elements to be
selected on the graphical screen. If this option (structure element e.g. nodes) is not checked
in the dialog box, then during graphical selection on the graphical screen structure nodes will
not be selected. If this option is checked ( symbol appears), then during graphical
selection on the graphical screen structure nodes will be selected.
2.2.5 Display of Structural Attributes
The Attribute Display dialog box is used to select the structure attributes that are to be
displayed on the screen. This dialog box may be opened by selecting the View/Display
command or by selecting the icon located in the lower left corner of the screen.
Once the View / Display option is selected, the dialog box shown below will be displayed.
Once a tab is selected, the appropriate option list is
activated. The following tabs can be selected from this
dialog box: Structure, Sections, Loads, Advanced, Finite
Elements, Hidden Lines, and Others.
Each tab contains a group of options that allow for the
presentation of the appropriate structure attributes on the
screen. For example, options appearing in the dialog box
once the Structure tab has been activated are shown above.
In order to display structure attributes on the graphical
viewer, one should indicate the selected attributes on the
tabs in this dialog box, and then press the Apply button.
The following buttons are shown in the lower part of the
Display dialog box:
All - pressing this button results in selecting all the
options in the Display dialog box (all the attributes
defined in the structure will be displayed on the screen)
None - pressing this button results in not selecting any of
the options in the Attributes Display dialog box (none of
the attributes defined in the structure will be displayed on
the screen)
structures.com
44
- pressing this button opens the Default Settings dialog box where the following
options can be chosen:
Save current settings as default if this option is selected, then the settings chosen in
the Display dialog box will be adopted as default ones
Restore factory defaults - if this option is selected, then the factory (input) default
settings are restored in the Display dialog box; the factory settings are the option
settings in the Display dialog box delivered together with the Robot program
Default - pressing the button results in displaying the basic attributes of the structure. The
following options belong to the default set:
on the Structure tab:
Structure
on the Other tab:
Structure axes
Descriptions of structural axis
Grid
Ruler
Objects out of plane
on the Finite Elements tab:
Panel Contours
Panel Numbers and Description
Complex panel descriptions
Characteristic points
Contour components
Finite Elements
the option None found on the Hidden Lines tab
and the symbol size option - allowing the selection of the size of the structure attributes
presented on the screen (size scale: 1-10) the default scale is equal to 3.
NOTE: Pressing the icon located in the lower left corner of the layout allows one
to restore the default structure attributes to be presented on screen.
2.2.6 Lists Used in the Program
While working in the Robot Millennium program, the user may face situations (support
definition, applying sections to bars, load definitions, etc.) where it may turn out helpful to take
advantage of the possibility of creating lists of certain quantities, aimed to simplify the process
of selecting nodes, bars, load cases, panels, and objects. The user may define lists of
quantities in several ways described below:
by specifying all the numbers to be contained in the list, e.g.:
2 3 6 7 12 14
by specifying in shortened notation the range of numbers to be included (by means of
commands "TO" and "By"):
6TO10 denotes the set of numbers: 6 7 8 9 10
6TO10By2 denotes the set of numbers: 6 8 10
by specifying in shortened notation the sequence of numbers to be included with the
commands Repeat and Step
4R3 denotes the sequence: 4 5 6 7 (1 is the default step),
4R3By2 denotes the sequence: 4 6 8 10 (2 is the value of step).
by specifying in shortened notation the sequence of numbers to be included with the
command "EXClude":
1TO58EXC44 49 52
denotes the following list of elements: 1TO43 45TO48 50 51 53TO58
The "EXClude" command may by used only once during list preparation.
The following syntax is valid when a list of structure components is generated by means of
the Extrude, Revolve, Extrude along polyline options:
objectnr_objectelement(list_of_object_elements),
structures.com
45
where:
objectnr - the number of the object that is to be extruded or revolved
objectelement - three possibilities are available in the current program version: side, edge
and object reference after modification (ref)
list of object elements - list of components (the rules presented above are valid for these lists
as well).
Here is an example of combining operations on lists:
2_ref(1,2,5), 1_side(3to7), 4_edge(5to8,11).
The syntax is identical with the description of object components on the graphical viewer.
2.2.7 Common Features of Dialog Boxes (Graphic Cursor,
Calculator)
In the Robot Millennium system various mechanisms have been introduced to make
structure definitions simpler and more efficient. According to the type of operation performed,
the mouse cursor changes its shape to:
hand - in the selection mode,
cross pointer - during node and bar definition,
shape of the appropriate feature when assigning various features (supports, sections,
releases etc.) to the structure bars.
Additionally, during structure node/bar definition the coordinates of the cursor location in the
graphic viewer are presented in the appropriate field: Nodes or Bars. The coordinates change
with each move of the cursor.
The active field is highlighted in the dialog box. For fields accepting only one value, the color
changes to green, yellow or red. The green background of the field indicates that the entered
value is correct whereas yellow or red indicates that it is incorrect. A yellow background
represents a value outside the recommended range that is still admissible whereas red
represents an inadmissible value. All edit fields in the program accept number formats set in
the Preferences dialog box (units, number format). In the case of fields where only one value
is to be entered, it can be entered with an arbitrary unit. Once the "=" key from the keyboard is
pressed, this unit will be automatically converted to the default unit used in the Robot
Millennium system.
The Robot Millennium system also provides the user with a calculator. It is accessible by
selecting the Tools/Calculator command or by double clicking on the edit field in the dialog
box. An arithmetic expression calculator is available in the edit fields. Once an expression is
entered in the edit field and the "=" key is pressed, its value will be calculated.
structures.com
46
2.3 Sign Convention
The positive orientation of forces and displacements agrees with the positive orientation of
coordinate system axes. Positive orientation of angles, rotations or moments in the local or
global coordinate system is determined on the basis of the right hand rule. This convention
defines signs of external forces, nodal forces, displacements and rotations. All these values
are used in structure definition, during structure calculations and results display.
A different convention can be assumed in the program to describe the signs of internal forces
operating within each element. The program sign conventions concerning internal forces
operating in bars and planar finite elements will be discussed separately in details in sections
3.16 and 3.17.
Sign Convention for Bar Elements
Positive directions of forces for the above described sign convention are presented
schematically in the drawing below. In the program, the sign convention for bar elements is
based on the convention used for sectional forces. According to this convention, sectional
forces are of the same sign, if they produce the same result at the end of a bar as the result
produced by positive nodal forces (the forces whose direction agrees with the direction of the
axes of the local coordinate system) applied at the beginning node of the bar. Therefore,
compression forces are positive, while the tension forces are negative. Positive bending
moments MY produce tension in the beam fibers that are located on the negative side of the
local coordinate axis z. Positive bending moments MZ produce tension in the beam fibers
that are located on the positive side of the local coordinate axis y.
The positive directions of force vectors in the above-described convention are presented
schematically in the figure below.
structures.com
47
Sign Convention for Planar Finite Elements
The local coordinate system for an entire finite element is not required, as every node has its
own local system. It is important, however, to know the orientation of the normal vector,
perpendicular to the finite element surface. Results obtained in program Robot Millennium
for planar finite elements do not take into account nodal forces. One can obtain only internal
forces and stresses in an element for planar FEs. They are determined on the basis of the
location with respect to the local normal vector and to the tangential to the cross-section.
Results obtained for planar FEs are presented in local coordinate systems that may be
defined and changed by the user in any moment of results presentation. For instance, positive
orientation of forces and the corresponding stresses in a node are presented in the figure
below for X axis as the reference direction.
Z
Y
X
Mxx
Myy
Nxx
Nyy
xx
xx
yy
yy
Nxy Nyx
6- and 8-node finite elements are defined in the following way in text file syntax: first - nodes,
and then nodes in the centers of particular finite element edges (see the figure below).
A different principle holds during presentation of results for 6- and 8-node finite elements: the
nodes creating an element are listed in the following order: node, node in element center,
node, etc. (see the figure below).
structures.com
48
Sign Convention for Volumetric Finite Elements
In the Robot program volumetric structures are modeled with the use of isoparametric
volumetric finite elements with displacement field approximation based on 1st order shape
functions. The sign convention for volumetric elements is presented schematically in the
drawing below. The convention is shown for stresses; the stresses displayed in the drawing
have positive signs.
structures.com
49
2.4 List of Shortcuts
In order to Press
select all Ctrl + A
copy a text or a drawing Ctrl + C
open a new project Ctrl + N
open an existing project Ctrl + O
start printing Ctrl + P
save the current project Ctrl + S
cut a text or a drawing Ctrl + X
repeat the last operation Ctrl + Y
paste a text or a drawing Ctrl + V
undo the last operation Ctrl + Z
display the 3D view of a structure (3D XYZ) Ctrl + Alt + 0
project a structure on XZ plane Ctrl + Alt + 1
project a structure on XY plane Ctrl + Alt + 2
project a structure on YZ plane Ctrl + Alt + 3
zoom in the structure view on screen Ctrl + Alt + A
display the initial view of the structure (defined by the initial angles and
scale)
Ctrl + Alt + D
exploded view of structure elements (on/off) Ctrl + Alt + E
zoom window Ctrl + Alt + L
turn on/off section drawing display Ctrl + Alt + P
screen capture Ctrl + Alt + Q
zoom out structure view on screen Ctrl + Alt + R
turn on/off section symbol display Ctrl + Alt + S
rotate continuously around the X axis Ctrl + Alt + X
rotate continuously around the Y axis Ctrl + Alt + Y
rotate continuously around the Z axis Ctrl + Alt + Z
delete a text or a drawing Del
call ROBOT Help system for the active option in the active dialog box F1
call text editor F9
reduce structure attributes (supports, numbers of nodes, bars, loads)
presented on screen
PgDn
enlarge structure attributes (supports, numbers of nodes, bars, loads)
presented on screen
PgUp
structures.com
50
2.5 Snap Settings
The cursor movement on screen depends on the cursor move mode. The parameters of
cursor movement may be set in the Snap Settings dialog box. The dialog box may be
activated by selecting:
Tools/Snap Settings command from the text menu
Snap settings icon located in the bottom left screen corner.
The upper part of this dialog box contains three basic modes of cursor movement:
Nodes - it allows the user to set the mouse cursor
only in the already existing structure nodes. It should
be noted that the cursor in this mode may be
snapped to objects like polylines, contours, etc. To
make the program do so, one should use the Objects
option.
Structure axes it allows the user to define nodes
only in the points of intersection between structure
axes defined by the user (NOTE: the axes must be
visible on screen).
Grid - it allows the user to define nodes only in the
grid points presented on screen (NOTE: the grid must
be visible on screen). The grid step (distance
between the points) may be modified by the user in
the Grid step definition dialog box.
The central part of the dialog box contains options allowing the user to determine the cursor
settings for objects (checking out the snap settings for bars, lines, polylines). There are two
snap settings available:
endpoint - the cursor is snapped to endpoints of bars and object segments. Switching
endpoints off results in making the Midpoint option inactive (it cannot be accessible
without setting the endpoints)
midpoint the cursor is snapped to the midpoints of bars and object segments
Advanced snap setting options are located below. They allow one to snap the cursor to
intersection points and perpendicular. Among the useful effects of the options, one may
mention the possibility of following vertical and horizontal lines, as well as of finding the points
of their intersection with bars or structure axes. The following modes are available:
perpendicular defines a perpendicular line from the starting point to bars and object
segments.
parallel defines a parallel line from the starting point to bars and object segments.
intersections intersections of bars with segments of objects and ends of brackets
intersections with structure axes - intersections with structure axes visible on screen
(NOTE: structure axes must be visible!)
structures.com
51
intersections with the grid - intersections with the grid displayed on screen (NOTE: the
grid must be visible!)
The bottom part of the dialog box contains three buttons:
Default - pressing this button results in selecting the basic snap settings. The basic snap
settings cover the following options: Nodes, Structure axes, Grid, Objects - Endpoint,
Advanced - Intersections.
All - pressing this button results in selecting all the options available in the Snap settings
dialog box
None - pressing this button results in selecting none of the options available in the Snap
settings dialog box.
structures.com
52
3. GENERAL RULES ON STRUCTURE MODEL
DEFINITION
3.1 Structure Types
In the Robot Millennium system 2-node bar elements used for definition of bar structure,
surface finite elements used for generation of finite element mesh for definition of plates and
shells and volumetric elements used during generation of finite element mesh for volumetric
structures are available. Structures containing different types of such elements can be
defined. The element type depends on the structure type. Currently, the following structure
types are available within the Robot Millennium system:
2D/3D frame and truss,
grillage,
plate and shell,
plane stress structure,
plane deformation structure,
axisymmetric structure,
volumetric structure (solid).
Apart from those mentioned above, the program contains a large library of typical bar, plate
and shell structures, the definition of which is simplified by entering several parameters (see
chapter 3.14).
3.2 Nodes, Bars
The first step in bar structure definition is the creation of nodes and bars. It is not necessary to
define nodes and bars separately; when defining a bar the beginning and end nodes are
automatically created. Therefore, only the bar definition method will be described below.
The option is available from:
the menu by selecting: Geometry/Bars or
the toolbar by selecting the Structure Definition icon
BARS layout (in the case of NODES and BARS layout the screen will be divided into
three parts: graphic editor used for structure definition, the Nodes or Bars dialog box and
a table displaying defined structure nodes or bars).
Once the option is selected, a dialog box shown below will be displayed on the screen.
structures.com
53
Apart from the information regarding the number, beginning and
end node of the bar, the following additional bar properties can
be set in the above dialog box:
bar type - it is used in code calculations of structure bars
(design, verification of structure members with respect to
the code criteria)
bar cross section
material (material is assigned to sections).
NOTE: Two buttons (...) (right of the Bar Type and Section
fields) are located in the upper part of the dialog box.
Pressing them results in opening the New Bar Type
or New Section dialog box in which a new bar type
or new section may be may be defined. The defined
bar types or sections are then applied to the
appropriate lists of bars or sections.
A bar element can be defined in several ways as described below:
1. Enter a bar number and the coordinates of its origin and end point in the dialog box (or
select bar properties, if necessary), then press the Add button.
2. Click with the left mouse button on the Beginning field, switch to the graphic viewer and
using the left mouse button click first on the point indicating the beginning of the bar
element and then on the point which will be its end.
3. Method combining the two approaches (text and graphic methods) are outlined above.
There are also two options available within the program: Node properties and Bar properties.
The Node properties option is used to present the basic data and calculation results for a
single structure node. The Node properties dialog box does not allow modification of node
parameters.
NOTE: The parameter presentation is possible only for a single node. If several
nodes are selected in the structure view, the Node properties dialog box
presents data concerning the node with the lowest number.
The option becomes available after highlighting (selecting) a structure node:
from the menu by selecting the following command: Results/Properties/Node properties
or
from the context menu (opened in the graphical viewer by clicking with the right mouse
button) by choosing Object properties command.
The dialog box bar presents, beside the dialog box name (Node properties), the following
data:
number of the selected node
number and name of the selected load case.
Node properties dialog box consists of three tabs: Geometry, Displacements and Reactions.
structures.com
54
The Node no. edit field allows for node selection:
graphically, by clicking the structure node
by typing the node number in the edit field.
NOTE: It is not possible to edit node number in this dialog box.
The Geometry tab presented on the figure above contains the basic information concerning
the selected node. Apart from the node number, the remaining tabs present displacement or
reaction values calculated for a selected node and load case. Displacements or reactions are
presented in a table form. If the active selection contains more than one case, the table
shows extreme values of displacements.
NOTE: The number of columns in the table of displacements or reactions depends
on the number of nodal degrees of freedom for a selected structure type.
The Bar properties option is used to present the basic data and calculation results for a single
structure bar. The dialog box presenting bar properties allows for modification of certain bar
properties (type, section, material).
NOTE: It is possible to present parameters only for a single bar.
The option becomes available after highlighting (selecting) a bar:
from the menu, by selecting the following command: Results/Properties/Bar properties or
from the context menu (opened in the graphical viewer by clicking right mouse button) by
choosing the Object properties command.
Selecting this option opens the dialog box which may consist of five tabs: Geometry,
Properties, NTM, Displacements and Code check. First two tabs present general information
concerning bar geometry and properties of bar cross section. The NTM and Displacements
tabs, in the upper part of the dialog box, contain the diagram of the quantity selected in the
Diagram field. The diagram may present a diagram of only one quantity. Diagrams of the
following quantities are available: forces FX, FY and FZ, moments MX, MY, MZ, stresses
Smax and Smin and displacements. The diagrams are redrawn if a structure load case is
changed.
NOTE: The number of available quantities depends on a structure type.
structures.com
55
Options available on the Code check tab in the Bar Properties dialog box are used for quick
verification of load capacity of a bar section.
NOTE: If no structure calculations have been performed (the structure view bar
displays the following message: Results (FEM): none or unavailable), the tab
is not available.
Contents of the Code check tab depend on a selected member type: the quantities presented
for steel, aluminum and timber member differ from those presented for an RC member
(calculation of theoretical (required) reinforcement area).
The table may present either values of a selected quantity (e.g. displacement) or extreme
values of a selected quantity.
If a mouse cursor is set in the at point table field, and then it is moved to the diagram of a
selected quantity located in the upper part of the dialog box, a vertical line will appear that
allows defining the value of the coordinate for which the relevant values will be presented in
the table
If the mouse cursor is set in the for a bar table field, and then it is moved to the structure view,
then, indicating other bar with the cursor will result in updating the content of the Bar
properties dialog box, according to the current selection.
3.3 Bar Sections and Materials
Once the structure geometry is defined, bar sections should be specified (if not defined during
the structure bar definition). The option for bar section definition is available from:
the Bars option where one should select the relevant section type in the Section field
the menu - by selecting: Geometry/Sections/Bar Sections or
the Structure Definition toolbar - by selecting the icon or
the SECTIONS AND MATERIALS layout.
Once the option is selected, a dialog box shown below will be displayed on the screen.
The dialog box consists of three main parts:
several icons located in its upper part,
a field containing the list of active sections,
a current selection field and standard buttons.
structures.com
56
The following icons are located in the upper part of the
dialog box:
= - add a new section type,
= - remove a selected section type from the list of active
sections,
= , , and - display a list of active sections as:
large icons, small icons, short list or a long list.
= - allows one to remove from the active list the profiles
that do not appear in a given example of bar sections.
= - allows one to save the section highlighted in the list of
active sections to a section database. There appears a Save
to database dialog box that allows one to select the
database where to save the section
= - allows one to open the Label Manager dialog box.
To describe the process of assigning bar sections a general method of attribute (supports,
offsets, bar type etc.) definitions will be illustrated, as it is the same in all cases.
The process of assigning sections (attributes) to structure bars is divided into two steps:
Section type (attribute type) definition
Click on the New Section icon. If the list of active sections is empty or a new section is
to be added to the active list, two situations are possible:
if any one of the section types is selected, clicking on the New Section icon opens a
dialog box for a new section type definition; the first tab will be called up with the
previously defined fields (except for the Label field) or default parameters will be set;
if any of the sections are selected, clicking on the New Section icon opens a dialog
box for a new section type definition and a tab appropriate for the selected section
type will be called up. All edit fields, except for the Label field will be filled in according
to the section type selected.
It is also possible to open the dialog box for a new section definition by double clicking on the
element from the list of active sections. The New Section dialog box will be opened using the
tab corresponding to the selected section type and all fields will be filled in. After modification
of appropriate parameters the new section type is added (updated) to the list of active
sections by clicking on the Add button or pressing the <ENTER> key. If the label cannot be
changed, the appropriate note will be displayed on the screen. This option in the dialog box
allows an easy modification of the section. The dialog box consists of five tabs: Standard,
User, Tapered, Compound sections and Value. The Compound sections tab is used to define
/ select compound sections (multi-chord sections). Compound sections consist of two or more
chords connected by means of battens or bracings. Such sections are applied as sections of
column piers (C-sections, I-sections, angles) as well as sections of truss bars (set of angles,
most often). Code verification of compound section resistance is carried out analogously as
for solid sections taking account of the equivalent stiffness. The equivalent stiffness includes
influence of battens and slenderness of a single chord. The resistance of the battens or
bracings themselves has to be checked in the verification, as well.
structures.com
57
Assigning a section to structure bars
There are several ways of assigning a section to structure bars (it is assumed that at
least one section type is on the list of active sections):
if no selection has been made before the Sections dialog box is open - assign the
section by selecting the appropriate section type from the active list. Move the cursor
to the graphic viewer to point on the bar (click with the left mouse button) to which the
section is to be assigned. The chosen section is marked (an arrow on the left side of
the section type symbol appears).
The cursor changes its shape to the icon of the chosen section when it is out of the
dialog box (on the graphic screen); when the cursor is on the edit viewer the structure
bar in the neighborhood is highlighted,
if the selection has been made before the Sections dialog box is open once the
dialog box is opened, the list containing this selection is entered into the Lines/Bars
group field. In order to assign the appropriate section to the bars from the list, select
the section type from the active list and press the <ENTER> key or the Apply button.
The section type has been assigned (NOTE: Selected bar list is removed from the
Lines/Bars edit field),
if the selection is to be made when the Sections dialog box is already open - first the
Lines/Bars edit field should be activated (locate the cursor therein and click with the
left mouse button). When the cursor is moved out of the dialog box (to the graphic
viewer) it will be in the selection mode. Therefore, the selection of any structure bar is
possible and the numbers of selected bars will be displayed in the Lines/Bars edit
field. In order to assign the appropriate section to the bars, select the section type
from the list and press the <ENTER> button or the Apply button.
(NOTE: Selected bar list is removed from the Lines/Bars edit field).
To delete an assigned section, the Delete Section (DELETE icon) should be used. It is
always available from the list of active sections in the Sections dialog box. Such a section
type cannot be modified; it is assigned in the same way as the section definition in the
structure.
structures.com
58
NOTE: The description presented below as well as the description of all tabs concern
steel sections (the tabs for timber and aluminum sections look similar). If a
concrete section type (e.g. RC column or RC beam) is selected in the
Sections dialog box, then the New Section dialog box will assume a
different form. Available cross section types include:
RC columns: rectangular, T-shaped, L-shaped, Z-shaped, regular-polygon-
shaped, circular, circle half, circle quarter
RC beams: rectangular, T-shaped, I-shaped.
For example, for an RC beam the New Section dialog box assumes the form shown in the
drawing below.
The dialog box presented above allows the user to:
choose the beam section type (rectangular, T-shaped, I-shaped, T-shaped with flanges
positioned lower) and determine section dimensions
specify section name; by default the program enters a section name composed of a few
letters denoting section type and dimensions of the cross-section
select section color.
The user may select one out of four types of the beam/continuous footing section:
rectangular, T-shaped, I-shaped and T-shaped with lower-positioned flanges. Depending on a
section that has been selected, the dialog box presents parameters determining the section
type chosen. The dialog box shown above determines parameters of a rectangular section.
Similar options are displayed in the dialog box if a T-shaped section is selected. Once a T-
shaped section with lower-positioned flanges is chosen, two additional tabs appear in the
dialog box: Slabs and Cuts.
Switching on the Use tapered section option enables assigning cross-section varying linearly
to a beam by specifying - in the h2 field - a value being the equivalent of the height on the
right end in the selected segment.
structures.com
59
The Reduction of mom. of inertia option allows the user to determine coefficients reducing
sections moments of inertia Iy or Iz in a section definition (RC beam and RC column). The
reduction is connected with a property of a given section and is not a global analysis
parameter. Reduced moments of inertia are displayed (in tables or in dialog boxes) as current
section properties. Reduced properties are considered in static calculations and are passed to
the design modules.
Reduction of moments of inertia for RC sections is applied in static calculations to take the
effect of section cracking into account. This method is allowed by, among others, USA codes
(UBC 1997 point 1910.11.1 or ACI 318-95 p.10.11.1).
The above dialog box contains the Elasto-plastic analysis button (this button is available on
the Standard and User tabs). The options provided in the dialog box that opens after pressing
this button allow defining parameters for the elasto-plastic analysis of a bar with a selected
section.
For a selected section type (e.g. I-section) the division of a section may be determined. The
division depends on a section type; it is usually determined by means of a division number
along the lengths of web and flanges. It is assumed that in case of standard sections, there is
no division along the wall thickness. In the current program version the following material
types are available: elastic-perfectly plastic and elastic-plastic with hardening. The stress -
strain properties for the material types listed are shown in the figures below. A value of the
limit elastic stress has been assumed based on the Re design strength of a given material,
defined in the material database.
Material model:
elastic-perfectly plastic elastic-plastic with hardening
If the elastic-plastic model with hardening is selected, then the E/E1 edit field becomes
available in which the user may enter a value of the plastic hardening parameter defined by
means of a quotient of the material stiffness value (Youngs modulus E) in the elastic range
and the stiffness in the plastic range (in this range the linear material model is adopted).
There are four unloading methods available:
elastic
plastic
damaged
mixed; after selecting this unloading method, the edit field for defining the a parameter
becomes accessible; 0< a < 1.
Additionally, on the SECTIONS/MATERIALS layout the Material dialog box appears. It is
also available after selecting the option from the menu (Geometry/Materials) or pressing
icon. The upper part of this dialog box contains a list of materials available within the program.
Below, a section list consisting of two columns is shown: labels of the defined sections are to
be found in the first column, whereas materials assigned to them are in the second. The list
displayed in the Material dialog box is identical to the list of active sections presented in the
Sections dialog box. Pressing the Save button results in opening a dialog box which allows
for saving material to current material database. The middle part of the Save to Database
dialog box contains a list of materials defined in the program. On entering this dialog box, the
program highlights all the materials that are not saved to the database. Pressing the Save
button results in saving selected materials to the current material database.
structures.com
60
To assign material to a given section:
select the appropriate section (by clicking on it with the left mouse button),
select material from the list of available materials,
press the Apply button.
While assigning sections to structure bars, materials are assigned to them.
3.4 Panels
Mesh of planar finite elements for plate/shell structures is defined in two stages. The first one
consists in defining areas where meshes of finite elements will be generated. The areas are
created by defining their edges (the area contours are defined by means of the Polyline -
contour option). Panels modeling ceilings and walls are defined in the indicated areas.
During panel definition, the panel is ascribed certain properties (thickness, reinforcement
type). The second step (following panel definition and commencement of structure
calculations) consists in automatic generation of a mesh of planar finite elements on the basis
of the parameters selected in the Job preferences dialog box (Meshing options).
Contours of plate/shell structures may be defined by means of the Polyline-Contour option,
available for the two structure types (plate and shell) from:
the menu by choosing: Geometry/Objects/Polyline-Contour or
the toolbar Structure Definition by selecting the icon.
Options allowing one to define methods of line definition are located in the Definition Method
field of the dialog box. Schematic drawings showing the line definition method are shown
below.
Line will be defined using two points: line
beginning and end.
Broken line definition (polyline). The line will
be defined by providing the consecutive points
on the line
Contour definition; it will be defined by
defining the consecutive points on the
contour.
structures.com
61
Once contours are defined, one should determine
panels to be included in the structure. This is done by
the option available from:
the menu using the Geometry/Panels command or
the Structure Definition toolbar by selecting the
icon.
Panel definition includes entering:
the panel number
panel edges (contour) and if defined, edges
(contours) of openings located in the defined panel
and edges of faces. It is possible in three ways:
by entering the internal point of the panel/hole or
by indicating the object number
by selecting the list of finite elements
panel reinforcement type
material (this field is not available; the material
defined for the selected panel thickness is presented
in it)
panel thickness.
If the Face option in the Contour Type field has been selected, then, all options in the
Properties field located in the lower part of the dialog box become unavailable. Selection of
this option causes the object that is being generated, to be defined as a face (without
assigning properties such as reinforcement type and thickness); such a object can be used
during generation of a volumetric structure (solid) - it may constitute a face of such a
volumetric object.
NOTE: Two buttons (right of the Reinforcement and Thickness fields) are
located in the upper part of the dialog box. Pressing them results in opening
the New Bar Type or New Section dialog box appropriately in which a new
panel thickness or plate and shell reinforcement type may be defined. The
defined thickness or reinforcement types are then applied to the appropriate
lists of thickness or plate and shell reinforcement types.
Once panels are defined and structure calculations started, the program creates a mesh of
finite elements according to the parameters selected in the Job preferences dialog box
(Meshing options), The FE mesh is visible only once the FE Mesh in the Display dialog box.
The procedure of creating an element mesh for a given contour may be repeated several
times; it is important to point out that the new planar element mesh will erase the old one.
Two planar FE types are available in the program:
triangular elements (3- or 6-node),
quadrilateral elements (4- 8-node).
In the Robot program it is recommended to use 3- and 4-node surface elements. If 6- and 8-
node surface elements are used for mesh generation, the following options may not work
correctly:
linear releases
Boolean operations (cutting off)
adjusting meshes between panels as well as between panels and bars.
structures.com
62
Functions used during FE mesh creation create the nodes inside the selected area first and
then assign the created nodes to the appropriate finite elements. Nodes inside the area
(contour) may be created through Delaunays algorithm of triangulation or by means of
Coons method.
NOTE: Examples of finite element mesh generation for plate / shell structures are
presented in the appendices at the end of the users manual.
3.4.1 Types of Planar Finite Elements
Delaunay's Triangulation Method
Delaunay's triangulation method may be used to create a FE mesh for any 2D surface. If
holes occur inside the domain, the user should define them as the contour edge. They will not
be taken into consideration during the FE mesh creation. Once Delaunay's method is chosen,
the user should select the contour for which the element mesh will be created.
An example of the FE
mesh created using
Delaunay's method is
shown on the drawing.
Contour Edges Delaunay's Triangulation
The following parameters may be defined for Delaunays method:
mesh generation method: Delaunays method alone or the additional node generation
method (Kangs method - emitters).
Emitters are the nodes defined by the user near which the FE mesh will be thickened. The
thickening parameters are given as Kangs parameters.
= H0 parameters defining the length of the first wave
= Kangs parameters (H
max
, Q)
Specific parameters of Kangs method represent:
1. H
max
- the length of the second-to-last wave before the end of mesh thickening;
2. Q - the relation of the length of the next-to-previous wave.
Coons' Method
Coons' surfaces are 3D surfaces spread over quadrilateral or triangular contours whose
opposite sides are divided into the same number of segments. The shapes of the created
elements correspond to the region on which the mesh is created. The general concept of this
method depends on connecting all points created on the selected contour edge with the
points laying on the opposite edge of the contour. The cutting point of each pair of horizontal
and vertical lines marks the end positioning of the node inside the region (see the drawing
below).
structures.com
63
Once the contour is selected, the user
should define the parameters of Coons'
method, which describe the FE mesh
shape (triangles, quadrilaterals, mixed
element type) as well as the division
parameters: division1 and division 2.
The division parameters describe
number of the elements which will be
created on the first (between the first
and the second contour corner) and the
second (between the second and third
contour corner) contour edge. The
contour edges opposite to the listed
contour sides will be automatically
divided, so that the division corresponds
to the division accepted on the first and
second contour edge. For triangular
regions, the edge division between the
third and first contour corner is the same
as that between the second and third
corner.
In quadrilateral regions, the division
between the third and fourth contour
corner is the same as between the
second and third corner. If the division,
e.g. between the third and fourth
contour top is larger than between the
first and second corner, then the initial -
given by the user- edge contour division
between the first and second contour
corner will be automatically increased.
Coons' meshing may be applied for flat
(2D) domains (contours are defined on
planes - see drawing above) regions as
well as for 3D surfaces (contours
defined in space - see drawing below).
The following parameters may be
defined for Coons method:
= the FE mesh type (Coons typology)
= mesh creation parameters (division1
and division2) explained earlier.
The program also provides the Base Mesh Points option used to determine panel points that
will be the basis for generating finite element mesh by means of the Coons method. The
option is available from:
the menu, by selecting the command: Analysis/Calculation Model/Base Mesh Points
the toolbar, by pressing the icon .
structures.com
64
Parameters of FE Mesh Generation
If the Modification button located in the Job preferences dialog box (Meshing Options
option see section 2.2.3 of the manual) is pressed or the Analysis/Calculation
Model/Meshing Options command in the menu is selected (panel must be selected as the
current structure type), the following dialog box will appear on screen.
NOTE: The Analysis / Calculation Model / Meshing Options command is available in
the menu for the following structure types: plate, shell and volumetric
structure.
New surface FE mesh parameters may be chosen in the above dialog box. One of listed
mesh generating methods may be chosen in the Available Meshing Methods field: Coons or
Delaunays method. Additionally, the degree of method implementation may be determined
for each case of mesh generation: never, rarely, often. The forcing ratio of the selected FE
mesh generation method may also be defined. If, for instance, the user selects Coons
method, defines implementation degree - as often, and forcing ratio - as forced, it would mean
that the mesh generation algorithm will force the mesh creation in the selected region
according to Coons method.
The finite element type used during element mesh generation may be chosen in the Finite
Elements field: 3-node triangles, 6-node triangles, 4-node rectangles, 8-node rectangles (see
figure below).
structures.com
65
Triangular 3-node and 6-node elements
Rectangular 4-node and 8-node elements
In the Robot program it is recommended to use 3- and 4-node surface elements. If 6- and 8-
node surface elements are used for mesh generation, the following options may not work
correctly:
linear releases
Boolean operations (cutting off)
adjusting meshes between panels as well as between panels and bars.
Also, the forcing ratio of the selected surface FE type may be defined. If, for example, 3-node
triangles and the forcing ratio as Any have been, it will mean that the mesh generation
algorithm will use any type of surface FE during mesh generation.
The automatic or the user-defined type of meshing may be selected in the Mesh Generation
field. For Coons method, one can define the following two parameters:
division 1 - the parameter defines the number of elements which were used on the first
contour edge (between the first and second peak). The edge of the contour opposite to
the mentioned contour side will be automatically divided so that the division corresponds
to the first edge contour.
division 2 - the parameter defines the number of elements which were used on the
second contour edge (between the second and third peak). The edge of the contour
opposite to the mentioned contour side will be automatically divided so that the division
corresponds to the second edge contour.
There is a possibility to specify the size of finite elements generated during mesh
generation. The Element size option enables it.
Moreover, this field contains the option that enables determining what type of volumetric
finite element mesh is to be generated: the slider allows selection in the range between a
coarse mesh and a fine mesh.
One of the following contour division types may be chosen in the Coons Method Parameters
field:
triangles in triangular contour
triangles and rectangles in triangular contour
triangles and rhombs in triangular contour
squares in rectangular contour
triangles in rectangular contour.
structures.com
66
Additionally, the forcing ratio of the selected contour type for Coons method of surface FE
mesh generation may be selected in this field.
The method of mesh generation may be selected in the Delaunays Method Parameters field:
Delaunays method only - selection of this option means that the mesh will be created
using Delaunays method alone.
Kangs method - selection of this option means that the FE mesh will be generated only in
the contour near the emitters according to the accepted parameters of Kangs method
(H0, Hmax and Q).
Delaunays and Kangs method - selection of this method means that mesh creation near
the emitters will occur according to Kangs method and outside of this contour - according
to Delaunays method.
If during mesh generation the Smoothing option is checked out, the program will also use an
algorithm for smoothing up the generated mesh of finite elements.
Options Used for FE Mesh Generation and Modification
Options are available after clicking the Option of FE mesh generation icon , located in the
upper toolbar. It results in opening the toolbar shown on the figure below
which contains the icons that allow the following:
- generation of calculation model, it means mesh generation for panels which have not
become frozen yet
- determining the points that will be the basis for mesh generation by means of the
Coons method
- opening the Meshing options dialog box for a selected panel
- freezing mesh for a given panel selection of this option means that during generation
of calculation model the mesh for that panel will not be modified
- defreezing mesh for a panel selection of this option means that the panel will be taken
into account during generation of a calculation model
- local mesh generation mesh will be generated only for selected panels (NOTE: usage
of this option causes mesh to become frozen)
- deletion of mesh for a selected panel
- definition of users emitters
- mesh consolidation - the window allows conversion of triangular elements to rectangular
ones for selected finite elements.
NOTE: Examples of finite element mesh generation for plate / shell structures are
presented in the appendices at the end of the users manual.
structures.com
67
3.4.2 Emitters, FE mesh refinement, consolidation and quality
Emitters are user-defined nodes in the vicinity of which the mesh of finite elements will be
refined. The option is important during calculations of plates/shells or volumetric structures,
when the user wants to obtain possibly precise calculation results in structure characteristic
points (supports, force application points, etc.). The option is available
by selecting the Analysis/Calculation model/Emitters command from the menu
by selecting the icon from the toolbar.
There are two methods of emitter definition. They depend on the manner of finite element
mesh increment:
constant increment - the method is used for plate or shell structures
variable increment - the method is used for volumetric structures.
Once the command is selected, the below-presented dialog box appears on screen (the
figure shows the dialog box after selecting the Variable increment option).
One of the following three options may be selected in this dialog box:
define a node with an emitter - the user defines a node
where to place an emitter with the user-defined coordinates
define an emitter in (an) already existing node(s) - definition
of (a) node(s) number(s) where (an) emitter(s) will be placed
delete an emitter - definition of (a) node(s) number(s) where
(an) emitter(s) will be deleted
The above presented options have the following parameters:
define a node with an emitter:
H0 - the initial length of mesh wave (the remaining parameters
of mesh generation around the emitter may be defined in
the Mesh generation options dialog box)
Coordinates - coordinates of the defined emitter node with the
defined H0 parameter
define an emitter in (an) already existing node(s):
H0 - the initial length of mesh wave (the remaining parameters
of mesh generation around the emitter may be defined in
the Mesh generation options dialog box)
node list - list of numbers of nodes where to locate emitters the
defined H0 parameter
delete an emitter:
node list - list of numbers of nodes where to delete an
emitter.
Once the Variable increment option is selected, the above dialog box provides access to all
the options that have been discussed for the constant mesh increment. Apart from that, the
following edit fields are available:
r1 - radius of the sphere in which the mesh will be characterized by the initial wave length
H0
structures.com
68
r2 - sphere radius within which the mesh density will be decreased (i.e. the mesh density
will be decreased within the zone between radiuses r1 and r2)
estimated element number in sphere r1 - the field is not accessible; the program defines
the number of elements after entering emitter coordinates and H0, r1 and r2 values.
During the finite element mesh generation, one may also take advantage of the following
options: mesh consolidation and mesh refinement.
Running the CONSOLIDATION option results in transforming the selected triangular elements
into rectangular ones (the number of elements diminishes). It is recommended for the user to
apply the CONSOLIDATION option after creating a mesh of finite elements by means of
Delaunay triangulation method. It results in the transformation of triangular elements into
rectangular ones that usually allow one to receive more precise calculation results. Before
running the CONSOLIDATION option, the user has to provide:
conversion coefficient - a parameter adopting values belonging to the range [1, +1]
list of elements for which consolidation will be performed.
Adopting the conversion coefficient value equal to +1 means that the created rectangles will
be created from triangular elements in all the possible places of the selected area (this,
however, may lead to the generation of rectangles with incorrect shapes and, consequently,
to incorrect conditioning of the relevant set of equations). Adopting the conversion coefficient
value equal to -1 means that only those triangular elements that will form rectangular
elements will be changed in the mesh of triangular finite elements.
The use of REFINEMENT option results in increasing the refinement of the mesh of finite
elements in the user-defined area. The created mesh of rectangular elements is divided into
smaller triangular or rectangular elements, depending on the adopted parameters. One can
refine a mesh of finite elements, once the following parameters are provided:
refinement type
list of elements for which the mesh will be refined.
There are three refinement types available in the program:
Simple - finite element edges will not be divided
Double - each edge of a finite element will be divided
into two parts
Triple - each edge of a finite element will be divided into
three parts
The dialog boxes provide also the FE Mesh Freeze option. If this option is activated, the FE
mesh will become frozen after performing the operation of refining the mesh for selected
panels. It means that the mesh will not be changed during the preparation of a structure for
calculations (generation of a calculation model for the structure).
If this option is not active, then, during the preparation of a structure for calculations, the FE
mesh may be changed for the panel on which the mesh has been refined, for the program will
assume the mesh generation parameters determined in the Meshing options dialog box.
Pressing the Mesh quality icon allows assessment of the quality of finite element mesh
created for selected panels.
structures.com
69
Each element has certain proportion coefficient determining its geometry quality, i.e. whether
the element is correctly conditioned or not. The coefficient adopts values from the range (0,1),
where 1 describes an element which is a square or an equilateral triangle. Lower values are
adopted by worse-conditioned elements, i.e. by those whose geometry diverge from square
or equilateral triangle. Two coefficients are globally checked for selected panels:
Q1 the weighted coefficient taking into account element significance with respect to its
area i.e. the greater the area of an element, the greater weight of its quality in the
coefficient
Q2 takes into account the number of both incorrect and correct elements without
considering their area weight.
Both elements adopt values from the range (0,1). If the coefficient approaches value 1, it
means that the quality of mesh is correct, and if it approaches value 0, it means that the
mesh is not satisfactory. Low value of Q1 coefficient indicates that mesh contains elements of
large area which are incorrectly conditioned. Whereas low value of Q2 coefficient informs that
mesh contains a great number of such (i.e. incorrectly conditioned) elements compared with
the total number of elements. Simultaneously, there is a possibility to find elements, whose
proportion coefficient is lower than certain value (the Precision field in the Mesh quality
dialog box.
NOTE: Examples of finite element mesh generation for plate / shell structures
including operations of mesh consolidation and refinement are presented in
the appendices at the end of the users manual.
3.5 Panel Thickness
One can assign thickness to planar structure elements (panels) by selecting:
the Panels option, and indicating the required thickness type in the Properties field
from the menu by selecting the: Geometry/Properties/Thickness or
from the toolbar Structure Definition by selection the icon
once the PROPERTIES layout is chosen - this layout is available for the following
structure types: plate and shell.
The New Thickness dialog box contains two tabs: Homogeneous and Orthotropic. The
Homogeneous tab can be used to define the following parameters:
thickness:
1. uniform of a value determined in the Th field
2. variable along a defined line (parameters are defined in appropriate fields for points
P1 and P2)
3. variable on a defined plane (parameters are defined in appropriate fields for points
P1, P2 and P3)
coefficient KZ value elastic support coefficient
material.
Additionally, uplift of a plate / shell from the ground for each direction may be defined. The
option is available only when the ground elasticity coefficient is defined. There are three
possibilities:
structures.com
70
None - there will be no uplift
+ - the axis will take place in the sense consistent with the axis sense (e.g. UX+)
- - the axis will take place in the sense opposite to the axis sense (e.g. UZ-).
The options located on the Orthotropic tab are used for taking structural plate/shell orthotropy
into consideration. Structural orthotropy refers to the lack of uniformity of plate stiffness in
different directions, due to, for example plate ribbing. Orthotropy takes account of differences
in stiffness in the mutually perpendicular directions, and it does not take account of lack of
material uniformity. All the geometrical heterogeneities are included only in stiffness matrices
of elements. A plate with the thickness defined in this way should be treated as a structure
with equivalent thickness with different stiffness in perpendicular directions.
NOTE: Local change of rib stiffness is not taken into account; the exact plate
geometry is not visualized and it is not taken into account during calculations
of reinforcement.
The dialog box for definition of orthotropic thickness type contains the following options:
Direction button pressing this button results in opening the Orthotropy direction
dialog box, in which the main orthotropy direction can be defined
drop-down list with the available types of predefined plate geometry (ribs, box floor,
grillage, material orthotropy); it is also possible to create a user-defined orthotropy matrix;
structures.com
71
after selecting a plate geometry type, the relevant edit fields are available, in which the
plate dimensions can be defined
the Display or Define button pressing this button results in opening the Stiffness
matrices dialog box
the Equivalent thickness option checking this option provides access to the edit field
which allow defining thickness values: Th, Th1, Th2. The equivalent thickness Th is used
to calculate a plate self-weight. The thickness values Th1 and Th2 are used to define the
equivalent thicknesses used for calculations performed in case of applying thermal
gradient loads. These thickness values are calculated automatically on the basis of the
plate geometrical parameters.
NOTE: Reinforcement calculations for this type of plates will not provide correct
results An algorithm of plate reinforcement should be implemented, which
would take account of a T-section or an H-section. Therefore, the
reinforcement calculations for this type of plates will be carried out as if they
were homogeneous plate with invariable cross-section.
Pressing the Elastic foundation coeff. button opens the Building soils - calculations of K
coefficient dialog box which serves as a calculator for computing a value of the elastic
foundation coefficient K for a layered soil.
As in the dialog box for bar section definition, the process of thickness definition for selected
panels has been divided into two stages:
definition of the active thickness type for the planar structure element (panel)
assigning thickness to panels.
To remove a thickness type assigned to an element defined in the structure, a zero thickness
should be used (DELETE icon), which is always present on the active list in the FE
Thickness dialog box. The zero thickness cannot be modified; it is assigned in the same way
as the procedure of assigning thickness to panels generated in the structure.
Once a thickness is assigned, its symbol is drawn in on the graphic viewer.
3.6 Solids (Volumetric Structures)
The option is used for generation of volumetric objects (solids). The option is available from:
the menu by choosing the command: Geometry/Solids or
the Structure definition toolbar by selecting the icon.
Once this option is selected, the dialog box shown below appears on the screen.
structures.com
72
Definition of a solid consists in determining:
solid number
solid edges (contours). It can be performed in two ways:
by indicating surface objects which determine solid volume
by defining a list of volumetric finite elements
by defining a list of surface elements that determine a solid contour; if the Delete
surface elements option is switched on, then surface elements defining the solid
contour will be deleted from a volumetric structure
solid properties.
NOTE: The lower part of the dialog box contains the () button (to the left of the
Properties field). Its pressing results in opening the Definition of solid
properties dialog box in which solid physical properties can be defined.
Defined types of solid properties are added to the list of active properties.
Solid volumes are created by specifying their edges (defining volume contours). Solid
definition can be performed using one of the two ways described below:
by defining a list of planar objects which create solid external area
by specifying the numbers of generated volumetric finite elements. It is possible to take
advantage of this option once a mesh of finite elements is generated.
NOTE: If a solid-like object is created by means of the option Creation with list of
volumetric elements (FE), then the selected finite elements cannot contain
inclusions (i.e. a solid cannot include openings). This problem can be solved
by creating two solids in such a way, so that neither of them includes the
entire opening by itself.
NOTE: If a bar is defined in a structure, whose node adjoins a node of a volumetric
element, then, the bar functions in the structure as if there was a pinned
connection in the node that connects the bar with the volumetric element.
structures.com
73
The Solid Properties option is used for defining properties of volumetric elements and
assigning them to structure volumetric elements (solids). The option is available from:
the menu by selecting the command Geometry/Properties/Solid Properties or
the Structure definition toolbar by selecting the icon
after selecting the PROPERTIES layout - this layout is available for the structure type:
solid (volumetric structure).
After pressing the Definition of solid properties icon in the Solid Properties dialog box, the
screen displays the dialog box presented in the figure below.
structures.com
74
To define a new type of solid properties, one should:
enter a label of a property type that is being defined
select a color for a property type that is being defined
enter a material model (in the present version elastic material is available)
define parameters characterizing a material model.
Pressing the Import from database button results in opening an additional dialog box, in
which a material available in material databases can be selected. Once such a material is
selected, the parameters are entered into the appropriate fields in the Definition of solid
properties dialog box.
After defining the above parameters and pressing the Add button a defined type of properties
will be added to the list of active types of solid properties.
3.6.1 Description of Volumetric Finite Elements
Volumetric structures in the Robot program are modeled with the use of isoparametric
functions.
The following element types are available: brick B8, wedge W6 and tetrahedron T4.
Shape functions and node numbering referred to 3D reference elements are presented in the
table below.
structures.com
75
Reference Element Shape Functions
Nen i
N N
i i
1,
elements 3D for ) , (

8 Nen
) 1 ( 1 )( 1 (
8
1

i i i
N
6 Nen
1 3)) mod )( 1 (( ) (
) ( ) 1 (
2
1
6
) (

i i k where
N N
T
i k i

6
3
6
2
6
1
1
T
T
T
N
N
N
4 Nen

4
3
2
1
1
N
N
N
N
Description of element geometry, evaluation of strains, stresses, stiffness and mass matrix as
well as element force vector for all enumerated element types is performed in the following
manner:
Element geometry
Element geometry is based on isoparametric mapping from reference element onto an
arbitrary element
structures.com
76
Nen a
a a
N
, 1
) ( ) ( x x
Displacement fields within the element
T
w v u ] , , [ u
Nen a
a a
N
, 1
) ( ) ( u u
Strain evaluation
T
yz xz xy zz yy xx
] , , , , , [

Nen a
a a
, 1
) ( ) ( ) ( u B u B ,
where B matrices are defined as:
], [
a
B B
x
N
z
N
x
N
z
N
x
N
y
N
z
N
y
N
x
N
a a
a a
a a
a
a
a
a
0
0
0
0 0
0 0
0 0
B
, Nen a , 1
structures.com
77
Shape function derivatives present in B matrix elements are defined as:
, ) (
1
z z z
y y y
x x x
N N
a T a
J J
x
Stresses (elastic case)
T
yz xz xy zz yy xx
] , , , , , [ ,
) (
o
Bu D ,
where e
o
are imposed strains (thermal dilatation, shrinkage), while D is a constitutive
matrix; constitutive matrix D (linear - elastic isotropic material) is presented below
) 1 ( 2
2 1
0
) 1 ( 2
2 1
.
0 0
) 1 ( 2
2 1
0 0 0 1
0 0 0
1
1
0 0 0
1 1
1
) 2 1 )( 1 (
) 1 (
sym
v
v
v
v
v
v
E
D
Sign Convention
structures.com
78
3.7 Operations on 2D Objects (Extrude, Revolve,
Combination) and on 3D objects (Fillet, Chamfer)
The system of Robot Millennium is equipped with options used to create simple solid-like
elements by extruding or revolving predefined two-dimensional objects. The Extrude option
is accessible:
from the menu command: Geometry/Objects/Extrude or
from the Objects: operations/modifications dialog box by pressing the Extrude button
(first, a defined object must be selected).
NOTE: The option is accessible only for the following structure types: plate, shell,
volumetric structure.
There are two ways of defining the operation of extruding a two-dimensional or three-
dimensional (solid) object:
the first consists in defining with triple coordinates the vector that determines the direction
and length of extrusion.
the second method of defining extrusion consists in determining the axis of the global
coordinate system along which the object will be extruded and the length of object
extrusion.
In order to carry out the operation of extruding an object through vector definition, one should:
1. define a two-dimensional object, e.g. a
rectangle (see the figure beside)
2. select the object
3. open the Extrude dialog box
4. define the parameters of the operation of
extruding the object; let us adopt the
following parameters for the defined
rectangle: the vector of extrusion - (0,10,0),
division number - 10, top and base options -
inactive, scale - 0.5, new object option -
inactive.
5. press the Apply button to carry out the
operation of extruding the predefined
rectangle. The operation yields the object
presented in the figure beside.
The Revolve option functions in a similar way. It is used to create simple surface or solid-
like elements by revolving (rotating in space) predefined two-dimensional objects with
respect to a selected axis. The option is accessible:
from the menu command: Geometry/Objects/Revolve or
from the Objects: operations/modifications dialog box by pressing the Revolve button
(first, a defined object must be selected).
NOTE: The option is accessible only for the following structure types: plate, shell and
volumetric structures.
structures.com
79
In order to carry out the operation of revolving a predefined two-dimensional object, the user
should define several parameters, namely: the parameters of the rotation axis (axis beginning
and axis end), the value of the rotation angle, number of divisions, top, base, scale, and new
object.
An example of defining the operation of revolving a square whose side equals 6 is presented
below. In order to carry out the operation of revolving a square, one should:
1. define a two-dimensional object, e.g. a
square (see figure A); let us assume
that the coordinates of its vertices are:
(0,0,-6), (0,0,0), (6,0,0), (6,0,-6)
2. select the object
3. open the Revolve dialog box
4. define the parameters of rotation; let us
assume the following parameters for
the square defined above: axis
beginning: (12,0,0), axis end: (12,0,-6),
rotation angle: -90, number of divisions:
10, top and base options inactive, the
scale: 1.0, new object option inactive
fig.A
5. press the Apply button to carry out the
operation of extruding the predefined
square. The operation yields the object
presented in figure B.
fig.B
When objects are generated by means of the above-described options, the program creates
also their components (side, base, edge). They are appropriately marked. The syntax of lists
for components of objects created in this manner is presented in chapter 2.2.6.
Object defined by extruding or revolving predefined two-
dimensional objects may be edited and modified by means
of the Operations and modifications on objects option.
The option is accessible from the menu, by selecting the
Edit/Substructure Modification/Object Modification.
The respective dialog box shown below is divided into four
parts.
the Object edit field, where the user determines the
number or identifier of the created/selected object
the dialog box part called Geometry/Subobjects
the dialog box part called Object modification list
the dialog box part called List of operations on object
modification.
When the Geometry/Subobjects button is pressed, the
Polyline-contour dialog box opens. In this dialog box, one
can define an object that will undergo modifications and/or
operations on determined modifications.
structures.com
80
The part of the dialog box called Object modification list contains options that allow one to
select modifications to be applied to an object and to define their parameters. There are three
types of modifications in the program: Extrude, Revolve, Along Polyline. Once a button
from the Add Object Modification group is pressed, the selected modification is added to the
list specifying the successive stages of modification applied to a two-dimensional object. The
operations of modifying a predefined object may also be deleted. Once an object modification
is selected from the list and the Delete button (beside the Object modification list) is pressed,
the selected modification is deleted from the list. Once an object modification is selected and
the Object Modification Parameters button is pressed, the Objects:
operations/modifications dialog box gets maximized to present the parameters of the
selected object modification (Extrude, Revolve or Along polyline). The parameters of the
above mentioned object modifications are discussed in the descriptions of Extrude and
Revolve options. Once the Apply button is pressed (object modification having been
defined), the selected modification is applied to the object in accordance with the parameters
adopted for the defined modification.
The part of the dialog box called List of operations on object modification contains options that
allow the user to select operations to be applied to the object modifications, defined in the
upper part of the dialog box. Moreover, the options allow the user to determine the
parameters of operations on modifications.
There are four types of object modifications available in the program: Translation, Rotation,
Scaling and Deformation. If one of respective buttons (in the Add Operation group) is
pressed, the selected operation on the modification (selected in the upper part of the dialog
box) is added to the list specifying the successive operations on the modification of the
predefined two-dimensional object. The operations on object modifications may also be
deleted. Once an operation on object modification is selected from the list and the Delete
button (beside the List of operations on object modification) is pressed, the selected
modification is deleted from the list. Once an operation on object modification is selected and
the Operation Parameters button is pressed, the Objects: operations/modifications dialog
box gets maximized to present the parameters of the selected operation on object
modification (Translation, Rotation, Scaling and Deformation).
Parameters of translation and rotation are the same as those of extrusion and revolution. The
operation of scaling allows one to determine the object scale coefficient(s) for particular
system axes. It the value of scaling is larger than 1.0, the object will be enlarged in the
selected direction. If the value of scaling is smaller than 1.0, the object will be reduced in the
selected direction. The operation of deformation allows one to translate characteristic points
of an object in the course of modification (e.g. in the case of a rectangular object, its vertices
will be translated). In other words, if the original object to be extruded was a square, one can
obtain, for instance, a trapezoid by adequate selection of parameters defining deformation to
be carried out on extrusion results.
Once the Apply button is pressed (object modification and operation on object modification
having been defined), the defined operations and object modifications will be carried out
according to the adopted parameters.
The Object combinations option allows one to create complex objects on the basis of
formerly-defined two-dimensional objects. It is available from the menu by selecting the
Geometry/Objects/Object Combinations command or pressing icon. The option is
available only for plate/shell structures and volumetric structures. Once the option is selected,
the dialog box shown below appears on screen.
structures.com
81
NOTE: The appearance of the above dialog box and available logical operations
depend on that whether the ACIS Kernel technology (see Preferences /
Advanced dialog box) or internal mechanisms of the Robot program will be
used to generate Boolean objects.
NOTE: Application of the ACIS technology is not recommended for the following
structure types: plate, plane stress structure, plane deformation structure and
axisymmetric structures.
The options included in the above-presented dialog box allows one to combine formerly-
defined two-dimensional objects (pipe surface, prism surface, arc, etc.) and volumetric objects
into complex objects. For two-dimensional objects, operations will be carried out on surfaces
and not on volumetric objects (three-dimensional objects).
The following operations on pre-defined objects are available in the program:
logical operations (Boolean algebra) with two operands - union, complement (separately
for surface structures and for volumetric structures), common part, exclusive alternative
logical operations (Boolean algebra) with one operand - union, common part, exclusive
alternative
cutting off.
The logical operations listed are available when the internal mechanisms of the Robot
program are applied. The Boolean operations using the ACIS kernel include the following
types of logical operations: intersection, union, partition, subtraction, inclusion and partition
(A-B) (B-A) (A*B) (the last three types are available only for double-operand operations).
For the Boolean operations to be performed with the use of the ACIS Kernel modules, the
ACIS geometric kernel option should be activated in the Preferences dialog box (NOTE: the
ACIS geometric kernel option is accessible in the dialog box, if the ACIS Kernel option is
available in the Robot program protection).
While applying the ACIS technology, the user should take note of the following limitations:
a Boolean operation already performed cannot be modified, in other words, once any
operation is performed (e.g. cutting out an opening in a solid), a user is unable to
change/correct the operation parameters; it can be solved by canceling the operation by
means of the UNDO option or by conducting additional Boolean operations to restore the
initial structure state
volumetric elements cannot cooperate with shell elements; it means that there is no
possibility of combining these elements (a shell cannot interpenetrate a solid and be
adjacent to a solid).
structures.com
82
If the ACIS geometric kernel option is switched on, it affects operation of the following options
available in the Robot program: Fillet 3D and Chamfer. The options are available in the
Geometry/Objects menu or by pressing the icons: and (respectively). The options
are used to create arcs (fillets) between the edges of faces forming a 3D object or chamfers
between two edges of faces forming a 3D object.
The current version of the Robot program provides access to the following edit operations on
objects with the use of ACIS technology:
Fillet 3D
Chamfer
Object combinations
Face skinning
Face offsetting
Object bending
Object stretching
Object twisting
Object warping.
Moreover, the following options have been complemented with operations on object faces
applying ACIS technology (their operation may start if only object faces are selected):
extrude
extrude along polyline
revolve
delete
all the available edit operations (inclusive of complex edit).
NOTE: In case of operations of combining planar objects (such as a plate or shell)
one should not apply the Coons meshing method; for this type of objects the
Delaunays method should be used.
In order to perform logical operations (with one or with two operands) one should:
define the number of the complex object that will be created out of the selected objects
choose operation with one or with operands
indicate the type of the logical operation (union, complement, common part, exclusive
alternative) see figures below
define the list of objects that will taken into consideration during the process of complex
object creation.
press the Apply button.
structures.com
83
The cutting off option allows one to determine the part of the surfaces of the selected objects
that is located on one side of the defined plane. The operation yields an object cut-off from
the selected objects by means of the defined plane; definition of direction allows one to
determine which part of the selected objects will be cut-off. The example of this operation is
shown on the figure below. Two objects A and B will be subjected to operation of cutting off.
The cutting line (plane) has been defined and the direction (an arbitrary point on one of the
line sides) has been selected. The object marked with gray color poses the result of the
operation.
structures.com
84
3.8 Supports
Assigning supports to structure nodes
is made possible from:
the menu by selecting:
Geometry/Supports or
the "Structure Definition toolbar -
by selecting the icon
the SUPPORTS layout.
Once the option is selected, a dialog
box shown beside will be displayed on
the screen.
The following support types can be defined:
support in a structure node
support defined along the line (segment), e.g. along the plate edge or volumetric structure
edge (it is available for the following structure types: plate, shell, volumetric structure)
support defined on a surface (it is available for volumetric structures).
As in the dialog box for the definition of other structural attributes, the definition of supports in
the structure is divided into two steps:
Support type definition
Click on the Support Definition icon. If the list of active supports is empty or a new
support is to be added to the active list, two situations are possible:
if none of the support types is selected, clicking on the New Support Type icon opens a
dialog box for a new support type definition; the first tab will be called up with the fields
defined previously (except for the Label field) or default parameters will be set; the
following support types are available in the program:
= fixed support (with the possibility of modeling the uplift effect)
= elastic support
= support with damping
= non-linear support.
Definition of a new support type consists in selecting the blocked degrees of freedom in
the node (UX, UY, UZ, RX, RY, RZ) with a possibility of selecting the direction of support
up-lifting, if necessary, determining values of the elastic base coefficients for an
appropriate direction (in case of elastic support) and defining the value of damping
coefficients (in case of support with damping) or a support with a defined non-linear
model. Supports may be defined in the global or local coordinate system.
The program provides the option that allows definition of non-linear behavior of supports,
releases and compatible nodes. It may be used for all structure types. The option may be
activated on the Non-linear tabs in the dialog boxes for definition of supports, releases
and compatible nodes. It is possible to define the non-linear force (moment) -
displacement (rotation) relationship for selected directions (degrees of freedom). These
relationships may be determined for individual directions independently (there are no
interactions). The following types of non-linear models are available in the current version
of the program: linear, bilinear, parabolic, parabolic according to EC2, perfectly plastic,
plastic with hardening, gap/hook as well as model defined by determining the function.
structures.com
85
Additionally, one may define support uplift for each direction. There are three possibilities:
none (there will be no uplift), + - uplift direction will agree with axis direction (e.g. UX+),
- - uplift direction will be opposite to axis direction (e.g. UZ-). If uplifting is defined for a
support (e.g. uplifting will be possible according to the orientation of Z axis, i.e. UZ+),
there also exists a possibility of determining the elastic base coefficient KZ for each
support. One should, however, remember that the elastic base coefficient will be
determined only for the orientation opposite to the one defined for up-lifting (i.e. for UZ-)
see the schematic figures below.
The New Support dialog box contains the Advanced button. Pressing the button results
in opening a dialog box with options used for defining specific support parameters
which are used during the process of definition and analysis of concrete structures.
These options are used in the module calculating R/C plate reinforcement area. A
support may be defined as:
- support defined in a node (this is the default type of support)
- support defined by determining column dimensions; there are two possibilities
available:
rectangular column - it is required to define width and height of the transversal
column cross-section (dimensions b and h)
circular column - it is required to define the diameter d of the transversal column
cross-section
- support defined by determining wall dimensions; it is required to determine wall
width b. Additionally, beside the Wall option there is also a drop-down list
containing available wall types (brick, concrete). The list is available after selecting
the Wall option. A defined wall type is taken into account only in design of plate
and shell reinforcement by means of reinforcing bars or wire fabrics; it does not
affect model calculation parameters.
There is the Elastic coefficients button provided on the Elastic tab; its pressing
opens the Building soils calculations of K coefficient dialog box which serves as
a calculator for computing a value of the elastic foundation coefficient K for a layered
soil.
if any of the supports is selected, clicking on the New Support Type icon opens a
dialog box for a new support type definition; a tab appropriate to the selected support
type will be called up. All edit fields, except for the Label field will be filled in according
to the support type selected.
It is also possible to open the dialog box for a new support definition by double
clicking on the element from the list of active supports. The Support Definition dialog
box will be open on the tab corresponding to the selected support type and all fields
will be filled in. After modification of appropriate parameters the new support type is
added (updated) to the list of active supports by clicking on the Add button or
pressing the <ENTER> key. If the label cannot be changed, the appropriate note will
be displayed on the screen. This option in the dialog box allows for an easy
modification of the support.
Assigning supports to structure nodes operation resembling that of assigning
sections to bars
To delete the assigned support, the Delete Support (DELETE icon) should be used. It is
always available from the list of active supports. Such a support type cannot be modified; it is
assigned in the same way as the support definition in the structure.
Once the support is assigned, an appropriate symbol will be displayed on the graphic viewer.
structures.com
86
3.8.1 Elastic Support Definition (Layered Soils)
The Robot program offers a possibility to calculate a value of the equivalent elastic coefficient
for a layered soil. The soil calculator enabling calculation of the equivalent coefficient for a
layered soil is installed on the disk during installation of the Robot program. The elastic
coefficient may be applied while defining the following:
elastic support
bar elastic ground
panel elastic ground.
The calculator allows computation of the elastic coefficient directly on the basis of a defined
profile of a layered soil. The calculator provides the possibility to:
define a soil profile taking advantage of the available soil database including soil
characteristics
save and read a complete soil profile defined by the user
calculate the elastic coefficient for a defined profile
forward a calculated value to the dialog boxes for support or ground definition
The option operates as an independent tool which enables calculation of the ground reaction
coefficient for a defined foundation and soil profile. A saved soil profile may be used
in the RC calculator and the continuous footing calculator.
The following assumptions have been adopted in the calculator design:
a soil subjected to action of forces works in the elastic state; it corresponds to ULS
in the allowable codes; it allows adopting principles of the linear theory of elasticity
a soil is an infinite elastic semispace within which material parameters are changed only
in planes parallel to the surface
calculations concern a rectangular foundation of infinite stiffness
a soil model is a discreet - layered model with a constant layer thickness.
The calculator may be activated in the following ways:
by choosing the Tools / Building Soils - Calculator command from the menu
by pressing the Building Soils icon located on the Tools toolbar
by pressing the calculator icon provided on the computer desktop
by selecting the Building Soils - calculations of K coefficient option included in the group
created during installation of the Robot program
after pressing the Elastic coefficient button provided in several dialog boxes of the
Robot program (New support definition, New elastic ground, New thickness
definition dialog boxes).
Once the calculator is activated, the dialog box (module) used to calculate the equivalent
coefficient, shown in the drawing below, is displayed on the screen.
structures.com
87
In the calculator the K coefficient is calculated based on the value of average stresses under
the foundation for an area unit. The equivalent KZ coefficient for a foundation with specified
dimensions is also computed.
To calculate the equivalent K coefficient for a layered soil, the user should follow the steps
listed below:
in the table located in the top part of the dialog box define parameters of the successive
soil layers (the soils available are the soils from the soil database selected in the Job
preferences dialog box of the Robot program) - consecutive soil layers will be presented
schematically in the bottom left part of the dialog box
Once the soil type is selected from the drop-down list provided in the Name column, the
level of a given soil layer should be determined; it is defined by two quantities: Level or
Thickness; the remaining parameters are read from the soil database
choose a foundation type:
spread footing with dimensions A x B; the KZ coefficient unit is (force/length);
a computed value KZ = K * A * B may be applied while defining the elastic coefficient in
the dialog box used for support definition
structures.com
88
continuous footing with the following dimensions: A (continuous footing length)
and B (continuous footing width); the KZ coefficient unit is (force/length^2); a computed
value KZ = K * B may be applied while defining the elastic coefficient in the dialog box
used for definition of an elastic ground type
foundation plate with dimensions A x B; the KZ coefficient unit is
(force/length^3); a computed value KZ = K may be applied while defining the elastic
coefficient in the dialog box used for definition of a panel thickness type
determine dimensions of a selected foundation type
once dimensions are defined, press Tab key or OK button - in the K = field a value
of the equivalent coefficient for a layered soil will be specified.
Pressing the OK button causes a computed value of the KZ coefficient to be forwarded to the
edit field located in the New support definition, New elastic ground or New thickness
definition dialog box (provided that the relevant dialog box is opened and the edit field for
defining elastic coefficient is accessible).
NOTE: The K coefficient value may be forwarded only to the KY, KZ edit fields in the
dialog boxes mentioned. It should be remembered that elastic coefficient
values should be passed to the appropriate dialog box, depending on the
foundation type selected.
A defined soil profile may be saved on the disk; pressing the Save as button allows saving a
profile in a file with *.mdb (database) extension. The File name field shows a name of the
current soil profile with a full access path. Pressing the Open button enables reading in a file
with defined parameters of a soil profile
3.9 Loads
It is recommended to use the Robot MillenniumLOADS layout to define loads applied to the
structure. The screen will be divided into three parts: a graphic viewer which allows structure
definition, a dialog box with different Load Types (shown below) and a table allowing the user
to define loads for the generated load cases.
In the dialog box, load cases for the created structure are to be
defined by the user. For each case, its nature, number and
name (the program suggests a default name) should be given.
Once the New button is pressed, the case will be defined and
added to the defined cases list found in the lower part of the
Load Types dialog box. Load case parameters may also be
changed. The Modify button is used to do this. The following
steps should be followed to modify load case parameters:
select a load case to be modified from the list of defined
load cases
change the number, name or nature of the load case
press the Modify button.
In Robot Millennium, it is possible to define the following load
cases: self-weight, dead, live, wind, snow, temperature,
accidental and seismic.
structures.com
89
Once the load cases have
been defined, the loads
acting in the defined load
cases. This may be done
in two ways:
move to the table located in the lower part of the LOADS layout which is used to define
structural loads acting in the given load cases.
To define the load acting in one of the defined load cases, the following should be done:
left click in the Case field and select the load case defined in Load Types dialog box from
the list.
for the selected load case, define the type of the load applied. A list of load types
available in the system will appear after clicking in a field in the Load Type column. The
following load types are available in Robot Millennium:
for bar structures: dead loads, nodal forces, uniform loads, trapezoidal loads, bar forces,
thermal loads, imposed displacements, dilatation and distributed moments
for plate and shell structures: dead loads, uniform planar loads, planar loads defined with
3 points, linear loads defined with 2 points, pressure loads, uniform planar loads on
contour, planar loads defined with 3 points on contour, and thermal loads.
to select structure bars/nodes to which a load will be applied, click in the field located in
the LIST column and select (graphically in the load editor or using the Select option from
the graphic editor context menu) the appropriate bar/nodes of the structure.
Once the load type is selected, the load table will change to correspond to the selected
load type (i.e. the table will contain only the columns necessary for the definition of the
selected load type). For example, for a uniform load on the 2D frame, the table consists of
columns:
- allowing for definition of loads in the X and Z directions (px and pz values)
- indicating whether the load should be applied in the local or global coordinate system
and whether or not the load should be projected.
open the Load dialog box which allows for the
definition of loads for the created load cases.
This option is available from the menu by
choosing the Loads/Load Definition command.
NOTE: The menu command Load Definition is not
active until at least one load case is
defined.
Once the option is selected, a dialog box shown
beside will be displayed on the screen.
The dialog box consists of four tabs: Node, Bar,
Surface and Self-weight and mass.
Once the Node tab is selected, the dialog box contains the following icons:
structures.com
90
- opens a dialog box for definition of nodal forces,
- opens a dialog box for definition of imposed displacements for support nodes,
- opens a dialog box that allows one to define the values of forces in a point of the designed
structure (the option is available for plate and shell structures)
- deletes a selected nodal load type. To remove a load from the structure, select the load
type to be deleted and indicate the nodes for which the load will be deleted.
Once the Bar tab is selected, the dialog box contains the following icons:
- opens a dialog box for definition of uniform loads,
- opens a dialog box for definition of trapezoidal loads,
- opens a dialog box for definition of concentrated loads along the elements length,
- opens a dialog box for definition of dilatation values,
- opens a dialog box for definition of thermal load,
- opens a dialog box for definition of surface load on bars,
type to be deleted and indicate the bars for which the load will be deleted.
The program provides the possibility of taking account of the load which is not applied on a
bar axis. If such geometrical offsets of a bar axis exist, the point of applied force is defined in
relation to an applied local coordinate system of an element. Loads applied to bars acting on
some eccentricity allow for definition of forces applied to a bar in a distance to a longitudinal
bar axis. Eccentricity (force distance to a longitudinal bar axis) is defined in a local coordinate
system of a bar.
Loads on eccentricity can be defined for the following types of loads: concentrated force and
moment acting in the point on the bar length (bar force), uniform load on a bar. The option is
accessible by pressing Load on eccentricity button that can be found in dialog boxes used
for defining the mentioned load types. Loads defined on eccentricity are reduced to a bar axis
(see the figure below presenting concentrated force load): concentrated force components (in
relation to local element coordinate system) Fx, Fy, Fz should be found. Next, additional
moments in a local system can be calculated: Mx = Fz*y Fy*z, My = Fx*z and Mz = - Fx*y.
structures.com
91
Once the Surface tab is selected, the dialog box contains the following icons:
- opens a dialog box for definition of uniform planar loads,
- opens a dialog box for definition of planar loads defined with 3 points,
- opens a dialog box for definition of linear loads defined with 2 points,
- opens a dialog box for definition of pressure loads,
- opens a dialog box for definition of uniform planar loads on contour,
- opens a dialog box for definition of planar loads defined with 3 points on contour,
- opens a dialog box for definition of thermal loads defined with 3 points,
- opens a dialog box for definition of linear load on edges
NOTE: In case of volumetric structures (solids), the Solid loads option appears in the
lower part of the dialog box. Switching it on means that the defined loads will
be determined for volumetric structures.
The program provides the possibility of defining some load types of load applied to part of a
panel. The Geometrical limitations option is used for it. The option is available by pressing the
Geometrical limitations button located in the dialog boxes used for defining: uniform surface
loads, surface load defined by three points, load with uniform and hydrostatic pressure and
structures.com
92
thermal load on finite elements. The option is also available in the Loads table (by pressing
the Limitations button) if the one of the above-mentioned load types has been selected
After defining the plane, one may determine a semi space or a panel (object) layer to which a
defined load will be applied. The figure below shows the layer of d thickness to which a load
will be applied.
Once the Self-weight and mass tab is selected, the dialog box contains the following icons:
- applies self-weight for the whole structure,
- opens the dialog box for definition of self-weight on selected elements,
- opens the dialog box in which body forces may be defined
- opens the dialog box in which centrifugal and angular acceleration forces may be defined
- opens the dialog box in which the values of nodal masses may be defined
- opens the dialog box in which the values of bar masses may be defined
New types of loads enabling definition of body forces and centrifugal and angular
acceleration forces allow analysis and design of marine industry structures (structures
associated with the industry of crude oil extraction from the sea bottom, e.g. components
of equipment of drilling platforms). The loads mentioned are the loads generating structure
inertia forces due to ascribed velocity or acceleration. This type of loads is applied for
marine structures where transport loads may be of great importance (for example a
structure lifted by crane or mounted on a ship). Loads of this type do not solve all the
problems of marine structure modelization, but make them to a great degree easier.
The body force load is a static load including added masses. This load generates forces
resulting from mass of elements and masses added in nodes or elements for a given
acceleration of the value a. The value of generated force equals F = m * a. The centrifugal
and angular acceleration forces are a static load which enables considering added
masses. This load generates centrifugal and angular acceleration forces induced by the
element mass and masses added in nodes or elements; the following forces are
generated:
centrifugal force for a given angular velocity V: Fr = m * v^2 * r
force tangent to the direction of motion at the point of a given angular acceleration a: Ft =
m * a * r, where r is a distance of a given mass node to the axis of the coordinate system
positioned in the central point of rotation (see the drawing below).
structures.com
93
The Robot program enables definition of masses (weights) added to nodes and bars as well
as conversion of loads to masses. Added masses are mainly considered in dynamic analyses
(dynamic, harmonic, spectral, seismic and time history), but they need to be taken into
account in static calculations during generation of forces for body loads and self-weight.
Therefore, in the current program version:
table of added masses is accessible if any load case has been defined
options for added masses (nodal and bar masses) are always available in the dialog box
in the mass table and in the Analysis Type dialog box on the Load to Mass Conversion
tab a full list of simple load cases is provided.
An added mass influences a given load case, when one of the loads listed below, generating
forces that result from added masses, is defined:
self-weight
body forces
centrifugal and angular acceleration forces.
It is often the case that a self-weight load is assigned to structure bars/panels before defining
all bars/panels included in the structure. As a result, a self-weight load is not assigned to the
bars/panels which were defined after applying the self-weight load; thus, a structure that is not
fully loaded with the self-weight is adopted in calculations. A similar problem may arise as
well, when an edit operation (translation, rotation, etc.) is performed with the activated Drag
option; the bars that have been generated due to that operation are not automatically loaded
with the self-weight load either.
In order to facilitate considering the self-weight for the whole structure, the following
attributes: Whole structure and Part of structure have been added to the load record
containing a self-weight load case. If the Whole structure option is set in a table, then during
generation of data for calculations a self-weight load will be automatically applied to all
structure bars/panels.
The Whole structure attribute may be defined in two ways:
in the Self-weight dialog box: if the user presses the icon used for assigning a self-weight
load to the whole structure (all bars/panels), it means that the self-weight load will be
automatically assigned to the whole structure
in the load table: by selecting the Whole structure option in a line where a self-weight
load case is defined (it is the default value of the attribute).
structures.com
94
The Robot program offers a possibility to generate loads resulting from prestress in concrete
elements (prestressed concrete). Losses in prestressed elements may be calculated
according to the following codes:
- American code ACI 318-99
- EuroCode 2 (ENV 1992-1 : 1999)
- Polish code PN-B-03264:1999
- French code BAEL 91.
After choosing one of the codes listed (the Job Preferences / Codes option), selecting a
concrete structure element and then the option Analysis / Analysis of Pre-stressed Elements
from the menu, an appropriate sheet of the ESOP system which enables calculation and
generation of the load due to prestress, is executed. Data from the structure element chosen
is passed to the sheet (element length as well as section dimensions); once calculations in
the sheet of the ESOP program are performed, the structure modification takes place (adding
load cases).
NOTE: For the connection between the programs Robot and ESOP to be
operational, both programs have to be installed on disk. If the ESOP system
is not accessible, then while trying to run the option Analysis / Analysis of
Pre-stressed Elements, a message with information that the ESOP system
has to be installed, appears on screen.
At present, three sheets concerned with prestressed (post-tensioned) structures are available.
These sheets include calculation and generation of loads due to prestress considering
immediate losses caused by:
- friction between the tendon and its duct
- anchorage slip
- elastic deformation of the concrete.
Details concerning operation of the sheets mentioned can be found in descriptions of these
sheets (Help) provided in the ESOP program.
The upper part of the dialog box contains information on the selected load case (name,
number) for which the load will be defined and the type of load to be assigned to
nodes/bars/panels in the structure.
Once the load type is set, the load can be assigned to structure nodes/bars in one of the
following ways:
1. Enter nodes/bars/panels numbers to the Apply To field and press the Apply button;
2. Add the defined load to consecutive nodes/bars/panels in the structure (the cursor will
change its shape to the load symbol);
3. Select the nodes/bars/panels graphically and press the Apply button.
If loads are defined in the LOADS layout of Robot program, then in the right bottom corner of
the graphical screen on which a structure is displayed a legend of defined load types already
defined for a structure is presented. The legend contains (the size of symbol description
depends on the selected font size): load symbol and units used during load definition. The
table below presents symbols which are used to designate individual load types.
SYMBOL LOAD TYPE
self-weight (in this case it is weight in the direction of Z axis the sense is opposite
to the axis sense)
uniform load
concentrated force
moment
structures.com
95
imposed displacement, dilatation
thermal load
surface load
moving load
masses (weights).
NOTE: Symbols of the defined load types may also be presented on the graphical
screen (in an arbitrary Robot layout), if the Symbols option located on the
Loads tab of the Display attributes dialog box is checked.
There is INFO field at the end of each load record in the load table. It allows adding
description to each load case (a user has a possibility of additional description of acting load
e.g. a load transferred from roof to beam).
In the Robot program positions of center of gravity and geometrical center of a structure are
determined in the following manner:
Geometrical center
i = x,y,z
Center of gravity
For a 3D structure:
the dead load acts in Z direction
Xc[0] = Sum(My(0,0,0))/Sum(Pz)
Xc[1] = -Sum(Mx(0,0,0))/Sum(Pz)
the dead load acts in X direction
Xc[2] = -Sum(My(0,0,0))/Sum(Px)
For plates and grillages:
Xc[1] = - Sum(Mx(0,0,0))/Sum(Pz)
Xc[2] = 0
For 2D frames and trusses
For plane stress / deformation structures
Xc[1] = 0
structures.com
96
Xc[2] = -Sum(My(0,0,0))/Sum(Px)
For axisymmetric structures
the dead load acts in Y direction
Xc[0] = 0
Xc[1] = 0
Xc[2] = -Sum(My(0,0,0))/Sum(Px)
3.9.1 Load Case Combinations
The Robot Millenniumsystem also allows the creation of combinations of the generated load
cases. To do this, select the Loads/Combinations option. Once the combination type and
combination nature is specified (USL, SLS, accidental), the combination name should be
given and the combination load cases along with their factors of safety should be defined.
Once the general combination type is chosen, the Combination dialog box, as shown below,
will appear on the screen.
All defined load cases are found in the field
located in the left part of the dialog box. In
order to create a load combination for
them, highlight the appropriate load cases
and press the button. The selected load
cases, with the appropriate factor of safety
assigned to the load type, will be added to
the field found in the right part of the dialog
box thus defining the combination (the
factor of safety may also be entered
manually in the Factor field). Pressing the
New button may create a new load case
combination. It is possible to modify the
existing load combination by pressing the
Change button. Modification is performed
in the same way as a combination
definition.
The current program version enables generating combinations containing moving load cases.
This option, however, involves the following limitations:
after generating a linear combination containing moving load cases, additional three
auxiliary cases are defined (as for code combinations); these cases are available on the
selection list of load cases; a combination generated by the user is accessible only in the
combination dialog box and table; the user may not request results for it, because results
for all the components are obtained (a combination defined with moving load cases
includes component cases, similarly as code combination)
quadratic combinations (COMB QUA) may not include moving load cases or
combinations that contain such cases; for quadratic combinations including moving load
cases results are unavailable
linear combinations may contain several moving load cases (moving load cases can be
nested); there is a possibility to apply quadratic combinations in linear combinations
including moving load cases.
structures.com
97
3.9.2 Code Combinations
The system of Robot Millennium is equipped with an option that allows one to form load
case combinations described by the selected code (the code should be chosen in the Job
Preferences: option Codes/Code Combinations). The option is available from the menu by
choosing the command: Loads/Code Combinations.
Automatic code combinations (ponderations) in the Robot Millennium system provide the
user with an easy method to define and calculate selected combinations. The above dialog
box is composed of six tabs: Cases, Combinations, Groups, Relations, Simplified code
Combinations and Selection. Once the structure creation parameters have been defined (load
cases, combination templates, and additional relations etc.), press the Calculate button. The
current code combination calculations depend on the indication of all possible (theoretically
allowable) combination cases. For a detailed analysis of code combinations SLS (ULS,
exceptions) cases are created, allowing to see results for each of the combinations separately
(composite combinations), as well as the composite combinations. These cases are used for
code combinations. For a tabular presentation of maximum and minimum value presentation,
cases marked SLS+, ULS- (SLS+, ULS-, ALS+, ALS-) are created. These allow to display the
appropriate extreme values. The selection of cases SLS, SLS+, SLS- makes no difference in
a graphic presentation.
The Cases tab shown in the figure is used
to select the load cases which have been
defined for the structure and which will be
included in creating code combinations. A
list of the defined load cases with assigned
natures is located in the Active Case
Selection field. Appropriate coefficients
used during the creation of the combination
are defined for each load case defined by a
selected nature. By default, all cases as
selected in the Selection of Active Cases (a
symbol appears by the name and
number of the case), that is, all will be
considered when creating the code
combinations. If any of the load cases
should not be considered when creating
the code combinations, click on it with the
right mouse button.
structures.com
98
The Combination Definition tab is used to
define the type of code combinations that
will be created.
Numeric procedures allow to calculate
numerous combination types (patterns)
described in the code files. Depending on
the combination method and coefficient
number these regulations are included in
the template used in various codes as
follows: requirements for dead load, live
loads, accidental load and seismic load
combinations. Which of the regulations the
program should consider is defined by the
code regulation file. NOTE: the current
version of the Robot program includes the
additional application PondEdit (in the
folder SYSTEM / EXE of the Robot
program) which enables edition of the
existing code regulation files or definition of
new regulations.
Just as during the active case number, the user may decide before calculating code
combinations, which of the proposed sets to disregard. Leaving all active will result in a full
combination list according to the indicated templates.
The Groups tab is used to define/display groups, that is sets of load combinations connected
by logical relations.
In the current version of the program, all dead loads are connected in one group of
combined (acting simultaneously) loads. Snow and wind loads create a separate group of
self-exclusive loads for which the occurrence of one case excludes the other. All other live
loads occur as loads that are not inter-related.
The program creates the above mentioned groups and relations automatically. This solution
resolved the problem of base combinations. For more advanced ones, options found on the
Relations tab may be used. This allows for creating logical operations on groups of cases
within natures. AND, OR (INCLUSIVE) and OR (EXCLUSIVE) will be the logical operators.
When creating the operations, the user may use brackets to join or exclude selected groups
of cases. The functioning of particular operators will be explained in a short example. Let us
assume, that the following three groups of load cases are defined in a structure: G1, G2 and
G3. The logical operators function in the following manner:
AND - this operator results in
simultaneous activation of all
loads (loads from load cases G1,
G2 and G3 will be applied
simultaneously to the structure);
one may represent it symbolically
as
G1 G2 G3
OR (EXCLUSIVE) - this operator
results in the situation where the
loads from particular load groups
will be mutually exclusive (either
loads from load case G1, or G2,
or else G3, will be applied to the
structure); one may represent it
symbolically as:
G1
G2
G3
structures.com
99
OR (INCLUSIVE) - this operator
results in allowing all the possible
combinations between groups of
load cases; one may represent it
symbolically as:
G1
G2
G3
G1 G2
G1 G3
G2 G3
G1 G2 G3
For more elaborate tasks (more than 10 load cases for which code combinations are created)
creating code combinations may be time consuming. For this reason, a mechanism for
defining simplified code combinations generating extreme combinations due to a specific
result or their combinations has been foreseen. Options found on the Simplified Combinations
tab are used for this purpose.
In such a case, the user must provide the list of points and define the deciding value
(indicated force, moment, stress). The program will decide which combinations meet such
conditions, and only those values will be saved as code combination for the chosen bar. In
the case of providing an interaction between two selected values, the system will try to select
the extents of those variables.
In the case of more complex tasks, creation of simplified code combinations may also be
carried out for selected structure nodes and/or bars. The options provided on Selection tab
are provided to realize such tasks:
all bars (values selected on the Simplified Code Combinations tab will be verified for all
bars of a structure) or bars from a given list (values selected on the Simplified Code
Combinations tab will be verified for the bars of a structure selected by the user; the list of
selected bars should be introduced in the appropriate field).
all nodes (values selected on
the Simplified Code
Combinations tab will be
verified for all nodes of a
structure) or nodes from a
given list (values selected on
the Simplified Code
Combinations tab will be
verified for the nodes of a
structure selected by the user;
the list of selected nodes
should be introduced in the
appropriate field).
structures.com
100
3.9.3 Moving Loads
The option for defining a moving load allows one to analyze a structure with a moving load
applied. The moving load may be defined by any combination of forces (the definition of loads
originating from vehicles may contain concentrated loads, linear loads and surface loads).
The option is available by:
selecting Load/Special loads/Moving command from the menu
by pressing the Moving loads icon in the toolbar.
Then, the dialog box shown below will appear on screen.
Moving loads are defined by indicating the vehicle and its route
over the structure. The vehicle is a set of forces with a defined
direction, value and position. Each step consists in moving the
vehicle from one position to another, while the set of forces
applied to elements is created for each position. That is why the
case of moving load is treated as a set of several static load
cases (describing successive positions of the vehicle).
The upper part of the dialog box contains the following icons:
= - allows one to add a new type of vehicle
= - allows one to delete a selected vehicle type from the active
list
= , , and - allow one to display the list of active
vehicles as big icons, small icons, short list or full list
= - allows one to delete from the active list all the vehicle types
that are not used for designing a particular structure.
The process of defining a moving load case in a structure may be divided into the following
stages:
definition and selection of the vehicle that will move over the structure; to select a vehicle,
one should highlight it in the list of active vehicle types
definition of a moving load case
To define a moving load case, one should type the number and name of this load,
pressing the New button results in creating a new (next) moving load case
definition of the route to be followed by the selected vehicle
To define a route to be followed by a vehicle, one should press the Define button. It will
result in opening the Polyline contour dialog box. Pressing the Parameters button
results in opening an additional dialog box where one may define route parameters
(multiplication coefficients for the loads belonging to the moving load, etc.)
The following two are the basic parameters defining a moving load:
Step the size of the step by which the vehicle moves over the structure (the step defines
successive vehicle positions); and
Load direction direction of the forces defining a moving load
structures.com
101
definition of the plane of load application; this option describes the method of
redistribution of forces defining a vehicle moving over the structure. There are two
possibilities:
Automatic - forces are automatically distributed over the closest elements from among structure
elements
Selection - forces will be applied to the closest elements (or nodes belonging to these elements)
taken from the <list_of_elements> available in the bottom right part of the dialog box; if
the Consider vehicle dimensions option is switched on, then bars are selected on which
a load induced by vehicle forces will be generated; this selection is determined by
dimensions of the vehicle contour: b width, d1 and d2 distances between the load
and vehicle contour from the front and from the back, respectively (parameters of
vehicle contour are determined for each vehicle together with a definition of a set of
loads in the vehicle definition dialog box).
During generation of loads on bars resulting from a vehicle load the program takes account of
all bars or of their selection determined on the Application plane selection list. These bars
are projected on the plane formed by a segment of the route polyline and a perpendicular
vector defined by the user as Load direction. If the Consider vehicle dimensions option is
switched on, then on the projection plane the vehicle contour is determined and selection of
bars in the projection is narrowed down to these bars that are contained within or intersect
with the vehicle contour. This type of limitation for selection of bars on which vehicle loads are
generated may help in situations when automatic bar search might cause certain problems.
Pressing the Apply button results in creating a new moving load case whose parameters are
defined in the above dialog box.
The vehicle route, vehicle, elements bearing the load originating in the vehicle, and a set of
loads - all these can be visualized after calculations of the structure are performed
(Visualization / Loads option).
The results obtained for the moving load case may be presented in two ways. The first
method consists in presenting the results of a static case for the moving load position
selected by the user. There are options available that allow one to change the position of the
moving load. The user may move the load step by step or take advantage of the animation of
the vehicle and the results for the moving load. The second method consists in presenting the
changes of the value of a selected quantity in a selected point, due to the movement of the
load over the structure, i.e. the presentation of the influence lines for a selected quantity (see
section 5.11).
NOTES CONCERNING THE USE OF MOVING LOADS FOR DIFFERENT STRUCTURE
TYPES
Bar structures
In the case of bar-type structure (FRAME, TRUSS, GRILLAGE), it is possible to apply a load
generated by a vehicle defined by means of concentrated and linear forces. Concentrated
forces are applied as bar load. In the case when a concentrated force is not directly applied
into a bar, the program operates according to an algorithm that resolves the force over the
neighboring bars. The linear load is modeled by 10 concentrated forces along the length of
the linear load.
For bar-type structures, one cannot use vehicles defined by means of surface loads.
Plate-shell structures
In the case of surface structures (PLATE, SHELL), one may use all types of vehicles (defined
by means of concentrated, linear and surface loads). A concentrated force is applied as a
geometrical load at a point, applied to surface elements, and it is resolved into bar elements.
Linear force is applied as geometrical linear load applied to surface elements. It is not applied
to bars. Surface force is applied as a geometrical contour load, applied to surface elements.
When using this type of load, one should pay attention to the place of contour load
application, for it is generated on each panel located within the contour limits. For the load
type to be applied only to selected panels, one should use the selection options in the
definition of moving load case.
structures.com
102
In the case of mixed structures (bar-shell), linear and surface loads are applied only to the
surface element. For a load originating in a vehicle to be applied directly to bars, one should
use concentrated forces in vehicle definition.
NEW VEHICLE DEFINITION
A vehicle is a set of forces which constitutes the load in a moving load case. The vehicle
moves along the defined route by a distance referred to as a step of the route in a moving
load case. Each individual position of the vehicle is saved in the successive case component.
A vehicle definition may include concentrated, linear forces or planar forces acting on a
rectangular contour.
Once the New vehicle
icon is pressed in the
Moving loads dialog box,
the dialog box shown
below will appear on
screen.
The dialog box allows one
to select the vehicle that
will be added to the active
list of vehicles located in
the Moving Load dialog
box. The dialog box is
divided into several parts.
There are two vehicle types available:
symmetric vehicles loads are defined as pairs of forces with a defined spacing,
symmetrically with respect to the vehicle axis at any point over the axis length; forces are
defined as vertical loads (possible horizontal forces caused by braking may be obtained
by defining values of route coefficients); vehicles of this type are used e.g. for modeling
bridge loads
arbitrary vehicles loads defined as forces at any point over the length of the vehicle
axis and in any distance from it; concentrated forces may be defined in the directions X,
Y, Z of the local coordinate system of the vehicle; route coefficients are not applied for
this vehicle type; this type of vehicles is used e.g. for modeling loads induced by
overhead traveling cranes.
The upper left corner of the dialog box contains the Vehicle selection field where one finds
two selection fields: Code (database) and vehicle name. Moreover, one finds here three
buttons:
New - it allows one to define a new vehicle; pressing this button results in opening an
additional dialog box where one can define the name of the new vehicle. There are two
situations possible: if the user presses the Add button after defining loads for the new
vehicle, the vehicle will be added only to the list of the active vehicles in the Moving
Loads dialog box; if the user presses the Save to database button after defining loads
for the new vehicle, there will appear a new dialog box where one may select the
database (folder) of vehicles where the new vehicle may be stored.
Save to database - allows one to save the new vehicle to a selected vehicle database;
pressing this button results in opening the dialog box where one may select the database
(folder) of vehicles where the new vehicle may be stored.
structures.com
103
Delete - allows one to delete a selected vehicle from the vehicle database.
NOTE: The user may define his own database of vehicles (users database). It is
possible in the Job preferences dialog box (Vehicles database option).
The upper right corner displays a schematic drawing of the selected vehicle.
The middle part of the dialog box contains a table presenting a description of the loads for the
selected vehicle. The program provides three types of loads: concentrated force, linear load
and surface load. Each parameter of the load defined for the selected vehicle may be edited
(changed).
The following parameters must be defined for individual load types:
concentrated
force
Symmetric vehicles:
F- value of a concentrated force
X - coordinate value of the point where the force is applied (along the vehicle
axis)
S - width of force spacing
Asymmetric vehicles:
FX, FY, FZ - values of the concentrated force
X - coordinate value of the point where the force is applied (along the vehicle
axis)
Y - coordinate value of the point where the force is applied (perpendicularly to the
vehicle axis)
linear load Symmetric vehicles:
Q- value of a linear load
X - coordinate value of the line along which the force is applied (along the vehicle
axis)
S - width of linear load spacing (only in the Y axis direction)
Dx - length of a segment along which the load acts (along the vehicle axis)
Dy - length of a segment along which the load acts (perpendicularly with respect
to the vehicle axis)
Compared to symmetric vehicles, instead of the S value there is
vehicle axis)
planar load Symmetric vehicles:
P - value of a planar load
X - coordinate value of the line along which the force is applied (along the vehicle
axis)
S - width of planar load spacing (only in the Y axis direction)
Dx - length of a rectangle side upon which the load acts (along the vehicle axis)
Dy - length of a rectangle side upon which the load acts (perpendicularly with
respect to the vehicle axis)
Compared to symmetric vehicles, instead of the S value there is
vehicle axis)
The lower part of the dialog box contains two fields:
Vehicle dimensions:
b - vehicle width
d1 - distance between the load and the vehicle contour (from the vehicle front)
d2 - distance between the load and the vehicle contour (from the vehicle back)
structures.com
104
Units:
forces - presentation of units in which the values of forces will be displayed (the units
may be modified in the Job Preferences dialog box)
length - presentation of units in which the values of length will be displayed (the units
may be modified in the Job Preferences dialog box).
ROUTE PARAMETERS
The option used to determine scaling coefficients that allow the user to obtain the real
(calculation) value of a load originating in a moving vehicle. The option is available by
pressing the Parameters button located in the Moving Loads dialog box.
NOTE: To open the coefficients definition dialog box, the user has to select a
polyline for which the load value multiplication coefficients will be defined.
Once the Parameters button is pressed in the Moving Loads dialog box, the dialog box
shown below appears on the screen.
The above dialog box contains a table presenting the following parameters of a route
(polyline):
Edge column. - number and name of the polyline defining the vehicle route
Gamma column - a vehicle may be revolved around the vertical axis; it results in changing
the position of forces (their direction remains unchanged); rotation is determined by
Gamma angle, defined in the same way as the user defines the Gamma angle while
describing the properties of bar elements
coefficient VL column - scaling coefficient for the vertical force (V) operating on the left; it
allows the program to multiply the load value (e.g. a concentrated force) to obtain the real
(design) value of the load
coefficient VR column - scaling coefficient for the vertical force (V) operating on the right;
it allows the program to multiply the load value (e.g. a concentrated force) to obtain the
real (design) value of the load
coefficient HL column - scaling coefficient for the transversal horizontal force (H)
operating on the left; it allows the program to multiply the load value (e.g. a concentrated
force) to obtain the real (design) value of the load (the horizontal load perpendicular to the
route direction)
structures.com
105
coefficient HR column - scaling coefficient for the transversal horizontal force (H)
operating on the right; it allows the program to multiply the load value (e.g. a concentrated
force) to obtain the real (design) value of the load (the horizontal load perpendicular to the
route direction)
coefficient LL column - scaling coefficient for the longitudinal horizontal force (L) operating
on the left; it allows the program to multiply the load value (e.g. a concentrated force) to
obtain the real (design) value of the load (the horizontal load parallel to the route
direction)
coefficient LR column - scaling coefficient for the longitudinal horizontal force (L)
operating on the right; it allows the program to multiply the load value (e.g. a concentrated
force) to obtain the real (design) value of the load (the horizontal load parallel to the route
direction).
If coefficients on the right and on the left have different values, then intermediate coefficients
within the width of a vehicle are interpolated depending on its width.
Additionally, the user may apply the options used to limit a vehicle position on a structure:
Vehicle position limit - route beginning and Vehicle position limit - route end. If the options are
switched on, then, at the beginning/end of the defined vehicle route, the program will
recognize the effect securing the vehicle wheels from sliding off the defined route (e.g. wheels
of an overhead traveling crane should not slide off the crane beam).
The lower part of the dialog box contains the Tolerance edit field. It is a parameter assigned
to a defined case of moving load analysis. Tolerance determines the conditions of generating
loads that originate from a defined vehicle on bars. Moving load tolerance is a parameter
assigned to a defined case of a moving load analysis. The tolerance defines maximal
distance from a bar, for which a concentrated force determined in a vehicle definition will be
applied directly to a bar without distributing forces on bars.
If the Moment originated from force eccentricity option is switched on, then a concentrated
moment load resulting from a concentrated force is generated and applied to a bar
considering the eccentricity.
3.9.4. Autoloader Moving Loads
The moving loads option described in the previous chapter enables analysis of loads caused
by a set of forces (vehicle) which is shifted - during the analysis - into successive positions
along a route defined on a structure. Thus a composed case is obtained which contains
components of each of the vehicle positions. This approach is convenient when analyzing the
vehicle movement along a single route. In case of bridge structures for which several
carriageways (routes along which a vehicle moves) may be defined, for the moving loads
option a user should define several moving load cases and analyze their combinations; it
results in obtaining a great number of combinations.
The Robot program provides access to the Autoloader module used to analyze bridge loads
with code requirements considered. To perform an Autoloader analysis of bridge loads,
influence lines (areas) are applied; an influence line is a two-dimensional form of a three-
dimensional influence surface. Based on an influence surface, a user may obtain load
distribution which results in the most adverse effect for a quantity described by a given
influence surface. Autoloader conducts such an analysis and provides the most unfavorable
load distribution for a given influence surface.
Autoloader works as an internal module of the Robot program. As a result of this module
operation, new load cases are generated with the most unfavorable load distribution for a
given influence surface.
The Moving loads - Autoloader option is available from:
the menu by selecting the command: Loads/Special Loads/ Moving - Autoloader
from the toolbar by pressing the Moving loads icon.
structures.com
106
Operation of the Robot-Autoloader interface has been divided into several stages (data
definition and analysis). The following stages may be distinguished:
definition of geometrical data and selection of an influence surface
definition of parameters for the Autoloader module
model generation and start of calculations for an auxiliary case with unit loads
generation of an input file and start of the Autoloader module analysis
generation of load cases in a model of the Robot program.
NOTE: Influence surfaces are generated for a unit load, which acts in the direction of
Z axis (sense: minus Z). Therefore, a deck should be defined in a horizontal
plane (parallel to the XY plane); the bridge loads obtained will act in the
vertical direction (in the direction of Z axis).
The above dialog box enables determining geometrical data and defining (selecting) an
influence surface (the first point from those listed above). Influence surfaces in the Robot
program are generated by means of the static method. It means that they are generated by
applying a concentrated unit load at successive points. Once all the unit cases are solved, an
influence surface taking results from successive cases is constructed.
Influence surfaces are defined identically as influence lines for the analysis of moving loads.
Both, influence line and surface are a diagram of a relationship between a certain result value
(e.g. force or displacement) and position of a unit force. A value of the influence line at a
given point is equal to the value of a selected result quantity while a unit force is positioned at
this point.
structures.com
107
In the Deck - surface - Object field a user may enter or indicate graphically an object of a
loaded surface on which an influence surface will be generated. Loaded surfaces are used to
determine a deck contour. These objects do not contribute to the structure stiffness, they are
only geometrical objects used for generating loads.
The Bars: Automatic / Selection option is used to determine the plane to which the load will be
applied; the option enables a user to specify the method of distributing the forces that define a
vehicle over structure bar elements. The are two options available:
Automatic - forces are distributed automatically on the closest elements taken from all
structure elements
Selection - forces are applied to the list - defined by a user in the edit field - which
contains the closest elements or nodes belonging to these elements.
In the Carriageway - line/arc field a carriageway position on the deck may be determined.
Carriageways are defined by means of a carriageway axis (line or arc) and its width. A user
may define several carriageways on a deck. However, it should be remembered that all
carriageways must be of the same type (straight segments or arc fragments).
After accepting a defined set of geometrical parameters and pressing the Apply button, the
Autoloader Parameters dialog box opens, which allows defining load parameters and
analysis performed in the Autoloader module (among other things, values of coefficients and
loads applied during analysis, detailed parameters for a selected national code); however, it
should be remembered, that to have the dialog box for parameter definition opened, a correct
set of geometrical data is required:
number of the deck object must be indicated
at least, one carriageway must be defined.
The Autoloader module enables bridge load analysis according to the codes listed below:
UK - Highways Agency Departmental Standard BD 37/88 Loads for Highway Bridges,
which includes BS5400 Part 2:1978
UK - Highways Agency Departmental Standard BD 21/97 Assessment of Highway
Bridges and Structures
Malaysia - JKR Highway Loading standard
Hong Kong - guidelines from Hong Kong, as an extension to BD 37/88 code
RU - RU chapter from BD 37/88 code
RAIL - Autoloader specific parameters, to be applied with railway loads
Australia - AUSTROADS standard
USA - AASHTO standard.
The following types of bridge loads can be analyzed in the Autoloader module:
normal loads consisting of planar loads (Uniformly Distributed Load - UDL) or linear loads
(Knife Edge Load - KEL); it should be noted that in various codes different terminology is
applied for this type of loads: BD 37/88 defines it as HA, AUSTROADS applies L44,
whereas JKR applies LTA
abnormal vehicle loads (loads resulting from a set of concentrated forces); the
terminology applied in codes varies: BD 37/88 determines vehicles as HB, AUSTROADS
as HLP, whereas JKR applies SV
SDL - superimposed dead loading.
structures.com
108
It should be noted that the program performs the analysis using certain degree of
approximation. It is only a tool and results of its analysis should be checked to ensure
correctness of a solution. The analysis consists in positioning the loads - by the program -
within the contour of an influence surface considering appropriate code coefficients
Influence surfaces are saved as a mesh of points with determined division, while values
between points are linearly interpolated. Thus, an analyzed vehicle position is determined
together with defining a certain step. A user may manipulate both, mesh division and vehicle
step. A coarse mesh may result in incorrect solutions. As the mesh becomes finer or the
vehicle step decreases, accuracy of solution grows. However, it should be remembered that it
requires longer calculation time, therefore, as in most cases of numerical calculations, a user
should select a solution being a trade-off between calculation costs and their accuracy.
3.10 Snow/Wind Loads
Once the Loads/Special Loads/Snow/Wind
command is selected from the menu, snow/wind
loads can be generated.
The dialog box shown beside will be displayed on
the screen. The options found in this dialog box
allow definition of snow/wind loads according to
American code ANSI/ASCE 7-98, Eurocode 1 (the
general code and several codes for individual
European countries see the list below), French
codes NV 65/N84 Mod.96 or NV65+Carte96 as
well as Algerian code DTR C2-47/NV99.
Eurocode 1 is available with National Application
Documents of the following European countries:
Austria, Belgium, Denmark, Finland, France, Spain,
Germany, Greece, Iceland, Ireland, Italy,
Luxembourg, the Netherlands, Norway, Poland,
Portugal, Sweden, Switzerland, UK.
NOTE: The SNOW/WIND LOADS option is available only for two structure types:
PLANE FRAME and PLANE TRUSS. Automatic generation of such loads is
impossible for other structures or when no structure has been defined.
The Snow \ Wind Load dialog box contains condition options of structures for which
snow/wind loads will be generated:
Envelope - Option for defining those structure elements for which snow/wind loads will be
generated. Numbers of selected nodes create the envelope.
Auto - Begins the automatic envelope generation. Parameters are set according to
the selected options for automatic envelope recognition.
Without
Parapets - If this option is active, an automatic envelope generation with no regard to
parapets will be performed.
Total depth - Option for defining the building depth (length). It is necessary for the definition
of global snow/wind coefficients of the structure.
NOTE: To assure proper generation of snow/wind loads this option MUST be used.
Bay spacing - Option for setting the bearing element spacing on the structure length. It is
necessary for accumulation of tributary loads.
NOTE: To assure proper generation of snow/wind loads this option MUST be used.
structures.com
109
To assure proper generation of snow/wind loads the envelope and the two basic building
parameters: depth and spacing must be defined.
The Parameters button is located in the lower part of the dialog box. It opens the dialog box
with detailed snow/wind load parameters. For example, once the detailed parameter option
dialog box is opened and the Global Parameters tab selected, the dialog box shown below
will appear.
The above dialog box contains the
basic options required by the
selected national snow/wind code
set in the Preferences module.
Once the option is chosen, a
corresponding icon appears in the
upper right corner of the box.
The dialog box has been divided
into several fields containing
groups of specific options. The
following options are located in the
upper part of the dialog box:
Exposure category, Building
category and Altitude above the
sea level. Below there are options:
Structure height (in the Structure
dimensions field) and Hurricane
prone region.
Three additional tabs are also located in the above dialog box. These are: Wind, Snow and
Permeability. These allow for the definition of snow and wind load parameters as well as
building permeability parameters according to the requirements of the selected code.
SELECTED REFERENCES - CODES
European Code Eurocode 1: Basis of Design and Actions on Structures
Regles N V 65 Et Annexes Regles N 84 Regles Definissant Les Effets De La Neige Et Du
Vent Sur Les Constructions Et Annexes, Eyrolles 1987
structures.com
110
3.10.1 Snow/Wind Loads in 3D
Once the user has defined
general structure parameters
indispensable for generating
snow/wind loads on a 2D
structure (i.e. envelope,
spacing, depth) and pressed
the Generate 3D button in the
Snow/wind loads dialog box,
there appears on screen the
dialog box presented beside.
It is titled Snow/wind loads
3D geometry.
Snow/wind loads in 3D are created in the following manner:
snow/wind loads are generated for a previously defined 2D frame
a 3D structure (frame) is generated on the basis of the 3D structure parameters
determined in the above-presented dialog box
the defined 2D frame is copied a determined number of times (according to the user-
defined spacing)
horizontal bars (purlins) are generated between consecutive frames; loads will be carried
from the structure cover through the purlins to frame bars.
Snow/wind 2D loads calculated for a two-dimensional frame (expressed as the value of force
per area unit) are gathered from the relevant surfaces and applied to the purlins as a uniform
load (value of force per length unit). For purlins, the load is gathered from a flat surface of the
area limited by the neighboring purlins or the edge of the surface (if there is no neighboring
purlin in this direction).
One can define the following parameters in the Location of frames field:
Location of frames - manner of positioning 2D frames (regular - constant distance
between frames, irregular - different distances between consecutive frames)
Number of frames - expected number of frames in the final 3D structure
Spacings - spacing between frames; the value of spacing is determined for regular
positioning of frames; for irregular positioning one must define n-1 spacings divided by
separators, where n is the number of frames.
In the Location of longitudinal elements field, one can select a bar in the structure (by means
of < and > buttons), and then, determine its parameters. By the side of the Bar field, there
appears the number of the selected bar. At the same time, the bar is highlighted in the field
presenting a schematic drawing of the 2D frame (a bar may also be indicated directly on the
2D frame drawing). The Location of longitudinal elements contains also the following options:
Section button - allows one to select a section to be used in creating purlins (longitudinal
beam)
structures.com
111
Location - relative or absolute positioning of purlins on the current bar of a 2D frame
Offsets - activating the option makes the program account for offsets during the creation
of purlin elements.
Pressing the Generate 3D button results in starting the generation of snow and wind loads in
3D.
3.10.2 Wind Loads Masts / Towers
The option is used to generate snow/wind loads on the following types of structures: truss
masts, columns and towers (high-voltage line support towers), transmission towers,
broadcasting masts, etc. The generation of snow/wind loads is started by:
selecting Loads / Special loads / Wind on towers command from the menu
pressing the Loads on Towers icon.
Once any of the above is performed, the dialog box shown below appears on screen.
NOTE: The Wind Loads on Towers option is available only for bar 3D structures:
FRAME 3D and TRUSS 3D. The base of the created structures must be an
equilateral triangle or a rectangle. Generation of wind loads for masts, towers
and columns is performed according to American EIA code and French code
NV 65.
The dialog box shown below consists of four tabs: General, Specific, Segments, and
Equipment (the options located on individual tabs depend on the selected code). On the
General tab, for instance, one may define the following parameters: state, wind velocity, wind
pressure and directions.
The central part of the dialog box
contains the Structure height field
allowing one to specify structure
height for calculation of a wind
pressure value.
The lower part of the dialog box
contains options that allow one to
select:
the manner of defining wind
pressure (there are three
possibilities: automatic, on the
basis of the selected base
wind velocity for a state that
has been set, manual, by
typing the value of the base
pressure and manual, by
defining wind velocity)
number and direction of
operation of wind cases
(according to code - assumes
structure symmetry; all
standard directions - the
option is useful in the case of
asymmetrical structures)
structures.com
112
manner of calculating wind pressure for particular structure elements (there are two
possible choices: constant - calculated according to the highest point of a segment;
variable - calculated for each element separately, according to the highest point of an
element).
The options located in the remaining tabs allow one to:
Specific tab allows one to define additional load parameters (icing, dynamic wind action,
etc.)
Segments tab allows one to define parameters of segments into which the designed
structure of the mast/tower type is divided (e.g. high-voltage line columns, broadcasting
masts, etc.)
Equipment tab allows one to define additional surfaces that have significance influence
on the load applied to the structure; this category covers antennas, tables, structure filling,
etc.
The lower part of the dialog box contains standard buttons (Close, Help and Cancel) and the
Generate button. Once the last is pressed, the program starts to generate wind loads on a
previously defined tower and to prepare the relevant calculation note.
SELECTED REFERENCES CODES
TIA/EIA STANDARD Structural Standards for Steel Antenna Towers and Antenna Supporting
Structures TIA/EIA-222-F (Revision of EIA/TIA-222-E), Telecommunications Industry
Association, June 1996
3.10.3. Snow/Wind Loads on 3D Objects
The operation of the option used to generate snow/wind loads for surfaces is similar to the
manner the generation of such loads for 2D frames is performed (see chapter 3.10). The
successive steps while generating snow/wind loads for 2D frames and their equivalents for
3D structures are as follows:
2D frame 3D structures
1. generation of envelope 1. definition of surface
2. definition of code parameters 2. definition of code parameters
3. generation of coefficients 3. generation /modification of coefficients
4. applying loads to structure bars 4. generation of surface loads (these surfaces - by
means of the surfaces defined - are transferred onto
structure bars).
The Snow and Wind 2D/3D option may be activated from:
the menu by selecting the option Loads / Special loads / Wind and Snow 2D/3D
the toolbar by pressing the icon.
NOTE: The option is available only for the following structure types: 3D Frame and
Shell (at present, for the French code NV65).
structures.com
113
To generate snow/wind loads on surfaces, the user should:
determine a list of surfaces on which snow/wind load cases will be generated (dialog box
above)
determine parameters used for automatic generation of snow/wind loads (once the
Parameters button is pressed in the dialog box above, the program opens the dialog box
where parameters of snow/wind loads may be defined); after defining parameters of the
snow/wind loads and pressing the Apply button, the Parameters dialog box closes and
the Snow/Wind Loads dialog box is displayed on the screen.
The Snow/Wind Loads dialog box presents a view of the defined structure and selection list
with generated cases of snow/wind load. The structure view (see the figure below) shows
loaded planes using the color legend. In the table contained in the dialog box all structure
faces are displayed. For these faces which have not been automatically loaded by the
program, load values equal zero.
structures.com
114
Once the Generate button is pressed, the above dialog box is closed and the loads are being
generated on surfaces. Loads are applied in the form of pressure on planar objects in the
local or global system. Afterwards, the loads are transferred onto bars in the identical manner
as while generating loads on bars from 3D objects; loads are transferred onto bars during
generation of a structure model.
The set of coefficients that are used to determine a snow/wind load is saved in the project so
that a user is able to edit the coefficients and parameters of a snow or wind load.
While discussing the definition of snow/wind loads, the Cladding option should be mentioned,
which is applied to define a new object type - a surface being the object that enables
transferring planar loads onto bars (however, it is a non-load-bearing object). The option
allows definition of bar structures with panel walls. An object like this facilitates considerably
load generation by providing the possibility to define real structural objects which do not carry
structure loads, like panel walls and roof cladding.
The Cladding option is available from:
the menu by selecting the command: Geometry/Additional Attributes/Cladding or
the toolbar by pressing the icon.
A surface object is generated in the form of a face with defined cladding. A surface is defined
in the same manner as a panel (by indicating an internal point or the list of linear objects).
The option allowing surface definition by assigning cladding to the face type object is
available for the following structure types: bar structures and shell structures. It is assumed
that for volumetric structures the face object behaves like a face of a volumetric structure; it
is not allowed to define loaded surfaces on such an object.
3.10.4. Wind Loads for the Structure with a Polygon Base (Prism)
The option is used to generate wind loads on axisymmetric structures (structures whose base
is a regular polygon). The option is available for the following codes:
American snow/wind code ANSI/ASCE 7-98
French snow/wind code NV65 (article 3).
The drawing below shows structure types for which wind loads may be generated.
NOTE: The program does not generate a finite element mesh on a surface with
cladding. It is an auxiliary object used to define loads.
structures.com
115
The option allows generating wind loads on selected surface finite elements and on panels. It
means that prior to activating this option, a user should select panels or finite elements on
which wind loads are to be generated.
Generation of snow/wind loads may be started after:
selecting the Loads / Special loads / Wind on cylinders command from the menu
pressing the Wind on cylinders icon.
NOTE: The Wind on cylinders option is available only for shell structures.
NOTE: Wind loads are generated based on American code ANSI/ASCE 7-98 or
French code NV 65.
Parameters defined for wind load generation for this type of structures are similar to
parameters discussed in the previous chapters (the parameters concern American snow/wind
code ANSI/ASCE 7-98 and French snow/wind code NV65).
structures.com
116
In the top part of the dialog box in the Wind directions field:
on the Generating list - the user may determine direction along the height of the defined
axisymmetric structure (cylinder); directions X, Y and Z of the global coordinate system
may be selected
the Wind options allow the user to define directions of the generated wind loads;
directions selected refer to the global coordinate system.
The following parameters may be defined in the bottom part of the dialog box in the Structure
type field:
for the American code:
Category - the list contains available categories of the calculated structure:
category I square
category II hexagonal or octagonal
category III round moderately smooth
category IV round rough (D/D=0.02)
category V round very rough (D/D=0.08).
For each category the Number of faces option is also available; from the list a number of
axisymmetric structure faces should be chosen.
for the French code:
Category - the list contains available categories of the calculated structure:
category I prism: 3 or 4 faces
category II prism: 5 to 10 faces without curved ribs
category III prism: 11 to 20 faces with/without curved ribs
category IV cylinder: round base with thin or thick ribs (sharp edges)
category V prism: more than 20 faces with/without curved ribs
category VI smooth cylinder with round base without ribs (glossy).
If category II has been selected, then the Number of faces option becomes available; the user
should select a number of axisymmetric structure faces from the list. For the remaining
categories the Number of faces option is inaccessible.
After defining basic parameters of axisymmetric structure and pressing the Parameters
button, the dialog box allowing definition of wind load parameters is opened on the screen.
This dialog box consists of three tabs:
structures.com
117
General
Specific
Segments.
The options provided on the tabs enable as follows:
General tab definition of basic load parameters (structure location, wind pressure,
structure reference level, etc.)
Specific tab definition of additional parameters of wind action (e.g. topographic factor
and gust effect factor for the American code or dynamic wind action for the French code)
Segments tab definition of parameters of segments into which the structure is divided.
The parameters given on these tabs depend on the selected snow/wind code.
In the bottom part of the dialog box there are standard buttons (Close, Help and Cancel) and
Generate button, which when pressed runs generation of wind loads on a defined structure
and calculation note.
NOTE: Since it is not specified in the ASCE 7-98 code how the resultant force should
be distributed over the structure surface, the method of distribution adopted
for the French wind load code NV 65 modifies 99 is applied in the program for
the ASCE 7-98 code.
3.10.5 Load Report (Loads Taken from Database)
At present, the program enables preparing a report of loads acting on individual structure
elements. It is a tool that enables adding up the loads applied to a given structural element.
There are serviceability and factored values of loads; serviceability values of loads come from
tables saved in the database, whereas factored values are obtained by multiplying
serviceability values by appropriate load factors.
In the program the option operates for surface loads resulting from dead loads. The dead
loads include loads caused by:
surface weight (e.g.: weight of roofing, cladding, insulation, etc.)
density (e.g.: weight of constructional materials, embankment materials, etc.).
The option serves as a handy calculator for preparing load reports. Calculated load values are
not transferred to load records. The user himself/herself defines load values for individual load
cases using values calculated in the dialog box below.
NOTE: After changing unit loads in the database, load values are not updated.
The Load Report option is accessible from:
the menu by selecting the command: Loads / Special Loads / Load Report
the toolbar by pressing the Load Report icon.
Once this option is selected, the dialog box presented below appears on the screen.
structures.com
118
The option is accessible for all the structure types available in the Robot program.
The left part of the dialog box includes options used for calculating as well as saving and
reading a load report:
list of set selection; it contains the list of saved load sets - this field also allows defining a
name of a new load set; a set name (label) is a set identifier - it is provided in a note, as
well; a set is read after selecting the set name from the list
Above the set selection list the following buttons are located:
New - pressing this button allows definition of a new set name (the current load set is
deleted)
Save - pressing this button enables saving a load set currently defined; when saving a
set, the user must specify its name
Delete - pressing this button enables deletion of a currently selected set from the list
Note - activation of the text editor containing data of the current load report (a note may
be used while preparing a printout composition)
in the Load report table data concerning the total load is displayed; successive table
records include:
load name
the Unit load value (density load or surface load) read from the database
the Layer thickness value - in the case of density, it may be edited - a default value
equals 10 cm (for surface loads, this field is inaccessible); section dimensions are the
units
value of the serviceability loading:
- for density it is the product (unit weight)*(thickness)
- for surface loads it is a unit weight
load unit: force/length^2
value of the Load factor Gf - edit field; default values are saved in and taken from the
database
value of the Factored loading = (serviceability loading)* (load factor)
unit: force/length^2
structures.com
119
the last table row contains the totals of serviceability and factored loading, i.e. the sum of
loads in the successive rows
under the table there are the options used for calculation of concentrated or linear forces
resulting from the obtained (serviceability and factored) surface load p.
- Concentrated load the product: A*B*p, unit: force
- Linear load - the product: A*p (B dimension is inaccessible), unit: force/length
- Surface load: equal to p (dimensions A, B are inaccessible), unit: force/length^2.
The right part of the dialog box includes options that allow support of the unit load database:
the Loads option group is used for selecting a load from the database; a selected load
may be transferred to the load report (located in the left part of the dialog box) by pressing
the < button; the Database field presents a name of the current database of unit loads
the Database field presents a name of the current database of unit loads; the current
database may be selected in the Job Preferences dialog box
the selection list used for selection of the appropriate data table with unit loads - the
contents of this list depend on the selected load type: Material weight, Element weight,
Soil weight or Variable loads; selection of e.g. material type depends on the database
contents: Concrete, Timber-derivative materials, Roofing, etc.
table for viewing and selecting unit load data - table elements cannot be changed; the
table contains columns saved in the load database: Material and Weight
A weight unit depends on the selected type; if it is material weight, then force/length^3 is
the unit; if it is element weight, then force/length^2 is the unit
pressing the Edit load database button opens the dialog box used for database edition;
the DATA Table dialog box contains options enabling the user to define a new record,
delete a record, modify a record (with the possibility of copying), select columns to be
displayed.
3.10.6 Automatic Definition of Loads Induced by Soil Pressure
The Robot program provides a tool for calculation of the soil pressure acting on structure
elements entrenched in soil, such as retaining walls, walls, etc. The pressure induced by soil
and by forces applied to a soil surface are taken into consideration.
A value of soil pressure may be modified depending on the work mode of a retaining
element:
pressure influenced by the presence of other objects in the case of a different object
located nearby, which reduces soil pressure
active pressure reduced pressure, if a retaining element has been displaced due to
action of external forces, in the direction compatible with the pressure induced by soil
passive pressure increased pressure, if a retaining element has been displaced due to
action of external forces, in the direction opposite to the pressure induced by soil.
For a load acting on a soil surface the linear distribution of stresses in soil is applied. The
option enables the user to:
define a soil profile taking advantage of the soil database containing soil properties
save and read a complete soil profile defined by the user and edit the soil database,
structures.com
120
define loads applied to a soil surface
calculate and display soil pressure diagrams
generate a load acting on panels or bars consistently with the calculated soil pressure.
A saved soil profile may be exchanged with other modules using soil profiles.
The Soil pressure option is available from:
the menu by selecting the command: Loads / Special Loads / Soil Pressure
the toolbar by pressing the Soil pressure icon.
Once this option is selected, the dialog box shown below is displayed on the screen
(depending on the objects chosen to which the pressure load will be applied bars or panels,
the dialog box includes slightly different data concerning the direction of load application).
The following parameters may be determined in the above dialog box:
number and name of the current load case; soil pressure loads will be generated for this
case
selection of the type of objects to which the load is applied: Panels or Bars
list of bars or panels (in the List of objects edit field) on which the soil pressure load will
be defined; take note that these objects will be loaded that are positioned below the Z
coordinate, which is assumed to be the soil level in the Parameters dialog box; the load
value depends on that how deep a given element is entrenched in soil
if a load is defined on bars, then it is necessary to specify a spacing value; since the soil
pressure is calculated as a surface load and in order to obtain a linear load on a bar it is
multiplied by the spacing width
direction of load action
structures.com
121
for panels the soil pressure is always defined as a load normal with respect to the
surface; the only parameter that the user may choose is that whether the load is defined
According to local surface system (according to the sense of the local Z axis) or Opposite
to local surface system
for bars the user may choose any load direction according to Global of Local system of
a bar.
Pressing the Apply button results in definition of a soil pressure load applied to a selected list
of elements according to the direction chosen. The load is defined for the currently selected
load case. Pressing the Close button closes the dialog box (without saving any parameters).
The Soil pressure (parameters) dialog box is opened after pressing the Parameters button
provided in the Soil Loads dialog box.
The dialog box consists of three tabs: Soils, Loads and Results. .
The Soils tab
The following parameters may be defined on this tab:
geometrical data concerning a retaining element and soil
- soil level; a value of soil level identifies the Z coordinate of the soil level position in a
structure model; this value may not be modified, since it corresponds to the level of the
first soil layer defined in the table
- angle of soil inclination to the horizontal direction
- angle of the retaining element deviation from the vertical
- distance to other object; the value should be defined if a reduction of pressure caused
by other objects located nearby (i.e. pressure influenced by the presence of other
objects) is to be considered
- underground water level is specified in global coordinates (with respect to the soil
level); the value should be defined if a reduction of pressure caused by the presence of
underground water is to be considered; a position of the underground water level is
indicated in the view of the soil cross section; it should be remembered that the
presence of underground water has effect on a pressure value due to reduction of soil
compaction and pressure induced by water
- data determining the soil work mode concerned with displacement of a retaining
element; the displacement is defined by means of the rotation of element , which is
assumed approximately to equal f/H (wall top displacement / element height); the soil
work mode may be determined as:
- active pressure Ka -> a < 0
- static pressure Ko -> = 0
- passive pressure Kp -> 0 < p
- limit value or intermediate value.
table used for definition of soil layers
The user should select a soil type on the drop-down list in the Name column and next,
determine a soil layer level by entering a value in the Level or Thickness column; the
table enables access to the soil database which is a default database determined in the
Job Preferences dialog box (the list of databases is editable after selecting the option:
Tools / Job preferences on the Databases tab / Soil Database)
a view with drawing of soil layers and depth scale; the Profile field contains the options
that allow saving and opening a user-defined profile; pressing the Save button opens the
dialog box for file saving; each profile is saved as a separate file in the MS Access
(*.mbd) program format; if the Open button is pressed, then the dialog box for opening
the *.mbd type file is activated; the Name edit field presents a path to the current file with
a soil profile.
Pressing the Edit soil database button enables edition of the current soil database.
The Loads tab
In the above dialog box external loads applied to soil may be defined. The loads are defined
in the table included in the bottom part of the dialog box. Each load is assigned a name and a
list of parameters depending on the load type. The following load types may be defined:
structures.com
122
linear load the load is described with the following data: name, distance x [length], load
intensity Q [force/length]
distributed load - the load is described with the following data: name, distance x1 [length],
distance x2 [length], load intensity P [force/length^2]
uniform load - the load is described with the following data: name, distance x [length],
load intensity P [force/length^2].
Moreover, there is a possibility to choose pressure distribution type connected with the
following codes:
French requirements: SETRA
Polish codes: PN-83/B-03010 and PN-85/S-10030
Russian code RD 31.31.27-81.
The Results tab
The tab presents:
in the right part of the dialog box the list of defined loads; the list is always composed of
a soil pressure load and user-defined cases resulting from loads applied to the backfill;
the list enables selection of cases that will be transferred to a model as a load
in the left part of the dialog box- the view of soil cross-section with a load diagram; the
diagram shows the case which is chosen on the list; while moving with the cursor on the
diagram, the user may read values at individual diagram points
messages in the case of incorrect data, provided in the field located under the list.
Pressing the Calculation note button opens a text editor with a note containing a set of data
and obtained pressure diagrams. Pressing the OK button closes the dialog box and starts
preparation for generating loads.
3.11 Cables
The Robot Millennium program provides the possibility of defining cable structures (i.e.
structures with a cable as the main load-carrying element). If one of the main dimensions of
an element is bigger than the two remaining ones and section rigidity with respect to bending
and torsion is small in comparison to tension rigidity, such element is regarded to be a cable.
The basic conclusion drawn from the above definition is that only tensile forces can be
applied to cables. However, in some cases small bending or torsional moments and shearing
forces can be applied to cables. The option that allows for attributing cable properties to
structure elements is available from:
menu by selecting: Geometry/Properties/Cables or
toolbar Structure definition by selecting the icon .
Defining a cable in a structure is similar to attributing a section to a bar or a support to a node.
The following cable parameters may be defined for a cable: label, color, cross-section,
(defines the area of a cable cross-section, material (selection of the material a cable will be
made of) as well as the following assembling parameters of a cable:
structures.com
123
stress - selecting this option allows one to define the normal stress (calculated with
respect to the cable chord) for an assembling case loads. The value of the stress should
be introduced in the appropriate field.
force - selecting this option allows one to define the cable parameter of Force Fo
(calculated with respect to the cable chord) for an assembling case loads. The value of
the force should be introduced in the appropriate field.
length - selecting this option allows one to define unloaded cable length for an assembling
case. The value of the length should be introduced in the appropriate field.
dilatation - if this option is selected, the cable dilatation (or relative dilatation, if the option
Relative is selected) will be defined for the assembling case. The value of dilatation
should be introduced in the appropriate field. Dilatation is the difference between the
unloaded cable length and the distance between the relevant nodes. If dilatation assumes
positive value, the cable length is larger than the distance between nodes; if it assumes
negative values, the distance between nodes is larger than the cable length.
NOTE: Options Stress, Force, Length and Dilatation are mutually exclusive and they
are not obligatory (if the user does not specify any of the parameters, the
length of the loaded cable equals the distance between nodes).
Theory of cable structures is based on the following assumptions:
loads and other external effects are of quasi-static type and constant in time,
for cables no bending moments and shearing forces are considered,
cable elements work in the elastic range (Youngs modulus E = const),
any loads can be applied, except for the moment loads,
large displacements u, but small gradients
du
dx
are admissible,
cable section area F is constant (F=const),
unloaded cable length = l.
Equations Governing the Problem
Lets consider a small sag cable (i.e. cable for which the angle between the tangent in any
cable point and a straight line joining its ends is small), loaded with an arbitrary load in its
plane. Lets consider an infinitesimal element in this cable - one, which is described in the
initial stage (first, assembly stage) by the load q0, temperature T0 and tension H0; the length
of this element is equal to dso (Fig. 1a). Once the load is applied to the cable (second, final
stage with the load q, temperature T and tension H), the length of an elementary cable
section equals ds (Fig.1b). Both stages, together with loads in both planes (xy and xz) are
presented also on the Fig. 2.
structures.com
124
Fig.1a
Fig.1b
Assuming a small cable sag value and taking into account that the total cable force must be
tangent to the cable, one can assign the appropriate cable elongation as a function of static
values only. Once the integration along the total cable length is done, the known formula for a
cable with a small cable sag value will be obtained. The cable chord elongation value can
be derived from (1).
Fig. 2
where:
A, B - beginning and end cable node,
EF - cable tension rigidity (where: E - Youngs modulus, F - cable cross section
area),
- coefficient of thermal expansion,
l - initial cable length (for unloaded cable),
- distance change between supports,
structures.com
125
- initial, internal cable shortening/elongation (regulation),
T - change in temperature,
Q(x) - function of shear force as for beam with pinned supports (according to
indexes: respectively in the Y-axis and Z-axis directions and for the initial and
final stage) - schematic drawing below (Fig. 3a),
Fig. 3a
N(x) - axial force function caused by the static load for a beam with fixed supports (during
the initial or final stage) - schematic drawing below (Fig. 3b).
Fig. 3b
One can also notice that in the cable equation (1), as opposed to traditional solutions applied
to cable calculations, axial force can vary along the cables length (in denominators of both
integration functions in the equation (1) the following functional components exist: [H+N(x)]
2
and [H0+N0(x)]
2
). It allows for more accurate results.
Cables in Robot Millennium
Cable element theory in the Robot Millennium system is based on the general theory of
cables with small value of a cable sag. According to this theory, cable rigidity is an implicit
function of the following parameters: cable tension rigidity (E*F), cable tension, cable support
displacements, transverse loading in both directions (p
y
, p
z
).
Due to the non-linearity of cable element, its definition in the structure requires applying
iterative methods of structure analysis.
Cable implementation capabilities in ROBOT MILLENNIUM:
Cable elements can be used together with elements of the following structure types:
PLANE FRAME, SPACE FRAME, SHELL,
All standard structure analysis types are admissible: Linear (in fact, it is a non-linear
analysis but no other non-linear effects, except for the non-linearity of cable elements are
taken into account), Non-linear (with regard to stress-stiffening effect), P-Delta,
Incremental, Buckling, Dynamic, Harmonic, Seismic; NOTE: Dynamic analysis will be
treated as linear with regard to the current rigidity,
structures.com
126
Offsets are admissible,
Material is defined as for a bar (Youngs modulus E is the only requirement; in case of a
cable dead load definition, additionally the unit weight RO should be entered, and in case
of a thermal load - coefficient of thermal expansion LX),
GAMMA angles defined as for bars (it is substantial only for the load description).
Limitations of cable element implementation:
For cable elements, a release definition is impossible since bending and torsional rigidity
do not apply to such elements.
Cable loads
The following load types apply to cable elements:
nodal loads
dead loads
uniform loads (constant or variable)
initial shortening/elongation (additional loads applied during the assembly stage)
temperature load
concentrated forces along the elements length.
The following load types are not allowed in case of cable elements:
moment,
uniform moment.
SYNTAX (data entered by the user in the text file)
PROperties
(<element list>) CABles AX=<section area> (E=<Youngs modulus>)
(RO=<unit weight>) [STRess = <s> | FORce = <h> | LENgth = <l>
| [ DILatation = <d> (RELative) ]]
where:
STRess - normal stress (calculated in regard to the chord) to be achieved for
the assembling load case
FORce - tension force (calculated in regard to the chord) to be achieved for
the assembly load cases
LENgth - initial cable length
DILatation - difference between the initial cable length and distance between
the support nodes (if it is a positive value - the length is bigger than
the distance between the nodes, if negative - the length is smaller
than the distance between the nodes)
RELative DILatation - ratio of the difference between the initial cable length and distance
between the support nodes to the distance between the support
nodes (if it is a positive value - the length is bigger than the
distance between the nodes, if negative - the length is smaller than
the distance between the nodes).
Assembling Load Case
Robot Millennium system includes the structure assembly stage (the program recommends
it to be the first load case). Syntax for such load case is given below:
CASe
ASSembly
[load description]
structures.com
127
For this load case:
for selected bars, the initial tension forces are to be specified by the user by entering one
of the values presented below in the text file syntax (PROperties command):
STRess = s0 (initial cable stress) or
FORce = t0 (initial tension force in the cable),
the initial cable length can be described by specifying the LENgth = l
0
, if it is to be
different from the default value LONG = x x y y z z
B A B A B A

2 2 2
which
is equal to the distance between the nodes,
the initial cable shortening/elongation can be specified by using the DILatation (RELative)
command
if there is no RELative key-word, DILatation is expressed in absolute values
if there the RELative key-word exists, the initial elongation is expressed as a fraction, i.e.
final cable length equals: L = LONG (1 + DIL),
all defined loads are applied (e.g. dead load, added masses),
temperature TX definition for cables in the assembly stage is possible,
displacements calculated for such load case describe initial geometry for remaining cases
in the structure analysis.
During the analysis of successive structure load cases in the state of equilibrium, the
assembling case loads applied to the structure are taken into account. Displacements
assigned to this case are used as the basis for further analysis. Predefined tension forces are
changed (which means that after the assemblage the cable will be anchored).
Load Cases after Anchorage
After completing the structure analysis, results for cable elements are similar to those
obtained for bar elements; however, some differences remain. The differences are described
below:
no shearing forces and moments can be obtained for cable elements,
for cable elements either the simplified deformation (assigned as for the truss bar) or the
exact deformation (described by the differential equation of the sag line), can be obtained
additional results for cable elements (as a consequence of the assembly stage) appear:
in cables, for which tension is required (in the ROBOT Millennium syntax: STRess or
FORce), the regulation value [m] needed for the required tension is assigned,
in other cables, force essential for assemblage is assigned.
Such results are useful at designing the assembly stage. They are available from the
Results module by selecting the command: Stresses / Parameters / Results for cable
elements in the assembly stage.
axial force (tensile) is calculated from the formula:
N FX FY FZ
2 2 2
,
where:
N - force applied along the cable tangent,
FX, FY, FZ - N force components projected on directions of successive axes of
the local co-ordinates system.
structures.com
128
3.12 Other Structural Attributes
Within the current version of the system it is possible to define additional structural attributes
such as releases, offsets, compatible nodes, rigid links, brackets, tension/compression bars
and bar types. Structure bar connections are rigid, e.g. displacement and rotation
compatibility for all bars converging at a given node is ensured. Rigid links can be released, if
necessary. The Geometry/Releases option allows the user to release selected degrees of
freedom at a given structure node. The following types of compatible nodes may be defined in
the program: rigid, elastic, with damping, unilateral as well as non-linear.
A new algorithm has been implemented into the program that allows one to calculate a
structure containing releases (the DSC algorithm option in the Job preferences / Structure
analysis dialog box). For any bar, for which any type of analysis has been defined, with the
following releases:
regular
unilateral
elastic
elastic and unilateral
the following operations are carried out:
a new node is generated in the structure (during the structure model generation)
the input element with the release is modified in such a way that the new node takes the
place of the old one in the element (the old node remains in other structure elements)
between the old and the new node the program creates the so called DSC element
(Discontinuity) see the figure below.
The DSC element is a 2-node element where the nodal forces are generated according to the
following formula:
) (
) (
1 2 2
1 2 1
u u T Tk f
u u T Tk f

T T
T T
where Ndl i k
i
, 1 ]; [ k is the stiffness vector ascribed to particular degrees of freedom.
The stiffness matrix of the element has the following structure
T T
T T
T k T T k T
T k T T k T
K
) ( diag ) ( diag
) ( diag ) ( diag
where T is the matrix of transformation from the local base to the global one, inherited from
the bar element, while diag(k) is the diagonal matrix created from the vector k.
The introduction of the DSC element allows one to define elastic releases in a bar element,
which was impossible in the hitherto implemented approach.
structures.com
129
The user may obtain the same values of displacements in bar intersection points for bar
structures where bars intersect. This is done by means of the Compatible nodes option. One
has to define so many nodes with identical coordinates as there are bars. Each node belongs
to a different bar. The option is available from the menu: Geometry/Additional
attributes/Compatible nodes. The following types of compatible nodes may be defined in the
program: rigid, elastic, with damping as well as non-linear.
Rigid links is used to model perfectly rigid parts of elastic structures (definition of a rigid body
within a structure). All nodes in such a connection have identical rotations and displacements
due to the rigidity of the entire connection. Displacements and rotations defined for a rigid link
may be limited to certain selected degrees of freedom. For instance, linear displacements
may be blocked, while rotations are allowed. The first node is called the master node
(MASTER), while the remaining nodes are slave nodes (SLAVE). The option is available from
the menu: Geometry/Additional attributes/Rigid links.
For some structure elements an offset definition is needed, e.g. the eccentric (non-axial)
connection of the structure bars is to be defined. It can be done using the
Geometry/Additional Attributes/Offsets option.
An offset may be defined automatically for structure bars; then the offset is determined by a
shift of a bar axis to section extreme dimensions. The axis position is selected by clicking on
the appropriate selection button located in the section drawing; there appears a description of
a bar section position, i.e. Axis shift and characteristics of the section position describing the
value of the axis shift in the local system, e.g.: -Vpy, Vz (axis shift: top left corner).
The program provides the possibility to define elastic ground for structure members. The
Geometry/Additional Attributes/Elastic ground of bars option is used for it. Calculations for
bars with a defined type of elastic ground are performed on a base of the classic algorithm of
Winklers elastic ground, but uplift of the bar from the ground is not permissible.
NOTE: Elastic ground can be defined only for some structure types. The accessible
directions of elastic ground as for applied type of structure (degrees of
freedom permissible for a selected structure type):
2D FRAME KZ
2D FRAME KY, KZ, HX
GRILAGE KZ, HX
PLATE KZ, HX
SHELL KY, KZ, HX
2D TRUSS definition of Winklers ground
is impossible
3D TRUSS definition is impossible
In case of plate/shell structures there is also the possibility of defining elastic ground
coefficient, however, a value of such a coefficient is determined in the dialog box used for
thickness definition for panel or shell (see chapter 3.5).
It may be necessary to define node brackets in some nodes of a bar structure. The respective
option is available from menu by selecting: Geometry/Additional Attributes/Brackets. Node
brackets are used in a structure to improve section properties of bars in node zones. They
make it possible for these zones to carry larger moments and transversal forces, while
smaller sections are used along entire bars. Brackets may be applied only to I-sections. There
are two bracketing methods:
by means of plates of determined dimensions
by means of appropriately cut section of a given bar
To define a bracket one should determine bracketing method, bracket dimensions (absolute
or relative values) and bracket type (upper, lower, at both sides). A bracket defined in this
way is saved under the label chosen by the user. Later on, it may be used many times in a
structure. Brackets of a bar are introduced separately for each bar end.
Brackets are taken into account in successive calculation stages in the following way:
structures.com
130
during static calculations: the bracketed part of a bar is assumed to have a section of
variable inertia whose dimensions are the result of bracket definition
during connection design: bracket dimensions are automatically transferred to the
connections module when a steel connection is defined. The modification makes it
necessary to update the dimensions of the defined bracket or to define a new one.
during code calculations: certain assumptions, concerning substitute section parameters
of a bar treated as a whole, are made during code verification of bracketed bars. Moment
of inertia of a substitute virtual bar is calculated on the basis of the length and inertia of
bar components, i.e. of the bracketed and non-bracketed bar segments. Substitute virtual
properties are used to calculate general stability of a bar. Stress verification in particular
points of a bar takes into account real section properties
Design, i.e. finding optimum member sections, takes into account brackets as well.
The Geometrical Imperfections option (accessible from the menu by selecting the command:
Geometry/Additional Attributes/ Geometrical Imperfections) allows defining initial geometrical
imperfections. Geometrical imperfections may be assigned to single bars or to a group of co-
linear bars (so-called superbars). Imperfections do not cause initial forces, structure stresses;
if imperfections are considered, it results only in changing the structure geometry. While
applying this option the assumptions adopted should be taken into account:
imperfections may be assigned only to bar elements
imperfections are assigned at the middle point of a single bar (or a co-linear bar group)
imperfections cause change of geometry of a bar or a bar group, however, the change of
geometry is obtained by creating calculation elements mapping a deformed shape
all results of structure calculations are displayed on a deformed structure (i.e. the
structure with geometrical imperfections considered); nodal displacements are
determined in relation to the initial geometry defined by the user.
The programs also allows definition of non-linear hinges which may be used in the structure
pushover analysis. The structure pushover analysis is a static, non-linear analysis, in which
the magnitude of structure load is increased incrementally according to the adopted load
pattern. Increasing the value of load enables finding structure weak elements and structure
failure modes. The pushover analysis tries to estimate structure real strength. The option is
available from the menu by selecting the command: Geometry/Additional Attributes/Non-
linear Hinges. The Definition of non-linear hinge model dialog box contains options that
allow defining a pushover curve diagram and parameters. Three types of non-linear hinge
are available in this dialog box: force-displacement, moment-rotation and stress-strain.
The Robot program allows for definition of bars in the structure, which can perform only in
tension or compression, truss bars (the option is used to define truss type bars in frame
structures; such a definition does not result in assigning releases to a frame element,
however, it changes the type of a finite element from a beam element to a truss one) as well
as bars for which the influence of shear forces on structure deformations is considered. The
option is available from the menu by selecting the Geometry/Additional Attributes/Advanced
Bar Properties command. In the Advanced Properties dialog box, the user should define in
the appropriate edit field (by entering the bar numbers using the keyboard) or graphically on
the screen - the bars performing only in tension/compression, truss bars or bars for which the
influence of shear forces will be considered in the calculations of structure deformations.
Steel/timber bar types are assigned during the structure definition yet this is not necessary for
structural analysis; it is used during steel/timber member (columns, beams, etc.) design. Bar
type contains all parameters needed for steel/timber element design, e.g. buckling length,
lateral buckling type, etc. Structure attribute definition (mentioned above) is identical to the
method of assigning sections to structure bars.
structures.com
131
While working in the Robot program, a user defines many labels describing bar, panel or
solid parameters; these are different sets of parameters used to determine physical or
mechanical properties or applied to design structure elements. Examples of such labels
include: bar sections, slab thicknesses, parameter sets used for design of steel structure
elements or applied to calculate reinforcement in RC structure elements as well as definitions
of supports, hinges, etc.
The Label Manager option is available from the menu by selecting the Tools/Label Manager
command; it enables performing the following operations concerned with the access to labels:
saving labels from the current project to the database
reading labels from the database to the current project
viewing the contents of label definitions
transferring the labels defined from one program installation to another or from one
program work station to another.
However, it should be remembered that member code labels (i.e. steel or aluminum member
type, timber member type, concrete member type) are converted to the current design code
for steel, timber or RC (theoretical (required) reinforcement) structures; therefore, member
types defined for one code should not be used for other codes since it may result in loss of
definitions of certain specific parameters. All bars in a given project are designed according to
the currently selected code. As regards a label of plate and shell reinforcement, it is saved for
a determined code of RC structure design (theoretical (required) reinforcement). Thus labels
of panel reinforcement type may be applied in different projects independently of the design
code selected by default.
3.13 Other Structure Components
Structural axes are very useful during the structure definition. This option is available from:
the menu by selecting: Geometry / Structural Axes or
the Tools toolbar by selecting the icon .
Once the option is selected, the dialog box shown below will be displayed on the screen.
structures.com
132
Structural axes create an additional grid that allows
you to define various structural elements and
indicates selected structural components (their
selection is possible). The axis grid creates points
in which the cursor can be placed during graphic
definition of the structure.
Structure axes are vital while moving within a
structure model. Using them allows quick display of
the selected work plane (the View dialog box) and
selection of elements positioned in structure axes.
Axis selection may be carried out using
descriptions positioned on either ends of a
structure axis. Clicking on the description of the
selected axis results in selection of all structure
elements situated on this axis.
A structural axis grid may be defined as:
rectangular grid in the Cartesian system (in the
ZX plane for 2D structures, in the XY plane for
3D structures with Z direction as the structure
height). The same axis grid as the one defined
in the XY plane is created for each level during
definition of multiple structure levels.
grid in the cylindrical or polar coordinate
system.
arbitrary axes (straight lines, rays and
segments).
For structural axes defined in the Cartesian system there are two available options of defining
coordinates (after pressing the Advanced Parameters button, the dialog box expands
showing additional options):
relative (the Axes relative to the point option is switched on) - if this option is selected,
then in the dialog box the Insertion point edit field becomes available; position of the
created structural axes is defined with respect to coordinates of the insertion point
absolute (the Axes relative to the point option is switched off) - if this option is selected,
then structural axes will be created based on the real values of coordinates of individual
structural axes (in the global coordinate system).
In the case of a grid created in the cylindrical or polar coordinate system, only definition in the
relative mode, i.e. with respect to the insertion point, is available (the option is accessible after
pressing the Advanced Parameters button).
Structure axes in the Cartesian system can be rotated as a whole. To do this, the user should
switch on the Axis / rotation angle option, choose the axis about which the axes will be rotated
and define a rotation angle. Axes defined in the relative manner are rotated with respect to
the insertion point.
The top part of the dialog box contains the list of defined sets of structural axes. A definition of
a new set starts with providing a set name in the Name field (the default set name is Structure
axis); a name of a new set is added to the list of defined structural axes.
There is the Axis manager button provided in the bottom part of the dialog box. Its pressing
opens the Manager of structural axes dialog box. This part of the dialog box contains, as
well, the New button pressing which enables defining a name of a new structural axis set.
structures.com
133
The options located in this dialog box enable management of structural axis sets. The top part
of the dialog box comprises defined structural axis grids - they are identified by names
ascribed to them while creating structural axes.
After switching on a set of structural axes ( symbol appears) and pressing the OK button,
structural axes of this set will be visible on the screen. When an axis set is switched off (
symbol disappears) and the OK button is pressed, then the axes of this set are not displayed
on the screen.
There are two buttons provided in the bottom part of the dialog box:
Delete - pressing this button deletes a selected set of structural axes
Delete all - pressing this button deletes all structural axis sets available in the dialog box
Definition is limited to the rectangular grid (for 2D structures - on the ZX plane and for 3D
structures - on the XY plane, where the Z direction represents the structure height). In the
case of a level definition, an axis grid as defined on the XY plane is created at each level.
The Structural Axis dialog box contains three tabs that define:
vertical axes (selection of X coordinates on the YZ plane),
levels (selection of Z coordinates on the XY plane),
horizontal axes (selection of Y-axes on the XZ plane).
In the case of 2D structures (2D frame and 2D truss) only the first two tabs are active; for
grillages only the first and the third one are available. For cylindrical grids the following tabs
are available in the top part of the dialog box: Radial, Angle and Z (the last tab is accessible
only for 3D structures).
Successive tabs are identical and the definition of levels and horizontal and vertical axes are
the same. The following edit fields must be filled: POSITION, NUMBER OF REPETITIONS
and DISTANCE. To generate the defined axes/levels click on the Insert button. Created
axes/levels will be listed in two columns containing the name of the axis/level (label) and
location in the global coordinates system.
When defining axes using the Arbitrary option the axis type: segment, semi-line or line should
be chosen. Next, two points defining an axis should be determined. If the points are defined
graphically by means of the mouse, then the axis is added automatically to the list of
generated structure axes. If the user enters point coordinates into the dialog box, then for the
axis to be added to the axis list the Insert button should be pressed.
If the Arbitrary option is switched on, then the additional button: Create axes from selected
bars/lines appears in the top part of the dialog box. Pressing this button causes generation
of structure axes based on the bars and lines (edges of generated 2D or 3D objects) selected
in a structure model.
structures.com
134
The name of the axis is to be selected from the Numbering field; for vertical axes the default
names are A,B,C..., for levels/horizontal axes: 1,2,3 ... .
The user has also a possibility to define his own levels/axes. To achieve this, one should:
determine the axis/level position
select Define option in the Numbering field
provide the description of the defined axis/level in the field located in the lower part of the
dialog box (it becomes active after selecting the Define option); for example once the
Level name is entered, the program will generate the following names: Level1, Level2,
etc.
press the Insert button.
Additionally, a structure axis may be given a name (it concerns, in particular, names of axes
in the direction of Z axis structure story) that corresponds to a position of structure axis,
e.g.: 4.0, 6.5, etc. To do this, the user should choose the Value option from the Numbering
list. The %v variable which is used to generate an axis name depending on the axis position
may also be applied to define users own name of levels.
To remove a single axis/level, select the appropriate item (axis/level) from the list and click on
the Delete button. To remove all axes/levels use the Delete All button. In order to single out
any of the axes/levels on the graphical screen, one should highlight the selected axis/level
included in the list and click the Single out button. There will appear the X symbol in the
third column corresponding to this axis, which means that the axis has been singled out
(presented as a thick line).
Axes from the list are always displayed in an alphabetical order, according to the ascending
order of the coordinates defining the axes.
A position of Cartesian structural axis can be modified jointly with all the structure parts
situated on it. To start modification of a structure axis, the user should locate the cursor on its
description (the description becomes highlighted) and after pressing the right mouse button,
select the Object Properties option from the context menu.
The program also enables selection of structure elements by means of a defined structure
axis grid. The option is available from the menu by choosing the command: Edit / Select
Special / Structure Axes.
In some cases, it proves very useful to apply the Numbering option that allows the user to
define numbering of nodes, bars, panels and objects. This option is made available by:
selecting the Geometry / Numbering command from the menu,
pressing the Numbering icon.
The options found in the dialog box allow the user to change
the numbering of nodes, bars, panels and objects defined in
a structure. The dialog box is divided into two definite parts:
Bars/Panels/Objects and Nodes. The same options are found
in both fields (Object Number, Step, Selection).
To change the numbering of objects in a structure:
define the beginning node number in the Node number
field or bar/panel/object number in the Object number
filed
define numbering step
select node/bar/panels/objects, whose numbering will
change
press OK button.
If no numbering conflict occurs (i.e. the new number is not
assigned to another node, bar, panel, object), the numbers of
the selected nodes/bars/panels/objects will be changed.
structures.com
135
NOTE: If, for instance, the numbering of bars 8, 11, 15, 20 is to be changed and the
following numbering change parameters are entered: Object Number equals
11 and Step equals 2, then, no numbering conflict will occur, as long as bars
13 and 17 have not yet defined in the structure.
There is also a possibility of performing the operation of geometrical renumbering. The
parameters of this type of renumbering are available after pressing the Parameters button in
the above-presented dialog box (the button becomes active no sooner than the Geometrical
renumbering option is selected).
In some cases, dimension lines are necessary for displaying the structural drawing. Such
lines can be added to the structure by using the Tools / Dimension Lines option. Once this
option is selected, a dialog box for definition of parameters for created dimension lines (line
type, its position, beginning and end, etc.) will be displayed on the screen.
3.13.1 Edit Operations
Robot program is equipped with many useful edit tools which facilitate users work during
defining and/or modifying a designed structure. These options, among others, include:
rotation, translation, horizontal and vertical symmetry (mirror), planar symmetry, axial
symmetry and scale. All the enumerated edit options are available from the menu (Edit/Edit
submenu) or icons in the Edit submenu .
There is also the Complex edition option available within the program, which is used to
combine edit operations (translation, rotation and scaling) for the previously selected structure
nodes/objects. The option is available from the menu, by selecting the Edit / Complex edit
command. To define complex edition of selected nodes or objects in a structure, one should
first decide which operations will be performed during complex edition; this option may consist
of the following edit options: translation, rotation and scaling. Complex edition may consist of
two or three mentioned operations set in an arbitrary order.
The following options may be found very useful during structure model definition: Divide,
Extend and Trim.
The Divide option is used to divide bars or object edges into smaller bars or edge segments.
The option is available:
by choosing the Edit/Divide command from the menu,
by pressing the Division icon on the toolbar.
Extend is based on pulling a selected bar or object to the bars or objects that define the
extend edges (extend limits). The option is available from:
from the menu by selecting the Edit/Extend command
from the toolbar by pressing icon.
Trim is based on indicating in a selected bar/object this part which is to be trimmed. The trim
part is located among bars/objects defining the trim edges. Bars or objects (arcs, circles,
polylines, etc.) can be the trim edges. The option is available:
from the menu by selecting the Edit/Trim command
from the toolbar by pressing the icon.
structures.com
136
3.14 Typical Structures
This option allows for selection/definition of typical structures (structure elements). It is
available in the Design 3D module, both from the text menu and icons:
after selecting the Library Structure icon
after selecting File / Library command from the menu.
The dialog box shown below will be displayed on the screen.
Library structures are organized in certain structure bases. To select a structure, first the
appropriate base has to be found. The dialog box presented above consists of a few icons
corresponding to the structure types. To select any structure type double-click on the
appropriate icon with the left mouse button. The following structure bases are available in the
current version of Robot:
typical structure library - frames, trusses, beams
typical structures additional base
plates and shells.
A couple of typical bar structures, most
frequently used in structure design are
available within the program:
continuous beam,
multi-story frame,
multi-bay frame,
grillage,
different types of trusses.
These bar structures can be configured in an arbitrary way within the template. The following
values are to be defined:
- length,
- height/width,
- number of fields/divisions/spans,
- roof girder slope angles (in frames)
The typical plate and shell structures have been defined in the plate and shell base:
rectangular plate building wall with openings
rectangular plate with
a rectangular opening
half of a circular plate
structures.com
137
rectangular plate with
a circular opening
half of a circular plate with
a circular opening
rectangular plate with ribs
half of a circular plate with
circular plate
shell structure - a rectangular
container
circular plate with
a circular opening
shell structure - cylindrical
container
circular plate with
shell structure - container in
the shape of a cut cone.
In addition, for surface finite elements, the following parameters are to be defined:
- thickness,
- mesh type,
- reinforcement type and geometry,
- etc.
A structure defined using the library structure base can be used as a component of a bigger
structure (appropriate options allows for inserting it into the already existing structure in a
convenient and accurate way) or be a separate structure.
3.15 Phase Structures
Among the most interesting options offered by the Robot system one can point to the
possibility of analyzing phase structures (structures designed in several technological stages).
The option is available from the menu. The following commands are used to activate it:
Geometry/Phases/Phase Selection
Geometry/Phases/Collecting Phases.
The program performs structure calculations separately for each phase. Results are obtained
for each structure design phase, for each one is treated as a separate structure. The program
performs automatically the analysis of a phase structure phase by phase. The user decides
which of the phases is active during the presentation of results, i.e. for which phase to present
calculation results.
The results of phase structure calculations may be viewed for each phase separately, or else,
they may be collected (combined into one phase) to compare results and create an envelope
of the results for the entire structure with the use of the results obtained for the selected
phases.
Once the Phase selection option is chosen, there appears on screen the dialog box shown
below (the dialog box already contains definitions of four phases). The following options are
available in the dialog box:
the Activation/Edition range field - the field where the names of defined structure phases
will be displayed; beside each of the defined phases, one finds its number and a field
allowing one to switch on/off the bar/element transfer to other phases (if the option is on
for a given phase - there appears the symbol - the defined bars/elements in the
currently selected phase will also be included in the phase for which the transfer has been
switched on; if the option is off, the bars/elements defined for the currently selected phase
will not be included in the given phase)
structures.com
138
the All windows option - if the option is off, then,
activation of a phase of the structure will refer to all
the viewers open in the program; if the option is off,
the phase activation will refer to the active viewer.
the New Phase field - the field where one can define
a new structure phase; one should provide a name of
the phase (the program automatically provides phase
numbers; the defined phase will receive the next
number) and press the Define button.
three buttons:
Activate - pressing it results in activation of the
selected (highlighted on the list) phase; one may also
do it by double-clicking the selected structure phase
Modify - pressing it allows one to change the name
of the selected (highlighted on the list) structure
phase
Delete - pressing it results in deleting the selected
(highlighted on the list) structure phase
NOTE: The title bar of the Robot program displays the name of the phase that is
currently active.
The Phase collection option is used to define results for a phase structure on the basis of a
selected final phase (base phase) which should typically be a complete structure. The option
results in gluing (combining the results) indicated phases and allows one to compare single
phases graphically. One should determine the Base phase (it should be an entire structure) in
the upper part. To define Added phases, one should select appropriate phase (the phase
name is marked by the symbol). Each phase may be easily added to or deleted from the
current list that provides the basis for phase collection.
NOTE: During definitions of successive phases of a structure, one should pay
attention to the numbering of structure bars; bars should have the same
numbers in the successive phases for the operation of phase collecting to
make any sense (in order to combine the results obtained for particular
phases for the same bars/elements).
The results for a structure collected in this way will be presented when one selects the
Glued structure / 1
st
phase before collecting phase in the Phase dialog box.
3.16 Surface Finite Elements
There are two types of surface finite elements available in Robot: plates and shells. They are
defined by following steps:
material definition
contour definition
panel and hole definition
support definition
load definition.
Material, supports, and loading definition are similar to definitions for bar and are described in
sections 3.3 through 3.5 respectively.
structures.com
139
A contour can be defined using following methods:
Contour definition
line,
polyline/contour,
arc, or
pre-defined geometric shapes such as a circle or rectangle.
When polyline/contour method is selected, it is necessary to come to a closure when the
corners of the panel are being defined. This is done by specifying the first corner twice: as a
first and last point of the shape.
Those points can be defined in two different way either using mouse or by specifying
coordinates of each of them in numerical way with the keyboard. A dialog box, called Lines,
serves the purpose.
Panel and opening definition
A panel can be defined either by selecting a single internal point belonging to the panel or by
selecting all the objects defining the shape of the panel (lines, arcs other polylines, etc.).
Panel properties can be assigned in the same dialog box or from text menu. Two properties
should be specified: the material from which the panel is made of and its thickness.
Optionally, the reinforcement, for example in case of reinforced concrete, can be also defined.
When a panel is modeled with an opening, this opening has to be defined first. Therefore, the
sequence of steps would be following:
Select opening for in the contour type field
Create a contour by choosing, for example, an internal point within a figure
defining the hole
Select panel in the contour type field
Create the panel by, for example, choosing an internal point within the panel figure
but outside of the hole.
structures.com
140
3.17 Coordinate System
3.17.1 Global Systems
Each structure's geometry is initially defined by node locations. Elements then can be defined
by indicating an appropriate beginning and end node list.
Node locations are determined by their coordinates in a right-handed Cartesian system, which
is the GLOBAL SYSTEM in which all calculations are made.
For plane structures, the XZ-plane is assumed for coordinate data (structure definition).
However, the graphical display is actually a 3D picture that can be rotated about any axis to
create different views.
3.17.2 Local Coordinate Systems
Each element defined in a structure has its own LOCAL COORDINATE SYSTEM, in which
section properties (moments of inertia, extreme fiber locations), and result forces are defined.
Furthermore, load cases and additional attributes such as end releases and offsets can be
defined using element local coordinate systems.
3.17.2.1 Bar Elements
The local system is a right-hand Cartesian type and has the following axis orientation for bar
elements:
the local x axis is always located along the element's longitudinal axis, having its
direction sense from the beginning node to the end node. The origin is "fixed" at the
beginning node, as shown in Fig. 4.
the local y and z axes are located according to the right-hand rule. These axes
typically represent the strong and the weak axes of bending respectively, depending on
the section profile. The default orientation of 3-D bar elements is shown in figure below.
The local system (y and z axes) together with the section profile can be rotated around the
element axis x, by setting the GAMMA angle.
NOTE: The default orientation of 2-D bar elements is treated as a special case. The
local z and y axes represent the strong and weak axes for plane frame
and plane truss elements (models defined in the XY-plane). Initially (for
GAMMA=0) the profile is set in a position which safeguards the structure with
the greater moment of inertia reacting the bending effects from allowed loads
(i.e. the Z-axis, being perpendicular to the XY-plane of the structure, is
assumed to be the strong axis of bending by default).
structures.com
141
X
Y
Z
y
z
x
A
B
3D
X
Y
x
y
A
B
2D
A - beginning node
B - end node
Fig. 4
Z
Y
Z
Y
Z
Y Y
Z
Y
Z
Z
Y
Z
Y
Z
Y
Z
Y
Z
Y
Z
Y
Z
Y
Y
Z
Y
Z
TF2
HT
TF1
Y
Z Z
Y
TH
D
HT
BF
TH
HT
TH
BF
TF
TW
TF
HT
BF
B1
B2
TW
3.17.2.2 Finite Elements
The local coordinate system for a finite element is not required as every node has its own
'local' system. However, it is required that the direction of the vector normal to the element
surface is known to allow proper definition of normal pressures.
- the normal vector is oriented perpendicular to the element surface, according to the
right-hand rule, when considering a sequence from the first to the second, and third
nodes of element. The local system and the normal vector are shown for 6 and 8-
noded finite elements in figure.
structures.com
142
e
2
e
1
1
e
2
e
y
x
z
1
2
3
4
5
6
1
2
3
4
5
7
8
6
z
y
x
z
y
x
z
y
x
z
y
x
z
y
x
z
y
x
n n
3.17.3 Global-local Systems Relations
In the definition of the relation between the global coordinate system and the local coordinate
systems, positive angles are defined by the right-hand rule. Robot Millennium uses a
common convention for the angle transformation, in which the angles ALPHA, BETA, and
GAMMA signify positive rotations around the axes of the global coordinate system. For
composite transformations the sequence of rotations is important. Thus, if the non-zero
angles are submitted, first the rotation around Z-axis (ALPHA) is applied, next the rotation
around Y-axis (BETA), and finally the rotation around X-axis (GAMMA). The first two angles
of rotation define exactly the local x-axis for an element. A profile's specific position (setting
the local y and z axes), is determined by the GAMMA angle.
X=X0
Y
Z
X1
Y1
Y2
Z2
y
z
X2=x
Z1
Y0
Z0
To understand the GAMMA angle
definition, consider a 3D-global to
local Cartesian system
transformation. First the rotation
around the Z axis (ALPHA angle) sets
the new X1-axis in line with the
projection of the element on the
horizontal plane. The next rotation lifts
the horizontally situated X-axis in its
final position. Finally the GAMMA
angle fixes the profile with its local
axes y, z in its final configuration. This
sequence of rotations is shown in the
figure to the left.
structures.com
143
If the element is situated
vertically (local x-axis parallel
to the global Z-axis) its
projection on the horizontal
global XY plane reduces to a
point. In such a situation the
ALPHA angle is assumed to
be zero, and only the BETA
angle rotation is applied to fit
the X-axis to the element axis.
The examples of different
GAMMA angles are given in
the figure to the right.
X
Y
Z
GAMMA=0
GAMMA=90
GAMMA=0
GAMMA=180
GAMMA=-15
GAMMA=-90
GAMMA=90
GAMMA=30
3.18 Sign Conventions
In general, the direction of positive force and displacement vectors is the same as the positive
axis direction. For the positive direction of angles, rotations, and moments in the external
coordinate system (global or local) the right-hand rule is used. This convention defines the
signs of external forces, nodal forces, displacements, and rotations. These are items used
during structure definition, analysis, and results review.
However, for internal forces that act within the element, a different this sign convention is
used. There are different definitions for the internal sign convention used for bar elements and
that used for finite elements.
3.18.1 Bar Elements
Positive directions of forces for the above described sign convention are presented
schematically in the drawing below. In the program, the sign convention for bar elements is
based on the convention used for sectional forces. According to this convention, sectional
forces are of the same sign, if they produce the same result at the end of a bar as the result
produced by positive nodal forces (the forces whose direction agrees with the direction of the
axes of the local coordinate system) applied at the beginning node of the bar. Therefore,
compression forces are positive, while the tension forces are negative. Positive bending
moments MY produce tension in the beam fibers that are located on the negative side of the
local coordinate axis z. Positive bending moments MZ produce tension in the beam fibers
that are located on the positive side of the local coordinate axis y.
The positive directions of force vectors in the above-described convention are presented
schematically in the figure below.
structures.com
144
3.18.2 Finite Elements
Robot does not display nodal forces for finite elements when printing results. Only internal
forces and stresses are displayed. They are defined in relation to the local normal and
tangential directions of the cross section. If n signifies the normal vector to the element
surface, s the tangent to the element surface, and z the outward normal to the element
surface, then these three vectors (n, s, z) will create a right-handed Cartesian system, the
positive forces, moments, and stresses acting on the given cross section should conform to
the appropriate directions of the n, s, z vectors.
Fs
Fn
Mn
Mns
Ms
Fz

s
n
n
s
z
This definition is displayed with positive
internal forces shown.
structures.com
145
3.18.3 Volumetric Finite Elements
3.19 Section Definition
The function of the Sections module is to allow the user to graphically define the geometry of
a cross-section, calculate the geometrical and cross-sectional properties, and store the
definition and properties to disk. These properties can be used in the analysis as well as the
design functions of Robot Millennium. The defined cross-sections can be:
solid (or thick-walled),
with or without holes,
homogeneous (one material) or composite (different material properties for different areas
of the cross-section),
with straight or curved contour segments,
taken from the catalog of profiles in the Sections Tables module, then modified, and
added to another contour.
The Sections module has two general options to define the cross-section and to calculate its
properties. These are the solid cross-section and the thin-walled cross-section options. Each
option operates slightly differently. The choice determines the graphical representation and
the calculation process used. The thin-walled cross-section option is not implemented in the
current version of Robot Millennium.
For solid or thick-walled cross-sections, the following properties are calculated and presented
in graphical and numerical form:
values of the true area (A
x
) and weighted area (A
x
),
structures.com
146
section circumference
position of the flexural centroid (Y Z
0 0
, ) or (Y Z
0 0

, ) in the user's coordinate system,
principal angle (Alpha) or (Alpha*),
inertias (weighted inertias): about the user axes ( , , ) I I I
Y Z YZ
or ( , , ) I I I
Y Z YZ

; or about
the central axes parallel to the user's coordinate system; or about an
inertia radiuses
shear capacity coefficients
first moments of area (in any coordinate system)
arbitrary central system and central principal axes ( , ) I I
y z
or ( I I
y z

, ),
torsion constant ( ) I
x

distance to the extreme fibers (V V V V
y py z pz
, , , )
,

shear reduced area (A A
y z
, ) for calculating the maximum shear stress in a beam.
Generation of the section requires following steps:
1. Define the geometry of the section graphically.
2. Define the physical material properties.
3. Perform the calculations.
4. Save the new calculation to the user section database.
The user can also generate the calculation note and verify it by performing a Stress analysis
with unit force loads applied to the section.
To define the geometry of the section, the user can use a combination of polygons, arcs,
rectangles or circles. These can be modified next by translating, rotating, or mirroring, as well
as by modifying any of their edges or vertexes. The final shape has to consist of contiguous
shapes and has to be a closed two dimensional figure.
Often during the edition of the section user wants to erase the overlapping portions of two
separate figures and unify it into a single section. This can be easily achieved with a
Standardization tool. The user has to click on the icon, then select all separate figures, and
finally click on the icon again to execute the "standardization".
Holes can be created simply by forming a figure within an existing section, using the tools
mentioned above.
The material of a section can be set by choosing Contour / Properties option and then by
choosing the material type in the material field as shown below.
structures.com
147
Here, the user can also modify the location of each
vertex of the section.
The user can also import a predefined section from
the database, like AISC for steel sections. This
option is available from the text menu under File /
Import from Databases.
The results can be achieved in two ways: by
choosing option from the text menu, Results /
Geometric Properties / Results or by clicking on the
Results icon. The results dialog box allows to view
all the characteristics of the created section and to
generate a calculation note which can be added to
the final project report in the printout composition.
The function of the Results option is to perform the calculation of the geometrical and
weighted (transformed) characteristics of the cross-section. For composite cross sections, the
weighted properties are marked with the sign "*". They are defined by the relation:
A
A
i i
i
b
where:
i - indicates the i - material of the composite section,
b - indicates the base material of the composite section.
The program output presents the
general results (the values of
quantities independent of the system
of coordinates) and results dependent
on the system and the kind of
coordinates (arbitrary, central,
principal).
Now the user can verify the section
properties by applying a unit force (as
well as unit moments in each direction
along and about the main axes). This
option, however, is only available for
the sections made out of the same
material.
structures.com
148
In the case of a thin-walled section the following values are presented in the graphical or
numerical form:
the section area (Ax)
positioning of the center of gravity (
Y Z
0 0
,
) in the global coordinate system
main angle (Alpha) - the angle of inclination of the first main axis in relation to the positive
direction of the axis Y of the principal coordinate system.
moments of inertia and deviation determined by the user in relation to the users global
coordinate system axes (I I I
Y Z YZ
, , ) as well as to the main, central axes ( Iy, Ix )
moment of inertia for torsion
distance of the extreme section fibers to the main and central axes
location of the bending center ( Y
c
, Z
c
) in the global coordinate system
weight per bar length unit (WU).
The following method is used in calculating characteristic properties of thin-walled section
geometry: the cross-section is assumed to be reduced to the central section line consisting of
points with ascribed mass m(s)= (s) (s)=1*(s), where (s) refers to the thickness of the
section wall, while s is a partial coordinate on the central line. A thin-walled section is treated
as a one-dimensional figure and divided into an arbitrary yet finite number of segments and/or
arcs.
The Results/Geometric properties/Calculation note option (available once calculations for
geometrical properties of a section are performed) activated a text editor, where one finds
data concerning the profile and the results of calculations made for geometrical properties.
structures.com
149
SELECTED REFERENCES
General
K.J. BATHE, E. WILSON, Numerical Methods in Finite Element Analysis, Prentice Hall,
New Jersey 1976
J.L. BATOZ, G. DHATT, Modelisation des structures par elements finis, Hermes, Paris 1990.
M.A. CRIESFIELD, Solution procedures for non-linear structural problems, 1984
T. GALEA, H. GACHON, Modele d'analyse nonlineaire des structures a barres methodes
d'approche du seuil de bifurcation, Construction Metallique, 2-1978
E. HINTON, D.R.J. OWEN, An Introduction to Finite Element Computations, Pineridge Press,
Swansea 1984
E. HINTON, D.R.J. OWEN, C. TAYLOR (Eds), Recent Advances in Non-linear Computational
Mechanics, Pineridge Press, Swansea 1982
T.J.R. HUGHES, The Finite Element Method - Linear Static and Dynamic Finite Element
Analysis, Prentice-Hall, New Jersey 07632
T.J.R. HUGHES, F. BREZZI, On Drilling Degrees of Freedom, Comp. Meth. In Appl. Mech.
and Eng. 72(1989), 105-121
W. KANOK-NUKUICHAI, Mathematical modelling of cable-stayed bridge, Structural
Engineering International, 2/92
J.S. PRZEMIENIECKI, Theory of Matrix Structural Analysis, McGraw-Hill, New York 1968
S.P. TIMOSHENKO, J.M. GERE, Theory of Elastic Stability, McGraw-Hill, New York, 1961
S.P. TIMOSHENKO, WOYNOWSKI-KRUGER, Theory of Plates and Shells, McGraw-Hill,
New York 1966
K. WASHIZU, Variational Methods in Elasticity and Plasticity, Pergamon Press, 1976
O.C. ZIENKIEWICZ, Finite Element Method, McGraw-Hill, London 1977.
Selected References Describing Methods of Finite Element Mesh Generation
AKHRAS, DHATT, "An automatic node recaballing schema for minimizing a matrix or network
bandwidth, IJNME'76, vol. 10, 787-97
AKIN, Contouring on isoparametric surfaces, IJNME'77, v.11, 893-7
AMEZWA el. el. 3, A general algorithm for the bidimensional Finite Element meshing, II
Congress Mundial Vasco, ed. ALEA t.I, 1988 p. 540-52
BARNHILL, BIRKOFF, GORDON, Smooth Interpolation in Triangles, J. of Approx. Theory
8,114-28, (1973)
BOWYER, Computing Divicklet tesselations, The Computer Journal, v.24, 2, 1981
BYKAT, Automatic generation of triangular grid: I - subdivision of a general polygon into
convex subregions, II - triangulation of convex polygons, IJNME'76, 1329-42
BYKAT, A note on an element ordering schema, IJNME '76, 194-8
BYKAT, Design of a recursive shape controlling mesh generator, 1375-90, vol. 19, 1983
CAVENDISH, GORDON, HALL, Substructured macro elements based on locally blended
interpolation, IJNME'83, v11, IJNME'77, 1405-21
COLLINS, A modified prefrontal routine, IJNME '77, 765-6
DUROCHER, GASPER, A Vertical Two-Dimensional Mesh Generator with Automatic
Bandwidth Reduction, Computer & Structures, 1979, v. 10, p. 561-75
EISENBERG, MALVERN, On FE Integration in Natural Co-ordinates, IJNME'73, v.7, N 4,
574-5
FREY, HALL, PORSCHING, An application of computer graphics to 3D FE, C&3 1979, v. 10,
149-54
GOLDEN, Geometric structural modelling: a promising basis for FE analysis, Computers &
Structures 1979, v. 10, 347-50
GORDON, Spline blended surface interpolation through curve networks, Journal of Math &
Mech. (JMM) '69, 931-52
GORDON, HALL, Transfinite Element Methods: Blending Function Interpolation over
Arbitrarily Curved Element Domains, Numer. Math, 21, 109-29
HABER, SHEPHARD, ABEL, GALLAGHER, GREENBERG, A General 2-D Graphical FE
Preprocessor Utilizing Discrete Transfinite Mappings, IJNME, 1981, vol. 17, 1015-44
HALL, LUCZAK, SERDY, Numerical Solution of Steady State Heat Flow Problems over
Curved Domains, ACM - TrMS'76 (Transaction of Mathematical Software), vol. 2, Nr 3, Sept.
76, p. 257-74
structures.com
150
HALL, Transfinite Interpolation and Applications to Engineering Problems, Theory of
Approximation with Applications, ed. LAW, SAHNEJ, 308-31
HENSHELL, Differences Between Isoparametric Assumptions and True Circles, IJNME,
1977, p. 1193-6
JOE, SIMPSON, Triangular meshes for regions of complicated shape, IJNME'86, v. 23, 751-
78
JOHNSON, Meshing - Beauty is only skin deep, BENCHmark, June 1993, p. 34-37
KLEINSTREUER, HOLDEMAN, A triangular FE mesh generator for fluid dynamic systems of
arbitrary geometry, IJNME'80, v. 15, 1325-34
LEE, LO, An Automatic Adaptive Refinement, FE Procedure for 2D Elastostatic Analysis,
IJNME 1992, vol. 35, P. 1967-89
LEWIS, ROBINSON, Triangulation of Planar Regions with Applications, The Computer
Journal 1987, vol. 21, n4, p. 324-32
LIU, CHEN, A 2D Mesh generator for variable order triangular and rectangular elements,
Computer & Structures, v. 29, Nr 6, 1033-53, 1998
LO, Generation of High-Quality Gradation Finite Element Mesh, Engineering Fracture
Mechanics, 1992,vol. 41, N 2, pp. 191 - 202
LO, Visualisation of 3D solid FE mesh by the method of sectioning, C&S, 1990, vol. 35, N 1,
pp. 63-8
LO, Volume discretization into tetrahedra - I. Verification and orientation of B surfaces, C&S,
1991, vol. 39, N 5, pp. 493-500
LO, Volume discretization into tetrahedra - II. 3D triangulation by advancing front approach,
Computers & Structures, 1991, vol. 39, N 5, pp. 501-11
LO, CHEUNG, LEUNG, An algorithm to display 3D objects, C&S'82, v. 15, No. 6, 637-83
McNEICE, MARCAL, Optimization of FE Grids based on Minimum Potential Energy,
Transaction of ASME (Journal of Eng. for Indus.), February 1973, p. 186-90
MEEK, BEER, Contour Plotting of Data Using Isoparametric Element Representation, IJNME,
1975, vol. 10, N4, P; 954-7
MOSCARDINI, LEWIS, CROSS, AGHTOM - Automatic Generation of Triangular and Higher
Order Meshes, IJNME, 1983, vol. 19, pp. 1331-53
OBERKAMPF, Domain Mappings for the Numerical Solution of Partial Differential Equations,
IJNME, 1976, vol. 10, p. 211-23
PARK, WASHAM, Drag method as a FE mesh generation scheme, Computers & Structures,
1979, vol. 10, p. 343-6
PEANO, PASINI, RICCIONI, SARDELLA, Adaptive Approximations in FE Structural Analysis,
C&S, 1979, vol. 10, p. 333-42
PERUCCHIO, INGRAFFEA, ABEL, Interactive Computer Graphic Preprocessing for 3D FE
Analysis, IJNME, 1982, vol. 18, p. 909-26
PERUCCHIO, SAXENA, Automatic Mesh Generation from Solid Models Based on Recursive
Spatial Decompositions, IJNME, 1989, vol. 28, pp. 2469-501
PIEGL, Hermite and Coons like interpolants using rational Bezier approximation form with
infinite control points, CAD, 1988, p. 2-10
PINA, An Algorithme for Frontwidth Reduction, IJNME, 1981, vol. 17, p. 1539-46
PISSANETZKY, KUBIK, An Automatic 3D FE Mesh Generator, IJNME, 1981, vol. 17, p. 255-
69
POURAZADY, RADHAKRISHAN, Optimization of a triangular mesh, Comp. & Structures,
1991, v. 40, Nr 3, pp. 795-804
PREISS, Checking the Topological consistency of a FE mesh, IJNME 1979, v. 14, 1805-12
RHEINBOLDT, Adaptative Mesh Refinement Processes for FE solutions, IJNME, v. 17, 649-
62, 1981
RUDD, Impacting the design process using solid modelling and automated FE mesh
generation, CAD'88, p. 212-60
SADEK, A Scheme for the Automatic Generation of Triangular Finite Elements, IJNME,
19780 vol. 15, p. 1813-22
SCHROEDER, SHEPHARD, Geometry Based Fully Automatic Mesh Generation and the
Delaunay Triangulation, IJNME, 1988, vol. 26, p. 2503-15
SIMPSON, Automatic Local Refinement for Irregular Rectangular Meshes, IJNM, 1979, v. 14,
1665-78
SLOAN, A Fast Stiffness Formulation for FE Analysis of 2D Solids, IJNME 1981, v. 17, p.
1313-23
structures.com
151
STELZER, WELZEL, Plotting of Contours in a Natural Way, IJNME, 1987, v. 24, 1757-69
STRICKLIN, HO, RICHARDSON, HAISLER, On Isoparametric vs. Linear Strain Triangular
Elements, IJNME, 1977, v. 11, 1041-43
WU, Techniques to Avoid Duplicate Nodes and Relax Restrictions on Superelement
Numbering in a Mesh Generator, Computers & Structures, 1982, vol. 15, n 4, p. 419-22
WU, ABEL, GREENBERG, An Interactive Computer Graphics Approach of Surface
Representation, Communication of the ACM, October 1977, vol. 20, n 10, p. 703
ZIENKIEWICZ, ZHU, Error estimates and Adaptivity - The Essential Ingredients of
Engineering FEM Analysis, BENCHmark, July 1989, p. 9-15
SPATIAL TECHNOLOGY, ACIS Geometric Modeler - Technical Overview, Printed in the
USA, 1996
SPATIAL TECHNOLOGY, ACIS 3D Toolkit - Technical Overview, Printed in the USA,
1996/97
structures.com
152
4. STRUCTURE ANALYSIS
4.1 Structure Calculations
Structure analysis may be run using one of the following methods:
1. Choose the command from the menu: ANALYSIS / CALCULATIONS
2. Choose the ANAYSIS icon from the toolbar
3. Select any of the RESULTS layouts (RESULTS DIAGRAMS or RESULTS MAPS)
4. Select values for calculation (reactions, displacements, forces, etc.) from the RESULTS
menu.
The program provides the option that protects against the loss of structure calculation results
(i.e. against displaying structure calculation status: Out of date), when after performed
calculations an operation is carried out in the program which changes the structure data
saved in an *.RTD file.
A global status of the result blockade has been implemented in the program. There are three
methods of setting it:
manually by the user - the Results menu contains the Frozen Results option, which may
be switched on/off; thus structure calculation results are blocked or unblocked,
respectively (NOTE: the option is available only when structure calculation results are
Available)
automatically - on the basis of settings in Job Preferences - the Structure Analysis tab
contains the option Automatic freezing of results; if the option is switched on, then after
each time structure calculations are performed (in other words, when the status of
structure calculation results changes to Available) structure calculation results are
automatically frozen; the option is switched on by default
semi-automatically for the action performed by the user - it concerns only unfreezing of
calculation results; if calculation results are frozen and the user performs any operation
that results in a change of data regarding the structure, then a message will appear on
the screen with warning about the possible loss of the available status of calculation
results; if it is accepted, it changes data regarding the structure and unfreezes calculation
results (if is not accepted, then the change in the structure will not be made and the result
status will not change).
It should be emphasized that if in the program any operation is performed that will lead to a
change of data regarding the structure, then the message with the warning will be displayed
on the screen (of course, if results are frozen). It means that for example, if there is a manual
combination defined (the operation after structure calculations is correct), then the warning
will also be displayed. Obviously, the user will be able to accept the warning and define a
combination, and next, freeze the calculation results manually.
structures.com
153
4.2 Available Analysis Types
In Robot Millennium, the user may define the parameters of various types of structural
analysis. At the beginning of the structure load definition, all structure load cases are
assigned the same type i.e. linear static. In the Analysis Type dialog box which opens once
the Analysis/Analysis Types option is chosen, the analysis type may be changed (e.g. to
nonlinear). New load cases may be created in this dialog box and calculations for analysis
types that do not require previous definition of the static load case (seismic or modal analysis)
may begin.
In the current Robot Millennium version, the following structure analysis types are available:
linear static
non-linear static (the P-Delta effect taken into consideration) - here, the non-linearity is
geometric
buckling (effects of the second order are not taken into consideration)
modal analysis (structure eigen vibrations)
modal analysis recognizing static forces the commonly used modal analysis
(calculations of structure eigen vibrations) does not recognize the influence of static
forces; to approach the real conditions of structure work, one may select in the Robot
program modal analysis recognizing static forces
harmonic analysis
seismic analysis (the following codes are available: French PS69, PS92 and AFPS,
European EC8 code, American UBC97 code, Italian DM16.1.96 code, Romanian P100-92
code, Algerian RPA88 and RPA99 codes, Moroccan code RPS 2000, Turkish seismic
code, Chilean seismic code NCh 433.Of96, Chinese seismic codes, Argentine seismic
code CIRSOC103, Greek code EAK 2000, the code released in USA: IBC 2000,
Canadian code NBC 1995, Russian codes: SniP II-7-81 and SniP 2001)
spectral analysis
Time History analysis non-linear time history analysis is also available
Pushover analysis
elasto-plastic analysis of bars (in the current program version this analysis is available
only for steel sections)
analysis of bars working only for tension/compression as well as analysis of cable
structures.
Overview:
It is assumed that the whole structure is divided into separate parts (finite elements).
Elements are connected together only by the common nodes. Deformation inside the element
is defined by linear combinations of the nodal displacements and the presumed functions
(shape functions). Thus, the whole internal energy of the system depends only on the
independent nodal parameters. The nodal displacements collected (assembled) for the whole
structure form the global vector of unknowns Q. On the basis of an appropriate variational
principle (for example the principle of virtual work) the equilibrium conditions can be
formulated. This leads to the well-known form of the equilibrium equations system:
structures.com
154
M Q'' + C Q' + K Q = F(t) - f(t,Q)
(G.1)
where:
K - the tangent stiffness matrix in the form of the sum of
element matrices
K = K
0
+ K
+ K
NL
, where:
K
0
- initial stiffness matrix (independent of Q)
K
- stress matrix (linearly dependent on the compression

intensity)
K
NL
- matrix of other components dependent on Q
C - damping matrix
M - mass matrix
Q - displacements (increments or total displacements)
Q' - velocities (the first time derivatives of the Q vector)
Q'' - accelerations (the second time derivatives of the Q
vector)
F(t) - external forces vector
f(t,Q) - unbalanced forces vector.
The user can choose the following structure analysis types:
Static analysis
The general system of equilibrium equations can be simplified when the additional
assumption is made that the loading is quasi-static. This means that the loads are applied so
slowly that the velocities and accelerations of the masses are close to zero, and that the
inertia and damping forces and the kinetic and damping energy can be neglected. Such a
reduced system describes a static state of a multi degree of freedom system. There are
generally two approaches to static problems - linear and non-linear.
Linear analysis
Linear analysis is the basic structural analysis type. It assumes that displacements and
rotations are small, that the material is perfectly elastic. This implies linearity of 'cause and
effect', so that the results produced by combining or factoring a basic load case by a factor
are equal to the results from the basic case multiplied by that factor . The stiffness matrix is
then constant and the system of equilibrium equations takes the form of K
0
Q = F, which can
be solved by a single reduction and back-substitution procedure.
The following load types can be defined during structure static calculations:
all types of static loads (concentrated forces nodal and at any point along the element
length, continuous loads constant and variable along the element length)
imposed support displacements and reduction/elongation of bar elements
thermal loads (constant or variable thermal field along the section height).
The displacement (stiffness) method is used to solve linear static problems.
The following are produced:
- node displacements,
- forces and stresses in elements,
- reactions in support directions,
- residual forces at nodes.
If linear static calculations of the structure are to be performed, the user does not have to
define any analysis parameters. Linear static is a default structure analysis type in the
program, i.e. if no other analysis type is defined, the program will conduct static calculations
of the defined structure. All load cases defined in the program are linear static cases.
structures.com
155
Non-linear analyses
The Robot program allows defining different types of nonlinear static analysis of the structure.
Nonlinear structure behavior may be connected with a single structure element (structural or
material nonlinearity) or it may result from a nonlinear relation between forces and
deformation of the entire structure (geometric nonlinearity).
There are three main sources of nonlinearity in the program:
- structural nonlinearity,
- geometric nonlinearity,
- material nonlinearity.
Structural nonlinearity results from either the activation/deactivation of selected structural
components (elements, connections, supports, or d.o.f) as determined by the force or
displacement state of the component, or the use of non-linear elements such as cables.
Therefore, non-linear systems show some features of linearity if no cables are present.
Positive multiplication of load gives an equal increase of results. However, other attributes of
linear systems are not adhered to. The system of equations is assumed to take the form of
(K
0
+ K
+ K
N
) Q = F(t) - f(t,Q),
as for the linear static analysis, but this a priori assumption must be verified a posteriori (after
the system is solved). This usually involves an iterative process. Structural nonlinearity is
automatically set when defined with the element or releases definition.
Geometric nonlinearity results from the non-linear theory applied to the equilibrium equation
formulation and solution (second order approach). In Robot Millennium, geometric
nonlinearities are divided into separate effects: stress-stiffening (influence of the internal
forces on the element stiffness) and P-delta (influence of the structure deformation on its
equilibrium). They can be considered independently since they are activated by separate
options. In addition, the loads may be applied in increments for each type of non-linear
analysis.
Material nonlinearity results from nonlinear material properties (nonlinear stress-strain
relation: it may include elasto-plastic, plastic materials or other nonlinear materials). At
present, material pseudo nonlinearity can be obtained in the Robot program by applying
cable elements, which have nonlinear stress-strain property.
All algorithms used while solving nonlinear projects assume that rotations are small, so that
the tangents and sines of angles can be replaced with the angles themselves during
calculations.
Description of Algorithms Used in Nonlinear Analysis
The Robot program provides two manners of solving a system of non-linear equations: the
method of direct iteration or the incremental method.
In the method of direct iteration, the entire vector of the right-hand part of the system (the load
vector) is applied only once, for it is assumed that the entire load is applied to the structure in
one, single step.
In the incremental method, the right-hand vector (load vector) is divided into n equal parts
called increments. Consecutive load increment is applied to the structure once the state of
equilibrium for the previous increment is achieved. The norm of unbalanced forces is
specified for each step, which allows monitoring of the structure force-deformation relations.
Example of the non-linear process within the incremental method is shown in the figure
below. Values used for non-linear calculations are displayed there.
structures.com
156
Definition of a Nonlinear Analysis Case
Nonlinear structure analysis can be defined for each static load case in the Analysis Type
dialog box that appears on the screen once the Analysis / Analysis Types command is
selected. To define a nonlinear load case, in the Analysis Type dialog box one should
choose a load case (it will be highlighted), for which the analysis type should be changed, and
then press the Set Parameters button. In the Nonlinear Analysis Parameters dialog box
(see the figure below) one should select the Non-linear analysis option or P-delta analysis
option for a given load case and accept changes that have been made.
The iteration process is controlled through defining nonlinear parameters in the Nonlinear
Analysis Algorithm Options dialog box (when Parameters button in pressed).
structures.com
157
To select one algorithm - out of three available in the program - for solving a non-linear
problem, one should set the following parameters:
for the INITIAL STRESS method: - Matrix K actualized after each subdivision - OFF
- Matrix K actualized after each iteration - OFF
for the MODIFIED NEWTON- - Matrix K actualized after each subdivision - OFF
-RAPHSON method: - Matrix K actualized after each iteration - ON
For the FULL NEWTON- - Matrix K actualized after each subdivision - ON
RAPHSON method: - Matrix K actualized after each iteration - ON.
The user may also use the BFGS modification procedure (Broyden-Fletcher-Goldforb-
Shanno). The algorithm of the BFGS method modifies the stiffness matrix during calculations.
The use of the line search algorithm may improve in certain cases the convergence of the
method.
In general, the quickest way to obtain the problems solution is to apply the INITIAL STRESS
method, while the calculations take the longest time when the user decides to select FULL
NEWTON-RAPHSON method. However, the greatest probability of obtaining convergence of
a method occurs in case of FULL NEWTON-RAPHSON method, while the probability is the
smallest in case of INITIAL STRESS method.
The program automatically checks the convergence of the process. The iteration process is
stopped once the state of equilibrium is achieved. Displacement increments dUn and
unbalanced forces dFn are then equal to zero (i.e. are smaller than defined precision
parameters for both values). Iteration process is stopped also in case of divergence. Lack of
convergence can be interpreted either as the numerical effect of structure overload or as a
result of numerical process instability (e.g. when the applied load is divided into a small
number of intervals). In such case, the number of load increments can be increased in the
structures.com
158
program, which usually helps the process to converge. It can also be done by changing
parameters of step-length reduction.
The parameters listed below, located in the Nonlinear Analysis Algorithm Options dialog
box, affect the course of non-linear calculations:
- load increment number is used during the division of a load into smaller segments. For
complex structures where the impact of non-linear effects is considerable, calculations
may not converge, if the analysis is carried out for the value of a load applied in one step.
The number of load increments affects the number of iterations; the greater the number of
increments, the greater the probability for the calculations to reach the point of
convergence,
- maximum iteration number in each load increment is used to control the calculation
process during one load increment,
- allowable number of increment length reduction (modification) defines how many times
the program may automatically change the number of load increments in case when
calculations do not reach convergence
- increment length reduction factor is used to modify the required number of load
increments. This is the so-called conditional option, used only when calculations do not
reach convergence for the currently defined parameters. If the problem does not reach
convergence, the program automatically reduces the size of load increment (depending
on the value of the defined coefficient) and continues calculations. This procedure is
repeated until convergence of results is reached, or else, until the iteration process
exceeds the allowable number of step length reductions.
If the Arc-length method is selected, then the following parameters need to be determined:
load increment number
maximum iteration number for one increment
maximum load factor max - the maximum value of the load parameter
node number, degree of freedom - number of a node located on the structure roof and the
displacement direction
maximum displacement for selected degree of freedom Dmax - the maximum value of
a displacement at a selected node.
The Arc-length method is applied during non-linear pushover analysis; it is strongly
recommended when non-linear structure attributes are defined in a structure model. The arc-
length method (displacement steering) should be applied when the incremental algorithms of
solving equations by force steering are not convergent.
In the Non-linear Analysis Algorithm Options dialog box there is also the Additional
criterions to stop analysis button available, which when pressed opens the Criterions to
stop analysis dialog box.
structures.com
159
In the standard non-linear analysis the load is applied with the increment d = 1.0 / X, where
X number of load increments; consequently, the maximum possible load factor () which
may be reached for convergent calculations, equals
max
=1.0.
Additional criterions of stopping the analysis provided in the dialog box presented in the
drawing above enable execution of non-linear analysis with the increment of the load
parameter determined by the user; the maximum load factor
max
is indefinite or may be
defined by the user.
The following criterions which allow stopping the analysis are available in the dialog box:
1A Structure collapse
1B Reaching load factor
2A Plasticity state
2B Reaching plasticity rate
3A Exceeded maximum displacement of any node
3B Exceeded maximum displacement of a selected node
There is a possibility to define more than one criterion of stopping the analysis; however, the
definition is limited to one criterion from a given group, i.e. 1, 2 or 3. Thus the user may select
for example 1A, 2A, 3B, but the definition 1A, 1B or 2A, 2B or 3A, 3B is not allowed.
Buckling analysis
The stress-stiffening effect can be taken into account in the formulation of linear systems. The
buckling analysis option investigates the influence of the load-level increasing as the stiffness
matrix softens. The buckling load coefficient is found through a linear eigenproblem solution.
This coefficient describes the load-level for which the stiffness matrix becomes singular. The
eigenvector defines the buckling shape associated with the current buckling load
(eigenvalue).
The buckling analysis solves the buckling eigenproblem and the following are determined:
- required number of structure buckling forms,
- critical forces, buckling lengths,
- global value of the critical load.
structures.com
160
Dynamic analysis
Various dynamic analyses can be performed for any of the available structure types. The
general assumptions of a linear theory are used, i.e.:
- small deformations,
- small displacements,
- linearly elastic materials.
The mass used for the dynamic analyses can be entered using the following methods:
- self weight,
- self weight plus concentrated, added masses
- weights resulting from forces the user may change all the formerly-defined forces to
masses, that can be applied during structure dynamic analysis. For example, if a structure
has been loaded with external forces (e.g. with self weight), the weights determined
based on these forces may be considered during structure dynamic calculations.
Modal analysis
The modal analysis determines all parameters for the basic modes of free vibration. These
parameters include eigen-values, eigen-vectors, participation factors, and masses.
The number of modes to be calculated can be entered directly, or by defining a range of
values for the free vibration parameters. The eigenvalues and mode shapes are obtained
from the following equation:
( K -
i
2
M ) U
i
= 0.
Harmonic analysis
In a harmonic analysis, the user defines structures and loads as in a linear static analysis.
Imposed forces are interpreted as excitation force amplitudes. Their frequency, phase angle,
and period are defined by the user. The equation of motion being solved in the harmonic
analysis (assuming that structure damping is neglected) is:
( K -
2
M ) Q = F.
Displacement, internal force and reaction amplitudes are produced by the analysis.
Spectral analysis
The analysis calculates all values from the modal analysis and also calculates the following
for each of the vibration forms:
- participation coefficients for the spectral analysis,
- value of the accelerating excitation spectrum,
- modal coefficients,
- displacements, internal forces, reactions and vibration combinations.
The equation of motion being solved in the spectral analysis is:
K Q + M Q'' = - M A.
Seismic analysis
The analysis calculates all values from the modal analysis and also calculates the following
for each of the vibration forms:
- participation coefficients for the seismic analysis,
- value of the seismic excitation coefficient,
- modal coefficients,
- displacements, internal forces, reactions and vibration combinations.
structures.com
161
Seismic analysis can be performed according to the following codes:
- UBC97 (the Uniform Building Code 1997)
- other non-US codes.
Time history analysis
Time history analysis is an analysis that allows one to obtain the structure reaction at the
selected time points for a defined lasting interaction (contrary to other analysis types available
in Robot that show the structure reaction in the form of amplitudes obtained for a single
moment).
The time history analysis consists in finding a solution of the following equation of the time
variable t:
M * a(t) + C * v(t) + K * d(t) = F(t)
with the following initial values d(0)=d0 i v(0)=v0, where:
M - mass matrix
K - stiffness matrix
C = * M + * K - damping matrix
- coefficient defined by the user
d - shift vector
v - velocity vector
a - acceleration vector
F - load vector.
All the expressions containing the (t) parameter are time-dependent.
The program uses the Newmark method or the method of decomposition to solve the above-
presented task. The Newmark method belongs to the group of algorithms that are
unconditionally convergent for the appropriately defined method parameters. The method
uses the following formulas for calculating displacements and velocity in the next step of
integration:
d(n+1) = d(n) + Dt * v(n) +
Dt
2
2
* [(1-2 ) * a(n) + 2 * a(n+1)]
v(n+1) = v(n) + Dt * [(1- ) * a(n) + * a(n+1)].
The parameters and control the convergence and precision of the results obtained by
means of the method.
The unconditional convergence is assured for

2 5 . 0 .
The algorithm used in the program introduces the two parameters, and , defined by the
user (in the preferences option or in analysis parameters). The default values are the
following: = 0.25 and = 0.5.
It is advisable to use the Newmark method in the case of short time histories, when a
concentrated load is applied to the structure (loads are distributed over small squares). Such
loads will induce a movement that will require a large number of eigenmodes to be described.
Therefore, the Newmark method will be more efficacious than the modal decomposition
method for this type of tasks. The Newmark method takes advantage of the initial equations
without any simplifications. The precision of the obtained results depends on the precision of
numerical integration of time equations, and, for the selected parameters , , it is defined by
the value of the time step. The method does not require the eigenproblem to be solved to
obtain the eigenvalues and eigenvectors. For long time histories, however, the method is very
time-consuming, for in the case of such tasks, calculations have to be performed for a large
number of time steps with the required precision.
The method of modal decomposition is a simpler method of obtaining the required solution. It
is based on the representation of structure movement as a superposition of the movement of
uncoupled forms. Therefore, the method requires the eigenvalues and eigenvectors to be
determined. The Lanczos method is recommended for this purpose. It should be followed by
structures.com
162
the Sturm verification. The method of modal decomposition takes advantage of reduced
uncoupled equations.
The equation (without damping) may take the following form:
MX KX P t

, (1)
where

P t P t
k
k
Ng
k
1

, Ng - number of load groups,

k
t - time history for the k
th
load
group.

X t q t V
i
i
m
i
1
(2)
By inserting equation (2) into equation (1) and recognizing modal damping and the conditions
of orthogonality

V MV
i j i j
,
,
,

V KV
i j i i j
,
,

2
one obtains the following equation
q q q p t
i i i i i i k
i
k
Ng

2
2
1

, i= 1,2,,m
where

p t P V t
k
i
k i k

,
,
i
- modal damping parameters,
i
- frequency for the i
th
form.
Each of equations is solved numerically with the precision of the second order. The resultant
displacement vector

X t for the defined time points t t t t

s
1 2
, ,..., is obtained after
introducing q t
i
, i=1,2,,m into equations (2).

It is worth noting the differences between the analysis types described in this chapter which
are available for a user within the program. Here are the most important differences between
similar analysis types: the "Moving load" analysis differs from Time History Analysis in the
following way: the first analysis type does not recognize the dynamic effects, while the
difference between harmonic analysis and Time History Analysis consists in the fact that the
first analysis type determines the structure reaction exclusively in the form of amplitudes, and
not in that of a time function.
There are the following potential and limitations of Time History Analysis:
the same structure and load types are available as in the case of linear statics
the function of load variability may be defined for an arbitrary static load case, except for
the moving load case (in order to model a dynamic impact of a moving load, one should
define successive vehicle positions in separate load cases and use the time functions
with the phase shift corresponding to the vehicle movement)
additional modeling options, available in the linear static analysis (such as releases,
elastic connections, rigid links, etc.), may also be used in Time History Analysis
the method allows for linear structure work; it means that one may not use non-linear
elements (cables, elements working only for compression/tension, supports and unilateral
releases)
case components of time history analysis may be used in combinations, after generating
an additional load case containing the results of analysis for a given component
the analysis assumes zero initial conditions, i.e. it is not possible to define imposed
displacements or velocities and accelerations
structures.com
163
time history analysis is solved only by means of the modal decomposition method, which
requires the modal analysis to be carried out first
only one time function may be used to determine time variability of loads of a given load
case, it is possible, however, to add (sum up) time functions.
In order to obtain satisfactory results for time history analysis case, the user is required to
carry out iterative analysis with multiple calculations for different case parameters. It means
the modal analysis has to be unnecessarily carried out again. In the case of large-scale
structure, the modal analysis itself may be time-consuming, not to mention the case of time
history analysis. Therefore, it is necessary to select cases for calculations or at least to mark
the modal analysis as calculated. This may also be useful in the case of seismic analysis.
The non-linear time history analysis allows obtaining response of the structure in which any
non-linear elements have been defined. Time history analysis consists in reaching a solution
of the following equation of the t time variable:
M * a(t) + C * v(t) + N (d(t)) = F(t)
with known initial values d(0)=d0 and v(0)=v0, where:
M - mass matrix
K - stiffness matrix
C = * M + * K - damping matrix
N - internal force vector (which is in a non-linear relation to the d shift
vector)
d - shift vector
v - velocity vector
a - acceleration vector
F - load vector.
A load vector is assumed as
n
i
i i
ext
t t
1
) ( ) ( P F , where n denotes a number of force
components, Pi - i-th force component, ) (t
i
- i-th time-dependent function. The excitation
may be expressed in the following form: ) ( ) ( t t
g dir
ext
MI F , where Idir denotes a
direction vector (dir = x, y, z) whereas ) (t
g
is an accelerogram.
NOTE: For the non-linear time history analysis the following simplification is adopted:
C = M.
To solve a non-linear task of time history analysis, the predictor-corrector approach is
employed (see Hughes T.R.J., Belytschko T. Course notes for nonlinear finite element
analysis. September, 4-8, 1995).
Pushover Analysis
The Pushover analysis is a non-linear, static structure analysis, which enables presenting - in
a simplified manner - structure behavior caused by different types of loads resulting from an
earthquake. The magnitude of structure load is increased incrementally according to the
adopted load pattern. Increasing the value of load allows finding structure weak elements and
structure failure modes. The pushover analysis tries to estimate structure real strength. While
determining this type of analysis, it is necessary to define non-linear hinges.
The following limitations have been introduced in the current version of the program:
all the non-linear properties which determine possible structure damage due to forces
resulting from an earthquake are concentrated in non-linear hinges
structures.com
164
non-linear hinges may be applied only in frame (bar) structures
non-linear hinges are treated as independent non-linear connections for each degree of
freedom at a selected node (the interaction between different degrees of freedom is
ignored).
The pushover analysis consists of several stages listed below:
defining non-linear hinges in a structure calculation model
assigning non-linear properties to the hinges (force-displacement or moment-rotation
diagrams)
performing structure modal analysis to determine one mode
defining a set of lateral forces (these forces depend on the mass matrix type used in the
modal analysis)
defining a control node as well as direction and value of the allowable displacement
determining parameters of the non-linear analysis
starting the non-linear analysis; the result of this analysis is the equilibrium state curve V
= V(D) where the shear forces are defined as the sum of reactions for a given direction
caused by the appropriate set of lateral forces
determining the capacity curve ) (
d
cap
a
cap
a
S S S , where
cap
a
S is spectral acceleration
and S
d
is spectral displacement
smoothing of the capacity curve
determining (step-by-step analysis) the performance point.
Elasto-plastic Analysis of Bars
The elasto-plastic analysis enables taking material non-linearity into account. However, it
should be added here that the assumed material non-linearity does not consider the change
of material rigidity depending on external factors, such as temperature; rheological issues
(change of material properties in time) are not included in considerations, either.
Below basic assumptions of the elasto-plastic analysis in the Robot program are presented:
the option operates both for 2D structures (2D frames, grillages) and for 3D structures
(3D frames)
only normal stresses resulting from longitudinal forces and bending moments are
considered (tangential stresses caused by transversal forces and torsional moment are
disregarded)
elasto-plastic analysis is performed for selected bars indicated by a user; it is assumed
that the analysis type is not changed globally for the entire structure since this type of
analysis is time-consuming and requires defining local conditions for a bar (division of a
section or bar, material model)
elasto-plastic analysis is performed only for structure bars.
In the current program version the following operation methods as regards the elasto-plastic
analysis have been adopted:
structures.com
165
analysis on the element level: adding global degrees of freedom
Bars are divided into smaller calculation elements. Additional nodes and calculation
elements are invisible to the user. A method of automatic division may be determined
using the Maximum element length option located in the Analysis Type dialog box (the
Structure Model tab). The user may also define a value of the division parameter applying
the Division of elements for elasto-plastic analysis option contained in the Job
Preferences dialog box.
In each element, stresses are calculated at three points (third-order Gauss quadrature is
applied here).
analysis on the section level: the layered structure approach, however, a uniform material
within a section is assumed
A section is divided into a set of layers (fibers); in a bi-axially bent section they may
referred to as zones. In each of the zones stresses are checked according to the
assumed model. Forces acting on the bar axis are obtained by integration of forces in all
section zones.
For each of the zones of section division the following parameters must be determined:
coordinates (y
i
, z
i
) of the point being the gravity center of a zone in the principal central
system of a section, area of Ai zone, Mi material ascribed to the zone, where i denotes
the zone number (i=1,, N).
The analysis is performed in such a manner that for each load increment, the program
calculates displacement increments at division points along the bar length. Afterwards,
based on the displacements, deformations at points in a section are computed.
Considering the function that describes the material model, for a given zone, stresses are
calculated at each point depending on the current deformations. Next, internal forces are
determined on the basis of stresses. In the end, internal forces at all points (zones) are
summed (integrated) in order to obtain internal forces in a bar.
material model: elastic-perfectly plastic or elasto-plastic with hardening: material behavior
- elastic and linear, in the plastic range - linear with hardening; the model is generated
based on the material data: Youngs modulus (E) and yield strength (Re).
In the plastic analysis the unloading method should be defined, as well. It determines the path
of material behavior once the yield strength is exceeded during unloading, when deformations
become smaller (deformation gradient is negative). There are four ways of material behavior
available: elastic, plastic, damage, mixed.
Results obtained due to the elsto-plastic analysis of bars:
Deflections
During elasto-plastic analysis the program calculates deflections directly within a bar. The
user obtains values of displacements and nodal rotations within the internal division of a bar.
To achieve deflections at any bar point, the linear interpolation between points of internal
division has been implemented.
Internal forces
Internal forces in a bar are calculated in the identical manner as for the linear analysis. Based
on the forces and moments in the initial node and the load along the bar length, the program
calculates internal forces at any point of a bar.
Only for the P-delta analysis a different algorithm is applied. In bar internal forces, the effect
of the eccentricity caused by a bar deflection is taken into account. Deflections are obtained
at points of bar internal division.
Stresses
In the elasto-plastic analysis normal stresses in a section are not distributed linearly; they are
calculated independently for each of the division zones. Some of the zones may fall within the
plastic range, whereas others may still belong to the elastic range of stresses. Therefore, the
user cannot determine unambiguously the stresses at a point along the bar length.
The table shows maximal and minimal values of stresses within a section. For elasto-plastic
sections separated stresses resulting from bending and those resulting from longitudinal
forces are unavailable. The accurate stress analysis within a section of elasto-plastic bars in
the Stress Analysis module is impossible.
structures.com
166
4.3 New Case Definition and Analysis Type Change
Once the Analysis /Analysis
Types command is chosen,
the dialog box shown below
will appear on the screen.
All previously defined
structure load cases will be
listed on the Analysis Types
tab.
New cases may be defined
or the structure analysis
type for the selected case
can be changed using the
dialog box.
To change the structure analysis type, select
the load case and press the Change Analysis
Type button. The dialog box shown below will
appear on the screen. The new analysis type
should be described within. Once the analysis
type is chosen and the OK button pressed, an
additional window will appear on the screen in
which the parameters of the selected analysis
type may be chosen. The new structure
analysis type will be entered in the Analysis
Type dialog box in the Analysis Type column.
structures.com
167
To add a new case, press the Define New
Load Case button in the Analysis Type dialog
box. The dialog box shown below will appear
on the screen. The new analysis type should
be defined below. Once the analysis type is
chosen and the OK button pressed, an
additional dialog box will be displayed on the
screen in which the parameters of the selected
type may be defined. The new structure
analysis type will be entered in the Analysis
Types dialog box in the Analysis Type column.
As an example, the definition method of time history analysis case will be presented. After
defining modal analysis for the structure and selecting the Time history analysis option in the
above dialog box, the Time History analysis parameters dialog box opens, in which
parameters of time history analysis can be determined.
The above-presented Time history analysis dialog box contains the following parameters:
the top part of the dialog box contains an editable Case field containing name of a time
history analysis case
structures.com
168
the Method field allows one to select a method of carrying out time history analysis; the
modal decomposition method is the default value; the field contains also the Parameters
button that opens the dialog box allowing one to determine detailed parameters of the
analysis method
edit field in the Time field:
Time step - the step of time variable for which the results are stored
Division - the number of time step divisions defining the frequency of storing the analysis
results
End - the end value of time variable for which the analysis is carried out
If Newmark method (linear time history) or Predictor-Corrector method (non-linear time
history) has been selected, then in the Division field the number of time step divisions
(time step of saving results) is specified to enable definition of the time step of integration,
i.e. time step of integration equals Time step / Division. When the division value equals 1,
the time step of saving results is identical as the time step of integration.
If modal decomposition method (linear time history analysis) is selected, the algorithm
calculates - for each mode - the maximum value of the time step of integration equaling
the value of period divided by 20 (such an operation is performed to guarantee stability
and precision of the obtained results). Thus calculated step value is divided by the
division value; the value received (e.g. step_1) is compared to the time step of saving
results. A smaller one of these 2 values mentioned (i.e. step_1 and time step of saving
results) is adopted as the time step of integration. However, note should be taken that if
the first one of these values (i.e. step_1) is to be applied in calculations, it is slightly
modified so that the time step of saving results is a multiple of this value.
unfolding list of the available simple static load cases or masses in directions X, Y or Z
unfolding list of the defined time functions and the preview of the diagram of the selected
function
Coefficient edit field
Phase shift edit field
Function definition button
The definition of a time function may be carried out in two ways in the Time function
definition dialog box:
by typing the values of the time point T [s] and the dimensionless value of the function
F(T) in the appropriate edit fields, which is to be followed each time by pressing the
Add button; the successive points belonging to the function are introduced into the
list that defines the function course
by pressing the Add expression button, which results in opening the dialog box
where one may define the course of the function by means of mathematical formulas
(sum, subtraction, multiplication, division, trigonometric functions, power function,
square root function).
a table containing the following columns: case-function-coefficient-phase, where:
Case indicates the number of the selected load case or mass direction
Function is the name of the time function selected for the given load case
Coefficient - the incremental coefficient for time function value for the given load case; the
default value of the coefficient = 1.0
Phase - phase shift of the time function for the given load case; the default value = 0.0.
The Analysis type dialog box contains five more tabs:
Structure model
structures.com
169
Load to mass conversion
Combination sign
Result filters
Buckling deformation.
The Structure model tab contains three options in the Node generation field:
Node generation at intersection of diagonals once this option is selected, an additional
node will be created at the intersection of sloping bars.
Node generation at intersection of vertical/horizontal bars - once this option is selected,
an additional node will be created at the intersection of vertical and horizontal bars.
Node generation at intersection of bars and finite elements - once this option is selected,
an additional node will be created at the intersection of bars and planar finite elements.
The mesh of planar elements will be adjusted to the position of the generated nodes.
Additionally, this field contains two options: a bar list which are to be omitted during node
generation at intersections and an object list which are to be omitted during structure model
generation.
The Assembling Case Definition for Cable Structures field contains an edit list allowing the
user to determine which among the load cases defined for a cable structure will be treated as
an assembling case (see chapter 3.11).
In the Tolerance of Structure Model Generation field, one may define the parameter
determining the precision of structure definition. Pressing the Precision calculation button
results in determining the precision of structure definition (default setting: 1mm; if the
precision is smaller than 1mm, the program displays the calculated value). The program
assumes a standard value of tolerance (1mm), but the user may define a new tolerance
value. Pressing the Generation of Calculation Model button results in creation of a structure
model. The program will generate bar-type or planar finite elements as well as intersections of
bars or bars and planar finite elements depending on which options are checked in this dialog
box.
The Load to mass conversion tab
contains options that allow the user to
convert static loads to masses used in
dynamic calculations.
The option allows the user to define
loads/masses once for all calculation
purposes. It is not necessary to define
separately static loads taken into
account in static structure analyses and
the masses taken into account in
dynamic structure analyses. On the
basis of the already-defined static loads
one may create masses to be used
during dynamic calculations. In order to
perform conversion of static loads to
masses used in dynamic calculations,
one should:
determine the cases for which conversion of loads to masses will be carried out (one
should provide the numbers of load cases, define the direction of conversion process and,
additionally, the multiplication coefficient to be applied to the value of the static load)
structures.com
170
determine the set of directions in the global coordinate system (X, Y and Z) along which
the masses will operate
determine the dynamic analysis case that will use the masses created from loads; once
the Add Mass to Global Mass option is selected, the masses created from loads will be
taken into account in all cases of dynamic structure analyses
press the Add button.
Beside the Add button, there are two more buttons that allow one to:
Delete - a load case selected from the list of cases to be converted to masses
Modify - the parameters of a load case from the list of cases to be converted to masses.
Conversion of loads to masses retains the differentiation of load types. In other words,
concentrated forces are automatically converted to concentrated masses, distributed loads to
distributed masses, and moments to rotational masses. The converted masses may be
viewed in the table of masses that can be called by selecting the following option from the
menu: Loads / Mass table. Mass values are presented in the table as weight values
(acceleration of gravity is applied here). To differentiate them from the user-defined masses,
the masses created due to the conversion process are marked in the table with the CNV
symbol in the MEMO field. The symbol informs also about the origin of a mass for the benefit
of the conversion procedure. NOTE: For shell structures the hydrostatic pressure load cannot
be converted to masses.
The options available in the dialog box shown below are used to define the sign of the
generated combination in the case of seismic and spectral analysis. One determines the case
of seismic or spectral analysis and the dominant mode (number of the mode that will be
dominant in defining combination sign).
If a user does not select any dominant mode (i.e. 0 mode is selected), then adopted
combination sign will be the same as the combination sign calculated by means of the formula
for seismic combination type.
In the bottom part of the dialog box one can select a default calculation type for seismic
combination:
CQC - Complete Quadratic Combination
SRSS - Square Root of Sums of Squares
10% - 10% double sum
2SM - double sum.
Result filters are the fifth tab in the Analysis Types dialog box. The options available in this
dialog box are used for global selection of results obtained for nodes, bars, etc. defined in a
structure.
In the above dialog box the following edit fields are located: case list, node list, bar list, list of
panels/objects and list of finite elements. These are the fields where numbers of cases,
nodes, bars, etc. can be typed for which results of structure calculations will be presented; in
other words in a result table only rows of a table assigned to typed numbers of cases, nodes,
bars, etc. will be presented, however, the results for other (not typed) cases, nodes, bars, etc.
will not be presented in a table.
NOTE: If the edit fields located on this tab are empty it denotes that calculations will
be performed for all load cases, nodes, bars, panels, objects and finite
elements defined in a structure.
The options provided on the Buckling Deformation tab are used to generate a structure model
which includes deformations induced by a selected buckling mode or linear combination of
modes (deformations do not cause initial forces or stresses in a structure; if deformations are
considered, it results only in changing the structure geometry).
structures.com
171
In order to apply the option, first the user should define buckling analysis case and perform
structure calculations. The change of structure geometry results in deletion of a deformed
structure; it is necessary to carry out structure calculations again (the calculations should be
carried out both for the initial structure - calculation of new buckling modes, and for a
deformed structure). All the results are displayed on the deformed structure, whereas nodal
displacements are specified in relation to the initial geometry defined by the user.
If the Consider buckling mode as initial deformation option is switched on, then the options in
the dialog box become accessible and parameters of structure deformation may be defined.
Pressing the Apply button accepts the choice (i.e. switching on or off deformations), whereas
pressing the Close button closes the dialog box without remembering changes.
The Parameters field includes the Case selection list containing buckling cases defined for a
structure. Based on a selected buckling case buckling deformations are determined. Below,
the Mode and Coefficient fields are provided, which allow the user to determine the number
of a buckling mode and the coefficient with which the selected mode will be included in a
linear combination.
Deformations can be scaled after defining a value in the Maximum displacement field; the
value enables scaling a selected mode or mode combination.
If the Omit case for structure with deformations option is switched off, then a given buckling
case will be calculated for a deformed structure; if this option is switched on, then the case
will be omitted during calculations.
In order to define buckling deformations in a structure, first the user should define a buckling
case and perform calculations. Next, applying the available options, the user should define
the initial deformation resulting from a selected buckling mode. After changing the structure
geometry, the result status changes to NOT AVAILABLE, therefore, structure calculations should
be performed again.
Thus, calculations for a structure with buckling deformations defined are conducted in two
stages:
stage 1 - calculations of an initial structure (without deformations)
stage 2 - calculations of a deformed structure.
Both stages are carried out automatically.
structures.com
172
SELECTED REFERENCES
K.J. BATHE, Finite Element Procedures in Engineering Analysis, Prentice Hall, New Jersey
1982
E.L.WILSON, An eigensolution strategy for large systems. Computers&Structures, Vol.16,
No. 1-4, pp.259-265, 1983
E.L. WILSON, A new method of dynamic analysis for linear and nonlinear systems. Finite
Elements in Analysis and Design, 1, 1985, 21-23, North-Holland
E.L.WILSON, Three dimensional dynamic analysis of structures, Computers and Structures,
Inc., Berkeley, California, USA, 1996
R.W.CLOUGH, J.PENZIEN, Dynamics of Structures, McGraw-Hill Book Comp., 1975, 634 p.
S. YU. FIALKO, Investigations of the Initial Imperfections Influence to Natural Vibrations of
Ribbed Conical Shells, Soviet Applied Mechanics, 1982, 18, N11, pp.118 - 122. (In Russian)
S. YU. FIALKO, Non-steady vibrations of ribbed conical shells under the influence of local
loads, Soviet Applied Mechanics, 1987, v23, N6, p. 547-552
S. YU. FIALKO, High-performance aggregation element-by-element iterative solver for large-
scale complex shell structure problems, Archives of Civil Engineering, XLV, 2, 1999, p.193-
207
S. YU. FIALKO, High-performance aggregation element-by-element Ritz-gradient method for
structure dynamic response analysis. CAMES (Computer assisted mechanics - engineering
sciences), IV, 2000
G. GAMBOLATI, G. PINI, F. SARTORETTO, An improved iterative optimization technique for
the leftmost eigenpairs of large symmetric matrices, J. Comp. Phys., 74: 41 - 60, 1988
G. GAMBOLATI, G. PINI, F. SARTORETTO, Accelerated simultaneous iterations for large
finite element eigenproblems, J. Comp. Phys., 81: 53 - 69, 1989
M. PAPADRAKAKIS, A partial preconditioned conjugate gradient method for large
eigenproblems, Comp. Meth. Appl. Mech. Eng., 62: 195 - 207, 1987
M. PAPADRAKAKIS, Solving large-scale problems in mechanics, John Wiley & Sons Ltd,
1993
S. BITZARAKIS, M. PAPADRAKAKIS, A. KOTSOPULOS, Parallel solution techniques in
computational structural mechanics, Comp. Methods Appl. Mech. Engrg. 1997, 148 , p.75-
104
T.J.R. HUGHES, M. FERENCZ, Implicit solution of large-scale contact and impact problems
employing an EBE preconditioned iterative solver, IMPACT 87 Int. Conference on Effects of
Fast Transient Loading in the Context of Structural Mechanics, Lausanne, Switzerland,
August 26-27, 1987
T.J.R. HUGHES, M. FERENCZ, J.O.HALLQUIST, Large-scale vectorized implicit calculations
in solid mechanics on a CRAY X-MP/48 utilizing EBE preconditioned conjugate gradients,
Comput. Meths. Appl. Mech. Engrg., 61
B. N. PARLETT, The Symmetric Eigenvalue Problem, Prentice-Hall, Inc., Englewood Cliffs,
N.J. 07632, 1980
V. E. BULGAKOV, M. E. BELYI, K. M. MATHISEN, Multilevel aggregation method for solving
large-scale generalized eigenvalue problems in structural dynamics, Int. J. Numer. Methods
Eng., 40: 453 - 471, 1997
V. E. BULGAKOV, Iterative aggregation technique for large-scale finite element analysis of
mechanical systems, Comput. Struct, 52: N4, 829-840, 1994
V. E. BULGAKOV, G. KUHN, High-performance multilevel iterative aggregation solver for
large finite-element structural analysis problems, Int. J. Numer. Methods Eng., 38: 3529-3544,
1995
structures.com
173
5. RESULT ANALYSIS
Once the structure calculations are completed, analysis of results may be viewed in two
forms: as graphics (diagrams or maps of the structure load effects) or in tabular form (tables
presenting components of displacements, reactions, internal forces, etc.). It is easier to review
the results by opening the RESULTS layout. Once the structure calculations are completed,
the screen will be divided into three parts in case of Bars and two parts in case of Surface
Elements (see the figures below): the graphic viewer where the designed structure is
displayed, the Diagrams or Maps dialog box and the Reactions table (in case of surface
elements, these tables can be opened from the text menu when necessary).
The program provides the option that protects against the loss of structure calculation results
(i.e. against displaying structure calculation status: Out of date), when after performed
calculations an operation is carried out in the program which changes the structure data
saved in an *.RTD file.
A global status of the result blockade has been implemented in the program. There are three
methods of setting it:
manually by the user - the Results menu contains the Frozen Results option, which may
be switched on/off; thus structure calculation results are blocked or unblocked,
respectively (NOTE: the option is available only when structure calculation results are
Available)
automatically - on the basis of settings in Job Preferences - the Structure Analysis tab
contains the option Automatic freezing of results; if the option is switched on, then after
each time structure calculations are performed (in other words, when the status of
structure calculation results changes to Available) structure calculation results are
automatically frozen; the option is switched on by default
semi-automatically for the action performed by the user - it concerns only unfreezing of
calculation results; if calculation results are frozen and the user performs any operation
that results in a change of data regarding the structure, then a message will appear on
the screen with warning about the possible loss of the available status of calculation
results; if it is accepted, it changes data regarding the structure and unfreezes calculation
results (if is not accepted, then the change in the structure will not be made and the result
status will not change).
structures.com
174
It should be emphasized that if in the program any operation is performed that will lead to a
change of data regarding the structure, then the message with the warning will be displayed
on the screen (of course, if results are frozen). It means that for example, if there is a manual
combination defined (the operation after structure calculations is correct), then the warning
will also be displayed. Obviously, the user will be able to accept the warning and define a
combination, and next, freeze the calculation results manually.
5.1 Diagrams
The Diagrams dialog box is used to display the structure deformations and the chosen
diagrams of internal forces and stresses in the bars. This option is available from the menu
by choosing the Results / Diagrams for Bars option or choosing the RESULTS -
DIAGRAMS layout.
The Diagram dialog box is made up of six tabs:
NTM
Deformation
Stresses
Reactions
Reinforcement
Parameters.
Two additional options are found in the lower part of the Diagrams dialog box: Scale and
Open a New Window. The first option is used to set the scale of the diagrams presented on
the screen. Once the Open a New Window option is chosen, a new additional window will be
displayed on the screen in which Diagrams of the selected values will be presented.
For example, after selecting the NTM tab, the dialog box will take the shape shown below.
Internal forces for which diagrams will be selected
may be chosen in this dialog box. Pressing the
Apply button results in displaying the diagrams of
the selected values in the graphic viewer.
structures.com
175
The user may select for presentation, displacements obtained during static structure analysis
as well as eigen vibration modes obtained for cases of dynamic structure analysis. Options for
animating the structure deformation diagrams presented on the screen are also available on
the Deformation tab. To activate the animation, two animation parameters should be
provided: the number of frames created and the number of frames per second. Once the
Start button is pressed, the program will prepare the animation of the selected value based
on the parameters assigned and begins the animation. During the presentation, a toolbar
appears on the screen with options to stop, pause, replay etc. It is also possible to save the
created structure deformation animation to an *.avi file. It is also possible to read and replay
the *.avi file in which the structure deformation animation was initially created. Once the
Stresses tab is selected in the Diagrams dialog box, one can select the components of
stresses obtained during the static analysis of a structure.
On the Reactions tab one may choose the following quantities for presentation:
reactions: reaction forces - values of reaction forces at supports and reaction moments -
values of moments at supports
residual values: residual forces sums of forces at individual structure nodes (check of
the equilibrium of forces in structure nodes) and residual moments sums of moments at
individual structure nodes (check of the equilibrium of moments in structure nodes).
Forces and moments are presented in the global coordinate system.
Once the Reinforcement tab is selected the dialog box shown on the figure below appears on
screen.
The dialog box allow one to select the quantities obtained once theoretical (required)
reinforcement calculations for R/C members have been performed:
the top-most field contains:
structures.com
176
theoretical (required) reinforcement top reinforcement (along the side of b length),
bottom reinforcement (along the side of h length)
real (provided) reinforcement - top reinforcement (along the side of b length), bottom
reinforcement (along the side of h length)
the second field from top contains:
theoretical (required) and real (provided) reinforcement ratio
the third field from top contains:
theoretical (required) and real (provided) stirrup spacing (for transversal reinforcement)
the first field from bottom contains:
number of top bars (top reinforcement) - along the side of b length
number of bottom bars (bottom reinforcement) - along the side of h length.
Once the Parameters tab is chosen the dialog box
shown below will appear on the screen.
The type of diagram presentation may be chosen in
the dialog box:
presentation method of the value descriptions
may be chosen from the Diagram Description
field: none, horizontal or vertical descriptions.
the Positive and Negative Values field allows
the user to decide whether the positive and
negative values will be differentiated by color.
the method of filling the diagram may be
defined in the Filling field as fence or filled.
Each diagram presented in the graphic editor may be printed. Two printing methods are
possible:
while in the graphic editor select the File / Print command; this will result in printing the
contents of the graphic viewer.
while in the graphic viewer, select the File / Screen Capture and then File / Printout
Selection.
Select desired printout components in the dialog box.
structures.com
177
NOTE: Detailed information about printing from Robot Millennium will be provided
in Chapter 8 of the manual.
5.2 Tables
Once the calculations are complete, the table containing reactions in the support nodes,
calculated during the structural analysis, will be opened. Tables with other structural results
(internal forces, displacements, stresses, etc.) may be shown on the screen by choosing
View/Tables or selecting the appropriate icon from the toolbar. An example of the table is
shown below.
Four tabs are located at the bottom of the table: Values, Envelope, Global Extremes and Info.
To scroll the tab field, click on the arrows in the appropriate scrolling direction, right or left, up
and down. Selecting the Value tab results in the presentation of the detailed values
(reactions, internal forces etc.) for all structure nodes/bars and load cases defined for the
structure. Once the Envelope tab is chosen, maximum and minimum values will be presented
at all structure nodes/bars. Selecting the Global Extremes tab displays the minimum and
maximum values from all the values obtained during the structural analysis. The Info tab
displays the nodes, bars and load cases for which values obtained during analysis will be
presented in the table.
By default, the results for all nodes/bars and load cases defined within the structure are
presented on the Value tab. To select the analysis of results which is most useful for the user,
press the right mouse button while the cursor is located in the table and choose the Filter
option from the context menu appearing on the screen. The dialog box shown below will be
open (this is an example of a dialog box with reactions obtained at the structures support
nodes).
structures.com
178
Selection of the
attribute for filtering:
node/bar, load case
List of the
selected
nodes/bars, load
cases
Selection of the attribute
type according to which
filtering will be performed
(for reactions table these
will be: support, rigid
connections, etc.)
Selection of the
attribute according
to which filtering will
be peformed
The range of results presented in the tables may be defined here. Nodes/bars for which the
results will be presented may be selected. This may be done using the buttons which are
found in the upper part of the dialog box (All, None, Inversion, Previous) or the options
found in the Attributes tab. The selected nodes/bars and load cases will be entered into the
appropriate field found in the upper part of the Filters dialog box.
The table containing the analysis results may be freely composed. Once the structure
calculations are completed, a table with the reaction values for support nodes will appear on
the screen. Additional columns containing data and the remaining analysis results may be
added into this table. To do this, press the right mouse button while the cursor is positioned
somewhere within the table and choose the Table Columns command from the menu. This
opens the dialog box shown below, where the values to be displayed in the table may be
selected. The dialog box is made up of several tabs (General, Displacements, Reactions,
etc.). Once one or more values are selected (the symbol will appear) on one of the tabs and
the OK button is pressed, the columns will appear in the table where the values selected by
the user will be displayed.
structures.com
179
The Supports tab is shown beside as
an example (none of the dialog box
options has been chosen).
Two options are found in the lower part
of the above dialog box:
columns will be added to the tables
- additional columns containing the
values selected in the above dialog
box will be added to the table;
columns will replace the existing
ones - the currently available
columns will be removed and
replaced by the columns
containing the values selected in
the above dialog box.
As with diagrams, the content of each table may be printed. Printing the tables may be
executed in one of two ways:
having the table active - select the File/Print command from the menu. This will result in
printing the contents of the current tab displayed in the table (NOTE: the table contents
may not be printed if the table is in the Edit mode);
having the table active - select the File/Screen Capture command from the menu (this
results in saving the contents of the table for use in a composed printout) and then the
File/Printout Composition command.
(NOTE: For detailed information on printout composition, see Chapter 8.)
It is also possible to copy all or a part of the table to a spreadsheet (Excel, Lotus etc.). To do
this, highlight part or the entire table, press Ctrl+C or the Copy icon and once the
spreadsheet is open, press Ctrl+V or the Paste icon .
Cost estimate belongs to interesting options whose results are presented in table form. The
option is used to estimate the cost of the designed structure. The option is available by
selecting the Tools/Cost estimate command from the menu. To obtain a cost estimate, one
should first define certain groups of sections (in general, steel sections differ in price
depending on the section type) and section protection layers in the Cost estimate dialog box.
Once the section groups and protection layer types are defined, one should assign sections to
appropriate groups. Assigning sections to groups and protection layers ends the process of
estimating costs. Cost estimate in the form of a table becomes available by pressing the
Apply button in the Cost estimate dialog box or by selecting the View/Tables command from
the menu and selecting the Cost estimate option in the Tables - data and results dialog box.
structures.com
180
5.3 Maps on Bars
The option is used to present for bar structures color maps obtained in the process of
analysis of sectional forces, stresses, deformations, etc. within a structure. The option is
available by selecting
Results/Maps on bars command from the menu.
Maps on bars icon in the toolbar
The below-presented dialog box, appearing on screen
after selecting the command, consists of four tabs:
NTM/Stresses, Design, Scale, Parameters.
For a structure, it is possible to present only one
quantity selected in the presented dialog box (the figure
shows the NTM/Stresses tab as an example).
On this tab, one can select the quantities for which
diagrams are to be presented:
in the Force components field: FX, FY, FZ, MX,
MY, MZ
in the Normal stresses field: maximum stresses
(from My and Mz), axial FX/AX
in the Shear/torsion stresses field: shear stresses
TY and TZ, torsion stresses T.
The bottom part of the dialog box contains the
Structure Deformation option. Switching it option on
results in displaying the deformation of the current
structure resulting from the applied load.
This part of the dialog box contains also the Normalize button. Pressing it results in
presenting the diagrams of a selected quantity in such a way that their scale will adjust to the
maximum and the minimum value of the quantity. On the Design tab one can select the
following quantities related to structure member design: ratio, member length, slenderness
Lay, slenderness Laz.
The options located in the Scale tab allow one to determine the set of colors and the range of
values for the selected quantity that will be presented in diagrams.
structures.com
181
n the Parameters tab of the Maps on bars
dialog box, one can select the manner of
presentation of maps on structure bars:
in the Diagram descriptions field, one may
decide how to present quantity
descriptions on the diagrams
none - if this option is selected, the
descriptions of the quantities available in
the Maps on bars dialog box will not be
presented
horizontal - if this option is selected, the
the Maps on bars dialog box will be
presented horizontally
vertical - if this option is selected, the
the Maps on bars dialog box will be
presented vertically
In the Diagram thickness field, one can
introduce a digit determining the ration of
the line used in map presentation to the
thickness of the line presenting structure
bars.
5.4 Maps (Panels)
The Maps dialog box is used to display the structure deformations and the chosen maps of
the internal forces, stresses and displacements across the surface elements. This option is
available by choosing the Results/Maps option or choosing the RESULTS - MAPS layout.
The Maps dialog box is consists of seven tabs: Detailed, Principal, Complex, Parameters,
Scale, Animations, Crosses.
Results obtained for surface finite elements are presented in local coordinate systems which
may be defined and modified by the user at any moment of result presentation. Results
obtained for structures which contain surface FE may be presented in the form of isolines or
color maps on selected elements. Only one value may be presented in a dialog box. In order
to display other value maps on the screen:
select the value and press the OK button. The map of the selected value will replace the
current map on the screen.
select this value and activate the Open a New Window option. A map of the selected
value will be presented in a new window; a map of the previous value will be kept on the
screen.
There are multiple options for the display of the maps. User has also ability to select the
direction of the cut calculated, the layer for which the load effects are displayed, choose
between maps and isolines. Deformation, descriptions, and smoothing of the maps can be
selected as well. Finally, the user chooses the range of color coding for the maps and
animates the deformation of the panels. Here is an example of the Scale tab within the Maps
dialog box.
The presented maps/isolines may be displayed with description of particular isolines. Maps
may be created with the option Smoothing. Surface FE results are defined Gauss points
located inside each element (values estimated in a common node of adjoining elements may
differ slightly in each element, and isolines may not be continuous). To obtain a smooth map
of selected values, select the with smoothing option. The Smoothing option may be used in
the following manner:
structures.com
182
no smoothing
global smoothing (on the entire structure)
smoothing within a panel - it does not recognize the edges between panels
smoothing according to selection - the operation covers only the selected elements.
There are three more options available:
with description - if the option is switched on, the maps will be presented with the
description of the values of particular isolines
with normalization - if the option is switched on, the maps of a selected quantity will be
automatically presented with in such a way that the scale will be adjusted to the maximum
and minimum value of the selected quantity.
with FE mesh if this option is switched on, then apart from a map of a selected quantity,
the program also presents generated finite element mesh.
Particular tabs in the Maps dialog box have the following functions:
the Detailed tab - the following values may be chosen for presentation: stresses,
membrane forces, moments, shear stresses, shear forces, displacements, rotations and
the main direction of the co-ordinate system to be used during the presentation of the
results for planar finite elements
the Principal tab - the following extreme values may be selected for presentation:
stresses, membrane forces, moments, shear stresses, shear forces
the Complex tab - the following reduced values may be chosen for presentation:
membrane forces, moments and stresses
the Parameters tab - the placement of presentation of results obtained for surface FE may
be defined in this dialog box
the Scale tab - change the map presentation parameters: the type of color palette, the
scale type (linear, logarithmic, arbitrary), maximum and minimum division number, in
which a map of the selected value will be presented (minimum and maximum value
inclusion may also be turned off) and define the map color change division number, the
colors using which the value map and values for the colors will be presented; it is also
worth mentioning here that the user may select the automatic scale; while maps are being
generated, it is often the case that the user loses, among other things, the information
concerning the sign of values presented; without help of the scale presented in the dialog
box, the user cannot indicate - on the map - the point at which the sign changes and
specify the ratio of minimum values to maximum values. An automatic scale is generated
in such a way so that negative values of a quantity presented on a map are marked with
cold colors (shades of blue), whereas positive values with warm colors (shades of red);
thus created map scale allows both locating the regions where the sign changes and
determining the ratio of extreme values. The automatic color scale is created once the
extreme values (vmin, vmax) for a given quantity are determined. Colors contained
between vmin and vmax are ascribed to values by means of the uniform division of the
scale (zero is one of the limit values on the scale).
the Deformations tab - presentation of the deformation of the currently designed structure
and running animation of the deformation
the Crosses tab - here, the results may be presented in the form of crosses for three
types of quantities: stresses, forces and moments.
structures.com
183
The drawing below presents an example of a displacement map for a plate, accompanied by
the Maps dialog box.
The X and Y axes are local and defined using the DIRECTION option on the Detailed tab.
This option allows to define the direction of the main local coordinate system which will be
used (x axis). The result values for all elements are calculated to this rotated coordinate
system. The direction may be defined by entering any vector defining the main direction
when selecting the results for surface FE. The selected vector is then projected onto an
element which in the end defines the positioning of the local x-axis. Only one limit exists - the
main vector can not be perpendicular to the element (i.e. parallel to the local element axis).
If the user selects this direction, all results will be equal to zero.
The typical situation is shown on the drawing below where all elements are located on a
cylindrical surface. Initially local x axes are parallel to the global X axis. These axes will be
redefined when using the main direction (reference direction) which is parallel to the global Y
axis.
structures.com
184
The below-presented pictures define the sign convention applied to forces for planar finite
elements.
structures.com
185
5.5 Panel Cuts
The Panel cuts option completes the Maps - Panels option. It allows the user to create
diagrams of internal forces and of displacements in planar finite elements. The diagrams are
drawn in arbitrary cuts defined by the user. For two-dimensional structures, the cut is done
along a straight line or a segment located in the plane in which the structure has been
defined. For shells, the cut is defined by an arbitrary plane in 3D space. The user may define
several cut diagrams for one structure.
The option becomes available by:
selecting Results/Panel cuts command from the menu.
selecting the Panel cuts icon from the toolbar.
The command activates the dialog box that allows one to define cuts and select the quantities
to be presented for the cuts. To create a diagram of any of the available quantities of internal
forces, stresses and local displacements in finite elements, it is necessary to define the line of
the cut and indicate which of the available quantities is to be displayed in the diagram.
The option allows one to define more than one plane (straight line) determining the cut. Thus,
diagrams may be drawn simultaneously for several cuts. Defining a new cut adds it to the
previously defined ones. Diagrams may be drawn as tangential or normal with respect to the
plane of the cut. The option also allows one to have the element-to-element transitions
smoothed out or not.
The below-presented dialog box - appearing on screen, once the option is selected - consists
of nine tabs: Definition, Cuts, Detailed, Extreme, Complex, Parameters, Diagrams,
Reinforcement and SLS.
The last two tabs refer to reinforcement calculation results for a defined plate or shell. One
may select for presentation calculated reinforcement area or spacing, and in case of design
code selection for plate/ shell including calculations according to SLS, quantities relating to
service state (crack width, etc.).
To display a diagram of any quantity on screen, one should:
define cuts through a plate or shell structure
select the quantity to be presented in a diagram
determine the diagram parameters
press the OK button. The diagram of the selected quantity will be presented on screen.
NOTE: In the case of a cut through panels (surface FE), directions are set in the
following manner: direction xx is tangential to the plane and perpendicular to
the cutting line, while the yy direction is tangential to the cutting line.
Cut Definition - Two-Dimensional Structure (Plate)
If one designs a two-dimensional structure, the plane of the cut must be parallel to OZ axis. Thus, it is
necessary to define only a straight line- the intersection of the cut plane and the OXY plane. For two-
dimensional structures, it is additionally possible to define a segment of determined length for which the
required cut is carried out. Such segments may form a polyline of arbitrary shape.
structures.com
186
The program provides many ways of defining a straight
line or a segment. The defining points may be introduced
graphically (by indicating the right nodes) or textually (by
providing in the relevant dialog box fields the coordinates
of the points or node numbers). The Definition tab
assumes the form presented in the drawing on the right.
In the case of two-dimensional structures, a cut may be
defined in three ways:
by defining 2 points - once this option is selected, one
should introduce (graphically or by writing them in the
fields) the coordinates of two points that define the
beginning and end of the segment for which the
selected quantity will be presented
by defining a line parallel to axis and a point - once
this option is selected, the cut definition will consist in
selecting an axis of the global coordinate system
(axis X or Y) with respect to which the cut is to be
parallel, and in defining the distance between the
plane of the cut and the beginning of the coordinate
system. If the cut definition is performed graphically,
after selecting the required axis, one should indicate
the node in the defined cut.
by defining a point and direction - once this option is
selected, a straight line is defined instead of a
segment. Graphical definition boils down to indicating
two points; keyboard definition consists in defining a
point belonging to the line and the direction of the line
(dx and dy).
The defined cut will be added to the list of cuts available on the Cuts tab.
Cut Definition - Three-Dimensional Structure (Shell)
If one designs a three-dimensional structure, the plane of the cut must have a general
definition. It means in the majority of cases the necessity to define one point belonging to the
cut plane and a direction of the required normal vector. The plane may also be defined by
three points not belonging to a straight line.
The program provides many ways of general definition of a cut plane. The defining points may
be introduced graphically (by indicating the right nodes) or textually (by providing in the
relevant dialog box fields the coordinates of the points or node numbers). The Definition tab
assumes the form presented in the drawing below.
In the case of three-dimensional structures, a cut may be defined in three ways:
by defining a vertical plane going through 2 points - once this option is selected, the
definition of a cut will consist in determining the plane parallel to Z axis of the global
coordinate system (NOTE: it must always be the Z axis!); the cut plane definition consists
in determining the coordinates of two points that belong to the plane. If the cut definition is
performed graphically, it suffices - after selecting an axis - to indicate the node through
which the cut is to go.
structures.com
187
by defining a plane parallel to one of the axis planes and a
point - once this option is selected, the definition of a cut
will consist in indicating a plane of the global coordinate
system (plane XY, XZ or YZ), with respect to which the cut
plane is to be parallel, an determining the distance
between the cut plane and the center of the global
coordinate system. If the cut definition is performed
graphically, it suffices - after selecting an axis to indicate
the node through which the cut is to go.
by defining three points once this option is selected, one
should define coordinates (graphically or by writing them
with the keyboard) of three points not co-linear that
univocally define the cut plane.
The defined cut will be added to the list of cuts available on the
Cuts tab.
The Cuts tab of the Panel cuts dialog box presents all the cuts
defined for a structure. Three items of information are
presented for each cut:
display of the cut switched on/off for the structure (if the
option is active, the cut will be presented on the structure
together with the selected diagrams of the indicated
quantities)
color of the cut and the diagram presented for the cut
cut name.
In the Detailed, Extreme and Complex tabs, one may select
quantities to be presented in the form of diagrams over the
defined cut lines. The Parameters tab contains options that
allow one to select a layer parallel to the central surface of a
surface structure for which the calculations results will be
presented.
The Diagrams tab allows one to select the manner of
presentation of diagrams on a structure (see the figure below):
in the Diagram descriptions field one may decide how the
descriptions of selected quantities will be presented on the
diagram (none, horizontal and vertical)
in the Positive and negative values field one may decide
whether positive and negative values of the presented
quantity will be marked with different colors
in the Diagram position field, one may determine the
position of the diagram with respect to the structure.
NOTE: When diagram descriptions are presented along cutting lines, maximum and
minimum values are presented for the diagram, as well as the value of the
integer for the selected component along the cutting line length.
structures.com
188
5.6 Maps on Solids
The option is used to present maps (isolines) of displacements, stresses and strains obtained
during analysis of a volumetric structure. The results may be presented in the form of a map
or isoline with values of the presented quantity. The results are presented on external face of
a solid or selected elements. The dialog box shown in the figure below, which appears on the
screen if this option is selected, consists of four tabs: Detailed, Principal, Scale and
Deformation. The option is available from:
the menu by selecting the command: Results/Maps on Solids
the toolbar by pressing the icon: Maps on Solids .
Results for volumetric elements (solids) may be presented in the tabular or graphical form.
Available result groups include:
a) Stresses:
- in the local coordinate system defined by a user (globally for the entire model)
- in principal directions, if a stress tensor assumes the diagonal mode
- as reduced values according to the selected hypothesis
b) Strains
- in the local coordinate system defined by a user (globally for the entire model)
- in principal directions, if a stress tensor assumes the diagonal mode
- as reduced values according to the selected hypothesis
structures.com
189
c) Displacements
- are available in the local coordinate system defined by a user (globally for the entire
model)
- in the global coordinate system as total displacements, i.e. length of a displacement
vector.
The lower part of the dialog box contains the options allowing selection of a graphical form for
result presentation (Maps, Isolines, Isosurfaces). Activation of the Values option results in
automatic activation of the With Description option; then, the values in element centers are
presented without drawing maps or isolines. The following forms of graphical result
presentation may be selected:
isolines - results obtained for volumetric finite elements will be presented in the form of
isolines on a solid external contour
maps - results obtained for volumetric finite elements will be presented in the form of
maps on a solid external contour
isosurfaces - results obtained for volumetric finite elements will be presented as surfaces
- of the same values - inside a solid (identically as isolines on panels which show the run
of a selected quantity of the same value, isosurfaces show such a run within a solid);
number of surfaces is determined by the number of scale colors. (NOTE: if there is a
large number of colors, the duration of isosurface generation will increase)
Both, maps and isolines may be presented in the form of averaged values between elements.
Averaging of values (smoothing) may be performed by selecting one of the options from the
list available in the bottom part of the dialog box:
No smoothing
Global smoothing
Smoothing within a solid
Smoothing according to selection
Smoothing according to properties.
The remaining options located in the lower part of the dialog box include:
With Description - switches on isoline descriptions or displays values in element center
With Normalization - switches on automatic calculation of extreme values for a map scale
or isolines
With FE mesh if this option is switched on, then apart from a map of a selected quantity,
the program also presents generated finite element mesh
Open new window with scale displayed - results in opening a new window which displays
only selected elements and objects. The window is divided into two parts containing view
and a scale legend.
In the above dialog box, on the Detailed tab the following quantities may be selected for
presentation: stresses, strains and displacements. Only one of the results can be selected on
all tabs.
On the Principal tab stresses and strains may be selected for presentation. Only one of the
results can be selected on all tabs. Reduced values are presented according to the selected
hypothesis, which are available on the list located in the lower part of the dialog box.
Presently, the following hypotheses are available:
Huber-Mises
Tensor invariant I1.
First tensor invariant is determined according to the following formula:
structures.com
190
zz yy xx
I
3 2 1
1
Second invariant equals:
where:
p = I1/3 average stress.
The reduced value according to the Huber-Mises hypothesis equals:
The Scale and Deformation tabs contain options that have been described for diagrams
(bars) or maps (bars, maps). In addition, the Deformation tab includes the Deformation in
structure scale option if this option is switched on, then real values of structure deformation
will refer to structure dimensions; this option is connected with the Scale factor option in this
edit field the user should specify a factor by which values of structure deformation will be
multiplied; if the Constant scale option is activated, the scale will be selected for all the
presented diagrams (the option is convenient when comparing the results obtained for
different load cases, etc.); this option is connected with the Map scale on 1 cm (in) option in
this edit field the user determines the number of centimeters (inches) to which one centimeter
in a drawing corresponds.
The difference between the options: Map scale on 1 and Deformation in structure scale can
be seen while zooming in/out a structure on the screen; for the Map scale on 1 option, the
size of deformation remains unchanged, however, for the Deformation in structure scale
option, a deformation becomes adjusted to dimensions of a structure on the screen.
Similarly as for plate/shell structures, the X and Y axes are local and defined using the
DIRECTION option on the Detailed tab. This option allows one to define the main direction of
the local coordinate system which will be used (x axis). The result values for all elements are
calculated to this rotated coordinate system. The direction may be defined by entering any
vector defining the main direction when selecting the results for finite elements. The selected
vector is then projected onto an element which, in the end, defines the position of the local x-
axis. There is only one limitation - the main vector cannot be perpendicular to the element
(i.e. parallel to the local element z axis). If the user selects this direction, all results will be
equal to zero.
The sign convention for volumetric elements is presented schematically in the drawing below.
The convention is shown for stresses; the stresses displayed in the drawing have positive
signs.
structures.com
191
5.7 Solid Cuts
The option is used for presenting maps on cuts through solids. Results are shown on the cut
surface. The dialog box in the drawing below, appearing on screen after selecting this option,
consists of five tabs: Definition, Cuts, Detailed, Principal and Scale.
NOTE: The last three tabs are similar to the tabs provided in the Maps on Solids
dialog box. The Detailed tab contains additionally the Internal forces in the
table option (see below).
The primary assumption of drawing maps on cuts through solids is that that on all sections
maps are drawn for the same result quantity. It is a main difference compared to diagram
presentation on panel cuts, where for each diagram a different result quantity may be
selected.
The equivalent resultant forces available in the result table are calculated based on the
following formulas:
structures.com
192
... - (Macauley brackets) or the positive part operator expressed as:
For results dependent on the local direction in a finite element, the direction is determined
independently from the cut plane. For example, if a stress map Sxx is displayed, then the
stress direction x will be defined by the user independently from the cut plane.
Thus, maps on cuts are only a different form of drawing maps for a quantity selected in the
Maps on Solids dialog box. Therefore, it is possible to display a map on a solid external
contour or on internal cuts through a solid.
When defining a direction for cuts through solids the Selection of Direction dialog box looks
as shown in the drawing below:
In the above dialog box the user may determine a direction of the first tangent axis T1
according to a selected vector or direction of the main coordinate system.
structures.com
193
The coordinate system in the cross section used in the Robot program looks as follows (see
the drawing below):
- N axis normal to the cross section plane - x axis
- T1 first tangent axis to the cross section plane - y axis
- T2 second tangent axis to the cross section plane - z axis.
With the directions listed the following stresses are associated: NN, NT1 and NT2.
The following cut definitions are allowed:
by means of a vertical plane defined by two points; such a definition generates a vertical
strip limited by two points in the plane parallel to the global Z axis
as a cut parallel to the selected principal plane (XZ, XY, ZY) in the global coordinate
system, passing through a defined point
by means of a plane defined by three points in space.
To define a cut for a structure, the user should:
select the cut definition method
determine cut parameters
assign a name to the defined cut
structures.com
194
select the cut color
press the New button.
The defined cut will be added to the list of available cuts on the Cuts tab. The options located
on this tab allow selection of displayed cuts.
The List of Cuts field shows all cuts defined for a structure. For each cut three pieces of
information are presented:
switching on/off cut presentation for a structure (if the option is switched on, then the cut
with maps of a selected quantity is presented for a structure)
color
cut name.
5.8 Stress Analysis (Bars)
Once the results of a structure analysis is obtained the stress analysis can be performed to
evaluate stresses in the longitudinal sections and the cross-section of the selected structure
bar. The Stress Analysis - Bars option is available from:
menu by selecting the Results/Stress Analysis/ Stress Analysis - Bars command
layout field by selecting the RESULTS/STRESS ANALYSIS - BARS layout.
Once the STRESS ANALYSIS - BARS layout located in the Results group is selected, the
screen is divided into four parts (in the case of a solid section): the Stress Analysis dialog
box and three viewers in which stress maps are presented (in the bar cross-section and two
longitudinal sections: XY and XZ). In the case of a thin-walled section, the screen is divided
into two parts: the Stress analysis dialog box where the numerical results of stress analysis
are presented for the section and the graphical viewer presenting the maps of stresses.
In the Stress Analysis dialog box, the following tabs may be selected: Cross Section,
Section XY, Section XZ, Point, and Bar. In the case of a thin-walled section, there appears an
additional tab: Table, presenting the values of stresses in table form (for characteristic points
of the thin-walled section). The lower portion of the dialog box is the same for all the tabs
listed. To obtain the stress distribution presented in the three viewers located in the left
portion of the screen:
select the structure bar and load case for which stress maps will be presented
select the stress type (normal, tangent, reduced); the section force components included
in calculations of this stress (FX, FY, FZ, MX, MY and MZ) will be automatically selected
enter internal force values for the selected bar (it is possible to change the internal force
value obtained during structure analysis)
select cut plane location.
Once these parameters are chosen and the Apply button is pressed stress values of the
chosen section will be presented in the dialog box, stress maps in two longitudinal sections
and the cross-section will be shown in the three viewers located in the left part of the screen.
NOTE: If one has selected normal stress SIGMA X for presentation, the program will
present the neutral axis in the form of a dashed line in the cross-section of
the solid bar.
structures.com
195
Pressing the Calculation Note button results in activating the word processor in which the
calculation note generated for the chosen bar will be presented.
If the parameters located in the Stress Analysis dialog box have been changed, (e.g. the
load case, stress type, etc.), then pressing the Apply button results in determining the stress
values and their maps for the new parameters.
In the menu of the Stress Analysis - bars module, the following should be pointed out:
Cutting Planes - allows for a graphic definition of the point cutting plane. Once the Edit/Cutting
Planes option is selected the shape of the cursor changes. When in the
appropriate dialog box, the cutting plane may be defined by pressing the left
mouse button. The cutting plane coefficients will be automatically entered in
the appropriate fields (values x=, y= and z=).
Point Values - allows for a graphic definition of coordinates of a point in which stressed will
be marked. Once the Edit/Point Values command is selected (cursor shape
will change to a view-finder) the appropriate bar point should be pressed
with the left mouse button. When shifting between the consecutive windows
(cross-section, longitudinal sections), the settings on the Point tab change
(the appropriate plane is chosen, the appropriate coefficients are entered).
View Attributes - allows to define the stress map parameters (scale for views, colors, etc.). A
dialog box is made up of several tabs in which the user defines stress map
presentation parameters.
3D Projection - allows one to view a 3D presentation of a selected structure bar, accompanied
by the presentation of an indicated stress type. There appears an additional
structure bar view (see figure below) with an additional toolbar containing
options that allow one to move, rotate, zoom the selected structure bar. The
option is also available in the side toolbar (in the STRESS ANALYSIS -
BARS layout): .
The following extreme stresses calculated in the selected cross-section and longitudinal bar
sections will be presented in the above dialog box:

x
normal stress (maximal and minimal)
shear stresses
stress from torsion

i
stress.
structures.com
196
The coordinate value for which the extreme stress value was obtained is given for each of the
listed stress values.
The same values will
be presented on the
Bar tab; they will be
calculated for the
whole bar.
The following stresses
indicated in the
selected bar point will
be presented on the
Point tab:

x
normal
stresses (maximal
and minimal)
shear stresses
stress from torsion

i
stress.
The example of the STRESS ANALYSIS - BARS layout is presented above.
5.9 Structure Stress Analysis
Once the structure analysis is completed in the Robot Millennium system it is possible to
determine stress maps obtained for the entire bar structure. The Stress Analysis - Structure
option serves that purpose and is available by:
selecting the command Results/Stress Analysis/Stress Analysis - Structure from the
menu
selecting the layout RESULTS/STRESS ANALYSIS - STRUCTURE.
After selecting the STRESS ANALYSIS - STRUCTURE layout the screen displays a window
presenting the structure view, result table in the lower part and the dialog box managing
stress analysis. This dialog box is used for selecting stresses and determining the manner of
stress graphical presentation.
The result window (table) of the structure stress analysis contains numerical stress values
shown in the tabular form. All typical stresses and user-defined stresses may be presented.
Stresses for individual bars are presented in the form of extreme values for selected load
cases. At the table end the global extremes for individual stress types are displayed together
with the information about the bars and cases for which these extremes occur.
The Stress Analysis dialog box consists of the following tabs: Stresses diagrams, Scale,
Parameters
This dialog box allows selecting a user-defined stress; a set of basic stress types: normal,
tangent, Mises, Tresca can be used for that purpose. For each stress type a set of forces can
be selected to be taken into consideration during calculations. It enables assessing influence
of individual internal forces on the bar ratio.
The dialog box allows selecting a set of stresses presented in the form of diagrams on the
structure, scale of diagrams, diagram colors and parameters. The Scale tab contains options
that enable selection of colors for a stress map used in the 3D structure view presenting
stresses.
structures.com
197
The lower part of the dialog box is shared by all the enumerated tabs. In order to obtain the
stress distribution presented both in the structure view and in the tabular form, the following
should be done:
1. select a structure load case for which stress maps will be presented
2. if the All bars option (located in the lower part of the dialog box) is switched on, then the
stresses will be presented for all structure bars; if the Selected bars option is switched
on, then it is possible to select bars for which calculations will be performed and maps/
diagrams will be presented
3. select a stress type (normal, tangent, reduced); the components of internal forces taken
into account in that stress calculations (FX, FY, FZ, MX, MY and MZ) will be selected
automatically
4. indicate the stress type to be presented in the form of diagrams.
Once these parameters are selected and the Apply button is pressed in the dialog box, the
program performs calculations and presents the stress values on the structure bars, whereas
the tabular result window (table) displays the values of the appropriate stresses.
If the parameters in the Stress Analysis dialog box have been changed (e.g. a load case or
stress type, etc. has been changed), then pressing the Apply button results in determining
stress values and diagrams and/or maps for the new parameters.
The Stress Analysis - structure table presents the stress values for selected structure bars.
The stress types to be displayed in the table are selected in the dialog box that appears once
the Table Columns option from the context menu (the right mouse button) is switched on. It is
possible to present minimal and maximal values for all available stress types and for user-
defined stresses.
At the end of the table a set of extreme values for the entire structure is displayed; for a given
stress type the following data is presented:
extreme stress value
case, for which the extreme value occurred
bar in which this stress value occurs
location of the extreme value on the bar length.
Table support (sorting, selection, etc.) is analogous as for similar tables in the Robot
program.
3 View Stress maps - the option available from the Results / Analysis menu allows
presenting a structure together with section shapes and accurate detailed stress maps on
these sections.
structures.com
198
NOTE: Stress calculations for a structure - due to the complexity of the accurate
stress calculations for an arbitrary bar section (it refers mainly to the stresses
caused by a torsional force and to locating extreme stress values in a
section), the calculation process may be time-consuming. Owing to that, a
certain facilitation has been introduced which enables considerable reduction
of stress analysis duration. Each section utilized in the structure while using
the Robot program, is analyzed in the long-lasting manner only once, its
subsequent use (also during every other work session with Robot) does not
cause time-consuming operations analyzing the section, therefore, in practice
stress calculation time is reduced to minimum.
The figure below presents the STRESS ANALYSIS STRUCTURE layout for the example of
bar structure.
structures.com
199
5.10 Global Analysis - Bars
This option serves to present variability of a selected parameter (displacement, internal
forces) for all the bars of the currently designed structure. The option is accessible from:
the menu, by selecting the Results/Global Analysis - Bars command
the toolbar, by selecting the icon.
This option presents in one figure the maximum and minimum values of selected resultant
parameters for each bar element. Once the option is selected, there appears an additional
window on screen, where the user may select the parameters that are to be presented.
When they are indicated, the program creates a global diagram for the selected bars. The
figure below shows an example of diagrams and tables for internal forces and extreme
stress.
The global diagram of selected parameters for all bars is presented in the upper part of the
window shown below. There is a possibility of changing the form of this diagram. Clicking the
right mouse button within the diagram activates the context menu containing the Diagram
Types option. Five diagram types are available: line, column, bar, 3D column, 3D bar.
The lower part of the above-presented window contains a table containing the following data:
column 1 - the parameters selected for presentation (internal forces, stresses, design
parameters);
structures.com
200
column 2 - value of the lower limit; this is the value that allows one to determine the
minimum value of the selected parameter and identify the bars for which the minimum
values have been exceeded;
column 3 - value of the upper limit; this is the value that allows one to determine the
maximum value of the selected parameter and identify the bars for which the maximum
values have been exceeded;
column 4 - list of bars for which the limit values of the selected parameters have been
exceeded;
column 5 - list of bars for which
the values of the selected
parameters fall within the
range defined by the upper
and lower limit values;
column 6 - color selected for
presentation of a given
parameter
column 7 and 8 - respectively,
the minimum and maximum
value of the parameter(s)
selected for presentation,
determined for all bars in
a structure.
It is possible for the user to select the values to be presented in the table and in the diagram.
Clicking the right mouse button within the diagram activates the context menu containing the
Table Columns option.
structures.com
201
Once the option is selected, there appears on
screen the Parameters of Presentation
Windows dialog box where one can select the
parameters to be presented within the global
analysis of structure bars. The dialog box
consists of three tabs: Forces, Stresses and
Design. The figure below shows the Stresses tab
as an example.
The value of upper and lower limits for particular
parameters can be defined in two ways:
textually, by introducing the values of upper
or lower limit into the corresponding table cell
graphically: having situated the mouse cursor
in one of the table cells (upper or lower limit
of the selected parameter), one should go to
the diagram field and determine the limit
value there (on the diagram, there appears a
horizontal line that represents the required
limit value).
5.11 Detailed Analysis
The option is used to present detailed results (diagrams, tables) for selected bars of a
structure. It is accessible by:
selecting the RESULTS/DETAILED ANALYSIS layout of Robot Millennium program,
selecting the Results/Detailed analysis command from the menu
clicking the icon on the toolbar.
NOTE: Before the option is activated, the user should select a bar or bars in the
structure for which detailed analysis is to be presented.
This option provides the possibility of presenting detailed diagrams and numerical results in
table form for particular bars of a structure. Once the option is activated, the screen is divided
into three main parts (see the figure below):
detailed analysis dialog box where one can select the quantities to be presented and the
manner of their presentation
a table where the numerical calculation results will be presented for selected bars
graphical viewer, where diagrams for the selected quantities will be presented along the
relevant bars of the structure.
structures.com
202
The table displays the values of quantities selected for presentation: global displacements,
values of internal forces and stresses at the beginning and end of an element, as well at
intermediate points indicated by the user in the Detailed analysis dialog box. The table
consists of three tabs:
Values - the tab presents the values in selected points of quantities selected for
presentation
Local extremes - the tab presents local extremes of quantities selected for presentation
for each bar subject to detailed analysis
Global extremes - the tab presents global extremes of quantities selected for presentation
for all the bars subject to detailed analysis. If only one bar is subject to detailed analysis,
local extremes are equal to global extremes.
The user may define the position of intermediate points for which the values of selected
quantities are to be presented. This is done on the Division points tab in the Detailed
analysis dialog box.
The graphical viewer presents detailed
user-defined diagrams of displacements,
internal forces and stresses obtained for
the active load case. The diagrams are
presented along the length of relevant
elements. The values of forces are
presented according to sign convention
(see chapter 2.3). The figure below
presents an example of diagrams. If the
diagram labels are inactivated, the
diagrams show the points of maximum and
minimum values of the quantity selected for
presentation.
structures.com
203
The Detailed Analysis dialog box appears on screen when the Detailed Analysis option is
selected. The dialog box consists of five tabs:
NTM
Stresses
Reinforcement
Parameters
Division points.
The first two tabs (NTM and Stresses) allow the user to select the quantities to be presented
for the bars selected from a structure. The values of the selected quantities (displacements,
elastic ground reaction, internal forces and stresses) will be presented graphically - in the
viewer, and in the textual form - in the table. One may of course indicate several quantities to
be presented simultaneously for the bars selected from a structure. The figures below show
the NTM and Stresses tabs in the Detailed Analysis dialog box.
The options located on the third tab: Reinforcement allow presentation of calculation results
for theoretical (required) reinforcement of structure R/C members. Theoretical (required) or
real (provided) reinforcement area, reinforcement spacing (stirrup spacing), reinforcement
ratio, etc. can be presented. The bottom part of the dialog box contains the Show required
and provided values on one diagram option. If it is on, one diagram for one quantity (e.g. top
reinforcement) will show two values: theoretical (required) and real (provided) (e.g. theoretical
(required) top reinforcement area and real (provided) top reinforcement area). If the option is
off, all the values are presented on separate diagrams.
The next two tabs in the Detailed Analysis dialog box allow the user to determine the way
the selected quantities are to be presented in the respective diagram and table. The
Parameters tab controls the manner of presentation of diagrams displayed on the graphical
viewer. The Division points tab allows the user to indicate the intermediate points for which
the values of selected quantities are to be presented in the table.
structures.com
204
The figures below show the Parameters and Division points tabs in the Detailed Analysis
dialog box.
The Parameters and Division points tabs allow the user to control the way of presenting the
descriptions of quantities presented in diagrams (no descriptions, horizontal, vertical), while
the Positive and negative values field allows one to ascribe colors that differentiate positive
and negative values in the diagrams of relevant quantities.
The Division points tab allows the user to indicate the points for which the values of relevant
quantities will be presented in the table. If the N points along bar length option is checked, the
user may define in the appropriate field the number of points (the beginning and end node of
the bar included) distributed in a regular manner along the length of the bar for which the
relevant values will be presented. The default value of N = 2, which means that the table will
present the values of the quantities selected for presentation only for the bar beginning and
end. If N = 3, the program will add a point in the middle of the bar (dividing it into two equal
parts) and present the values of relevant quantities for the point. If the Characteristic option is
checked, the user will be able to define a point on the selected bar (by specifying its absolute
or relative distance to the bar beginning) for which the program should display the values of
the relevant quantities. The successive points selected for presentation will be accompanied
by the following descriptions:
AUTO - points generated automatically
USER - user-defined points
ZERO - characteristic points on the bar (places where the value of the selected quantity
equals zero and the extremes of the selected quantity).
The lower part of the Detailed Analysis dialog box (Division points tab) contains three
buttons:
Refresh - the button refreshing the list of points (if changed by the user) for which the
values of indicated quantities are to be presented
Delete - the button deletes points on a bar for which the values of indicated quantities are
no longer to be presented
Add - the button adds points on a bar for which the values of indicated quantities are to
be presented.
structures.com
205
If the Open a new window option is checked, there will appear a new window on screen
where the diagrams of quantities indicated in the Detailed analysis dialog box will be
presented.
5.12 Influence Lines
A moving load is defined by selecting a vehicle and determining its route over the structure.
The vehicle is treated as a set of forces with a defined direction, size and position. In each
step, the vehicle is moved from one position to another; the set of forces applied to structure
elements is created for each position. Therefore, the moving load case is treated as a set of
several static load cases (successive load case for the next position of the vehicle).
For a moving load case, the user may draw the route of the vehicle over the structure.
Moreover, one may display the loads originating in the vehicle and display the resultant
values, separately for each position of the vehicle or for the entire sequence of positions
(animation of the moving load and of the resultant values).
The results obtained for the moving load case may be
presented in two ways. The first method consists in
presenting the results of a static case for the moving
load position selected by the user. The user may
move the load step by step or take advantage of the
animation of the vehicle and the results for the moving
load. The second method consists in presenting the
changes of the value of a selected quantity in a
selected point, due to the movement of the load over
the structure, i.e. the presentation of the influence
lines for a selected quantity.
Now, we will present the possibility of creating
influence lines of a selected quantity. This is done by
Influence Line option that may be activated by:
selecting the Results /Advanced /Influence Line
command from the menu
pressing the Influence line icon from the
toolbar.
Once this option is selected, the dialog box shown
beside will appear on screen.
In the case of bar structures, only two tabs are available in the above dialog box, namely:
Nodes and NTM. While in the case of plate and shell structures, the following tabs are also
available: Detailed, Extreme, Complex and Parameters. The tabs allow one to select the
quantities for which the influence lines may be presented.
To present the influence lines for a selected quantity, one should:
define the range (from and to options); pressing the All button results in recognizing all
the defined positions of the moving load
determine the point for which the program will create an influence line of a selected
quantity; this is done by means of the following options:
structures.com
206
- elements - number of an element/bar, for which influence lines of the selected
quantity will be selected
- position - coordinate (relative) of the point for an element/bar for which influence lines
of the selected quantity will be selected
in the dialog box tabs, select the quantities for which an influence line will be created
for a plate and shell structure: determine the area for which influence lines will be created
press the OK button.
A new dialog box will appear on screen (see figure below), presenting the influence line for
the selected quantity. The newly-opened dialog box may be divided into two parts:
table presenting the numerical results of calculations; the following data will be presented
in the order of listing:
first column - the number of the bar (element) for which the influence line will be created,
point position on the bar (element), name of the moving load case, position of the moving
load, position of the moving load for it successive positions over the structure
two or three columns (depending on the structure type) - coordinates of the vehicle
position in the global coordinate system
further columns - values of the selected quantities for which the influence line will be
created.
graphical viewer presenting the diagrams of influence lines of the selected quantities.
The figure shows an example
of influence lines for force FZ
and the bending moment MY.
5.13 Reduced Results for Panels
The option allows one to obtain results for panels in a quick and simple manner. The results
may be used for other calculations, such as calculations of reinforcement that is to be used in
walls aimed to maintain the stability of a building that undergoes the operation of wind and
seismic impact. In order to carry out such calculations, one has to know the reduced forces
along different cross-sections.
WALL ELEVATION
REDUCED FORCES
structures.com
207
The cross-sections for which the reduced forces are to be calculated may have the following
orientation: horizontal - at the wall base (Section AA and Section BB), or vertical (Section CC
and Section DD).
In order to obtain a system allowing one to calculate reduced forces that would be both simple
for the user and quick in its use, the walls should be divided into rectangular panels according
to the scheme presented below:
The reduced forces should be calculated only for two-dimensional panels with the shape of a
convex quadrilateral. The reduced forces will not be calculated for the following types of
panels:
panels created by means of the following edit options: Extrude and Revolve
three-dimensional panels (curved)
panels that are not quadrilaterals
concave quadrilateral panels (not convex)
panels with variable thickness.
The values of reduced forces should be obtained for 6 cross-sections that are described in
the scheme below.
Position of allowable cross-sections for reduced results
In order for cross-sections 1, 3, 4 and 6 to be defined correctly within panels (no defined
coordinates), one should move the cross-sections by the Delta value with respect to the main
panel nodes N1, N2, N3 and N3. The delta value may equal the tolerance value applied in the
generation of the calculation model.
structures.com
208
One obtains the following point positioning:
Point M1 is the center of segment N1-N2.
Point C is the center of segment M1-M3 or is the center of segment M2-M3
Point M1 is the point of intersection between the line (M1,M3) and the panel edge.
CALCULATION OF REDUCED FORCES
The coordinate system for results is identical as the coordinate system for results used for
panel cuts.
The beginning of coordinate system is located in point Pr (Reference point) that is identical
with M1, M2, M3, M4 or C, depending on the required cross-sections.
Points Po and Pe will be identical with Points N1 and N4 for cross-section 1.
Points Po and Pe will be identical with Points M1 and M3 for cross-section 2.
Points Po and Pe will be identical with Points M2 and M4 for cross-section 5.
Calculation of NRx and MRz
Sign convention for normal forces and reduced
bending moments (NRx and MRz)
e
o
P
P
xx X dyy . N NR
e
o
P
P
xx Z dyy . yy . N MR
MRz moment has a positive value when it results in tension of fibers on the positive side of
axis yy.
Calculation of other components of results
e
o
P
P
xy Y dyy . N TR
structures.com
209
e
o
P
P
xx Y dyy . M MR
Sign convention applicable to MRy moment is identical as that applicable for Mxx moments:
MRY moment has a positive value, when it results in tension of fibers on the positive side of
the local axis z for panels.
e
o
P
P
XX z dyy . Q TR
Calculation of reduced stresses sigma and tau
The components are necessary to check the reinforcement introduced in RC bracing walls.
2
6
C
Z
C
X
L e
MR
L e
NR
sRo
2
6
C
Z
C
X
L e
MR
L e
NR
Re s
)
e
L ( e
TR
tR
C
Y
2

where:
e - panel thickness
Lc - section length.
Reduction e/2 along section length allows one to recognize reinforcement cover at the ends of
bracing walls.
Calculation of section length Lc
The information is necessary to check the reinforcement introduced in RC bracing walls.
po pe po pe po pe
P P dyy L
z z y y x x
e o
P
P
C
e
o

2 2 2
where:
Xpo, Ypo and Zpo are absolute coordinates of point Po
Xpe, Ype and Zpe are absolute coordinates of point Pe.
Calculation of panel height Ht
The information is necessary to check the reinforcement introduced in RC bracing walls.
For horizontal sections 1, 2 and 3, height Ht will be calculated on the basis of the following
formula:
4 3 4 2 2 1 6 5 4 N N , ' M ' M , N N max L , L , L max H C C C t
For vertical sections 4, 5 and 6, height Ht will be calculated on the basis of the following
formula:
3 2 3 1 4 1 3 2 1 N N , ' M ' M , N N max L , L , L max H C C C t
The results in the tables are presented in the same way as in other tables (e.g. in tables of
reactions, displacements, forces, etc.)
structures.com
210
The header of the first column contains, depending on the indicated load case, the selected
loads:
Panel / Cut / Case
Panel / Cut / Case / Component
Panel / Cut / Case / Mode.
Cut description is carried out with respect to the numbers of the main panel nodes: N1, N2,
N3 and N4.
Cuts 1, 3, 4 and 6 receive the following descriptions: N1-N4, N2-N3, N1-N2 and N3-N4.
Cuts 2 and 5 receive the following descriptions: N1~N2-N3~N4 and N1~N4-N2~N3.
5.14 Time History Analysis / Advanced Analyses Diagrams
and Tables
Results of time history analysis may be presented in graphical form of diagrams, maps,
structure deformations. The diagrams may be presented for the envelope or for each of the
time components. After selecting the additional load case (+/-) generated for envelopes, the
program displays envelopes. After selecting the main case, the results of a single component
with particular time steps become available. After selecting the Results/ Advanced/ Time
History Analysis - Diagrams option from the menu, the program displays the dialog box shown
in the figure below:
The graphical results of time history analysis are presented in the form of time-dependent
diagrams of a selected quantity in an indicated case of time history analysis. The diagrams
are shown in a new graphical viewer: "Time History Diagrams". It presents diagrams and the
table containing their description.
The dialog box presented in the figure above contains the following options:
in the Diagram definition field:
buttons:
structures.com
211
Add - pressing the button results in opening the dialog box for defining a new function
A diagram is created for the selected, single resultant quantity. The quantity must be
selected in one of the available tabs of the Diagram definition dialog box (NOTE: the
number of available tabs depends on the type of the analyzed structure: frame, plate,
etc.). Additionally, one has to determine the element (bar) or node for which the diagram
will be prepared. In the case of surface elements, the values of forces and stresses are
read in the centers of the elements.
The name of function diagram is created automatically: the name components are
presented below:
(case_number)_(quantity_name)_(option_name)_(element_or_node_number)/(position_a
long_bar).
The diagram name may be changed. One cannot define a function diagram if none of the
quantities is selected from among those presented in the above dialog box, or if no
diagram name is defined.
Modify - pressing the button results in opening the dialog box for defining a function
(allows one to change function name or its value that is currently selected in the drop-
down list)
Delete - pressing the button results in deleting the definition of the function currently
indicated in the list
the bottom part of the dialog box contains two panels: one contains the defined diagrams
(Available diagrams), the other contains the diagrams selected for presentation
(Presented diagrams). Between the panels, one finds standard button for transferring
table contents:
> - pressing the button results in transferring a selected diagram to the right-hand panel
>> - pressing the button results in transferring all diagrams to the right-hand panel
< - pressing the button results in deleting a selected diagram from the right-hand panel
<< - pressing the button results in deleting all diagrams to the right-hand panel.
The panel presenting the defined diagrams contains two types of functions: firstly, all the
functions defined by the user and contained in the drop-down list located in the top part of
the dialog box; secondly, the time functions defined as given for time history analysis in
the Analysis Types dialog box (they are by default transferred from load case definition)
Open in new window - switching on the option results in opening a new graphical viewer
with the relevant diagrams.
After defining diagrams and transferring them to the Presented diagrams field in the Time
history analysis dialog box, one may display a viewer with the selected diagrams by
pressing the Apply button. The figure below shows an example of a diagram for a case of
time history analysis.
structures.com
212
The top part of the viewer contains diagrams of indicated quantities, superimposed on one
drawing. The values of abscissa are the values of time variable. The bottom part of the viewer
contains a table with the description of diagrams and the values of extremes for the selected
quantities.
Placing the cursor in the table and followed by pressing the right-hand mouse button opens
the context menu that contains the Columns option. Selecting the option opens the dialog box
that allows one to select the quantities to be presented in the table located below the
diagrams of time history analysis.
The context menu contains also the following options (in the Diagrams - properties menu):
Display main grid lines - switching on/off the presentation of the main grid lines in the
diagram for the case of time history analysis
Display intermediate grid lines - switching on/off the presentation of the intermediate grid
lines in the diagram for the case of time history analysis
Automatic range - switching the option on results in adjusting the ranges on the
coordinate axes of the diagram to the range of time variability of the indicated quantity
User-defined range - allows one to define a time span for which the program will present
a diagram for the case of time history analysis.
For the case of time history analysis, apart from the main load case, the program creates two
auxiliary load cases, containing the (+) top and (-) bottom envelope. Selecting the main load
case makes available the results for particular components of the complex case.
NOTE: Due to the fact that a large number of results may be arrived at for a large
number of time steps, it is recommended to reduce the content of the opened
tables of results by means of the options located in the Filter results tab in the
Analysis types dialog box.
If no single complex case of time history analysis is selected, the table will show results for
auxiliary cases of the top (+) and bottom (-) envelope.
If, however, a single complex case of time history analysis is selected, the results are
available for particular components. The first table column presents the following data:
Node Case Component Time (s)
for example
1 Time history
analysis
2/100 0.0
The tables of results for time history analysis for bars and surface elements, the relevant
quantities are presented in the same way as in the table of nodes. The first table column
contains the number of case component and the step of the time variable.
The results of time history analysis may also be presented after selecting the
Results/Advanced/Diagrams option from the menu. The option enables defining and
presenting diagrams for cases of non-linear (elasto-plastic analysis of members), time history
and PushOver analyses (in the menu of the non-linear analysis table this is the option:
Results/Non-linearity\Plasticity/Diagrams, whereas in the menu of the pushover analysis table
this is the option: Results/Advanced/Pushover Analysis - Diagrams). Diagrams allow the user
to present any result values (e.g. displacements, internal forces, stresses) collected in
successive steps/increments of non-linear and PushOver analyses or in successive time
increments for time history analysis. Results may be displayed as a function of successive
increments (iteration steps or time steps) and as a function of other quantities. Diagrams may
be displayed for a single load case or for several selected cases. If different types of load
cases are chosen (the selection includes e.g. cases of non-linear, time history as well as
PushOver analyses), diagrams may be displayed only for one analysis type.
structures.com
213
NOTE: The user may present on a diagram n different quantities (shown on the
vertical Y axis) as a function of only one quantity provided on the horizontal X
axis.
NOTE: If several load cases have been selected, then each of the defined diagrams
is displayed for successive cases (in other words, n different diagrams are
generated); the range on the X axis is determined by the <min,max> values
of all the load cases, as regards the range on the Y axis, it is defined by
<min,max> of all the load cases (correspondingly to each scaling type).
For the elasto-plastic analysis there is a table available containing basic information about
parameters and results for this analysis type.
For members with the elasto-plastic section identical analysis results are accessible as for
members with standard sections: deflections, internal forces and stresses at any point over
the member length. Afterwards, elasto-plastic members may be verified and designed in the
program using the calculations for steel codes.
Apart from that, the section plastic ratio is available for bars with elasto-plastic section. It is
the ratio of the area of plasticized section part to the total section area. It assumes values
from 0.0 (for an elastic section) to 1.0 (for a fully plastic section). The plastic ratio is presented
in the form of maps on bars.
The Data for plastic process history dialog box, opened from the result table for elasto-
plastic analysis (the table opens after selecting the option: Results/Non-
linearity\Plasticity/Plastic Process History Table), consists of several tabs: Plasticity, Forces
and Stresses (the Forces and Stresses tabs are identical as for member quantities).
The Plasticity tab is shown in the drawing below.
structures.com
214
In this dialog box the following quantities may be selected for table presentation: load factor
and plastic ratio.
A list presented in a table may be additionally filtered; the filtering criterion is a state of the
section plastic process (value of the plastic ratio):
beginning of plastic process value of the plastic ratio equals 0.0
plasticity - value of the plastic ratio equals 1.0
value of the plastic ratio is always greater than the value defined (a value entered in the
edit field has to be greater than zero and less than 1.0).
structures.com
215
6. STRUCTURE DESIGN
6.1 Steel/Aluminum Design
Designing elements of steel structures can be carried out in the Robot Millennium system
according to the requirements of several steel codes. The below presented list contains the
steel codes currently available in the program:
American code LRFD and LRFD (new edition)
American code ASD
American code EIA
Canadian code CAN/CSA-S16.1-M89
British codes BS 5950 and BS 5950:2000
Eurocode 3 (there are several codes available with the following National Application
Documents: French, British, German, Belgian, Dutch, Swedish and Finnish)
French code CM66
French code Add80
Dutch code NEN6770/6771
Italian code CNR-UNI 10011
South-African code SABS 0162-1:1993
German code DIN 18800
Spanish code MV 103-1972 (NBE EA-95)
Swedish code BSK 99
Norwegian code CNS 34.
Russian code SNiP-II-23-81.
The design code for aluminum structure elements is also available within the program; this is
Al76 French code. The way a design of aluminum structure elements is performed is identical
to the one used during a design of steel structure elements.
The structure design process is divided into three stages; first the structure geometry and
applied loads are defined, then internal forces and displacements are calculated and finally
code requirements are verified and successive structural members designed. The design is
applied to either single members or groups of members and calculations can be executed in a
similar way.
Depending on the selected steel code, the contents of the list of parameters defined before
the member design can vary, but the basic definitions remain the same no matter which code
has been selected.
The following definitions apply:
structures.com
216
MEMBER Single structure member to be verified or modified in the module. The most
common bar types are columns, bolts, purlins and bracing.
A member used during verification/design can be defined as a single element or
a sequence of consecutive elements creating a column, beam, etc.
GROUP Member list. A group of structural members for which the same section is
assigned. Once the verification/design is completed, a section appropriate for all
members in the group (regardless of differences in internal force values for these
members or design parameters) will be selected. Groups are defined in order to
limit the variety of sections in the designed structure.
Once the STEEL\ALUMINUM DESIGN layout is selected, the screen will be divided into three
parts: graphic viewer for structure presentation and two dialog boxes: Definition and
Calculations.
The Definition dialog box contains two tabs: Groups and Members (see the drawing below).
Once the members and groups are defined, verification for a single member or a group will be
carried out. Clicking on the Parameters button located on the Members tab results in opening
the Parameters dialog box (the contents of which depend on the selected steel code). In this
box, code parameters such as buckling length, buckling parameters, lateral-buckling
parameters, etc. can be defined, determined by the selected steel code (see the drawing
below).
Among the interesting options to be found in Robot, one can count the possibility of
automatic design by means of parameterized tapered sections. The option is available by
pressing the Parameterized sections button located in the Definitions dialog box. The
option is available both for steel sections and for timber sections. The shape of the dialog box
depends on the material selected for the designed bar (steel or timber). The user has two
section types at his disposal (the relevant selection is performed in the Section types field in
the right-hand part of the dialog box):
steel sections timber sections
I-section
box section (rectangular pipe)
rectangular section
double rectangular section
The Section definition field of the Parameterized sections dialog box allows one to define
the dimensions of steel or timber sections. To start designing a new section, one should press
the New button. There will appear a new line where one should type the relevant dimensions.
Pressing the Delete and Delete all buttons results, respectively, in deleting the highlighted
section from the list or deleting all the sections from the list. There are two types of sections:
tapered section (dH)
tapered section (auto).
In the case of timber sections, one has also the invariable option at his disposal. For the
defined sections to be recognized in calculations, one should "move" them to the Sections
recognized in calculations field.
structures.com
217
An example of the Parameters dialog box for the LRFD steel code is presented on the figure
below. The name of the selected member type is given in the member type field (any name
may be entered here). The length of a bar may be defined in the fields: Bar Length Iy or Iz.
There are two ways of defining it:
once the Real option is selected, the value entered here is interpreted as the length
once the Coefficient option is selected, the value is interpreted as the coefficient by which
the actual length should be multiplied to get the correct value i.e. entering the value 0.25
means that the appropriate length is equal to 1/4 of the actual length.
In the Buckling Length coefficient field, the buckling lengths of a member may be defined in
both directions. The actual length of a member (or, the sum of the component element
lengths) is automatically entered in the appropriate fields.
The buckling length coefficient depends on the support conditions of the member end nodes
in the buckling plane. The buckling length of a member may also be defined in the Buckling
Diagrams dialog box that may be opened by pressing the icon representing the selected type
of member-buckling model. Typical member support patterns are located here; once one of
them is selected; the coefficient value will be accepted or calculated automatically.
The present version of the program allows one to define the parameters of brackets that
reduce both buckling length (in both directions) and lateral buckling length (separate for upper
and lower flange). Thanks to this, one may easily read the sets of bending moment in
characteristic points of a bar during analysis. The options provided in the program allow one
also to define buckling length coefficients and/or lateral buckling length coefficients of
segments between brackets.
The option is available by pressing the icon located in the Buckling Diagrams dialog
box.
The program provides also the
possibility of defining the
buckling coefficient of the main
bar on the basis of the
parameters of adjoining bars. It
is available by double-clicking on
one of the icons
/ /
located in the Buckling
Diagrams dialog box. The
parameters of adjoining bars
may be defined in the below-
presented dialog box (here, for 3
adjoining bars).
In the above dialog box, essential information pertaining to the bar adjoining the main bar may
be given. The following values should be defined for the main bar in appropriate fields:
the number of the consecutive structure bar (in the second column, the section of the
selected structure bar is automatically entered); the user may enter the bar number in the
appropriate edit field or perform a graphic selection on the screen.
position (placement) of the bar in the structure; two situations are possible: a vertically
placed section or a horizontally placed section .
In the case of several codes (ADD8, Eurocode3, NEN6770/6771) another parameter is
included: the support method for the other end of the adjoining bar. The available support
types for the other end depend on the selected steel code requirements.
The program provides also the possibility of defining manually the parameters of adjoining
bars. In the respective dialog box, one can define these parameters manually.
structures.com
218
In the Lateral Buckling Parameters field, one can select options used during lateral buckling
verification: lateral buckling type, load level and lateral buckling length coefficient. Pressing
the appropriate icon opens the dialog box for defining the appropriate parameters.
The Lateral Buckling Type option is used to define the appropriate lateral buckling code
parameters depending on the static model of the member. According to the code
requirements, one of the models considered in the code must be accepted. The models in the
shape of icons represent the appropriate support positions from the code. Pressing the last
icon signifies resignation from including lateral buckling effects in the calculations.
For a proper calculation of the lateral buckling coefficient, it is necessary to define the lateral
buckling length. Due to the possibility of affixing the upper or lower flange separately and the
occurrence of compressive stresses in either flange under various load cases, two lateral
buckling lengths are available. The proper lengths are controlled by the coefficient by which
the base length of the member should be multiplied to obtain the lateral buckling length. The
length Iz is taken as the base length. The coefficient value may be entered directly or an icon
of the typical support conditions case may be selected for which the coefficient will be
automatically chosen.
Once the More button is pressed, an additional dialog box appears on the screen where one
may define the load parameters and the section parameters. Once the Service button is
pressed, there appears a dialog box where one may define allowable displacement limit.
The name of the selected bar is given in the Member Type field. The bar length may be
entered in the Member Length ly or lz fields.
This may be done in two
ways:
once the Real option
is selected, the
entered value is
interpreted as the
length
once the Coefficient
option is chosen, the
value is interpreted
as the coefficient by
which the actual
value should be
multiplied to obtain
the appropriate
length. E.g. entering
a 0.25 value means
that the appropriate
length is equal to 1/4
of the actual length.
For a simultaneous definition of several members of differing actual lengths, e.g. additional
supports equally spaced, the second method mentioned is very convenient. If the set
parameters are to be saved as a category, entering the length this way is essential. The
buckling length coefficient depends on the end support condition of the bar nodes in the
buckling plane. The buckling length may also be defined in the Buckling Diagrams dialog
box opened by pressing the icon representing the selected buckling model type. Typical
schemes are found here; once one is selected, the coefficient value will be accepted or
calculated automatically.
The icons in the dialog box are divided into two groups: the first one contains typical (code)
methods of member support and corresponding values of the buckling coefficients, whereas
the other contains icons of options used for the calculation of the buckling coefficient for
columns of multi-story frames. The colors of the icons represent: red (A) - calculations for
sway frames, blue (B) - for non-sway frames.
structures.com
219
The buckling is always considered in the calculations if a compression force appears in the
member even if it is negligible in comparison to the other internal forces. The program does
not perform analysis determining if buckling effects should be disregarded or not on its own. If
the user wants to eliminate buckling effects from the calculations, the last icon must be
chosen. It represents the option to disregard buckling in the calculation process.
Options used during the lateral buckling
verification: parameters Lb and Cb may
be selected in the Lateral Buckling
Parameters field. Pressing the
appropriate icon for the Cb parameter
opens the dialog box for defining the
appropriate parameters.
In the dialog box presented beside,
calculation options for steel members or
groups of members are to be selected.
The Option of Verification field allows
the user to select:
Member Verification - verification according to the member list based on consecutive
and independent calculations for each member. The procedure is based on considering
intermediate points on a member along with the load case present, verifying that it is
suitable by checking against the worst case scenario according to the relevant codes of
practice. A number of points are considered during calculations and a list of load cases is
specified. In other words, verification is based on examining whether certain sections
(accepted before internal force calculations) meet the code requirements. Such selection
determines whether the member will be considered satisfactory, unsatisfactory or
unstable.
Group Verification - group verification is based on consecutive and independent
calculations (see Member Verification) for each member in the group. Material properties
set for the group are taken into account.
Group Design - group design is based on examining the previously adopted set of
sections determined by Group Definition and eliminating those that do not meet the code
requirements. Consecutive sections are eliminated until the first section meeting the
requirements is found. The described process is carried out separately for each family of
sections belonging to the analyzed group. Code calculations are performed for each
section at consecutive intermediate points on the member, consecutive load cases,
consecutive elements of the given member and all members belonging to the group. If the
given section does not meet the code requirements for a certain intermediate point, the
load case or element of the member in the group is eliminated and the next section from
the list is chosen. This process continues until all sections from the list are eliminated. To
start calculations in the design mode at least one group has to be defined. Design may be
carried out for many groups (the process is run for each group separately).
If the Optimization option is switched on and the Options button is pressed, there appears an
additional Optimization options dialog box on screen that allows one to select the following
optimization options to be used during calculations of member groups:
weight - switching this option on results in recognizing section weight in optimization, i.e.
the program will look for the lightest section in the group from among the sections that
meet the code-defined criteria
structures.com
220
maximum section height - switching this option on results in recognizing maximum section
height in optimization; the user may define the section height in the edit field located to
the right of the option
minimum section height - switching this option on results in recognizing minimum section
height in optimization; the user may define the section height in the edit field located to
the right of the option
minimum flange thickness - switching this option on results in recognizing minimum flange
thickness in optimization; the user may define the flange thickness in the edit field located
to the right of the option
minimum web thickness - switching this option on results in recognizing minimum web
thickness in optimization; the user may define the web thickness in the edit field located to
the right of the option.
The lower part of the dialog box contains the Calculations for the entire set of sections option.
Switching it on results in activating during calculations the procedure that searches the entire
section database for the optimal section (this is especially important when the database
contains sections not arranged in the ascending order, i.e. each successive section is bigger
than the previous one).
The lower part of the Calculations dialog box contains the Configuration button, pressing
which results in opening the Configuration dialog box used to define the parameters applied
during the verification of a steel member. One may define the following calculation
parameters in this dialog box:
calculation points; they can be determined in two ways:
1. by defining the number of points along a member length (the points are evenly
distributed along a member length) - the Point number option
2. by defining the coordinates of the characteristic points; in order to do that one should
switch on the Characteristic points option and press the Options button; it results in
opening the Calculations in Characteristic Points dialog box (the points of maximal
values of internal forces, etc.)
efficiency ratio defines the coefficient by which the plasticity resistance limit will be
multiplied (increasing/decreasing the limit of plasticity)
maximum slenderness; if the option is switched on, the slenderness of a member is
verified. Moreover, one may determine the allowable value of member slenderness.
In the middle part of the dialog box one may choose internal forces to be taken into account in
member code calculations; switching on an option (e.g. Fx) results in the Fx force being
considered in calculations. Moreover, one may select the units for the presentation of results
of member design process. The results may be presented in the units used in the indicated
steel code, or in the units used in the Robot system.
The lower part of the dialog box contains the list for selection of a load case (dead load), for
which the defined displacements will be treated as structure initial deflections. The option
Take the deflections from the following case into consideration must be switched on then.
The lower part of the Calculations dialog box contains two fields: Loads and Limit State. In
the first one, the following options are located:
load case list - field for displaying the load cases taken into account during calculations.
The load case number can be entered therein.
load case selection - opens an additional dialog box (Case Selection) in which load
cases to be considered during calculations can be chosen.
Calculations can be run for ULS and SLS.
structures.com
221
Clicking on the Calculations button results in steel bar verification or design, according to the
parameters set in the Definition and Calculations dialog boxes. Once the calculations are
completed, the Short Results dialog box will be displayed on the screen. The dialog box
consists of two tabs: Results and Messages. The latter tab presents comprehensive
information regarding warnings and errors that occurred during structure member
verification/design. Clicking on the section shown on the short list on the Results tab will result
in displaying the Results dialog box.
There are some differences concerning the result display for various calculation options:
Verification of the List of Members - only one line for each member or group of members
is displayed according to the list of members of member groups. Sections that meet the
code criteria are marked with the symbol, while those that dont meet the code criteria
are marked with .
Member Group Design - three consecutive sections from each family of the selected
section group are displayed. The section in the central line meets the code criteria.
Section names proceeded by the and symbols indicate those cross-sections which
either dont meet the code requirements or meet the requirements with excessive
reserves. The profiles that meet code criteria are marked with the symbol, while those
that do not - with the symbol. The unstable sections are marked with one of the
following icons: or . The first one is applied to an unstable member or group of
members, while the second one indicates that a member or a group of members is
unstable and the efficiency ratio is larger than 1.0.
Member Group Optimization (the Optimization option is active in the Group Design) - the
Short Results dialog box contents are the same as for group design. Additionally, at the
beginning of the appropriate line the symbol is displayed indicating the optimum
section (if it exists). Below, one can see a view of the Short results dialog box for
designing member groups with the optimization option active.
Calculations can be also carried out for user-defined section forces (not calculated by the
program itself). This can be done using the Manual Calculations option that is available from
the Analysis / Bar Structure Design / Steel\Aluminum Members Design menu. Member
verification/design can be carried out.
An example dialog box with simplified results is presented below.
structures.com
222
Once the Calculation note button is pressed, an additional dialog box will appear on screen.
Here one can select the type of the calculation note to be presented.
If the Table printout option is selected, the program will
print a simple calculation note containing a table with the
basic information concerning the designed or verified
members or groups of members (the note will be
presented in the same form as in the Short results
dialog box). On the other hand, if the Full printout option
is selected, the calculation note will contain all the code-
defined conditions checked during
calculations/verification of members or groups of
members. The calculation note will be created for
selected members or groups of members. The user may
select members or groups of members in the List edit
field (all members of groups of members are contained in
the field by default).
Selecting the Table screen capture option means that the table with the basic information
concerning designed or verified members or groups of members - presented in the Short
results dialog box - will be screen-captured (the user should define the name of the screen
capture in the Label edit field). It will be possible to use the table in printout composition. The
name of the table screen capture will be available on the Screen Captures tab in the Printout
composition dialog box.
6.1.1. Detailed Analysis - Eurocode3
The option allows one to carry out detailed analysis of I-section members, rolled or welded.
The detailed analysis of a member can be activated by pressing the Detailed button located
in the Detailed result dialog box for the EC3 code.
structures.com
223
The program allows one to carry out the following calculations on the basis of the
requirements defined in EUROCODE 3:
Web stability for shearing
The program analyzes each panel (the space between neighboring ribs) at 11 points to
find the greatest value of the shearing force. For the identified force, the program carries
out verification according to paragraph (5.6)
The results are presented for each panel at the points where there appeared the most
straining stresses.
Web stability for compression (chapter 5.7 of the code)
Verification of the web under a concentrated force is carried out only when the force is
applied to a point where there is no rib (the web is not ribbed)
The program carries out an analysis aimed to determine if the force or reaction results in
web compression. The compression effect results from a force (reaction or an adjoining
bar) in the following cases:
- a force (reaction or an adjoining bar) at the top and the negative force/reaction value
- a force (reaction or an adjoining bar) at the bottom and positive force/reaction value
Otherwise, the program does not carry out calculations (the force is disregarded)
Calculations are carried out for all load cases defined by the user in the Automatic loads
field
If there are several concentrated forces (within one load case), the program automatically
sums them up
In the case of a combination or code combination, if there are several forces originating
from different load cases, the program sums them up while recognizing at the same time
the relevant combination coefficients
The results are provided for each force that results in compression of the web without
ribs.
NTM interaction (chapter 5.6.7 of the code)
The program analyzes each panel at 11 calculation points
Calculations are carried out for all load cases defined by the user in the Automatic loads
field
Results are shown for the point of a panel where there appears the most straining stress.
Compressed flange stability (chapter 5.7.7 of the code)
The program verifies compressed flange stability at 3 equidistant points (flange beginning,
center and end)
The program checks the geometrical condition of the compressed flange stability
according to the formula (5.80).
Transversal ribs stability
The check is carried out at the points where stiffeners are located
If a concentrated force is applied directly to a rib, the force itself will be taken for
verification. If no concentrated force is applied directly to a stiffener, the program
calculates the force that compresses the stiffener and does so on the basis of formula
(5.63). If the concentrated force results in stretching the stiffener, then calculations for the
stiffener are not performed
Multiple forces applied to a stiffener are combined in an analogous manner as in the case
of concentrated forces applied to a web without stiffeners.
Once the Detailed analysis option is run, the program reads all the concentrated force for the
load case that turned out to be decisive during the verification if the member complies with the
code. If, for instance, the load case DEAD1 turned out to be most exigent during member
structures.com
224
verification, then, all the concentrated forces covered by the load case will be read by the
program. If, however, a combination or code combination turned out to be decisive, the
program will read all the concentrated forces covered by the simple load cases included in the
combination. The governing load case is automatically introduced into the Automatic loads
edit field. The automatically-recognized concentrated forces are visualized in the graphical
viewer, together with the force number and its coordinates. The coordinate system is located
by default in member beginning.
A user can verify if the concentrated forces are recognized correctly in the Concentrated
forces tab.
The program is set by default in the automatic mode (the Loads / Automatic option). It means
that the calculations will recognize the load cases formerly defined by the user. All the
external and internal forces will be read for the current load case(s), and only the forces will
be recognized during calculations.
If the user wants to introduce a new load case, he must enter the manual mode.
The Detailed analysis dialog box is divided into several parts:
the top part of the dialog box contains a schematic representation of the member
containing member dimensions; the following icons are located in the bottom right-hand
corner of the drawing:
switching on/off of the presentation of ribs, rib numbers, panel numbering
and distances between ribs
switching on/off of the presentation of the forces operating in the top flange
of the member, their numbering and the distance between the forces
switching on/off of the presentation of the forces operating in the bottom
flange of the member, their numbering and the distance between the forces
switching on/off of the presentation of supports/adjoining bars located in the
top flange of the member, their numbering and the distance between
supports/adjoining bars
switching on/off of the presentation of supports/adjoining bars located in the
bottom flange of the member, their numbering and the distance between
supports/adjoining bars
the part containing four tabs: Stiffeners, Forces, Supports/Bars and Internal forces
the fields Verification and Loads.
The Stiffeners tab
Once the Detailed analysis is run, ribs are calculated automatically at the following member
points:
at the points where support have been defined
at the points where concentrated forces have been applied
at the points where adjoining bars have been identified
if the distances between bracings are excessive and do not meet code-defined
requirements.
It is possible to modify the parameters of the automatically-defined ribs. All the automatically-
defined ribs may be modified, added or deleted at will.
The Forces tab
A user can verify if the concentrated forces are recognized correctly using the Forces tab.
The program is set by default in the automatic mode (the Loads/Automatic option). It means
that the calculations will recognize the load cases formerly defined by the user. All the
external and internal forces will be read for the current load case(s), and only the forces will
be recognized during calculations.
If the user wants to introduce a new load case, he must enter the manual mode. There is no
possibility in the automatic mode to change the position of load application along an element,
neither is it possible to modify its value.
structures.com
225
The Supports/Bars tab
If a user-defined support has been identified in the beam in question, the support will be
automatically introduced into the list of supports and drawn in the graphical viewer. During
calculations, the program will verify the beam, recognizing the information concerning the
reactions in the support.
(NOTE: the assumption is correct, if the reaction is perpendicular to the longitudinal
axis of the beam).
For instance, for a beam with pinned supports and a concentrated force P applied at the
center, the program will identify two support at beam ends and it will recognize the reaction
value R = P/2 in its calculations.
If no supports are defined along the beam, the program will verify if a vertical bar adjoins the
beam in question (a bar that may potentially serve as a support). If there is an adjoining bar,
the program suggests a support at this point. The value of the reaction transferred to such
support will be calculated on the basis of the analysis of the shearing force at this point.
In all other cases, the program suggests by default the position of supports at beam ends.
The program automatically identifies the places where there are adjoining (subordinate)
beams, that transfer loads to the beam in question. During calculations, the program finds the
force transferred by the bar(s) by analyzing the distribution of the shearing force over the
beam in question. For instance, while analyzing a cantilever beam with the load P applied at
its end, the force being transferred from a subordinate beam, the program will identify the
place where the force has been transferred and calculate the force P on the basis of the
shearing force diagram at this point.
The Internal forces tab
In order to define a new load case, one should select the Manual option available in the
Loads field. From now on, the user may introduce arbitrary concentrated forces, define ribs in
arbitrary locations and determine internal forces, corresponding to the defined load. The
values of internal forces should be defined in the Internal forces tab.
If the user has already entered the Manual mode, the calculation will be carried out only for
the manually defined load case. It is not possible to run calculations for both an automatically-
defined load case and a manually-defined one.
SELECTED REFERENCES (DESIGN OF STEEL STRUCTURES)
AKBAR R. TAMBOLI, Steel Design Handbook - LRFD Method, The McGraw-Hill Companies
Inc., 1997
CHARLES G. SALMON, JOHN E. JOHNSON, Steel Structures - Design and Behavior, Third
Edition by HarperCollins Publishers Inc., 1990
JEAN MOREL, Calcul des structures metalliques selon lEUROCODE 3, Eyrolles, 1994
JEAN MOREL, Structures Metalliques - Guide de calcul CM66 - Additif 80 - Eurocode3,
Eyrolles, 1999
Steelwork Design Guide To BS5950: PART 1: 1990 - Third Edition, Volume 1 - Section
Properties, Member Capacities & Volume 2 - Worked Examples, The Steel Construction
Institute in association with: The British Constructional Steelwork, 1992
T J MAC GINLEY & T C ANG, Structural Steelwork - Design to Limit State Theory, Second
Edition, Reed Educational and Professional Publishing Ltd, 1987,1992
IOANNIS VAYAS, JOHN ERMOPOULOS, GEORGE IOANNIDIS, Anwendungsbeispiele zum
Eurocode 3, Ernst & Sohn, 1998
WARREN C. YOUNG, Roarks Formulas For Stress & Strain - Sixth Edition, The McGraw-Hill
Companies Inc., 1989.
structures.com
226
6.2 Design of RC Structure Elements
The purpose of the following modules: R/C beam design, R/C column design, Foundation
design, Continuous footing design and Deep-beam design is to define, calculate and
design concrete beams and columns in a structure (the module allowing R/C plate design is
described in chapter 6.6). The current version of the program allows design of structural
elements according to:
American codes ACI 318/99 and ACI 318/99 (metric) - beams, columns, spread footings
British code BS 8110 - beams, columns, spread footings
Eurocode 2 (Belgian NAD, Italian NAD) - beams, columns
French RC codes B.A.E.L. 91 and B.A.E.L. 91 mod.99
Singaporean code CP65 available modules as for BS.
The modules mentioned above can be used in two ways. They can be treated as an integral
part of the Robot Millennium system (with a link to other modules which are responsible for
structure definition and data transfer) or as an independent module (i.e. stand-alone) for R/C
structure element design only.
During structure design the user can easily enter, display and change the structure and
design parameters. If the obtained results do not meet the users criteria, calculations can be
repeated for different values of design parameters or for different cross sections.
The following chapters explain how to deal with modules for R/C design and address both
approaches.
If modules for R/C structure design are used as a part of Robot program, then having
calculated the forces operating within a structure and having shifted to the stage of designing
particular structure elements, a dialog box shown in the figure below appears on the screen
(this is the dialog box for R/C beams). The shape of the dialog box depends on the module
that has been activated (R/C beams or R/C columns.
In case of R/C beams it is possible to get to the R/C beam design module after selecting a
beam or a group of co-linear spans by calling the Analysis / R/C beam design command from
the main menu.
Then the program transfers the relevant loads to the beam automatically. The loads are
displayed in the Loads dialog box.
After entering the beam design module, there appears a dialog box allowing one to determine
the type of loads to be transferred: Simple cases or Manual combinations. In case of the
Polish code, there appears an additional field that allows one to define the participation of
structures.com
227
variable long-term loads, required for the calculation of deflection. If one selects the Simple
cases option, the programs transfers the loads for which the program internally creates code
combinations according to the regulations prepared for R/C structures, available in the CFG
folder in the file with *.rgl extension.
If one selects the Manual combination option, calculations are performed for the combinations
defined in the Robot program. Apart from that, there appears the list of all manual
combinations and the user may select them.
The Grouping type field contains options that allow automatic grouping of elements according
to certain criteria:
according to story - once this option is selected, the program - based on the structure
geometry - divides the structure into stories and on their basis creates levels in the
calculation tree in the beam and column modules ascribing automatically elements to the
relevant level
according to geometry (in the current version the option is available only for columns) -
once this option is selected, elements of the same geometry are treated as one
calculation element; afterwards, it is designed for many combination groups resulting from
loads acting on individual elements - in consequence, all the elements are designed for
the most unfavorable arrangement of loads.
The bottom part of the dialog box contains the Supports of the grillage beam button.
Pressing the button results in opening the dialog box. The option is used to determine which
of the adjoining elements are to be treated as supports for the indicated beam. The settings
have direct bearing on the type and shape of reinforcement in the beam in question.
The Supports of the grillage beam button is active when there are other beam-type
elements adjoining the beam in question. The bottom part of the dialog box displays a table
containing a list of adjoining elements (bar number accompanied by a section label).
Supports in the form of columns are identified automatically and there is no possibility of
switching it off.
In case of R/C columns, it is possible to get to the column design module taking account of
loads in column nodes after selecting a beam or a group of columns and by calling the
Analysis / Design of RC Structure Elements / RC Column design command from the main
menu.
Then the program transfers the nodal loads to the column automatically. The loads are
displayed in the Loads dialog box.
After entering the column design module, there appears a dialog box allowing one to
determine the type of loads to be transferred: Simple cases or Manual combinations.
Apart from the load values and natures, the Group field is filled up in the table of loads. It
contains the number of the bar from which a load is transferred
If a group of columns is to be designed together (which means that calculation result in one
column type for all the columns of the group), one should carry out group selection of the
columns and call the R/C column design option. Then, the dialog box for loads displays as
many load groups as there are selected columns. After calculations one obtains for each load
group a resultant column, capable of resisting the load belonging to each group. Identical
column geometry (section and height) and identical column support conditions are the
requirements to be met for the group to be calculated.
The modules used for design of RC structure elements also provide access to the
reinforcement calculator (the option is activated by selecting the following option from the
menu: Structure/Reinforcement/Reinforcement Calculator
structures.com
228
The bar diameters and reinforcement areas are given in the units chosen in the Preferences.
The Calculator allows calculation of the following values (d denotes the reinforcing bar
diameter):
reinforcement area:
(for example:
7*d 12 = 7.92 cm2
7*d 12 + 5* d 16 = 17.97 cm2
7*d 12 + 5*d 16 + 8*d 10 = 24.25 cm2
... )
the required number of reinforcing bars (for example 44/d14 = 29 bars)
the required number of reinforcing bars of a predetermined diameter (e.g. 18 and 12 mm)
with an additional condition given that the number of bars of both diameters is
approximately equal (for example: 44 /d 18 /d 12 = 12*d 18.0 + 12*d 12.0)
the required number of reinforcing bars of a given diameter (e.g. 18 and 12 mm), so that
the bars with the diameter of 12 mm make up a predetermined percentage of all bars
(e.g.: 44 /d 18 /d 12 %25 = 16 * d 18.0 + 5 * d 12.0)
the difference between the given area (i.e. 44 cm2) and the sum of the given reinforcing
bar areas (e.g.: 44 - 5* d 12 = 38.35 cm2).
6.2.1 RC Beam Design
The R/C beam design module allows for definition, calculation and design of continuous
beams (both single and multi-span). Load cases may consist of vertical concentrated forces,
uniform loads or additional support moments. Rectangular and T-sections are admissible (in
addition, various beam slab connection types are taken into account).
The R/C beam design module can be called up in the following way:
select the R/C beam design from the structure type vignette (see chapter 2.1) - the
module will work as an independent program (stand-alone) with no link and data transfer
with other Robot Millennium parts
once the structure is defined, select (by highlighting in the graphic viewer) the appropriate
list of bars creating the beam and choose the R/C beam design command from the menu.
The BEAM layout will be opened and geometry and member loads along with
corresponding results will be loaded into the code module. The screen will be divided into
two parts: the viewer displaying the designed beam and the viewer for the presentation of
the entire structure.
structures.com
229
The description of the R/C beam design process presented below applies to the second
method of calling the R/C beam design module; the main differences between the two
approaches will be pointed out.
Once the module is chosen, the selected beam structure will appear in the upper part of the
screen. Its geometry, applied loading and static analysis results will be loaded to the R/C
beam design module (whereas in the case of the module working independently the user
must define the geometry and loading and run static analysis). The beam geometry and loads
can be modified by using the following options:
beam section geometry - available by selecting the Geometry/Section Type command
from the menu or pressing the Section Type icon:
elevation dimensions of the beam span - available by selecting the Geometry/Dimensions
command from the menu or pressing the Dimensions icon:
loads - available by selecting the Geometry/Loads command from the menu or pressing
the Loads icon: .
Section geometry can be
defined/modified in the dialog box
presented below. To modify the
span(s) select the ones, whose
section dimensions are to be
changed. Selected spans will be
highlighted.
Rectangular, I-sections and T-
sections are admissible (in
addition, various beam-slab
connection types are taken into
account). Dialog box contents
(parameters) vary according to the
selected type. The dialog box
beside contains only the General
Parameters tab for rectangular
sections. Similar parameters are
available for the T-section and I-
section types. Once the T-section
with lower-positioned flanges is
selected, the dialog box will
contain two additional tabs: Slabs
and Cuts.
The program assigns automatically names to sections of RC beams/columns. The first letter
B or C corresponds to beam or column, while the following one determines the shape of the
section. It is followed by the characteristic dimensions. For instance, BR 30x50 denotes a
rectangular beam section, for which b=30 and h=50.
Elevation dimensions for consecutive beam spans can be modified in the Elevation
Definition dialog box. As in the case of cross section dimensions, the dimension
definition/modification applies only to the selected span.
Once the Load option is selected the LOAD layout of the Robot Millennium system (in the
case of a version integrated with other modules) or appropriate tables to define the loads (in
case of the module working as a stand-alone programs). The selection of the Loads option
allows the definition of loads applied to an R/C beam. This is done by:
left-clicking on the Load Definition icon or,
selecting the Structure / Loads command from the menu
choosing BEAM - LOADS.
structures.com
230
The dialog box shown below will appear on the screen.
To define R/C beam loads:
1. In Spans, enter the number of spans to which the load will be applied; if one writes A (all)
in this field, all the spans will be selected,
2. From the Nature list, select the load nature (dead, live, snow, wind, etc.),
3. From the Load Category list, select the type of the applied load (continuous,
concentrated, and surface).
4. Define the values of the selected load type (the icons in the upper part of the dialog box
present schematically available load types) and press the Add button.
The left side of the load dialog box depends on the load type selected. The schematic icon
representing the selected continuous or concentrated load type is displayed in the upper left
corner of the dialog box. Icons allowing the selection of the load type are located on the right
side.
In the case of a planar load, the Coordinates field contains the following options: alpha, beta
and y; the adjacent area contains the field allowing one to define the value of p load (load per
unit of area). This load is transformed into a distributed load. The following planar load types
are available: loads gathered from the plate surface between two beams. One should define
the distance y from which the loads are gathered and two types of loads gathered from the
plate surface is delimited by four beams.
Below the Load Type icons, is the Relative Dimensions field and, in the case of having
selected a continuous load type, the Dimension Chain option. If the Relative Dimensions
option is active - the symbol appears, then during the load position definition, the relative
point coordinates will be used from the interval <0,1> will be used. If the option is not active
then the point coordinates will be given in the units selected by the user. In the case of the
continuous load - the Dimension Chain option (if this option is active - the symbol appears)
is available. During continuous load definition the so-called dimension chain will be used. This
means that the continuous load values p2 will be applied at a point with the coordinate
(x1+x2) load value p3 at a point with the coordinate (x1+x2+x3) etc. If the option is inactive,
then the load p1 is applied at a point with the coordinate x1, load p2 at a point with the
coordinate x2 etc.
The fields Co-ordinates and Value fields are located in the middle portion of the dialog box. In
the case of a concentrated load, two other options, x1 and F or M are located here.
In the case of a continuous load, the Co-ordinates field contains the options: x1, x2, x3, x4.
The Value field contains the options: p1, p2, p3.
structures.com
231
NOTE: When a continuous load is selected, not all the fields allowing co-ordinate
load value definition are active.
For each beam the following parameters can be defined:
openings (by selecting the Geometry/Openings command from the menu or pressing the
icon: ) - a dialog box for definition of openings for a selected beam span will be
displayed.
story parameters (story level, fire resistance, cracking and exposure rating) - by selecting
the Analysis/Story Parameters command or pressing the icon - the options included in
this dialog box depend on the selected code of RC structure element design
calculation options (by selecting the Analysis/Calculation Options command or pressing
the icon ) - the options included in this dialog box depend on the selected code of RC
structure element design
reinforcement parameters (by selecting the Analysis/Reinforcement Parameters
command or pressing the icon ) - the options included in this dialog box depend on
the selected code of RC structure element design.
NOTE: Once any changes are made to the beam geometry or its loads, analysis
results become unavailable. To obtain appropriate results for the modified
beam, the structure needs to be updated (so the changes can be saved).
This is possible by selecting the Results/Structure Updating command or
pressing the icon . Once the geometry is updated, the entire structure has
to be re-calculated.
Beam calculations and design can be started once all parameters have been assigned. It can
be carried out by either selecting Analysis/Calculations option or pressing the icon .
Internal force envelopes (moments and shear forces) and displacements are calculated for
the defined loads.
To view design results the following Robot Millennium system layout needs to be called up:
BEAMS - RESULTS
BEAMS - REINFORCEMENT.
Once the BEAMS - RESULTS layout is selected, the screen will be divided into two parts: the
left part containing the Diagrams field for graphic presentation of results and the right one
containing the Diagrams dialog box. This dialog box allows you to display results in a
tabulated format and select values for which diagrams in the left part of the dialog box will be
presented (the options included in the Diagrams dialog box depend on the selected code of
RC structure element design). Among those values section forces for available limit states,
reinforcement areas and deformations can be found.
structures.com
232
Sample diagrams
for the multi-span
beam are
presented in the
figure.
Once the BEAMS - REINFORCEMENT layout is selected, the screen will be divided into four
parts: a field in which beam elevation along with calculated reinforcement is presented, a field
containing reinforcement in the beam section, a summary table with description of
consecutive reinforcement rods and the Reinforcing bars dialog box presenting parameters
of selected bars.
The program provides also several options allowing one to edit the reinforcing bars to be used
in an RC beam:
starter bars - the option allows the user to determine the parameters of the reinforcing
bars that connect a beam to a column. The option is accessible by selecting the Structure
/ Reinforcement / Starter bars command in the main menu or the Starter bars command
from the context menu
division of reinforcing bars - the option allows the user to define the points of division of
RC beams. It is accessible by selecting the Structure / Reinforcement / Divide reinforcing
bars command from the menu. There is also the Connect reinforcing bars option available
parameters of reinforcing bars - the option is used to present the parameters of
reinforcing bars, determined during RC beam design.
The following options are also available for all RC structure element design modules:
Visibility it is used to select reinforcing bars to be presented in the graphical viewer in
the calculated RC structure elements (command: Structure / Reinforcement / Visibility or
icon ).
One may chose to present the four main types of RC structure element reinforcement:
main reinforcement, transversal reinforcement, structural reinforcement and
reinforcement of openings (only for beams and deep beams).
Translation it is used to perform the operation of translation on the formerly selected
reinforcing bars in RC structure elements (command: Structure / Reinforcement /
Translation).
Stirrup spacing it is used for manual modification of the stirrup spacing in a beam. (the
Structure / Reinforcement / Stirrup Spacing command). NOTE: The option is available in the
menu only after stirrups are selected in a calculated RC beam.
structures.com
233
After completing calculations, the results can be presented in the form of calculation notes
(Results/Calculation Note option). The Robot Millennium system text editor containing data
on the designed beam and calculation/design results will be displayed on the screen.
Once the Results/Drawings option is selected or the icon is pressed, the Robot program
will activate the FINAL DRAWING layout presenting a working drawing for the calculated and
designed beam. The selected beam span(s) will be presented on the drawing. The working
drawing of a beam will be presented on screen in the form which corresponds to the adopted
drawing parameters (see chapter 6.2.5).
6.2.2 RC Column Design
The R/C column design module allows you to calculate, pre-dimension and verify R/C
columns. Axial forces and moments in both directions are admissible. The following cross
section types are available: regular (rectangular, round or regular polygon) and irregular (T-
section, Z-section, semicircle etc.).
The R/C column design module can be called up as follows:
select the R/C column design from the structure type vignette (see chapter 2.1) - the
module will work as an independent program (stand-alone) with no link and data transfer
with other modules. The Robot Millennium system is responsible for the structure
definition.
once the structure is defined, select (by highlighting in the graphic viewer) the appropriate
list of bars creating the column and choose the R/C column design command from the
menu. The COLUMNS layout will be open and geometry and member loads along with
corresponding results will be loaded to the code module. The screen will be divided into
two parts: the viewer with the designed column elevation display and the viewer with the
column section display.
The description of the R/C column design process (presented below) applies to the situation
when the module works as a stand-alone program.
The screen will be divided into two parts: a viewer containing the column elevation display
and the viewer with the column section.
Column definition is based on the following data:
column elevation dimensions - available by selecting the Structure / Dimensions
command or pressing the icon ;
type and dimensions of the column cross section - available by selecting the Structure /
Section Type command or pressing the icon , after which the dialog box presented
below will be displayed on the screen. The section type (rectangular, round, T-section, Z-
section, L-section, regular polygon, semicircle or quadrant) should be specified and
dimensions of the selected section type are to be entered (they will be displayed on the
schematic drawing in the upper right corner of the dialog box).
structures.com
234
The program assigns automatically names to sections of RC beams/columns. The first
letter B or C corresponds to beam or column, while the following one determines the
shape of the section. It is followed by the characteristic dimensions. For instance, CR
30x50 denotes a rectangular column section, for which b=30 and h=50.
column buckling model - available by
selecting the Structure/Buckling Length
command or pressing the icon , after
which a corresponding dialog box will
appear and buckling models for both
directions (Y and Z) can be specified and
selection of the sway/non-sway structure
type can be made. Once the icon indicating
the buckling model is selected, an
additional dialog box will appear (as shown
below) in which the appropriate column-
buckling model can be selected. The
options included in this dialog box depend
on the selected code of RC structure
element design.
Buckling models presented in the above dialog box depend on a selected RC code. They
are based on the following code recommendations:
ACI 318 nomograms in the notes concerning the point (in ACI 318-95 points 10.12;
10.13)
structures.com
235
BAEL since there are no code guidelines, they are based on nomograms included
in EC2 in the point Slenderness of Isolated Columns (in ENV 1992-1-1 (1991) point
4.3.5.3.5 figure 4.27 formula 4.60)
BS 8110 - point 3.8.1.6
Eurocode 2 Belgian NAD based on nomograms included in EC2 in the point
Slenderness of Isolated Columns (in ENV 1992-1-1 (1991) point 4.3.5.3.5 figure 4.27
formula 4.60),
PN-B-03264 - Annex C.
Values ascribed to the models are the simplified representation of typical cases.
In the last three options code formulas and nomograms are directly applied.
After double-clicking on the icons shown in the drawing above, the program opens the
dialog box where node rigidity may be defined.
applied loads - available by selecting the Structure/Loads command or pressing the icon
, after which a dialog box containing a table for column load definition will be displayed
on the screen. The following data is included: case name, nature, group, axial force value,
shear force value and bending moments applied to the column.
Moreover, there is a possibility to add automatically the loads coming from the upper
column and beam (a column may be positioned with respect to a beam; names of beam
supports are selected and columns are associated with a beam support successive
records being the reactions of successive simple cases are entered to the dialog box for
loads).
As in the case of R/C beams, the following parameters can be defined:
story parameters (story level, fire rating, cracking and exposure rating), by selecting the
Analysis/Story Parameters command or pressing the icon - the options included in
this dialog box depend on the selected code of RC structure element design
calculation options (by selecting the Analysis/Calculation Options command or pressing
the icon ) - the options included in this dialog box depend on the selected code of RC
structure element design
reinforcement parameters (by selecting the Analysis/Reinforcement Parameters or
pressing the icon ) - the options included in this dialog box depend on the selected
code of RC structure element design.
structures.com
236
Column calculations and design can be
started after defining all parameters. It
can be done either by selecting the
Analysis/Calculations or pressing the
icon . Once the calculations are
completed the dialog box shown below
can be displayed on the screen. The
ratio within column cross-section is
presented in the Intersection dialog
box shown below.
The upper part of the dialog box contains a list of all load combinations taken into account
during column design process. For a selected combination the following values are displayed:
column section with neutral axis, compression and tension area and corresponding safety
factors. After calculations, the worst load combination (design combination) is presented in
the dialog box.
NOTE: If the above dialog box contains identical combinations with different values
of internal forces calculated for these combinations, it means that the values
of internal forces have been calculated for different sections along the column
length.
To view design results the following Robot Millennium system layouts are to be called up:
COLUMNS - RESULTS
COLUMNS - REINFORCEMENT.
Once the COLUMNS - RESULTS layout is selected, the screen will be divided into two parts
(see the drawing below).
structures.com
237
This dialog box contains N-M interaction curves for a given load combination. The left part of
this box is used to display the 3D N-Mx-My interaction surface, whereas the right one is for
the N-M interaction curve. The latter is the intersection of the 3D-interaction surface and the
N-M plane containing currently analyzed load combinations.
Once the COLUMNS - REINFORCEMENT layout is selected the screen will be divided into
four parts: a field containing the column elevation display along with calculated reinforcement,
a field with reinforcement in the column cross section, a summary table with description of
consecutive reinforcement rods and the Reinforcing bars dialog box.
As in the case of RC beams, the program provides also the Parameters of reinforcing bars
option used to present the parameters of reinforcing bars, determined during RC column
design.
Additionally, apart from the Visibility and Translation options (which have been discussed with
reference to R/C beams), the column design module contains the Stirrups arrangement and
Dowel Bars options. The option is used for manual definition of the shape of transversal
reinforcement of a column or dowel bars (at the level of a column cross-section). The options
are applied after calculations of column reinforcement are performed.
After completing calculations, the results can be presented in the form of calculation notes
(Results/Calculation Note option). The Robot Millennium system text editor containing data
on the designed column and calculation/design results will be displayed on the screen.
designed column. The working drawing of a column will be presented on screen in the form
which corresponds to the adopted drawing parameters (see chapter 6.2.5).
6.2.3 Foundation Design
The Foundation design option allows the user to define, analyze, and design rectangular
spread footings as well as continuous footings (to be placed under concrete wall). Footings
can be either concentrically or eccentrically loaded. The footing stability and soil bearing
stress distribution are also checked.
The main features of the foundation design option are:
- customization of units, display formats, and material properties,
- possibility of defining soil parameters and footing geometry,
- interactive input of footing parameters,
- possibility of defining constraints e.g. position of a pier on the footing or footing and/or pier
thickness,
- possibility of defining as many load cases as required.
The foundation design includes:
- automatic footing dimensioning,
- verification of sliding,
- verification of overturning and minimum bearing (contact) area of footing surface,
- verification of footing thickness,
- dimensioning according to entered shape proportions or dimensions,
- accounting for ground water level,
- on-line visualization of results,
- evaluation and detailing of footing and pier reinforcing,
- summary of earthwork, concrete, formwork, and reinforcing quantities.
The foundation design module is divided into following system layout:
- definition (of geometry and type)
- load
- soil
- results
- reinforcement.
During the definition, the type and the basic dimensions of the footing and pier are defined.
structures.com
238
The user can choose between the spread footing and continuous footing to be placed under a
concrete wall, as well as spread or continuous footing on lean concrete. One may also specify
foundation geometry (rectangular spread or continuous footing, rectangular spread or
continuous footing to be placed under two columns, spread or continuous footing of a tapered
section). There is an option to specify the constraints/limitation according to which the design
would be performed (geometry optimization). The following limits are currently allowed: all the
geometric dimensions, shape of the plan of footing, type of eccentricity, offset of the column,
and the adjacent footing condition. The definition also includes the type of the connection
between the column/wall and the footing (see the dialog boxes below). In the case of
selecting a foundation on lean concrete, the below-presented dialog box presents additional
tab (Lean concrete) that allows one to define the geometry of lean concrete.
Together with the program, the soil database is provided; it enables direct application of
correlational relationships described in the appropriate code. Once the basic soil
characteristics are defined, the remaining parameters are calculated automatically in the
table. Basic soil parameters may be changed; fields in the soil table, except for the fields
containing basic parameters, are inaccessible.
The load system layout consists of the following windows: dialog box (to enter/modify the load
on the footing, see picture), graphical window (with the picture of the current footing), tables
(to enter/modify the current load), and schematic for the footing location.
structures.com
239
In the dialog box beside, one can
define the loads applied to a
foundation. The list of the
available load categories contains
two items: the foundation load
described above and the backfill
load. Several load natures are
available: dead, live, snow, wind,
and seismic. Three load types are
available: axial load, load with a
normal force, a bending moment
and a shear force, load with a
normal force, a bending moment
and a shear force acting in two
directions. Depending on the load
selected, appropriate edit fields
allowing to define the force values
will appear.
The direction, type of load (dead, live, wind, snow, seismic), and category (foundation,
backfill) can be specified. The load factors are determined on the basis of the type of the load.
In the soil module user can specify the soil conditions: all the necessary levels (with regard to
the reference level), for example ground water level, backfill height, pier level; type of soil and
its parameters.
To get the results, the FOUNDATION - RESULTS layout has to be chosen. It includes
graphical windows with a diagram of soil stresses and a plan of the footing, design
combinations, combination coefficients checks, and global coefficients checks.
The last layout in the foundation design module is the reinforcement section. It consists of
longitudinal and transverse sections, isometric view of the footing, and tables with the
characteristics of layout and type of reinforcing steel used in the design. All the concrete and
reinforcement characteristics can be set in Calculation Options.
As in the case of RC beams and columns, one may define the following parameters:
calculation options (by selecting the Analysis / Calculation options command or pressing
the icon)
reinforcement parameters (by selecting the Analysis / Reinforcement parameters
command or pressing the icon).
structures.com
240
Once all foundation parameters are determined, one may start calculations and design of the
defined foundation. One may do this in two ways: by selecting the Analysis/ Calculations
command from the menu or by pressing the icon. This results in activating the
FOUNDATIONS RESULTS layout of the Robot system. The computer monitor will be
divided into two parts: the graphical viewer showing the view of a spread and continuous
footing and the FOUNDATIONS RESULTS dialog box.
Design of a foundation covers:
checking the pressure under spread footing
checking resistance to sliding
checking resistance to overturning
checking the foundation uplift
recognition of seismic dispositions (checking the sliding and foundation uplift)
checking the effects of punching / shearing
determination of an adequate reinforcement in the spread footing and in the footing-
column connection
distribution of the determined reinforcement in the footing and in the footing-column
connection
determination of total quantities of concrete, formwork and reinforcement.
NOTE: During calculations of a footing on lean concrete, the program does not check
the conditions of resistance against sliding and overturning and conditions of
shearing and punching for the lean concrete.
After completing foundation calculations, the results can be presented in the form of
calculation notes (the Results/Calculation Note option). The Robot Millennium system text
editor containing data on the designed foundation and calculation/design results will be
displayed on the screen.
designed foundation. The working drawing of a foundation will be presented on screen in the
form which corresponds to the adopted drawing parameters (see chapter 6.2.5).
6.2.4. Continuous Footing Design
The Continuous footing design module allows calculation, initial design and verification of
continuous footings (to be placed under a group of columns).
Continuous footing design may be started after choosing from the selection vignette the
structure type (compare chapter 2.1) of continuous footing design the continuous footing
design module will operate as an independent program (stand-alone) without connection
(data exchange) to other parts of the Robot Millennium system.
The options included in this module operate in the same manner as in case of RC beam
design. The only new option is the Soils option. The option is used to define layers of the soil
located under a continuous footing. The option becomes available once:
the Soils icon is pressed
the Structure / Soils command is selected from the menu
the CONTINUOUS FOOTING - SOIL layout is selected.
structures.com
241
After defining basic soil properties, all other parameters are calculated automatically and
displayed in the table. The name field includes the list of predefined soils. Once one of them
is chosen, the table is filled with data.
The table presents only these soil properties which are applied during calculations for a
continuous footing.
The basic parameters of a soil may be changed; after accepting the new values, the
remaining parameters will be automatically calculated and displayed in the table. The table
fields, except for the fields containing basic parameters, are inactive.
The continuous footing module allows saving (the Save button enables it) a defined soil
profile to be used in the calculators. The profile is saved as MS Access (*.mdb) database.
NOTE: Manual modification of a file directly in the MS Access program is not
recommended as data necessary for correct operation of the application can
be easily deleted.
The soil profile contains all the data about soil parameters and may be freely transferred
between work stations and used in other modules of the Robot Millennium program and
calculators.
A soil located under a continuous footing may be divided into segments characterized by
different soil layers. It is illustrated in the figure presented below. The segment geometry is
defined by determining coordinates of the segment beginning and end.
6.2.5 Deep Beam Design
The Deep beams design module allows one to define, calculate, and design deep beams
(one- and multi-span) design is performed according to French BAEL code. It is
characteristic for a deep beam that the height of its cross-section is much greater than the
width of the cross-section. Deep beam definition is similar to the definition of an RC beam
(see section 6.2.1). The following loads may be applied to a deep beam: vertical concentrated
forces, continuous load, additional concentrated support moments. The loads may be applied
to the top or bottom surface of the deep beam. A beam may have a rectangular or T cross-
section (different types of connection between ceiling plate and the beam are allowable).
Design of RC beam may be commenced by selecting the Deep beam design from the starting
vignette (compare section 2.1). The deep beam design module will work as an independent
(stand-alone) program, not connected to other Robot Millennium system parts (no data
exchange). To define a deep beam, one should:
define the geometry of deep beam section (Structure / Section type command or Section
type icon)
define the elevation dimensions of a deep beam span (Structure / Dimensions command
or Dimensions icon)
define loads (Structure / Loads command or Loads icon)
structures.com
242
define openings in the designed deep beam, if need be.
As in the case of RC beams and columns, one may define the following parameters:
calculation options (Analysis / Calculation options command or the icon)
reinforcement parameters (Analysis / Reinforcement parameters command or the
icon).
Deep-beam calculations may be performed by means if two methods:
based on the recommendations included in the French BAEL code - static calculations
are performed according to the simplified method described in BAEL 91 code (annex E1);
reinforcement calculations are carried out according to the method described in BAEL 91
code (annex E5); limitations of the method: the maximum difference in height between the
neighboring spans equals 1 m, the distance between an opening and the deep-beam
edge cannot be less than the opening width; once deep-beam calculations by means of
this method are completed, results are presented in the similar form as in the case of the
results obtained for RC beams
applying the Finite Element Method - based on the deep-beam geometry the program
generates automatically a model for the finite element method. The length of a single
finite element side reaches approximately 20 cm. The first support is generated as a
pinned one, whereas the remaining ones are generated as roller supports allowing
displacement in the X direction. In static calculations the mechanism of automatic
generation of code combinations is applied. The theoretical (required) area of
reinforcement is calculated using the analytical method. The real (provided) reinforcement
is generated in zones that are defined - by default - identically as for the BAEL method
with the possibility of changing them later on. The reinforcement area in each zone is
assumed by default to be the maximum area from all the finite elements located within a
given zone (with the possibility of the area subsequent reduction). Once deep beam
calculations by means of this method are completed, results are presented in the form of
isolines (similarly as results for RC plates).
One of the features distinguishing deep beams consists in the possibility of reinforcing them
by means of wire fabrics. The Robot Millennium program provides a database of wire fabrics
that may be used for calculating reinforcement for deep beams. Once the Edit database
button is pressed (on the Wire fabrics tab of the Calculation options dialog box opened for
deep beams), there appears an additional dialog box (viewer) shown below. The viewer
presents data concerning the available wire fabrics.
structures.com
243
The wire fabrics viewer presented above is divided into two parts:
toolbar with icons
table presenting the data of the available wire fabrics.
The following data are presented for each wire fabric type: the first three table columns
provide wire fabric number, information concerning the decision of taking the wire fabric into
account during calculations (if the option is selected - - the relevant wire fabric will be
taken into account; if not - it will not) and the wire fabric name. The successive table columns
provide the following information on the wire fabrics: reinforcement cross-section [cm2/m],
reinforcement spacing and diameter [mm], data concerning bar ending parts and, if needed,
the lap splice.
6.2.6 Final Drawings
The final drawings of the reinforcement calculated for RC elements constitute a separate
Robot layout. This Robot layout contains specific options aimed at facilitating the
manipulation of drawings. The most important ones will be discussed here.
Normal view (menu: View) - when a drawing of any element is called, the program enters
automatically the drawing layout and opens the normal view. It is the general view of the
contents of the entire drawing page. It does not allow one to insert, delete or modify the
contents of the drawing. The mode under discussion is useful during the preparation of
arrangement and composition of drawings for the final print-out format
Page set-up (menu: View) - this is the mode that allows one to correct the arrangement
and size of elements of a drawing (the so-called viewport). Each element of a drawing is
provided with handles at the corners that allow one to carry out edit operations. Once the
modifications are completed, one should go to another display mode. Then, the program
will regenerate the drawing and adjust the drawing contents to the newly-defined sizes of
its elements.
Drawing components (menu: View) - this is the mode that displays the range of drawing
elements and their contents. By indicating a given element of a drawing (it gets
highlighted in red), one makes its contents ready to undergo editing process. Within the
active area, one may carry out the following operations:
change of the drawing scale and section position
editing of the text (after indicating the text with the cursor and highlighting it in yellow)
by calling the Edit text option from the context menu, available by pressing the right-
hand mouse button
deleting a text - after highlighting a text, one may delete it by pressing the DELETE
button
moving a text; after highlighting it, one should click the text, which changes the cursor
shape to an arrow, by means of which one may move the text within the drawing
element (viewport)
editing the dimension (after indicating the relevant dimension with the cursor and
highlighting it in yellow); by hooking the cursor at the end of a dimension, one may
change its length, together with the value describing the dimension. In the case of
dimension chains, the neighboring dimensions are changed together with the edited
one
deleting + moving the dimension lines (the option functions according to the same
principle as text editing). In the case of moving a dimension, one may only move it
parallel to the original position.
adding a line, circle, text or dimension.
NOTE: After completing the editing process, one should press the ESC button in order to be
able to go to editing of another drawing element (viewport).
structures.com
244
Final print-out format (menu: View) - the option allows one to view the currently displayed
drawings in a single common sheet of paper. Once the option is called, the program
switches to another operation mode. The default final format is set to A4. In order to
change the format size, one should call the Page setup option from the File menu and
indicate the required paper format. Then, all the loaded drawings will be arranged
automatically.
Automatic drawing arrangement (menu: View) - the option is responsible for automatic
arrangement of drawings in a big format and it operates in cooperation with another
option, described above. If the automatic drawing arrangement does not meet the user's
expectations, it is possible to switch the option off. After indicating a required drawing
(one should go to the Normal view) and with the left mouse button pressed, one obtains
the possibility to move the drawing.
NOTE: In order to situate a drawing precisely in place, one should carry out the translation
operation with the Ctrl button pressed.
Undo, redo (menu: Edit) - the options allow one to undo or redo the last operation. One
should remember, however, that they bring, as a consequence, the loss of the possibility,
for instance, of scaling the drawing elements (viewports) or inserting a summary
reinforcement table
Cut, paste (menu: Edit) - these are standard options, operating for the entire drawing. By
means of the options, one may cut any drawing and paste it in a different location, in a
different page. The operation is of particular use during the arrangement of drawings in
bigger formats, when the number of pages is larger than 1.
Drawing (menu: Insert) - if a drawing has been saved formerly as a project component,
the option allows one to call and insert a drawing or a list of drawings. One should
remember, however, that, after inserting a formerly saved drawing, it is not possible to
scale it, neither is it possible to include it during the creation of a summary reinforcement
table
New page (menu: Insert) - calling the option results in inserting a blank page with the
format defined in the Page setup settings.
Table (menu Insert) - calling the option results in inserting a table for the drawings in a
larger format. The table provided with the program (the default.lay file in the USR folder)
is an example of a table; it may be modified or the user may define a new table by means
of the PloEdit program
Summary reinforcement table (menu: Insert) - the option should be called after
completing the changes of drawing arrangement in big format. It functioning consists in
creating a summary reinforcement table, referring to the active drawings. When the table
is being created, the program carries out automatic renumbering of all the reinforcement
positions in the drawings.
PRINT-OUT IN LARGE FORMAT
In the case of a printer that does not support certain formats, the Page setup dialog box
displays the formats that are supported by the currently installed printer. The remaining
formats (not supported by the peripheral device) will be displayed, but their description will be
shown in gray fonts. It is possible to carry out drawing arrangement for the "gray" formats, but
the preview and printing will be proceeded by the appropriate message: "The selected format
is not supported by the active printer".
structures.com
245
If there are printer controllers installed that service a printer that is not attached (physically) to
the computer, the available formats defined by the installed peripheral device will be
described in red. The preview and printout will also be proceeded by an appropriate message.
Final drawings are presented on screen according to the parameters accepted in the
Drawing parameters dialog box. The option is used to select parameters of display, drawing
presentation and detail drawing presentation for RC structure elements. The option may be
run in one of the following manners:
by pressing the Drawing parameters icon
by selecting the Analysis / Drawing parameters command from the menu.
The options available in the dialog box are used to define parameters of display and
presentation of particular drawing fragments and the manner of behavior of drawings.
The dialog box consists of four tabs: General, Reinforcement description, Scale and
Reinforcement table.
The above dialog box allows one to select the general drawing template. The first letters of
the standard names of templates provided with the program refer respectively to:
bm - beams/continuous footings
bc - columns
bf - foundations
bs - slabs
bw - deep beams.
All the standard templates are located in the CFG folder created during Robot installation and
they bear the *.plo extension. In order to modify an existing drawing template or open a new
template, one should run the PLOEDIT program - the editor of drawing templates. The
program is installed during Robot installation.
Once a template is selected for the module allowing one to design elements of RC structures,
the right-hand part of the dialog box presents a preview of the template. Apart from template
selection, the tab allows one also to set the mode (manner) of creating a drawing. Selecting
the first mode (Open new) results in the presentation of only the drawing of the selected
concrete element or the list of elements. Each time a drawing is called again, the former
drawing is removed.
structures.com
246
Selecting the Add drawing to the list mode changes the functioning and operation of
drawings. If the option is selected, a drawing will not be removed after calling again a new
one, but it will be added to a common list. The option is designed to allow one to create
(compose) drawings of different elements (beams with columns, etc.) and to arrange them in
a common, large-format sheet.
The Reinforcement description tab allows one to determine the manner in which
reinforcement elements are described. Apart form the number of reinforcement position -
always switched on a drawing may display the following elements:
Number (of bars) + diameter - displays the number of identical bars and their diameter,
apart from the reinforcement position number
Length - if the option is switched on, the program displays information concerning the total
length of the selected reinforcement element
Spacing - in the case of beams, the spacing refers only to the splice reinforcement; in the
case of columns, the spacing refers to stirrups in the central part of a column; in the case
of foundations, the description of reinforcement spacing refers to the main reinforcement
of the foundation; in the case of deep beams and slabs, the description of spacing refers
to each reinforcement element in the form of a bar
The Longitudinal reinforcement description on the section option is created principally for
beams. In the case of columns and deep beams, the program always displays
descriptions of longitudinal reinforcement along a section; the option is not used in the
case of footings and slabs.
The options provided on the Scale tab allow one to impose the manner of scaling of particular
drawing elements:
if The same for all pages option is switched on, the same scale for entire pages of the
drawing of one element are imposed; for instance, in the case of a beam comprising
several spans that are presented each on a separate page, the option imposes identical
scaling for all pages containing the drawing of the selected beam
if The same for elevation and section option is switched on, the program draws the
element section and its view in identical scale
if The same for both directions on elevation option is switched on, the program will not
apply two different scales to present an element along its length and along its height.
The options provided on the Reinforcement table tab allow one to define parameters of
displaying the reinforcement table, i.e. to add or remove a selected column of the table.
6.3 Design of RC Bars
Modules for Beams, Columns and Foundations - included in the hitherto available version
of Robot allow one to determine the required reinforcement area and the spacing of
reinforcing bars within the cross-section of an concrete structure element.
The Members - required reinforcement (Design of concrete members) module allows one to
calculate a theoretical (required) area of reinforcement for selected members. The option is
accessible by selecting:
MEMBERS - REQUIRED REINFORCEMENT layout, included in the layout group titled
DESIGN
the Analysis / Design of RC Structure Elements / RC Member Design / Calculations
command from the menu.
structures.com
247
The module is available for the following codes:
Eurocode 2
Eurocode 2 (French NAD)
Eurocode 2 (Belgian NAD)
Eurocode 2 (Dutch NAD)
Eurocode 2 (Italian NAD)
Eurocode 2 (German NAD)
Eurocode 2 (Finnish NAD)
American code ACI 318/95 and ACI 318/99 (metric)
Canadian code CSA A23.3-94
British code BS 8110.
French codes BAEL91 and BAEL 91 mod. 99
Dutch code NEN 6720
Spanish codes EH91, EHE98.
Russian code SNiP 2.03.01-84
Romanian code STAS 10107/0-90
Singaporean code CP65.
MEMBER is the main calculation element in the Members - required reinforcement
module. Usually it is a structure element of a certain type e.g. RC beam or RC column.
Definition of a member type allows one to perform appropriate calculations of the theoretical
(required) reinforcement according to the relevant code requirements. In some cases a
member can be defined as a chain of successive members entered during the structure
definition. The option used for definition of RC member type in a structure is available:
from the menu by selecting: Geometry/Code parameters/RC member type
from the toolbar, by selecting the icon.
The process of member type definition in a structure is identical to the manner of defining
other structure attributes. The type of an RC member is correlated with the relevant RC code
that has been indicated as the basis for the design of structure bars. While working with the
given code, one can edit and use only the members defined according to the code. The
shape taken by the dialog box for defining RC member type is also correlated with the type of
the member to be defined. The dialog box is different when one defines a member type
belonging to the category of concrete beam, and its shape changes when one defines a
member type belonging to the category of concrete column.
Before starting calculations of reinforcement area for members one should first determine
calculation parameters. The option is available from:
from the menu by choosing Analysis / Design of RC Structure Elements / RC Member
Design / Calculation Parameters command
on the DESIGN/ MEMBERS - REQUIRED REIFORCEMENT layout, from a toolbar by
pressing the icon.
The process of calculation parameter definition for structure members is identical to the
manner of defining other structure attributes.
When the New set of calculation parameters icon is pressed in the Calculation parameters
dialog box, a dialog box consisting of three tabs appears on the screen:
General
Longitudinal reinforcement
Transversal reinforcement.
structures.com
248
In the above dialog box (as an example a dialog box for an ACI code is shown), parameters
needed for RC element design that are not connected with its geometry such as: steel and
concrete characteristics, used member types, covers, etc. are grouped. Remaining code
parameters dependent on geometry are defined in the RC member type dialog box. The
contents of particular tabs of the Calculation parameters definition dialog box are
correlated with the selected RC code.
In the dialog box shown in the figure above, on the General tab two main fields are located:
concrete parameters distinguished with a frame and additional parameters required in
calculations. On the Longitudinal reinforcement tab, steel parameters, reinforcing bar types
and cover to the edge or bottom reinforcement axis can be defined. On the Transversal
reinforcement tab steel parameters for transversal reinforcement, reinforcement type and its
parameters can be defined.
Starting theoretical (required) reinforcement calculation for RC member results in opening of
the dialog box presented in the figure below.
structures.com
249
When creating a structure model, the user defines geometrical parameters of concrete beams
and columns (buckling parameters, allowable values of deflection and displacements at
member ends). The code parameters of concrete structure elements (steel and concrete
parameters, types of reinforcing bars) are defined in the Calculation parameters dialog box.
The above dialog box contains the following options:
in the Calculation type field:
- Design
- Load capacity check (the option is currently not available).
in the Calculations for field, elements taken into account in calculations:
- members
- bar groups (the option is currently not available);
- Lists of elements taken into account during calculations can be defined in three ways:
- manually typing member numbers in an appropriate edit field
- opening selection dialog box by means of ... button
- indicating elements in the viewer containing a structure view
Design cases:
- code combinations
- case lists for each of the analyzed limit states;
A set of active fields differs depending on a selected code. The selection fields for
code combinations are active only when such combinations were previously created
and calculated. A case list may be entered in an appropriate edit field or in the
Selection dialog box activated by means of ... button.
Number of calculation points for beams; it can be defined in two ways:
- by determining a number of calculation points along a beam length (min. = 3, max.=
100)
- by specifying every what length unit calculations are to be performed every; as a
beginning point one adopts the point corresponding to the options accepted in the
RC beam definition dialog box.
Results for calculations of theoretical (required) reinforcement area for RC members are
available in table form; the results can also be presented in the form of diagrams along the
bar length (see chapter 5.1).
Once the process of RC member design is completed the RC Member Calculations: Report
dialog box, presented in the figure below, is displayed on the screen.
The above dialog box presents the following information:
structures.com
250
list of designed RC members
list of members for which the calculations have been performed correctly
list of members for which warnings occurred during calculations
list of members for which an error occurred in calculations
additional notes.
The Calculations performed for bars field displays the numbers of RC members that
underwent design; these must be the members such as RC beam or RC column, since these
are the only elements for which a theoretical (required) area of reinforcement can be
determined in RC members.
Next three fields available in the dialog box present short information concerning the run of
RC member calculations. They divide the members into groups of members for which the
calculations have been completed with the same result:
the first field Results of calculations for bars are correct displays numbers of members for
which no warnings or errors occurred during design
the second field Results of calculations for bars contain warnings groups the members for
which warnings occurred during RC member calculations. Note should be taken that in
this dialog box a warning is treated in a more general way than in the result table for RC
members. Thus, a warning may concern both exceeding the maximal reinforcement ratio
(in the table it is presented in red color - code requirements are not fulfilled) and
information that maximal allowable stirrup spacing has been applied (in the table, only the
information in the Remarks column occurs).
the third field Results of calculations for bars contain errors groups the members for which
errors occurred during RC member calculations; for these members the calculations have
not been performed. In the result table such members are indicated by entering the word:
error. Errors during calculations may result from:
incorrect member definition; the following data may be inconsistent: section, member
type and reinforcement parameters. All data should pertain to the same code and
parameters describing an RC member. It is not allowed to assign sections to the
member types that do not correspond to them; it also concerns sections of the same
geometry
code requirements that make calculations impossible (exceeding of allowable
slenderness, maximal shear force or moment).
If errors or warnings occurred during RC member calculations, then in the lower part of the
dialog box a message appears which informs a user how to obtain information regarding
errors or warnings that have occurred.
If at least one of the designed elements has been a beam bent with respect to the Z axis or a
beam bent biaxially, then in the lower part of the dialog box a message will appear informing
how the results for such elements are presented in the result table.
In the bottom part of the dialog box the Reinforcement change button may be located. It is
accessible only if calculations of RC member reinforcement have been performed according
to a code which enables calculation of deflections and if at least one of the bars has not been
verified due to deflection. Pressing this button opens the Change of reinforcement dialog
box. The option enables semi-automatic correction of the calculated reinforcement in
elements for which admissible deflection has been exceeded.
Presently, the program provides the possibility of deflection calculation (for the serviceability
limit state) in the case of the following RC codes:
American RC codes ACI 318/99 and ACI 318/99 metric
British RC code BS 8110
Eurocode 2 (with different national application documents)
French RC codes BAEL 91 and BAEL 91 mod.99
Polish RC codes PN-84/B-03264 and PN-B-03264 (1999).
structures.com
251
The top part of the dialog box contains a bar table; the following data is presented in
individual table columns:
bar number
information if a bar has been verified or not
name of adopted reinforcement parameters
current bar deflection
admissible deflection
proportion ratio of the current deflection value to the admissible deflection value.
The table enables selection (multiselection) of bars and sorting in columns.
In the middle part of the dialog box there are options grouped in the fields: Method of
reinforcement change and List of load cases. The Method of reinforcement change field
includes three options:
Proportional to required reinforcement area the ratio of required reinforcement is
increased by the calculated values in such a manner so that the proportions of top area
and bottom area are preserved
Change of required reinforcement area if in a given bar section the (top or bottom)
required area is a non-zero value, then it is increased by the calculated value
Change of number of reinforcement bars - if in a given bar section the number of bars (at
the top or bottom) is a non-zero value, then it is increased by the calculated number of
reinforcing bars.
structures.com
252
Depending on the option chosen, in the edit field provided under the above listed options, the
user should specify:
dA= ..... [%] increment of the required area ratio
dA= ..... [cm2] area increment
dn= ..... increment of bar number.
Values given in this edit field stand for increment of the appropriate quantity in relation to the
values already existing. Pressing the Apply button causes recalculating and saving
appropriate values for selected bars; if one bar has been chosen, then values available in the
table provided in the bottom part of the dialog box are updated.
The List of load cases field presents the list of load cases (the field is inaccessible) that have
been used during calculations of deflection for the serviceability limit state.
The bottom part of the dialog box contains a table with information concerning reinforcement
area for a selected beam (if in the top table several bars have been chosen, the table in the
bottom part of the dialog box is empty). Any value may be changed in the table. The table
comprises the following data:
positions along the bar length
top and bottom required (theoretical) reinforcement
top and bottom number of bars
ratio of (required) reinforcement
rigidity.
Note should be taken that:
if a new value of the required reinforcement ratio is specified, then new required
(theoretical) areas of reinforcement and new number of bars are calculated
if a new value of required (theoretical) area of reinforcement is specified, then a new
value of reinforcement ratio and bar number are calculated.
Calculations are carried out for areas resulting from a number of bars. After pressing the
Verify button calculations for selected bars are performed. Once the calculations are
completed, the program updates information in the table. If verification has proceeded
correctly, the icon in the table is changed; the list of elements for which conditions have not
been satisfied, is refreshed only on opening the dialog box, while working in the dialog box;
only the results for the existing list are refreshed correspondingly.
To change reinforcement of RC members, the user should:
perform calculations of required (theoretical) reinforcement for RC members; once they
are completed, the RC Member Calculations: Report dialog box is displayed on the
screen
press the Reinforcement change button in the RC Member Calculations: Report
dialog box; the Change of reinforcement dialog box opens then
select bar(s), choose the method of reinforcement change
press the Apply button
press the Verify button.
After calculations the user should check coefficients in the table presented in the top part of
the dialog box. These operations should be performed as long as necessary to obtain an
intended value of deflection for all the bars.
Theoretical (required) reinforcement of RC members obtained after calculations is interpreted
in the following manner:
structures.com
253
1. Longitudinal reinforcement
For biaxially bent columns with the rectangular, T-shaped, L-shaped or Z-shaped sections,
the areas of reinforcement should be interpreted as follows:
As1= As2 = Reinforcement along b
As3= As4 = Reinforcement along h
For columns whose cross section is described by: a regular polygon, circle, semicircle or
circle quarter, the areas of reinforcement should be interpreted as follows:
As1 = Reinforcement along b - evenly distributed along the edge
For rectangular beams bent biaxially:
As1= Bottom reinforcement As2 = Top reinforcement
As1= Bottom reinforcement (Z axis) As2 = Top reinforcement (Z axis)
For elements with reinforcement in both planes, the results should be interpreted as follows:
The areas are treated independently - they have no common parts in the corners. The corner
area of section reinforcement is assigned to the area resulting from bending in the Y plane.
structures.com
254
2. Transversal reinforcement:
theoretical (required) stirrup spacing (Stirrup spacing) - spacing of the defined stirrups
required for a given section
real (provided) stirrup distribution - the distribution which is assumed for a given section
once the element has been divided into N equal parts (defined earlier in the Calculation
Parameters dialog box) and once the distribution in each of the zones has been
calculated
transversal reinforcement type/ distribution - presents the stirrup type and distribution in
the number of zones defined earlier in the Calculation Parameters dialog box. The
reinforcement type is described by the number of bars and their diameter linked by the
letter f or a steel grade (according to the code). The designation 5f8 (4HA8, 4T8) denotes
four-legged stirrups made from the bars of 8. The description of the distribution includes
- for each zone - the number of stirrups and their spacing linked with multiplication sign,
the zones are linked with the addition sign. The following description:
20*4.0+10*8.0+20*4.0 denotes three zones of stirrup distribution: the first one and the
third one comprise 20 stirrups each spaced every 4.0 units of section dimension, the
middle zone comprises 10 stirrups spaced every 8.0 units of section dimension. The valid
units are the units according to which calculations were performed.
NOTE: In the result table a designing combination denotes such a combination for
which, assuming the reinforcement calculated on the basis of all the
combinations, the greatest section efficiency ratio is obtained. For thus-
understood designing combination internal forces are presented.
A designing combination shows the most unfavorable set of forces for a given section,
however, it may by no means be used in its design.
SELECTED REFERENCES (DESIGN OF RC STRUCTURES)
General Part
P.CHARON, Calcul des ouvrages en beton arme, Eyrolles, Paris 1986
V.DAVIDOVICI, Formulaire du beton arme, Le Moniteur, Paris 1996
J.EIBLE (ED.), Concrete Structures Euro-Design Handbook, Ernst & Sohn, Berlin 1994/96
J.G.MACGREGOR, Reinforced Concrete Mechanics and Design, Prentice Hall, New Jersey
1988
structures.com
255
EC:
A.W. Beeby, R.S.Narayanan, Designers' Handbook to Eurocode 2 Part 1.1: Design of
concrete structures, Thomas Telford, London 1995
BAEL:
J.PERCHAT, J. ROUX, Pratique du BAEL 91 Cours avec exercices corriges, Eyrolles, Paris
1998
H.THONIER, Conception et calcul des structures de batiment, Presses de lecole nationale
des Pony et chaussees, Paris 1992
BAEL Regles techniques de conception et de calcul des ouvrages et constructions en beton
arme, suivant la methode des etats limites, Eyrolles, Paris 1992
ACI:
Buiding Code Requirements for Structural Concrete (ACI 31-95) and Commentary (ACI 318R-
95), ACI, Farmington Hills 1995
E.G.NAWY, Reinforced concrete: a fundamental approach, Prentice Hall, New Jersey 1996
S.K.GHOSH, D.FANELLA, B.RABBAT (ED.), Notes on ACI 318-95, Portland Cement
Association, Illinois 1996
BS:
Structural Use of Concrete. BS 8110, BSI, London 1998
Handbook to British Standard BS 8110:1995. Structural Use of Concrete, Palladian
Publications Ltd, London 1987
CH.E.REYNOLDS, J.STEEDMAN, Examples of the design of reinforced concrete buildings to
BS8110, E & FN Spon, London 1992
W.MOSLEY, J.BUNGEY, Reinforced Concrete Design, McMillan Education Ltd, London 1987
F.KONG, R.EVANS, Reinforced and Prestressed Concrete, Van Nostrand Reinhold (UK),
Berkshire 1987
EH/EHE:
Instruccion para el proyecto y la ejecucion de obras de hormigon en masa o armado EH-91,
Ministerio de Fomento, 1998
Instruccion de hormigon estructural (EHE), Ministerio de Fomento 1999
R.FERRARAS, Manuel de hormigon armado, Colegio de ingenieros de caminos, canales y
puertos, Madrid 1999.
6.4 Steel Connection Design
Once the CONNECTIONS layout is selected,
the screen is divided into four parts: a graphic
viewer for structure presentation, a field for
steel connection verification display, the
Connection Definition dialog box (presented
below) and the dialog box where the defined
connections will be presented.
Design of connections between steel structure
elements can be carried out in the Robot
Millennium system according to the
requirements of the two steel codes:
French code CM66
Eurocode 3.
structures.com
256
There are also American steel codes ASD and LRFD as well as the Polish code PN-90/B-
03200 available in the program, however, verification of a steel connection is performed in the
ESOP program. After selecting American codes or the Polish code (the option Job
Preferences / Codes), choosing the bars to be connected and creating a connection (i.e.
pressing the Create button in the Connection Definition dialog box), an additional dialog box
is displayed on screen, in which the connection type should be selected. Once the connection
type is selected and the OK button pressed, the appropriate sheet of the ESOP system
relevant for verification of the selected connection type, is activated.
Sections of the bars forming a connection as well as the forces acting on a connection are
forwarded to the sheet.
NOTE: For the link between the Robot and ESOP programs to be operational, both
programs needs to be installed on disk. If the ESOP system is not available,
then while trying to create the link (verification according to the requirements
of American codes or the Polish code), the user receives the message
informing that the ESOP system has to be installed.
The following connection types are available in Robot:
beam-beam connection column base - column connection: concrete base.
frame knee connection
beam-column connection
, ,
,
connection with angles (indicated in the table by
ANGLE)
column-beam (two-sided)
connection
tube connection (indicated in the table by: TUBE).
column base - column
connection: fixed
,
connection using gusset plate: for a single bar,
internal node and truss chord node (indicated in
the table by: GUSS)
column base column
connection: pinned
To start connection definition in the structure:
select the bars forming the connection (selected bars will be highlighted in the graphic
editor)
press the Create button in the Connection Definition dialog box. The appropriate type
(frame knee, beam-beam, column-base etc.) will be automatically assigned to the
connection. Additionally, appropriate options for the selected connection type will appear
in the Connection Definition dialog box.
define appropriate parameters for the selected connection type.
The following parameters can be defined for the selected connection types:
frame knee, beam-beam, column-beam
connections:
bar section parameters
base parameters
stiffener parameters
anchorage parameters
concrete, weld, and wedge parameters
pipe connection:
connection type (unilateral, bilateral)
parameters of adjoining bars (cross
braces)
bracket parameters
weld parameters.
column-base connection:
reinforcement parameters
bolt parameters
stiffener and plate parameters
depth - for concrete base.
connection gusset plate:
connection type (welded, bolted) and its
parameters
bar parameters, etc.
structures.com
257
connection with angles:
angle parameters
bolt parameters
cuts dimensions, distances, etc.
NOTE: If a connection has been defined, there is a possibility of modifying the
connection type. This is accomplished by means of the icons available in the
Connection type change toolbar :
change to pinned column base change to column-to-beam connection
change to concrete column base change to pipe connection
change to frame knee change to gusset plate connection (single
bar)
change to connection with angles change to gusset plate connection
(internal node)
change to fixed column base change to gusset plate connection (truss
chord node)
change to beam-to-beam connection
change to column-to-beam connection
(two-sided)
Once the Calculations icon from the toolbar or
the Analysis/Calculations command from the menu
is selected, the dialog box shown on the right will
be displayed on the screen.
Once the load cases to be considered during
connection calculation are selected (currently, only
the Connection Verification option is available) and
the Calculations button is pressed, the internal
connection forces will be analyzed and the
connection will be verified.
Each of the available connection types can be
verified manually. In this case the Analysis/Manual
Calculations option from the menu should be used.
Once this option is selected, the dialog box shown
on the right will be displayed on the screen.
structures.com
258
To start connection calculations, once the connection and its parameters are defined, either
click on the Connection viewer and select Analysis/Calculations or choose the
Analysis/Manual Verification option.
NOTE: In case of connection verification the dialog box appearing on the screen
depends on the connection type.
After entering the connection internal force values, a given connection type will be verified.
NOTE: In case of a pinned column base compared to a fixed column base, the above
dialog box has a slightly different form.
The dialog box shown below contains the basic data and verification results of connections
defined in the structure. The Connection Definition - data/simplified results is presented in
the lower portion of the screen (under the Connection Definition) dialog box after selecting
the CONNECTIONS layout.
After calculating (verifying) the connection, the Connection Definition - data/simplified
results dialog box (shown below) will contain the following data:
connection number
connection type: beam-to-beam connection (B-T-B), knee connection (frame knee, beam-
to-column connection or beam-to-column (two-sided) connection) (KNEE), fixed column
base (FIX), pinned column base (PINN), concrete column base (CONC), connection with
angles (ANGLE), pipe connection (TUBE) and connection using gusset plate (GUSS)
bars that make up the created connection
node, for which the connection has been created
ratio indicator defining the maximum straining of connection elements (bolts, sheet,
welds, etc.). The number determines the least favorable relation between the ratio within
the connection element and the capacity of the connection element. Once the connection
is calculated, the table column displays information indicating whether the connection
meets the code requirements (symbol ), or does not meet the requirements (symbol
).
After pressing the Note button, there appears an additional dialog box in which the user may
select the calculation note type (a simple one contains the table with basic info about the
connection, whereas the full one contains all the code requirements checked during
connection calculations).
If the bar section in a connection has been changed, then, pressing the Change button allows
one to up-date the geometry of the designed structure (bar section will be exchanged in
structure model). One should pay attention to the fact, that such change usually results in the
necessity of recalculating the structure.
In the current program version the user may also assign a defined connection to other
structure nodes (it must be a similar connection type). In order to do that, the user should:
structures.com
259
select (highlight) the connection to be copied
select appropriate structure bars
while being in the connection viewer, select the following option from the menu: Geometry
/ Connection Copy.
Once these operations are performed, the connection (together with the current parameters)
is copied; sections, materials and angles of inclination relevant to this new connection are
now taken into account.
Further on there are drawings presented that show forces occurring in steel connections as
well as directions of force operation for all connection types available within the Robot
program.
CONNECTION TYPE DIRECTIONS OF FORCE OPERATION
WITHIN CONNECTION
beam-beam, frame knee, beam-column,
beam-column (two-sided)
angle connection
Gusset plate: single bar
Gusset plate: internal node
Gusset plate: truss chord node
structures.com
260
Fixed column base
Pinned column base
Pipe connection
6.5 Timber Design
The structure design process is divided into three stages: first the structure geometry and
applied loads are defined, then internal forces and displacements are calculated, and finally
code requirements are verified and successive structural members are designed. This section
of the manual is concerned only with the third stage. We assume here that the load effects in
the structure are already calculated. Also please note this section is very similar to the steel
design section; therefore, if the reader is familiar with steel design in Robot Millennium, he
or she can skim the contents and focus on the differences only. In the current version of
Robot Millennium two timber codes are available: French timber code CB71 and Eurocode
5.
User can either verify or design the modeled structure. The design is applied to either single
members or groups of members and calculations can be executed in a similar way.
Depending on the selected timber code, the contents of the list can vary, but the basic
definitions remain the same no matter which code has been selected.
The following definitions apply:
MEMBER Single structure member to be verified or modified in the module. The most
common bar types are columns, beams.
A member used during verification/design can be defined as a single element or
a sequence of consecutive elements creating a column, beam, etc.
GROUP Member list. A group of structural members for which the same section is
assigned. Once the verification/design is completed, a section appropriate for all
members in the group (regardless of differences in internal force values for these
members or design parameters) will be selected. Groups are defined in order to
limit the variety of sections in the designed structure.
structures.com
261
Once the TIMBER DESIGN layout is selected, the screen will be divided into three parts:
graphic viewer for structure presentation and two dialog boxes: Definition and Calculations.
The Definition dialog box contains two tabs: Groups and Members (see the pictures below).
Once the members and groups are defined, verification for a single member or a group will be
carried out. Clicking on the Parameters button located on the Members tab results in opening
the Parameters dialog box (the contents of which depend on the selected timber code
available in Robot i.e. either Eurocode5 or French timber code CB71). The basic set of code
parameters includes buckling length, buckling parameters, lateral-buckling parameters,
rigidity, fire resistance parameters, etc.
Among the interesting options to be found in Robot, one can count the possibility of
automatic design by means of parameterized tapered sections. The option is available by
pressing the Parameterized sections button located in the Definitions dialog box (see
section 6.1).
The name of the selected bar is given in the Member Type field. The bar length may be
entered in the Member Length ly or lz fields. This may be done in two ways:
once the Real option is selected, the entered value is interpreted as the length
once the Coefficient option is chosen, the value is interpreted as the coefficient by which
the actual value should be multiplied to obtain the appropriate length. For example
entering a 0.25 value means that the appropriate length is equal to 1/4 of the actual
length.
For a simultaneous definition of several members of differing actual lengths, e.g. additional
supports equally spaced, the second method mentioned is very convenient. If the set
parameters are to be saved as a category, entering the length this way is essential. The
buckling length coefficient depends on the end support condition of the bar nodes in the
buckling plane. The buckling length may also be defined in the Buckling Scheme dialog box
opened by pressing the icon representing the selected buckling model type. Typical schemes
are found here; once one is selected, the coefficient value will be accepted or calculated
automatically.
The buckling is always considered in the calculations if a compression force appears in the
member even if it is negligible in comparison to the other internal forces. The program does
not perform analysis determining if buckling effects should be disregarded or not on its own. If
the user wants to eliminate buckling effects from the calculations, the last icon must be
chosen. It represents the option to disregard buckling in the calculation process.
Options used during the lateral buckling verification: lateral buckling type, load level, and
lateral buckling coefficient for lower and upper flanges.
In the dialog box presented below, calculation options for timber members or groups of
members are to be selected.
The Option of Verification field allows the user to select:
structures.com
262
Member Verification - verification according
to the member list based on consecutive and
independent calculations for each member.
The procedure is based on considering
intermediate points on a member along with
the load case present, verifying that it is
suitable by checking against the worst case
scenario according to the relevant codes of
practice. A number of points are considered
during calculations and a list of load cases is
specified. In other words, verification is
based on examining whether certain sections
(accepted before internal force calculations)
meet the code requirements. Such selection
determines whether the member will be
considered satisfactory, unsatisfactory or
unstable.
Group Verification - group verification is based on consecutive and independent
calculations (see Member Verification) for each member in the group. Material properties
set for the group are taken into account.
Group Design - group design is based on examining the previously adopted set of
sections determined by Group Definition and eliminating those that do not meet the code
requirements. Consecutive sections are eliminated until the first section meeting the
requirements is found. The described process is carried out separately for each family of
sections belonging to the analyzed group. Code calculations are performed for each
section at consecutive intermediate points on the member, consecutive load cases,
consecutive elements of the given member and all members belonging to the group. If the
given section does not meet the code requirements for a certain intermediate point, the
load case or element of the member in the group is eliminated and the next section from
the list is chosen. This process continues until all sections from the list are eliminated. To
start calculations in the design mode at least one group has to be defined. Design may be
carried out for many groups in which case the described process is run for each group
separately.
The lower part of the dialog box contains two fields: Loads and Limit State. In the first one, the
following options are located:
load case list - field for displaying the load cases taken into account during calculations.
The load case number can be entered therein.
load case selection - opens an additional dialog box (Case Selection) in which load
cases to be considered during calculations can be chosen.
Calculations can be run for ULS and SLS.
Clicking on the Calculations button results in timber member verification or design, according
to the parameters set in the Definition and Calculations dialog boxes. Once the calculations
are completed, the Simplified Results dialog box will be displayed on the screen. Clicking on
the section shown in the short list will result in displaying the Results dialog box.
Similarly as for steel members, the detailed analysis is also available for timber members; it
can be activated by pressing the Detailed button provided in the Detailed results dialog box
for EC5 code. The program enables performing additional calculations based on the
requirements described in Eurocode 5:
with transversal compression considered (coefficient Kc,90)
with openings considered (coefficient Khol: see: Eurocode 5:Book1 IV-5-8)
with the shape of a beam ending considered (coefficient Kv: see Eurocode:Book1 IV-5-7).
structures.com
263
Each of the member analyses listed is performed independently, thus each of them may be
run separately.
Calculations can be also carried out for user-defined section forces (not calculated by the
program itself). This can be done using the Manual Calculations option that is available from
the Timber Member Design menu. Member verification/design can be carried out.
Configuration button allows for selecting a number of points at which a member is calculated.
These
points are chosen equidistant along the member.
SELECTED REFERENCES (DESIGN OF TIMBER STRUCTURES)
EUROCODE 5 - Calcul des structures en bois. Partie 1-1: Regles generales et regles pour les
batiments. Norme P21-711
STRUCTURES EN BOIS AUX ETATS LIMITES - Introduction a l'Eurocode 5. STEP1 -
Materiaux et bases de calcul, SEDIBOIS. Union nationale francaise de charpente,
menuiserie, parquets, 1997
STRUCTURES EN BOIS AUX ETATS LIMITES - Introduction a l'Eurocode 5. STEP2 - Calcul
de structure, SEDIBOIS. Union nationale francaise de charpente, menuiserie, parquets, 1996.
6.6 Plate and Shell Reinforcement
Robot Millennium allows for calculating of reinforcement areas for plates or shells. Plate and
shell reinforcement type parameters depend on the selection of the code used during plate
and shell design. A list of currently available plate and shell reinforcement codes includes:
American codes ACI 318/95 and ACI 318/95 (metric)
British code BS 8110
Eurocode 2 (with French, Belgian, Dutch, Italian, Finnish and German NADs)
Canadian code CSA A23.3-94
French codes BAEL91 and BAEL 91 mod.99
Dutch code NEN6720 (VBC 1995)
Spanish codes EH91, EHE98
Russian code SNiP 2.03.01-84
Polish code PN 84/B-03264 and Polish code PN-B-03264 (1999)
Romanian code STAS 10107/0-90
Singaporean code CP65.
The option for definition of reinforcement parameters is available for only two types of
structures (plates or shells) by:
selecting the command from the menu Geometry / Code Parameters / Reinforcement
Type or
pressing the icon.
The New reinforcement type dialog box is available by pressing the New icon on the Plate
and Shell Reinforcement Type dialog box. The dialog box consists of three tabs: General,
Materials and Reinforcement. The General and Reinforcement tabs are identical for all codes,
whereas the shape of the Materials tab depends on the selected code of plate and shell
reinforcement.
structures.com
264
NOTE: In case of selecting some of the concrete codes, the Reinforcement
parameters dialog box consists of four tabs, as additionally, it includes the
SLS Parameters tab (it allows calculation of cracking and deflection for
plate/shell structure).
Same as in the dialog box used to define other structure attributes (supports, sections, etc.),
the reinforcement type definition process has been divided into two stages:
reinforcement type definition
assigning reinforcement types to panels.
RC plate design can be started in two ways:
choosing from the vigniette of structure type selection (compare chapter 2.1) RC plate
design a plate design module will be working as a stand-alone program without
connection (data exchange) with other parts of Robot Millennium system
once a structure is defined, one should select in it (by highlighting in a graphic editor)
an appropriate panel (plate) list, and then choose the following command from the menu:
Analysis / Design of RC Structure Elements / RC Plate and Shell Design / Required
Reinforcement. It will result in running the RC PLATES - REQUIRED REINFORCEMENT
layout and reading geometry, loads and obtained results into a code module. The screen
will be divided into three parts: edit viewer with a designed plate view and two dialog
boxes: Plate and Shell reinforcement and Reinforcement.
Description of RC plate design will be presented for a second case of calling a module of RC
plate design. The option for calculating plate or shell reinforcement is available once the RC
PLATES REQUIRED REINFORCEMENT layout located in the RC Plates group is
selected. A dialog box shown below appears at the bottom of the screen.
The upper part of the dialog box
contains: name of the code used for
determining reinforcement type
parameters of a panel as well as the
number of the version of plate and
shell reinforcement module installed in
the Robot system.
To begin calculations of the plate or
shell reinforcement,
in the Case lists field - define the load cases taken into account during calculations for the
individual limit states (ULS, SLS and ALS - accidental state). One should enter the
numbers of load cases in the appropriate fields or press the () button and in the
Selection dialog box one should select load cases or combinations of load cases for the
individual limit states
if code combinations were previously created then activating an appropriate option in the
Code Combinations field ( symbol appears) determines the limit state, for which the
generated code combinations will be included in calculations
determine the method of calculating shell or plate reinforcement; the following calculation
methods are available in the program: analytical method, the method of equivalent
moments (NEN) and Wood&Armer method of equivalent moments
select in the graphical viewer the panel(s) for which reinforcement calculations will be
performed; if no panel is selected, calculations will be performed for all panels.
In Robot the width of cracking is calculated independently for two directions defined by axes
of reinforcement. This is an approach analogous to the simplified methods presented in
relevant literature. The implementation of the method not related to codes results from lack of
relevant recommendations concerning plates with cross reinforcement.
structures.com
265
The algorithm of calculations is based on the formulas enabling calculation of cracking width
for beam elements. Calculations are carried out on the cross-section with reinforcement
resulting from Ultimate Limit State for all the forces caused by the loads defined as SLS or
appropriate SLS combination. The moments recognized in calculations of Serviceability Limit
State are equivalent moments calculated according to the selected calculation method:
analytical, NEN or Wood&Armer. The analytical method for Serviceability Limit State does not
recognize actions of m
xy
moments. Thanks to the implementation of NEN or Wood&Armer
method, one may recognize the m
xy
moments in calculations by increasing the moments m
xx
and m
yy
. Wood&Armer method is recommended for calculations of plates with cross
reinforcement among others by ENV 1992-1-1 EUROCODE 2 (Annex A.2.7).
The calculated cracking width whose value is presented in the table of results is the maximum
value obtained from all the analyzed load cases.
The algorithm for calculating deflections of RC plates is based on the use of calculations of
an isotropic elastic plate, with the stiffness of the cracked element recognized. Originally, the
deflections are obtained from the calculations performed according to the finite element
method (FEM). Then, they are modified. Deflections may be identified with displacements
only for not deformed supports. In the shell module (3D) while calculating RC plate deflection,
the displacement of the least displaced support is subtracted from displacements of each
element. It means that the deflections are measured from the plane parallel to the not
deformed plate surface that passes through one support point of the deformed plate.
One should pay attention to displacements of the remaining supported corners of a plate.
The calculation algorithm used in the Robot program is based on the assumption that it is
possible to obtain an RC plate deflection by means of multiplying its elastic deflection by the
coefficient defining stiffness change.
u
D
B
u
b

where:
u
b
- displacement of an RC plate
u - elastic displacement of a plate
D - elastic stiffness of a plate
B - equivalent stiffness of an RC plate.
Once the FEM calculations and the calculations of reinforcement resulting from Ultimate Limit
State are performed, the program determines the stiffness for every finite element. The
calculations are conducted for two directions of reinforcement. They recognize the phase of
work of an RC element and the ratio of long-term loads to total loads. As a result, one obtains
two stiffness values - different in most cases - for each finite element. As further calculations
require one value of stiffness, it is obtained by averaging (weighted average) of the
component stiffness values. The weight is the ratio of moments acting on a given element in
both directions.
Afterwards, the average stiffness of elements is calculated. The assumption that the plate
stiffness equals average stiffness of elements causes the calculated deflections to be smaller
than real ones.
The next calculation step is to find an element with the least stiffness. If the plate stiffness
identical to this element stiffness is adopted for calculations, the deflections will be overrated.
Since the searched value lies between thus calculated values, the following may be adopted:
0 . 1 0 . 0
) )( 1 ( ) (
min

B B u u
b
The alpha coefficient has been determined on the basis of data concerning deflections of
typical RC plates and equals 0.25.
If the Reinforcement adjustment option is switched on, the program increases the
reinforcement area during calculations to increase the stiffness of the element, which, in
consequence, leads to reduction of deflections. In both directions the reinforcement is
distributed in inverse proportion with respect to stiffness. In the case when it is not possible to
fulfill the condition of maximum deflection defined by the user, the table of results will display
appropriate warning (red background of the result cell). The program does not have any limits
set on reinforcement ratio (other than the code-defined ones), so one should pay attention to
the economic aspect of the solution provided.
structures.com
266
The central part of the dialog box contains the Globally averaged design forces option. It
appears in this dialog box for the results of calculations of plates and shells are discontinuous
for the forces in nodes of finite element mesh (if four finite elements converge in one node, a
different force value is determined for each element). If the Globally averaged design forces
option is switched off, the results in nodes are averaged within a given panel for which
reinforcement is being calculated. If the option is switched on the results in nodes are
averaged for all panels.
NOTE: If the Globally averaged design forces option is switched on, one should be
careful while calculating reinforcement for plate structures consisting of
panels that are not located in one plane, for global averaging may result in
the situation where quantities that do not correspond to each other are
averaged for edges of such panels.
Two options are located in the in the bottom part of the dialog box:
display error messages - if this option is active, possible reinforcement calculation error
warnings will appear on the screen during reinforcement calculations.
reduction of forces (at supports or above columns)
For plate and shell elements supported at points (by means of the available support types or
columns), values of moments and stresses near the support points may be considerably
greater than at the remaining points of the plate. It may cause calculation of incorrect
reinforcement in the vicinity of supports and columns; to avoid that, such connections may be
modeled by means of the rigid links available in the program or by applying the Reduction
near supports option.
Reduction of values near supports consists in replacing the result values obtained in the
vicinity of supports and columns with the average value obtained in the neighborhood of these
supports. Three stages may be distinguished in the option operation:
definition of support nodes the program reduces values only near the nodes considered
as supported; these are the nodes at which:
- a column type support (rectangular or round) is defined with non-zero dimensions
advanced parameters for supports
- bar elements are connected with finite elements (but only when the other bar end is
not connected with the same finite element, either as it is the case for a
strengthening beam); moreover, for a node to be considered as supported, it must be
adjoined by at least one finite element
At nodes where supports at points (without dimensions) and beam (linear) supports are
defined, values are not reduced.
definition of a reduction radius if reduction of selected values is performed at a given
node, then it should be determined which values should be disregarded; the values
ignored include the values at this node and at centers of elements that adjoin this node;
apart from that, the program looks for nodes positioned in the distance lesser than the
reduction radius values at these nodes will also be disregarded and replaced with the
calculated average value; the reduction radius is determined in the following manner:
- for a support defined at a node the reduction radius equals half the length of the
diagonal of a rectangular support or half a diameter of a round support
- for a column adjoining a node - the reduction radius equals half the length of the
diagonal of a rectangle with dimensions equaling the column width and height.
If one node is adjoined by several supports with different values of reduction radiuses,
then the greatest of the calculated radiuses is adopted in calculations.
calculation of the average value once the reduction radius value is calculated, the
program searches nodes positioned in the distance (from the supported node) lesser than
the determined radius value; the result values in all these nodes and in centers of finite
elements that adjoin them are disregarded; these values are replaced with the average of
the result values on the edge of the disregarded region when calculating this average
the averaging parameters are taken into account; if a node is adjoined by an element, in
structures.com
267
case of which at least one of the nodes is positioned inside the reduction circle, then a
value from this element is also ignored while averaging.
Once the averaged value is calculated, it substitutes for all the values disregarded in the
vicinity of a given support node.
The following can be selected in the Reinforcement dialog box that is located in the right part
of the RC PLATES REQUIRED REINFORCEMENT layout:
calculated reinforcement areas and spacing: Top Ax, Bottom Ax, Top Ay, Bottom Ay, Top sx,
Bottom sx, Top sy, Bottom sy and Transversal At.
Theoretical (required) areas of reinforcement obtained during the design of a plate/shell
structure may also be presented in the form of reinforcement crosses. The crosses of the
reinforcement area or the spacing of the reinforcement will be presented if one selects the On
option.
In case of concrete code selection which allows calculations considering service limit state,
there is additional tab: SLS accessible in the Reinforcement dialog box. One can select the
following quantities to be presented on this tab (depending on a code not all quantities are
available): cracking width in both directions (ax and ay), deflection f.
Once calculations of theoretical (required) reinforcement are performed one can shift to the
RC PLATES PROVIDED REINFRCEMENT layout. The options available in the Results
dialog box allow presentation of the results of reinforcement area calculations: a presentation
of real (provided) reinforcement zones and a manual modification of these results. The
Results dialog box consists of two tabs: Bending and Punching.
structures.com
268
The dialog box shown in the figure above consists of three main parts:
reinforcement zone definition: automatic / manual (working mode definition)
selection of lists of solutions
presentation fields of real (provided) reinforcement for a selected solution.
Definition of reinforcement zones
The program allows to determine real (provided) reinforcement zones in two modes:
automatic and manual.
The automatic mode allows to generate reinforcement zones and calculate a real (provided)
reinforcement area on the basis of the accepted reinforcement parameters, calculation
options and on the basis of the calculated earlier theoretical (required) reinforcement areas.
Reinforcement zones are determined by the optimizing algorithms. When the calculations are
completed, a user can select a solution from a list of the available solutions proposed in a
program.
To optimize reinforcement zones a solution taking into consideration several factors should be
found. For wire fabric reinforcement the optimization aims are as follows:
cut number to achieve the appropriate cuts
mass of reinforcing wire fabrics
usage ratio for wire fabrics.
For a bar reinforcement a spacing modularity, an assortment decrease of the used bars or a
used steel mass is preferred.
A manual mode allows a user to define real (provided) reinforcement zones. In this case the
program only selects the appropriate reinforcing bars and wire fabrics (on the basis of the
calculated theoretical (required) areas, reinforcement parameters and calculation options).
After selecting the Manual option of a zone definition and moving to a graphical viewer, a
mouse cursor changes its shape (to a cross shape), that allows to insert the zones.
A definition method of reinforcement zones is similar to a definition of rectangular contours.
The definition consists in determination of two points position. The first click with a left mouse
button results in the first corner definition and moving to a rectangle dragging mode. Clicking
again results in a opposite corner definition. When the zone geometry is defined, a zone
name should be given and press the Apply button.
The Delete button results in deleting the current selected zone from a list, while the Delete
All button results in deleting all zones from a list.
Solution list
A solution list allows to select one of several, possible solutions, proposed by the program.
The solutions are sorted out according to an optimization coefficient. This coefficient is the
weighted average of the parameters being the optimization aim. It should be taken into
consideration that the solutions making reinforcing works easy are preferred more to these
requiring less steel amount.
When wire fabric reinforcement option is selected, the solution list includes: a number and
type of used wire fabrics, percent of wire fabric consumption and a total mass of wire fabrics
with wastes.
For the bar reinforcement the following items are given on the solution list: estimate survey for
the weight of all bar types and survey for total steel weight. In these results the mass being a
result of the demanded bar laps and a structural reinforcement.
Presentation and modification of reinforcement zones
The fields located in the Zone list table facilitating a presentation and modification of
reinforcement zones. For a selected zone, the coordinates of the left bottom and right top
zone corner are determined.
structures.com
269
The selected zone is highlighted in the table presenting the reinforcement values and in the
viewers presenting the top and bottom reinforcement zones. The active zone can be modified
by a user.
NOTE: Any zone modification during the operation in the automatic mode results in
moving to the manual node.
The options located on the Punching tab allow to:
view and define the punching verification points
group the verification points (unify the geometry)
assign the supports (columns) the geometrical properties of the head
view the punching calculation results.
Viewing and adding the defined by the user verification point
If in a structure the point supports are defined, they are automatically entered on the point list
and denoted by the letter S with consecutive ordinal numbers. For each support type the
following things can be read:
coordinates in the Position field
number on the node to which a support is assigned (in the Node number field)
maximum reaction value of a support in the Maximum punching force field
support geometry if such one is defined during the support definition in the Advanced
dialog box.
To display data concerning a selected verification point, one should only select its name
located on the point list.
Independently from supports, verification points can be defined by the user.
To define a new verification point, one should press the New button in the Verification points
field. Each time when the operation is completed, the verification point denoted by letter P
and the consecutive ordinal number is added to the point list. Once a verification point is
chosen, the fields allowing definition of the point position, load geometry and values of
punching force are active. These values are saved (no additional operation to confirm them is
needed).
To delete the added verification point, one should select it and press the Delete button. The
points, which are the supports defined in a structure geometry (denoted by the letter S)
cannot be deleted.
Support (column) heads
The program allows to define heads over the supports (columns), that are taken into
consideration in the punching calculations. To perform it, the Head option should be active
(the fields with the dimensions are available) and the head values should be defined. For the
head on the rectangular support they are the lengths of the head sides on the level where
head meets the slab (denoted by the letters a and b) and the head height denoted by the
letter h. For the heads over the circular supports it is the d head diameter and h head height
(for a circular head) or the lengths of the head sides (denoted by the letters a and b) - for a
rectangular head.
When the support belongs to a group, the head is applied to all elements of this group.
Point grouping
Both, the additional verification points and supports can be grouped to modify the geometry in
the easier way. The points may be grouped manually by selecting on the name list and
pressing the > button; it is also possible to group all the supports automatically: one should
press the >> button.
structures.com
270
The additional verification points can be grouped if they have the same geometry type. In a
case of supports, a compatibility of the support dimensions is necessary. If the conditions to a
compatibility of the support dimensions are not satisfied, the verification points or supports not
compatible with the first one on the list are deleted in a moment of the group confirmation.
When the points with different dimensions are grouped, the dimension values are accepted
on the basis of the first defined point in the group. When a new point is added to the group, its
dimensions are automatically changed to the compatible ones with the group dimensions.
Grouping the supports with different head dimensions is done by analogy.
When the group is defined, any change for the arbitrary group component relates to the entire
group and it is modified.
Presentation of calculation results
In the table presenting the results of the punching analysis for each verification point the
following values are presented:
admissible punching force calculated by the program according to the appropriate code
requirements
generalized design force defined in the Maximum punching force field for the additional
verification points or read from FEM calculation results for supports
critical circumference calculated on the basis of code requirements
reinforcement, which includes:
reinforcement range from the member center in both perpendicular directions L1 and L2
circumference of reinforcement zone (if required)
total area of reinforcement
number and diameter of bars calculated on the basis of the total area and settings in the
options for punching
safety factor being a ratio of a maximum punching force and admissible punching force.
Punching analysis results for individual points are presented in the color corresponding to the
calculation result:
in blue for points that fulfill the conditions for punching and do not require reinforcement
in green - for points that fulfill the conditions for punching and require reinforcement
in red - for points that do not fulfill the conditions for punching despite applying
reinforcement.
The critical circumference is presented graphically on the Slab reinforcement layout as a
green line around columns. The range of punching reinforcement is shown graphically in a
final drawing of a slab (formwork).
The Punching tab may include additional options; it depends on the selected code of RC
structure design.
ACI Code
Once this code is selected, in the above dialog box the Type option is also available which for
each support determines its position: within slab, on slab edge or in slab corner. The support
type is used during calculation of the admissible punching force [ACI 318-99 11.12.2.2].
EC2 Code
Once this code is selected, in the above dialog box the option is also available. For each
support the parameter determines position of the support: within slab, on slab edge or in slab
corner. A value of this parameter is used during calculation of the punching force [ENV 1992-
1-1 EC2 4.3.4.3].
structures.com
271
Once real (provided) reinforcement calculations for a plate are completed one can present
calculation results in the form of a calculation note (the Results / Calculation note option). The
program will display the text editor of the Robot Millennium system containing designed
plate data and obtained calculation and design results.
Once the Results / Drawings option is selected from the menu or the icon is pressed the
Robot program will activate the FINAL DRAWINGS layout which will present a final drawing
of a calculated and designed plate. The shape of the final drawing of a plate presented on the
screen corresponds to the accepted drawing parameters (see chapter 6.2.5).
The analytical method of calculating plate and shell reinforcement area implemented in Robot
is based on the conception presented in the A.Capra and J-F. Maurys article titled Calcul
automatique du ferrailage optimal des plaques et coques en beton arme, Annales de lInstitut
Technique du Batiment et des Travaux Publics, No.367, Decembre 1978.
Calculations procedure
The calculations procedure is based on the fundamental assumption that, if the reinforcement
values Ax and Ay - corresponding to two perpendicular directions x and y - are given, one
may adopt an equivalent reinforcement in any other direction, calculated by means of the
following formula:
) ( sin * ) ( cos *
2 2

y x n
A A A
where: ) , ( n x
The values of sectional forces (moments and membrane forces) M
n
, N
n
may be obtained from
the following transformational formulas:
) 2 sin( * ) ( sin * ) ( cos *
2 2

xy y x n
M M M M
) 2 sin( * ) ( sin * ) ( cos *
2 2

xy y x n
N N N N
Thus, the below-presented inequality formulates the condition of correct reinforcement, i.e.
the reinforcement that is able to carry the internal forces in an arbitrary section:
) , ( ) ( sin * ) ( cos *
2 2
n n n y x
N M A A A
where the function (M
n
, N
n
) refers to the value of reinforcement required to carry the forces
calculated for the direction n - M
n
, N
n
.
Inequality ) , ( ) ( sin * ) ( cos *
2 2
n n y x
N M A A
determines on the plane (Ax, Ay) the area of admissible values of reinforcement Ax, Ay
(half-plane). If such area is determined for a sufficiently dense set of directions n (the
program assumes that control is performed every 10), one obtains the area of admissible
values Ax, Ay.
The reinforcement adopted by the program is the minimal reinforcement (it yields the minimal
sum of surfaces Ax+Ay).
The program Robot Millennium provides also the possibility of using the method of
determining equivalent moments devised by Wood and Armer (supplement to the European
code [ENV 1992-1-1 EC2 Design of Concrete Structures Appendix 2, point
A.2.8 Reinforcement in Slabs]).. The detailed description of the method may be found, for
instance, in R.H.Wood The reinforcement of slabs in accordance with a pre-determined
field of moments, Concrete, February 1968, August 1968 (correspondence).
Calculation procedure
For a selected direction x (and the corresponding perpendicular direction y) one calculates
two types of design moments M*: the lower ones (positive, causing mainly tension in the
bottom parts) and the upper ones (negative, causing tension in the upper parts). The
general procedure takes the following form:
structures.com
272
Determination of the lower moments M
xd
*, M
yd
*:
M
xd
* = M
x
+ |M
xy
|
M
yd
* = M
y
+ |M
xy
|
However, if M
x
< -|M
xy
| (i.e. the calculated M
xd
* < 0)
M
xd
* = 0
M
yd
* = M
y
+ |M
xy
2
/M
x
|.
Similarly, when M
y
< -|M
xy
yd
* < 0) (*)
M
xd
* = M
x
+ |M
xy
2
/M
y
| (*)
M
yd
* = 0 (*)
If any of thus obtained moments M
xd
*, M
yd
* is smaller than zero, one should assume the zero
value (the design moments for tension in the upper layers are determined further on in the
text).
Determination of the upper moments M
xg
*, M
yg
*:
M
xg
* = M
x
- |M
xy
|
M
yg
* = M
y
- |M
xy
|
If M
x
> |M
xy
xg
* > 0)(*)
M
xg
* = 0 (*)
M
yg
* = M
y
- |M
xy
2
/M
x
| (*)
Similarly, when M
y
> |M
xy
yg
* > 0)
M
xg
* = M
x
- |M
xy
2
/M
y
|
M
yg
* = 0.
If any of thus obtained moments M
xg
*, M
yg
* is bigger than zero, one should assume the zero
value (such moments would design the lower reinforcements, which is already guaranteed by
the formerly calculated lower moments M
xd
*, M
yd
*)
NEN Method
The method of design for the given equivalent moments is adopted in the Dutch code NEN
6720 (section 7.3.2).
Calculation procedure
The following algorithm is a next simplification of Wood and Armer algorithm.
Determination of the lower moments M
xd
*, M
yd
*:
M
xd
* = M
x
+ |M
xy
|
M
yd
* = M
y
+ |M
xy
|
Determination of the upper moments M
xg
*, M
yg
*:
M
xg
* = M
x
- |M
xy
|
M
yg
* = M
y
- |M
xy
|
Assumed stress state
The above quoted methods allow one to simplify calculations in the case of pure flexural
state (bending of plates). There is no ready simplified algorithm for a complex state (shells)
which implies additional membrane forces (N
x
N
xy
, N
y
). It seems that the implementation of
the analytical approach is the only admissible procedure.
However, as the modeled shells frequently work as plates (when the membrane forces are
negligible), the program still provides the possibility of selecting a simplified method of
calculating design moments on which the longitudinal forces N
x
, N
y
are superimposed.
It should be stressed that this solution is justified only for relatively small membrane forces
and it is the user who takes the responsibility for a particular case of implementing this
procedure.
Comparison between methods
In the performed tests, the differences between the results obtained by calculating
reinforcement by means of the analytical method and the simplified ones have not exceeded
5% of reinforcement area (the simplified methods yielded a slightly greater maximum
reinforcement).
structures.com
273
7. SECTION DATABASES
In the Robot Millennium program, the user has at his/her disposal several bar section
databases. All data pertaining to the section geometry is stored in the standard bar section
databases. The user may make use of them by providing the name of the appropriate
database. The user may also define his/her own section database in which the most often
used sections will be stored.
Once the Tools/Section Databases option is chosen from the Robot menu or the icon
pressed a viewer, used for the presentation of information about standard sections and
performing operations in section databases will appear on the screen as shown on the
drawing below.
The Section Database viewer shown on the above drawing is divided into several parts:
view of the selected section cross-section on which the basic markings and section
dimensions are shown.
dialog box in which a list of basic properties values of the selected section is shown.
table in which all the section data is given.
The table presents the following data concerning sections:
section name
basic section dimensions (dim1, dim2, dim3) - depending on the section type, one column
(dim1, and in the remaining, a zero value is entered), two columns or three columns are
filled
structures.com
274
section shape type
all available dimensions and geometric properties of the section (dimensions d_1, d_2, ...,
angle a_1, a_2 are additional geometric section parameters, while dimensions P1_L,
P1_T, P2_L, P2_T, P3_L, P3_T, P4_L, P4_T are the dimensions of a cross-shape
section)
section symbol - contains a full description of section geometry, when the profile is not
standard
The final four table columns provide the following information: thin-walled (if there
appears symbol, it means that the section is a thin-walled one), material (if there
appears symbol, it means that a material has been defined together with the section),
material description (contains a complete description of materials in a section), points
(contains a description of characteristic points of a section).
The section database viewer allows the user to:
present the dimensions and section properties located in the selected section database
open a new section database by selecting the File/Open Existing Database command or
selecting the Open icon
declare a new section database
copy sections between databases.
The Robot Millennium program provides also an option that allows the user to design his
own bar sections and calculate their properties. The option is available by selecting the
TOOLS/SECTION DEFINITION layout . The basic functions of the Section Definition
module are:
possibility of a graphic definition of the bar cross-section
calculation of its geometric and weighted values (area, moments of inertia, center of
gravity, etc.)
saving the section to the users database along with calculated values.
The details of section definition are presented in 3.18.
Once a section is defined one can save the section to a section database (into users
database). The option is available in the SECTION DEFINITION layout:
from the menu, by selecting the following command: File/Save to databases
from the toolbar, by pressing the icon.
After selecting this option, the dialog box shown in the figure below appears on the screen.
structures.com
275
To save a section to a database one should:
determine a name: maximum four letters (NOTE: Using numbers is not permitted)
define dimensions describing a section (real numbers determining information about a
section):
= in case when one dimension is given it should be determined in the following way:
Dimension 1
= in case when two dimensions are defined it should be determined in the following way:
Dimension 1 and Dimension 3.
The Section type option allows definition of the section nature/type: it is used during structure
design. The units given in the lower part of the dialog box concern section type (they are
specified in the current units of section dimensions).
structures.com
276
8. PRINTOUTS
Completing technical documentation of projects is a very important stage in the design
process. Therefore the Robot Millennium system provides a variety of possibilities for
printout composition to meet the needs of the user.
8.1 Calculation Notes
A wide range of calculation notes generated according to predefined templates has been
introduced in the Robot Millennium system. Such notes are created after installation,
structure calculation and structural member design. Such notes contain all the data entered
by the user and especially in the case of notes after the structure design and calculation
results (internal force values, reinforcement components for R/C structure design, connection
parameters for connection verification, etc.).
Printout parameters are to be set in the Page Setup dialog box (see chapter 8.3).
8.2 Printout Composition
Printing from any viewer is possible after selecting the File / Print option. Once this option is
selected, only the active graphic viewer contents will be printed.
A novelty in the Robot Millennium system is the possibility of printout composition according
to user needs. Once the File / Printout Selection option or the respective icon is
selected, the dialog box shown below will be displayed on the screen.
The above dialog box allows for composing a printout of all previously created objects. The
printout composition is complex since the entire user-defined set will be printed as opposed to
single table contents or one graphic viewer.
The above dialog box contains four tabs:
structures.com
277
Standard - standard template with basic data on a given structure (structure view,
information on nodes, bars, sections and loads), calculation results obtained during
structure analysis (reactions, displacements, internal forces, stresses and the eigenvalues
for the dynamic analysis) as well as information concerning design/verification of steel
structure elements (bars and connections)
Screen Captures - field in which the viewer screen captures are placed. The screen
capture option can be run by selecting the File/Screen Capture command from the menu.
Once it is selected, the small dialog box shown below will be displayed on the screen.
Pressing the OK button results in saving the viewer contents under a given name to the
left panel (Screen Capture tab).
NOTE: Table content can be captured any time, except for the case when the Edit
tab is selected.
Templates - allows composition configuration, including change of standard set
configuration.
Simplified printout - the tab allows one to compose simplified printout covering data and
calculation results.
The lower part of the dialog box contains the following buttons:
Page setup opens the Page setup dialog box
Preview Presents the composed printout
Print starts printing the composed printout
File saves the components of the composed printout
Close Closes the Printout composition dialog box.
If the Preview button is pressed, there will appear the preview of the currently composed
printout. Two of the options contained in the menu are worth being discussed in detail, as
they may turn out very useful during the preparation of the documentation for a designed
structure. Once the preview is activated, the mouse cursor is in the zoom in/out mode
(symbolized with ). If one presses the button located in the menu, the cursor mode will
change to selection and edition. When the cursor is in this mode, double-clicking an object in
the print preview results in opening the viewer that allows the user to edit the view of the
structure or a table. If the user modifies the view of the structure (e.g. adds a new bar to the
structure) or of the table (e.g. adds a new column to the table) and presses the Return to
preview button, the changes introduced in the view of the structure or table will be
automatically included in printout composition (visualized in print preview). Returning to the
zoom in/out mode is carried out by pressing the button in the menu of print preview.
The Full/Simplified buttons are located at the end of menu. They allow one to select the
mode of the printout composition component preview. Choosing the simplified mode results in
quick printout preview opening. Simplified preview presents only placing of printout
components on pages - in tables the structure analysis results may not be presented. Printout
preview will always be opened in the mode selected by a user on the toolbar shown in the
figure above.
structures.com
278
8.2.1 Standard Tab
The Standard tab of the Printout Composition dialog box has been divided into two panels:
left and right. The left panel presents particular objects selected for printing (the contents of
which depend on the chosen tab), whereas the right one is used to compose a complete
printout that may contain all the objects specified in the left panel. To prepare the printout
composition move objects from the left panel to the right one using two buttons: All and Add.
Once the All button is pressed, all the objects from the left panel are transferred to the right
one whereas just pressing the Add button moves only the highlighted object from the left
panel. Below the Add and All buttons, the user finds the Standard button. If the user
composes a printout that fits his needs and he would like to make the composition a standard
one, he should press the Standard button. All the printout components contained in the right
panel will be transferred to the left panel of the Standard tab.
NOTE: If the user defines his standard printout components (i.e. moves the elements
from the right panel to the left panel on the Standard tab by pressing the
Standard button), the new standard arrangement will overwrite the one
provided by the producer together with the program. It will become
impossible to restore the previous standard.
There is a small toolbar over the right panel. It consists of several icons that allow one to:
-
delete from the right panel all the printout elements, hitherto composed by the user
-
delete from the right panel selected printout elements, hitherto composed by the user
-
moving a selected element up in the composed printout
-
moving a selected element down in the composed printout
-
activate the preview of the selected composed-printout elements (if no element is selected in
the right panel, the preview will be activated for the entire composed printout)
-
opening preview of selected printout composition components in the simplified or full mode.
Choosing the simplified mode results in quicker printout preview opening (NOTE: simplified
preview presents only placing of printout components on pages e.g. in tables the structure
analysis results may not be presented)
-
activate printing of the selected composed-printout elements (if no element is selected in the
right panel, the printing will be activated for the entire composed printout)
Once the Insert Page Break option is selected (marked by ), an object added to
the right panel will begin on a new page. The command from the context menu (displayed
below) can also be used to insert the page break. To do so, click the right mouse-button while
the cursor is in the right panel.
Insert blank page - adds to composed printout a blank page before
a selected (highlighted) object contained in the right-
hand panel.
Note before - runs the text editor of the Robot Millennium system,
which allows one to add texts to appear before an
object selected (highlighted) from the right panel, i.e.
contained in the composed printout
Title - runs the text editor of the Robot Millennium system,
which allows one to add texts to appear as the title of
an object selected (highlighted) from the right panel,
i.e. contained in the composed printout
Moreover,
the presented context
menu contains several
other options that
allow one to:
Note after - runs the text editor of the Robot Millennium system,
which allows one to add texts to appear after an object
selected (highlighted) from the right panel, i.e.
contained in the composed printout
structures.com
279
8.2.2 Screen Captures Tab
The Screen captures tab of the Printout Composition dialog box has been divided into two
panels: left and right. The left panel presents the names of the screens captured by the user.
The screen capture may be carried out by means of the File/Screen capture command from
the menu.
NOTE: It is also possible to capture the contents of a table, except when the Edition
tab is selected in the table.
The right panel contains the complete printout composed by the user from the object available
on the left panel. It is possible to add defined screen captures to the printout by clicking the
All and Add buttons. Pressing the All button results in transferring all the elements from the
left panel to the right one, whereas pressing the Add button results in adding to the printout
only the screen capture that is selected (highlighted) in the left panel.
There is a small toolbar over the right panel. They were described in the former chapter.
8.2.3 Templates Tab
The Templates tab in the Printout Composition dialog box is divided into two panels. The
left one contains the templates of entire printouts provided with the program. The templates
contain structure views/drawings, appropriately arranged data/results tables. Some templates
have the node/bar selection activated. The predefined templates are presented in the left
panel in blue. They cannot be removed, although the user can overwrite them. If the user
creates a template and ascribes it an already existing name (the name will change the color
to black - the color of user-defined templates), an then deletes the template, the name will not
be deleted (color will change back to blue) and the predefined contents of the template will be
restored.
The middle part of the dialog box contains the following buttons:
New - once this button is pressed, a new line in the Available templates field will be
added and a new template name can be entered there. Once an appropriate name is
inserted and Enter is pressed, the all the component contained in the right panel of the
dialog box will be saved to the template.
Open - once this button is pressed, components defined in the selected template will be
transferred to the right-hand panel of dialog box.
Save - once this button is pressed, all the components contained in the right-hand panel
will be saved to the template of the defined name.
When a template is being opened (after the Open button is pressed), a small dialog box -
presented below - will appear on screen. If Standard Font of Component Titles option is
inactive (this is the default setting), the font of the titles of particular printout components will
be identical with the one of the opened template. If the option is active, the titles will be
presented with the font defined by the user in the Preferences dialog box.
structures.com
280
8.2.4 Simplified Printout Tab
Once the Simplified printout
tab is selected, the Printout
composition dialog box
takes the form presented on
the figure below.
To create a template of
simplified printout, the user
should:
select the quantities
that should be
contained in the
composed printout
filter nodes, bars, cases
or modes for which the
printout will be created
activate sorting, if
necessary
press the Save template button, which will result in activating Templates tab of the
Printout composition dialog box and locating the selected elements of simplified printout
in the right panel. The user-defined template of simplified printout may be saved.
Below, one finds a presentation of successive stages of composing simplified printout.
STAGE 1
The left part of the dialog box contains the elements and quantities that may be included in
the printout. Here one finds: nodes, bars, quantity survey, loads, combinations, reactions,
displacements, forces and stresses. The following three types of table printout are possible
for reactions, displacements, forces and stresses: for the Values tab, for the Envelope tab and
for the Global extremes tab. The following options are not available in the current version of
the program: Structure view, Diagrams, Detailed analysis.
On the left of each of the above-mentioned options, one finds a selection field which informs
whether the option is to be included in the simplified printout or not. The field may assume
one of the following values:
The element / quantity will not be included in the simplified printout
the element / quantity will not be included in the simplified printout for complete
selection
the element / quantity will not be included in the simplified printout with the user-
defined selection of nodes, bars, cases or modes.
STAGE 2
If the third possibility is selected, the radio buttons located on the right of the above
mentioned options will become useful. They may be included in the simplified printout. If an
element or quality is included in the simplified printout with the user-defined filtering, the
appropriate button is activated and the relevant edit fields are available for the user to define
lists of nodes, bars, cases or modes.
There are two ways to filter nodes, bars, cases or modes for which the printout is to be
composed:
inserting numbers of nodes, bars, cases or vibration modes in the relevant edit field
pressing the appropriate button, which results in opening the Selection dialog box.
structures.com
281
STAGE 3
If the Sorting option is active, all the tables available for simplified printout will be sorted
according to the user-defined criterion. The present version provides the possibility of sorting
only according to the defined groups of nodes and bars.
8.3 Page Setup
Once the File/Page Setup command is selected or the Page Setup button in the Printout
Composition - Wizard dialog box is pressed, the dialog box shown below will be displayed
on the screen. It allows you to define parameters for the page setup.
The dialog box contains four tabs: Page, Margins, Header/footer and Parameters:
Page tab - used for definition of paper size (A4, Letter, etc.), orientation (portrait, landscape)
and the current page number.
Margins tab - as in each text editor, used for definition of margins: top, bottom, left, right,
gutter and margins from the edge to the footer and the header.
Headers tab - used to specify whether title pages, headers, footers and tables of contents are
to be displayed. A few header and footer presentation options can be chosen
here: none, frame and separation lines. However, the Robot Millennium
system provides standard headers and footers which can be modified once
the appropriate Edit button is pressed for header, footer, or title page.
Apart from the Edit buttons, there are also Restore buttons beside the following options: Title
page, Header, Footer and Table of contents. Pressing these buttons results in restoring the
default values in the files defining the printout title page, header, footer and table of contents,
respectively. The Restore button located in the Frame field restores the default settings of
printout frames.
The user may save the values of the parameters he adjusted on the Header tab, in order to
use his personal header settings. It is done by means of the Select Template option, where
one can choose a file containing user-defined settings of parameters available on the Header
tab. Saving modified parameter settings is carried out by writing a name in the Select
Template field, followed by pressing the Save button. It is also possible to remove a template
from the list by selecting it and pressing the Delete button.
Parameters tab - contains all variables defined in the system, accompanied by their names.
structures.com
282
APPENDICES
structures.com
283
APPENDIX 1
Conversion of Loads to Masses
Introductory remarks:
Loads, in great majority, originate from the gravity (masses). Thus dynamic calculation needs
these masses to be taken into account. To enable the user an easy conversion of static loads
(gravity loads) into masses the special command MASses ACTive was applied into text file
analyzer.
This will allow the users to define load only once for the purpose of static analysis and then to
use them to create mass distribution over the computational model of the structure to perform
any dynamic analysis.
Command needs two elements to successful conversion. The first is the set of directions in
which masses are be active. Usually all global directions (X, Y, Z) are used, because only in
specific calculations inertia acts not on all of them. The second is the inertia magnitude. This
is defined by the static load case number, and the direction of the loads, which are be taken
into account during conversion. Additionally, an extra coefficient may be given to multiply the
load value.
The character of the load is automatically transformed into the masses: concentrated forces
are transformed into concentrated masses, moments - into rotational inertia, distributed forces
- into continuous masses.
Syntax:
ANA [ DYN | MOD | TRAN | HAR | SEIsmic | SPEctral ].,( concerns all the dynamic analysis
types)
CASe (#<number> <name>)
MASess ACTive [X/Y/Z]
[X|Y|Z ] (MINus|PLus) <case_list> COEfficient=<c>
NOTE: The above syntax has been introduced into the text file only (is erased from
the data file after its successive saving within the program).
General principles:
Let = (x) be a function of the mass density distribution within given element while N(x) be
the nodal interpolating function matrix (shape function matrix). As a base of further treatment
consistent mass matrix of an element will be created according to the following general
formula (1.1) :

e
T
e
d x N ActDir x x N M
e

(1.1)
where:
active is direction global the i when

inactive is direction global the i when
ActDir
i
, 1
, 0
0 0
0 0
0 0
3
2
1
,
The global direction activity flags are set by MASess ACTive [X/Y/Z], i.e. any direction is
active if specified. This is the consequence of general ROBOT style of the mass treatment,
where some components of inertia forces may be neglected during the analysis.
Mass matrix will be created from all the loads belonging to all load cases specified in
<case_list> acting on current element/node according to the following rules:
Each load record from specified case is converted to the mass separately and independently
from other loads and masses. Only simple load cases (no combinations !) may appear on the
structures.com
284
list (but in one dynamic case the list of static cases may be given to be converted into
masses).
Total mass matrix is created as a sum of mass matrices from all above load components and
from predefined mass due to dead weight of the structure and/or specified element masses.
Thus also part of a mass matrix originated from loads will be submitted to diagonalization
and/or negligence of rotational inertia part if specified by CON|LUM, ROT setting.
The value of the density function in given point is created as the value of the projection of the
current force vector f on the vector of uniquely and obligatorily specified global direction

Z
Y
X
1 , 0 , 0
0 , 1 , 0
0 , 0 , 1
) | ]{ | | [
if
if
if
v PLus MOins Z Y X
T
x f v x
T
(1.2)
Only positive values are taken into account in each integration point, thus
c x x * , 0 max (1.3)
NOTE: Above rules are intended to allow an easy selection of loads originated from
gravity. As nothing like default gravity direction exists, signed direction
specification must be done by the user.
All directions used to define load to mass conversion must be acceptable for current general
structure type, thus PLAte accepts only Z direction, for other plane types only X and Y will be
accepted. Incompatible direction specifications will be ignored or error will be reported. 3D
structural types accepts all global directions.
Example:
Consider a beam, loaded centrally by the gravity load Fy= -120 kN. Let the static case shown
below has a number 3. To calculate free vibration modes of this structure in the case number
10, taking into account this mass (Fx=Fy= 12 232 kg) one can use the following command:
ANA MOD=3 MAS=CON
CAS #10 modal
MASses ACTive X Y
Y MINus 3
Details of conversion for different load types
Loads acting on beam elements
uniform element load
[Px=<px.>/Py=<py>/Pz=<pz>] (LOCal/GLObal) (PROjected) ([R=<r>])([R=<r>])
The load density vector is transformed to global directions as specified by setting :
(LOCal/GLObal) (PROjected) ([R=<r>]), taking into account (PROjected) flag as for load
treatment, then uniform mass distribution is set according to (1.2) (1.3)dead load
Dead load is converted to mass equivalently to the uniform element load
NOTE: This operation should be used with caution, as mass originated from dead
load of the structure is taken into dynamic calculations automatically (if only
material density is greater than 0).
structures.com
285
variable element load
(X=<x1>)[ P=<p1>] ((JUSque)(X =<x2>)[P=<p2>] ) (R=<r>) (LOCal/GLObal) (RELative)
(PROjected)
load is transformed to global directions as specified by setting :
(LOCal/GLObal) (PROjected) ([R=<r>])
then uniform mass distribution is set according to (1.2)(1.3)
NOTE: Rule (3) implicate the following treatment of variable sign load, for each load
record (component) separately (not for the total load being the sum of all
loads acting on given element), as shown in Fig.1.1.
- ignored
+ valid
Fig.1.1
concentrated element force
[X=<x>] [F=<f>](R=<r>) (Local)(RELative)
The total mass m concentrated in a point x
0
is evaluated from global representation of force
vector f as follows:
m = max( 0,
T
f ) * | c | (1.4)
Consistent mass matrix is then evaluated, as if mass distribution would be represented by
Diracs delta function leading to:
M
e
= N
T
(x
0
) m [ ActDir ] N(x
0
) (1.5)
concentrated element moment
[X=<x>] [F=<fc>] (R=<r>) (LOCal)(RELative)
As mass direction specification does not concern directions of rotational inertia, thus separate
rule should be established to perform the conversion between concentrated element moment
and rotational inertia of a certain body attached to the element.
Vector style transformation of <fc> is performed according to (R=<r>) (LOCal) settings to
obtain a vector I referred to element local co-ordinate system. To omit necessity of
inconsistent vector style transformation (while tensorial one should be used), load should be
given as LOCal and no R=<r>, otherwise the warning will be issued.
There is assumed that element local co-ordinates coincide with principal axis of inertia of the
body, thus I
T
= [ I
XLoc
, I
YLoc
, I
ZLoc
] represent principal inertia moments in element local co-
ordinates. From this results the following modeling limitation:
structures.com
286
Correct situation
Incorrect situation, modeling impossible
Fig.1.2
distributed element moment
[M=<m>] (LOCal)
In this definition, <m> is a vector, which, after vectorial style transformation to element local
co-ordinate system, represents densities of rotational inertia referred to element local axis per
element length.
All notions as for concentrated element moment, (see Fig. 1.2), holds.
Loads acting on surface elements
uniform element load
[Px=<px.>/Py=<py>/Pz=<pz>]
Load density vector is evaluated, then transformed to mass density according to (1.2)(1.3)
dead load
Dead load is converted to equivalent uniform load and further treatment as above.
NOTE: This operation should be used with caution, as mass originated from dead
load of the structure is taken into dynamic calculations automatically (if only
material density is greater than 0).
variable element load
[P=<p1>] AU <n1>( [P=<p2>AU<n2> ([P=<p3> AU<n3>))
In each integration point load density is evaluated, then transformed to mass density
according to (1.2)(1.3), see Fig. (1.1). Enhanced integration rules are used with
NGAUS= 3x3 for Q8,
= 7 for T6,
= 2x2 for Q4
= 3 for T3
variable load inside a contour
[P=<p1>] AU <n1>( [P=<p2>AU<n2> ([P=<p3> AU<n3>)) PROjected DIRection <v> _
CONtour <l_node>
structures.com
287
In each integration point load density is evaluated, then transformed to mass density
according to (1.2)(1.3), see Fig.(1.1). In the case when not whole area of the element belongs
to the contour, fully automatic integration over the up to 100x100 point mesh is performed
within element, to reach required accuracy of integration. Thus using this option may
sometimes slow down the mass matrix evaluation process.
variable load along the line
LIN
<n1>[P=<p1>] Jusque <n2> (P=<p2>) ( [LOCal (GAMma=<gamma>)] )
Only translational force may be converted to element mass distributed along the line.
3-point Gauss type integration rule is used on each in segment of the line crossing the
element. In each integration, load density vector is transformed to global co-ordinate system,
then treated according to (1.2)(1.3) to evaluate mass distribution along the line.
concentrated load on auxiliary point
NODe (auxiliary)
F=<f> ( [R=<r>] )
Only translational force may be converted to element mass ( for beam elements moment -
rotational inertia conversion was allowed, here is prohibited). Force vector <f> is transformed
if necessary to global co-ordinate system and then treated according to (1.2),(1.3) to evaluate
the mass value attached to the point within the element, then the mass matrix is evaluated
using (1.5). The element to which mass will be attached is searched automatically.
Nodal loads
concentrated force
NODe
F=<f> ( [R=<r>] )
Force vector <f> treated according to (1.2), (1.3) to evaluate the nodal mass value.
concentrated moment
NODe
F=<c> ( [R=<r>] )
As mass direction specification does not concern directions of rotational inertia, thus separate
rule should be established to perform the conversion between concentrated nodal moment
and rotational inertia of a certain body attached to the node.
Vector style transformation of <fc> is performed according to (R=<r>) setting to obtain a
vector I referred to global co-ordinate system. To omit necessity of inconsistent vector style
transformation (while tensorial one should be used), no LOCal should be given as and no
R=<r>, otherwise the warning will be issued.
There is assumed that global co-ordinates coincide with principal axis of inertia of the body,
thus
ZLoc YLoc XLoc
T
I I I I , , represent principal inertia moments in global local co-
ordinates.
NOTE: This rule is different than those used in case of concentrated mass attached
to beam element.
structures.com
288
APPENDIX 2
New Bar Elements (Non-linear Analysis in the Robot Program)
Designations adopted:
E - Youngs modulus
G - shear modulus
- Poissons ratio
fd - limit of elasticity
Ax - cross section area
Ix - torsional constant
Iy - moment of inertia - bending in XZ plane
Iz - moment of inertia - bending in YZ plane
ky, kz - correction coefficients for shear rigidity in Y and Z directions
L - bar length.
1. Preliminary remarks and assumptions
The following assumptions have been adopted for bar (beam) elements:
Uniform formulation for 2D and 3D (2D & 3D frames, grillages)
Uniform element allowing for material and/or geometrical non-linearity
Standard displacement degrees of freedom at 2 extreme nodes
T
z y x z y x
u u u ] , , , , , [ } , { u d
Use of the following is allowed:
- shear deformation included (Timoshenkos model)
- tapered cross section - only for geometrical non-linearity.
- Winklers ground
There are 2 levels of geometrical non-linearity available:
Non-linearity (second order theory)
P-DELTA the most accurate theory possible - large displacements and rotations
(incremental approach with geometry update - Updated Lagrange Description)
Assuming small displacements and absence of physical non-linearity, for the limit, the results
are identical as for standard linear elements
In the material non-linearity analysis the layered model and the constitutive stress-strain
principle for the uni-axial stress-strain on the point (layer) level are applied
Shear and torsion states are treated as linearly elastic and have to be uncoupled from axial
forces and bending moments on the cross section level
Non-linear releases and hinges may be defined only as DSC elements
All types of element loads are allowable (identically as for standard elements). However, it is
assumed that nodal forces acting on a structure are determined at the beginning of the
process (i.e., the changes in the transfer of element loads onto nodes resulting from
geometrical or material non-linearity are ignored)
Apart from the elasto-plastic element, it is also possible to generate elasto-plastic hinges in
selected bar cross sections as an extension of the non-linear hinges option (see point 5)
structures.com
289
2. Geometry, kinematics and strain approximation
Geometry, sign convention for forces, displacements, stresses and strains
Basic kinematic relationships
In the element local system, in the geometrically linear range, the generalized strains E on the
cross section level are as follows (symbol
x
), ( indicates calculation of differential along the
direction of the bar axis):
T
z y z y ox
} , , , , , {
where:
Axial strain in the bar axis:
x ox
u,
Curvatures:
x z z
x y y
,
,

Average angles (strain):

y x z
z x y
w
v

,
, ,
Unit torsion angle:
x x
,
Displacement approximation
In view of the possibility to consider shear influence and consistence of results obtained for
the linear element, so-called physical shape functions considering shear influence have been
implemented.
structures.com
290
2D bars:

10 9 8 7
10 9 8 7
3 1
6 5 4 3
6 5 4 3
2 1
0 0
0 0
0 0 0 0
0 0
0 0
0 0 0 0
, ) (
h h h h
h h h h
h h
h h h h
h h h h
h h
x N Nu u
Shape functions and their derivatives are expressed by the formulas:
] 3 ) 1 ( 2 [
) 2 1 (
1
] ) 1 ( [
) 2 1 (
10
] 6 6 2 [
) 2 1 (
1
] 2 3 2 [
) 2 1 (
1
9
] 3 ) 2 ( 2 ) 1 ( [
) 2 1 (
1
] ) 2 ( ) 1 ( [
) 2 1 (
8
] 6 6 2 [
) 2 1 (
1
)] 2 1 [(
) 2 1 (
1
7
] 6 ) 1 ( 2 [
) 2 1 (
1
] 3 ) 1 ( 2 [
) 2 1 (
1
6
] 12 6 [
) 2 1 (
1
] 6 6 [
) 2 1 (
1
5
] 6 ) 2 ( 2 [
) 2 1 (
1
] 3 ) 2 ( 2 ) 2 1 [(
2 1
1
4
] 12 6 [
) 2 1 (
1
] 6 6 [
) 2 1 (
1
3
/ 1 2
/ 1 1 1
,
2 3 2
2 3 2
2 3 2
2
2
2
2
2
2
2

L
L
L
L
L
L L
L
L L
L
L
h h i
x i i
where:
L
x

2 2
6
,
6
GAL k
EI
GAL k
EI
z
y
y
z
for planes XY and XZ, respectively.

Kinematic relationships for the matrix notation (the geometrically linear theory)
In general, when considering the influence of imposed strains
} , , {
T
z
T
y
T
o
o
E
Increment of generalized (sectional) strains:
structures.com
291
o
Loc L
u B E
Glo Loc
u T u ,
T - global local transformation matrix
2D:
2
1
10 6 9 5 8 4 7 3
6 5 4 3
2 1
, , 0 , , 0
, , 0 , , 0
0 0 , 0 0 ,
u
u
x x x x
x x x x
x x
y
z
ox
h h h h h h h h
h h h h
h h
3D:
2
1
2 1
10 6 9 5 8 4 7 3
10 6 9 5 8 4 7 3
6 5 4 3
6 5 4 3
2 1
0 0 , 0 0 0 0 0 , 0 0 0
0 , 0 , 0 0 0 , 0 , 0 0
, 0 0 0 , 0 , 0 0 0 , 0
, 0 0 0 , 0 , 0 0 0 , 0
0 , 0 , 0 0 0 , 0 , 0 0
0 0 0 0 0 , 0 0 0 0 0 ,
u
u
E
x x
x x x x
x x x x
x x x x
x x x x
x x
z
y
z
y
ox
h h
h h h h h h h h
h h h h h h h h
h h h h
h h h h
h h

where:

T
z y x z y x z y x z y x
T
z y x z y x
u u u u u u D
u u u u D
} , , , , , , , , , , , { : 3
} , , , , , { : 2
} , {
2 2 2 2 2 2 1 1 1 1 1 1
2 2 2 1 1 1
2 1

u u u
Strains at a point (layer)
Given the generalized strains
z y ox
, , of a cross section, the
xl
strain or its increment
xl
at any point of the cross section l - of the coordinates yl, zl, is calculated as
l z l y ox xl
y z
T
l l
T
l xl
y z } , , 1 { ; v E v
finally, strain increment in the layer:
) ( ) (
o T
l
o T
l xl
E u B v E E v
structures.com
292
3. Stresses and internal forces within an element
The constitutive principle on the point level
It is adopted in the general incremental form, where current stresses
1 n
x
are defined as a
function of stress for the last equilibrium
n
x
and current strain increment with imposed
(thermal) strains considered,
) , (
1 1
xl
n
xl
n
xl
F

based on the function = f() which describes the relationship in the process of active loading
and on the specification of the principle of unloading and reloading. In particular, it may be the
elasto-plastic principle with linear hardening and the specified principle of unloading, i.e. (a)
elastic, (b) plastic, (c) damage, (d) mixed. For elastic unloading the passive and active
process is performed along the same path = f(). For the remaining ones, it is performed
along the straight line determined by the beginning point of a given unloading process
} , {
UNL UNL
and the unloading module
UNL
D defined as
D UNL P UNL M UNL
n n
n
D UNL P UNL
aD D a D d
e
D c E D b
_ _ _ _ _
) 1 ( : ) ( ; : ) ( ; : ) (

.
n
e is a remembered strain, for which the current active process has started, commenced after
exceeding 0 by stresses with the unloading ( 0
1
e ) assumed.
For the analysis it is necessary to provide the current stiffness assumed to be a derivative
x
D
Calculation of forces and cross section stiffness values
On the cross-section level, the vector of internal forces (stress resultants) is composed of:
T
x z y z y x
T
y z x
M Q Q M M N D
Q M N D
} , , , , , { : ) 3 (
} , , { : ) 2 (
States of shear and torsion

ST
are treated as linearly elastic and not conjugated with the
state of axial /bending forces on the cross section level.
structures.com
293
x
n
x
n
x
z z
n
z
n
z
y y
n
y
n
y
GI M M
GA k Q Q
GA k Q Q
1
1
1
Compression/tension states
NM
are generally treated as conjugate when applying the
layered approach. However, as long as the elastic state is guaranteed, i.e. until the current
generalized strains fulfil the following elastic state condition:
1
ELA Z
Z
ELA Y
Y
ELA ox
ox
,
where:
)) /( ( min )); /( ( min ); / ( min
l l dl
l
ELA Z l l dl
l
ELA Y l dl
l
ELA ox
y E f z E f E f ,
the cross section is treated as elastic and the layered approach is not activated.
z z
n
z
n
z
y y
n
y
n
y
o
n
x
n
x
EI M M
EI M M
EA N N
1
1
1
Once violation of the elastic state condition is asserted, stresses induced by axial strains and
bending are calculated separately for each layer and on their basis sectional quantities are
calculated
Nlayer
l
l l l
z
y NM
Nlayer
l
l l
n
xl
n
z
Nlayer
l
l l
n
xl
n
y
Nlayer
l
l
n
xl
n
x
A
M
M
N
y A M
z A M
A N
1
1
1 1
1
1 1
1
1 1
v
Stiffness on the level of D cross section is calculated as follows:
in the elastic state as:
D = diag {EA, EIy, EIz, KyGA, kzGA, GIx)
After exceeding the elastic state condition as:
ST
NM
D
D
D
0
0
where:

Nlayer
l
l l l l
l l l l
l l
l l
Nlayer
l
T
l l l l NM
y z y y
z y z z
y z
A D A D
1
2
2
1
1
v v D
} , , {
x z y ST
GI GA k GA k diag D
structures.com
294
Nodal force vector and element stiffness matrix
They are calculated by means of the standard formulas applying Gauss quadrature
(Ngauss=3).

NGAUSS
iG
iG iG iG iG
T
L
T
dJ W x dx
1
0
) ( B B f

NGAUSS
iG
iG iG iG iG iG
T
L
T e
dJ W x x dx
1
0
) ( ) ( B D B DB B K
4. Geometrical non-linearity
The following configurations are taken into consideration:
0
B - initial configuration
n
B - reference configuration (the last one
for which equilibrium conditions are satisfied)
1 n
B - current configuration (iterated)
An entry point for the element formulation is the virtual work principle saved in the following
form for displacement increments:
u dV e F dV C dV
V
ij
n
ij
V
n
ij kl ijkl ij
n
ij

,
1
where: strain increment while moving
n
B to
1 n
B , e, constitute its parts -
correspondingly: linear and non-linear with respect to the displacement increment u,
whereas is a stress referring to the reference configuration and Cijkl is a tensor of
tangential elasticity modules.
The Non-linearity option
It corresponds to the non-linear formulation, i.e., to the second order theory. Since material
non-linearity is possible, the incremental formulation is being introduced, however, without
modification of element geometry.
Kinematic relations
Strain increments in the matrix notation:
g H g u e E
N
T
Loc
2 / 1
where:
T
x z x y x x x x x
w v u } , ; , ; , ; , ; , ; , { g then the displacement increment gradient g = u
x
, N
structures.com
295
whereas ) 3 (
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 1 0 0
0 0 0 0 1 0
0 0 0 0 0 0
); 2 ( ;
0 0 0
0 1 0
0 0 0
D D
N
H is a selection matrix.

L
N
T
s
L
T
L
NL
n
L
n
ext
n n n n T
ext
n n
L Loc
dx N
dx
dx
0
0
int
1
int
1 1 1 1 1 1 1
) ( H K
DB B K
f f f u K B f f
K K K
Algorithm on the element level

The element geometry is not modified, the local-global transformation is performed with the
use of initial transformation matrix T
0
Glo Loc
u T u
0
,
0
2 / 1 E Hg g u E
T
Loc
calculation of generalized strains

E , (
1 1 n n n
calculation of stresses (internal forces)
) (
1
n
K K

stress rigidity matrix
NL
n
L
n
ext
n
Loc
n
int
1
int
1 1 1
f f f f
Loc
T
Glo
f T f
0
K K K
L Loc
T K T K
0 0
Loc
T
Glo
P-DELTA option
It is a certain variant of bar description allowing for large displacements. The approach of the
updated Lagrange description is applied here.
structures.com
296
) 3 (
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0
0 0 0 0
0 0 0
), 2 (
0 0 0
0
0
) (
0
1
0
int
1 1 1 1 1
D
N M
N M
M M N
D N M
M N
dx
dx
dx
z
y
z y
y
y
L
n T
s
L
T
L
n
ext
n n T
ext
n n
L Loc
O
O

K
DB B K
f f B f f
K K K

Modification of the element geometry, local-global transformation performed using the current
transformation matrices T T
1
,
n n
Glo
n
Loc
u T u ,
0
) ( E u E
Loc NL
calculation of generalized strains

E , (
1 1 n n n
, calculation of stresses (internal forces) according to point 3
) (
1
n
K K

, stress rigidity matrix for current stresses
int
1 1 1

n
ext
n
Loc
n
f f f
geometry modification
Loc x
n n
u L L
1
updated element length
, , Eulers angles defined identically as for the element of the
coordinates } , , , 0 , 0 , 0 { w v L
n
, 2 / ) (
1 1 x x

); , , ( T T transformation matrix for the angles , ,
T T T
n n

1
Loc
T n
Glo
f T f
1
Transformation to the global system
K K K
L Loc
T K T K
1 1
n
Loc
T n
Glo
Once the equilibrium is achieved on a given increment step the stored data describing
element geometry is overwritten:
L L
n n 1
T T
1
n n
structures.com
297
5. Elasto-plastic hinges
Alternatively, the elasto-plastic work of a structure can be modeled by introducing non-linear
hinges at selected bar cross sections. Characteristics of a hinge represented by a 2-node
DSC element are defined applying the cross section analysis algorithm described in point 3,
assuming that the role of generalized strains E is played by mutual node displacements (with
respect to bar local directions) divided by the adopted (fictitious) element length (L) that
equals the minimum cross section height, which act as the element volume dV=L. Forces
and displacements of newly-generated nodes of the DSC element constitute global degrees
of freedom, in other words, they do not undergo condensation.
- calculation of generalized strains in a cross section
Glo
u B E
- calculation of internal forces (stress resultants) and cross section rigidity according to point
3.2
) , (
) , (
1
1
E D D
E

n n
n n
- calculation of forces (reactions on bar ends) and DSC element rigidity
L
L
n T n
n T n

B D B K
B f
1 1
1 1
where:
) 3 ( ;
) 2 ( ;
1 0 0
0
0
; ,
3 3
3 3
D
D c s
s c
L L
x
x

T 0
0 T
T T
B
structures.com
298
APPENDIX 3
Theoretical Background of the Dynamic Methods
Implemented in the Robot Millennium Program
THE DYNAMIC ANALYSIS METHODS
IN ROBOT Millennium
THE SHORT INTRODUCTION INTO
THE BASIC IDEAS AND ALGORITHMS
Introduction
This document presents the description of the dynamic analysis methods applied in Robot
Millennium. The theoretical background details and examples are included in the appendices
- to simplify the first reading. This document is not a users manual and is not aimed at
familiarizing the user with the details of Robot interface. The aim of the current document is to
expound the main ideas, which are realized into this program.
Most of the dynamic methods in Robot Millennium is based on modal analysis results. It is
necessary to understand that modal analysis methods depend on a selected type of solver.
For skyline solver the following methods are available: block subspace iteration (BLSI)
method, subspace iteration (SI) method
1
, Lanczos method and basis reduction method. The
methods available for sparse direct solver
2
include: block subspace iteration (BLSI) method,
Lanczos method and basis reduction method. Whereas for iterative solver the following
methods are available: modified Lanczos method (pseudo mode - see 3.5 and appendices
3A, 3B), Ritz-gradient (PCG_Ritz) method and preconditioned conjugate gradient (PCG)
method.
Sparse direct solver (SPDS) is a specific form of Gauss elimination. Application of such a
method is strongly recommended for analysis of medium-sized and large-scale problems (10
000 - 200 000 equations)
3
. It is a good alternative for iterative solver.
3.1. The Modal Analysis Methods
The modal analysis comprises two basic approaches. The eigenproblem analysis
0
2

k k k

M K k = 1,2,,N
(3.1)
1
SI method has been developed in Robot as a first method of solving an algebraic eigenproblem. Such
a method is slow and it is strongly recommended to apply BLSI method or Lanczos method instead of it.
2
Temporary sparse direct solver involves the following limitations:
Sturm sequence check, Upper limits option, accelerations by shifts during BLSI are not available for
sparse direct solver. A missing eigenvalue can be partially controlled by means of the BLSI method.
The method is not available for modal analysis recognizing static forces
3
In the current version for small problems, when all matrices can be allocated in RAM without block-by-
block disk storage, skyline technique is faster, because usual matrix-vector product procedure acts
faster than EBE procedure in SPDS technique. The situation is different, when size of the problem does
not allow full allocation of all matrices in RAM SPDS technique is then drastically faster. That
disadvantage of SPDF technique for small problems will be eliminated in the next version.
structures.com
299
is produced by the definition of the eigenvalues
k
and eigenvectors
k
. It is the first
approach, familiar to engineers. The second approach consists in generation of basis vectors

N
q q q Q

,..., ,
2 1
(3.2)
and search of the Ritz approximations
k
~
,
k
~
(k=1,2,.,N). It is based on the idea similar to
the method of Load depended Ritz vectors, proposed by E.L. Wilson [1, 3] and applied into
SAP2000. This approach is applied for seismic analysis and it is a powerful method when
there are great difficulties with obtaining sufficient mass percentage (see section 3.5).
The methods of modal analysis (first approach), which are concerned with definition of the
eigenvalues
k
and eigenvectors
k
, are discussed in this section.

Block subspace iteration (BLSI) method, subspace iteration (SI) method, selective
orthogonalization Lanczos method and basis reduction method (see appendix 3A) are used
when direct solvers (skyline or SPDS) are selected. Subspace iteration method is usually
slow. Therefore, application of BLSI or Lanczos method is strongly recommended for
analysis of medium-sized and especially large-scale problems, when a large number of
eigenpairs is required. Basis reduction method may prove very effective for an experienced
engineer, however, it requires additional information about basis nodes and appropriate basis
directions.
Preconditioned conjugate gradient (PCG) method is used in the first approach, when iterative
solver is selected. Such approach may be very efficient, when small number of eigenmodes
are extracted (at the most 5). It should be used for wind analysis rather than for seismic
analysis. PCG method may be used for estimation of the lowest eigenmode when a large-
scale problem is met.
The second approach (which is realized by Pseudo mode Analysis mode) is presented in
section 3.5.
The block subspace iteration method (BLSI) [1,3] is more general than Lanczos method,
since it allows implementation of all types of mass matrices (see section 3.2) and is capable
of analyzing separate structures. Iterations in a block of the constant size with immediate
exclusion of converged vectors and addition of new start ones usually ensures faster
computations compared to usual subspace iteration method [1-3]. Just as Lanczos method,
BLSI can be applied to extract a large number of eigenpairs (till 100 200).
The subspace iteration method (SI) can be used for analysis of all types of mass matrix [4]
and for analysis of separate structures, however, in case of a large number of required modes
(approximately N > 10), this method is still very time-consuming, especially for large-scale
problems.
The Lanczos method [12,16,17] is a powerful method allowing one to obtain a large number
of eigenpairs (N ~ 20 500 and more). Although it is preferable for large-scale problems, it
involves the following limitations:
It is impossible to analyze separate structures
The mass matrix type should be lumped with rotations or consistent
It is impossible to disregard the material density of (in this case it suffices to assign a
fictitious small density to avoid these limitations).
The basis reduction method [5] is known as the improved Rayleigh-Ritz method [4] or
Bubnov-Galerkin method for discrete systems. This algorithm allows one to get approximate
values of the first few eigenpairs if the user has some information about them. This method
requires assigning of a master degree of freedom (MDOF) in order to get the reduced system.
Thus, the user can control the process of creating the reduced model. It is a powerful tool for
users who have some experience in dynamic analysis of structures and deal with the same
type of structures whose behavior is known. This method allows one to exclude undesirable
structures.com
300
degrees of freedom (DOF) from the reduced model and to reduce the initial complex problem
with a large number of DOF to a reduced form - with a considerably smaller number of DOF.
The experience concerning structure dynamic analysis shows that some problems can be
encountered by the user when the automatic reduction methods (BLSI, SI and Lanczos
methods are taken into account) lead to a very complex computation process. For example,
the local vibration modes of single bars can lead to such serious problems for these methods,
since the computation process seeks the eigenpairs automatically without any selection. It
should be noted that for the majority of cases in real structures these local vibrations will be
restricted by some constraints that are not taken into account in the FEM model, or their
contribution will be inessential to the overall system motion. Usually, the mass percentage is
very small for such local vibrations. The usage of exact methods in this case will lead to the
above-mentioned difficulties, however, the implementation of the approximate basis reduction
method can simplify the computation process considerably.
The preconditioned conjugate gradient (PCG) [9-13] method is available for iterative
solver. Application of such a method is recommended for extracting of a small number of
eigenpairs when a large-scale problem is met. Implementation of the Pseudo mode (see 3.5)
with modified Lanczos method or PCG_Ritz method is recommended if it is necessary to
determine a large number of modes while running seismic or spectral analysis and iterative
solver is selected.
The Ritz-gradient (PCG_Ritz) method [8] is available for iterative solver in pseudo mode. It
allows one to produce an approximate solution in the terms of Ritz vectors. It is a very fast
method for seismic and spectral analysis of medium size (10 000 100 000 equations).
Modified Lanczos method is an extension of Lanczos method when iterative solver is
applied. It acts like a usual Lanczos method in the pseudo mode, however, being different
from Lanczos method for direct solvers, it does not require factorization of a stiffness matrix.
Instead of that the principles of preconditioned gradient method are implemented. Such
approach is the most robust among all the dynamic methods of iterative solver, though it often
appears not to be the fastest one.
Details of all dynamic methods are presented in appendix 3A.
3.2. Mass Matrix Types
The Lumped without rotations, Lumped with rotations and Consistent mass matrices of
dynamic analysis can be applied to a structure.
The Lumped without rotations and Lumped with rotations are the diagonal mass matrices.
These types of mass matrices require minimum computational effort.
The Consistent mass matrix appears when the user wishes to consider a system with the
distributed parameters. It is commonly believed that a consistent mass matrix describes
inertial properties of a structure more exactly than the lumped one. However, in most cases
the lumped mass matrix provides a good approximation, since it is obvious that the inertial
parameters can be presented less precisely than the stiffness ones. In fact, that kinetic
energy is described as displacements of a structure, but potential energy is expressed
through spatial derivative of displacements. It is a well-known fact that approximation error
increases considerably during each differentiation [4]. Thus, for continual objects (solid,
shells, plates), it is possible to approximate the mass parameters less precisely than the
stiffness ones for the same mesh.
Usually, Hermit polynomials are used as shape functions for bars. It is an exact solution for
most of the static problems and the dynamic problems when lumped mass matrix is
considered. However, exact solutions for dynamic problems of a bar with distributed masses
belong to the class of Krylov functions (it is a specific combination of hyperbolic and
trigonometric functions). It enables the stiffness parameters in such case to be presented
approximately when Hermit polynomials are used simultaneously with a consistent mass
matrix. (Let us take note that, in fact, it is not intended for implementing a different type of
shape functions for static and dynamic problems). Therefore, for most cases it is not a great
benefit to complicate the dynamic model by the use of distributed mass parameters, since the
approximate solution with consistent masses occurs instead of the exact solution for an
approximate model (lumped masses).
structures.com
301
Moreover, usually own masses of bar structural elements (girders, columns, etc.) are
negligible compared to masses of walls and roof (dead load), which are taken into account
through the technique of conversion of dead loads to masses. Such non-structural masses
usually reduce the effects of distributed element masses.
All that was mentioned above leads to the following conclusion: for most practical cases the
lumped mass matrix ensures a sufficiently precise approximation of structure inertial
properties. It should be remembered that a consistent mass matrix requires considerable
computational efforts, if a large-scale problem is analyzed. It should be certain that
implementation of a consistent mass matrix will be justified before selection of such a type of
matrix for analysis.
It is assumed that the mass matrix must be Consistent, if the rigid links are used into
computation model.
If sparse direct solver or iterative solver is applied, element-by-element (EBE) technique is
used for computation of matrix-vector product. It means, that the consistent mass matrix can
never be assembled, however, all operations are performed only on the element level. For
skyline solver, a consistent mass matrix is assembled and stored in the same way as a
stiffness matrix. For small problems (at the most ~3000 equations) skyline technique is faster,
although it still drastically time-consuming when the size of a problem increases.
It is possible to use the additional concentrated masses (added masses), and to convert static
loads to masses.
When the Lanczos, PCG_Ritz or modified Lanczos method (iterative solver) are selected,
only the Lumped with rotation and Consistent mass matrices are available.
3.3. Upper Limits
It is possible to calculate all eigenvalues and eigenmodes that do not surpass the user-
defined value. This value is treated as the upper limit. When the upper limit is activated,
Robot searches
*
2 1
,....., ,
n
, where
*
is the upper limit. The algorithm works in
two steps. The Sturm sequence check is performed in the first step. It defines the number of
eigenvalues n which is smaller than the upper limit. In the next step, the algorithm
generates n eigenpairs, each one smaller than the upper limit.
Lanczos method and BLSI method are recommended for the type of analysis that uses upper
limits, since usually it is necessary to obtain a large number of eigenpairs.
The criterion of mass participation percentage (see section 3.4) is ignored when upper limit
is activated.
For example, such problem may arise when a user deals with the French seismic code PS-
92. It is required that all frequencies smaller than 33Hz should be taken into consideration.
3.4. Mass Participation Percentage
It is possible to activate the criterion of mass participation percentage. The mass percentage
for each mode (k=1,2,,N) is defined as

dir
dir
k dir k
M m /
2
,
,
where
dir
T
k
dir
k
I

M ,
dir
k
is the mass participation factor for k eigenmode,
dir
I
is the
vector of unit translation into the direction (dir = X,Y,Z),
dir
T
dir dir
I I M

M is the total mass
into direction dir,
T
k
is k-th eigenmode, 1
k
T
k

M .
The mass percentage for direction dir equals M%dir

dir
N
k
dir
k
M
1
2
. It defines the contribution
of all the modes involved in motion of a structure in the considered direction.
structures.com
302
If the Modal Analysis mode is selected and the mass percentage for a specified maximum
number of nodes is smaller than required, the user receives a message about the
unsatisfactory mass percentage, while calculations are continued without any corrections.
It is necessary to set the Seismic or Pseudo mode Analysis mode to ensure automatic
search for the required mass percentage. Details are presented in section 3.5.
3.5. Analysis Modes
The following dynamic analysis modes (regimes) will be presented in this section: Modal,
Seismic, Pseudo mode.
Several seismic codes (UBC-97, French code PS-92) require that sum of masses for each
direction (or for horizontal directions only) should be not less than 90%. Some hard
problems may arise, when it is very difficult to achieve the required sum of masses due to
small contributions of a large number of the lowest modes. Usually, this problem is caused by
local character of the lowest modes. The Seismic mode and Pseudo mode are presented to
improve the situation of such difficult problems. The effectiveness of such approaches is
illustrated by appendix 3C. Lanczos method for direct solvers is available for these two
modes. Modified Lanczos method and PCG_Ritz one are available for pseudo mode, when
iterative solver is selected.
1. Modal mode
This mode constitutes a well-known approach implemented in the previous versions of
Robot.
Available methods: BLSI, SI, Lanczos and Basis Reduction Method for direct solvers and
PCG method for iterative solver.
Convergence criteria for direct solvers: iterations will stop when tol
k
i
k
i
k
i

/
1
,
where i= 1,2,N, k is the iteration number, N - number of modes (defined by the user ). The
basis reduction method does not produce the convergence check, since it is not the iterative
approach, although it is a kind of Ritz method. It needs increasing the number of master
DOFs to improve the result precision.
Convergence criteria for PCG method (iterative solver):
tol
k k k k k k k

M M K M r / / , where N k
k k
,... 2 , 1 ,
2

The details are described in appendix 3A.
Upper Limits is the lower bound value for period, frequency and pulsation;
if this parameter is different from 0, all the sequential eigenpairs from 0 to Upper Limit will be
computed.
Mass% - the mass percentage (sum of masses for all computed modes for each direction)
Sturm check is a verification of skipped eigenpairs between 0 and shift parameter
and consists in counting the negative elements on the diagonal of the decomposed shifted
matrix
1
) (
M K
It is very expensive procedure for large-scale problems. Let us take note that for seismic and
spectral analysis it is not necessary to get the continuous spectra of eigenvalues. It is
important only to ensure the sufficient modal mass percentage for each direction. If such a
condition is fulfilled, completeness of the basis is ensured. Take note that implementation of
BLSI method allows one to produce the partial verification of continuity of eigenvalue spectra
without running Sturm check see description of BLSI method.
structures.com
303
Number
of
Modes
Upper
Limits
Mass% Program behavior
N 0
(inactive)
0
(inactive)
Sturm check is checked. Such option ensures that skipped
frequencies for first N modes are missing. It is available for
direct solvers when BLSI, SI or Lanczos methods are applied.
It is not available for basis reduction method and for all
iterative solver methods. Define N first sequential
eigenmodes. Sturm check is performed. If skipped
frequencies are detected, user receives a warning about the
number of skipped frequencies.If the user answers:
Yes, then iteration process is continued while the number
of skipped eigenpairs is determined. Afterwards, Sturm
check is repeated.
No, then converged eigenpairs are saved as a final result
and the next case is calculated
Cancel, then iterations are continued while all skipped
frequencies are determined. Warning is ignored.
Sturm check is unchecked - Sturm check is not performed
N
Inactive
(due to
active
Upper
Limit)
It is available only for direct solvers and for BLSI, SI, Lanczos
methods. It is not available for basis reduction method and for
all iterative solver methods.
Sturm check is performed at the start of computations the
number of frequencies N1 which are contained between zero
and Upper Limit is obtained:

1
... 0
2 1 N
If (N1 > N), user is warned about the number of frequencies
N1. If the user answers:
Yes, it means: compute

1
... 0
2 1 N
No, it means: stop computations
If(N1 <= N), compute

1
... 0
2 1 N
without
any warnings.
In both cases it is possible to derive a number of converged
eigenpairs higher than N1, but they will be saved as final
results only when:

1
... 0
2 1 N
. All the
converged egenpairs that are larger than
, will be lost.
N Inactive
(due to
active of
mass%)
Active:
0<
mass%<
=
100%
It is available for all methods of direct solvers. It is not
available for iterative solver. If mass% is not satisfactory, a
relevant message appears. No corrections are performed.
Otherwise, computations are performed in the same manner
as for the first case.
2. Seismic mode
Such mode is available only for direct solvers skyline or sparse.
In case of seismic and spectral analysis, it is not important to use the sequentially ordered
eigenpairs, since only the eigenpairs that contribute considerably to the seismic response
(e.g. they have significant mass participation factor) should be taken into account. Thus, the
Sturm Check is not performed.
structures.com
304
In general, the Lanczos method usually ensures convergence of the number of eigenpairs
considerably greater than N eigenpairs in the sequentially increasing order. When the user
wants to restore the skipped eigenvalues, it is necessary for him/her to get the number of
converged frequencies considerably greater than the first N desired frequencies. For
example, it is typical that the Lanczos method yields the following convergent frequencies
21 18 13 12 10 1
, , , ,
When the user requires the sequentially ordered eigenpairs, he will obtain only the first 10
ones. The last 4 eigenpairs are simply thrown away, as well as the correspondence mass
contribution. The essence of the proposed Seismic mode is to take all converged eigenpairs
into account (not only the first sequential ones). It ensures a bigger sum of masses compared
to Modal mode.
Available methods: Lanczos Method.
Convergence criteria: tol r r
i i i i i i

/ ;
1 2
M K ;
i
is i-th eigenvector; there is

adopted tol=1.0e-02 for current version.
Upper Limits are ignored.
The current mass percentage is defined as an average value of M%x, M%y, M%z for 3-D
problems and as a minimum value of M%x, M%z for 2-d problems (M%x, M%y, M%z are the
current sum of masses for x, y, z direction, respectively). This strategy is explained by the fact
that it is usually very difficult to ensure sufficient mass percentage for vertical direction. It is
possible to verify the mass percentage for each direction in the final results.
Nmodes Upper
Limits
N Inactive Inactive
(0)
Compute the N unsequenced eigenpairs. Sturm check is not
performed. The number of converged eigenpairs always
equals N.
N Inactive Active:
0<
mass%<
=
100%
N is ignored. The computations continue, until the current
mass percent is not smaller than the demanded mass%, or
the user is prompted by the appropriate message to stop the
computations, or the number of converged eigenpairs
achieves the maximum available value. This value set is
defined internally and it is adopted to equal 100 in the current
version.
After each 20 Lanczos steps program re-computes the
number of the converged eigenpairs and modifies the
current mass percentage. The user receives a warning
message about the achieved mass percent. The user may
respond to the message in the following ways:
Yes: continue computations during the next 20
Lanczos steps and display this message again, if
required mass percentage is not achieved (and the
number of converged eigenpairs does not exceed the
maximum available value)
No: save the converged eigenpairs as final results
and pass them to the next case
Cancel: ignore all warnings in the future and continue
computations.
3. Pseudo mode
Such option is available both the direct and iterative solvers.
This mode is recommended only for the seismic and spectral analysis when the Modal and
Seismic modes are still very time consuming. The Modal mode and Seismic mode use the
eigenmodes as the basis vectors for presenting seismic response and it is possible to require
a very large number of eigenmodes to ensure a sufficient sum of masses for some difficult
problems. The Pseudo mode rejects this idea and generates the Ritz approximations to the
lowest eigenpairs by means of Lanczos vectors for direct solvers or Ritz-gradient method [8]
and modified Lanczos method for iterative solver. This is a more effective way of operation in
structures.com
305
most cases, because a smaller number of basis vectors is required than in case of modal
mode. It was shown by E.L.Wilson [1-3]. Pseudo mode is similar to the Load depended Ritz
vectors, proposed in [1-3] and applied in SAP2000. It should be noted that the French code
PS-92 admits the application of the scientifically-based approaches to add some systems of
basis vectors to the existing eigenmodes, in order to increase the sum of masses. The details
of pseudo mode approach and it effectiveness are presented in appendices 3B and 3C,
respectively.
Upper Limits are ignored.
The current mass percentage is defined as an average value of M%x, M%y, M%z for 3-D
problems and as a minimum value of M%x, M%z for 2-d problems (M%x, M%y, M%z are the
current sum of masses for x,y,z direction, respectively). This strategy is explained by the fact
that it is usually very difficult to ensure sufficient mass percentage for vertical direction. It is
possible to verify the mass percentage for each direction in the final results.
Nmodes Upper
Limits
N inactive Inactive
(0)
It is available for both direct solvers and the iterative one
Generate N basis vectors to define the work subspace. Save
N basis vectors to use them for seismic and spectral analyses.
This regime is recommended.
N Inactive Active:
0<
mass%<
=
100%
It is available only for direct solvers
Generate N basis vectors to define the work subspace. Save
these basis vectors which suffice to satisfy the given mass%.
The number of saved basis vectors is less than N,
if mass% < 100%
3.6. Spectral Analysis
The Response Spectra Method is applied for seismic and spectral analysis. The main idea of
this method consists in decomposing a structure of multiple degrees of freedom (MDOF) into
a system of the single-degree of freedom (SDOF) oscillators. Then, the program calculates
the response for each of these independent oscillators and the statistical summation of the
extreme responses for each oscillator by means of SRSS, CQC, ten percent and double sum
methods [3, 21].
The eigenmodes define this system of SDOF oscillators, when Modal mode or Seismic mode
are applied. The pseudo mode basis vectors define this system of SDOF oscillators when
Pseudo mode is applied (see section 3.5).
The introduction of the pseudo mode requires a new approach towards response evaluation
for each mode. The classic approach is as follows:
i i a dir
dir
i i
T S k x

M K ) (
max ,
( 3.1.1 )
where K, M stiffness and mass matrices, mass participation factor, Sa spectra
acceleration, T period, i- mode number, k - coefficient of spectrum scaling, dir index of
input seismic motion direction (dir = X,Y,Z), x displacement vector for maximum reaction of
i-mode.
Now, the following is applied (see appendix B):
i
i
i a dir
dir
i i
T S k x
~
1
) (
2 max ,
, ( 3.1.2 )
where
i
~
denotes a basis vector (it is not necessary for
i
~
to be an exact approximation of
i
- exact eigenvector of 0
2

i i i

M K ),
i
-approximation of exact eigenvalue
i
.
It is possible to show that (3.1.1) gives exactly the same solution
max , i
x
as (3.1.2), if
i i

~
(
i
=
i
). However, (3.1.2) is applicable not only for direct solvers, but also for
iterative solver, since it does not require the resolution procedure corresponding to the
structures.com
306
stiffness matrix K. This way is faster than (3.1.1) and allows one to control results more safely
(sum of forces sum of reactions).
The below-presented formula is obtained from (3.1.1)
i
i
i a dir
dir
i i i a dir
dir
i i
T S k T S k x
2
1
max ,
1
) ( ) ( M K ( 3.1.3 )
The modal response vector
max , i
x
describes the extreme response of the correspondence

SDOF oscillator. The next step should be dedicated to defining the final response of the
MDOF structure by means of statistic averaging between modes and between seismic input
directions.
Robot (early versions) allows one to assign several statistically independent seismic input
directions with their own scale multipliers in one load case. The statistical averaging between
directions is produced by means of sum of absolute values and square root of sum of
squares combinations within each mode. The corresponding options are defined in the Job
Preferences.
The sum of absolute values option gives the following:
i
i
i a
dir
dir
dir
i i
T S k x
2
max ,
1
) (
( 3.1.4 )
The square root of sum of squares option makes averaging of the
max , i
x
of corresponding
seismic input motion directions like
Z i Y i X i i
x x x SRSS x
, , , max ,
, ,

.
i
i
i a
dir
dir
dir
i i
T S k x
2
2
max ,
1
) (
( 3.1.5 )
It is possible to show that each component of
max , i
x
is the SRSS combination of

correspondence components of
X i
x
,
,
Y i
x
,
,
Z i
x
,
,
where i =1,2,,N indicates the number of mode or pseudo mode.
The SRSS or CQC combination between modes (or pseudo modes) is applied to obtain the
final response of the considered MDOF structure after the averaged modal response vectors
max , i
x
have been obtained, i =1,2,,N.

The modal averaged response vectors
max , i
x
, i =1,2,,N are the same for sum of absolute

values and square root of sum of squares options if the single seismic input direction has
been defined for the current load case (For example, Kx=Kz=0, Ky=1).
Robot (version 12.2 and later) saves the above-mentioned procedure of averaging modal
responses between seismic input directions, however, it allows one to carry out the best
approach. It is recommended to define a single seismic input direction for each load case,
and then to apply either the SRSS combination between directions (it corresponds to the
American Regulatory Guides) or the so-called Newmark combinations (corresponding to the
French seismic code PS-92 and the Eurocode-8).
Let us illustrate new capabilities of the following typical example.
In this case (single seismic input motion for each load case), the typical values for scale
multipliers will equal
Kx=1; Ky=Kz=0 for dir = X (load case S_X)
Kx=0; Ky=1; Kz=0 for dir = Y (load case S_Y)
Kx=Ky=0; Kz=0.7 for dir = Z (load case S_Z; the vertical motion intensity is assumed to
equal 2/3 of the horizontal motion intensity)
Three load cases are defined for each statistically independent seismic input motion. The
modal response for each mode will be the same as (3.1.2) (i = 1,2,,N; dir = X,Y,Z ).
Then, it is necessary to define the averaging factor over all modes due to each seismic input
direction:
dir
i dir
R SRSS R or
dir
i dir
R CQC R
structures.com
307
where
dir
i
R - some factor (displacement, force, stress,) for the i-th mode due to seismic
input motion into direction dir which corresponds to the modal response
max , i
x
(obtained from
(3.1.2));
dir
R is the result of SRSS or CQC combination over all considered modes (pseudo modes).
Then, the averaging over all active seismic input directions according to the chosen option is
produced:
either SRSS combination:
2 2 2
z y x
R R R R
or Newmark combinations:
z y x
z y x
z y x
R R R R
R R R R
R R R R

The Spectral Analysis options enable definition of arbitrary spectrum of seismic input motion.
3.7. Seismic Analysis
The Response Spectra Method is applied for seismic and spectral analysis. The seismic
analysis is run on the basis of spectral analysis (see section 3.6), however, the spectra
accelerations
i a a
T S S are generated to correspond to a selected seismic code, instead
of being assigned by the user (as it is done for spectral analysis).
The UBC-97 seismic code is available in Robot (version 12.0 and the later ones). The
Response Spectrum Analysis is run in accordance with Sections 1631.5.1 1631.5.3 of the
1997 Uniform Building Code. It is possible to fulfill the requirements of Section 1631.5.4
(Elastic Response Parameters may be reduced ) by means of combination mechanisms
of Robot (see Help). The basic shear components Vx, Vy, Vz, overturning moment
components Mx, and My and torsion moment Mz (it is assumed that axis OZ is vertical) - all
are presented in the Reactions table in the line sum of forces, both for each modal
response and for SRSS and CQC combinations between modes. The following seismic codes
are available in the program:
UBC97
PS 69 R. 82
AFPS
PS 92
RPA 88
DM 16.1.96
EC 8
IBC 2000
P100 92
Turkish code
Chinese codes
Argentine code
Chilean code
EAK 2000.
structures.com
308
3.8. Selective Filters
This approach is based on the following observation: only the modes that have a relatively
significant mass participation factor make considerable contributions to the seismic response
of a structure. Therefore, it suffices to take only these modes into account. The remaining
modes (with small mass participation factors) can be ignored during seismic analysis. The
number of derived modes is usually considerably greater than the number resulting from
mass percentage assessment. Thus, the disk space and computation time may be saved, if
only the modes with significant mass participation factors are selected.
Two ways can be used: create a list of accepted modes for each seismic input direction
(seismic case) on the base of results obtained for the previously performed modal analysis
and assign a mass limit bound value for modal mass percentage (all modes with modal
mass percentage below that limit will be ignored). The first way is more efficient, although it
requires running modal analysis previously. The second way allows one to apply filters in the
same program run with spectral and seismic analysis, however, it usually occupies more disk
space and involves greater computational effort.
Let us consider another example. The results of modal analysis are presented below, in table
3.1, where the seismic cases are defined in the following manner: Dir_X (Kx=1; Ky=Kz=0),
Dir_Y (Kx=0; Ky=1; Kz=0) and Dir_Z (Kx=Ky=0; Kz=1)
Table 1
Mode number Mass particip.
UX (%)
Mass particip.
UY (%)
Mass particip.
UZ (%)
Period
1 0.05 12.01 0.004 0.803
2 67.43 0.06 0.005 0.705
3 0.002 0.08 0.07 0.686
4 0.001 0.008 0.009 0.650
5 25.4 0.07 2.06 0.590
6 0.09 68.5 5.05 0.540
7 0.08 10.3 0.06 0.490
8 0.07 0.06 0.56 0.460
9 0.05 0.07 30.56 0.420
10 0.08 0.06 0.25 0.380
11 0.06 0.01 26.7 0.270
Let us assume that we take all modes with mass participation factor greater than one percent
into account. The corresponding mass participation values are given in the table. Let us take
note that if seismic input directions are assigned as ( 1 0 0) for Seism_X case, the modes
with significant mass participation values for directions UY, UZ do not contribute to seismic
response at all (see section 3.6):
i
i
i a dir
dir
i i
T S k x
~
1
) (
2
max ,
,
where dir = X, Y, Z input seismic direction;
dir
i
x
max ,
-maximum response for i mode;

dir
i
-
mass participation factor;
i
T Sa - spectra acceleration;
i
~
- i-eigenvector or basis vector (in
case of the pseudo mode). The scalar multiplier on the right side of the above
formula
2
1
) (
i
i a dir
dir
i
T S k
defines the contribution of i-mode to the seismic response of dir

direction. In this case, where Ky = Kz = 0, considerable contributions will be made by modes
2 and 5. The remaining modes do not contribute to seismic response, due to zero Kdir
multiplier (dir = Y, Z) and to small mass participation values for dir=X direction. It is possible to
show - in the same way - that for case Dir_Y it suffices to take the modes 1, 6, 7 into
account, while for case Dir_Z the modes: 5, 6, 9, 11.
structures.com
309
Thus, by means of filters the program may take only the relevant modes into account - 2 for
Dir_X case, 3 for Dir_Y case and 4 for Dir_Z case - without a significant loss of mass
contributions. Let us take note that we would be forced to apply the 11 modes for each case if
we do not use the filters.
This approach enables one to reduce computation time for large-scale dynamic problems (as
well as disk space requirements and the amount of data to be post-processed) without
significant reduction of the result precision compared to the traditional method (when the
selective filters are not used).
For example, the large-scale problem PJG203 contains 34 266 equations (bandwidth equals
990 after optimization). The corresponding FE model is presented in appendix 3D see
Fig.A1. The 25 eigenpairs with the consistent mass matrix and 3 seismic cases were to be
calculated. The computation time still reaches approximately 50 hours on the Pentium PRO
(64 MB RAM, 200MHZ). The required disk space exceeded 1GB. Moreover, there arose a
problem with steel design module, caused by insufficient disk space. (To compute the SRSS
and CQC combinations, it was necessary to store the data of 25 modes multiplied by 3
seismic cases comprising a large number of degrees of freedom for all factors
displacements, internal forces, stresses). Application of selective filters allows the program to
solve this problem successfully.
3.9. Harmonic Analysis
The following definition of steady reaction of a structure to the action of a single harmonic
load is produced:
t F t F sin

where is the pulsation of the excitation load. The behavior of a structure is described as
F X

M K
2
,
where X
- amplitude value of the displacement vector.

3.10. Time-History Analysis
The modal decomposition (superposition) method is realized in Robot. It is based on the
representation of a structure movement as a superposition of the movement of uncoupled
modes. Therefore, the method requires the eigenvalues and eigenvectors to be determined.
The Lanczos method is recommended for this purpose. The method of modal decomposition
takes advantage of reduced uncoupled equations. It is an appropriate approach to analyze
the dynamic response of structures subjected to long-term action of dynamic loads (for
example, non-steady loading caused by working in-line equipment or seismic action).
Mathematical background and particularities of application are presented in [3,4,6].
The equation (without damping) may take the following form:

Ng
k
k k
t P X X
1
) (

K M (3.11.1)
where Ng - number of load groups, ) (t
k
- given time history for the k-th load group.

N
i
i i
t q t X
1
) ( ) (

(3.11.2)
where
i i
t q
), ( - correspondingly i-th normal co-ordinate and mode (eigenvector or Ritz

vector). Substitution (3.11.2) to (3.11.1) and addition of damping terms leads to following
uncoupled modal equations [3,4,6]

Ng
k
k i i i i i i i
t p q q q
1
,
2
) ( 2 , (3.11.3)
structures.com
310
where
i k k i
t P t p

), ( ) (
,
,
i
modal damping parameter (usually 2 . 0 05 . 0
i
; when
1
i
it indicates critical damping limit between oscillation motion and aperiodic motion),
i
- natural vibration frequency (pulsation), i=1,2,,N
Each of equations is solved numerically. Secondorder method with automatic selection of
integration step is applied. The resultant displacement vector for the defined time points
s
t t t t ,..., ,
2 1
is obtained by means of substitution of ) (
s i
t q in (3.11.2).
Modal decomposition method can be applied for analysis of seismic response. In such a case
equation of motion takes the following form
) (t I X X
g dir
M K M
(3.11.4)
and appropriate uncoupled modal equations
) ( 2
2
t q q q
g
dir
i i i i i i i
(3.11.5)
where
i dir
dir
i
I

, M - mass participation factor for i-th mode and seismic input direction
dir. Each mode must be normalized as follows: 1 ,
i i

M . Finally all results
(displacements, velocities, accelerations, internal forces, reactions, etc.) are stored only for
the defined time points
s
t t t t ,..., ,
2 1
. The high-performance post-processor allows one to
analyze time-history analysis result both in the diagram and table modes. The diagram mode
displays selected factors (displacement, acceleration, velocity, reactions, shear forces,
bending moments, etc.) for chosen DOF (degrees of freedom) and presents the deformed
shape of a structure in the selected the time point. The table mode allows one not only to see
the corresponding values, but also to search automatically the maximum and minimum values
among the response factors over all stored time points.
3.11. Modal Analysis Taking Static Forces into Account
The linear small vibrations with respect to the equilibrium static state induced by a given static
load is considered. The static forces are known to have influence on the natural vibration
frequencies. The usual Modal Analysis does not take such influence into consideration,
however, Modal Analysis taking static forces into account does.
Full non-linear equations describe the motion of the relatively static equilibrium state of a
system, induced by the given static loads.
b (t) x (t) x (t) x

L K M (3.12.1)
where M, K - mass and stiffness matrices, (t) x
L - non-linear operator, b (t), x

-
displacement vector and load vector. Linearization procedure consists of the following:
) ( x x (t) x
d st
t

(3.12.2)
where
st
x
is a part of the common solution which describes the static equilibrium state and
) ( x
d
t
is a vector of small dynamic displacements. Non-linear operator can be presented as

decomposition of Taylor('s) series
) ( x ) x ( x ) ( x x (t) x
d st st d st
t t
s

K L L L (3.12.3)
where
i
s
x
(...)
) x (
st
L
K

is a stress-stiffness matrix, which is a Jacobian and takes the
action of static forces into account. Thus, the following yields:
structures.com
311

b x x
b x x (t) x ) x ( (t) x (t) x
st st
st st d st d d

L K
L K K K M
s
(3.12.4)
The first expression is a result of linearization of appropriately small dynamic displacements
(note: ) ( x
st
t f
, so 0 x
st
) and the second one describes the non-linear static

equilibrium state. Therefore, small dynamic motion with respect to the static equilibrium state
is as follows:
0 (t) x ) x ( (t) x (t) x
d st d d

s
K K M (3.12.5)
Let us substitute
t i
d
e (t) x

. An eigenvalue problem originates from (3.12.5)

0 x
2

i i i st s

M K K (3.12.6)
where
i
- eigenvalue;
i
- eigenvector.
The computations are performed in two stages:
1. Linear (3.12.7) or non-linear (3.12.8) analysis of static stress-strain state induced by a
given static load
b x

st
K (3.12.7)
b x x
st st

L K , (3.11.8)
where
st
x
- unknown vector of static state, b
- vector of given static forces (static load

vector), K - stiffness matrix, b , x

st
L - non-linear operator. The static load vector b
may
be a result of the combination of several static loads. It should be noted here that linear
approach does not satisfy exactly the non-linear equilibrium equation (3.11.8). Thus,
vector
st
x
for the static equilibrium state is a result of an approximate solution and the
stress-stiffness matrix
st s
x
K contains an error. If the considered structure is

sufficiently stiff and non-linear effects appear poorly, such approximation seems to be
correct. Otherwise, it is necessary to solve the non-linear static problem (3.11.8) (that
technique is not covered by the manual). Obviously, the linear approach (3.2.17) is faster
than the non-linear one(3.11.8). In case of linear approach, it turns out that
G G K ) x ( x
st st s

, where G is a geometrical stiffness matrix.
2. Eigenvalue analysis (3.12.6)
The positive values of
i
( 0
i
) are known to represent to stable equilibrium states,
negative values ( 0
i
) unstable ones, whereas zero value ( 0
i
) corresponds to
lack of stability (buckling).
The loss of positive definiteness of matrix
st s
x
K K means that static load exceeds

its critical (buckling) value. A relevant message appears. The convergence will be lost
during the run of non-linear static problem (3.11.8). It is recommended to interrupt
computations because the following calculations is still senseless.
Only the non-linear approach is available for structures containing cable and tension-
compression elements.
structures.com
312
Example
Let us consider the example shown in the figure below.
Fig. 3.11.1
There is N - static load. The following expression describes behavior of such a system:
0
2
2
2
2
4
4
t
w
F
x
w
N
x
w
EI , ( 3.11.9 )
where w - bending displacement, - material density, F - cross-section area.
The solution will be searched as:

a
x
We t x w
t i

sin , ( 3.11.10 )
After substitution ( 3.11.10 ) to ( 3.11.9) the following is derived:
2
0
2
2
1
EI
a
n
N
, ( 3.11.11 )
where EI
a
n
N
cr
2

- buckling load,
0
- eigenvalue for N = 0 (result of usual Modal
Analysis). Finally,
cr
N
N
1
0
, ( 3.11.12 )
where - eigenvalue for the system subjected to action of a static load N. This result is
presented graphically in Fig. 3.11.2:
structures.com
313
0
Fig.3.11.2
The dependence , where is a load parameter, for a real structure is usually
more complex than presented by the expression ( 3.11.12 ) (see [ 1,22 ]).
Appendix 3A
Eigenproblem Solution Methods
The user should understand that at present the best universal method of solving an
eigenproblem does not exist.
0
2

i i i

M K , i=1,2,,n ( A1 )
where K is the stiffness matrix, M is the mass matrix,
i
is the eigenmode and

i
is the
pulsation. The term the most preferable means that for most of the problems such a method
will use up less resources (the computation time and HD storage) than any other one.
However, it does not exclude different situations in case of other tasks. Then, application of
other method is recommended. The present version of ROBOT covers several methods of
solving a generalized eigenproblem (A1). Each of them involves its own advantages and
disadvantages. Below, we present some recommendations to be considered while choosing
analysis method. We hope that, in the majority of cases, they will lead to the required results
in the best way.
The subspace iteration (SI) method is realized exactly as described in [4], therefore, the
description of this method is not included here.
Lanczos method
The Lanczos method [12,16,17] is a powerful robust approach used for solving large-scale
eigenvalue problems (A1). It is available when direct solvers (skyline or sparse) are selected.
This approach allows one to get the required first n eigenvalues and eigenmodes with any
desired precision. The greater number of the required eigenpairs is obtained, the more
significant advantages of Lanczos method become. However, the approach involves several
limitations, namely:
structures.com
314
1. The three-diagonal matrix T should not be decomposed. It means that it is impossible to
analyze a structure which consists of two or more unconnected substructures. In such a
case, either each substructure is considered separately, or another approach is
implemented (for example, block subspace iteration (BLSI) or the basis reduction
methods).
2. The mass matrix M should be taken as Lumped with Rotations or Consistent.
3. Zero density is not allowable.
The Lanczos Method uses reduction to the three-diagonal matrix T
j
T
j
MQ MK Q T
1
, ( A2 )
where
j j
q q q

,..., ,
2 1
Q - the rectangular matrix Neq x j, and Neq is the number of
equations, j - number of Lanczos steps,
j
q
- j-th Lanczos vector. The expression

j j j j j j j
q q q q

M K
1
1 1
( A3 )
generates the next Lanczos vector
1 j
q
, and defines the current line of T matrix

j j

4 3 3
3 2 2
2 1
T
Thus, the following reduced eigenproblem is obtained:
0
j
k
j
k
j
k
s s

T , k=1,2,,j ( A4 )

j
k
j
k
/ 1
2
, where
j
k
is the j-th approximation to
k
, k=1,2,,n, n is the required number
of eigenpairs. The algorithm will continue computations (to increase the j - number of Lanczos
steps), until the required accuracy is achieved for all required eigenvalues.
The selective orthogonalization procedure supports the required level of orthogonality
between Lanczos vectors
j
q
which ensures safety and numerical stability of the

computational process. We employ economic methods to provide selective orthogonalization
and to solve the reduced eigenvalue problem (A4) by double QR-iterations with shifts.
The source eigenvectors are determined by the following formula
j
k j
j
k
s
Q , k=1,2,,n ( A5 )
The details are presented in [12,16,17].
structures.com
315
Basis Reduction Method
The basis reduction method [4,5] is known as the Improved Rayleigh-Ritz method [4]. In [5]
such method is presented as a discrete variant of Bubnov-Galerkin method. It is available for
direct solvers (skyline and sparse). This algorithm allows one to get approximate values of the
first few eigenpairs. It is required to assign the master degree of freedom (MDOF) in order to
get the reduced system. Thus, a user is able to control the creation of a reduced model. It is a
powerful tool for users who have some experience in dynamic analysis of structures and deal
with the same type of structures whose behavior is known. This method allows one to exclude
undesirable degrees of freedom (DOF) from the reduced model and to reduce the initial
complex problem with a large number of DOF to a reduced form - with a considerably smaller
number of DOF. The experience of structure dynamic analysis shows what problems can be
encountered by the user when the automatically reduction methods (SI, BLSI and Lanczos
methods are taken into account) lead to a very complex computation process. For example,
the local vibration modes of single bars can lead to serious problems for these methods,
because it seeks automatically the eigenpairs without any selection. It should be noted that
for the majority of cases of real structures these local vibrations will be restricted by some
constraints that are not taken into account in FEM model, or their contribution will be
inessential to the overall system motion. Usually, the mass percentage is very small for such
local vibrations. The usage of exact methods in this case will lead to the above-mentioned
difficulties, however, the implementation of approximate basis reduction method can simplify
the computational process considerably.
This method has the following limitations.
1. The user has to assign the MDOF: the master nodes and master directions. It is assumed
that only displacements (not rotations) may be assigned as the master degrees of
freedom.
2. The algorithm is implemented for any type of mass matrix, however, Lumped without
rotations mass matrix type is most advantageous with respect to computation time.
3. The Sturm sequence check is not available. There is only one way to explore the check
convergence, namely, increase the number of MDOF (assign additional master nodes
and master directions), solve this problem once more and compare the eigenvalues.
Such a method transforms the source large-scale eigenvalue problem for FEM
0
~ ~
2

i i i

M K
(A6)
model (A1) into eigenvalue problem for reduced model
0
2

m f (A7)
where {f} - the influence matrix, {m} - the generalized mass matrix for a reduced model,
n
...
2
1
,
*
1
~
j
n
j
j
X
(A8)
Where n is a number of degrees of freedom of a reduced model. The basis for such
transformations is a static solution obtained for appropriate unit states: unit nodal forces are
applied consequently in each master node, in the selected master direction. A large-scale
static problem is solved for n right-hand sides:
,
*
i i
T X

K n i ,..., 2 , 1 (A9)
where
i
T
- load vector which corresponds to i - unit load. The user has to assign master
nodes and master directions. All demanded operations will be performed by the program.
The reduced eigenvalue problem is solved by Jacobi method, which leads to the approximate
frequencies
i
, and modes
~
, i=1,2,,n. The details of this approach are presented in [5].
structures.com
316
Block Subspace Iteration Method
The block subspace iteration method (BLSI) is developed to solve a generalized eigenvalue
problem (A1). It is available for direct solvers (skyline and sparse). It is a powerful robust
approach. Application of this method is strongly recommended when a large-scale problem
arises and it is necessary to obtain a large number of eigenpairs (more than 10). BLSI method
can be applied for analysis of separate structures. All types of mass matrices (Lumped
without rotations, Lumped with rotations and Consistent) during modal analysis are
available. The application area of this approach is limited by the modal mode. The seismic
mode and pseudo mode are still available if Lanczos method is chosen.
The Sturm sequence check is performed to detect the skipped eigenvalues. BLSI method
controls the continuity of converged eigenvalues. Discontinuity of converged eigenvalues
indicates presence of skipped eigenvalues. However, continuity of converged eigenvalues
does not provide a strict assurance that skipped eigenvalues are missing. Nevertheless,
experience of numerous computations indicates that for the most cases Sturm sequence
check does not detect skipped eigenvalues, when BLSI method ensures the continuity of
converged ones. Thus, a great advantage of such a method is the possibility of avoiding the
time-consuming Sturm check procedure, if a full warrant of missing skipped eigenvalues is
not needed. If discontinuity of converged eigenvalues is met, the following message appears
(see Fig A1).
The main idea of BLSI method [1-3] consists in simultaneous vector iterations in the subspace
of the fixed size. Each converged vector is removed from work subspace (block) and instead
of it a new start vector is added. The orthogonality of the converged vectors is ensured on
each iteration step.
Application of the shift acceleration procedure [1,4] is recommended during modal analysis
when the slow convergence occurs:
0
2

M K
, (
A10 )
where M K K
, - shift value. At beginning of the analysis = 0 is assumed. The

automatic update of shift value is made, if new converged eigenvalues do not appear through
the accepted number of control iteration steps. For example, let us accept the number of
control steps equal to 5. Then 5 converged eigenvalues appear after 4 iterations. The shift
value remains = 0. On the next iteration step 3 eigenvalues converge. The shift value
remains = 0. Then, throughout 5 iteration steps no eigenmodes converge. The algorithm
detects slow convergence again, adopts
2
8
, updates M K K
and factorizes
the updated shifted matrix
K . Then after 2 iteration steps converge 2 eigenmodes. The

shift value remains
2
8
. Then during following 5 iteration steps do not converge any
eigenvalues. Algorithm again detects a slow convergence and takes
2
10
, updates
M K K
and factorize updated shifted matrix
K . And so on.
structures.com
317
Fig. A1 Discontinuity of converged eigenvalues is detected while running BLSI method. Yes
computations will be continued until the next converged eigenvalue appears. Afterwards, the
next verification will be performed. No results will be saved and computations will proceed
to the next case. Cancel computations continue as long as the full continuity of converged
eigenvalues is ensured. Warning will be ignored.
These are not all the recommendations: apply the shift accelerations or do not do it. The user
should remember, that application of the appropriate shift is a most powerful tool of
convergence acceleration. Otherwise, each factorization of the updated
K matrix may be a
time-consuming procedure, especially for a large-scale problem. Thus, the final decision
about application of shift should be made on the basis of experience and intuition of the user.
The following example illustrates the benefit of shift application. The computation model is
shown in Fig.A2. There are 50 eigenmodes extracted by BLSI method. Solver skyline is
selected. Tolerance 1.0e-09 is accepted. It turns out that convergence beginning with 38
mode is still so slow that for 20 minutes of computations no results are obtained. Once
acceleration by shift (update of a shift was accepted over each of the 5 unconverted iteration
steps) has been activated, computation time still amounts to 50 seconds. Obviously, it is
possible to present numerous examples, when application of shift reduces the number of
iterations, however, it increases the computation time. We recommend activation of the shift
accelerations, when conventional approach (shifts are turned off) leads to a large number of
iterations at some stages of the BLSI run.
structures.com
318
Fig.A2 Spatial frame structure
Modified Lanczos Method
It is an adjustment of Lanczos method in pseudo mode to iterative solver. Usual Lanczos
method requires factorization of a stiffness matrix (see A3). When a large-scale problem is
met, factorization of the stiffness matrix is still very time-consuming. In case of large
problems (over 100 000 equations), usually, not only factorization of a stiffness matrix
requires enormous computational effort, but also solution of an equation set of appropriately
factorized matrix is expensive.
The modified Lanczos method is based on the iterative approach. It allows one to avoid
storage, assembling and factorization of a large-scale stiffness matrix. Evaluation of each
Lanczos vector requires approximately as much computational effort as solution of a static
problem with single right-hand side. As regards the pseudo mode, it reduces the required
number of Lanczos vectors compared to the modal mode which is applied when running the
modified Lanczos method.
Iterative solver AEBEIS (see [7,8]) is applied for generation of Lanczos vectors. It is
recommended to use ICCF (incomplete Cholesky factorization) technique for both multilevel
aggregation preconditioning [7,8,18-20] and usual (non-multilevel) one. It ensures fast
operations during evaluation of matrix-vector product and fast resolution of correspondence
preconditioning. It should be noted that tolerance adopted for iterative solver (Job preferences
| Structure Analysis | Parameters) determines the precision of evaluation of Lanczos vectors.
Usually it is sufficient to accept
1.0e-04. The greater Number of Modes will be taken into account, the closer low Ritz vectors
will be to the corresponding eigenmodes and the fuller sum of modal masses will be
achieved.
See Help (about parameters of iterative solver for static and dynamics) for more details.
structures.com
319
Preconditioned Conjugate Gradient Method (PCG)
PCG method [9-13] is recommended for definition of a small number of eigenmodes in the
modal mode, when iterative solver is applied. It may prove very useful for assigning a wind
load or for checking a few low modes, obtained by PCG_Ritz method. All types of
preconditioning (see Tools / Job Preferences / Iterative / Parameters), defined for static
analysis, are available. All types of mass matrix (consistent, lumped with rotations and
lumped without rotations) can be used.
Preconditioned conjugate gradient method is based on direct minimization of Rayleigh
quotient

k k
k k
k
x , x
x , x

M
K
(A11)
by means of the gradient approach where: k number of iteration,
k
- corresponding
approximation of an eigenvalue. The gradient approach searches such value of parameter
k
, which ensures the minimum value of

k
from (A11):
k k k k
p x x
1

(A12)
where
k
p
is a vector of conjugate direction. Search of the appropriate value of

k
[see 9-13]
leads to:

ad bc
mb nd
nc ma ad bc mb nd
n
m
d
c
b
a
k
k k
k k
k k
k k
k k
k k
2
4
x , x
x , x
p , p
p , x
p , p
p , x
2

K
M
M
M
K
K
Preconditioning B is applied to accelerate the convergence
1 1 1

k k k
z r Bz (A13)
Gradient direction is defined as
) x , x /( ) x x ( 2 g
1 k 1 k 1 k 1 k 1 1 k

M K M
k
(A14)
New conjugate direction is defined as
k 1 k 1 k
p z p

k

(A15)
where

k
1 k 1
g ,
g ,

k
k
k
z
z

.
structures.com
320
Iterations are performed until
tol
k k k

1 k 1 1 k 1 k 1 1 k 1 1 k
x / x x x / r

M K M M , (A16)
where tol is a desired tolerance. Usually tol = 1.0e-02 ensures very good precision for
engineering purposes. It should be remembered, that the convergence ratio (A16) is
computed in very strong norm (see the part covering precision of computations). The
tolerance mentioned above provides precision of eigenvalues not worse than 1.0e-04.
When a first eigenpair is converged, it is stored as a final result, and iterations begin to
compute the next one. The orthogonalization procedure of the previously defined
eigenvectors on each iteration step is employed to avoid doubling of eigenpairs. Such
process is applied until all desired eigenpairs are obtained.
The most efficient way of convergence acceleration for PCG method is implementation of
good preconditioning. All types of preconditioning presented for iterative solver are available
for PCG method. It is strictly recommended to apply the multilevel preconditioning [18-20] or
non-multilevel preconditioning with ICCF [9-12] smoothing from AEBEIS solver [7,8]. For
more details see Help.
Ritz-gradient Method (PCG_Ritz)
PCG_Ritz method [8] is a fast method of definition of a set of Ritz vectors in pseudo mode
when iterative solver is selected. Such an approach may be very fruitful for seismic and
spectral analysis of a medium-size structures including (10 000 60 000) equations.
It is based on generation of the orthogonal system of basis vectors. The gradient approach
with multilevel aggregation preconditioning on the basis of element-by-element technique is
applied to minimize the Rayleigh quotient for each step of basis vector preparation. It ensures
evolution of the consequence basis vector toward the lowest eigenmode without aggregation
and decomposition of a large-scale stiffness matrix. Such method is often more effective for
dynamic response analysis compared to classic modal superposition method, especially for
seismic response analysis. The proposed method allows one to apply arbitrary types of finite
elements due to aggregation approach and ensures fast solution and inexpensive
requirement concerning disk storage, caused by using EBE technique. This method is
particularly effective when the consistent mass matrix is used.
The given eigenvalue problem is as follows:
0 M K (A17)
where K, M are the stiffness and mass matrices respectively, is the eigevector and is
the eigenvalue. The procedure of evolution of the basis vectors set
n
x x x ,..., ,
1 0
toward the
lowest eigenmode will be described. The preconditioned gradient approach is applied to
minimize the Rayleigh quotient

k k
k k
k
x Mx
x Kx
,
,
(A18)
where n k , 0 , k is the evolution step number; 1 n is the number of basis vectors, which
define the size of the subspace
n
span x x x ,..., ,
1 0
; N n 1 , where N is the number
of degrees of freedom for the considered problem (A17). It very often appears that the
considered eigenvalue problem is ill-conditioned. In such a case the evolution of the
consequence basis vector
k
x toward the lowest eigenmode will be very slow. The
preconditioning operator B is applied to improve such a situation. The expression
k k k
z r Bz denotes the resolution of a given equation set of correspondence vector
k
z ,
structures.com
321
where B is a preconditioning operator and
k k k k
Mx Kx r is a corresponding residual
vector.
The basis vectors satisfy the following conditions of orthogonality:
n p k
p k
p k
p k
,..., 2 , 1 , 0 , ,
, 0
, 1
,
x Mx (A19)
The source large-scale eigenproblem (A17) is reduced to the subspace eigenproblem
0
,
2
,
q m q k
j i j i

(A20)
The matrices of subspace projection are defined as
j i j i
k x Kx ,
,
and
U x Mx
j i j i
m ,
,
, where U is a unit matrix.
The Ritz vectors
1 2 1
,..., ,
n
v v v for the derived basis vectors
n
x x x ,..., ,
1 0
and the
corresponding approximations of frequencies
1 2 1
,..., ,
n
are utilized for superposition of
the structural dynamic response.
The procedure of the evolution of the basis vectors n k
k
,..., 1 , 0 , x toward the lowest
eigenmode is very close to the corresponding step of preconditioned gradient iteration
method for eigenproblem solution. It is a well-known fact that convergence of the
preconditioned iteration methods depends considerably on the properties of the
preconditioned operator B. This operator should be positively definite; it allows inexpensive
solution
1 1
k k
r Bz and satisfies the condition number 1 ) (
1
K B C
in the best possible

manner.
The last requirement in case of Ritz-gradient method ensures good approximation of the low
part of eigenmodes.
Such a method is available only for the multilevel iterative approach, which ensures a good
quality of preconditioning. Both EBE (element-by-element) preconditioning technique and
ICCF technique one are used. The quality of Ritz vectors generated in such way depends
considerably on the properties of the preconditioning operator B (see A13 and [8]). Since the
coarse level model approximates well the low vibration modes, the Ritz vectors on the fine
level are a good approximation of the corresponding eigenvectors (see [8]). Thus, the quality
of results, obtained using such a method, depends considerably on the ability of the coarse
level model to maintain similarity to the given FEM model (so-called fine level). Usually a
single aggregation level ensures good approximation. When the number of aggregation levels
is larger than one, quality of results is not guaranteed. It is a main limitation of this method
application for analysis of a large-scale problem, when the number of equations exceeds
~60000.
If preconditioning matrix B = K (the coarse level is identical to the fine level), the proposed
Ritz-gradient method passes exactly to Lanczos method (see [8]). Mathematical background
is presented in [8].
Modal Analysis Precision of Computations
The generalized eigenvalue problem is defined as
0 M K , (A17)
where K, M are the stiffness and mass matrices respectively, , - eigenpairs (natural
vibration mode and eigenvalue). Two types of residual vectors are defined:

M K r (A18)

1
M K f
, (A19)
structures.com
322
where
, are really computed eigenpairs which contain some computational errors. The
first expression defines the residual vector in terms of forces and the second one in terms of
displacements.
Four different criteria are used to estimate the computational error of eigenvectors.
1.
/ M r . It is a very stiff criterion. Usually 01 . 0 means that first four digits in

the eigenvalue are defined exactly. It is applied only for PCG method, when iterative
solver is selected.
2. , r . It is a slightly softer criterion than the previous one. It is applied for modified
Lanczos method, when iterative solver is selected.
3. / f . It is a soft criterion because convergence of displacements in given
realization of FEM is usually faster than convergence of internal forces. It is used for
BLSI, SI and Lanczos method, when direct solvers (skyline or sparse) are selected.
4. tol
k
k k

1
, where
1
,
k k
- two successive eigenvalues on k, k-1 iteration steps
and tol denotes the tolerance for eigenvalues, adopted in the Modal Analysis Parameters
dialog box. It is used as intermediate criterion while the following methods are being run:
BLSI, SI and Lanczos method (skyline or sparse direct solvers) in the modal mode. Such
a criterion is not robust, however, it is very fast. The use of (4) allows one to reduce
considerably computation time for BLSI, SI, Lanczos methods, especially for large-scale
problems. When an eigenvalue analysis is completed, criterion (3) is applied as the final
verification of precision. The user should look at the Precision column in the table where
the value / f is presented. If insufficient precision is met for an eigenmode, it is
necessary to repeat the eigenvalue analysis with a higher tolerance for eigenvalues tol.
The following table summarizes the considerations mentioned above. The symbol N/A means
that the corresponding convergence check is not produced. Results of the final verification are
obtained only once and are presented in the Precision column of the table. The
convergence check during computations is performed several times.
Direct solvers Iterative solver
Type of
criterion
BLSI,
SI,
Lanczos
method
Basis
reduction
method
Modified
Lanczos
PCG_Ritz PCG
During
computations tol
k
k k

1
N/A N/A N/A

/ M r
Final
verification
/ f
N/A
, r
N/A

/ M r
It should be noted, that Lanczos method for seismic mode produces convergence check
/ f through each of the 20 Lanczos steps. Basis reduction method and PCG_Ritz
method are the Ritz methods. Since it is not the iterative approach, verification of precision is
not performed.
If precision of some modes after computation appears to be insufficient, it is necessary to do
the following:
structures.com
323
Direct solvers Iterative solver
BLSI,
SI,
Lanczos
method
modal mode
Basis
reduction
method
Modified Lanczos PCG_Ritz PCG
Decrease
tol from
Modal
Analysis
Parameters
dialog box
Increase the
number of
basis nodes
and basis
directions
Increase the number
of modes;
Decrease the tol in
Iterative Solver
Parameters dialog
box
Increase the number
of modes;
decrease the number
of aggregation levels;
increase the number
of inner iterations
Decrease tol in
Modal Analysis
Parameters dialog
box
Appendix 3B
The Pseudo Mode Approach
The source motion equations for seismic loads take the following form:
t I t P X X
dir

M M K (B1)
M K, - the stiffness and mass matrices;
dir
I
- direction unit vector; t - time history of the ground acceleration

The solution is searched as:

N
i
i i
q t t X
1
(B2)
where
i
q
are the basis vectors of the size Neq number of equations of the source FE
model.
These vectors should satisfy the following requirements:
kinematics and static boundary conditions;
linear independence
basis completeness.
It is possible to adopt either Lanczos vectors or any vectors obtained for unit concentrated
nodal forces (basis reduction method for direct solvers or Ritz-gradient PCG_Ritz method).
The subspace projection is described as:

N
dir
T
i
j i ij
j i ij
ij ij
q q q Q
t I Q q t P P
q q m
q q k
P m k

,..., ,
; } {
); , ( } {
); , ( } {
}, { } { } {
2 1

M
M
K
(B3)
Let us note that
is of the N size; the Q matrix is of the NeqxN size.

The subspace equations (B3) will be solved by means of decomposition of the eigenvectors
(on the subspace which is defined by
N
q q Q

,...,
1
).
structures.com
324
; 0 } { } {
, ) (
2
1

k ij k k ij
N
k
k k
v m v k
where v t

(B4)
Let us note that decomposition (4) is an exact expression, because the k subscript assumes
values from 1 to N over the entire size of subspace Q. The substitution (B4) to (B3) leads to
the uncoupled set of equations

1 } {
,..., 2 , 1
) ( ) (
2

k ij
T
k
dir
T T
k k k k
v m v
N k
t I Q v t t

M
(B5)
Let us consider

N
s
T
k s s k
T T
k
q v Q v
1
,
~

(B6)
Thus, (B5) can be presented as
t I t t
dir
T
k k k k

M
~
) ( ) (
2
(B7)
where k=1,2,,N
Let us apply the response spectra method to the uncoupled equations (B7)
) (
~ 1
2 max , k
dir
a dir
T
k
k
k
S I
M (B8)
where ) ( ,
max , k
dir
a k
S are the maximum response for k subspace mode and spectra
acceleration function, respectively. Let us substitute (B8) to (B4) and then to (B2):
) (
~ ~ 1
) (
~ ~ 1
2 2 max , k
dir
a k
dir
k
k
k
dir
a k dir
T
k
k
k
S S I X
M (B9)
It should be noted that for Neq N :
T
k
~

T
k
,
k

k
where
k
,
T
k
are the exact eigenpair of the source FEs eigenproblem

0
2

k k k
K

M (B10)
and
dir
T
k
dir
k
I
M
~ ~
dir
T
k
dir
k
I

M , where
dir
k
is the mass participation factor for k
eigenmode.
Conclusions
1. The proposed approach does not require the good approximations
T
k
~

T
k
,
k

k
. It is possible that
T
k
~
,
k
approximate the
T
k
,
k
with the arbitrary precision.
2. The proposed approach is not worse than the well-known modal decomposition
(superposition) method. Both of these methods constitute particular cases of the common
projection method and converge to the exact solution when N Neq. (This statement is not
valid exactly for response spectra method, because this method is of the statistical character and its
convergence to the exact solution takes place when N . It is obvious that for Neq N there will
arise a computational conflict due to the fact that the number of basis vectors cannot exceed Neq. Therefore, it
is possible to get different solutions for different basis vectors when N = Neq. It is a particularity of the response
spectra method. For methods other than the response spectra method, the full convergence takes place when
N= Neq. The convergence for the case of response spectra method should be considered in the statistical
structures.com
325
sense only, because this method constitutes a statistical approach aimed at getting an average solution once
the time history has been lost. Due to that the term exact solution should be applied with caution).
3. Which basis is better: Ritz vectors or eigenvectors? There should be such a basis which
will ensure better convergence to the exact solution (N = Neq) by means of the smallest
number of basis functions N. The question should be solved by examining practical
examples.
4. Usually, the first part of pairs
T
k
~
,
k
provide good approximations to the
correspondence eigenpairs
T
k
,
k
(It is possible to determine the precision of each pair
in the Precision column of the output listing). Only the last part of pairs
T
k
~
,
k
provide
bad approximations to the exact eigenpairs and may be considered to be pseudo modes
(from the viewpoint of the French seismic code PS-92).
5. The use of the entire subspace
N
q q Q

,...,
1
ensures fast increase of mass
percentage. In case of Modal and Seismic modes, only a part of this subspace is used.
Appendix 3C
The examples of application of the Seismic mode and Pseudo mode
1. It is obvious that there are many kinds of seismic and spectral problems where it is
difficult to get sufficient (70%- 90%) percentage of masses. It is possible to solve the so-
called good problems using well-known methods: the user assigns an arbitrary number
of N modes and computes the first sequential N modes by means of the Modal mode.
However, for hard problems this approach may turn out to be unreachable. Let us
consider, for example, the Coreal or Museum problems. They are FEM models which
have been prepared by French engineers. A mass percentage for different number of
defined eigenmodes is presented. The Modal mode (Lanczos method) is used.
The convergence history for problem Coreal
Number of Converged Modes Mass %
44 < 1%
62 12%
75 38%
89 60%
116 74%
154 77%
179 80%
The convergence history for problem Museum
Number of Converged
Modes
Mass %
41 20%
106 40%
119 42%
For example, 80% means that for two directions sums of masses are not smaller than the
given number. The Modal mode generates eigenmodes, while either the adopted mass
percentage is achieved or the assigned top limit number of modes is exhausted. The final
solution of the response spectra method is obtained in the form of statistical
superposition of eigenvectors.
2. In both problems under consideration, the number of degrees of freedom is smaller than
2000. Such problems are considered to be the small ones (with respect to the number of
degrees of freedom). In case of medium-sized and large-scale (tough) problems, it is
possible that both the Modal and the Seismic mode remain still inapplicable in practice,
due to the expansive character of the computation process. In such cases, application of
structures.com
326
the Pseudo mode is recommended. Below, the convergence history for Coreal and
Museum problems is presented.
The convergence history for problem Coreal
Number of Basis Vectors Mass %
10 58%
20 67%
40 70%
80 80%
The convergence history for problem Museum
Number of Basis Vectors Mass %
10 60%
20 66%
40 71%
NOTE: It is necessary to define approximately 130 150 Lanczos vectors to obtain
80 eigenvectors for the Modal mode. The computational effort in the Pseudo
mode is considerably smaller (approximately 3 5 times).
The convergence of results in the Pseudo mode is illustrated in the next problem. The
following table presents the resultant max/min values of CQC combination for the Modal
mode and Pseudo mode :
Mode UX (cm) UY (cm) UZ (cm) Mass percent.
Modal 5.52002e-00 5.88293e-00 5.83013e-00 81%
Pseudo, Nvect = 10 5.58710e-00 5.89055e-00 5.00224e-00 80%
Pseudo, Nvect = 20 5.52937e-00 5.88870e-00 6.08661e-00 91%
Nvect the number of basis vectors.
Appendix 3D
Examples Application of Modal Analysis Methods to Solution of Large-scale
Problems
Example 1 presents the comparison of computation time for BLSI and Lanczos methods.
Different numbers of eigenmodes are considered. Applied solver: sparse direct solver. Mass
matrix type: Lumped with rotations.
structures.com
327
Fig. D1. Model of a hotel. Number of nodes: 6 359; Number of elements: 7 264;
Number of equations: 37 806
Table D1. Result Comparison
BLSI Lanczos
f (Hz) Precision F (Hz) Precision
2,794e-001 1,082e-006 2,794e-001 5,680e-015
1,388e+000 1,389e-004 1,388e+000 2,199e-013
1,520e+000 7,847e-004 1,520e+000 8,193e-012
1,644e+000 2,469e-004 1,644e+000 4,497e-013
1,747e+000 2,691e-004 1,747e+000 5,455e-014
1,776e+000 3,092e-004 1,776e+000 9,127e-013
1,806e+000 3,153e-004 1,806e+000 6,621e-013
1,818e+000 6,383e-004 1,818e+000 3,656e-012
2,622e+000 1,565e-003 2,343e+000 2,047e-011
2,634e+000 1,383e-003 2,622e+000 1,223e-005
Since Sturm check is not performed, skipped eigenpair occurs between 8
th
and 9
th
modes,
when BLSI is applied.
Table D2. Duration of mode extraction (in seconds)
Method 10 modes 50 modes 100 modes
BLSI 735 6029 23572
Lanczos 1472 12637 25271
Factorization of matrix: 841 s.
structures.com
328
Advanced methods (BLSI, Lanczos) on the basis of powerful sparse direct solver allows one
to produce - in the short time - not only matrix factorization, but also to extract a large number
of eigenmodes. Computations were performed on P-350 (256 MB RAM) computer.
Example 2 illustrates application of different methods to the solution of PJG203 problem. The
model contains rigid links which leads to the use of a consistent mass matrix. 25 pseudo
modes were extracted.
Fig. D2. PJG203 problem. Number of nodes: 5 945; number of element: 11 471; number of
rigid links: 22; number of compatible nodes: 302; number of equations: 34 266
structures.com
329
Table D3. Computation time, disk storage, the first ten frequencies and corresponding
precision for different methods.
Method Time (s) HDD (MB) Frequencies (Hz) Precision
Skyline 61 633 597 1.175e+000
1.337e+000
1.454e+000
2.445e+000
2.445e+000
2.628e+000
2.829e+000
3.033e+000
3.209e+000
3.595e+000
8.043e-015
1.025e-013
1.031e-013
1.712e-006
5.566e-006
6.331e-008
3.538e-001
3.052e-005
9.086e-005
4.498e-003
Sparse 4 435 99 1.175e+000
1.337e+000
1.454e+000
2.445e+000
2.445e+000
2.628e+000
2.825e+000
3.033e+000
3.209e+000
3.595e+000
3.522e-012
2.689e-011
1.159e-010
1.735e-006
5.639e-006
6.419e-008
3.520e-001
3.034e-005
9.938e-005
4.386e-003
Modif. Lanczos 3 459 24 1.175e+000
1.337e+000
1.454e+000
2.445e+000
2.445e+000
2.628e+000
2.791e+000
3.033e+000
3.209e+000
3.595e+000
3.719e-004
3.891e-004
6.601e-004
1.454e-003
1.875e-003
2.946e-003
3.364e-003
3.923e-003
2.175e-002
1.580e-001
PCG_Ritz 1 521 24 1.266e+000
1.350e+000
1.467e+000
2.445e+000
2.446e+000
2.446e+000
2.805e+000
3.035e+000
3.381e+000
3.566e+000
N/A
PCG_Ritz is not
an iterative
method, it is a
Ritz method.
Therefore,
pseudo modes
are obtained
instead of the
exact
eigenmodes
1. The following parameters have been used for modified Lanczos method: iterative solver
AEBEIS (multilevel mode); ICCF preconditioning; 2 aggregation levels; 4 inner iterations,
tol = 1.0e-04 precision of generation of Lanczos vectors.
2. The following parameters have been adopted for PCG_Ritz method: iterative solver
AEBEIS multilevel; ICCF preconditioning; 1 aggregation level; 4 inner iterations
3. The converged frequencies are marked in yellow. If the precision (see Appendix 3A) of a
relative norm of a residual vector for a given mode is less than 5.0e-02, such a norm can
be considered as fully converged to the corresponding eigenvector. Therefore, the
frequencies marked in yellow are considered to be the exact values for a given discrete
model. Error estimation for PCG_Ritz method is presented in table D4. It can be
concluded, that results obtained by PCG_Ritz method provide sufficiently good
approximation for engineering purposes and such an approach can be used for fast
estimation of seismic behavior.
structures.com
330
4. The advanced methods allow one to reduce drastically the computation time and disk
storage requirements without any serious loss of result correctness.
Table D4. Error estimation for PCG_Ritz method
exact frequency PCG_Ritz results Error (%)
1.175e+000 1.266e+000 7.7
1.337e+000 1.350e+000 1.0
1.454e+000 1.467e+000 0.9
2.445e+000 2.445e+000 0.0
2.445e+000 2.446e+000 0.0
2.628e+000 2.446e+000 7.0
2.791e+000 2.805e+000 0.5
3.033e+000 3.035e+000 0.1
3.209e+000 3.381e+000 5.4
3.595e+000 3.566e+000 0.8
Computations were performed on PC-450 (128 MB RAM) computer.
Example 3. A thin square plate with a consistent mass matrix clamped along one edge is
considered. The 4-noded shell element and mesh 128x128 (number of equations Neq is
99072) are used. The 40 pseudo modes are extracted.
The computation time and disk space storage for Lanczos method (skyline solver and sparse
one), modified Lanczos method (iterative solver) and Ritz-gradient (PCG_Ritz) methods are
presented in table D5.
Table D5. Computation time and disk storage requirements for several methods.
Method Time (s) HDD (MB)
Lanczos, solver skyline 141 559 7 367
Lanczos, solver sparse 15 615 157
Modified Lanczos, iterative
solver
18 978 0 in core
PSG_Ritz 9 651 192
First ten frequencies, obtained by Lanczos method (for both skyline solver or sparse solver)
and modified Lanczos method (iterative solver, ICCF preconditioning, 3 aggregation levels, 8
inner iterations, tolerance 1.0e-03 for generation of Lanczos vectors were adopted) are
identical. As the precision of computations was very high, it was possible to consider such
values to be exact within the given discrete model. They are adopted as etalon values for
error estimation for frequencies obtained by PCG_Ritz method. The multilevel approach with
ICCF preconditioning (one aggregation level, 4 inner iterations) was selected when PCG_Ritz
method was being applied. The corresponding results are presented in table D6.
Table D6. Comparison of frequencies for PCG_Ritz and Lanczos methods
Frequencies by
Lanczos (Hz)
Precision Frequencies by
PCG_Ritz (Hz)
Error (%)
3.722e+000 5.918e-014 3.725e+000 0.08
9.112e+000 4.474e-014 9.115e+000 0.03
2.282e+001 2.424e-012 2.284e+001 0.09
2.915e+001 5.866e-013 2.915e+001 0.00
3.315e+001 1.795e-013 3.318e+001 0.09
5.801e+001 2.373e-011 5.803e+001 0.03
6.565e+001 3.028e-011 6.571e+001 0.09
6.873e+001 6.907e-014 6.875e+001 0.03
7.602e+001 1.549e-012 7.609e+001 0.09
9.949e+001 3.302e-013 9.953e+001 0.04
structures.com
331
Example 4. Large building is presented in Fig. D3
Fig. D3. Large building. Number of nodes: 26126, number of elements: 30272,
number of equations: 155920.
Linear static analysis (a single load case) and extraction of 10 eigenmodes are considered.
The tolerance adopted for iterative solver equals 1.0e-04. For skyline solver Lanczos method
is used. For sparse direct solver both Lanczos and BLSI methods are applied. Modified
Lanczos method is used for iterative solver (multilevel method with 3 aggregation levels, 4
inner iterations, preconditioning ICCF). Mass matrix type: lumped with rotations is adopted.
Table D7. Computation time and disk space for several methods
Method Disk space
(MB)
Linear static
(single rhs) (s)
Extraction of 10
eigenmodes (s)
Total
time (s)
Skyline 5 702 136 065 65 052 203 878
AEBEIS 52 2 442 25 793 28 355
Sparse
(Lanczos
method)
773 10 253 24 500 35 762
Sparse NS
(BLSI method)
773 10 253 11 534 22 604
structures.com
332
Conclusions
Advanced methods: BLSI, Lanczos on the base of sparse direct solver and high-performance
iterative solver AEBEIS with ICCF preconditioning are the powerful tools for solution of large-
scale linear static and eigenvalue problems. They reduce considerably the computation time
and disk storage requirements compared to conventional skyline solver. Ritz-gradient method
PCG_Ritz is a fast approach which allows one to estimate seismic behavior of the given
structure. When a single aggregation level is accepted, the corresponding results, obtained by
PCG_Ritz method, are close to the ones obtained by Lanczos or BLSI method.
References
1. E.L.Wilson. An eigensolution strategy for large systems. Computers&Structures, Vol.16,
No. 1-4, pp.259-265, 1983.
2. E.L. Wilson. A new method of dynamic analysis for linear and nonlinear systems. Finite
Elements in Analysis and Design, 1, 1985, 21-23, North-Holland.
3. E.L.Wilson, Three dimensional dynamic analysis of structures, Computers and Structures,
Inc., Berkeley, California, USA, 1996.
4. R.W.Clough, J.Penzien. Dynamics of Structures. McGraw-Hill Book Comp., 1975, 634 p.
5. Fialko S. Yu. Investigations of the Initial Imperfections Influence to Natural Vibrations of
Ribbed Conical Shells, Soviet Applied Mechanics, 1982, 18, N11, pp.118 - 122. (In
Russian)
6. Fialko S. Yu. Nonsteady vibrations of ribbed conical shells under the influence of local
loads, Soviet Applied Mechanics, 1987, v23, N6, p. 547-552.
7. Fialko S.Yu. High-performance aggregation element-by-element iterative solver for large-
scale complex shell structure problems. Archives of Civil Engineering, XLV, 2, 1999.
P.193-207.
8. Fialko S.Yu. High-performance aggregation element-by-element Ritz-gradient method for
structure dynamic response analysis. CAMES (Computer assisted mechanics
engineering sciences), IV, 2000
9. Gambolati G., Pini G., Sartoretto F., An improved iterative optimization technique for the
leftmost eigenpairs of large symmetric matrices, J. Comp. Phys., 74: 41 60, 1988.
10. Sartoretto F., Pini G., Gambolati G., Accelerated simultaneous iterations for large finite
element eigenproblems, J. Comp. Phys., 81: 53 69, 1989.
11. M. Papadrakakis. A partial preconditioned conjugate gradient method for large
eigenproblems, Comp. Meth. Appl. Mech. Eng., 62: 195 207, 1987.
12. M. Papadrakakis, Solving large scale problems in mechanics, John Wiley & Sons Ltd.,
1993.
13. S.Bitzarakis, M.Papadrakakis, A.Kotsopulos. Parallel solution techniques in
computational structural mechanics. Comp. Methods Appl. Mech. Engrg. 1997, 148,
p.75-104.
14. Hughes T.J.R., Ferencz M. Implicit solution of large-scale contact and impact problems
employing an EBE preconditioned iterative solver, IMPACT 87 Int. Conference on Effects
of Fast Transient Loading in the Context of Structural Mechanics, Lausanne, Switzerland,
August 26-27, 1987.
structures.com
333
15. Hughes T.J.R., R.M.Ferencz, and j.O.Hallquist. Large-scale vectorized implicit
calculations in solid mechanics on a CRAY X-MP/48 utilizing EBE preconditioned
conjugate gradients, Comput. Meths. Appl. Mech. Engrg., 61
16. Hughes Th. J. R. The Finite Element Method. Linear Static and Dynamic. FEM Analysis.
17. Parlett B.N., 1980. The Symmetric Eigenvalue Problem. Prentice-Hall, Inc., Englewood
Cliffs, N.J. 07632.
18. Bulgakov, V.E., Belyi, M.E., Mathisen, K.M. Multilevel aggregation method for solving
large-scale generalized eigenvalue problems in structural dynamics, Int. j. Numer.
Methods Eng., 40: 453 - 471, 1997.
19. Bulgakov, V.E. Iterative aggregation technique for large-scale finite element analysis of
mechanical systems, Comput. Struct, 52: N4, 829-840, 1994.
20. Bulgakov, V.E., G. Kuhn. High-performance multilevel iterative aggregation solver for
large finite-element structural analysis problems, Int. j. Numer. Methods Eng., 38: 3529-
3544, 1995.
21. Regulatory Guide
22. I.Ja.Amiro, V.A.Zarucky, V.N.Revutsky, Yu.V.Skosarenko, A.I.Telalov, S.Yu.Fialko.
Oscillations of Ribbed Shells of Revolution, Kiev, Naukova Dumka, 1988, 169 p. (In
Russian).
23. S.Yu.Filako. Natural Vibration Modes of Ribbed Conical Shells, Soviet Applied
Mechanics, 1984, v20, N11, p. 1033-1037. (In English)
structures.com
334
APPENDIX 4
Instability Types in Robot
Instability Types That May Occur During Structure Analysis
The structure analysis can issue the following instability messages:
type 1 - there is zero value element on the diagonal of stiffness matrix
type 2 - there is zero value element on the diagonal of inverted stiffness matrix
type 3 - some values of stiffness matrix element are very disproportional.
The first, second and sometimes the third type of the instability is usually caused by
mechanical instability of a structure, for example a part of the structure is a mechanism or the
structure support parameters (number, type and positions of supports) are not sufficient. The
third type can also occur due to the fact that there are big differences in section profiles of
some elements.
All types of instabilities can be met for skyline solution method. For frontal method only the
first and third type can occur. If the message is displayed for the frontal method it is
suggested to recalculate the example with the skyline method selected. In the case of skyline
solver the number of nodes and degrees of freedom where the instability has been found is
precisely indicated.
Iterative solvers do not report instability messages and the instability of the model may result
in slow convergence of analysis. In such a case thorough verification of the model prior to
calculations is advised.
The solution method can be changed in the Job Preferences dialog box (Structure Analysis
option).
structures.com
335
APPENDIX 5
Pushover Analysis
Non-linear pushover analysis
Introduction
The static nonlinear pushover analysis is a simplified approach that allows an engineer to
understand the performance of a building structure for various design earthquakes. The
following restrictions apply to the current version:
All non-linear properties, which define the eventual structural failure during design
earthquake, are concentrated in so-called nonlinear pushover hinges. Other nonlinear
effects (longitudinal forces, P-Delta effect, traction-compression bars, etc.) can to be
consider together with non-linear hinges, but its do not play a decisive role in the damage
behavior of building structures.
The non-linear hinges can be considered only for frame elements of structures and for
non-linear restrictions. The wall elements, modeled by shell finite elements (FE) and solid
ones, are not covered by the option.
Non-linear hinges are considered an independent non-linear links for each degree of
freedom in a given node. The interaction between different degrees of freedom is
disregarded. For example, bending of a bar in one plane does not depend on bending in
another plane and longitudinal forces. Such limitation will be removed in the next version.
Location of each non-linear hinge is defined by the user.
Pushover analysis consists of the following stages:
Define non-linear hinges in a computational FE model.
Assign non-linear properties to hinges (force-displacement or moment-rotation diagrams)
Run modal analysis to activate a mass matrix (1 mode is sufficient).
Define the lateral forces specimen. Note: Push lateral forces depend on the mass matrix
type
Assign a control node and direction and ultimate push displacement value (when
displacement in the control node and direction exceeds such a threshold, a non-linear
analysis is completed).
Assign parameters to non-linear analysis.
Run step-by-step non-linear analysis. An equilibrium state curve shear forces versus
controlling displacement V = V(D) is a result of a non-linear analysis. Shear forces are
defined as a sum of reactions for a given push direction caused by the corresponding
lateral force specimen.
structures.com
336
Converse the equilibrium states curve V = V(D) to ADRS format (acceleration
displacement response spectra) derive the capacity curve ) (
d
cap
a
cap
a
S S S , where
cap
a
S is a spectral acceleration and S
d
is a spectral displacement.
Perform smoothing of a capacity curve. The smoothed capacity curve is used for the
performance point analysis.
Consider the reduction of the vibration period due to hysteretic damping caused by strong
non-linear deformations of a structure. Conversion of the capacity curve to the axis
effective damping versus period ) (T
eff eff
.
Step-by-step search of the performance point as an intersection point between the
capacity curve ) (
d
cap
a
cap
a
S S S and the selected demand curve.
Lateral load definition
It is often the case for 3-D structures that the first vibration mode produces a local vibration of
a small part of a structure (local vibrations of single or several bars, local mode of a plate,
etc.). Such a vibration mode is usually not representative for the analysis of a seismic
response as it makes little contribution to the seismic motion (its modal mass percentage is
small). Therefore, following algorithm is applied to prepare a push mode a shape vector,
which predefines a displacement vector during the pushover analysis. In general, the future
consideration is based on [2] with partial adaptation to computer implementation.

dir dir dir
I
~ ~
M K , where K, M - stiffness and mass matrices respectively, I
dir
-
unitary direction vector (in case of which units are located on the positions of translation
displacements of direction dir = x y,z, whereas on all other positions zeros are located;
x, y, z - push directions);
dir
is taken as a push mode.
Normalize the push mode:
dir dir

~ 1
, where
dir dir

~
,
~
M
Mass participation factor:
dir dir dir
I , M
Search lateral forces as:
dir dir dir
F M , where
dir
is a scalar multiplier. Shear force

dir dir dir dir dir dir dir dir
I I F V , , M .
Therefore,
dir
dir
dir
V
1
and
dir dir
dir
dir
V F
M
1
. Assume V
dir
=1 as V
dir
plays a
role of a load parameter in the non-linear pushover analysis and only the spatial
specimen of lateral forces is of interest to us at the moment. Thus,
dir
dir
dir
F
M
1
. In
the following part the dir subscript will be disregarded since each push direction requires
a specific pushover analysis.
structures.com
337
Equilibrium states curve. Nonlinear algorithm.
Characteristics of non-linear hinges are usually complex (see [1,2]) and contain degradation
branches. It often causes the tooth-like shape of equilibrium states curves. To overcome such
difficulties the arc-length algorithm is applied. The pushover analysis dialog-box appears.
Fig. 10.5.1 Pushover analysis parameters: Node number, direction - number of nodes and
direction in which the controlling displacement is set; Maximum displacement the maximum
value of a controlling displacement to be pushed; Non-linearity parameters - see non-linear
analysis.
If non-linear analysis or P-delta check boxes are checked, the nonlinear properties of each
finite element are taken into account in addition to non-linear hinges. Otherwise, only non-
linear hinges and other specific non-linear elements (traction-compression, cable and non-
linear connections) define the non-linear properties of a structure.
Capacity curve. ADRS format
Thus, equilibrium states curve V = V(D), where V is a shear force and D is a controlling
displacement, is derived. The next step is the convergence of equilibrium states V = V(D) to
so-called acceleration-displacement response spectra format (ADRS):
) ( ) (
d
cap
a
cap
a
S S S D V V , where S
a
, S
d
- spectral acceleration and spectral
displacement. The function ) (
d
cap
a
cap
a
S S S is a capacity spectrum. The conversion
procedure consists of the following: for each point {D,V} V = V(D) the corresponding point
for capacity spectra ) ( ,
d
cap
a
cap
a a d
S S S S S is derived using the formula
g
V
g M
V
W
V
S
tot
a
2

(10.5.1)
where
tot
M
2
- mass percentage for the push mode,

dir dir
dir
tot tot
I I M M , M - total
mass of a structure; W - structure weight; g - ground acceleration.
structures.com
338
D
d
D
S
, (10.5.2)
where
D
- such component of the push mode vector , which corresponds to (is assigned
the same node and direction) the controlling displacement D.
In general, V = V(D) is usually a non-linear function. The ) (
d
cap
a
cap
a
S S S equation is also a
non-linear function (see Fig.10.5.2). Each point on such a curve is associated with period T.
Evolution of non-linear deformations leads to the change of a free vibration period. It is
obvious that T=const on ADRS diagram is a straight line which passes through the origin of
coordinates. Therefore, for all points of the linear part of a capacity spectrum period is the
same and is denoted by T
lin
. This value is based on the solution of the linear equation set
KX
dir
= F
dir
, where K is a stiffness matrix, describing the linear behavior of a structure, and F
dir
is a specimen of lateral forces. Let us denote: D
lin
- component of the solution vector X
dir
for
which the degree of freedom is identical as the one for the controlling displacement D; V
lin
-
sum of reactions (shear force) caused by the action of F
dir
. According to [2],
g S
S
T
a
d
2 (10.5.3)
After substitution with (10.5.1), (10.5.2) to (10.5.3) it yields:
D lin
lin
lin
V
D
T
2 (10.5.4)
structures.com
339
Fig. 10.5.2 Capacity spectrum in ADRS format
HYSTERETIC DAMPING. THE ) (T
eff eff
CURVE
The appearance of non-elastic deformations give rise to hysteretic damping. The
corresponding areas, shown in Fig.10.5.3, illustrate the energy dissipated per loop and the
maximum strain energy. If the vibration of a given system is associated with the vibration of a
system of single degrees of freedom with viscous damping, then the equivalent viscous
damping per full hysteresis loop is as follows:
0
0
4
1
S
D
E
E
, (10.5.5)
where: E
D
energy dissipated per loop; E
S0
- maximum strain energy. The energy dissipated
per loop E
D
and the maximum strain energy E
S0
can be defined based on the consideration of
Fig. 10.5.3:
A
d a D
ABK Area dS S E
0
) ( 4 ; ) (
0
OAB Area E
S
(10.5.6)
Effective damping is defined as:
05 . 0
0

eff
(10.5.7)
where: 0.05 is the viscous damping, - the factor is adopted from [2] according to Fig.8-15.
structures.com
340
Fig. 10.5.3 Area of a curvilinear figure, restricted by the capacity curve ) (
d
cap
a
cap
a
S S S and
the AKO lines, is a of the hysteresis parallelogram area and and illustrates the energy
dissipated by damping per loop. The area of the OAB triangle presents the maximum strain
energy. AK is parallel to the constant period line T
lin
= const.
Therefore, a capacity curve can be presented not only as ) (
d
cap
a
cap
a
S S S , but as
eff
=
eff
(T), since each point of ) (
d
cap
a
cap
a
S S S corresponds to the period T (see equation
10.5.3) and the effective damping
eff
(see 10.5.5-10.5.7).
The integral of (10.5.6) is evaluated numerically for each point ) (
d
cap
a
cap
a
S S S A .
Application of the arc-length algorithm allows one to derive very complex capacity curves with
tooth-like reversible branches, caused by degradation paths of non-linear hinge
characteristics (see Fig. 10.5.4). Therefore, the capacity curve ) (
d
cap
a
cap
a
S S S is subjected
to the smoothing procedure prior to numerical evaluation of (10.5.5-10.5.7). A smoothed
capacity spectrum curve is defined on a regular mesh; it is a single-valued function unlike the
initial capacity curve. Such property is very important for correct evaluation of (10.5.5-10.5.7).
Thus, for the following analysis only the smoothed capacity spectrum curve is used.
structures.com
341
Fig. 10.5.4 Capacity spectrum and smoothed capacity spectrum curves. Based on the real
example.
Demand curve. Selected demand curve. Performance point
When moving along the non-linear capacity curve ) (
d
cap
a
cap
a
S S S ( ] , 0 [
MAX
d d
S S , where
} sup{
d
MAX
d
S S is defined by the last point of an equilibrium states curve), period T and
effective damping
eff
(T) are changing. Therefore, each point on the capacity curve
) (
d
cap
a
cap
a
S S S (
eff
=
eff
(T)) defines the corresponding reduced (demand) response
spectra curve ) (
d
red
a
red
a
S S S . The
elastic
a A
red
a
S SR S in the constant acceleration range
of a spectrum and
elastic
a V
red
a
S SR S in the constant velocity range of a spectrum (see Fig.
8-14 from [2]). The
elastic
a
S is the acceleration spectra from the elastic response spectrum (5%
damped).
structures.com
342
Fig. 10.5.5 Evaluation of capacity, reduced and selective demand spectrum curves. The
performance point is an intersection of capacity and selective demand curves.
The point ) ( } , {
d
red
a
red
a d
red
a
S S S S S , defined by given S
d
from capacity spectra
) (
d
cap
a
cap
a
S S S , we call as a selected point. Therefore, the motion along the capacity curve
) (
d
cap
a
cap
a
S S S give rise to the collection of selected points } , {
d
red
a
S S for ] , 0 [
MAX
d d
S S ,
which are called the selected demand spectrum curve ) (
d
dem
a
dem
a
S S S . The intersection
between the capacity curve and selected demand one defines a performance point. It is
explained by the scheme presented in Fig. 10.5.5. Until the points } , {
cap
a d
S S belong to the
linear part of a capacity spectrum diagram, T = T
lin

eff
= 0.05 = 5%. Point 1 is projected to
elastic response spectra as 1. Thus, demand spectra for such points are a part of elastic
response spectra from zero to the limit of the linear behavior. For non-linear part of capacity
spectra it is necessary to define: } , { , , } , {
red
a d V A eff
cap
a d
S S SR SR T S S .
Points 2, 3 define the corresponding projections 2,3 on the appropriate reduced spectrum
curves. The collection of such points creates the selective demand spectra curve. The
intersection between the capacity and selective demand curves defines a performance point.
Pushover curve dialog box
Press Results / Advanced / Pushover curve to display the pushover curve dialog box. Choose
the Displacement - reaction sum option to show the shear force - controlling displacement
diagram V = V(D) (Fig. 10.5.6.A). Selection of the capacity spectrum option (Fig. 10.5.6.B)
results in computations of capacity spectrum curve ) (
d
cap
a
cap
a
S S S , smoothed capacity
spectrum curve,
eff
=
eff
(T) curve, selected demand spectrum curve ) (
d
dem
a
dem
a
S S S and
in search of a performance point. The Normal and Smoothed options allow one to display
capacity spectrum curve and smoothed capacity spectrum curve, respectively. If the Selected
demand spectrum option is switched on, it results in the appearance of a selected demand
structures.com
343
spectrum which is computed on the basis of the seismic coefficients C
a
, C
v
and smoothed
capacity spectrum curve.
A B C
Fig. 10.5.6 The Pushover curve dialog box
Structure damping parameters allows one to set the structural behavior type (see [2]) and
assign the kappa-factor (see 10.5.7) according to [2, Fig. 8-15] or to the user choice. If a user
wants to assign user own dependencies = (
0
), the Other option should be switched on
and the default values of ,
0
should be corrected. The meaning of Point1 and point2 is
presented in the figure below.
structures.com
344
Fig. 10.5.7 Interpretation of point1 and point 2
In the current version viscous damping is assumed as 5% (constant value).
Auxiliary grid parameters allow one to display the lines of a constant period (period values are
assigned in the appropriate edit boxes) and the reduced spectra curves for given effective
damping (effective damping values are assigned in the appropriate edit boxes). Such a curve-
linear grid simplifies orientation on the S
a
, S
d
plane.
The performance point coordinates are presented under a separator if a performance point is
found. Otherwise, zeros values appear. To display all the performance point coordinates the
right mouse button should be clicked and the Table columns option selected. Other useful
graphical operations are available when the right mouse button is pressed.
The Damping-effective period option allows one to display the
eff
=
eff
(T) curve
(Fig.10.5.6.C). Take note that all points from the linear part of a capacity spectrum curve are
mapped to a single point of
eff
=
eff
(T) with the coordinates: T = T
lin
,
eff
= 0.05.
References
[1] FEMA 273, 1997, NEHRP Guidelines for the Seismic Rehabilitation of Buildings,
Developed by the Building Seismic Safety Council for the Federal Emergency Management
Agency (Report No. FEMA 273), Washington, D.C.
[2] ATC-40, Seismic evaluation and retrofit of concrete building, 1996.
structures.com
345
Arc-length controlling method for non-linear analysis
When the top point of an equilibrium states curve is met, the force-controlling incremental
algorithm is faulty.
Fig. 10.5.8 Typical view of equilibrium states curve with top points
The force controlling approach can be applied when 0 <
1
, where
1
corresponds to the
top point 1. If >
1
the force controlling iterative process is still non-convergent.
The arc-length algorithm allows one to pass all branches of the equilibrium states without any
serious problem. The normal plane method [1,2] is applied. The non-linear algorithm with the
developed arc-length strategy is presented below.
Input parameters:
max
- maximum value of a load parameter; D
max
- maximum value of a
controlling displacement; NoSteps number of assumed increments; NoIter number of
equilibrium iterations; tol_F tolerance for the residual vector norm, tol_L tolerance for a
load parameter.
Start initialization
= 0
Loop over load increments: n = 0, 1,
n n
i
d d
d
R
0
1
0
0
0
~
0
0
structures.com
346
where: ) (
1 1
i
n ext
i
n i
d N F R - residual vector,
i
n 1
- current value of a load
parameter,
ext
F - external load, ) (
1
i
n
d N - vector of internal forces;
i
n 1
d - current
displacement vector.
Loop over equilibrium iterations: i = 0, 1, 2, < NoIter
ext L
i
n
iteration each on matrix gent update or i if
F d K
d K K

) (
) _ _ _ _ tan _ _ _ 0 (
1
i
i
i
i
n
i
n
i
n
ext
i
i
n ext
i
n i
i over loop break L tol and F tol e convergenc Check
i if
d R d K
F
R
d N F R
~ ~
_ _ _ _ _ _ _ : _
) (
) 0 (
1
1
1 1
1 1

Set
i

Update
i
i
n
i
n
1
1
1
1
,
d to the next iteration

i
i
n
i
n
L i i
i
n
i
n

1
1
1
1
1
1
~
d d d d
End loop over i
nt displaceme g controllin D
n over loop break D D or if
i
n
_ _
_ _ _ _ ) _ _ (
*
max
*
max
1
1

End loop over n
The arc-length strategy sets the increment of a load parameter on each iteration step. At
the start of the solution (n=0; i=0)
L
T
L
S NoSteps d d 1 _ ; /
0 max 0
is
adopted where S is an arc-length increment. At the start step of each iteration (i=0; n>0)
L
T
L
S d d 1 /
0
and when i>0, the normal plane method gives
0
0
~
1
~
L
T
L
L
T
i
i
d d
d d

. The Fig.10.5.9 illustrates the normal plane method with the matrix
update in each increment (identically as in the modified Newton-Raphson method).
structures.com
347
Fig.10.5.9 The normal plane method
Let us denote:
0
0
d
- tangent line vector;
i
i
n
d
- normal line vector. The

constraint for computation of the load increment
i
is defined according to the normal plane
method plane which is normal to the tangent line on zero iteration step. The condition of
orthogonality is as follows: 0 n

or 0
0 0

i i
d d where
L i i i
d d d
~
and
0
0 0 L
d d ;
0
L
d -
L
d for zero iteration (i=0).
Such a condition of orthogonality allows one to define i , when i= 1, 2,
It is possible to demonstrate that the determinant of |K| = 0 when a singular point (the limit
top point or the bifurcation point) is achieved. The following condition is fulfilled:
i i
R d K
~
. When a given point of a plate load parameter controlling displacement
belongs to the equilibrium state curve, 0
~

i
d K because the residual vector 0
i
R (the
equilibrium conditions are satisfied exactly). The last expression constitutes a homogeneous
linear equation set. Thus, if at some point |K| = 0, it means that except for the trivial solution
0
~

i
d there exists a nontrivial one. The determinant |K| changes the sign if a singular
point is passed. The developed algorithm controls changes of the |K| determinant sign. If the
singular point is passed an appropriate warning informs the user that the current equilibrium
state is unstable.
Parameters for the arc-length method can be set in the dialog box shown below.
structures.com
348
Fig.10.5.10 Parameters of the arc-length method
Where: load increment number - NoSteps; maximum iteration number for one increment -
NoIter; maximum load factor -
max
; node number, degree of freedom - assign a node number
and direction for a controlling displacement; maximum displacement for a selected degree of
freedom - D
max
; relative tolerance for residual forces tol_F; relative tolerance for
displacements tol_L.
The arc-length method is used in the non-linear pushover analysis and is strongly
recommended when an FE model includes non-linear connections. The example illustrates
the capabilities of the arc-length method that allow the user to obtain automatically such a
complex equilibrium state curve (Fig.10.5.13) caused by degradation branches of non-linear
hinge characteristics (Fig.10.5.12).
structures.com
349
Fig. 10.5.11 Example of a frame structure loaded by lateral seismic forces
Fig. 10.5.12 Bending moment-rotation characteristic of non-linear hinges.
structures.com
350
Fig. 10.5.13 Equilibrium states curve. Tooth-like paths are caused by degradation branches of
non-linear hinge characteristics.
References
1. E.Hinton, NAFEMS. Introduction to nonlinear finite element analysis, Glasgow, 1992
2. E.Ramm, Strategies for tracing non-linear responses near limit points. Non-linear finite
element analysis in structural mechanics, (Eds. W.Wunderlich, E.Stein and K.J.Bathe),
Springer-Verlag, New York, 1981
structures.com
351
APPENDIX 6
Generation of Surface Finite Element Mesh (Plates and Shells)
- Examples
Coons method
The method consists in creating Coons surfaces over a contour whose opposite sides are
divided into identical number of segments. The opposite contour sides are linked by straight
lines, so that the intersecting lines create finite elements. Coons method is used in Robot for
3D areas and for flat rectangular or triangular contours. In the case of panels with openings, it
is recommended to use Delaunay method.
Options controlling contour division Division 1 and Division2
The Mesh Generation field of the above-presented Meshing Options dialog box contains the
following parameters responsible for contour division:
Division 1 defines the number of segments along the second edge
Division 2 defines the number of segments along the first edge.
Numbering of contour edges is determined by the order in which they are drawn. The first
edge drawn receives number 1, and every next one receives nest numbers (the numbering
step: 1). E.g. if one introduces a rectangular contour as shown below and defines
Division1 = 8 and Division2 = 4, then, edge 2 will be divided into 8 segments, while edge 1
into 4 segments. This, in turn, will determine the division of the opposite edges edge 4 will
be divided into as many segments as edge 2 (i.e. 8 segments), while edge 3 into as many
segments as edge 1.
Contour and the mesh generated for the following parameters: Division1 = 8 and Division2 =
4
If the compatibility conditions indicate that the initial user-defined division is insufficient, the
program will automatically increase the number of segments along a given edge. E.g. let us
define the following parameters Division1 = 4 and Division2 = 4 for two neighboring panels
that share one edge. Then, let us increase the parameters for the right panel - to obtain
Division1 = 6 and Division2 = 6 and leave the left panel as it is. We see that the program
increased the division for the left panel to maintain compatibility along the shared edge.
structures.com
352
Parameters responsible for the type of generated mesh
Apart from setting the values of contour edge division, the program allows one to control the
type of generated mesh. The options used for selecting surface type are located in the Coons
Method Parameters field. While defining parameters in this field, one should remember to
select appropriate finite elements. Otherwise for instance, if one sets Panel division type:
Square (Rectangular contour) and Triangle in the Finite Element type field, one will obtain a
triangular mesh instead of the expected mesh consisting of quadrilateral elements.
Examples:
Parameters shared by all examples:
Available Meshing Methods
Coons: Often
Forcing Ratio: Recommended
Example 1
Mesh Generation field
Division1 = 4, Division2 = 5
Coons Method Parameters field
Panel Division Type: Triangle in Triangular Contour
Finite Elements field
Type: 3-node Triangles
structures.com
353
For the settings: Panel Division Type: Triangle in Triangular Contour and Triangles and
Squares in Triangular Contour, the program creates regular meshes (each side of a triangular
panel is divided into the same number of segments). Therefore, if one introduces different
values of Division1 and Division2 parameters, the program will use the greater one (in our
case, Division2 = 5).
NOTE: If one selects quadrilateral in Finite Elements the field for this type of mesh,
the forcing ratio must be smaller or equal to the forcing ratio for the mesh
type. Otherwise, the program will generate a mesh consisting of either
quadrilateral or triangular elements (i.e. such a mesh as if we chose Panel
Division Type: Triangles and Squares in Triangular Contour).
Example 2
Panel Division Type: Triangles and Squares in Triangular Contour
Forcing Ratio: Proposed
In this case the finite element forcing ratio (Proposed) is smaller than the mesh type forcing
ratio (Recommended). This is due to the fact that triangular and quadrilateral elements must
be used here. Thus, imposing one element type (triangular) would lead to generation of a
mesh consisting only of one type of elements.
Example 3
Panel Division Type: Triangles and Trapezoids in Triangular Contour
Type: 4-Node Quadrilaterals
structures.com
354
We can observe here a situation similar to the one presented in the previous example. If we
used triangular elements with the forcing ratio higher than the mesh type forcing ratio, we
would obtain a mesh consisting solely of triangles. As we want to have a mesh consisting of
mixed elements, we may select quadrilaterals as finite element type. In this case, setting the
forcing ratio higher than for the mesh type will not result in the generation of solely
quadrilateral elements. However, it may happen that the mesh will not be generated at all.
Therefore, it is better to set a lower forcing ratio.
It can be observed in the figure above that here, as opposed to the former mesh types
(Triangles, Triangles and Squares), one edge may be divided in a different manner. The
process of dividing edges is carried out in such a way that edge 2 is divided into Division1
segments; the mesh orientation is responsible for assigning the Division 2 to edge 3 or 1 (or
to both of them). The orientation is determined in the following manner: the vertex from which
the mesh radiates is located in the vertex of the triangle where the angle is the biggest.
This is best illustrated by the next Example 3 triangular panels with identical meshing
parameters, differing only in edge numbering.
Example 4
Panel Division Type: Triangles and Trapezoids in Triangular Contour
structures.com
355
Example 5
Panel Division Type: Squares in Rectangular Contour
Forcing Ratio: Any
Forcing Ratio: Any
structures.com
356
Here, relatively low forcing ratio values (Any) are sufficient, for the area is regular. As it was
mentioned before, one should pay attention to finite element type. If one selected triangular
elements with the lowest forcing ratio (None), then, even the value Forcing Ratio: Forced
would not guarantee for this mesh type the generation of quadrilaterals. The case of Panel
Division Type: Triangles in Rectangular Contour is analogous.
Example 6
Panel Division Type: Triangles in Rectangular Contour
Forcing Ratio: Any
Forcing Ratio: Any
Example 7
Type: 4-Nodes Quadrilaterals
Forcing Ratio: Any
structures.com
357
In the case of plates with one edge defined as an object with imposed division (e.g. an arc),
one should remember that the program will not generate a smaller division than that
determined in the object definition. For instance, the figure above presents a contour with one
side defined as an arc with the division equal to 5 (Arc Parameters set Angle: 5 in the
Discretization field). Although Division1 = 2 in the meshing options, the program generates 5
segments. Only when the division number exceeds the number of divisions defined for the
object, the program generates a modification of edge 2 and, consequently, edge 4. This is
due to the fact that the arc consists of a given number of nodes linked by segments, and the
meshing algorithm adjusts the division to the number of nodes during the finite element
generation. Thus, creation of a division smaller than the one resulting from the arc definition
would involve deleting the existing nodes. This, however, is a forbidden operation.
Delaunay triangulation method and Kang method
Delaunay method
Triangulation method consists in dividing a (2D) surface into a mesh of triangular elements.
The Delaunay method copes very well with openings in contours, yet one has to define them
as contour edges first. Only one parameter - Division1 is required for mesh generation when
this method is used. The contour division is carried out in the following manner: the program
creates a square whose perimeter is identical as the perimeter of the area under
consideration. Then, each square side is divided into Division1 segments, i.e. the base length
is determined to assure even distribution of the edges of the contour in question. The figure
below illustrates well the described situation. For both panels the parameter Division1 = 4. By
dividing the square with perimeter identical with the perimeter of the rectangular plate, the
program determines the segment length (2.5). Thus, the rectangle in question is divided into
16 identical segments (16*2.5 = 40 = plate perimeter).
structures.com
358
In the Delaunay method, one can control the type of generated mesh by setting the options
available in the Finite Elements field. One may select here the type of finite element and
define the coefficient controlling conversion of triangular elements into quadrilateral ones
(Note: The Forcing Ratio option in the Finite elements field is not recognized for Delaunay
method). The conversion coefficient is rather important, as one obtains a smaller amount of
elements due to conversion into quadrilaterals. Additionally, the results obtained by means of
quadrilateral elements are usually more precise. The coefficient is a weighted quality,
assuming values from the following range (1, +1):
-1 means that the program will modify only the triangles that add up to shapes resembling
squares
+1 means that the program will create square elements wherever it is possible (NOTE: it
may lead to the generation of improperly conditioned elements)
NOTE: In order to be able to convert triangular elements, one has to select
quadrilateral elements in the Finite Elements field.
Examples:
Parameters shared by all examples:
Available Meshing Methods
Delaunay: Often
Example 8
Division1 = 5
Coefficient: -1
structures.com
359
Example 9
Division1 = 5
Coefficient: +1
Examples 8 and 9 illustrate well the influence of the conversion coefficient. For the conversion
coefficient equal to 1, only triangles that did not border with plate edges will be converted
into squares, for they added up to squares. On the other hand, for the conversion coefficient
equal to +1, the program converted triangles wherever it was possible. It should be added,
that conversion is carried out after mesh generation, i.e. it belongs to post-processing.
If one wants to generate a mesh consisting only of triangles, it suffices to select triangular
elements in the Finite Elements field, which means that the conversion coefficient will not be
recognized.
structures.com
360
Example 10
Division1 = 5
The coefficient does not have any influence because triangular elements have been selected.
Delaunay + Kang
The Kang method consists in refining a generated mesh in the vicinity of emitters. Emitters
are special nodes defined in such places in a structure that require increased calculation
precision. There are two types of emitters available:
user-defined emitters the option allows the user to define his own emitters and it is
available after checking out the Emitters: User option in the Delaunay Method Parameters
field in the Meshing Options dialog box. Emitters are defined in the Analysis / Calculation
Model / Emitters dialog box:
structures.com
361
default emitters emitters are created automatically in the corners and beside openings;
the option is available after setting Emitters: Default.
Mesh refinement by means of Kang method consists in generating a wave propagated from
an emitter into the panel area. The following parameters are available:
H0 first wave length (directly by the emitter)
Hmax last wave length
Q coefficient determining the relation between the next and the former wave. It means
that the lengths of successive waves compose the following sequence H0, H0*Q, H0*Q
2
,
..., Hmax For obvious reasons, Q coefficient must meet the following condition Q > 1.
structures.com
362
Example 11
Division1 = 5
Coefficient: +1
Delaunay Method Parameters
Emitters: Default
Delaunay + Kang: H0 = 0.5, Hmax = 3, Q = 1.35
As the default emitter is selected, the program generates waves in plate corners. Outside the
area of refined mesh, the Delaunay method is used. Additionally, the conversion coefficient
has been defined to assume value +1, which assures maximum conversion of triangular
elements into quadrilateral ones. Setting the following parameters H0=0.5, Hmax=3, Q=1.35
resulted in the generation of six waves with the following wavelengths: 0.5, 0.68, 0.91, 1.23,
1.66, 2.24, 3.03, which is best presented in the close-up of the left bottom corner.
Parameters: H0=0.5, Hmax=3, Q=1.35. Wavelengths: 0.5, 0.68, 0.91, 1.23, 1.66, 2.24, 3.03
structures.com
363
In order to generate a mesh with user-defined emitter, one should remember to switch on the
Emitters: User option in the Delaunay Method Parameters field of the Meshing Options dialog
box. In order to define a user-defined emitter, one should select the Analysis / Calculation
Model / Emitters command from the main menu. Emitters are defined by indicating a
node/point within a panel and introducing the first wavelength H0. The remaining parameters,
i.e. Q and Hmax are defined in the Meshing Options dialog box. The example presented
below contains an emitter defined by the user in the left bottom corner of the plate. The
default emitters are switched off.
Example 12
Division1 = 5
Coefficient: -0.6
Emitters: Default
Delaunay + Kang: H0 = 0.2 (defined in the Emitters dialog box), Hmax = 1000, Q = 1.2
As default emitters are switched off, the H0 parameter from the Meshing Options dialog box
does not influence the user-defined emitter. Setting Hmax = 1000 means that the generated
Kang wave propagates towards the inside of the plate.
Example 13
Pole Available Meshing Methods
Coons: Often (checked out)
Delaunay: Often
Coons Method Parameters
Coefficient: -0.8
structures.com
364
Emitters: Default
Delaunay + Kang: H0 = 0.3, Hmax = 1000, Q = 1.2
The above-presented example shows the use of global settings for mesh generation: the
above dialog box is opened by selecting the Tools / Job Preferences / Meshing Options /
Modification command from the main menu. The above-presented settings were used to
generate meshes for two square plates, i.e. Coons mesh was generated on the left panel,
while Delaunay mesh was used for the right-hand one. The meshing options were set in such
a way as to avoid imposing only one of the available methods, i.e. the forcing in the Available
Meshing Methods field is set to Proposed. Therefore, the program automatically recognizes
regular areas (left panel) and uses Coons method there, while it uses Delaunay method in
every case where there appear irregularities (right-hand panel with an opening).
structures.com
365
If one does not want to use Coons method, it suffices to set Never in the combo box beside
Coons in the Available Meshing Methods field and leave all the remaining parameters as
before. The above settings assure that the program will generate meshes on both panels
according to Delaunay method.
Identical effect can be obtained by forcing the Delaunay method, i.e.:
Let us see how such a mesh is influenced by the coefficient defining conversion of triangular
into quadrilateral elements. Let us modify its value from -0.8 to 0.5, and leave the remaining
parameters (forcing of Delaunay method for both panels included) as before.
structures.com
366
Due to setting new coefficient value, the program generated regular mesh with well-
conditioned quadrilateral elements. In this case, it is optimal to set the coefficient value equal
to 0.5.
The use of mesh consolidation and refinement - examples
Field Available Meshing Methods
Delaunay: Often (checked out)
Division1 = 5
The above settings result in generating Delaunay mesh consisting solely of triangles.
Now, we perform mesh consolidation. To do so, one should select the entire panel and open
the Mesh Consolidation dialog box by means of the Analysis / Calculation Model / Mesh
Consolidation command from the main menu. Then, one should set the conversion coefficient
structures.com
367
to 0.4 and uncheck FE mesh freeze option, so that it would be possible to modify the
structure later in the design process. Once the selected parameters are confirmed, one
obtains the mesh shown below.
Let us refine the mesh elements located in the plate corner (see the figure above). To do so,
one should select the elements and open the dialog box for mesh refinement by means of the
Analysis / Calculation Model / Mesh Refinement command from the main menu. The numbers
of the indicated elements will appear in the Element list field. Then, one should select double
refinement, i.e. set Mesh refinement type: Double. As before, the option for FE mesh freeze
should be unchecked.
As the above-presented figure shows, the sides of selected elements has been divided into
two segments, due to which each quadrilateral has been replaced with four smaller ones.
Simultaneously, to maintain continuity of the degrees of freedom, the program divided the
elements neighboring with the selected ones.
Now, generate the calculation model to obtain a new mesh consisting solely of triangles.
Select the Analysis / Calculation Model / Generation command from the main menu. Select
the entire panel and perform conversion into quadrilaterals by means of the options available
in the Mesh Consolidation dialog box. This time, assign the coefficient value 0.8.
structures.com
368
Now, let us refine the obtained mesh without dividing the sides of finite elements. To do so,
one should select the entire panel, open the Mesh Refinement dialog box and set there Mesh
refinement type: Simple. Confirming the defined parameters results in generation of the
below-presented mesh.

Complete Manual Training 14

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Complete Manual Training 14

Încărcat de

Drepturi de autor:

Formate disponibile

ROBOT MILLENNIUM v 16.

5 - USERS MANUAL UPDATE, May 2003

- stress matrix (linearly dependent on the compression

X t for the defined time points t t t t

, i=1,2,,m into equations (2).

active is direction global the i when

Average angles (strain):

for planes XY and XZ, respectively.

States of shear and torsion

Algorithm on the element level

, are discussed in this section.

is i-th eigenvector; there is

describes the extreme response of the correspondence

is the SRSS combination of

have been obtained, i =1,2,,N.

, i =1,2,,N are the same for sum of absolute

-maximum response for i mode;

defines the contribution of i-mode to the seismic response of dir

- amplitude value of the displacement vector.

), ( - correspondingly i-th normal co-ordinate and mode (eigenvector or Ritz

L - non-linear operator, b (t), x

is a vector of small dynamic displacements. Non-linear operator can be presented as

) and the second one describes the non-linear static

. An eigenvalue problem originates from (3.12.5)

- unknown vector of static state, b

- vector of given static forces (static load

K contains an error. If the considered structure is

K K means that static load exceeds

is the eigenmode and

- j-th Lanczos vector. The expression

, and defines the current line of T matrix

which ensures safety and numerical stability of the

, - shift value. At beginning of the analysis = 0 is assumed. The

K . Then after 2 iteration steps converge 2 eigenmodes. The

and factorize updated shifted matrix

, which ensures the minimum value of

is a vector of conjugate direction. Search of the appropriate value of

in the best possible

/ M r . It is a very stiff criterion. Usually 01 . 0 means that first four digits in

- direction unit vector; t - time history of the ground acceleration

is of the N size; the Q matrix is of the NeqxN size.

are the exact eigenpair of the source FEs eigenproblem

- mass percentage for the push mode,

d to the next iteration

- tangent line vector;

- normal line vector. The

S-ar putea să vă placă și