International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014 Available online at http://www.ijsrpub.

com/ijsres ISSN: 2322-4983; 2014 IJSRPUB http://dx.doi.org/10.12983/ijsres-2014-p0014-0028

Full Length Research Paper Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems
Ibrahim Medany1, Mostafa Shazly2, Mahmoud G. El-Sherbiny1*
1

Mechanical Design and Production Engineering Depart., Cairo University, Giza, Egypt, 2 Mechanical Eng. Depart., The British University in Egypt, Al-Shorouk City, Egypt,
*Corresponding Author: Tel: 00201006044616 Fax: 0020237746016, mgsherbiny@yahoo.com Received 30 October 2013; Accepted 27 December 2013

Abstract. Environmental regulations regarding industrial pollution are getting tougher to keep high air quality. The design of proper system to meet specific requirements relays on selection of different filtration components. Components are available in a wide range of designs to meet various requirements of air filtration level, quantity, and characteristics of the contaminants to be removed. An expert system package for the selection of dust collectors components is developed. The developed package utilizes database as well as knowledge base to aid the selection process based on practice and past experience. Industrial case studies are demonstrated to test the developed package and compare its results against the current systems design. The 3 developed package in one case study estimated a minimum air flow rate of 60000 m /hr, which is 30% higher than the

existing value and a static pressure drop due to ducts of about 1900 Pa which is 30 % lower than the current value leading to improved air quality. In another case study the package results indicated the need to minimize the value of air flow to be 2300 m3/hr, which is 50% lower than the current value and increase the external static pressure to 1700 Pa which is 70% higher than the existing design value leading to better filtration efficiency.
Keywords: Air Pollution, Dust Collection, Expert Systems, Design

1. INTRODUCTION In every application, fine filtration is strongly demanded, and air quality standards are much higher than they were 50 years ago. The investment in gas filtration of all kinds is about one sixth (16%) of the total investment in filtration and its related separations, worldwide. Air filtration devices remove

contaminants from an air or gas stream and are available in a wide range of designs to meet various in air filtration requirements. Degree of removal, quantity and characteristics of the contaminant, conditions of the air or gas stream, fire safety, and explosion control are among the factors that affect the selection process.

Fig. 1: Elements of a Local Exhaust System

14

Medany et al. Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems

For particulate contaminants, air cleaning devices are divided into two basic groups: air filters and dust collectors (Rajhans et al., 2004; Sutherland, 2008). Air filters are designed to remove low dust concentrations of the magnitude found in atmospheric air. They are typically used in ventilation, airconditioning, and heating systems where dust concentrations seldom exceed 0.025 grains per cubic meters of air and are usually well below 0.0025 grains per cubic meter. Industrial contaminant concentrations will vary from less than 4 to 4000 grains per cubic meter of air or gas. Therefore, dust collectors are, and must be, capable of handling concentrations 100 to 20,000 times greater than those for which air filters are designed (Rajhans et al., 2004). In industrial ventilation, only removing the larger dust particles from the air stream may be necessary for cleanliness of the structure, protection of mechanical equipment, and employee health (ASHRAE, 2008). Particles are

classified in three general categories (coarse, fine and ultrafine) and are derived from dust, construction activities, printing, photocopying, manufacturing processes, smoking, combustion and some chemical reactions in which vapors condense to form particles. These can be categorized as dust, smoke, mist, fume and condensates (Heinsohn et al., 2003). The dust removal process can involve quite high solid concentrations or low concentrations that will affect the dust collector type. For example, at the highest concentrations, the usual first step is a cyclone, which removes suspended solids quite efficiently, and collects them into an easily recyclable state (Sutherland, 2008; Thomas et al., 1989). Generally, there are five types of dust collectors for particulate contaminants, namely, Gravity Settling Chamber, Dry Centrifugal Collectors, Wet Collectors, Electrostatic Precipitators, and Fabric Collectors (Rajhans et al., 2004; Sutherland, 2008).

Fig. 2: CAD Program Flow Chart

Local exhaust systems are employed to capture air contaminants near their point of generation or dispersion. The local exhaust hood is the point of entry into the exhaust system and is defined herein to include all suction openings regardless of their physical configuration. The primary function of the hood is to create an air flow field that will effectively capture the contaminant and transport it into the hood as shown in Fig 1. Hoods have wide range of physical configurations but can be grouped into two general

categories: Enclosing and Exterior (Rajhans et al., 2004; Mahidol University, 2006; Jornitz, 2006). To move air in ventilation or exhaust system, powered air moving device such as a fan or an ejector are used. Selection of an air moving devices can be a complex task and the designers are required to take advantage of all available information from engineering societies as well as from individual manufacturers (Rajhans et al., 2004, Michigan, 1972). The objective of the present work is to develop an expert system package which can be implemented in a

15

International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014

user friendly Computer Aided Design (CAD) program for the selection of dust collectors and their components. Dust collection systems are usually designed as a construction from individual components such as cyclones, filters, ducts, fans and other joining elements for the severe loads generated from industrial processes where the air or gas to be cleaned originates in local exhaust systems or process stack gas effluents.

Expert systems have been used extensively over wide range of applications ranging from medical and industrial to business applications. Expert systems have the advantages of their consistency, completeness, reproducibility, and efficiency. However, they cannot adapt to changing environments, unless their knowledge base is changed, and when any errors occur in the knowledge base, will lead to wrong decisions (Arlington, 1983; Giarratano and Riley, 2004; Eronen and Zitting, 2001; Gamal et al., 2010; Beanlands, 1994; Bohanec and Rajkovic, 1990).

The expert system's reasoner operates over well-defined domain rules. These rules can be thought of as IF-THEN statements. Once the IF part of the statement is satisfied the THEN part is computed which can introduce additional information (Stefik et al., 1982; Ratajkosk et al., 2001; Clancey, 1983, Rahim and Ahmad, 2006). It also manipulates the knowledge base which can recursively cause more rules to be applied causing what is called as forward chaining (Eronen and Zitting, 2001). The objective of the present work is to develop an expert-based-computer aided design and component selection for dust collection systems. The developed system will make use of the current design approaches in dust collection system and collected data from experts and past studies that will aid in the selection criteria and decision making.

Fig. 3: Main Screen

2. COMPUTER AIDED DESIGN OF DUST COLLECTION SYSTEM Visual Basic program has been developed to assist in the design of dust collection system. The program is subdivided into eight main parts as shown in Fig 2. The developed package allows the user either to build a new system from scratch, edit or adopt previously designed systems as shown in Fig 3. The data used in

both cases are kept in the expert system memory as such that the user has the ability to make decisions to achieve better design for the dust collection system. 2.1. Application and Dust Types The first step in the design process is to choose the application type and the information about the dust to be collected. The interface screen for this purpose

16

Medany et al. Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems

contains three sections: Application Type, Dust Type and Dispersion Type as shown in Figs 4 and 5. The package allows the user to select the dust and particle types. If explosive and sticky dusts are to be removed they must be considered in the design process. The user in this case has to check or uncheck the appropriate check boxes in the window. Finally, the package allows the user to select the Dispersion Type which includes released with practically no velocity into quiet air, active generation into zone of rapid air motion etc. In all cases the built-in expert system presents application examples and recommended capture velocity. The package and the built-in expert system guide the users throughout the selection process. For example, a warning message appears when a non-recommended duct velocity is used, as shown in Fig 6. Other examples include the frequency of the same application type previously used, concentration dust load, duct velocities (main and sub), as illustrated in Fig 7. 2.2. Dust Collector Selection Based on the previously selected application and dust type, the built-in knowledge base recommends the type of dust collector to be used as shown in Fig 8. The package guides the user to determine the frequently used systems through on Often and Seldom usage keys. 2.3. System Design and Local Hoods The interface of System Design and Local Hoods consists of No. of Extraction Points, System Hood and Duct Design System as shown in Fig 9. First, the user must feed in the number of extraction points of the sources of emission to be removed. Then click on Design System Button to draw the single line diagram of dust collection system including Local Hoods, Branch Ducts, Main Duct, Dust Collector, Fan, The Duct Connecting Dust Collector and Fan and Fan Outlet Duct Chimney.

These input data will be used in Hood calculations and Duct design. Hood calculations data include hood type, air flow, hood static pressure, and branch pressure loss and duct. Duct design system includes duct between First Hood and Dust Collector, duct between Dust Collector and Fan, and Fan outlet duct as shown in Fig 9. 2.4. Hood and Duct Design

The package allows the user to select the hood type according to source and direction of emission. This includes, for example, slot and flanged slot hoods, and other types of hoods as shown in Figs 10 and 11. First, the interface allows the user to select hoods successively and then specify the hood type as shown in Fig 10. The user then feeds in the parameters in two steps as shown in Fig 10. The first step is to specify input parameters for selected hood (hood width, hood length, face diameter, and taper angle of hood shape) based on the application and the geometry of the pickup point. The second step is to input data for pressure losses (straight duct length from hood to branch including flexible or rigid ducts, duct material, fitting losses coefficients as shown in Fig 12). The results of these steps are the Air Flow, Hood Static Pressure and Duct Diameter as illustrated in Fig 13. Fig 14 shows an example of the expert system; applications where a warning message appears if calculated slot velocity is greater than 10 m/s.
The process of hood design is followed by duct design. The input in this case includes ducting parameters between Hood 1 and Dust Collector, Duct between Fan and Dust collector and Duct of Fan out/stack/chimney. This also includes the duct length and fitting loss in the duct branch as shown in Figs 15-17. The results of duct design process are duct diameters; air flow in the branch and main ducts, and the pressure losses.

17

International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014

Fig. 4: Flow Chart of Dust and Application Type

Fig. 5: Dust and Application Type

18

Medany et al. Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems

Fig. 6: Warning Message for the Duct Velocity

Fig. 7: Dust and Application Type

Fig. 8: Dust Collector Parameters

19

International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014

Fig. 9: System Design and Local Hoods

Fig. 10: Hood Parameters

20

Medany et al. Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems

2.5. System Results and Report The design of the complete system is reported by the package in the form of design report that includes all the necessary information to install the dust collection system. This includes the External Static Pressure (Pa.) at the ultimate operating conditions and Total Air Flow Rate (Q, m3/hr) as shown in Fig 18. If the external static pressure exceeds 3000 Pa, the system

should be redesigned by increasing the number of the extraction points of the design and a two fans solution may be more appropriate to minimize the fan and dust collector size. The fan selection process requires the Total Static Pressure and Total Air Flow Rate. This process requires the knowledge of the pressure drop across the filters. This information varies from supplier to another and must be obtained for the filters to be installed in the designed system.

Fig. 11: Hood Parameters- Other Hood Types

3. RESULTS AND DISCUSSIONS The above developed package is used to re-design existing dust collection systems for Paper-Winder Machine and a Fertilizers Crusher Room. The purpose of this section is to demonstrate some examples of industrial applications where the developed package can show its effectiveness. The developed package can save time and enable decisions to be made based on past experiences and practices. 3.1. Paper-Winder Machine Case Study Significant amount of dust are produced during reeling, unwinding, and slitting of paper rolls. This

dust causes many problems to printing machines and can impair the printing quality. Moreover, it affects working environment, causes health problems, and increases fire risk. Working with coated or calendared paper rolls decreases the amount of dust but at higher cost. The winder area is usually cleaned with pressurized air causing clouds of fiber dust to rise in the air and therefore, employees and operators need respiratory masks in order to survive the work place environment.

21

International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014

Fig. 12: Hood Parameters-Fitting Coefficients (Elbow)

Fig. 13: Hood Parameters - Complete Calculations

22

Medany et al. Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems

Fig. 14: Slot Velocity Warning Message Table 1: Dust Collection System Report for Paper-Winder Machine
Application Name: Winder of Paper Machine Version 1 23/12/2011 Date: Coal, Mining and Power Plant Main Application: De-dusting, air cleaning Sub Application: Heavy: 12 and up Concentration Dust Load (gr/m3): Yes No Explosive Dust or any spark from application: Dust Sticky: Average Industrial Dust Dust Type: 20 20 Main Duct Velocity (m/s): Branch Duct Velocity (m/s): Paper Tissue Dust Particle Type: Released at high velocity into zone of very rapid air motion Dispersion Type: 2.5 Capture Velocity (m/s): Venturi Collector (Wet Collector) Dust Collector Type: 1900 External Static Pressure (Pa.): 60,000 Total Air Flow (m3/h): No. of Extraction Points: 15

The existing dust collection system uses Venture Wet Scrubber consisting of 15 extraction points with fan capacity 45000 m3/hr having external static pressure 2000 Pa. The existing dust collectors were expected to reduce emissions to air but they were not efficient enough. There were also problems with dust, particularly during the winding process. After winding, some of the dust was trapped between paper sheets and some accumulated on the machine frame and fell back onto the paper. The present package was used to re-design the current collection system. The summary for the input data and results are shown in Table 1, which contains the information about the process and the pickup points and preferred pickup velocities. The results showed that the minimum air

flow rate is 60000 m3/hr, which is 30% higher than the existing value. Also the static pressure drop due to ducts is 1900 Pa which is 30 % lower than the current value. These results indicate that the system needs higher capacity of air flow at almost the same external static pressure. These results are different from the existing system design because the computed capture and hood face velocities (based on the knowledge base built in the expert system) were higher than the current values of the existing system. Also the distance from hood face to the farthest source is reduced to achieve better capture of dust and higher efficiency. As a result of the modified design, the air quality around the machines and working area is expected to be better.

23

International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014

Fig. 15: Duct System Data between Hood 1 and Dust Collector Complete Calculations

Fig. 16: Duct System Data Duct between Fan and Dust Collector Complete Calculations

24

Medany et al. Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems

Fig. 17: Duct System Parameters Duct between Fan and Dust Collector Complete Calculations

Fig. 18: System Results

25

International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014

Table 2: Dust Collection System Report for Crusher Room

Application Name: Version 2 Crusher Room of Fertilizer 24/12/2011 Date: Chemicals Main Application: Crushing, grinding Sub Application: Moderate: 5 to 12 Concentration Dust Load (gr/m3): No No Explosive Dust or any spark from application: Dust Sticky: Average Industrial Dust Dust Type: 20 20 Main Duct Velocity (m/s): Branch Duct Velocity (m/s): Fertilizer Particle Type: Released at low velocity into moderately still air. Dispersion Type: 0.5 Capture Velocity (m/s): Intermittent-Shaker Collector (Fabric Collector) Dust Collector Type: 1700 External Static Pressure (Pa.): 2300 Total Air Flow (m3/h): 8 No. of Extraction Points:

3.2. Fertilizer Crusher Room Case Study An Agricultural Fertilizer Company has an Underground Crusher Room. The Crusher Room consists of a Crusher, Screw Conveyors, Bucket Elevator, Hooper and Vibrator Screen. Due to the crushing process, the room is filled with foggy air with high dust concentration; and as a result the dust leaves the room to other working areas of production. The existing installed dust collection system servicing the crusher room area consists of eight extraction points, Intermittent-Shaker Fabric Collector, a fan with a capacity of 4400 m3/hr and an external static pressure 1000 Pa. The dust collection system efficiency is 85 % approximately and was satisfactory for the client. The computer aided design program was used to check the design of this particular dust collection system. The input data and results are summarized in Table 2. The results indicated the need to minimize the value of air flow to be 2300 m3/hr, which is 50% lower than the current value. Also the external static pressure due to duct is 1700 Pa which is 70 % higher than the existing design value. These results are different from the existing system design because the capture velocity was lower than the existing system. The knowledge base built in the expert system attempted to minimize the collection of product and minimize the size of fan and motor thereby reducing the power consumption and running cost.

4 CONCLUSIONS The present work introduced a proceducer for designing Dust Collection System, using an expert system based package for the selection of dust collectors and their components. The developed package utilizes database as well as knowledge base to aid the selection process. Expert data form previous experiences and practices are introduced to help in the selection of the components of the Dust Collection system, leading to maximizing the efficency and minimizing the cost of maintenance. The developed expert based program can save time and enable decisions to be made based on past experiences and practices. The examples considered in this study showed that the proposed designs suggested by the package support by the expert system is different from the current ones. The causes of these differences between the existing designs and the new designs are illustrated in two main points: Air Flow quantity and Static Pressure value. The distance from hood face to farthest point of emission sources should be minimum (near or close) and the designer must consider other hood types parameters, where the required air flow of suction is calculated rather than by the using the traditional equation (Q=V(volume) x A(area)) of hood. The static pressure loss of exhaust hood, flexible ducts , duct surface finish must also be considered in the design.

26

Medany et al. Expert-Based-Computer Aided Design and Component Selection for Dust Collection Systems

REFERENCES Arlington H (1983). Field Performance Measurement of Fan Systems. Air Movement and Control Association Inc., AMCA Publication 203-90, ASHRAE (2008). Air Cleaners for Particulate Contaminants, HVAC Systems and Equipment. Beanlands GA (1994). The Application of Expert Systems to Environmental Impact Assessment. Bibliography, GEBEC Consultants, Halifax. Bohanec M, Rajkovic V (1990). Expert system for the Decision making. Sisemica, 1(1): 145-157. Clancey JW (1983). The Advantages of Abstract Control Knowledge in Expert System Design. AAAI-83 Proceedings, pp. 74-78. Eronen P, Zitting J (2001). An expert system for analyzing firewall rules. Proc. 6th Nordic Worksh. Secure IT Systems, Technical report IMM-TR-2001-14, Technical Univ. of Denmark, Nov 2001, pp. 100107. Gamal MM, Hasan B, Hegazy F (2010). A Security Analysis Framework Powered by an Expert System. Int. JComp Sc and Secur (IJCSS), 4(6). Giarratano J, Riley G (2004). Expert Systems Principles and Programming. Course Technology, Fourth Edition. Heinsohn RJ, Cimbala JM (2003). Indoor Air Quality Engineering: Environmental Health and Control of Indoor Pollutants (Drugs and the Pharmaceutical Sciences). CRC Press, 1st Edition. Jornitz MW (2006). Filter Construction and Design. Adv Biochem. Engine/Biotechnology, vol 98, pp105-123.

Mahidol University (2006). Polishing and Buffing. Industrial Hygiene and Safety. Michigan L (1972). Industrial Ventilation A Manual of Recommended Practice. American Conference ofGovernmental Industrial Hygienists-Committee on Industrial Ventilation, 12th Edition. Rahim AZ and Ahmad BA (2006). A Web Based Expert system for PreDiagnosing Gestational Diabetes. Bachelor Degree thesis, University Technology MARA, Apr., 2006. Rajhans G, Paulson KM, Cleary WM (2004). Industrial Ventilation A Manual of Recommended Practice, American Conference of Governmental Industrial Hygienists (ACGIH), 25th Edition. Ratajkosk M (2001). Artificial Intelligence and Dispute Resolution. Health and Safety Executive (HSE), Evaluation of an expert system for the interpretation of serial peak expiratory flow measurements in the diagnosis of occupational asthma in a field trial, Contract Research Report No. 450. Stefik M, Aikins J, Balzer R, Benoit J, Birnbaum L, Hayes F, Sacerdoti E (1982). The organization of expert systems: A prescriptive tutorial. Xerox Corporation Report, Palo Alto Research Center. Sutherland K (2008). Filters and Filtration Handbook. Butterworth-Heinemann, Fifth Edition. Thomas HK, David YH, William TF (1989). Dust Collector Recirculation for Industrial Processes.Principal Investigators, ASHRAE.

27

International Journal of Scientific Research in Environmental Sciences, 2(1), pp. 14-28, 2014

Ibrahim Medany is currently a Project Manager for HVAC and Filtration (Dust Collection) Projects at Hammam Industries & Co. Company. He received his B.Sc. in Mechanical Production and Printing Technology from Akhbar Al-Youm Academy for Engineering, Printing and Press Technology in Egypt in 2006 and his M.Sc. in Mechanical Design and Production Engineering from Faculty of Engineering Cairo University in 2012.

Dr. Mostafa Shazly is an Associate Professor at the Mechanical Engineering Department, Associate Director for the Centre of Advanced Materials, the British University in Egypt. He received his B.Sc. and M.Sc. in Mechanical Design and Production Engineering from Faculty of Engineering, Cairo University in 1996 and 1999, respectively. He received his Ph.D. in Mechanical Engineering in 2005 from Case School of Engineering, Case Western Reserve University, OH, USA.

Prof. Mahmoud G. El-Sherbiny received the B.Sc. degree in Mechanical Engineering with honors from Cairo University, Egypt, in 1968, the M.S. from Cairo University on a cooperation program with the University of Denmark at Copenhagen in 1972, and Ph.D. degree from the Department of Aeronautical and Mechanical Engineering at the University of Salford, U.K, in 1975. He joined the department of Mechanical Design and Production Engineering at Cairo University since 1968 and promoted to Professor of Machine Design and Engineering Tribology in 1986. He is the founder of the Egyptian Society of Tribology (EGTRIB) in 1987, Tribology and Spare Parts Center (TSPC) at Cairo University in 1988, The Scientific journal of the Egyptian Society of Tribology in 2002, and The Arab Federation for Metrology (AFM) in 2007. He was appointed as Vice Dean for Curriculum and Students affairs in 1994, Dean of Engineering at Cairo University 1995-2001, and President of the National Institute of Standards 2005-2006. He published 56 reviewed articles in world known journals and holds 11 registered patents.

28