LEVEL C

~~~~~~~~~~~ .

A THEORY FOR AERODYNAMI C FORCES AND MOMENTS

Georgia Institute of Technology

June, 1978
Approved for public release
distribution unli.ited

H
~

~~~~~

~~~~

___________________

_ _ _ _ _ _ _ _ _ _

~~~~~~~~~~

d~~~~~

____________________________________________________________

~~~~~~~~~~~~~~~~

~~~~~~~~~

;~ .-- -.-.--

Unclassified

.
nh.r .0
SECURITY CLASSIF .ATION OF TIllS PAGE (INisa 0.1
~~~~~~~~~~~~~~~~~

~~~~~~

1. REPORT NUMIER
_

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

READ ZHSTRUCTIONS
i,vwMrjriu~~rI A. ~I IVN ~~~~~~~~
BEFORE COMPLETIN G FORM
2. GOVT ACCUStOM NO 3.. RLCIPIENt S C A T A L O G NUMBER
Ilj ~~~~ 4 I

) (~

4. T IT LE (. d subtSlI
~

-~~
,

RYF OR ANROD

1RCES

S.

~ rr 7J

~~~~~rimJ
~chnical ~ eP)~~~~~~~
T W e~ ro

1MG ORG. REPORT MU

S. CONTRACT OR GRANT _NuMBER(S)

1. AUThOR (s)

OO~~4~ l5~ C~24

j
*0. PROGRAM ELEMENT. PROJECT . T ASK

9. PERFORMING ORGANIZATION NAME AND ADDRESS

Geor gia Institute of Technology /

NR061 226

Atlanta, Georgia 30332

II. CONTROLLING OFFICE NAME AND ADDRESS

Fluid Dynamics Program

Off ice of Naval Rese arch
Department of the Navy
Arlington , VA

22217

Jun

IS. MONITORING AGENCY NAME A ADORES$(It dIU.,.n V (,uu ConeeoUind OW c.)

5. TYPE OF REPORT & PE~~ OO COVERED

_
_

*2.

REPORT DATE

I,.

PIUM5

78

R OF PAGES
~~

48

*5. SECURITY CLASS. (of ff1. r porf)

Unclassified
IS.. DECLASSIFICATI ONFOOWNGRA OING

SCHEOUI
.E

N/A

______________________________________________

IS. DISTRIBUTION STATEMENT (oI thIs R.pc.I)

Approved for public release, distribution unlimited.

*7.

I
U
.
,
n
I
DISTRIBUT ION STATEMEN1 (of VA. .A.1r.cV inb red In Blo ck 20. If d

*5.

SUPPLEMENTARY NOTES

1~~

K EY WORDS (Ce.tInu. on v.5.,..

Ire
. , Rupoil)

t.

\
20

i
~

sId . IV n.c..zoiy ind Id.ntIty by block nuoib.r )

Aerodynamics
Aeordynam2.c Force
Aerod~naa c Moment

~~~

ST R A CT (ConIInIA en ,.v.1s. .Id. H rrec.S.WF end I.ftntUy by bloc k n~~~b..1

Several general formulas relating aerodynamic forces and m~~ents act ing on
finite solid bodies imuersed in a fluid to the t ime -variation of vorticity
moment integrals are presented. The formulas are valid for two- and three
dimensional flows and are shown to form a theory for aerodynamic forces and
moments which encompasses much of the existing aerodynamic theories.

DD

~~~~~~~~~~~

1473

EDITION OF INOV SS IS OBSO*.. ETE

clps~ ifi.d
Ufl~

/ 53

~~~~~~~~~~

S
i ,
i,
SECURITY C L A S S I FI C AT I O N OF THIS PAGE (

~~~~~~~~~~~~~~~~~~~~~~
~~~~ .

~~~~~~~~ -~~~~~~~i

~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~ ~

flsei Fs r, dI
~ ~

r
~ ~~

~~~

~: TTTI

.::: :: :

~~~

I
I
1i
iu
iir
:
:
~:

:
I
i 1J~~
LI~~T:

TABLE OF CONTENT
ABSTRACT
I. INTRODUCTION

II. VORTICITY DYNAMICS

III.

SELECTED THEOREMS AND FORMULAS FOR THE KINETIC ASPECT

14

1. Asymptotic Behavior and Effective Extent of Vorticity Fluid .

2. Principle of Total Vorticity Conservation
TwoDimensional
3.

Flow s

Stress Outside the Vortical Regions

2.
3.
4.
5.
V.

17

24

Principle of Total Vorticity Conservation Three Dimensional

Flows

BiotSavart s Law
Asymptotic Behavior of Velocity Field
Velocity Integrals in Large Bounded Regions
VelocityMoment Integrals in Large Bounded Regions

24
25

28

31
36

AERODYNAMIC FORCE AND MOMENT

40

1. Aerodynamic Force
2. Moment ~f Aerodynamic Force
3. Suninary

40
43
44

VI. REFERENCES

~ CtSSO~
MIIS

br

jUST~FlCATIO ~

I
1

14
23

IV . SELECTED THEOREMS AND FORMULAS FOR THE KINEMATIC ASPECT

1.

v4 iite ~5etIOS ~~
on
1
~~~t~
~

.,
~~

.-

~~~

--

-- :-- .,

-~~~~

~~-

_ _J _ _ _
~~
~

ABSTRACT
Several general formulas relating aerodynamic forces and moments
acting on finite solid bodies immersed in a fluid to the timevariation
of vorticitymounent integrals are presented . The formulas are valid for
two and threedimensional flows and are shown to form a theory for aerodynamic forces and moments which encompasses much of the existing aerodynamic
theories .

_______

~~ ---

~~~~~~~

.
-

~~~~~~

i
iI
~~.
~~ ~ T~ W r r ~~~~..~_4_
~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~

H,
I. INTRODUCTION

The problem of predicting aerodynamic forces and moments acting on

finite solid bodies immersed in and moving relative to a fluid has occupied
the center stale of aerodynamic research from the end of the nineteenth
century onwards. In fact, it is this focal problem that distinguishes
the science of aerodynamics from other branches of theoretical fluid mechanics.
Studies of the motion of the fluid relative to the solid bodies of course
represent a fundamental aspect of aerodynamics. In the majority of aerodynamic
applications , however, such studies do not represent ends in themselves.
Rather, they are undertaken in recognition

of the fact that the motion

of the fluid is ultimately responsible for the forces and moments exerted
on the solid bodies by the fluid .
H

It is known that the problem of finding analytical solutions to equations

of fluid motion associated with solid configurations of practical importance
p r esen t s cons ide r able , often ins urmou n table , mathematical d i f f i c u l t i e s .

Historically , therefore, the most remarkable advances in aerodynamics

were brought about by aerodynamicists who perceived approaches for the
prediction of aerodynamic forces and moments that avoid , as much as possible ,
entanglement with the details of the fluid motion. For example, the
KuttaJoukowski theorem, i.e. the circulation theory for twodimensional
steady motion , permits the lift force acting on a solid body to be determined
from a knowledge of the circulation about the body . For the case of a
twodimensional body with a sharp trailing edge, the Kutta condition ,
requiring the ~ear stagnation point of a potential flow to be located

at the trailing edge, is known to yield acceptably accurate values of

the circulation, provided that the flow is steady and does not separate
over an appreciable region around the solid body.

.-

i L . b~~_ ~~~_~~ _ ~~

-.- - .--- ---

.-.~~~~~~ ~

With the KII ta condition,

,,-* .~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~c..~

L - ~.

.#

. *

,-

~~~~~~~~~~~~ _ ~-..

~A ~~~~~

2
the problem of predicting the lift force for such a flow is reducible to

that of solving an integral equation (Ref. 1).

The unknown function of this

integral equation is a distribution of singularity (sources , sinks, and vortices) over the body surface. There is no need to know the fluid motion away
f r om th e bod y except that it is reasonably represented by a potential flow

about the body .

Extensions of the circulation theory to threedimensional and to
unsteady flows

are generally based on the concept of bound vortices

and free vortices . The bound vortices are considered to move wi th the
solid body. The free vortices ar e con side r ed to be shed f r om the sol id
body either because of the Helmholtz theorem , which states that a vortex

f ilament cannot beg in or terminate wi thin the fluid domain , or because the
total circulation of the entire fluid system is required to be zero . These
free vortices are located at finite distances from the solid bodies and
they make a finite contribution to the fluid motion near the solid surfaces .
This contribution is quantitatively determinate once the spatial distribution
of the free vortices is known . This distribution is dependent on the con
p lex processes of sheddi ng of vortices and subsequent transpor t of free
vortices in the f l u i d .

To avoid the detailed computation necessary for an

accurate determination of this distribution, most previous authors have

prescribed the motion of free vortices in a simple manner . For example ,
with Prandtls liftingline theory , it is usually assumed that the free

vortices remain stationary relative to the freestream .

The circulation theory and its extensions permit reasonably accurate

i~ 1

predictions of the l i f t force for some solid configurations and flow environ
ments . The scope of applicability of this theory and its extensions has not
been precisely established . It is wellknown , however

that considerable

uncertainties exist regarding the value of the circulation to b~ prescribed

~~

~~~~~~~~~~~~~~~~~~~~

,
~~~~~~~~~~~~~~

-.~--,-

.-

____

-.
~~~~~~~~~~

in cases where the solid body does not possess a sharp trailing edge , where
appreciable regions of flow separation exist , or where the motion of the solid
is timedependent .

These uncertainties arise mainly because assu mp ti on s , or

hypotheses, utilized in the development of the theory are often adhoc and ,
c on seque n tl y, the theory is not readily interpreted as an approximation of
all y
a specific phjsical phenomenon. For example, the bound vortex is us u

described in wellknown treatises as replacing an airfoil (or a wing) .

The vorticity , defined as the curl of the velocity , is twice the angular
velocity in a solid region . If the airfoil is not rotating, then clearly
it does not possess vorticity . The bound vortex is therefore not an approximation of the airfoil. To the experienced aerodynamicist, the remarkable
agreement between the predicted and measured l i f t forces under certain circumstances po in ts to the inevitable conclusion , that the circulation theory is a

reasonable approximation of the physics of the problem under these circum

stances .

It was suggested (Ref. 2) that the bound vortex does represent the

airfoil plus its boundary layers. The fact that such an interpretation wus
emphasized so very recently (in 1976 ) has motivated the principa l investigator
to examine critically the historical development of the circulation theory,
together with its more recent extensions , with the hope of facilitating
the interpretation of the circulation theory and thus contributing to
i ts f u r ther developmen t for threedimensional flows, unsteady flows, and
separated flows. In the course of this study, seve r al general formulas

relating aerodynamic forces and moments to the timevariation of vorticity

moment integrals were uncovered . It became evident that these formulas have

far reaching consequences in the realms both of theoretical aerodynamics and of

computational aerQdynamics. In particular , it was shown that the formulas form
a theory for aerodynamic forces and moments. This theory encompasses much of
the existing aerodynamic theories. For example , the circulation theory for

en

~~~~~

______________

1*

-u

~ 7~~&

. ~

- . ~~~~~~
~~

*1

-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
*

steady lift and its various extensions are readily interpreted as various levels
of approximation of the general theory. The purpose of this report is to present

this general theory.

A distinguishing feature of the present theory is that the concept of bound
vortex, or that of singularity eleme~Lts such as sources, sinks, and vortex filaments, is not embodied in the genera~. formulas forming the theory . Rather , the

actual vorticity distributions of the fluid and of the solid, the latter being re
lated to the rotational motion of the solid , enter these formulas. The starting

point of the present theory is a rotational flow analysis. Consequently, the

theory is applicable to viscous flows. This freedom from bondAge is important
in the interpretation of the various aerodynamic theories. For example, it permits

a precise definition of the circulation about a twodimensional solid in the case

where an appreciable region of separation exists. While the KuttaJoukovski theorem
predicts zero drag, the general formula does relate the timevariation of a vorti

city-moment integral to a nonzero drag, including the profile drag. The formulas
predict an unsteady drag without the customary energy or apparent mass consideration .
These general formulas, in fact , clearly point out the basis principles for mini
mizing the drag and for maximizing the lift. Many of the measures proposed for dragreduction and for liftaugmentation are readily interpretable on the basis of
these principles.

The present research deals only with incompressible flows ,

although the basic princ iples described here are certainly applicable to

compressible flows as well .

In recent years , extensive efforts have been in progress at many research
institutions to develop numerical methods for the solution of aerodynamic problems .
An ultimate goal of these efforts is to make available methods of predicting
aerod ynamic forces and moments that are more accurate and that possess a wider

scope of validity than the circulation theory. These efforts are divisible into
two major categories.

~~~~

~~~~~~~

.I*I

.L

.
. :

In one category, numerical methods are being developed

~~~~~~~~

- --

- --~~_ _ _ _ _
1*
-

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

for the solution of inviscid flow equations . The free vortices, are assumed to

convect , but not to diffuse , with the fluid . The solid body is represented by

a singularity distribution . Conceptually, these numerical methods utilize the

basic assumptions of the circulation theory. They relax the restrictions of
analy tical methoO based on the circulation theory through detailed compu
tation . For example, the linearization procedure introduced in classical
studies of the unsteady twodimensional airfoil problem (e.g. Ref. 3) is no
longer necessary if numerical methods (e.g. Ref. 4) are employed. These
numerical methods are of course expected to be subject to the wellknown

limitations of the inviscid flow assumption . In this regard , the availability of the general formulas are expected to offer clearer interpretation
of these numerical methods and better definition of. their scope of application.

In the second ca tegory , nume rical me thods are being developed for the
solution of differential equations governing viscous flows. Impressive
progress has been made in recent years in the numerical solution of two
dimensional laminar separated flow problems as well as in the establishment

of turbulence models for separated flows. For threedimensional separated

flows , the development of numerical methods is hindered by excessive computation requirements (Ref. 5).Methods that possess superior computational

efficiency are therefore of critical importance . Du r in g the past few

years, the principal i~vestigator and his coworkers have developed a

new numerical approach which permits the confinement of the solution field
to the vortical region of the flow (Ref. 6). In comparison to available
finitedifference and finiteelement methods, all of which require the
solution field to include the potential region in addition to the vortical

region, the new approach requires a dras tically smalle r number of grid
points.

The computation requirements using this new approach, called the

integrodifferential approach, are consequently drastically smaller than

~~~~~~~~~~~~~~~~~

I~

~~

i_ __ .

6
those using other methods (Ref. 7,8) .

The new approach uses the vorticity

vector as a field variable in place of the pressure . The general formulas

presented in this report provide a convenient means of computing the aero-

dynamic forces and momenta directly from vorticity distributions . That is ,

using these general formulas, it is no longer necessary to compute first the
pressure and shear stress distributions on the solid surfaces from the value

and normal gradient of vorticity on the surface (Ref. 8) and then the integra ted forces and moments. In addition , the general formulas suggested
several promisi ng techniques for minimizing the required computation .

Several important theorems of fluid dynamics are utilized in the den

vation of general formulas for aerodynamic forces and moments. Similar
theorems and formulas are given in many wellknown textbooks on aerodynamics,
e.g. Refs . 9 and 10.

These earlier the or ems and fo rmulas , however , are

tradi tionally considered in the context of an infinite limitless fluid ,

i.e., an infinite fluid with no internal boundaries . The present theorems

and formulas are valid in the presence of internal boundaries that repre
sent solid surfaces.

The practical importance of this more general validity

needs no emphasis sin ce the int erac tion be twee n the solids an d the fluid is
indeed what the subject of aerodynamics is about.

The subjec t of this r epo r t received some a ttention in th e 1950s, nearly

a quarter of a century ago.

Phillips (Ref. 11) presented a formula relating

the fluid momentum to an integral of vorticity moment for a twodimensional

flow associated with a cylinder in translation . The general formulas

presented in this report are valid for two and threedimensional flows
associated with one or more finite solid bodies of any arbitrary shape

exec ut i ng any p rescribed stead y or timedependent translation and/or ro

tation . Moreau (Ref. 12) presented formulas for a limitless fluid and
for a portion of fluid subject to certain order condi t ions a t infinity.

_
_
-

~~~~~~~~~~~~~~~~~~~~~~

I t

Truesdell (Ref. 13) comeented that Moreau emphasized his application to a

limitless fluid , all but a finite interior part of which is in irrotational
or circulation preserving motion. In this connection we should beware of

the extremely strong order conditions at infinity required in order to

get simple results , order conditions , indeed , which possibly may never

be satisfied . In this report , formulas are rigorously derived for the

viscous flow of fluids past finite bodies, using order conditions at infinity that are shown to be satisfied under quite general circumstances.
The applications of the theory formed from these formulas will be presented
in future reports .

I
I

.-

- -
~~~

..- .

~~~~ -

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I
- I ~~~~~~~

.
~~ I.

_________
_ _ _
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

_
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _
_ ~~~~~~~~~~~~~~~~
-~~~~~~~~
~~~~~~
~~~~~~~~~~~~~~~~~~

,
,

__________

_
,~ -~ .rI .

_ _ _ _

~ ._ p p

p~~~~

pi

~~~~~~~~~~~

II. VORTICITY DYNAMICS

The timedependent motion of an infinite incompressible fluid with

uniform viscosity relative to one or more imeereed solid bodies is considered

in the present study . The solid bodies are initially at rest in the fluid
also at rest and are located within finite distances from one another.
Subsequent prescribed motionsof the solid bodies induce a corresponding
motion of the fluid . At lange time levels after the motion has initiated ,
if the solid bodies move uniformly at a constant translations-I velocity

relative to the freestream , then the possibili ty of an asymp totic stead y

flow exists. Al ternatively, the possibility of a timedependent flow
involving periodic vortex shedding, as evidenced by the wellknown Karman
vortex street behind a circular cylinder , also exists.

In the present

work , a steady flow, when it exists , is considered to be approached asymptotically at large time levels after the initiation of the solid motion.
If the solid bodies do not move uniformly, or if the solid motion is time
dependent, then the motion of the fluid is necessarily timedependent.
The familiar differential equations describing the timedependent
fluid motion are the continuity and NavierStokes equations :
+ 4

V .v

+
.4

where v, p, p, and

-
(v .V )v

1+

(lii)
2+

~ V p +v V v

(112)

V are respectively the velocity vector , the pressure ,

the density , and the kinematic viscosity of the fluid , p being a constant
in the present study. For simplicity, the kinematic viscosity of the
fluid is considered to be uniform in this report. It is not 4fficult

~~

- - ---
~

9
to generalize the analyses given here to flows where the viscosity is
not uniform. Such a generalization, however, is not essential to the
purpose of the present work.
In this report, the region occupied by the fluid is designa ted

R ~
f

A coordinate sys tem with its origin located within finite distances from

B~, is used. Unless otherwise

all solid surfaces , collectively designated by

specified , this coordinate system is considered to be at rest relative

to the freestream . The fluid region R

is bounded internally by 8
f
5
and externally by a close boundary
at infinity. The region occupied
~
by the jth solid body is designated R , which is bounded externall y
3
by B~ . The limitless reg ion jointly occupied by all the solid bodice
and the fluid is designated

I~,,.

It is convenient to introduce the vorticity vector

+

V x v

defined by
(113)

and tQ consider the vor ticity transport equation

4

3w

~~~~~

v x (vxw )+ v
+ 4

V ~~w

(114)

obtained by taking the curl of both sides of Eq. (112) and using Eqi .
(u i) and (113) .
-

.
,

The set of equations (lIi), (113), and (114) replaces the set

of equations (11 1) and (112) . There are several reasons for using

in the formulation of the problem. In the first place, the circulation

~:

.
~

theory for the l i f t force suggests that the vorticity of the flow, which

ultimately should be traced to the circulation, is responsible ~or the

forces and moments exerted by the fluid on the solid. Secondly , as shown

in Ref erence 6 , the use of the vorticity vector , which is intimately connected

--

~~~~~~~~~~~~~~~~~~~~

_ _ _ _

10
with viscosity effects , permits the solution field for the incompressible
flow problem to be confined to the viscous region only.
4

set of equa tions in term? of w

Thirdly, the

decomposes conveniently into a kinematic

aspect and a kiuetic~ aspect, each aspect constituting an entity by itself.

The first feature stated above provided the motivation for the present

effort in developing general formulas relating aerodynamic fo~ces and

moments to the timevariation of vorticity-moment integrals. The advantages

offered by the second feature have been extensively studied by this author
and his coworkers in a series of previous articles in the context of computation

methods. Mast of the numerical results obtained thus far have been for
twodimensional incompressible laminar flows , both timedependent (Refs. 6, 7 , 8)
and steadystate (Ref. 14). Some results have been obtained recently,
-

however, for relatively simp le turbulent flows (Ref. 15) . In addition ,

extensions to compressible flow problems have been suggested (Refe . 16, 17) .

The importance of the third feature in the present work is due to

the fact that the physical processes of flow development are clearly delineated
once the overall problem is decomposed into its kinematic and kinetic
aspects .

In particular , with vort icity as a field variable , considerable

insi ght is gained by examining the differential equations describing these

i thout emp loy ing detailed mathematical or numerical analyses .
two aspects , w
The kinema tic aspec t of the problem concerns the relationship between
the wox ticity distribution at any given instant of time and the velocity
distribution at the same instant. The differential equations describing
this aspect are the cont inuity and vorticity definition equations (ilI )
and (114) . Since the density of the solid bodies does not undergo appreciable
change , the continuity equation (Iti ) is obviously valid within the solid
regions

as well as in the fluid region R~ . If , within a solid region

one def ines a vorticity field according to Rq. (113), then the kinematics

~~ ~~~~,.,

~~~ ,-- ~~~~~~~~~~~~~~~~~

---~~~~ ~~
~S
~~~ -

.
~ ~

-S.-
~------ 5 ,
__..* .
~

, .~~~~~ .

U
~~
f _S
5-

~~

~~~~~~~~~~

~~~~~~~

~~~~~~~ *

...,.
.

. ..:

~~

~~~~~ ~~ TJ

~~~

1.1
of the solid bodies and the fluid are described by the same differential
equations. The stressstrain relation., which differentiate the fluid
from the solid bodies kinetically, do not enter the kinematic relation
betveeen the velocity field and the vorticity field.

As a consequence ,

the solid bodies and the fluid can be treated together as one kinema tical
system.
The recognization of the fact just mentioned makes it relatively
simple to derive the kinematic theorems and formulas presented in the
report. For these theorems and formulas, the region of interest is limitless
.ud ~h.differential equations leading to these theorems and formulas

are linear. In fact , all previously available kinematic theorems and

formulas derived for an infinite unlimited fluid are imsediately appli cable
to the present situation of an infinite fluid with one or more liasersed
solid bodies.

This fact is not well recognized. The treatment of the

solid bodies and the fluid together as one kinematical system has not received
emphasis in the literature. In this report , kinematic theorems and formulas
are derived by treating the solid bodies and the fluid together. The
author has , in addition , rederived each of these formulas and theorems
by considering only the limited fluid region which is bounded intercally
by solid surfaces. The presence of boundaries makes the derivations lengthy
and algebrai call y tedious .

In this report , only a few of these rederivation.

are presented to demonstrate the validity of treating the solid bodies

and the fluid together as one kinematical system.
The kinetic aspect of the problem is concerned with the development
-

of the vorticity field with time. This aspect is described by the vorticity

transport equation (114). This equation is nonlinear in the sense that

the first term on its righthand side involves the product of

4.

w, and

and

is kinematically a function of w. This equation is valid

only in th e fluid domain , which is limited.

&

~~~~ .

..._ [~~~~~~

V
___ _ _ _ .
. J _,._- &
~~

- - ~~~~

S
. ~~~~

S~

~~

U~- -
. ~~

~
~~
~~~~~~

5 - - - - 5 . _
.
. ..
~ a.

~~
--

,
-

~~-

~~~~~~

- - -

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

12

i
Recause of the nonlinearity of the differential equation and the
necessity of treating a limited region , the analysis of the kinetic aspect
of the problem presents greater mathematical difficulties than the analysis
of the kinematic aspect. Under certain conditions, it is possible to
specify the vorticity field appromiastely without actually solving the
vorticity transport equation .

It is then only necessary to deal with

the kinematic aspect of the problem which, as stated earlier, is described

by linear differential equations applicable to the entire limitless region
Under these circumstances, the aerodynamic forces and momenta are
Sometimes obtainable in a relatively simple manner.
The available literature on classical theories for aerodynamic forces
and moments shows that the avoidance of the kinetic part of the problem
has bee~ an essential ingredient of these theories. There have been many

recent efforts , via numerical methods, to treat the kinetic aspect either
par tiall y, e.g., on the basis of the inviscid flow equations , or fully .
These efforts can benefit from a clear understanding of the physical processes
involved in the development of the vorticity field in the fluid . A significant
amount of information already exists in the literature on this topic (e.g.
See Ref. 10). Those feature. of vorticityfield development that are
pertinent to the present work are described briefly below.
For an inviscid fluid , the last term in Eq. (114)vanishes and the
vorticity is convected with the fluid in the sense that the vorticity
flux

W .ds associated with each material element

ds moving with the

fluid remains a constant for all times. This wellknown theorem of Rel~ iolta,

a proof of which is available in many textbooks, e.g. Ref. 18, is modified

in the case of a real fluid by the process of vorticity diffusion. According

to Eq. (114), changes in the vorticity flux take place only by diffusion .

_
5

-5---

~~

Si_ _S._
~

~~~~~~~~

- ~~~~~~~~

~~~~

~~~~~

-- ~~~
*

1
:
~

5
- .

13

~~~~~
Vorticity flux cannot be created or destroyed in the interior of a fluid.

For the problem under consideration , the vorticity is obviously everywhere

zero prior to the impulsive start of the motion of the solid bodies.
The interior of the fluid domain therefore can become vortical only if
vorticity d~iffuse. across the boundaries of the fluid region. Consequently ,
i~~~diately after the onset of the motion, the vorticity is ever ywhere
zero in the fluid except at the boundaries in contact with the solid bodies.
That is , the fluid motion ii ediately after the onset of the motion has
a nonzero tangential velocity relative to the solid bodies at the solid
boundaries.

The discontinuity in tangential velocity constitutes a sheet

of concentrated vorticity (vortex sheet) at the boundaries. At subsequent

time levels, this concentrated vorticity spreads into the interior of
the fluid domain by diffusion and, once there, is transported away from
the boundaries by both convection and diffusion. At the same time, the
noslip condition provides a mechanism for the continual generation of
vorticity at the boundaries. The general flow pattern therefore contains
vor tical regions surrounding the solid bodies and vortical wakes trailing
the solid bodies. Outside of these vortical regions and wakes the flow
is essentially free of vorticity and therefore irrotational . In particular,
if the flow Reynolds number is not small, then the vorticity spreads by
diffusion only a-short distance from the boundaries before it is carried
away with the fluid by convection. Therefore a large region of the fluid ,
ahead and to the side of the solid bodies, is essentially free of vorticity
and irrotational .
In the next two Chapters, a number of theorems and formulas are derived
using the above described kinematic and kinetic characteristics of the
flow.

- - 5- -

~~~~~

T ~~~~ TF ~~~~~Y . _ . _ _ ~J

p
.

14

III. SELECTED THEORE1~S AND FORMULAS FOR THE KINETIC ASPECT

1.

Asymptotic lehavior and Effective Extent of Vorticity Field.

For the present problem, the vorticity decays exponentially

wi th increasing distance from the origin at large distances for

all finite time levels after the onset of the motion and the effective
extend of the vortical region is finite. The finite extent of the
vortical region is a consequence of the fact that the vorticity
is transported in the fluid by finiterate processes

and cannot

be created in the interior of a fluid domain. This fact as well

as the asymptotic behavior of the vorticity distribution at large
distances from solid bodies are established below by considering
the fundamental solution of the diffusion equation.
The fundamental solution F of the diffusion equation, i.e. the
Greens function for an infinite unlimited region, can be expressed as
(Ref. 18 and 19)

-!

t;

P(

r and

where

,
,
~~~~

t )
0

are position vectors and

of the problem, i.e., d

~~~~~~~

exp {

4 v(t t

(1114)

is the dimensionality

1, 2, and 3 respectively for one , two , and

threedimensional problems. For the present problem , the fundamental

solution represents the vorticity distribution at the time level

t resulting

from the diffusion of a concentrated vorticity which is of unit strength

and is located at the point

_ _ _ _

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I~~~~~~~~~~

__

-5--

~~~~~~~~~~~~~~~~~~~~~~~~

~~

at the time level

t ,
0

with

..
~

5--

t ct .
0

_S

5 - - S _ S5-_

~ 1~ *~~ - *~~

~~~~~~

_ _ _ _

J~~

V.

________________________

-_

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

15
If , at the time level
region

t , the vorticity is nonzero, only in an elemental

o
located at
a n d the val ue of vo r t icity in dR ~, is

dR

0
then the vorticity distribution at the subsequent time level t
If , at the time level

unlimited ~.egion
time level

~~~~

t , the vorticity distribution is known in the

o
then the vorticity distribution at a subsequent

is expressible as an integral :

is

(i ,

t)

L
~

t )dR
0 0

(1112)

where the subscript for dR

indicates that the integration is performed
0
in the

space.

Equation (1112) is valid in an infinite unlimited region in the

absence of convection. The form of the fundamental solution F clearly
shows that, if the vorticity is nonzero at the time level

t only within
o
finite distances from the origin, then the vorticity at any subsequent
finite time level

approaches zero exponentially with increasing distance

from the origin, at large distances. Therefore, the vortical region

is effectively confined to a finite region at any finite time level

t.

For a fluid region bounded internally by solid surfaces and in which

convective process is present , Eq. (III2) needs to be generalized. It
is obvious that the convective process, being one of finite rate, does
n ot alt er the abo
ve conclusions regarding the effective extent of vortical
regions and the asymptotic behavior of vorticity at any finite time level .
Similarly , the pre senc e of solid surfaces in the fluid provides a mechanism
for introducing vorticity at the boundaries of the fluid region and , as
long as these boundaries are within finite distances from the origin ,
.
-

the introduction of vorticity does not alter the above conclusions . For

~~~~~~~~~~~~~~~~~~~~ ~~~

5- -

~~ ~~~~~

~~~~~~~~~~

TTTTI~~
16

twodimensional flows, the generalized version of Eq. (1112.)is expressible

as (Refa . 17 and 20):
t)

dt~

dR~

(
:
w 0~~s~ ~FdR0 + vl dt
aj ~~ ~

~~~r ) .

~ 0dB 0

(1113)

where the subscript a indicates that the variables, differentiations,

+

and integrations are in the r0, to space, e.g. ~ ~~~~~ t 0 ) . The

second and third integrals of Eq. (1113) represent respectively the convective
process and the effect of solid boundaries. Each of the integrands in
the integrals of Eq. (1113) is directly proportional to F and/or
-

F.

At any given time level t and at large distances r from the origin,
F and ~ F decays exponentially with increasing r. It follows that,

for any problem in which B5 consists of finite surfaces located within

finite distances from the origin, if the vorticity decays exponential

with increasing

r at large r

for all time levels previous to t , then

the vorticity decays exponentially with increasing r also at the time

level

t.

In fact , for such a problem, if the vorticity is confined within

finite distances from the origin at any given instant of time , then the
subsequent finite time levels. For the present problem involving an impulsively
started motion, the vorticity is nonzero imaediately after the onset
of the motion only on the solid surfaces.- Consequently, the statements
made in the first paragraph of this Section are true for twodimensional
flows. It can be similarly shown that the se statemen ts are als o true
for threedimensional flows.

- -- -~~~~~~~~~~
~ ~~

--- .I . T TE.
~ ~ ~~~~
~~~~~~
-

~~~~~~~~~~~~~~~~ -~~~~~~

vorticity decays exponentially with increasing r at large r at all

.1

--

17
2.

TvoDimenional Plows.

Principle of Total Vorticity Conservation

It shall be shown that, for the flow of an incom pres sib le flui d pas t
solid bodies, the total vorticity in the infinite unlimited apace occupied
jointly by the fluid and the solid bodies is invariant with respec t to
t ime , prov ~ded that an order condition for the vorticity at infinity

is

satisfied . This order condition is that the vortici ty approaches zero

as

11

, where n > d , d being the dimensiouality of the problem.

The above statement will be referred to as the princip le of total

vorticity conservation and is expressible mathematically as
d

provided that

wdl
as

approaches r~~

As noted earlier

~~~ ,

(1114)

with

n> d.

the solid bodies and the fluid can be treated together

as one kinematic system in the present prob lem. For threediaenaiona l

fl ows , since the vorticity field is solenoidal and is effectively confined
to a finite region, all vorticitylines in the combined system form closed

curves . Conseqently, one has

I-

w d1

11

In the next Chapter, a proof of Eq. (1115) is presented treating the

solid bodies and the fluid as separate kinematical systems. Clearly, Eq. (iII-4~
follows directly from Eq. (1115). Thus the total vorticity is not only

c onserve d , it must always be zero in threedimensional flows.

For twodimensional flows, the vorticity lines are directed perpendicular
to the plane of the flow. The vorticitylines extend to infinity in the

-.

.
~~~~~~~~-~~~~~ - - ____ _

.5___ ______ _._____ ___ _ _

5-- & _ s~~~~~~~ -

.-

--5

5-

-5- -

5-
* I
7
. S ~ S .. -

~
.

- 5-

~~~~~~~~~~~~~~~~~~~~~~

- -

-5-

--

18
direction perpendicular to the flow, and they do not form closed curves
in the plane of the flow. The total vorticity of the fluid and the solid
bodies, that is, the integral of vorticity in the infinite unlimited plane
of the flow , is still conserved . This principle of total vorticity conservation
for twodi~ ensional flows is not usually discussed in standard treatises
on fluid dynamics. A proof of this principle for twodimensional flows
is given below on the basis of the kinetics of the problem.
For twodimensional flows, the vorticity transport equation (
114)
can be rewritten as
-

where

111 6

-vV x x ~

denotes a substantial derivative.

The time rate of change of the total vorticity of the fluid is

5.

ff(t)

.
~~~~

where

R (t )
f

dR

~~
R (t)
f

(
7
)

dR

is the entire region occupied by the fluid and is a function

of the time.
Placing Eq. (1116)
one obt ain s

~
where

into Eq. (1117) and using Stoke s theor e

m,

J
R (t)
~~

w dR

~~

and the boundary at infinity

~4~~~~~#

(111-8)

v f (Vxw )xndB
R(t)

is a unit outward directed normal vector~

In Eq. (1118) , the boundary

-~~~~r .

--

:. :L- - : : .

- -
~

consists of the

4~~~d

B , to Eq.(I1I 5)

B~,. The contribution ~

~~~~~~~~~~~~~~~~

--5

boundary

UL ft

. 1.

~~~~~~~
~~~~~~~
~~~~~~

~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~

- -

19

r ~~ for large

approach zero as

is zero provided that w

r , with

n >2.

exponentially wi th incretaing r for large

~ approaches zero
r , the above order condition at large r is satisfied for the present

Since

problem. The boundary B in Eq. (1118) can therefore be replaced by

3~ .

the solid boundary

Equation (112) can be rewritten as

h_ v x
~~
~~ ~~
term
in Eq. (1119) with
each
product
of
vector
Taking the
-

integrate the resulting equation around B8, one has

~~ x~ dB

~~~

~~ px~ dB

V~

(111

9)

and

x~ )
~~ dB

(111-10)
-

The first integral on the righthand side of Eq. (111 10) is zero
by virtue of the aing levaluedness of pressure on

B~ . This fact can

also be shown by using ~he Stoke s theorem and the fact that the curl

of the gradi ent of any scalar function is zero . Combining Eqs. (1118)
and (11110) therefore gives

.-

~~~

w dR

R (t )
f

5
B

x~ dB

(111li)

bound externally by B . With the no

5
~
on B~ is identical
slip condition, the substantial acceleration
Consider now the region

I:

for the solid bodies and for the fluid . Using Stokes s theorem and the
fact that the outward normal vector for
f or

R f~ one obtains :

~
!
~~~~~~ J~~~~

u dR
f
R (
t)
f

-.

is directed opposite to that

R
5

5-

_ -

(t )

~ x (~~ ) dR

~~~~~ ~~~~~~~

~~

(11112)

~~-

20

.1

R j~ the velocity vector is given by

In the solid region

v. +
3

*i .

x r

(11113)

where

is the rectilinear velocity of the solid body

and

is directed
~j
perpendicular to the plane of the flow . The vorticity in the solid body
is its rotational velocity . For two-dimensional problems,

is readily obtainable by taking the curl of the velocity vector as

given by Eq. (11113) and is

+

(U

(111 14)

~~~~

Using vector differential identities , it is simp le to show tha t

in each of the solid regions R~ . One thus obtains from

Eq. (111-12)
-

w dR
~~
R ( t)

d
~~~~~

t)
R (

dR

(111-15)

It states that the

Equation (111 15) is equivalent to Eq. (1114) .

rate of change of the total vorticity is equal in magnitude and opposite

in sign to the rate of change of the total vorticity in the solid bodies
or , equivalently , the total vorticity in the infinite unlimited region
is zero.

of the solid bodies are

If the rotationa l velocities

pr .scribed functions of time , then the rate of change of the total vorticity
in the fluid can be calculated usin g the foll owin g sim ple f ormu la :
d

I
f

~~~~~~~~~~~~~~~~~~~~~
s

~~~~~~~~~~~~~~~~~~~~~~~~~~

t-

~~~~~~~~~~~~~~~~

~~~~~~~~~~~~

5--.-

-~~~ 5-

~~~~~~~~~~~~~~~~~~~~~~~~~

(11116)

~~~~~~~~~~~~

-~

5--

~~~~~~~~~~~~~~~~~~~~~

- -

~~~~~~~~~~~

~~~~

~~~~~~~~~~~~~~

.~~

21
where

is the total number of solid bodies present , and

is the

j.

size of the solid body

Equation (11115 ) can be integrated with respect to time, yielding

.1

w dR+J
w dR
R (t)
R (t)
f
8

where A

(11117)

is a constant vector .

For a motion starting from rest , the total vorticity in the combined
fluid and solid regions is zero before the onset of the motion. Consequently,
A

0 and one obtaLns (1115) , which states that the vorticity tn th e

combined solid and fluid regions , i.e., the infinite unlimited region,
is always zero.

Thus, if the solid motion is prescribed at any time level ,

the total vorticity in the fluid is easily calculated from

+

J ~~w dR

,
(11118)

~~
~~~~~

~~~~~

There are several conceptual differences between the principle of

total vorticiy conservation discussed here and the usual understanding

of invariance of vorticity integral with respect to time.

In the p r esen t work , the solid regions are included in the evaluation
of the total vorticity . The integrands of Eqs. (1114 )and (1115) are
piecewise continuous and the integrals converge . The meaning of the total

vorticity in the unlimited infinite space jointly occupied by the fluid

and the solid bodies is unambiguous. While previous discussions of the
invariance of the total vorticity are usually made in the context of an
inviscid fluid or of an unlimited infinite fluid region (in the absence

of internal boundaries), the present study utilizes the noBlip condition

at the solid boundaries and permits the presence of such b~ indaries in

- ,

5-

~~~~ -

---5 --- - -

..
~~
.
. ~~~~~~~~~~~~~~~~~

_,

~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
i
~~
~~~~~~ ~~

I.

-~~

~~~~~~~~

- _,~~~~

IL U W :
~~~~~~~i

22
the fluid.

In Ref . 21, Section 8.2, a formula similar to Eq. (11118),

but specialized to a single solid and including an additional term representing the contribution of the velocity at a surface enclosing the solid, is
presented . In the present work, it is shown that this contribution is
absent if the surface is sufficiently distant from the solid bodies and
the order condition for the vorticity vector is satisfied. The formula
(11118) is derived for one or more solid bodies in the present Chapter .
Ref. 21 also gives a formula for the rate of change of the total
vorticity in the fluid . That formula contains a term representing conduction

of vorticity through the solid boundary .

It is pointed out that further

dynamic (kinetic) equations are needed to evaluate this term. The present
r esult , Eq. (11116) , shows that this term is given simply by the rate
of change of the total vorticity of the solid bodies.
In Ref. 10, it is shown that in threedimensions the total vorticity
is zero in a region which contains the fluid region and a region extending
beyond the actual boundaries. The present results show that the proper
extension of the fluid region is simply the solid regions in which the
correct vorticity values to assign are the actual vorticity of the solid
bodies . For two-dimensional flows, the literature emphasizes the possibility
of the existence of a nonzero circulation around closed paths at large
distances from the solids. According to the present results, this possibility
does not exist for a real fluid at any finite time level after a motion
An evaluation of the conduction of vorticity through

solid boundaries is not necessary since this conduction process conserves

the total vorticity in the infinite unlimited region occupied jointly
by the fluid and the solid bodies.

~~~~~~~~~~~~~~~~

--

______

~
~~~~~
- - --- --5 -

--

~~~~

- - - -

has started from rest.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--

--

--

~~~~

-- - - - -

23
3. Stress Outside the Vortical Regions.
Outside the vortical regions , the vorticity i . zero and the viscous
stress is absent. The momentum equation (112) simplifies to
p

(p +pv 2/~ )

(11119 )

The absence of vorticity implies the existence of a scalar potential

such that

(11120)

Placing Eq. (11120) into Eq. (11119) and integrating the resulting
equation in space giv es

.i
~~

2
2.
~

+ f(t)

(11121)

This wellknown equation for unsteady inviscid pressure will be utilized

to derive general formulas for aerodynamic forces and moments in Chapter V.
I

It should be noted that the scalar potential is stngle-valued . For three

dimensional flows past finite solid bodies, the region in which
is singlyconnected mnd therefore

$ is independent of path. For two

dimensional flows, the region in which

H oweve r , the cyclic constant for

exists

exists is multiplyconnected.

$ is zero since the vortical regions

are of finite extent and the total vorticity is zero.

i - I

-.

-~~~~ - - -

- -

-)

5 - -

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

fd~~~~~ &~~~ 5-

t ~1
~

F- ,

P
24

SELE CTED THEOREMS AND FORMULAS FOR THE KINEMATIC ASPECT

IV.

The pr oof ~ of the theorems and formulas for the kinematic aspec t
of the eztetnal flow problem utilize only the kinematic relationship between
the velocity and the vorticity fields and the order conditions for the
vorticity field.
1. Principie of Total Vorticity ConservationThreeDimensional Flows.
+

Let n, b 1, and b be a righthanded set of orthogonal vectors on

2

I~ .
l~ are tangential unit vectors. Let the velocity
~~~
components in the n , b 1, and b d irec tions be ~~ Vbl , and b2 respectively.
2
~
With the noslip condition , these velocity components are identical on
the boundary

B for the fluid and for the solid bodies . The normal component of the
5
3b
vorticity vector on the boundary B is given by
2
and is
~l

continuous across B . That is w

5
to that for the solid body.
Using the fact that

a on

a2

B
for the~ fluid
8

is

identical

is solenoidal , one obtains through vec tor

differential identities :

(
IV-l)
4

where a is an arbitrary constant vector .

.

The divergence theorem then gives

~~~dR

(Iv2)

S t

.5 ~~.~~~~~~~~~
_
_

___________________________

~~~~~~

- _

~~~~~~~

r
~

25

Le t R be the fluid region Rf. Th.boundary

B~ and a boundary at infinity. The contribution

of the solid boundary

of the boundary at infinity is zero provided that

as

r~~ for lar ge

$ then consists
approaches zero

r , with n>3. This order condition is cer tainly

met in the pt~esent problem for which

decays exponentially with

for large r. Therefore, th.boundary B in Eq. (

iV2) can be rep laced
by

B 3.
Since the normal component of

replace

~5 Continuous across

in the righthand side of Eq. (

IV2) by

i~ , one may

where
~~~~
th
is the angular velocity of the ) solid body. One then has
~~~~. 1

J
R

~~

dR

N
2 E

(
~~~

Consider now the region R

bounded externally by

~)dB

(
Iv3)

in Eq. (1112 ) to be the region R

~

B~ . One has
2

( ~i~ .)dB

2 J?i~dR

(I v4 )

Placing Eq. (
IV4)into Eq. (
IV3) and noting that the normal vector
-

in

Eq. (IV3) is directed outward from R while it is directed inward from R in Eq.
f
~
(IV4 )
, one obtains
-

~~

Rf

dR

2 E

t~ R~
3

(
Iv5)

which is equivalent to Eq. (1115).

BiotSavart s Law.

2.

It is shown in Ref. 6 that , by using the fundamental solution of

the Poisson s equation , the kinematics of th. problem, i.e. Eqs. (Ili)

and ( II3 ), is expressible in the fotm of an integral r.presen~a t ion:

T 5 ~~~~~~~ ~~~~~~~~ _ _ .
-i
_

~~~~

.
~~~~~~~~~~~~~~~~~~~~~~

______________

~~~~_ ;
~
~~~~~~

~~~~~~~~~~~~~~~~~~~~~~

- -5 -

.
~~

~~~~ .

~~~~~~~~

26
v(r , t)

where A

and d

prob lem A
and d

4ir

~~~~

(r0r j~

f w x ( r 0 r)

R ~~~
I O

rI

d1
0

(o x )x (t -r
0 ~ dB
~
~~

( IV-6 )

are constants depending on the dimensionality of the

and d

3 for threedimensional problems, A

Zn

2 for twodimensional probiemsand the subscript o indicates

that the variables and the integrations are in the r

0 space , i.e.
etc.
+

Let the region R be the fluid region R

+
N

+ 4

w(r~,, t)

f bounded internally by

the solid boundary B and externally by a boundary at infinity. With

9
a coordinate system attached to the fresetream , the velocity at infinity
is zero and the contribution of the boundary integral to the velocity
field is therefore limited to the solid boundary
The contribution of the solid boundary

is expressibl e as

B~

to the velocity field

; P) dB~
~

p
~~ ((~~.
~~ )
~~ (~~,x~~~)
3

wher e P

35.

(t vi)

is the fun damen tal solu tion of th e Poi sson s equation and is

defined by
1

4 Irr 0I
~

for threedimensional problems

in

for twodimensional problems

Using vector identities , Eq. (

IV7) can be rewritten as

:
,
~~

_
~~~~~~~~~~~

_______________________________________

~~~~~~~~~

~~~~~~~~~~~~ ~~~~~~~~~~

:
~~

___________________

_
_______

~~~~~~~

--

--

~13

tI

27

Consider now the solid region

R~

+~~

~~~~~~~~~

P~ x~~,dB~
0

( Iv-8)

bounded externally by

B~ . Using

the divergence theorem and Stoke s theor em , and noting that the outward
-

normal for the region R j is directed opposit. to

~~~~,

one obtains from

tVa)
Eq. (

R~

))+
~

.(P
(
~ ~~~

)
z(P ~;
~ ) dR ~,

Using vector differential identities, the integrand in Eq. (

IV9)
can be reexpressed as

x ~~ ) x~ P

~~~

~~~

~~P)

(tvb )

it is ea sy to show that

~~P

2
V P ~ 0 for

~~

Thus, in the fluid region R ~ one has ,

f
a

~~~~~

~~~ d

dR
0

One therefore obtains from Eq. (IV -6),

+

;(r , t)
+

+ 4

x (r r)

( tv il)

r r~~ ~~~
0

- ~~~~~~~~~ --- ~~~~~~~~~~~~

~~~

~~~~~~~~

~~~~~~~~~~~~~~~~
-

.
--- -

~~
- --

~~~~~~~~~~~~~~~~~~~~

--

:;

~~~

- --U--

--

- ~~

- -

28

I CA? 3
THIS PAGE IS BEST ~U~~XT! PR&~~ ~ ~

,a~u ooj ~ rui~usH~ I0DDC

where

lx

~~~~~~~~

is the unlimited infini te region jointly occupied by the fluid

Iv1l) is an expression of the BiotSavart s

and the solid bodies. Equation (
law for a distribution of vorticity in an unlimited infinite region.
In fact , by considering the solid bodies and the fluid together as one
tv il) can be imeediately written down without
kineaatical systein, Eq. (
going through the intermediate steps de*ling with the boundary conditions
on

B .
3
3. Asymptotic Behavior of Velocity Field.
To examine the asymptotic behavior of the velocity fi eld at large

IVll) as
distances from the solid bodies, one reexpresses Eq. (
+

v(
t)
~~

V z~ w pdR
0 0
,

For threedimensional problems , expressing the function

+

a Taylor series about the point r0 0 , one obtains

-

r r 0:

r;0

as

t~~ o i~~~ t r 0 ]

-I
V~~~~~~~)
0 r r
~ 0 ~ r0 oJ

...

Ls4
~

f ~v o

+~~~~ + .

~~~~~~~~~

10 1

(IV12)

Or
r _r~
0

+ terms of order

r~~ ,

(
1vU)

,(
IV12), and (
One has from Eqs. (IV8 )
IV13), the following expr ession
for the velocity vector

~~~~~~~~~

~~~~~~

r~~~~~~

~~~~~ q

-.
~-

~~~~~~~~~

--

~~~~~~~

.
~~~~~~~

~~~~

IS PAGE ~~ B~~T ~~~~~~

TO DRQ
OM ~~PI FUR~I1SU~~
4 +

v(r, t)

(V

-~~~

W ~dR~

terms of order r

~~~~~~~~~

- ---

--

~~~~~~~

29

~~~~~~~~~~~

.V .l ) w
dR
~~ xf (r4,~~~
~
0 0
+

(
rv-14)

0 n.
,
4

The first term on the righthand side of Eq. (

IV4) is zero because
the total vorticity in the infinite unlimited region is zero, i.e. Eq.(IV5)
or Eq. (1115).
Using the identity
V0 ((
rd

)
~~~

!
)

( ~~~~~~
~

~~~

(~ !
.
~~

(Iv-l5)

~~

the second term on the righthand side of Eq. (

IV14) is rewritten , with
the help of the divergence theorem, as
V!. r )
-~~ ex~~4f (
(

- .
~~

(V !

~~~~~~

W) r )dR

) r (W .n )dB
f(~. V I
o o o oI
oc

~~

The last integral in the above expression vanishes if w

r
-

11

(Iv-l6)

approaches zero as

for large r , with n> 4. This condition is certainly met , as discussed

in Chapter III. The remainder of the above expression can be rewritten

as
x( - x
~~ ~~~
l~~~(
l 4 )
y
.a 0

or
Li

(Iv-17)

)
~~ ,

where

~~~=

~~~~~~~~

-=~~

a is the total first moment of the vorticity

0

*
~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~

~~~~

~~~~~~~~~~~

-
~~~~~~

defined by

~~

~~~~~~~~ ~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~
US1i Z) TO DD,Q ~~~~~~~~~~~~~~
~OPIFUR~ ~

.1r
+

30

x ~,~ dR
0

(
tvia)

Equation (IV14) is th us expressible as

I

(i~, t)

~~~~~~~~

.
(* !

terms of order r ,
~~ ~ 4

The first term on the righthand side of Eq. (

IV18) is of order
Provided that
is not zero ,
is of order r 3 for large r.
For twodimensional problems , the function
as a Taylor series about the point
in

r r

( o~

in

lii

r
0

~~~

In

0:

(
IVl9)
r 3.

can be expressed
~lro ~r I

terms of order r~~ , ~~> 3

Following similar procedures as that given above for threedimensional

problems , one obtains for twodimensional problems:

~~~~~~~~~,

t)

(
)x ~~~~ dR
~~ ln!

+ terms of order

tt

~~~~~~~~~~~~~

~ ln ~ )~ 0dR0

n>3

(Iv 2o)

The first term on the righthand side of Eq. (

IV20) vanishes because

.4

of Eq. (
11118). The second term may be rewritten , by noting that

1
~,(r 0.~ ln ~~)

+ +
1) a
r0(w
~ .~ In ;

(V in 1) x (r
~~

and that in twodimensional problems

_
_

0 x

- -

----

~~~~ ~~~~~~~~~~~~~~~~~~~~~~~

--

~~~~
-- - -

___

~~~~~~~
- -

~~~~~
-

31

THIS PAGE IS BEST QUPIIIIT!PS~~~~~*1

~~

.
.
.
.
7R03& OOk Y F1JR.NISH~~ TODDC ~~_
-

as
~~~ ~~

in !
x ~~~~~~~ -

2
i~

f.

(
~~~

)(
~ ~ ln

)
~~

The first term on the righithand side of the above expression vanishes
for r * 0. One thus has, for twodimensional problems
(r t)

hV(

in 1

)
~~~,

Therefore, provided that

is not zero, ~

for large r.

r T~, n> 3

terms of order

(tV21)

is of order
-

The above conclusions for twodimensional and three-dimensional problems

can be combined into the statement that , provided that
is of order r~~ for large

is not zero ,

being the dimensionaliey of the

problem .
4.

Velocity Integrals in ~~~ge Bounded Regions

Using Eqs. (ILi) and (

113 ) and. vector differential identities ,
one has, for three-dimensional problems,
The divergence theorem and its corollaries then give

f(
R

.1
1
~

~ x W2v )dR

)tt
(
(
~~~

The above equation can be reexpressed as

J~ dR

~~

dR

J rx W

)- )dB
(
~~.
~ ~

5
3

iV-23)
(

~)dB

I
_

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-,

- -

1
~ ~~~~~~~~~~~~~~~~~~~~~~

s--

_
_
_
.
s
- __ _
~~~~~~~~~~~~~~~~~~~~~~~~~ .

_
_,u--

~~~~~~~~~~~~~~~~~~~~~~~~~~

- --- -

~~~~~~~~~~~~~~~
~
-~------ _ _ _ _ _ _ _ _ _ _ _ _ _ _
~ -.-

--

---

~~~~~~~~~~~~~~~~~~

.
.-

- -

.~~.

.: ~

~~~~~~~~~~

~~~

- I .

~~~
-

--

THIS PAGEIS BESTQUAIitTY PR&C~XC~~ J

r~o~ OGPY F

j. Equation

occupied by the solid body

Let R be the region

32

..~~~~~~-

ISHED TO DDO

(IV23) gives
1

B.

r x (n x v)dB

3..

r x

~~~~~~

.dR

1.

v.dR

(1v24)

in Eq. (
IV23) to be R the part of fluid region
1

Consider now R

bounded internally by the solid surface

but finite spherical surface

and externally by a barge

L. Let thia spherical surface be- designated

Let L be sufficiently large so that

encloses all the solid

B in Eq. (Iv-23) now consists of the solid

bodies present . The boundary

boundary

B and the spherical boundary B~ . With the no-slip condition ,

~
(
the contr ibution of B
Iv24). One
5 to Eq. IV23 ) is given by Eq. (

then has , noting that for

the solid where as in
J
~~~

N
j Ib

vdR

Eq.

.1

R th:normal vector
f

( IV24 )

R.
3

v~dR

is directed into

is directed into the fluid ,

ii

.1

(r xW )dR

R
.1

~~

.
,

R.
3

J1

zU~~~ dR

x(n x

)dB

(
Iv-25)

Equation (
IV25) can be reexpressed as
-~~~~

vdR

f (r xW) dR.

~~r x(n x v)dB

~~~~~~~~~~

___________________________________

~~~~~~

1
~~~1

:
,~~~~

,, I

~~~~~~~~~

T~~ PAGE IS BEST QUADft!PMCT C*~~~

FROM COPY FURNISHED TO DDO ~~~~~

~~

where

is the spherical region r ~ L and includes the fluid region

R ~ as well as all the solid regions R~ .

f
Let

L be sufficiently large so that the total vorticity outside

is negligible and the velocity on

is accurately giva~by the first

term on the righthand side of Eq. (IVl9). The integral over the surface
in Eq. (
IV26) is finite and expressible in terms of the vorticitymo.ent
integral ci as shown below .
Let

i,

+
j,

and k be a set of ri ght-banded normal unit vectors in

a Cartesian coordinate system (x, y, z), in the directions x , y, and

z respectively. The corresponding spherical coordinates (r,e .
unit vectors C
x

and
r~ e9 ,

r S in O C o s

On BLI

one has

One thus has

the first term of Eq. (

lV19) gives a velocity vector
v

x , y, and z by

r Cos e

Also , on

with

Sin eSin

are related to

r x (n x v) a .

8iiL

(2 Cos e~e
r

L and
~r

Sin9 e~ )

~r

S1n 0 4
-

~~

e
8

ci
2 .( S in O cos O Cos i
~
~
8~L

sin ecoe e sin~ j Sin 2 e

ii)

Placing the above expression into the sutface integral of Eq. (IV-26) ,
one obtains

.
~~

as the value of that integral . Noting that the first

int g al on the ri.ththsnd sid. of Eq. (

IV2~)~s
~ ~

:~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~

11 0 1

fl

---

______

-~~~~~~~

I.

1
.

vdR

~~

ci0

(
Iv27)

IV27 ) is valid for any sufficiently

It should be pointed out that Eq. (
large but finite value of L. Although Eq. (IV-26) is independent of
the size of ILL , one may not consider L

integral

5 vdR

to be infinitely large. The

is in fact indeterminate if

Rc~

scomponent of this integral over the region

a~

0. For example , the

r- L

is, according to the

IV19),
first term of Eq. (
iT

~~~~

IxO

O
~0

r~ L

(2Cos 0 Sin 8)Sin0drdl3d$

This integral gives

(c080 Cos3 0 )

1i~ )

which is indeterminate.
It is simple to show that the integrals of the x and ycoinponents
of

over the infinite region

r L are also indeterminate . Also ,

the integrals of the velocity moment r x v over the infinite region

+

r> L are indeterminate. To show that these latter integrals do not diverge,
terms of order

in Eq. (IV19) must be considered .

IV22) is valid only for threedimensional problems. For two

Eq. (
dimensional problems , one has, instead of Eq. (
IV22),

~~~~~~~~~~~~~

- -

__________________

The twodimensional version of Eq. (Iv26) is therefore

4
where

vdR

,
+

where

(cose er
211L 2

+ Sin

eee)

are unit vectors in the cylindrical coordinate system

and

(r , e)given by

3 one has r
L
-

Le

rCo.0

rSine

and

e . Thus one has

r

r x (n x v)

a
0

SinO e0

+ Co
~~~)

Placing this expression into the surface integral of Eq. (IV29) ,

-~~

one finds this integral to be

for twodimensional problems ,

~~ ~

Equation (IV29) therefore gives,

i- - -

(Iv-29)

the first term of Eq. (IV21) gives a

direction. On 5
L
velocity vector

n
~~

~~~~~~~

I r x ( n x v)dB

is a circle of radius L bounded by IL .

L
Let the flow be in the xy plane, with the xaxis selected to be

in the a

I ( r xw )dR

_____
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

_ _ _ _ __
_
_ _ _ _ _
_
_

I
j

vdR

1~~~~~~~~~
ci
2

Equation (
IV30) is independent of the size of

is finite. The integral

vdR

is indeterminate if

(
IV-30)
R~ , as long as L
~iO

because

the first term on the righthand side of Eq. (IV20) is of order r 2.

Ref. 11 gives a formula similar to Eq. (
iV30) but with the coefficient
for

4 in equation (IV30), replaced by 1.

That formula is derived

by considering a finite cylindrical voltine centered on the saxis , bounded

by two planes z

0 and

6, and using Eq. (IV23).

The derivation

of the formula neglected the contribution of the last integral in (IV-29)

(See the equation between Eqs. A3 and A4 of Ref. 11). The formula is
in error for any finite cylinders. As it turns out, however, if the cylinder
is infinite in length, then the formula in Ref. 11 is correct,

as shown in Chapter V.
5.

VelocityMoment Integrals in Large Bounded Regions.

In this Section , a formula for the velocitymoment integral is derived .

Using the equation

+

rxv

1 2
. r ~W
~~~

1+

2+

~ Vx (r v)

(IV31)

one obtains , with the help of Stokes Theorem

x

dR

Li-

2
2
~ dR 4 fr ( x~~)dB
4 fr
R
B
-

(IV-32)

-~

~~

--

~~~~~~~~~

-----

~~~~~~~~~~~~~~~~~~~

37

where B is the boundary of

the portion of the fluid region bounded i nterna ll y

Let 3. b e

by the solid boundaries

i f

-f

2
r ~~ ii
3f.

-4

L~. 0~ has from Eq. (IV-32)

and externally by

-4 f

Let R be R~

K.

fr

2(

x~n ) d B _ 4

(IBj

in Eq. (IV32), one obtains

(
~~r~ ~~ x ~ )dB

(
~~

x ~)dk +

4 fr 2~~

dR

(Iv-34 )
-

IV34) into Eq. (

IV33) and noting that the normal vector
Placing Eq. (

in these two equations are directed opposite to each other , one obtains
upon simplification

i$

r x v dR

1
~

f - r2 ~ dR
W

2(v x +n) dB
~~r ~~~~
EL

1
~

(IV 35)

include the solid regions R~ and the portion of the fluid

where R,
1~

R .
f
be a spherical region , or a circular region in two dimensions,
Let E
L
-centered on the origin with the radius L. Let L be sufficiently large

region

4,

so that the total vorticity outside EL is negligible . One has then

~~~~~~~~~~~~~~~~~~~~~

-. --- -- ~~~~
~~~~~~~~~
~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

j r~~,x ~~dB

L2 5~~~X ~ dB

L2

dR

(Iv

36)

The last integral in Eq. (

IV35) is therefore zero because of the
principle of total vorticity conservation discussed in Section 1112 .
One obtains therefore , upon noting r2 w

where

dR

(
IV-37)

is the integral of the second moment of vorticity defined by

is negligible outside

r~

2w
0d90

(
IV38)

IV37) relates velocitymoment integral to an integral

Equation (

of the second moment of vorticity . This equation is valid for both two
and threedimensional problems.

IV35) vanishes can also

The fact that the last integral in Eq. (
be shown by noting that if L is sufficiently large, then a scalar velocity

potential exists on

and is singlevalued . The boundary integral

in Eq. (IV35) can therefore be expressed as

L2

5
B

~
V~~ x n d B
+

Using Stokes theorem, this integral is expressible as

-.-

___________________________________________________________

-__ u__

- - -- .---

--

~~

I
-

~~~

S ~~~

*
-

Pr

~~~~~

, -

~~~~~~~~~~~~~~~~~~~ ! _

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

L2

~--~ .~~~~

f~~ x V ~~dR
E
L

and is zero because the incegrand, being the curl of the gradient of a
function, is zero.

It is not difficult to show that the terms of the asymptotic expression

f or

IV21) are all expressible as gradients of a scaler

v , Eqs. (IV19) and (

function. For example, for twodimensional flows, the firs c term on the

IV21) and the next term, being resp .ctively of

righthand aide of Eq. (
orders r

2 and

r 3 , are important. The first term is already written

in the form of a gradient of a scalar function , namely

~~~~

(
~~~

.
Vln~ .c10)

The next term can be shown to be expressible as

cos (:90_29)]
2 [
~~- x ~~~oro

~~~~

~~~~~

~~~

x~

dR0

09
woro2{
~ ~~~o ~~~~

_ _ _ _ _ _ _ _

___________

_
__
__
__
__
_
__
__
_
__
_
_
_

I~

.
~~p

~~~~~~~~~~

1 T

I:ziiI
~ i
~~
40

V. AERODYNAN IC FORCE AND MOMENT

In this Chapter , the theorems and formulae presented in Chapters III
and IV are utilized to derive general formulas relating the aerodynamic
forces and moments to the vorticitymoment integrals.
1.

Aerodynamic Force

Consider the control volume

the fluid occupying the region

EL

bounded externally by

R and the solid bodies occupying the

f

regions R . The momentum theorem gives

1
-

8L and containing

vcn
-j
~ ~ f ~~

EL

~~~~.
~ )dB
j

(v-i)

where F

t is the total force acting on the system within EL and ~

is either the fluid density p or the solid density P , depending on
~
the particular region of interest.
On account of the asymptotic behavior of v the last integral in

Eq. (vi) is negligible for sufficiently large values of L. The integral

EL

over

can be written as sums of integrals as follows:

P ~~

The total force

force

~~~~R- j 1
~

~~~~~

+
~~ ch1(
f

~~

acting on the system in

EL

consists of the

(V3)

~~~~~

I
-

TL,~ .

_________________

(v-z)

acting on the boundary 8 and the external force

L
+

~~

~~~~~

-f--

.-

~~

~~

i .~~
~ ~~~
- ---- -- ~~~~~~~~~~~~~~~
~~~~~~~~~~~~
~
~ ~

~
.- ~-

Pr

- ---- -- -

~~ - iLL
~~~~

~~~~

~~~~~~~~~~~~~~

pui p

p~~~~~~

41

is the force exerted f r oht outside the system

he external force 7e
T

and acts on the solid bodies . The total force acting on the solid bodies
is F + F , where F is the aerodynamic force exerted by the fluid on
5
the solid bodi~ s. Newton s second law of motion gives
+
F

~ ,1

ut
~~

f P.v.dR

(V4)

Placing Eqs. (V3) and (V4) into Eq. (vI) one obtains
+

d
FL P d fvda + 3j ~~~~
fPv 3dR
3
~~ ~~
+

(V 5)

EL

For sufficiently large L, the shear stress is negligible on

4.

and F~

is expressible in terms of p

as given by Eq. (11121). That

15

F
L

p 1p ~ dB

The asymptotic behavior of v

.

I f(t )J
i_
+
~
~
~~ 3
.1[

shows that the term

(v 6)

does not contribute

to the last integral in Eq. (V6). The function f(t), being independent
of position, also does not contribute to Eq. tvb ). One thus has
-:

-p 5 ~~ ~ dB -p
~~
1

f ;dR
EL

N
+

J~~~ dR

~~ Rj

IV1 9), one obtains for threedimensional

Using Eqs. (11120) and (
fbi.

~~~~~~~~

~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-T _ _

~~~.

_ _ _ _ _

~~~~~~~~~~~~

~~~~~~~~~

-
~~~~~

(v-i)

~~~ ~ - ---

42

ndl

)n
.
~~ dB

(
~~~ ~~

EL

Exp ressing the last integral in a spherical coordinate system with

directed in the k direction, one obtains

~d3

2
Cos e Site d~d0

~~

4
Using Eqs. (IV27) and the above equation, one obtains from Eq. (V-7)
the folloving expression for aerodynamic force in threedimensional flows.
N

1 dci0 +
~P ~~~~~~

~~~~

~~ fv 3di

(V 8)

IV21), one obtains

Pot twodimensional flows, usi ng Eqs. (11120) and (
+

~~~

ndB

21t

Cos

~~~

(v-9)

~~

Using Eq. (
IV30) and the above equation, one obtains from Eq. (V-i)
the following expression for the aerodynamic force in twodimensional
flows:

- i- - .
_

~~~~~

- -
_

+ P

~~~~~~~~~~~~~~~~

..i

__4._;_~_

(vb )

f . dR

- -----

~~

-- -

~~~~~~~

~~~~~~~~

~~~~~~~~~~~~~~~~~~~
~~~
.
t

.*

- ------.

~~~~~~~

F !

--

Moment of Aerodynamic Force

2.

The theorems of moment of momentum gives

4

x F~

~~~~~

EL

4.

~~r x v)dR +0

4 -4

x v)(v.n)dB

is a position vector describing the line of action of the total

where I~

force F .
t
The last integral in Eq. (vli )is negligible for sufficiently large
4.

L. The total moment consists of the moment

EL

x F

e
e
x L is negligible. One thus has
~

and the externally applied moment

on 8L Consequently the moment

+

1 xP
a

1
~

1e X F e * P ~~

or

4.

I ( r x v )dR
j ~~

~~

dt

(vli)

EL

x FL acting on the boundary

The shear stress is negligible

~~~ x v ~dR -4

N
J r x v jda

~~~~~~~

N
+

.(r
jP

~~

~~~~

v )dR
3

(vl2)

The total moment acting on the solid bodies consists of the externally
applied moment and the moment exerted by the fluid on the solid bodies.

This total momentum is equal to the t ime rate of change of angular momentum

of the solid bodies. This rate of change of angular momentum is given

One therefore obtains the following expressions

by the last term of Eq. (V- lI) .

for the first moment of aerodynamic force in both twoand threedimensional

~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~ ---~~~

T :I
~~~ ~

_ _
_ _
_ _

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

L~~~~ 4

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
_ _ _ _ _ _

~~~~

~~

p-

--~~~~~

__________________________

- - - -~~~ - - - -~~~~~~~~~~~~~~~~~~~~~~

problems , by using Eq. (IV-37)

N

~~

~~~~~~~~

+ P

~~

f1 x

(V43)

.dR

where

1 is a position vector describing the line of action of the aerodynamic

force

F.

Equations (V5) and (VlO)express the aerodynamic force exerted by

a fluid on solid bodies imersed in the fluid as integrals of the first

moment of the vorticity vector plus an inertia term. Equation (Vi3)
expresses the first moment of the aerodynamic force as an integral of
a second moment of the vorticity vector plus a moment of inertia term.
Higher moments of the aerodynamic force can be related to intgrils of
higher moments of the vorticity vector .
3.

Sumeary.

The preceeding results form a theory for aerodynamic forces and moments
which encompasses much of the previous aerodynamic theories. As discussed
in Chapter U of this report, this theory deals with unsteady motions
of a fluid which is at rest at certain initial instant of time . A steady

flow , if it exists , is considered to be approached asymptotically at large

time levels. It has been shown that under this circumstance the effective
vortical regions surrounding and trailing the solid bodies are of finite

extent at any finite time level.

The theory comprises of the following

three statements:
a.

The combined total vorticity of the fluid and the solid bodies

is zero.

b.

The aerodynamic force acting on the solid bodies is expressible

as the time derivative of an integral of the first moment of the vorticity

plus a inertia term.

I

I
~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-

The moment of aerodynamic force acting on the eolid bodies is

c.

expressible in terms of the time derivative of an integral of a second

moment of the vorticity plus an moment of inertia term.

These three statements are expressible mathematically as follows:
fwdR

(1115) , (III

where 1 , ii the infinite unlimited region jointly occupied by the fluid and
all the solid bodies.

~~
where

T ~t

~~~

x~~ m

j~ l

~~~

dR

R.

(v-8),(V-9)

is the dimensionality of the problem.

1 x

~E

.1 r 2 ~ dR

p ~
3l

~ 5

~x

(V 43)

dR

~6

~~~~

It I:t .

_
_

~~~~~

V i

~~~~

46

VI . REFERENCES
1.
2.

Hess, J. and Smith , A. 14. 0., Calculation of Potential Flows about Arbitra
Bodies; Progress in Aeronautical Sciences, Vol . 8, Pergamon Press , N. Y . ,
1967, pp. 1138.
Sears .V . k., Unsteady Motion of Airfoil with BoundaryLayer Separation,

AIAA Journal, Vol . 14, No. 2, 1976 , pp. 216220.

3.

Von Karman , T
h., and Sears V. R., Airfoil Theory for NonUniform Motion,
J. of the Aero. Sciences, Vol . 5, No. 10, 1938, pp. 379390.

4.

Ceising , J. P. , Nonlinear TwoDimensional Unsteady Potential Flow with

Lift , Journal of Aircraft, Vol . 5 , No. 2, 1968, pp. 135143.

5.

Vu , J. C., Prospects for the Numerical Solution of General Viscous Flow

Problems, Proceeding of the Viscous Flow Symposium , Lockheed Georgia
Company LG7YEROO44, 1976, pp. 39104 .

6.

Vu, J. C. and Thompson, J. F., Numerical Solutions of TimeDependent

Incompressible NavierStokes Equations using an IntegroDifferential
Formulation, , Computers and Fluids , Vol . 1, No. 2, 1973, pp. 197215.

7.

Vu, J. C., Numerical Boundary Conditions for Viscous Flow Problems ,

AIAA Jou rn al , Vol . 14, No. 8 , 1976 , pp. 10421049 .

8.

Vu, J. C., Sampath, S., and Sankar, N. L., A Numerical Study of Viscous
Flows around Airfoils, Proceedings of AGARD Fluid Dynamics Panel Sy.posiu
On Unsteady Aerodynamics, 1977, in print .

9.

Lamb , H., Hydrodynamics, Dover, N. Y., 1945.

10.

Batchebor , G. K., An Introduction to Fluid Dynamics, Cambridge Univerait

Press , 1967.

11.

Phillips , 0.14., The Intensity of Aeolian Tones , J. Fluid Mech .,

Vol. 1, 1956 , pp. 607624 .

12.

Mo r eau , J. ~J., Bilan Dynamique dun Ecoulement Rotationnel,

13.

Truesdell C., The Kinematics of Vorticity, Indiana University

Press, 1954.

14.

Vu. J.C., and Wahbah , 14. 14., Numerical Solution of Viscous Flow
Equations using Integral Representations , Proceedings of the Fifth
International Conference on Numerical Methods in Fluid Dynamics
Lecture Series in Physics, Vol . 59, SpringerVerlag, 1976 , pp. 448
453.

J. Math. Puree App i., Vol . 9, No. 31, pp. 355375 and Vol . 9, No.
32 , pp. 178 , 1953.

15.

Vu , J.C., and Sugavanaa, A Method for the Numerical Solution of

Turbulent Flow Problems, AIAA Paper No. 77649 , 1977 , 13 pages.

L
i

~~~~~

.- .- -~~

_________________________

~ ~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
V

~~~~~ .--- ~~- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-~~

k fl ~~i
~

______

..
~~~~~~~

I ~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

47

-.

16.

Vu, J.C., Integral Representations of Field Variables .for the

FiniteElement Solution of Viscous Flow Problems , Proceedings
of the 1974 International Conference on Finite Element Methods
in Engineering, Clarendon Press, 1974.
-

17.

Vu, J.C ., Finite Element Solution of Flow Problems using Integral

Representation, Proceedings of Second International Symposium
on Finite Element Methods in Flow Problems, International Centre
for Computer Aided Design, Conference Series No. 2176, June, 1976.

18.

So erfeld , A., Mechanics of Deformable Bodies Academic Press,

1950.

19.

Carslaw, R.S. and Jaegen , J.C., Conduction of Heat in Solids,

Oxford University Press, 1959.

20.

Vu. J.C., Integral Representation of the Kinetics of Viscous

Flow Problems, Georgia Institute of Technology.Notes, 1974.

21.

Lagerstrom, P.A., Laminar Flow Theory, Section B of Theory of

Laminar Flow , F.K. Moore, Editor, Princeton University Press,
1964.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~

- -- - - - -

r ~~~T

DISTRIBUTION LIST FOR UNCLASSIFIED

REP ORTS AND REPRINT S ISSUED UNDER
CON RACT
aO/
TASK I I R O W- ,?,.2~~
TECHNICAL

~-Zc ~C-- O~,Z~~

Al]. addresses receive one copy unless otherwise specified

Technical Library
Building 313
Ballistic Research Laboratories
Aberdeen Proving Grou~td, MD 21005

Office of Na va l Research
Attn : Code 1U21P (ONRL )
8)0 N. Quincy Street
Arlington , VA 222 17
6 copies

Dr. P. D. Benne tt
External Ballistic Laboratory
Ballistic Research Laboratories
Aberdeen Proving Ground , MD 2W03

Dr. J. L. Potter
Deputy Director, Technology
~on Karma n Gas Dynamics Facility
Arnold Air Force Station, T1~ 37389

Mr. C. C. Hudson
Sandia Corporation
Sandia Base
Albuquerque , NM 81115

f
*

Professor J. C. Vu
Georgia Institute of Technology
School of Aerospace Engineering
Atlanta , GA 30332

Professor P. J. Roache
Ecodynamice Research
Associates, Inc.
P. 0. Box 8172
A1buquerque,-.!
~( 87108

Library
Aerojet-General Corporation
~352 North Irwindale Avenue
Azusa, CA 917(i2

Defense Documentation Center

Cameron Station , Building 5
Alexandria , VA 22314

Dr. K. C. Wang
Martin Marietta Corporation
Martin Marietta Laboratories
1450 South Rolling Road
Baltimore, MD 21227

Scientific and Technicci

Information Facility
P. 0. Box 8757
Baltimore/Washington International
Airport , MD 2124 0

NASA

Dr. J. D. Shreve, Jr.

Sandia Corporation
Sandia Base
Al buquerque , NM 81115

12 copies

Library
Navel Academy
Annapolis , MD 2l4u2

Dr. S. A. Berger
Uni~ersity of California
Department of Mechanical Engineering
Berkeley, CA .94720

Director , Tactical Technology Office

Defense Advanced Research Projects
Agency
14.0 Wilson Boulevard
Arlington, VA 222 1)

Professor A. J. Chorin
University of California
Department of Mathema t ics
Berkeley, CA 9472J

Office of Naval Research

Attn: Code 211
8~)0N. Quincy Street
Arlington, VA 22217

Professor K. Rolt
University of California
Department of Mechanical Engineering
Berkeley, CA 94720

Office of Na ,al Research

AtCn : Code 438
8~ J N. Quincy Street
Arlington, VA 22217

-~~~~~ --

---

-.~

Dr. H. a. Chaplin
Code 1600
David V. Taylor Naval Ship Research
and Development Center
Bethesda , MD 20084

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~ ~~.1
~~

~~

~~~~~~~

~~~~~~~

--

- -

~~~~~~~~~

- - -

- -~~~

.
-~~~

-- - ,---
~~~
~

-~

Pate 2
Dr. Hans Lugt
Code i84
David V. Taylor Naval Ship Research
and Development Center
Bethesda , MD 20084
-

Dr. Yrancois Prenkie)

.

Code 1802.2
Dtvid V. Taylor.Naval Ship Researc h
and Development Center
Bethesda , MD 20084
Dr. G. H. lager
Department of Aerospace Engineering
Virginia Polytechnic Institute and
State University
Blscksburg, VA 24061

Professor A. H. Neyfeh
Department of Engineering Science
Virginia Polytechnic Institute end
State University
Blacksburg, VA 24061
Mr . A. Rubel
Research Depar tment
Gr~~~an Aerospace Corporation
Bethpage, NY U714

:-

Oommsnding O fficer
OftLce of Naval Research Branch Office

666 Summer Street, Bldg.114, Section D

-

Boston , MA 02210

Dr. G. Hall
State University of Nev York at Buffalo
Faculty of Engineering and Applied
Sciences
Fluid and Thermal Sciences Laboratory
Buffalo , NY 14214
-

Dr. B. J. Vida,l
CALSPAN Corporation
Aerodynamics Research Department
P. 0. Box 235
Buf falo , NY 14221

Professor B. F. Probstein
Department of Mechanical Engineering
Massachusetts Institute of Technology
Cambridge,MA 02139

Comeandiag Officer
Office of Naval Research Branch Office

Code 753

Naval Weapons Center

China Lake, CA 93555
Mr..3. Marshall

Code

4063

Naval Weapons Center

Chi na Lake , CA 93555

Professor R. T. Davis
Department of Aerospace Engineering
University of Cincinnati
Cincinnati, OH 45221
Library MS 603
NASA Levis Research Center
21000 Brookpark Road
Cleveland , OH 44135
Dr J D. Anderson., Jr.
Chairmen , Department of Aerospace
Engineering
College of Engineering
University of Maryland
College Park, MD 20742
Professor W. L. Melnik
Department of Aerospace Engineering
University of Maryland
College Park , MD 20742
-

Professor 0. Burggraf

Department of Aeronautical end

Astronautical Engineering
Ohio State University
1314 Kinnear Road
Columbus , OH 43212

Technical Library
Java]. Surface Weapons Center
Dahlgren Laboratory
Dahigren, VA 22448
Dr. F. Moore
Naval Surface Weapons Center
Dahlgren Laboratory
Ds.blgren, VA 22448
Technical Library 251131

LTJ Aerospace Corporation

P. 0. Box 5907
Dallas, TX 75222

=
~~~~~~~~~~~~~~~~~~

~~~~

:~T:

_________________________________________

Page 3
Library, United Aircraft Corporation

Professor B. A. Caughey
Sibl ey School of Mechanical and
Aerospace Engineering
Cornell University
Ithaca , NY 14850

Research Laboratories

Silver Lane
East Hartford, CT 06108
-

Technical Library
AVCO-Everett Research Laboratory
2385 Revere Beac~h Parkway
Everett, MA 021149

Professor E. L. Reeler
Sibley School of Mechanical and
Aerospace Engineering
Cornell University
Ithaca , NY 14850

Professor 0. Moretti
Polytechnic Institute of New York
Long Island Center
Department of Aerospace Engineering
and Applied Mechanics
Route 110
?armingdale, NY 11735

Professor S. F. Shen
Sibley School of Mechanical and
Aerospace Engineering
Ithaca, NY 14850
Library

Professor S. 0.Rubin
Polytechnic Institute of New York
Long Island Center
Department of Aerospace Engineering
and Applied Mechanics
Route 110
Permingd.ale, NY 11735

Midwest Research Institute

1425 Volker Boulevard

Kansas City, MO 641i0
.

Dr. V. R. Briley
Scientific Research Associates, Inc .
P. 0. Box 1e98
Glastonbury, CT 06033

Dr. B. H. Murman
Flow Research, luc.
P. 0. Box o4o
~8031
Kent, WA 9

Professor P. Gordon
Calumet Campus

Dr. S. A. Orezag
Cambridge Hydrodynamics, Inc.
54 Baskin Road
Lexington, MA 02173

Department o f Mathemat ics

~

Purdue University
H~~~ond, IN 46323

MS 185)
Library (
NASA Langley Research Center
Langley Station
Hampton , VA 23665

Dr. P. Bradshaw
I~nperial College of Science end
Technology
Department of Aeronautics
Prince Consort Road
London SW? 2BY, England

Professor A. Chapmann
Chairman , Mechanical Engineering
Department
William H. Rice Institute
Box 1892
Houston , TX 77001

Professor T. Cebeci
California State University,
LongBeach
Mechanical Engineering Department
Long Beach , CA 90840

Technical Library
Naval Ordnance Station
Ind ian Head , MD 20640
1
~~~~

&~

- ~ i __ ~~~ .a

Dr . M. H. Hafez
Flow Research , Inc.
P. 0. Box 5040
Kent, WA 98031

Mr. 3. L. Hess
Douglas Aircraft Company
3855 Lekevood Boulev6rd
Long Beach , CA 90305

~~~~~~~~~~~~~~~~~~~~~~~~~~~~

- .
- ~~

__
_
_

-----.-.- ~ .~~~-.

~~

~~

C~~~~~~~~__

.
, - -,
~- ,~~~~
I
V

-~~

. -~~ ~
~~
~~~
~~~

P..-

Page 14

Dr H. K. Cheng

Dr S S 8tahara
Nielsen Engineering & Research, Inc .
5lOClyde Avenue
Mountain View, CA 914043

University of Southern California,

University Park
Department of Aerospace Engineering
Los Angeles , CA 90007

Engineering Societies Library

345 East 47th Street

New York , NY 10017

Professor J. D. Cole
Mechanics and Structures Department
School of Engineering and Applied
Science
University of California
Los Angeles, CA 90024

Professor A. Jsmeeon
New York University
Courant Institute of Mat hematical
Sciences
251 Mercer Street
New York , NY 10012

Engineering Library
University of Southern California
Box 77929
Los Angeles , CA 90007

-t

Professor G. Miller
Department of Applied Science
New York University
26 36 Stuyvesant Street
New York, NY 10003

Dr. C. -M. Ho
Department of Aerospace Engineering
University of Southern California,
University Park
Los Angeles, CA 90007

Office of Naval Research

New York Area Office
715 Broadva~ 5th Floor
New York, NY 10003

Dr. T. D. Taylor
The Aerospace Corporation
P. 0. Box 92957
Los Angeles, CA 90009

Dr. A. VaglioLaurin
Department of Applied Science
2636 Stuyvesant Street
New York University
New York , NY 10003
-

C~~~anding.Officer
Naval Ordnance Station
Louisville, KY 140214
Mr . B. H. Little, Jr.
Lockheed-Georgia Company
Department 7274, Zone 369
Marietta, GA 30061

Professor B. B. Ranch
Ph.D. Program in Mathematics
The Graduate School and University
Center of the City University of
New York
33 West Ie2nd Street
New York , NY 10036

Professor E. B. 0.!ckert
University of Minnesota
2141 Mechanical Engineering Building
Minneapolis, MN 55455

Librarian, Aeronautical Library

National Research Council
MOntreal Road
Ottawa 7 Canada

Library
Naval Postgraduate School
Mon terey , CA 93940

Lockheed Missiles and Space Company

Technical Information Center
3251 Hanover Street
Palo Alto , CA 94304

Supersonic-Gas Dynamics Research

Laboratory
Department of Mechanical Engineering
McGill University
Montreal 12, Quebec, Canada

--

~~~~~~~~~~~~

----

~~

Page 5
Co~ nanding Officer
Office of Naval Research Branch Office
1030 Esst Green Street
Pasadena , CA 91106

Professor L. Sirovich
Division of Applied Mathematics
Brown Un ivers ity
Providence, RI 02912

California Institute of Technology

Engineering Division
Pasadena , CA 91109

Dr. P. K. Del CR1/2178)

TRW Systems Group, Inc.
One Space Perk
Redondo Beach , CA 90278

-.

Library

Jet Propulsion Laboratory

148oo Oak Grove Drive
Pasadena, CA 91103

Redstone Scientific Information Cerr

Chtef , Document Section
Army Missile Ccminand
Redstone Arsenal , AL 35809

Professor H. Liepnann
Department of Aeronautics
California Institute of Technology
Pasadena , CA 91109
Mr. L. I. Chasen , MGR-MSD Lib.
General Electric Company
Missile and Space Division
P. 0. Box 8555
Philadelphia, PA 19101

Library and Information Services

General Dynamics-CONVAIR
P. 0. Box 1128
San Diego, CA 92112

1402 South 36th Street

Phoen ix ,.AZ 85010

Technical Library
Naval M
issile Center
Point Mugu , CA 93042

Dr. B. Magnus

Professor S. Bogdonoff
Gas Dynamics Laboratory
Department of Aerospace end
Mechanical Sciences
Princeton University
Princeton, NJ 08540

Office of Naval Research

Sen Francisco Area Office
One Halildie Plaza , Suite 601
Sen Francisco, CA 94].
02

Mechanical Sciences

Princeton University

Princeton , NJ 08540

The RAND Corporation

1700 Main Street
Santa Monica , CA 90401

Dr. J. N . Yates
Aeronautical Research Associates
of Princeton , Inc .
50 Washington Road
Princeton, NJ 08540

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-~~ .-- ~~~

Mr. T. Brundage
Defense Advanced Research Proj ects

Agency

Professor s. i. Cl~ieng
Depar tment of Aerospace and

~~

General Dynamics-CONVAIR
Kearny Mesa Plant
P. 0. Box 80847
San Diego , CA 92138

Research and Development Field Unit

APO 1146,Box 271
San Francisco , CA 962146

U.S. Army Research Office

P. 0. BOx 1221 1
Research Triangle , NC 27709
Editor, Applied Mechanics Review
Southwest Research Institute
8500 Culebra Road
Sen Antonio , TX 78228

Mr. P. Dodge
Aires.arch Manufacturing Company
of ~~ 1zona
Division of Garrett Corpor ation

~~~~~~~~~~~~~~

Page 6
Dr. B. J. Hakkinen
McDonnell Douglas Corporation
Department 222
P. 0. Box 516
St. Louis,MD 63166

Dr. P. B. Rubbert
Boeing Aerospace Company
Boeing Military Airplane Development
Organization
P. 0. Box 3707
Seattle,WA 981214

H. Yosbihara
Boeing Aerospace Company
P. 0 Box3999
Mail Stop 141-18
Seattle , WA 98124
Dr

Librarian
Naval Surface Weapons Center
White oa~ lAboratory
Silver Spring, MD 20910

Library
Institute of Aerospace Btudiea
University of Toronto
Toronto 5, Canada
-

Dr. J. M. Solomon
Naval Surface Weapons Center
White Oak Laboratory
Silver Spring, MD 20910
Professor J. H. Ferziger
Department of Mechanical Engineering
Stanford University
Stanford, CA 94305

Professor K. Karamc~het i
Department of Aeronautics and
Astronautics
Stanford University
Stanford, CA 914305
-

Professor N . van Dyke

Department of Aeronautics and
Astronautics
Stanford University
Stanford, CA 94305

Professor V . B. Sears
Aerospace end Mechanical Engineering
University of Arizona
Tucson ,AZ 85721
Professor A. R. Seebass
Department of Aerospace and Mechanical
Engineering
University of Arizona
Tucson, AZ 85721
Dr. K. T. Yen
Code 3015
Naval Air Development Center
Warminster , PA 189714

Dr . N. Majinuth
Rockwell International Science Center
10149 Camino Dos Rios
P. 0. Box 1085
Thousand Oaks, CA 91360

Mr. R. Feldhuhn
Naval Surface Weapons Center
White Oak Laboratory
Silver Spring, MD 20910
*

Dr. B. P. Heiniech
Honeywell, tue.
Systems and Research Division
Aerospace Defense Group
23145 Walnut Street
St. Paul, MN 55113

Air Force Office of Scientific Rese.rc

(sR z 4)
Building 11410, Bofl ing Afl
Washington, DC 20332

Professor 0. Bunemarm
Institute for Plasma Research
Stanford Univereity
Stanford, CA 94305

Chief of Research and Development

Office of Chief of Staff
Department of the Army
Washington, DC 20310

Engineering Library
cD nne11 Do11as Co~~orat ion

Library of Co
eec
Science and Tec)mo1o~ Division
Washington, DC 20540

P. 0. Box 516
Bt. Louis, MO

63166

ii
-

~~~~~~~~~~~~~~~

- --

~~~~~~~~~~

Page~~

~~

Code RE)
Director of Research (
National Aeronautics and Space
Administration
600 Independence Avenue, SW
Washington, DC 20546

Chief o~ Aerodynamics
AVCO Corporation
Missile Systems Division
201 Lowell Street
Wilmington,MA 01887

Library
National Bureau of Standards
Waahington,.DC 20234

Reeearch Library
AVCO Corporation
Missile Systems Division
20].Lowell Street
Wilmington , MA 0188?

National Science Foundation

Engiueering Divis ion
1800 0 Street , NW
Washington, DC 20550

APAPL (APRC )
AR
Wright Patterson, APE , OH

Mr. U. Koven
AIR 03E
Naval Air Systems Coemand
Washington, DC 20361

Dr. Donald J. Harney

AFFDL/FX
Wright Patterson APE , OR

145433

14511 33

Mr. B. Sievert
AIR 320D
Naval Air Systems Co~~~nd
Washington , DC 20361
:

Technical Library Division

AIR 6OIe

Naval Air Systems Cc~ nand

Washington , DC 20361
Code ~627
Naval Research Laboratory
Washington, DC 20375
-

SEA 03512

Naval Sea Systems Command

Washington, DC 20362

..

SEA 0903
Naval Sea Systems Cr~ ~nd
Washington, DC 20362

-1

Dr. A. L. Slafkosky
Scientific Advisor
Comsandsut of the Marine Corps
( Code AX)
Washington, DC 20380

Director
Wespone System s Evaluation Group
Washington, DC 20305

_____________

~~

- _~ ______ ~ _*~~ _

. -~~~~~~~-~~~~~ -~

~~~~~~~~~ -

--

- ~~

- - - - -

- -

~~~~~~~~~~~~~~~~~~~~~~~~~

..

~~~~~

- -

q
~~~~

i ~~~~~ ~~

~~