World Applied Sciences Journal 16 (Special Issue of Applied Math): 82-92, 2012

ISSN 1818-4952
IDOSI Publications, 2012

Corresponding Author: D.D.Ganji, Faculty of Mechanical Engineering, Babol University of Technology, P.O. Box: 484,
Babol, Islamic Republic of Iran
82
Two-dimensional Viscous Flow through Permeable Walls
with Expanding or Contracting Gaps

1
D.D. Ganji,
1
M. Sheikholeslami,
2
H.R. Ashorynejad and
3
F. Shakeri

1
Faculty of Mechanical Engineering, Babol University of Technology, Babol, I.R. Iran
2
Faculty of Mechanical Engineering, University of Guilan, Rasht, Islamic Republic of Iran
3
Faculty of Mechanical Engineering, Shahid Bahonar University of Kerman, Kerman, I.R. Iran

Abstract: In this study, the problem of laminar, isothermal, incompressible and viscous flow in a
rectangular domain bounded by two moving porous walls, which enable the fluid to enter or exit during
successive expansions or contractions, is solved using a series of based analytical techniques named
Homotopy Perturbation Method (HPM), Differential Transformation Method (DTM) and Optimal
Homotopy Analysis Method (OHAM). The concepts of these methods are briefly introduced and their
application for this problem is studied. These methods are very useful and applicable for solving nonlinear
problems. Then, the results are compared with numerical results and the validity of these methods are
shown. Comparison between obtained results shows that HPM is more accurate and acceptable than two
other methods. After this verification, we analyze the effects of some physical applicable parameters on
velocity profile, pressure drop and shear stress to show the efficiency of HPM for this type of problems.
Graphical results are presented to investigate the influence of the non-dimensional wall dilation rate () and
permeation Reynolds number (Re) on the velocity, normal pressure distribution and wall shear stress. The
obtained results, in comparison with the numerical solution results, demonstrate remarkable accuracy. The
present problem for slowly expanding or contracting walls with weak permeability is a simple model for
the transport of biological fluids through contracting or expanding vessels.

Key words: Permeation reynolds number wall dilation rate Homotopy Perturbation Method (HPM)
Differential Transformation Method (DTM) Optimal Homotopy Analysis Method (OHAM)

INTRODUCTION

Studies of fluid transport in biological organisms
often concern the flow of a particular fluid inside an
expanding or contracting vessel with permeable walls.
For a valve vessel exhibiting deformable boundaries,
alternating wall contractions produce the effect of a
physiological pump.
The flow behavior inside the lymphatic exhibits a
similar character. In such models, circulation is induced
by successive contractions of t wo thin sheets that cause
the downstream convection of the sandwiched fluid.
Seepage across permeable walls is clearly important to
the mass transfer between blood, air and tissue [1].
Therefore, a substantial amount of research work has
been invested in the study of the flow in a rectangular
domain bounded by two moving porous walls, which
enable the fluid to enter or exit during successive
expansions or contractions. Dauenhauer and Majdalani
[2] studied the unsteady flow in semi -infinite expanding
channels with wall injection. They are characterized by
two non-dimensional parameters, the expansion ratio of
the wall and the cross-flow Reynolds number.
Majdalani and Zhou [3] studied moderate to large
injection and suction driven channel flows with
expanding or contracting walls. Using perturbations in
cross-flow Reynolds number Re, the resulting equation
is solved both numerically and analytically.
Boutros et al. [4] studied the solution of the
Navier-Stokes equations which described the unsteady
incompressible laminar flow in a semi-infinite porous
circular pipe with injection or suction
Through the pipe wall whose radius varies with
time.The resulting fourth-order nonlinear differential
equation is then solved using small-parameter
perturbations.
Most scientific problems and phenomena are
inherently in form of nonlinearity. Except a limited
number of these problems, most of them do not have
exact solution. Therefore, these nonlinear equations
should be solved using other methods. Therefore, many
different methods have recently introduced some ways
to obtain analytical solution for these nonlinear
problems, such as the homotopy perturbation method,
the differential transformation method and optimal
homotopy analysis method.
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
83



Fig. 1: Two-dimensional domain with expanding or
contracting porous walls

The Homotopy Perturbation Method (HPM) is one
of the well-known methods to solve nonlinear equations
that are established by He [5, 6]. This method has been
used by some authors in [7-9]. It was shown by many
authors that this method provides improvements over
existing numerical techniques.
One of the other semi-exact methods which does
not need small parameters is DTM, first proposed by
Zhou [10], who solved linear and nonlinear problems in
electrical circuit problems. Chen and Ho [11] developed
this method for partial differential equations and Ayaz
[12] applied it to the system of differential equations.
This method has already been applied successfully to
solve some problems [13-15].
One of the newest analytical methods is the
Optimal Homotopy Asymptotic Method (OHAM)
which is not dependent upon a small parameter.
Recently, a few authors used this method to solve
nonlinear equations arising in engineering problems
[16, 17].
Homotopy Analysis Method, Homotopy
Perturbation Method and Differential Transformation
Method are proposed and used in this study.

Flow analysis and mathematical formulation:
Consider the laminar, isothermal and incompressible
flow in a rectangular domain bounded by two
permeable surfaces that enable the fluid to enter or exit
during successive expansions or contractions. A
schematic diagram of the problem is shown in Fig. 1.
The walls expand or contract uniformly at a time-
dependent rate a
*
. At the wall, it is assumed that the
fluid inflow velocity V
w
is independent of position. The
equations of continuity and motion for the unsteady
flow are given as follows:


* *
* *
u v
0
x y

+

(2.1)


* * * * 2 * 2 *
* *
* * * *2 *2
u u u 1 p u u
u v [ ]
t x y x x y

+ + + +


(2.2)


* * * * 2 * 2 *
* *
* * * *2 *2
v v v 1 p v v
u v [ ]
t x y y x y

+ + + +


(2.3)

In the above equations u
*
and v
*
indicate the
velocity components in x and y directions, p
*
denotes
the dimensional pressure, , and t are the density,
kinematic viscosity and time, respectively. The
boundary conditions will be:


* * *
w
*
* *
*
* *
a
y a(t): u 0, v V
c
u
y 0 : 0, v 0
y
x 0: u 0

(2.4)
Where
w
a
c
V

s the wall presence or injection/suction coefficient, that
is a measure of wall permeability. The stream function
and mean flow vorticity can be introduced by putting:


* * * *
* * *
* * * *
* * * 2 * 2 *
* *
* * *2 *2
v u
u , v ,
y x x y
u v [ ]
t x y x y




+ + +

(2.5)

Due to mass conservation, a similar solution can be
developed with respect to x
*
[18] Starting with:


* * * *
y * *
2
* * *
* *
y
v x f
v x f (y,t)
, u
a a
vf (y,t) y f
v , y , f
a a y

(2.6)

Substitution Eq. (2.6) into Eq. (2.5) yields:

* * * * * * * *
* * * * * *
y t y x y y y y y
u u u v u vu + + (2.7)

In order to solve Eq. (2.7), one uses the chain rule
to obtain:


* * * * * * * 2 1 *
yyyy yyy yy yyy y yy yyt
f (yf 3f ) f f f f a f 0

+ + + (2.8)

With the following boundary conditions:


* *
yy
* *
y
a t y 0 : f 0,f 0
a t y 1: f Re,f 0


(2.9)

Where
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
84

aa
(t)

is the non-dimensional wall dilation rate defined
positive for expansion and negative for contraction.
Furthermore,

w
aV
Re

is the permeation Reynolds number defined positive for
injection and negative for suction through the walls.
Eqs. (2.6), (2.8) and (2.9) can be normalized by putting:


* * * *
u v f
, u , v , f
aa a a Re

(2.10)

And so:


'
xf xf f
, u , v , c
c c c Re

(2.11)


IV ''' '' ''' ' ''
f (yf 3f ) Ref f Ref f 0 + + +

(2.12)

The boundary conditions (2.9) will be:


''
'
y 0 : f 0,f 0
y 1: f 1,f 0


(2.13)

The resulting Eq. (2.12) is the classic Berman's
formula [19], with = 0 (channel with stationary
walls).
After the flow field is found, the normal pressure
gradient can be obtained by substituting the velocity
components into Eqs. (1.1)_(1.3). Hence it is:


1 1
y
*
2
w
p [Re f ff Re (f yf ) ]
p
p
V

+ + +

(2.14)

We can determine the normal pressure distribution,
if we integrate Eq. (2.14). Let p
c
be the centerline
pressure, hence:


c
p(y) y
1 1
p 0
dp [Re f ff Re (f yf)]

+ + +

(2.15)

Then using
2
ff (f ) / 2

and (f yf ) (yf) + , the
resulting normal pressure drop will be:


2
1 1 1
n
f
p Re f (0) [Re f Re yf]
2

+ + (2.16)

Another important quantity is the shear stress. The
shear stress can be determined from Newton's law for
viscosity:

* *
2 * *
* * *
3 x y
v x f
(v u )
a

+ (2.17)

Introducing the non-dimensional shear stress
*
2
w
V

w, we have:


xf
Re

(2.18)

Analysis of the homotopy perturbation method: To
illustrate the basic ideas of this method, we consider the
following equation:

( ) ( ) A u f r 0 r (3.1)

With the boundary condition of:


u
B u, 0, r
n
_


,
(3.2)

where A is a general differential operator, B a boundary
operator, f (r) a known analytical function and G is the
boundary of the domain O.
A can be divided into two parts which are L and N,
where L is linear and N is nonlinear. Eq. (3.1) can
therefore be rewritten as follows:

( ) ( ) ( ) L u N u f r 0, r + (3.3)

Homotopy perturbation structure is shown as follows:

( ) ( ) ( ) ( ) ( ) ( )
0
H , p 1 p L L u p A f r 0 1 + 1
] ]
(4)

Where:
( ) [ ] r,p : 0,1 R (5)

In Eq. (3.5), p[0,1] is an embedding parameter
and u
0
is the first approximation that satisfies the
boundary condition. We can assume that the solution of
Eq. (3.4) can be written as a power series in p, as
following:


2
0 1 2
p p ... + + + (3.6)

And the best approximation for solution is:


p 1 0 1 2
u lim ...

+ + + (3.7)
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
85

Implementation of the method: According to the
so-called homotopy-perturbation method (HPM),we
construct a homotopy Suppose the solution of Eq.(3.4)
which has the form of:


IV IV IV ''' ' '
0 1
''' ' ''
(1 p)(f f ) H (P)(f (yf 3f )
Ref f Ref f ) 0
+ + +
+
(4.1)

We consider as follows:


n
0 1 i
i 0
f ( ) f ( ) f ( ) ... f ( )

+ +

(4.2)

with substituting f from Eq. (4.2) to Eq. (4.1) into and
some simplification and rearranging based on powers of
p-terms, we have:


0 IV
p : f 0

'' '
0 0 0 0
f (0) 0,f (0) 0,f (1) 1,f (1) 0 (4.3)


1
p :
IV ' '' ''' '' '''
1 0 0 0 0 0 0
f Ref f Ref f 3 f yf 0 + + +

' ' '
1 1 1 1
f ( 0) 0,f (0) 0,f(1) 0,f (1) 0 (4.4)

.
.
.

Solving Eqs. (4.3) and (4.4) with boundary
conditions, we have:


2
0
1 3
f (y) y y
2 2
+ (4.5)

7 5 3
1
1 1 1 9 6
f ( ) 3( Rey y ) ( Re )y
840 30 6 140 5
1 1
( Re )y
140 10
y + +
+ +
(4.6)

The solution of this equation, when p1, will be
as follows:


0 1 2
f ( ) f (y) f ( y) f (y) ... + + + (4.7)

Differential transform method: Basic definitions and
operations of differential transformation are introduced
as follows.
Differential transformation of the function () is
defined as follows:

( )
( )
0
k
k
d f 1
F k
k! d

]
(5.1)

In (5.1), () is the original function and F(k) is the
transformed function which is called the T-function (it
is also called the spectrum of the () at =
0
, in the k
domain). The differential inverse transformation of F(k)
is defined as:

( )
k
0
k 0
f F(k)( )

(5.2)

By Combining (4.1) and (4.2) () can be obtained:

( )
( )
0
k k
0
k
k 0
d f ( )
f
d k!

1

1

(5.3)

Equation (5.3) implies that the concept of the
differential transformation is derived from Taylors
series expansion, but the method does not evaluate the
derivatives symbolically. However, relative derivatives
are calculated by an iterative procedure that is described
by the transformed equations of the original functions.
From the definitions of (5.1) and (5.2), it is easily
proven that the transformed functions comply with the
basic mathematical operations shown in below. In real
applications, the function () in (5.3) is expressed by a
finite series and can be written as:

( )
N
k
0
k 0
f F(k)( )

(5.4)

Equation (5.4) implies that

( ) ( )
k
0
k N 1
f F(k)( )

Table 1: Some of the basic operations of differential transf ormation
method
Original function Transformed function
f ( ) g( ) h( ) t F[k] G[k] H[k] t
n
n
d g ( )
f ( )
d

(k n)!
F[k] G[k n]
k!
+
+

f ( ) g( )h( )
k
m 0
F[k] F[m]H[k m]

f ( ) sin( ) +
k
k
F[k] sin( )
k! 2

+
f ( ) cos( ) +
k
k
F[k] cos( )
k! 2

+
f ( ) e

k
F[k]
k!

( )
m
F( ) 1 +
m(m 1)...(m k 1)
F[k]
k!
+

m
f ( )
1 , k m
F[k] (k m)
0 , k m

'

World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
86

is negligibly small, where N is series size.
Theorems to be used in the transformation procedure, which can be evaluated from (5.1) and (5.2), are given
below.

Solution with differential transformation method: Now Differential Transformation Method into governing
equations has been applied. Taking the differential transform of Eqs.(2.12)and (2.13) with respect to and
considering H = 1 gives:


( )
( )
( )
k
m 0
k
m 0
k
m 0
(k 1)(k 2)(k 3)(k 4)F[k 4] [m](k m 1)(k m 2)(k m 3)F[k m 3]
3 (k 1)(k 2)F[k 2] Re F[k m](m 1)(m 2)(m 3)F[m 3]
1 m 1
Re (k m 1)F[k m 1](m 1)(m 2)F[m 2] 0, [m]
0 m 1

+ + + + + + + + + +
+ + + + + + + + +

+ + + + +
'

(6.1)


0 1
F[0] 0, F[1] a , F[2] 0,F[3] a (6.2)

where F(k) are the differential transforms of () and a
0
,a
1
are constants which can be obtained through boundary
condition, Eq.(2.13). This problem can be solved as followed:

0 1
F[0] 0, F[1] a , F[2] 0, F[3] a, F[4] F[6] F[8] 0

0
3
F[5] a
20

2 2
1 1 0 1
3 1 1
F[7] a Rea Rea a
280 70 140
+ +
3 2 2
1 1 0 1
2 2 2
0 1 0 1 0 1
1 1 1
F[9] a Rea Rea a
2240 560 1120
1 1 1
a a a Rea Rea a
1260 1260 2520


(6.3)
.
.
.

The above process is continuous. By substituting equations (6.3) into the main equation based on DTM, it can
be obtained that the closed form of the solutions is:


3 5 2 2 7 3 2
0 1 1 1 0 1 1 1
2 2 2 2 9
0 1 0 1 0 1 0 1
3 3 1 1 1 1
F( ) a a a Rea Rea a a Rea
20 280 70 140 2240 560
1 1 1 1
Rea a a a a Rea Rea a ...
1120 1260 1260 2520
_ _
+ + + + + +

, ,
_
+

,
(6.4)

By substituting the boundary condition from Eq.(2.13) into Eq.(6.4) in point = 1 the values of a
0
,a
1
can be
obtained.


2 2 3 2
0 1 1 1 0 1 1 1
2 2 2 2
0 1 0 1 0 1 0 1
3 3 1 1 1 1
F(1) a a a Rea Re a a a Rea
20 280 70 140 2240 560
1 1 1 1
Rea a a a a Rea Rea a ... 1
1120 1260 1260 2520
_ _
+ + + + + +

, ,
_
+

,
(6.5)


' 2 2 3 2
0 1 1 1 0 1 1 1
2 2 2 2
0 1 0 1 0 1 0 1
3 3 1 1 1 1
F(1) a 3a 5 7 a Rea Re a a 9 a Rea
20 280 70 140 2240 560
1 1 1 1
Rea a a a a Rea Rea a ... 0
1120 1260 1260 2520
_ _
+ + + + + +

, ,
_
+

,
(6.6)
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
87

By solving Equations (6.5),(6.6) gives the values of
a
0
,a
1
. By substituting obtained a
0
,a
1
into Eq.(6.4), it can
be obtained the expression of F().

FUNDAMENTALS OF OPTIMAL HOMOTOPY
ASYMPTOTIC METHOD (OHAM)

Following differential equation is considered:

( ) L(u(t)) N(u(t)) g t 0, + + B(u) = 0 (7.1)

where L is a linear operator, is an independent
variable, u(t) is an unknown function, g(t) is a known
function, N(u(t)) is a nonlinear operator and B is a
boundary operator. By means of OHAM one first
constructs a set of equations:


(1 p)[L( ( , p)) g ( ) ] H(p)[L( ( ,p))
g( ) N( ( ,p))] 0
B( ( ,p)) 0
+
+ +

(7.2)

where p[0,1] is an embedding parameter, H(p)
denotes a nonzero auxiliary function for p0 and
H(0) = 0, (,p) is an unknown function. Obviously,
when p = 0 and p = 1, it holds that:


0
( , 0 ) u ( ), ( ,1) u( ) (7.3)

Thus, as p increases from 0 to 1, the solution (,p)
varies from u
0
() to the solution u(), where u
0
() is
obtained from Eq. (2) for p = 0:


0 0
L(u ( )) g( ) 0,B(u ) 0 + (7.4)

We choose the auxiliary function H(p) in the form:


1 2 2
H(p) pC p C ... + + (7.5)

where C
1
, C
2
, are constants which can be determined
later.
Expanding (,p) in a series with respect to p, one has:


i 0 k i k
k 1
( , p, C) u ( ) u ( , C) p , i 1,2,...

(7.6)

Substituting Eq. (7.6) into Eq. (7.2), collecting the
same powers of p and equating each coefficient of p to
zero, we obtain set of differential equation with
boundary conditions. Solving differential equations by
boundary conditions u
0
(), u
1
(, C
1
), u
2
(, C
2
), are
obtained. Generally speaking, the solution of Eq. (7.1)
can be determined approximately in the form:

m
(m)
0 k i
k 1
u u ( ) u ( , C )

% (7.7)

Note that the last coefficient C
m
can be function of
. Substituting Eq. (3.7) into Eq. (3.1), there results the
following residual:


(m) (m)
i i i
R( , C) L(u ( , C) ) g( ) N(u ( , C)) + + % % (7.8)

If R(, C
i
) = 0 then
(m)
i
u ( , C) % happens to be the
exact solution. Generally such a case will not arise for
nonlinear problems, but we can minimize the
functional:


b
2
1 2 n 1 2 m
a
J ( C, C, . . . , C) R ( , C, C, . . . , C )d

(7.9)

where a and b are two values, depending on the given
problem. The unknown constants C
i
(i = 1,2,,m) can
be identified from the conditions:


1 2
J J
... 0
C C



(7.10)

with these constants, the approximate solution (of order
m) (Eq. (7.7)) is well determined. It can be observed
that the method proposed in this work generalizes these
two methods using the special (more general) auxiliary
function H(p).

SOLUTION WITH OPTIMAL HOMOTOPY
ANALYSIS METHOD

In this section, OHAM is applied to nonlinear
ordinary differential Eq. (2.12) According to the
OHAM, applying Eq. (7.2) to Eq. (2.12):


IV IV IV ''' ' '
0 1
''' ' ''
(1 p)(f f ) H (P)(f (yf 3f )
Ref f Ref f ) 0
+ + +
+
(8.1)

where primes denote differentiation with respect to .
We consider and H
1
(p) as following:

2
0 1 2
f f pf p f + +


2
1 11 12
H (p) pC p C + (8.2)

Substituting and H
1
(p) from Eq. (8.2) into Eq.
(8.1) and some simplification and rearranging based on
powers of p-terms, we have:
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
88

0 IV
p : f 0


'' '
0 0 0 0
f (0) 0,f (0) 0,f (1) 1,f (1) 0 (8.3)

1
p :
IV IV ' '' ''' '' IV '''
1 0 11 0 0 0 0 0 0 0
f f C ( Ref f Re f f 3 f f y f ) 0 + + + +
' ' '
1 1 1 1
f ( 0) 0,f (0) 0,f(1) 0,f (1) 0 (8.4)

Solving Eqs. (8.3) and (8.4) with boundary conditions:


2
0
1 3
f (y) y y
2 2
+ (8.5)


7 5
1 11
3
11 11
11 11
1 1
f ( ) 3C ( Rey y )
840 30
1 9 6
( C Re C )y
6 140 5
1 1
( C Re C )y
140 1
y
0
+
+
+ +
(8.6)

The term of
2
(y) is too large that is mentioned
graphically. Therefore final expression for (y) is:


0 1 2
f ( ) f (y) f (y) f (y) + + (8.7)

From Eq. (7.8) by Substituting () into Eq. (8.1),
R
1
(y, C
11
, C
12
) and R
2
(y, C
21
, C
22
) are obtained and J
1
and J
2
are obtained in the flowing manner:

2
1 11 12 1 11 12
0
J ( C , C ) R (y,C , C ) dy

2
2 21 22 2 21 22
0
J (C , C ) R (y,C , C )dy

The constants C
11
, C
12
obtain from Eq. (8.10) in the
particular cases:
When Re = 5, = 1:

11 12
C 0.94688,C 0.195

By substituting Eq. (8.11) into Eq. (8.7), an
expression for (y) is obtained.

RESULTS AND DISCUSSION

The objective of the present study was to apply
HPM, DTM and OHAM to obtain an explicit analytic
solution of laminar, isothermal, incompressible viscous
flow in a rectangular domain bounded by two moving
porous walls, which enable the fluid to enter or exit

Table 2: Constant values with different non-dimensional parameters
Re a0 a1
1 1 1.586849 -0.68113
1 2 1.696949 -0.93264
2 1 1.585146 -0.67823

Table 3: The HPM, DTM, OHAM and numerical solution results for
f(y) when = 0, Re = 1
f(y)
-----------------------------------------------------------------------
y NM HPM DTM OHAM
0.00 0.000000 0.000000 0.000000 0.000000
0.05 0.086267 0.086267 0.084731 0.097431
0.10 0.171760 0.171760 0.168765 0.193764
0.15 0.255721 0.255721 0.251416 0.287920
0.20 0.337421 0.337421 0.332018 0.378868
0.25 0.416176 0.416176 0.409935 0.465637
0.30 0.491359 0.491359 0.484572 0.547345
0.35 0.562408 0.562408 0.555383 0.623212
0.40 0.628838 0.628838 0.621875 0.692581
0.45 0.690244 0.690244 0.683621 0.754934
0.50 0.746307 0.746307 0.740259 0.809906
0.55 0.796792 0.796792 0.791502 0.857300
0.60 0.841553 0.841553 0.837137 0.897093
0.65 0.880522 0.880522 0.877030 0.929448
0.70 0.913710 0.913710 0.911122 0.954712
0.75 0.941200 0.941200 0.939433 0.973425
0.80 0.963137 0.963137 0.962059 0.986309
0.85 0.979721 0.979721 0.979164 0.994267
0.90 0.991199 0.991199 0.990982 0.998368
0.95 0.997854 0.997854 0.997811 0.999830
1.00 1.000000 1.000000 1.000000 1.000000

during successive expansions or contractions (Fig. 1).
Also, error percentage is introduced as followed:

NM a
NM
%Error
f( ) f ( )
100
f ( )

where ()

is a value that obtained using different

analytical methods.
Figure 2 shows the comparison between numerical
method and other analytical solutions results for (y)
when = Re = 1. It verifies that, there is acceptable
agreement between the numerical solution obtained by
four-order Rung-kutte method and these methods.
Table 2 and 3 confirm the last conclusion.
Comparison between obtained results showed that
HPM is more accurate and acceptable than two other
methods, as can be seen in Fig. 3 and Table 4.
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
89



Fig. 2: Comparison between numerical method and
other analytical solutions for (y) when =
Re =1

0 0.2 0.4 0.6 0.8 1
0
0.15
0.3
OHAM
%
E
r
r
o
r
y
HPM
DTM


Fig. 3: Comparison between error percentages of HPM,
DTM and OHAM for (y) when = Re =1

After this validity, results are given for the velocity
profile, normal pressure distribution and wall shear
stress for various values of permeation Reynolds
number and non-dimensional wall dilation rate.
Figure 4 illustrate the behavior of (y) (or uc/x)
for different permeation Reynolds number, over a range
of non-dimensional wall dilation rate. For every level of
injection or suction, in the case of expanding wall,
increasing leads to higher axial velocity near the
center and the lower axial velocity near the wall. The
reason is that the flow toward the center becomes
greater to make up for the space caused by the
expansion of the wall and as a result, the axial velocity
also becomes greater near the center.

Table 4: The results of HPM, DTM, OHAM and numerical solution
for f(y) when = 1, Re = 2
f(y)
-----------------------------------------------------------------------
y NM HPM DTM OHAM
0.00 0.000000 0.000000 0.000000 0.000000
0.05 0.080088 0.080088 0.079173 0.086080
0.10 0.159633 0.159633 0.157837 0.171440
0.15 0.238097 0.238097 0.235491 0.255368
0.20 0.314954 0.314954 0.311636 0.337174
0.25 0.389691 0.389691 0.385788 0.416193
0.30 0.461819 0.461819 0.457477 0.491799
0.35 0.530872 0.530872 0.52625 0.563412
0.40 0.596412 0.596412 0.591678 0.630506
0.45 0.658037 0.658037 0.653355 0.692621
0.50 0.715375 0.715375 0.710901 0.749366
0.55 0.768097 0.768097 0.76397 0.800432
0.60 0.815908 0.815908 0.812245 0.845595
0.65 0.858559 0.858559 0.855444 0.884723
0.70 0.895837 0.895837 0.893323 0.917782
0.75 0.927572 0.927572 0.925674 0.944843
0.80 0.953638 0.953638 0.952328 0.966079
0.85 0.973944 0.973944 0.973157 0.981772
0.90 0.988441 0.988441 0.988071 0.992311
0.95 0.997119 0.997119 0.997021 0.998189
1.00 1.000000 1.000000 1.000000 1.000000

Table 5: The comparison among error percentages of HPM, DTM
and OHAM for f(y) when = Re = 2
f(y)
-------------------------------------------------------------------
y HPM DTM OHAM
0.00 0.000000 0.000000 0.000000
0.05 3.35E-06 0.016579 0.086579
0.10 1.19E-06 0.008088 0.098088
0.15 1.13E-06 0.005590 0.105590
0.20 1.26E-06 0.023750 0.133750
0.25 1.12E-06 0.045427 0.165427
0.30 9.41E-07 0.069424 0.199424
0.35 8.64E-07 0.094337 0.224337
0.40 7.35E-07 0.118603 0.248603
0.45 6.08E-07 0.140557 0.270557
0.50 5.19E-07 0.158494 0.288494
0.55 4.14E-07 0.170769 0.300769
0.60 3.66E-07 0.175905 0.305905
0.65 4.30E-07 0.172741 0.302741
0.70 2.51E-07 0.160611 0.290611
0.75 2.52E-07 0.139584 0.269584
0.80 2.62E-07 0.110756 0.240756
0.85 1.00E-07 0.076631 0.196631
0.90 1.42E-07 0.041614 0.151614
0.95 4.18E-07 0.012641 0.092641
1.00 2.00E-07 2.0E-08 0.000000
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
90

0 0. 2 0.4 0.6 0. 8 1
0
0.4
0.8
1.2
1.6
0
0.5
+0.5
f(y)
'
y

0 0.2 0.4 0.6 0.8 1
0
0. 4
0. 8
1. 2
1. 6
0
0.5
+0.5
f(y)
'
y

(a) (b)
Fig. 4: (y)changes shown over a range of at (a) Re = -5 (b) Re = 5

0 0.2 0.4 0.6 0.8 1
-2.4
-2
-1.6
-1.2
-0.8
-0.4
0
0
0.5
+0.5
P
r
e
s
s
u
r
e
d
r
o
p
y

0 0.2 0.4 0.6 0.8 1
0
0.4
0.8
1.2
1.6
0
0.5
+0.5
P
r
e
s
s
u
r
e
d
r
o
p
y

(a) (b)
Fig. 5: The pressure drop in the normal direction (p
n
) changes shown over a range of at (a) Re = -1 (b) Re = 1

0 0. 2 0.4 0.6 0. 8 1
0
0.8
1.6
2.4
3.2 0
0.5
+0.5

x
w

0 0.2 0.4 0.6 0.8 1
-3.2
-2.4
-1.6
-0.8
0
0
0.5
+0.5

x
w

(a) (b)
Fig. 6: Shear stress changes shown over a range of at (a) Re = -1 (b) Re = 1
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
91

The pressure distribution in the normal direction
for various permeation Reynolds numbers over a range
of non-dimensional wall dilation rates, are plotted in
Fig. 5.
Figure 5 shows that for every level of injection or
suction, the absolute pressure change in the normal
direction is lowest near the central portion.
Furthermore, by increasing non-dimensional wall
dilation rates the absolute value of pressure distribution
in the normal direction increases.
The wall shear stress (
xf (1)
Re

) for permeation
Reynolds number Re = -1 and Re = 1 over a range of
Non-dimensional wall dilation rates, are plotted in
Fig. 6. We can observe from Fig. 6 that the absolute
shear stress along the wall surface increases in
proportion to x.Furthermore, by increasing non-
dimensional wall dilation rates the absolute value of
shear stress increases.

CONCLUSION

In this research, the HPM, DTM and OHAM were
successfully applied to find the analytical solution for
laminar, isothermal, incompressible and viscous flow in
a rectangular domain bounded by two moving porous
walls, which enable the fluid to enter or exit during
successive expansions or contractions. The accuracy of
the methods is considerable. The figures and tables
clearly show the high accuracy of three methods to
solve this special and such other problems in
engineering. Also from these figures, we can find that
HPM is more accurate than two other methods in this
case.

Nomenclature
F
c Injection/suction coefficient
F Transformed of f
p Pressure
p
n
Pressure drop in the normal direction
Re Permeation Reynolds number
u,v Velocity components along x, y axes, respectively
V
w
Injection velocity

Greek symbols
Kinematic viscosity
Non-dimensional wall dilation rate
shear stress
Fluid density

REVERENCES

1. Chang, H.N. and J.S. Ha, J.K. Park, I.H. Kim and
H.D. Shin, 1989. Velocity field of pulsatile flow in
a porous tube. J. Biomech, 22: 1257-1262.
2. Dauenhauer, E.C. and J. Majdalani, 1999.
Unsteady flows in semi-infinite expanding
channels with wall injection. In: AIAA paper,
99: 3523.
3. Majdalani, J. and C. Zhou, 2003. Moderate-to-large
injection and suction driven channel flows with
expanding or contracting walls. ZAMM Z. Angew.
Math. Mech., 83: 181-196.
4. Boutros, Y.Z. and M.B. Abd-el-Malek, N.A.
Badran and H.S. Hassan, 2006. Lie-group method
for unsteady flows in a semi -infinite expanding
or contracting pipe with injection or suction
through a porous wall. J. Comput. Appl. Math.,
197 (2): 465-494.
5. He, JH., 2000. A review on some new recently
developed nonlinear analytical techniques. Int. J.
Non-linear Sci. Numer Simul., 1: 51-70.
6. He, J.H., 2005. Homotopy perturbation method
for bifurcation on nonlinear problems. Int. J.
Non-linear Sci. Numer Simul., 8: 207-208.
7. Ganji, Z.Z. and D.D. Ganji, 2008. Approximate
solutions of thermal boundary-layer problems in a
semi-infinite flat plate by using Hes homotopy
perturbation method. Int. J. Nonlinear Sci. Numer.
Simul., 9 (4): 415-422.
8. Ganji, D.D., H.R. Ashory Nezhad and A.
Hasanpour, 2011. Effect of variable viscosity and
viscous dissipation on the Hagen-Poiseuille flow
and entropy generation. Numerical Methods for
Partial Differential Equations, 27 (3): 529-540.
9. Ganji, D.D. and A. Sadighi, 2006 Application of
Hes homotopy-perturbation method to nonlinear
coupled systems of reaction-diffusion equations.
Int. J. Nonlinear Sci. Numer. Simul., 7 (4):
411-418.
10. Zhou, J.K., 1986. Differential transformation and
its applications for electrical circuits. Wuhan.
China: Huazhong University Press; [In Chinese].
11. Chen, C.K. and S.H. Ho, 1999. Solving partial
differential equations by two dimensional
differential transform method. Appl. Math.
Comput., 106: 171-179.
12. Ayaz, F., 2004. Solutions of the systems of
differential equations by differential transform
method. Appl. Math. Comput., 147: 547-567.
13. Khan, N.A., A. Ara, A. Yildirim and E. Ylkl,
2010. Approximate Solution of Helmholtz
Equation by Differential Transform Method.
World Applied Sciences Journal, 11 (12):
1509-1511.
14. Yildirim, A. and Y. Glkanat, 2010. HPM-Pad
Technique for Solving a Fractional Population
Growth Model. World Applied Sciences Journal,
11 (12): 1528-1533.
World Appl. Sci. J., 16 (Special Issue of Applied Math): 82-92, 2012
92

15. Mohyud-Din, S.T. and A. Yildirim and Y.
Glkanat, 2011. Analytical Solution of Nonlinear
Thermoelasticity Cauchy Problem. World Applied
Sciences Journal, 12 (12): 2184-2188.
16. Marinca, V. and N. Herianu, 2008. Application of
optimal homotopy asymptotic method for
solving nonlinear equations arising in heat
transfer. Internat. Commun. Heat Mass Transfer,
35: 710-715.
17. Marinca, V., N. Herianu, C. Bota and B. Marinca,
2009. An optimal homotopy asymptotic method
applied to the steady flow of a fourth-grade
fluid past a porous plate. Appl. Math. Lett., 22 (2):
245-251.

18. Majdalani, J., C. Zhou and C.A. Dawson, 2002.
Two-dimensional viscous flow between slowly
expanding or contracting walls with weak
permeability. J. Biomech., 35: 1399-1403.
19. Berman, A.S., 1953. Laminar flow in channels
with porous walls. J. Appl. Phys., 24: 1232-1235.