Applied Mathematics of Fluid Flow in Porous Media

APPLIED MATHEMATICS OF FLUID FLOW IN POROUS MEDIA
Erdal Ozkan, Ph.D.
Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2005
Chapter 1 INTRODUCTION
These notes explain how fluid flow in porous media can be translated into a mathematical statement and how mathematical analysis can be used to find answers to transient flow problems. It should be emphasized that this is a very broad area, which is common to many other disciplines, such as heat conduction in solids and groundwater hydrology. The objective of these notes is to introduce the fundamentals, present examples, and guide the interested reader to the relevant references. The first chapter presents and introduction of the derivation of flow equations in porous media. Most physical phenomena in the domain of transient fluid flow in porous media can be described, in general, by partial differential equations (PDE). Together with appropriate boundary conditions and sometimes simplifying assumptions, the PDE leads to an initial boundary value problem (IBVP) that is solved to find a mathematical statement of the resulting flow in the porous medium. This section discusses, briefly, the statement of the IBVP for transient fluid flow in porous media.
Chapter 2 EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

This chapter explains how fluid flow in porous media can be translated into a mathematical statement. Most physical phenomena in the domain of transient fluid flow in porous media can be described in general by partial differential equations (PDE). Together with appropriate boundary conditions and sometimes simplifying assumptions, the PDE leads to an initial boundary value problem (IBVP) that is solved to find a mathematical statement of the resulting flow in the porous medium. This chapter discusses, briefly, the statement of the IBVP for transient fluid flow in porous media. 2.1. Equations of Transient Fluid Flow in Porous Media
r In essence, fluid motion in porous media can be specified by the knowledge of the velocity vector, v , and the r r density of the fluid, , as a function of the position (x, y, z) and time t; that is, v = v (x , y , z , t ) and = (x , y , z , t ) . Relative to the fixed Cartesian axes, the velocity vector can be written as
r r r r v = v x ix + v y i y + v z iz ,
(2.1)
r r r where, v x , v y , and v z are the velocity components and i x , i y , and i z are the unit vectors in the x, y, and z
directions, respectively. The physical law governing the macroscopic fluid-flow phenomena in porous media is the conservation of mass, which simply states that mass is neither created nor destroyed. The mathematical formulation of this rule is developed by considering the flow through an arbitrary closed surface S lying entirely within a porous medium of porosity, , that is filled with a fluid of viscosity, (Fig. 2.1).
S n V dS
Fig. 2.1 Arbitrary closed surface S in porous medium. The conservation of mass principle states that the difference between the rates at which fluid enters and leaves the volume V through its surface S must equal the rate at which mass accumulates within V . The total mass within the volume V at any time t is given by
M=
dV .
V
(2.2)
Then, the time rate of change of mass within S is
Chap. 2
EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

dM d = dt dt
dV = t ( ) dV ,
V V
(2.3)
which, by the conservation of mass, is equal to the rate at which mass enters V through the surface S . Let us, now, consider the differential surface element dS shown in Fig. 2.1. The mass entering the volume V r r r r through dS at the normal velocity, v n , in a time increment t is dM = v n dS t , and the total mass of the fluid passing through S during t is
M = t
v n dS .
S
r r
(2.4)
Note that the surface integral in Eq. 2.4 takes into account both influx and outflux through the surface of the volume; that is, M is the difference between the masses entering and leaving the control volume during the time increment, t . Then, the mass rate entering the volume V through its surface S can be written as
q= M dM = = t dt
v n dS .
S
r r
(2.5)
By the conservation of mass principle, equating the right-hand-sides of Eqs. 2.3 and 2.5, we obtain
t ( ) dV + v n dS 0 .
V S
r r
(2.6)
r A more useful relation is found by using the divergence theorem, which states that the flux of v through the r r closed surface S is identical to the volume integral of ( v ) (the divergence of v ) taken throughout V; that is,
v n dS = ( v ) dV .
S V
r r
(2.7)
Here, is the gradient operator, which, for example, in three-dimensional Cartesian and cylindrical coordinates, is given, respectively, by
and
r r r ix + i y + iz x y z
(2.8)
r i r r i + i z . ir + r r z
(2.9)
Using the relation in Eq. 2.7, Eq. 2.6 can be recast into the following form:
( ) + v dV 0 . t
V
(2.10)
If the functions involved in the argument of the integral in Eq. 2.10 are continuous, then the integral is identically zero if and only if its argument is zero (because the volume integral in Eq. 2.10 is identically zero for any arbitrarily chosen volume). Then, we obtain the following relation that is known as the continuity equation:
Chap. 2
r ( ) + ( v )= 0 . t
(2.11)
Equation 2.11 is a partial differential equation that is equivalent to the statement of the conservation of mass for fluid flow in porous media. For practical purposes, however, we prefer to express Eq. 2.11 in terms of pressure r because we cannot directly measure density and velocity. To express density, , and velocity, v , in terms of pressure, we use an equation of state and a flux law, known as Darcys Law, respectively. A useful equation of state that relates density to pressure is the definition of isothermal compressibility, c, for a fluid given by 1 c= p . T (2.12)
If c is a constant (compressibility of most reservoir liquids may be considered as constant), then, Eq. 2.12 can be integrated to yield
= 0 exp[c( p p 0 )] ,
(2.13)
where the subscript 0 indicates the conditions at the datum. Similarly, the compressibility of the porous rock, c f , is defined by
cf =
1 p
(2.14)
and the total system compressibility, c t , is given by
ct = c + c f .
These definitions of compressibility are useful to recast Eq. 2.11 in terms of pressure.
(2.15)
Darcys Law for fluid flow in porous media is a flux law. Neglecting the gravity effect, it is expressed by the following relation:
r k v = p .
(2.16)
In Eq. 2.16, is the viscosity of the fluid and k is the permeability tensor of the formation given by
k k = k k
k k k
k k , k
(2.17)
where , , and are the directions and k ij is the permeability in the i direction as a result of the pressure gradient in the j direction. If we use Eqs. 2.13 2.16 in Eq. 2.11, we obtain an alternate statement of the conservation of mass principle for fluid flow in porous media as follows:
Chap. 2

k p k 2 p + c (p ) = ct t .
(2.18)
Equation 2.18 is the partial differential equation that governs transient fluid flow in porous media. As it stands, however, Eq. 2.18 is not very useful to obtain practical solutions because of the nonlinearity displayed in the second term of the left-hand side of Eq. 2.18. For liquid flow, the viscosity, , is constant and this equation can be linearized by assuming that the pressure gradients ( p ) are small in the reservoir and the compressibility of the reservoir liquids, c, is in the order of 10 5 or smaller. Then, the second term in the left-hand side of Eq. 2.18 may be neglected compared to the remaining terms and the following linear expression is obtained:
(k p ) = c t p t
(2.19)
Eq. 2.19 (or 2.18) is known as diffusivity equation. Using the Cartesian coordinates for example and assuming that the coordinate axes can be chosen in the directions of the principal permeabilities, k in Eq. 2.19 may be represented by the following diagonal tensor:
kx k = 0 0 0 ky 0 0 0 . kz
(2.20)
Then, Eq. 2.19 can be written as
kx
2 p x 2
+ ky
2 p y 2
+ kz
2 p z 2
= ct
p . t
(2.21)
Note that if we multiply each coordinate, j = x , y , or z , by the material balance, k is usually chosen to be for an isotropic domain given below:
3
k k j , where k may be chosen arbitrarily (to satisfy
k x k y k z ), Eq. 2.21 may be transformed into the diffusion equation
2 p
p =0 t
(2.22)
where is the diffusivity constant defined by
ct
(2.23)
If the same transformation is also applied to the boundary conditions, which will be discussed in the next section, the problems in anisotropic reservoirs may be transformed to those in isotropic reservoirs provided that the system is infinite or bounded by planes perpendicular to the principal axes of permeability. In all other cases, this transformation distorts the bounding surfaces and the c(p )2 term in Eq. 2.18 may not be negligible. In these cases, an expression similar to Eq. 2.19 may be obtained from Eq. 2.18 in terms of pseudopressure, m, as follows: For the flow of gases, the assumptions of small fluid compressibility and pressure gradient may not be appropriate
Chap. 2

2m x
2
kx
+ ky
2m y
2
+ kz
2m z
2
= ct
m . t
(2.24)
Here, the pseudopressure is defined by1
m( p ) = 2
Z dp ,
0
(2.25)
where Z is the compressibility factor. To define a complete physical problem, Eq. 2.21 (or 2.24) should be subject to initial and boundary conditions discussed in the next section.
2.2. Initial and Boundary Conditions
The solution of the diffusivity equation (Eq. 2.21) should satisfy the initial condition in the porous medium. The initial condition is normally expressed in terms of a known pressure distribution at time zero; that is,
t 0
lim p(x , y , z ,t ) = f (x , y , z ) .
(2.26)
The most common initial condition is the uniform pressure distribution in the entire porous medium; that is, f (x , y , z ) = p i . The boundary conditions are specified at the inner (wellbore) and outer boundaries of the reservoir. These are usually in the form of prescribed flux or pressure at the boundaries. The condition of prescribed flux can be formulated as
p = g (t ) , n
(2.27)
where is the surface of the boundary and n indicates the outward normal direction of the boundary surface. The prescribed flux condition may be used at the inner and outer boundaries of the reservoir. The most popular use of the prescribed flux condition at the inner boundary is for the production at a constant rate. In this case, the function, g (t ) , is related to a constant production rate, q. At the outer boundary, the prescribed flux condition is usually used to indicate impermeable boundaries [ g (t ) = 0 ] and leads to a pseudosteady state under the influence of boundaries. For some applications, pressure may be specified at the inner and outer boundaries. In this case, we have
( p )
= h (t ) .
(2.28)
When used at the inner boundary, this condition represents production at a constant pressure, p wf ; that is aquifer, which usually leads to steady state in the reservoir.
h(t ) = p wf . At the outer boundary, specified pressure, p e , is usually due to injection or influx from an adjacent
It is also possible to have mixed-type of boundary conditions. These usually correspond to interface conditions in porous media. More details about the common boundary conditions for the diffusion equation may be found in Ref. 2.
Chap. 2
2.3. Assumptions and Limits
In the derivation of the diffusivity equation given by Eq. 2.19, some assumptions have been made. These assumptions determine the limits of applicability of the solutions obtained from Eq. 2.19. One of the most important assumptions involved is the continuity of the properties involved in Eq. 2.19 (this was required to obtain Eq. 2.19 from the more general integral form in Eq. 2.10). Therefore, sharp changes in the properties of the reservoir rock and fluid (such as faults and fluid banks) should be incorporated in the form of boundary or interface conditions in the solution of Eq. 2.19. The second important assumption is that Darcys Law describes the flux in porous media. This assumption is valid at relatively low fluid velocities that may be appropriate to describe liquid flow. At high velocities (when Reynolds number based on average sand grain diameter approaches unity), such as those observed in gas reservoirs, Darcys Law is not valid.3 In this case, Forcheimers equation4 that takes into account the inertial effects should be used. In petroleum engineering, it is a common practice to consider the additional pressure drop because of non-Darcy flow in the form of a pseudoskin because it is usually effective in a small vicinity of the wellbore. Therefore, in these notes, we will not consider non-Darcy flow in the reservoir.
Chapter 3 BESSELS EQUATION AND BESSEL FUNCTIONS

As will be shown in Chapter 4, the Laplace transform of the diffusion equation in radial coordinates yields the modified Bessels equation and its solutions are obtained in terms of Bessel functions. Our objective in this chapter is to introduce Bessel functions and discuss some of their properties to the extent they are encountered in the solutions of common petroleum engineering problems. 3.1. Preliminary Definitions A differential equation of the type
d2y dz 2
1 dy z 2 v 2 + y = 0; z dz z2
v0
(3.1)
is called a Bessels equation of order v. A solution of Bessels equation of order v is called a Bessel function of order v. A differential equation of the type
d2y dz 2
1 dy 2 z 2 v 2 + y = 0; z dz z2
v0
(3.2)
is called a modified Bessels equation of order v. Note that Eq. 3.2 is obtained by substituting z for z in Eq. 3.1. Of particular interest to us is the case where = ki so that Eq. 3.2 becomes
d2y dz 2
1 dy k 2 z 2 + v 2 y = 0; z dz z2
v0.
(3.3)
Eq. 3.3 is referred to as the modified Bessels equation of order v in the petroleum engineering literature. A solution of the modified Bessels equation of order v is called a modified Bessel function of order v.
3.2. Solutions of Bessels Equations and Bessel Functions
There are many methods of obtaining or constructing Bessel functions.5 Here, we will only present the final form of the Bessel functions that are of interest to us. If v is not a positive integer, then the general solution of Bessels equation of order v (Eq. 3.1) is given by
y = AJ v (z ) + BJ v (z ) ,
(3.4)
where A and B are arbitrary constants and J v (z ) is the Bessel function of order v of the first kind given by
J v (z ) =
m =0
m! (v + m + 1) .
( 1)m (z 2)v + 2m
(3.5)
Chap. 3
BESSEL'S EQUATION AND BESSEL'S FUNCTIONS
In Eq. 3.5, (x ) is the gamma function defined by

(x ) = e t t x 1 dt .
(3.6)
If v is a positive integer, n, then J v and J v are linearly dependent and the solution of Eq. 3.1 for this case is written as
y = AJ n (z ) + BYn (z ) .
(3.7)
In Eq. 3.7, Yn (z ) is the Bessel function of order n of the second kind and is defined by
J (z ) ( 1)n J v (z ) Yn (z ) = lim v . vn vn
(3.8)
Similarly, if v is not a positive integer, the general solution of the modified Bessels equation of order v (Eq. 3.3) is given by
y = AI v (kz ) + BI v (kz ) ,
(3.9)
where I v (z ) is the modified Bessel function of order v of the first kind defined by
I v (z ) =
(kz 2)v + 2m . m! (v + m + 1) m =0
(3.10)
If v is a positive integer, n, I v and I v are linearly dependent. The solution for this case is
y = AI n (kz ) + BK n (kz ) ,
(3.11)
where K n (z ) is the modified Bessel function of order n of the second kind and is defined by I (z ) J v (z ) K n (z ) = lim v . v n vn (3.12)
Of special interest to us, are the modified Bessel functions of order zero and one. Below, we discuss some special features of these functions.
3.3. Modified Bessel Functions of Order Zero and One
Modified Bessel functions of order zero and one are related to each other by the following relations:
dI 0 (z ) = I 1 (z ) dz
and
(3.13)
dK 0 (z ) = K1 (z ) dz
(3.14)
Chap. 3
10
Figure 3.1 shows these functions graphically.

4
BESSEL FUNCTION I0(z), K0(z), I1(z), K1(z)
K1 K0
2
I0 I1
1
0 0 1 2 3
Fig. 3.1 Bessel Functions of order zero and one.
For small arguments, the following asymptotic expansions may be used for the modified Bessel functions of order zero and one:5
I 0 (z ) =
( z / 2 )2 m 2 m = 0 (m!)
(3.15)
I 1 (z ) =
2m (z / 2)2 m 1
m=0
(m!)2
(3.16)
z ( z 2 )2 m K 0 (z ) = ln + I 0 (z ) 2 2 m =1 (m!)
1 1 1 + 2 + 3 + L +
1 , k
(3.17)
where = 0.5772K , and

z I ( z ) 2m ( z 2 )2 m 1 1 + K 1 (z ) = ln I 1 (z ) + 0 1+ + +L+ 2m z 2 3 (m!)2 2 m =1 (2 )
1 . k
(3.18)
Also, for large arguments, the following relations may be useful:

I v (z ) = 4v 2 1 4v 2 9 4v 2 1 4v 2 9 4v 2 1 4v 2 25 4v 2 1 + + L 1 2 3 8z 2 z 2!(8 z ) 3!(8 z ) for arg z < 2 ez
) (
)(
) (
)(
)(
)(
(3.19)
and
K v (z ) = 4v 2 1 4v 2 9 4v 2 1 4v 2 9 4v 2 1 4v 2 25 4v 2 1 e z 1 + + L . 2z 8z 2!(8 z )2 3!(8 z )3 for arg z < 3 2 .
) (
)(
) (
)(
)(
)(
(3.20)
Chap. 3
11
Based on the relations given by Eqs. 3.15 3.20, the following limiting forms may be written:
z 0
lim I 0 (z ) = 1 lim I 0 (z ) =
(3.21) (3.22) (3.23) (3.24)
z 0
lim I 1 (z ) = 0 lim I 1 (z ) =
z 0
lim K 0 (z ) = ln e z 2 lim K 0 (z ) = 0
(3.25) (3.26) (3.27) (3.28)
z 0
lim K 1 (z ) =
z 0
lim z K 1 (z ) = 1
and
z
lim K1 (z ) = 0 .
(3.29)
These relations are useful in the evaluation of the asymptotic behavior of transient pressure solutions.
Chapter 4 LAPLACE TRANSFORMATION

Integral transforms are useful in the solution of differential equations. A special form of the linear integral transforms, known as the Laplace transformation, is particularly useful in the solution of the diffusion equation. This chapter first presents the definitions and the operational properties of Laplace transforms. The applications of the Laplace transformation are demonstrated, next, by considering practical transient fluid flow problems.
4.1. Definitions and Properties The Laplace transformation of a function F (t ) , denoted by L{F (t )}, is defined by
L{F (t )} = e st F (t ) dt ,
(4.1)
where s is a number whose real part is positive and large enough for the integral in Eq. 4.1 to exist. In this chapter, we will use the notation that a bar over the function indicates the image or the Laplace transform of the function; that is,
p (s ) = L{p(t )} .
(4.2)
The following fundamental properties of the Laplace transformation are useful in the solution of common transient flow problems. a) Transforms of derivatives:
dp(x , y , z , t ) L = sp (x , y , z , s ) p(x , y , z , t = 0 ) , dt
n d p (x , y , z , t ) n n 1 n 2 dp (x , y , z , t ) L = s p (x , y , z , s ) s p (x , y , z , t = 0 ) s n dt dt t =0
(4.3)
n 3
d 2 p(x , y , z , t ) dt 2
t =0
LLLL
d n 1 p (x , y , z , t ) dt n 1
t =0
(4.4)
n n d p (x , y , z , t ) d p (x , y , z , s ) L . = n dx dx n
(4.5)
b) Transforms of integrals:
t 1 L p (t ) dt = p (s ) . s 0
(4.6)
Chap. 4
c) Substitution
LAPLACE TRANSFORMATION
13
L{p(kt )} =
1 s p , where k k
p (s ) = L{p(t )} ,
(4.7) (4.8)
L e at p (t ) = p (s + a ) .
d) Translation
L{H (t t 0 ) p(t t 0 )} = e st0 p (s ) ,
(4.9)
where H (t t 0 ) is Heavisides unit function defined by

0 H (t t 0 ) = 1 e) Convolution
t L p1 ( ) p 2 (t ) d = p1 (s ) p 2 (s ) . 0
for t < t 0 for t > t 0
(4.10)
(4.11)
For the Laplace transformation to be useful, inverse Laplace transformation must be uniquely defined. We use the symbol L1 to denote the inverse Laplace transform operator; that is
p (t ) = L1 {p (s )} .
(4.12)
In this operation, p (t ) represents the inverse (transform) of the Laplace domain function p (s ) . A uniqueness theorem of the inversion guarantees that no two functions of the class have the same Laplace transform.6 The class is defined as the set of functions F (t ) that are sectionally continuous on each bounded interval over the half line t > 0 , defined by
F (t i ) = 1 [F (t i 0) + F (ti + 0)] 2
(4.13)
at a finite number of points t i where they are not continuous, and F (t ) < Me t for any constants M and . The most rigorous technique to find the inverse Laplace transform of a Laplace domain function is the use of the inversion integral6 but its discussion is outside the scope of this chapter. For petroleum engineering applications, a simple table look-up procedure is usually the first resort. Table 4.1 shows an example table of Laplace transform pairs that may be used to find the Laplace transforms of real-space functions or the inverse Laplace transforms of the Laplace domain functions. Fairly large tables of Laplace transform pairs may also be found in Refs. 6 and 7. The relations given in the Laplace transform tables may also be extended to more complex functions by using the fundamental properties of the Laplace transforms noted above. When a simple analytical inversion is not possible, numerical inversion of a Laplace domain function is an alternate procedure. Many numerical inversion algorithms have been proposed in the literature. For the inversion of the transient flow solutions in Laplace domain, the numerical inversion algorithm suggested by Stehfest8 is the most popular algorithm.
Chap. 4
14
The Stehfest algorithm is based on a stochastic process and suggests that an approximate value, p a (T ) , of the inverse of the Laplace domain function, p (s ) , may be obtained at time t = T by
p a (T ) = ln 2 T
V p(s )
i i =1
s =i
ln 2 T
(4.14)
where
Vi = ( 1)(N 2 )+ i
min(i , N 2 ) k = (i +1) 2
k N 2 (2k )! . [(N 2) k ]! k!(k-1)!(i-k )!(2k-i )!
(4.15)
In Eqs. 4.14 and 4.15, N is an even integer. Although, theoretically, the accuracy of the inversion should increase as N tends to infinity [ p a (T ) should tend to p (T ) ], the accuracy starts decreasing because of round-off errors when N becomes large. Normally, the optimum value of N is determined as a result of a numerical experiment. As a reference, however, the range of 6 N 18 covers the most common values of N for transient flow problems. It should also be noted that the Stehfest algorithm is not appropriate for the numerical inversion of oscillatory and discontinuous functions. In these cases, a more complex algorithm due to Crump9 may be used.
Table 4.1 Laplace transform pairs.
F (t ) = L1 { f (s )} F (t )
a
(t > 0, Re s > 0)
L{F (t )} = f (s )
(t > 0, Re s > )
st
e
0
F (t )dt
a s n! s n +1
tn
(n positive integer )
tk
(k > 0)
1t e at
(k + 1) s k +1
sin at
cos at
sinh at cosh at
H (t a )
( s (s a (s s (s
(a 0) (a 0) (a 0)
1 (s a )
a s2 + a2
2
+ a2 a2 a2
2
2
) ) ) )
s
e as s
{exp[ a {exp[ a
erf a 2 t
ln t +
[ ( )]
(4t )]} t 2 (4t )]} (2t )

(
exp a s
( ) [1 exp( a s )] s
K0 a s
(ln s ) s
= 0.5772...)
Chap. 4
15
In some cases, obtaining asymptotic solutions for small and large values of time may be of interest. These asymptotic results may be obtained without inverting the full solution into the real time domain. The limiting forms of the full solution as s and s 0 correspond to the limiting forms in the time domain for short- and longtime, respectively. The inversion of the limiting forms may be easier than the inversion of the full solution.
4.2. Example Applications
We, now, consider some examples to demonstrate the use of Laplace transformation in the solution of transient flow problems.
Example 4.1: Transient flow toward a fully penetrating vertical well in an infinite homogeneous reservoir of uniform thickness h and initial pressure p i .
This problem may be formulated most conveniently in the radial coordinates. The diffusivity equation governing fluid flow in porous media is given, in radial coordinates, by 1 p 1 p , = r r r r t where p = p i p and Eq. 4.16 is in consistent or Darcy units. The initial condition is
p (r , t = 0 ) = 0 ,
(4.16)
(4.17)
which means that the pressure is uniform and equal to p i initially throughout the reservoir. The outer boundary condition for an infinite reservoir is
p (r , t ) = 0 ,
(4.18)
which physically means that for every time, t, there is a large enough distance, r, in the reservoir where the initial pressure, p i , has been preserved. The inner boundary condition depends on the production conditions. Assuming that the well is produced at a constant rate, q, for all times, we have
qB p = . r 2kh r r = rw
(4.19)
Note that the inner boundary condition given in Eq. 4.19 is simply a restatement of the flux law (Darcys Law given by Eq, 2.16) at the surface of the wellbore. Equations 4.16 4.19 define the initial boundary value problem to be solved to obtain the transient pressure distribution for the given system. Application of the Laplace transforms to Eq. 4.16 yields
1 p 1 p 2 p s + = p , r = r r r r r r 2
(4.20)
or rearranging we obtain
Chap. 4
1 p 2 p s + p = 0 . r r r 2
16
(4.21)
Note that in obtaining the right hand side of Eq. 4.20, we have used the initial condition. Similarly, Eqs. 4.18 and 4.19 are transformed into the following forms, respectively:
p (r , t ) = 0
(4.22)
and
qB p = . r 2khs r r =rw
(4.23)
Comparing Eq. 4.21 to Eq. 3.3, we recognize Eq. 4.21 as the modified Bessels equation of order zero. Then, the solution of Eq. 4.21 may be directly written from Eq. 3.11 as follows:
p (s ) = C1 I 0
s r + C2 K 0
sr .
(4.24)
The constants C1 and C 2 in Eq. 4.24 are obtained from the boundary conditions. The outer boundary condition (Eq. 4.22) indicates that C1 = 0 [because lim I 0 (x ) = , the only way to satisfy Eq. 4.22 is to set C1 = 0 ]. Therefore,
x 0
p (s ) = C 2 K 0
sr .
(4.25)
From Eqs. 4.23 and 4.25, we obtain

p = C 2 s rw K1 r r r = rw
s rw =
qB , 2khs
(4.26)
which yields
C2 =
1 qB 2kh s s rw K1
s rw
).
(4.27)
Then, the solution for the transient pressure distribution is given, in the Laplace transform domain, by
p (s ) = K0 s r qB . 2kh s s rw K 1 s rw
(4.28)
To complete the solution of the problem, Eq. 4.28 should be inverted into the real time domain. The real inversion of Eq. 4.28, however, is not available in terms of known functions. One of the options is to use Stehfests numerical inversion algorithm8 discussed above. The dashed line in Fig. 4.1 represents the numerical inversion of the solution in Eq. 4.28. Another option is to find an approximate inversion. One of these asymptotic forms is known as the linesource solution and commonly used in transient pressure analysis.
Chap. 4
1E+2
17
FINITE WELLBORE RADIUS
p, atm
1E+1
LINE SOURCE
1E+0
q = 736 s.cc/sec B = 1.25 r.cc/s.cc k = 0.05 D h = 914.4 cm c = 1.47E-4 atm-1 rw = 15.24 cm r = 15.24 cm = 1 cp = 0.15
1E-1 1E-1 1E+0 1E+1 1E+2 1E+3 1E+4 1E+5
t, sec
Fig. 4.1 Finite wellbore radius (Eq. 4.28) and line-source (Eq. 4.29 or 4.34) solutions (Example 4.1).
To obtain the line-source approximation of the solution given in Eq. 4.28, we assume that the radius of the wellbore is small compared to the other dimensions of the reservoir. Thus, if we assume rw 0 and use the relation given in Eq. 3.28, we obtain
rw 0
lim
s rw K 1
(
(
s rw = 1 .
(4.29)
Using this relation in Eq. 4.28, we obtain the line-source solution in Laplace domain as follows:
p (s ) = qB K 0 s r . 2kh s
(4.30)
The inversion of Eq. 4.30 can be accomplished by using a Laplace transform table. From Table 4.1 (or from the tables in Refs. 6 and 7), we have
L1 K 0 k s =
{ ( )}
k2 1 exp 4t 2t
for
k >0.
(4.31)
Using Eq. 4.31 and the Laplace transform property noted in Eq. 4.12, we obtain the following inversion of Eq. 4.30 in the real time domain:
r 2 dt qB . p(r , t ) = exp 4t t 4kh 0
(4.32)
Making the substitution u = r 2 (4t ) and noting the definition of the exponential integral function, Ei(x ) , given by
Ei(x ) =
e u du . u
(4.33)
we obtain the line source solution as follows:
Chap. 4
qB r2 . Ei 4kh 4t
18
p(r , t ) =
(4.34)
Figure 4.1 shows a comparison of the results computed from Eq. 4.28 (finite-wellbore radius) and Eq. 4.30 or 4.34 (line-source) for an example data set noted in the figure. The two solutions yield different results at early times but become the same at later times. In fact, it can be analytically shown that the long-time approximation of the finitewellbore radius solution (Eq. 4.28) is the same as the line-source well solution. To show this, we note that the longtime approximation of the solution in the time domain corresponds to the limiting form of the solution in the Laplace domain as s 0 . Then, using the property of the Bessel function given in Eq. 3.28, we can show that
lim p (s ) = K0 s r qB qB K 0 s r . = 2kh s lim s rw K1 s rw 2kh s
s 0
s 0
)]
(4.35)
Example 4.2: Transient flow in a closed cylindrical reservoir initially at uniform pressure p i . Constant rate production from a fully penetrating vertical well.
Fluid flow in cylindrical porous media is described by the diffusion equation in radial coordinates given by 1 p 1 p . = r r r r t The initial condition corresponding to the uniform pressure distribution equal to p i is
p (r , t = 0 ) = 0 ,
(4.36)
(4.37)
and the inner boundary condition for a constant production rate, q, for all times is
qB p = . r r 2 kh r = rw
(4.38)
The closed outer boundary condition is mathematically represented by zero flux at the outer boundary that corresponds to
p = 0. r r = re
(4.39)
The Laplace transforms of Eqs. 4.37 4.39 yield, respectively,

1 p 2 p s + p = 0 , r r r 2
(4.40)
qB p = , r r 2 khs r =rw
(4.41)
and
Chap. 4
p =0. r r = re
19
(4.42)
(Note that the initial condition given by Eq. 4.37 has been used to obtain Eq. 4.40). Because Eq. 4.40 is the modified Bessels equation of order zero, its general solution is given by:
p (s ) = C1 I 0
s r + C2 K 0
sr .
(4.43)
Using the outer boundary condition given by Eq. 4.42, we obtain

p = C1 s I 1 r r = re
s re C 2 s K 1
s re = 0 ,
(4.44)
which yields
C 2 = C1
( sr ) K ( sr )
I1
e 1 e
(4.45)
and thus
p (s ) = C1 I 0

sr +
K0
s r I1 K1
s re . s re
)(
(4.46)
Using the inner boundary condition given by Eq. 4.41 yields

p = C1 s rw I 1 r r r = rw
s rw
s rw K 1
) ( K ( sr )
s rw I 1
e 1
s re qB . = 2 khs
(4.47)
From Eqs. 4.47 and 4.45, we obtain the coefficients C1 and C 2 as follows:
C1 = qB 2kh s s rw I 1
[ ( s r )K ( s r ) I ( s r )K ( s r )]
e 1 w 1 w 1 e
K1
s re
(4.48)
and
C2 =
qB 2kh s s rw I 1
[ ( s r )K ( s r ) I ( s r )K ( s r )] .
e 1 w 1 w 1 e
I1
s re
(4.49)
Substituting C1 and C 2 given by Eqs. 4.48 and 4.49, respectively, into Eq. 4.43, yields
p (s ) =
I 0 s r K1 s re + I 1 s re K 0 s r qB . 2kh s s rw I 1 s re K1 s rw I 1 s rw K1 s re
[(
) (
) ( ) ( ) ) ( ) ( ) (
)]
(4.50)
The inverse of the solution given by Eq. 4.50 may not be found in the Laplace transform tables. van Everdingen and Hurst10 provided the following analytical inversion of Eq. 4.50 by using the inversion integral.
Chap. 4
20
p (r , t ) =
n =1
3r 4 4r 4 ln re 2r 2 1 e e 2 e rw r r r 2 J1 n e J 1 ( n )Y0 n Y1 ( n )J 0 n . r r r w w w 2 t exp n 2 rw r 2 2 e n J1 n J 1 ( n ) rw
2 re2 r 1 t r ln 2 2 rw 4 re2 rw 4 re2 rw
qB 2 2 2 2kh re rw
(4.51)
In Eq. 4.51, 1 , 2 , etc. are the roots of

re re Y1 ( n )J 1 n r J 1 ( n )Y1 n r = 0. w w
(4.52)
The solution given in Eq. 4.50 may also be inverted numerically by using the Stehfest algorithm.8 Figure 4.2 shows the result of the numerical inversion of Eq. 4.50.
1E+4
1E+3
p, atm
q = 736 s.cc/sec B = 1.25 r.cc/r.cc k = 0.05 D h = 914.4 cm c = 1.47E-4 atm-1 rw = 15.24 cm r = 15.24 cm = 1 cp = 0.15
re = 152.4 cm
re = 1524 cm
re = 15240 cm
re
oo
1E+2
1E+1
1E+0 1E-1 1E+0 1E+1 1E+2 1E+3 1E+4 1E+5 1E+6 1E+7 1E+8
t, sec
Fig. 4.2 Bounded reservoir solution (Example 4.2, Eq. 4.50). Example 4.3: Flowing wellbore pressure of a fully penetrating vertical well with wellbore storage and skin in an infinite reservoir.
Here we revisit the case in Example 4.1 and add the effect of a skin zone around the wellbore. We also assume that the production rate is specified at the surface so that the storage capacity of the wellbore should be taken into account. Before we present the initial-boundary value problem, it may be useful to define the skin factor and the surface production rate. Using van Everdingen and Hursts thin skin concept10 (vanishingly small skin zone radius), the skin factor is defined by
S= p (rw + ) p wf p r r r = rw = p (rw + ) p wf q sf B 2kh
(4.53)
Chap. 4
21
where q sf is the sandface production rate, p (rw + ) denotes the reservoir pressure immediately outside the skin zone boundary, and p wf is the flowing wellbore pressure measured inside the wellbore. Rearranging Eq. 4.53, we obtain the following relation for the flowing wellbore pressure.
p . r r = rw
p wf = p (rw + ) S r
(4.54)
When the production rate is specified at the surface, it is necessary to take into account the fact that the wellbore can store and unload fluids. Then, the surface production rate, q , is equal to the sum of the wellbore unloading rate,
q wb , and the sandface production rate, q sf ; that is,
q = q wb + q sf ,
where
q wb = 24C dp wf dt B
(4.55)
(4.56)
and
q sf = 2kh p . r B r r = rw
(4.57)
In Eq. 4.56, C is the wellbore storage coefficient. Substituting Eqs. 4.56 and 4.57 into Eq. 4.55, we obtain the following expression for the surface production rate. 24C dp wf 2kh p = q. r B dt B r r = rw (4.58)
The mathematical statement of the problem under consideration is similar to that in Example 1, except the inner boundary condition should be replaced by Eq. 4.58 and Eq. 4.54 should be incorporated to take into account the skin effect. Then, the initial-boundary value problem is defined by the following set of equations in the Laplace domain:
1 p 2 p s + p = 0 , r r r 2 p (r , t ) = 0 , q 24C 2kh p sp wf = , r B B r r = rw s
(4.59) (4.60) (4.61)
and
p . r r = rw
(4.62)
The general solution of Eq. 4.59 is
Chap. 4
p (s ) = C1 I 0
22
(4.63)
( (
(
s r + C2 K 0
sr .
The use of the condition in Eq. 4.56 requires that C1 = 0 . Therefore,
p (s ) = C 2 K 0
sr .
(4.64)
Using Eq. 4.63 in Eq. 4.62 yields
p wf = C 2 K 0
s rw + + SC 2 s rw K 1
s rw .
(4.65)
From Eqs. 4.61, 4.64, and 4.65, we obtain

24C C2 s K 0 B
[ ( s r +) + S
w
s rw K 1
s rw +
)]
1
2kh C 2 s rw K 1 B
s rw =
q , s
(4.66)
which yields
C2 = qB 2kh s s rw K1
s rw +
24C s K0 2kh
[ ( s r +) + S
w
s rw K 1
s rw
)]
(4.67)
Then, substituting Eq. 4.67 for C 2 in Eq. 4.65, we obtain the solution for the transient pressure distribution in Laplace transform domain as follows:
p wf =
qB 2kh s s rw K1
K0
24C s rw + s K0 2kh
s rw + + S s rw K1
[ ( sr
( s r ) +)+ S s r K (
w w
1
s rw
)]
(4.68)
The real inversion of the solution in Eq. 4.68 has been obtained by Agarwal et al.11 by using the inversion integral. It is also possible to numerically invert Eq. 4.68. Figure 4.3 shows the results of the numerical inversion of Eq. 4.68 by using the Stehfests algorithm.8 Also shown in Fig. 4.3 are the logarithmic derivatives of p wf . These derivatives are computed by applying the Laplace transformation property given in Eq. 4.5 to Eq. 4.68 as follows
dp wf L = sp wf . dt
(4.69)
Here, we have used p wf (t = 0) = 0 . To obtain the logarithmic derivatives, we simply note that
dp wf d ln t =t dp wf dt
(4.70)
Chap. 4
1E+3
23
p and d p/dln t, atm
q = 736 s.cc/sec B = 1.25 r.cc/r.cc k = 0.05 D h = 914.4 cm c = 1.47E-4 atm-1 rw = 15.24 cm

1E+2
PRESSURE
r = 15.24 cm C = 3.93 cc/atm S=5 = 1 cp = 0.15
1E+1
DERIVATIVE
1E+0 1E-1 1E+0 1E+1 1E+2 1E+3
t, sec
Fig. 4.3 Wellbore storage and skin solution (Example 4.3, Eq. 4.69).
Example 4.4: Pressure buildup solution with wellbore storage and skin following a drawdown period at a constant rate in an infinite reservoir.
This example is similar to the case in Example 4.3 except at time t p , the well is shut in and pressure buildup begins. The system of equations to define this problem is as follows: 1 p 1 p , = r r r r t
p (r , t = 0 ) = 0 , p (r , t ) = 0 ,
(4.71) (4.72) (4.73)
24C dp wf 2kh p = q1 H t tp , r B dt B r r = rw
)]
(4.74)
where H t t p is the Heavisides unit function (Eq. 4.10), and
p . r r = rw
(4.75)
Note that the right-hand-side of the boundary condition in Eq. 4.74 accounts for a constant surface production rate, q , for 0 < t < t p and for shut-in ( q = 0 ) for t > t p . Taking the Laplace transforms of Eqs. 4.71 4.75, we obtain
1 p 2 p s + p = 0 , r r r 2 p (r , t ) = 0 ,
(4.76) (4.77)
Chap. 4
q 24C 2kh p st sp wf = 1 e p , r B B r r = rw s
24
(4.78)
and
p . r r = rw
(4.79)
The general solution of Eq. 4.76 is
p (s ) = C1 I 0
(
(
s r + C2 K 0
sr .
(4.80)
The condition in Eq. 4.77 requires that C1 = 0 . Therefore,
p (s ) = C 2 K 0
sr .
(4.81)
From Eqs. 4.81 and 4.79, we obtain
p wf = C 2 K 0
s rw + + SC 2 s rw K 1
s rw .
(4.82)
Substituting the results in Eqs. 4.81 and 4.82 into Eq. 4.77, we have
24C C2 s K 0 B
[ ( s r +) + S
w
s rw K 1
s rw +
)]
2kh C 2 s rw K 1 B
s rw =
) q (1 e ), s
st p
(4.83)
which yields
C2 = qB 2kh s s rw K 1
s rw
24C s K0 + 2kh
(1 e )
st p
[ ( s r +) + S
w
s rw K 1
s rw
)]
(4.84)
Using Eq. 4.84, we obtain the following solution in the Laplace transform domain that covers both the drawdown and buildup periods:
p wf =
K 0 s rw + + S s rw K 1 s rw 1 e p qB 2kh 24C s s rw K 1 s rw + s K 0 s rw + + S s rw K 1 2kh
[ ( (
[ (
)] (
st
s rw
)]
(4.85)
As a remark here, we should note that the 1 e
st p
) term contributed by the discontinuity at time t = t
causes
difficulties in the numerical inversion of the right hand side of Eq. 4.85 by using Stehfests algorithm.8 As suggested by Chen and Raghavan,12 these problems may be solved by noting that
L1 f (s ) 1 e
{ (
st p
)} = L
t
{f (s )}t L1 {f (s )}t t p ,
(4.86)
and applying the Stehfests algorithm term by term to the right hand side of Eq. 4.86. Figure 4.4 shows example results obtained by the numerical inversion of Eq. 4.85.
Chap. 4
25
1E+3
1E+2
p and d p/dln t, atm
1E+1
q = 736 s.cc/sec B = 1.25 r.cc/r.cc k = 0.05 D h = 914.4 cm c = 1.47E-4 atm-1 rw = 15.24 cm r = 15.24 cm C = 3.93 cc/atm S=5 tp = 1000 sec = 1 cp = 0.15
DRAWDOWN
BUILDUP
1E+0
1E-1 1E-1 1E+0 1E+1 1E+2 1E+3 1E+4 1E+5
t, sec
Fig. 4.4 Drawdown and buildup results with wellbore storage and skin solution (Example 4.4, Eq. 4.85).
Chapter 5 GREENS AND SOURCE FUNCTIONS

Greens and source functions have been used in petroleum engineering to solve two- and three-dimensional transient flow problems that may result from complex well geometries, such as partially penetrating vertical and inclined wells, hydraulically fractured wells, and horizontal wells. Before starting the introduction of these techniques, it may be useful to clarify the nomenclature. In our terminology, a source is a point, line, surface, or volume at which fluids are withdrawn from the reservoir. Strictly speaking, fluid withdrawal should be associated with a sink and the injection of fluids should be related to a source. We, however, use the term source for both production and injection with the convention that a negative withdrawal rate indicates injection. Greens and source functions are closely related. A Greens function is defined for a differential equation with specified boundary conditions (prescribed flux or pressure) and corresponds to an instantaneous point source solution. A source function, on the other hand, is the solution of the given differential equation with specified boundary conditions and source geometry. The details of the theory and application of Greens and Source functions for the solution of transient flow problems in porous media may be found in Refs. 2 and 13 20. Here, a brief account of the use of these techniques will be presented. For this presentation, the introduction of the fundamental solution and point source concepts is useful.
5.1. Fundamental Solution of Diffusion Equation The fundamental solution, (M , M ,t , ) , of diffusion equation for fluid flow in porous media satisfies the following differential equation:
2 (M , M ,t , )
(M , M ,t , ) = (M , M ,t , ) , t
(5.1)
where (M , M , t , ) is a generalized (symbolic) function,15 called the (Dirac) delta function, defined based on its following properties:
t2
(M , M ,t , )dM d =
t1 D
1 0
for t1 < t , < t 2 and otherwise
M , M in D
(5.2)
and
t2
t1 D
(M , M ,t , ) (M , )dM d =
(M ,t ) 0
for t1 < t , < t 2 and otherwise
M , M in D
(5.3)
The delta function is symmetric in M and M and t and . In this formulation, the delta function represents the symbolic density of a unit-strength, concentrated source located at M and acting at time . In physical terms, this source corresponds to an infinitesimally small well (located at point M ) at which a finite amount of fluid is withdrawn (or injected) instantaneously (at time ). Therefore, the solution of Eq. 5.1 (the fundamental solution) is
Chap. 5
GREEN'S AND SOURCE FUNCTIONS
27
also known as the instantaneous point source solution. Formally, the point source solution corresponds to the pressure drop, p = p i p , at a point M and time t in an infinite porous medium (reservoir) because of a point source of unit-strength located at point M and acting at < t .
5.2. The Source Function Solutions of the Diffusion Equation
The point source solution was first introduced by Lord Kelvin16 for the solution of the heat conduction problems and extensively discussed by Carslaw and Jaeger.14 The point source solution is usually obtained by finding the limiting form of the pressure drop due to a spherical source as the source volume vanishes. To demonstrate the derivation of the instantaneous point source solution, let us consider the transient flow of a slightly compressible fluid of constant compressibility and viscosity toward a spherical source of radius r = a in an infinite, homogeneous, and isotropic porous medium. Because of the spherical symmetry of the physical problem, we can conveniently express the governing equation of fluid flow in porous media in spherical coordinates as follows:
2 p 2 p 1 p + = r r t r 2 for r > 0 . (5.4)
Assume that the initial pressure drop satisfies
p (r ,t = 0) =
p i 0
for 0 < r < a , for r > a
(5.5)
and we have the condition that
p (r = 0 , t ) = 0 .
Upon substitution of u = rp , Eqs. 5.4 5.6 become, respectively, u 1 2 u , = t r 2
rp i for 0 < r < a u (r , t = 0) = , for r > a 0 and
u (r = 0 , t ) = 0 .
(5.6)
(5.7)
(5.8)
(5.9)
The solution of the problem described by Eqs. 5.7 5.9 is given by14
p =
r2 r 2 rr rr r exp exp exp exp 2t dr . 2 r t 4t 0 4t 2t
p i
(5.10)
If we expand the exponential terms in the integrand in Eq. 5.10 in powers of r and neglect the powers of r higher than four, we obtain
24(t )3 2
4a 3 p i
r 2 a 2 r 2 exp 4t 1 + 40t t 6 .
(5.11)
Chap. 5
28
~ denotes the volume of the liquid released Note that 4a 3 3 = V where V is the volume of the spherical source. If q as a result of the change in the volume of the source, V , which is caused by the change in pressure, p i , then ~ = V . Also using the definition of compressibility, c = (1 V )(V p ) , we obtain q ~ = cVp . Then, we can q i i show that
~ q 4a 3 p i = . c 3 Substituting Eq. 5.12 into Eq. 5.11, we obtain,
(5.12)
8 c (t )3 2
~ q
r 2 a 2 r 2 1 + . 6 exp 4t 40t t
(5.13)
~ remaining constant, Eq. 5.13 yields the point source If we let the radius of the spherical source, a , tend to zero, q solution in spherical coordinates given by
p =
8 c (t )3 2
~ q
r2 exp 4t .
(5.14)
~ , instantaneously This solution may be interpreted as the pressure drop at a distance r due to a volume of fluid, q ~ (c ) is the strength of the source which is the withdrawn at r = 0 and t = 0 . Consistent with this interpretation, q ~ (see Eq. pressure drop in a unit volume of substance caused by the instantaneous withdrawal of a volume of fluid, q 5.12).
Instantaneous Point Source in an Infinite Reservoir: A more general solution for an instantaneous point source of ~ (c ) acting at t = in an infinite, homogeneous, but anisotropic reservoir was presented by Nisle21 as strength q follows:
p (M , M ,t ) =
8c x y z [ (t )]3 2
~ q
~ (M M )2 exp . 4(t )
(5.15)
In Eq. 5.15, M and M indicate the locations of the observation point and the source, respectively. For a three M = (x , y , z ) , M = ( x , y , z ) , and dimensional Cartesian coordinate system, ~ = (x x )2 + ( y y )2 + (z z )2 with , , and representing the diffusivity (M M )2
x y z x
constants (defined in Eq. 3.1.23) in the x , y , and z directions, respectively.

~ (t ) from time 0 Continuous Point Source in an Infinite Reservoir: If the fluid withdrawal is at a continuous rate q to t , then the pressure drop due to a continuous point source in an infinite reservoir is obtained by distributing the ~ ( ) (c ) over a time interval 0 t and is given by point sources of strength q
p (M , M ,t ) =
1 ~ q ( )S (M , M ,t ) d , c
(5.16)
~ (c ) = 1] , instantaneous point source in an infinite reservoir; where S (M , M , t ) corresponds to a unit-strength [q that is,
Chap. 5
29
S (M , M , t ) =
8 x y z [ (t )]3 2
~ (M M )2 exp . 4(t )
(5.17)
Instantaneous and Continuous Line, Surface, and Volumetric Sources in an Infinite Reservoir: Similarly, the ~ (M ) (c ) over a line, surface, or volume, , in an infinite distribution of instantaneous point sources of strength q w reservoir leads to the following solution corresponding to the pressure drop due to production from a line-, surface-, or volumetric-source, respectively.
p (M , M , t ) =
1 c
~ (M ) S (M , M ,t ) dM . q
w w w w
(5.18)
~ (M ) is the instantaneous withdrawal volume of fluids In Eq. 5.18, M w indicates a point on the source ( w ) and q w per unit length, area, or volume of the source depending on the source geometry. For example, the pressure drop due to an infinite line source at x , y and z may be obtained as follows:
p (x , x , y , y ,t ) =
8c x y z [ (t )]3 2
+
( x x )2 x + ( y y )2 y exp 4(t )
2 ~ (x , y , z ) exp (z z ) dz q 4 z (t )
(5.19)
If we assume that the flux is uniform along the line source and the source strength is unity ~ (x , y , z ) (c ) = q ~ (c ) = 1] , then we can write the infinite line source solution in an infinite reservoir as follows: [q
S ( x , x , y , y , t ) = ( x x )2 x + ( y y )2 y . exp 4(t ) 4 x y (t )
1
(5.20)
As another example, if we consider an infinite plane source at x = x , y , and z in an infinite reservoir, we can write
p (x , x ,t ) =
8c x y z [ (t )]3 2
+ +
( x x )2 exp 4 x (t )
2 2 ~ (x , y , z ) exp ( y y ) y (z z ) z dy dz q 4(t )
(5.21)
which also leads to the following uniform-flux, unit-strength, infinite plane-source solution in an infinite reservoir:
S ( x , x , t ) = (x x)2 exp . 2 x (t ) 4 x (t )
1
(5.22)
If the fluid withdrawal is at a continuous rate from time 0 to t , then the continuous line-, surface-, or volumetricsource solution for an infinite reservoir is given by
Chap. 5
t
30
1 p (M , M ,t ) = c
~ (M , )S (M , M ,t ) dM d . q
w w w
0
(5.23)
Source Functions for Bounded Reservoirs: The source solutions discussed above can be extended to bounded reservoirs. Especially when the reservoir boundaries consist of impermeable and constant pressure planes, the method of images provides a convenient means of generating the bounded reservoir solutions by using the infinite reservoir solutions. The method of images is an application of the principle of superposition which states that if f 1 and f 2 are two linearly independent solutions of a linear partial differential equation and c1 and c 2 are two arbitrary constants, then f 3 = c1 f 1 + c 2 f 2 is also a solution of the partial differential equation. Below we present examples of source functions in bounded reservoirs.
1) Instantaneous point source nearby a single linear boundary. To generate the effect of an impermeable planar boundary at a distance d from a unit-strength, instantaneous point source in an infinite reservoir (see Fig. 5.1), we can apply the method of images to the instantaneous point source solution given in Eq. 5.17 as follows:
S (M , M , t ) = 1 ( y y )2 y + (z z )2 z exp 4(t )
8 x y z [ (t )]3 2
2 (x x 2d )2 ( x x ) + exp exp ( ) ( ) t 4 4 t x x
(5.24)
It can be shown from Eq. 5.24 that (S x ) x = d = 0 ; that is, the bisector of the distance between the two sources is a no-flow boundary. Similarly, to generate the effect of a constant pressure boundary, we use the method of images and the unit strength, instantaneous point source solution (Eq. 5.17) as follows: S (M , M , t ) = 1 ( y y )2 y + (z z )2 z exp 4(t )
8 x y z [ (t )]3 2
2 ( x x 2 d )2 ( x x ) exp exp 4 x (t ) 4 x (t )
(5.25)
y
M 1 = (x , y , z )
p x = 0 M 2 = (x + 2d, y , z )
x d d
Fig. 5.1 Application of the method of images to generate the effect of a linear boundary.
2) Instantaneous point source in an infinite slab reservoir. Considering the geometry shown in Fig. 5.2 A, we can generate the solution for a unit strength, instantaneous point source in an infinite slab reservoir with impermeable boundaries at z = 0 and h as follows:
Chap. 5

( x x )2 x + ( y y )2 y exp 4(t )
31
S (M , M , t ) =
8 x y z [ (t )]3 2
2 (z + z 2nh )2 (z z 2nh ) + exp exp ( ) ( ) t 4 4 t z z n =
(5.26)
which, by using Poissons summation formula given by14

( 2n e )2 n 2 2 t t cos n , exp 1 + 2 exp = 2 4t e e e n = n =1
(5.27)
may be transformed into the following form: S (M , M , t ) = ( x x )2 x + ( y y )2 y exp 4(t ) 4h x y (t ) . 2 2 n z (t ) z z 1 + 2 exp cos n cos n h h h2 n =1 1
(5.28)
n = 2 z = 2nh
+ +
+ -
+ +
z
n = 1 z = 2nh
+ +
+ -
z=h z = z' z= 0 z = - z'
+ +
SOURCE SINK
+ -
+ +
n = 1 z = -2nh
+ +
+ -
n = 2 z = -2nh
+ +
A - NO-FLOW BOUNDARIES
+ B - CONSTANT PRESSURE BOUNDARIES
+ +
C - MIXED BOUNDARIES
Fig. 5.2 Application of the method of images to generate the solutions for infinite slab reservoirs.
Following similar lines, if the slab boundaries at z = 0 and h are at a constant pressure equal to p i , we obtain
Chap. 5

(x x )2 x + ( y y )2 y exp 2ch x y (t ) 4(t ) ~ q
32
S (M , M , t ) =
n 2 2 z (t ) z z exp sin n sin n h2 h h n =1
(5.29)
Similarly, for the case where the slab boundary at z = 0 is impermeable while the boundary at z = h is at a constant pressure equal to p i , the following solution may be derived:
S (M , M , t ) = (x x )2 x + ( y y )2 y exp 2ch x y (t ) 4(t ) ~ q
(2n 1)2 2 z (t ) z z exp cos (2n 1) cos (2n 1) 2 h 2h 2h n =1
(5.30)
3) Instantaneous point source in a closed parallelepiped. The same ideas may be used to develop solutions for reservoirs bounded by planes in all three directions. For example, if the reservoir is bounded in all three directions; that is, 0 x x e , 0 y y e , and 0 z h , and the bounding planes are impermeable, then we can use Eq. 5.17 and the method of images to write
S (M , M , t ) = 1 8 x y z [ (t )]3 2 (x x 2nx )2 (x + x 2nx )2 e e exp exp + 4 x (t ) 4 x (t ) n =
2 ( y + y 2ny e )2 ( y y 2ny e ) + exp exp ( ) ( ) t t 4 4 n = y y
2 (z + z 2nh )2 (z z 2nh ) + exp exp 4 z (t ) 4 z (t ) n =
(5.31) which, by using Poissons summation formula (Eq. 5.27), may be recast into the following form:
S (M , M , t ) =
k 2 2 x (t ) x x 1 cos k cos k 1 + 2 exp 2 xe ye h xe xe xe k =1
m 2 2 y (t ) y y 1 + 2 exp cos m cos m 2 y y ye e e m =1
(5.32)
n 2 2 z (t ) z z 1 + 2 exp cos n cos n 2 h h h n =1
4) Instantaneous infinite plane source in an infinite slab reservoir with impermeable boundaries. The instantaneous point source solutions given by Eqs. 5.28 5.30 may be extended to different source geometries by using Eq. 5.18. For example, the solution for an instantaneous infinite plane source at z = z in an infinite slab reservoir with impermeable boundaries is obtained by substituting Eq. 5.28 for S in Eq. 5.18. This yields
Chap. 5

(x x )2 x + ( y y )2 y exp 4(t ) 4h x y (t )
+ +
33
p (M ,t ) =
1 c
~ ( x , y ) q
n 2 2 z (t ) z z 1 + 2 exp cos n cos n dx dy 2 h h h n =1
(5.33)
~ (x , y ) (c ) = q ~ (c ) = 1] , we obtain the following instantaneous Assuming a unit-strength, uniform-flux source [q infinite plane source solution in an infinite slab reservoir with impermeable boundaries:
S (M , M , t ) =
n 2 2 z (t ) z z 1 cos n cos n . 1 + 2 exp 2 h h h h n =1
(5.34)
5) Instantaneous infinite slab source in an infinite slab reservoir with impermeable boundaries. Following similar lines, we can obtain the solution for an instantaneous, infinite slab source of thickness h p located at z = z w ( z w is the z-coordinate of the mid-point of the slab source) in an infinite slab reservoir with impermeable boundaries as follows: 1 p (M ,t ) = c
+ + z w + h p
zw h p
(x x )2 x + ( y y )2 y exp 4(t ) 4h x y (t ) ~ ( x , y , z ) q
n 2 2 z (t ) z z 1 + 2 exp cos n cos n dz dx dy 2 h h h n =1
(5.35)
~ (x , y , z ) (c ) = q ~ (c ) = 1] , then Eq. 5.35 yields If we assume a uniform flux slab source [q
S (M , M , t ) =
hp 4h 1 + h h p
n exp 1
n =1
n 2 2 z (t ) h2
h z sin n p cos n z cos n w . h 2h h
(5.36)
6) Uniform-flux, continuous, infinite slab source in an infinite slab reservoir with impermeable boundaries. Solutions for continuous plane and slab sources may also be obtained as indicated by Eq. 5.16 (or Eq. 5.23). For example, the solution for a uniform-flux, continuous, infinite slab source in an infinite slab reservoir with impermeable top and bottom boundaries may be obtained by substituting the right hand side of Eq. 5.36 for S in Eq. 5.16 and is given by
p (M ,t ) =
~h q p
1 + 4h ch 0 h p
n exp 1
n =1
n 2 2 z h2
h z sin n p cos n z cos n w d . h h 2h
(5.37)
Note that the same solution could have been obtained by substituting the unit strength instantaneous point source solution given by Eq. 5.28 for S in Eq. 5.23.
Example 5.1: Transient flow toward a partially penetrating vertical well of penetration length h p in an infinite,
homogeneous, slab reservoir of uniform thickness h and initial pressure p i with impermeable top and bottom boundaries. The geometry of the well-reservoir system of interest is shown in Fig. 5.3. The solution for this problem can be obtained by assuming that the well may be represented by a vertical line source. Then, starting with Eq. 5.23 and
Chap. 5
34
substituting the unit-strength, instantaneous point source solution in an infinite slab reservoir with impermeable ~ (c ) = 1 ] for S , we obtain the following solution: boundaries [Eq. 5.28 with q 1 p (M ,t ) = c
t zw +h p 2
0 zw h p
(x x )2 x + ( y y )2 y exp 4(t ) 4h x y (t ) 2 ~ (z , ) q
n 2 2 z (t ) z z 1 + 2 exp cos n cos n dz d 2 h h h n =1
(5.38)
If we assume that the strength of the source is uniformly distributed along its length (this physically corresponds to a ~ (z , ) = q ~ = q h where q is the uniform-flux distribution) and the production rate is constant over time; that is, q p constant production rate of the well, then Eq. 5.38 yields (x x )2 x + ( y y )2 y exp 4 0
t
p (M , t ) =
q 4ch x y
4h 1 + hp
n 2 2 z 1 exp n h2 n =1
h z sin n p cos n z cos n w d h h 2h
(5.39)
hp z y x
hp / 2 h zw
Fig. 5.3 Geometry of the well reservoir system for a partially penetrating vertical well in an infinite slab reservoir with impermeable boundaries (Example 5.1).
Chapter 6 THE USE OF GREENS AND SOURCE FUNCTIONS IN SOLVING UNSTEADY FLOW PROBLEMS
As discussed in the previous chapter, the conventional development of the source function solutions uses the instantaneous point source solution as the building block with the appropriate integration (superposition) in space and time. In 1973, Gringarten and Ramey13 introduced the use of the source and Greens function method to the Petroleum Engineering literature with a more efficient method of developing the source solutions. Specifically, they suggested the use of infinite plane sources as the building block with Newmans product method.22 In this Chapter, we discuss the use of Greens and source functions in solving unsteady flow problems in reservoirs as introduced by Gringarten and Ramey.13
6.1. Greens Function for Transient Flow in Porous Media The Greens function for transient flow in a porous medium is defined as the pressure at M (x , y , z ) at time t due to an instantaneous point source of unit strength generated at point M (x , y , z ) at time with < t and the porous medium being initially at zero pressure and the boundary of the medium being kept at zero pressure or impermeable to flow.13,14 If we let G (M , M , t ) denote the Greens function, then it should satisfy the diffusion equation; that is,
G = 2 G t for
(t > ) .
(6.1)
Because G is a function of t , it should also satisfy the adjoint diffusion equation,

G + 2 G = 0 for
( < t ) .
(6.2)
The Greens function also has the following properties: i)
G is symmetrical in the two points M and M ; that is, the Greens function is invariant as the source and the observation points are interchanged.
As t , G vanishes at all points in the porous medium; that is, lim G (M , M , t ) = 0 , except at the
t
ii)
source location, M = M , where it becomes infinite, so that G satisfies the delta function property
lim (M ) G (M , M ,t ) dM = (M )
(6.3)
where D indicates the domain of the porous medium and (M ) is any continuous function. iii) Because G corresponds to the pressure due to an instantaneous point source of unit strength, it satisfies
Chap. 6
THE USE OF GREEN'S AND SOURCE FUNCTIONS IN SOLVING UNSTEADY FLOW PROBLEMS
36
G(M , M ,t ) = 1 .
D
(6.4)
iv)
G or its normal derivative, G n , vanishes at the boundary, , of the porous medium. If the porous medium is infinite, then G vanishes as M or M .
Let p (M , ) be the pressure in the porous medium and G (M , M , t ) be the Greens function. Let D denote the domain of the porous medium. Then, p and G satisfy the following differential equations:
p = 2 p for M in D
(6.5)
and
G = 2 G for M , M in D .
(6.6)
Then, we can write

( pG ) = G p + p G = G 2 p p 2 G
(6.7)
or
t
0 D
( pG ) dM d =
(G
0D
p p 2 G dM d ,
(6.8)
where is a small positive number. Changing the order of integration and applying the Greens theorem,
f (M ) dM =
n dS ,
(6.9)
where D and indicate the volume and boundary surface of the domain, respectively, S denotes the points on the boundary, and n indicates differentiation in the normal direction of the surface , we obtain
( pG ) =t dM ( pG ) =0 dM =
D D 0
p G dS d . n n
(6.10)
Taking the limit as 0 and noting the delta-function property of the Greens function (Eq. 6.3), Eq. 6.10 yields
p (M , t ) =
pi (M )G (M , M ,t ) dM +
p G dS d , n n
0
(6.11)
where p i (M ) = p (M , t = 0 ) is the initial pressure at point M . In Eq. 6.11, the boundary, , of the porous medium consists of two surfaces: the inner boundary that corresponds to the surface of the wellbore, w , and the outer boundary of the reservoir, e . Then, Eq. 6.11 may be written as
Chap. 6
p p G G G dS d + G dS d . p p p (M , t ) = p i (M )G (M , M , t ) dM + n n n n 0 w 0 e D
37
(6.12)
As required by the fourth property of the Greens function noted above, if the outer boundary of the reservoir is impermeable, [ (p n )e = 0 ] or at infinity, then G vanishes at the outer boundary; that is, G ( e ) = 0 . Thus, Eq. 6.12 becomes
p (M ,t ) =
pi (M )G (M , M ,t ) dM +
p G dS d . n n
0
w
(6.13)
Similarly, if the flux, (p n ) w , is specified at the inner boundary, then the normal derivative of the Greens function, (G n )w , vanishes at that boundary. This yields
pi (M )G (M , M ,t ) dM p (M ,t ) =
G n dS d .
0
w
(6.14)
Note that if the initial pressure, p i , is uniform over the entire domain (porous medium), D , then, by the third property of the Greens function noted above (Eq. 6.4), we should have
p (M )G(M , M ,t ) dM = p .
i i D
(6.15)
Also, the flux law for porous medium (Darcys Law) requires that the volume of fluid passing through the point on the inner boundary surface, w , at time t be equal to Mw
~ (M ,t ) = k p (M w , t ) . q w n
Substitution of Eqs. 6.15 and 6.16 into Eq. 6.14 yields
(6.16)
p (M , t ) =
1 c
~ (M , ) G (M , M , t ) dS d , q
w w
(6.17)
where p (M , t ) = p i p (M , t ) . Not surprisingly, Eq. 6.17 is the same as Eq. 6.23 because G in Eq. 6.17 is the instantaneous point source solution of unit strength denoted by S in Eq. 6.23.
~ (M ,t ) , is uniformly The expression given in Eq. 6.17 may be further simplified by assuming that the flux, q w distributed on the inner boundary surface (wellbore), w . This yields
p (M ,t ) =
1 ~ q ( ) S (M , M w ,t ) d c
(6.18)
where
Chap. 6
S (M , M w ,t ) =
38
G (M , M ,t )dM .
w w
(6.19)
Note that the integration in the right hand side of Eq. 6.19 represents the distribution of instantaneous point sources over the length, area, or volume of the source (well) and S denotes the corresponding instantaneous source function. In the previous section, we discussed the conventional derivation of the source functions starting from the basic instantaneous point source solution. In 1973, Gringarten and Ramey13 suggested the use of infinite plane sources as the building block with Newmans product method.22
6.2. Newmans Product Method and Source Functions
Newmans product method22 may be stated for transient flow problems in porous media as follows:13 If a wellreservoir system can be visualized as the intersection of one- and/or two-dimensional systems, then the instantaneous source or Greens function for this well-reservoir system can be constructed by the product of the source or Greens functions for the one- and/or two-dimensional systems. For example, an infinite line-source well at x = x , y = y , and z + in an infinite reservoir may be visualized as the intersection of two infinite, (one-dimensional) plane sources; one at x = x , y + , and z + , and the other at x + ,
y = y , and z + . Then the instantaneous source function for this infinite line-source well, S (x , x , y , y , t ) , may be obtained as the product of two infinite, (one-dimensional) plane sources, given by
S ( j , j , t ) =
( j j )2 exp 2 j (t ) 4 j (t ) 1
for
j = x , y , or z ,
(6.20)
as follows:
S ( x , x , y , y , t ) = ( x x )2 x + ( y y )2 y exp ( ) 4 x y (t ) 4 t 1
(6.21)
As expected, this solution is the same as Eq. 5.20 that was obtained by integration of an instantaneous point source in an infinite reservoir. For a radially isotropic reservoir ( x = y = r ), Eq. 6.21 yields
S (r , r , , , t ) = 1 d2 exp 4 r (t ) 4 r (t )
(6.22)
where d is the distance between the line source and the observation line in the x y plane (see Fig. 6.1) and is given by
d 2 = (x x )2 + ( y y )2 = r 2 + r 2 2rr cos( ) .
(6.23)
Chap. 6
z
LINE SOURCE OBSERVATION LINE
39
'
z M ' (x',y',0)
r'
z M (x,y,0)
Fig. 6.1 Geometry of a line source in three dimensional Cartesian and radial coordinates.
Similarly, intersecting three infinite instantaneous plane sources (or a line source and a plane source), we can obtain the instantaneous point source solution in an infinite reservoir as follows:
S ( x , x , y , y , z , z , t ) = 1 8 x y z [ (t )]3 2 (x x )2 x + ( y y )2 y + (z z )2 z . exp ( ) 4 t (6.24)
Instantaneous plane sources in slab reservoirs can be generated by using the plane sources in infinite reservoirs and the method of images as discussed in Section 5.2. Similarly, the instantaneous slab sources can be obtained by integrating plane sources over the thickness of the slab source (see Section 5.2). Table 6.1, compiled from the work of Gringarten and Ramey,13 presents the basic instantaneous source functions in infinite reservoirs and Table 6.2 shows the corresponding geometries of the source-reservoir systems. The basic instantaneous source functions given in Table 6.2 may be used to construct the source functions that represent the appropriate well geometry by Newmans product method.
6.3. Asymptotic Approximations for Basic Source Functions
Gringarten and Ramey13 have also presented the approximating forms of the instantaneous linear sources and the time ranges for these approximations to be valid. The approximate solutions are very useful to obtain expressions for pressure distributions at early and late times and identify the flow regimes during the corresponding time periods. Table 6.3 presents the short- and long-time approximating forms for instantaneous linear sources and their time ranges. Below, we present some example applications of the product solution method and the derivation of the approximate solutions for pressure distributions.
6.4. Example Applications Example 6.1: Transient flow toward a partially penetrating vertical fracture of vertical penetration h p and horizontal penetration 2 x f in an infinite, homogeneous, slab reservoir of uniform thickness h and initial pressure
p i with impermeable top and bottom boundaries.
The geometry of the well-reservoir system of interest is shown in Fig. 6.2. If we approximate the fracture by a vertical plane of height h p and length 2 x f , then the corresponding source geometry may be visualized as the intersection of an infinite plane source at y = y in an infinite reservoir (Source I in Tables 6.1 and 6.2), an infinite slab source of thickness 2 x f at x = x in an infinite reservoir (Source IV), and an infinite slab source of thickness
h p at z = z w in an infinite slab reservoir of thickness h (Source VIII). Then, by Newmans product method, the appropriate source function can be obtained as follows:
Chap. 6
Table 6.1 Basic instantaneous source functions.13
40
I. Infinite Plane Source
INFINITE RESERVOIRS ( x x )2 1 exp 2 x (t ) 4 x (t )

(x x )2 x + ( y y )2 y or exp 4(t ) 4 x y (t ) 1 d 2 = ( x x )2 + ( y y )2 1 d2 exp where 4 r (t ) d 2 = r 2 + r 2 2rr cos ( ) 4 r (t ) 8 x y z [ (t )]3 2 1 (x x )2 x + ( y y )2 y + (z z )2 z exp 4(t )
II. Infinite Line Source
III. Point Source IV. Infinite Slab Source
V. Closed Boundaries VI. Constant Pressure Boundaries VII. Mixed Boundaries
x f + ( x x ) x f ( x x ) 2 1 + erf exp u 2 du erf where erf (z ) = 2 ( ) ( ) 2 t 2 t x x 0 INFINITE PLANE SOURCES IN INFINITE SLAB RESERVOIRS n 2 2 (t ) 1 n x nx x cos cos 1 + 2 exp 2 xe xe xe xe n =1
(
z
2 xe 2 xe
n 2 2 (t ) nx nx x sin exp sin 2 xe xe xe n =1
VIII. Closed Boundaries IX. Constant Pressure Boundaries X. Mixed Boundaries

8
(2n + 1)2 2 (t ) (2n + 1)x cos (2n + 1)x x cos exp 2 xe xe xe n =1 INFINITE SLAB SOURCES IN INFINITE SLAB RESERVOIRS n 2 2 (t ) 2x f nx f 2xe 1 nx nx x exp sin cos cos 1 + 2 x e x f n =1 n xe xe xe xe
n 2 2 (t ) nx f 1 nx nx x sin exp sin sin 2 xe xe xe n =1 n xe 4
(2n + 1)2 2 (t ) (2n + 1)x f (2n + 1)x cos (2n + 1)x 1 x sin exp cos 2 2xe xe xe n =1 2n + 1 xe
Table 6.2 Source geometries for basic instantaneous source functions.13
Chap. 6
INFINITE RESERVOIRS II. Infinite Line Source
x=0 x' x
41
I. Infinite Plane Source

x=0 x' x
IV. Infinite Slab Source

x=0 x' x
2xf
INFINITE PLANE SOURCES IN INFINITE SLAB RESERVOIRS V. Closed Boundaries VI. Constant Press. Boundaries VII. Mixed Boundaries
x=0 x' x xe
x=0
x'
xe
x=0
x'
xe
INFINITE SLAB SOURCES IN INFINITE SLAB RESERVOIRS VIII. Closed Boundaries IX. Constant Press. Boundaries X. Mixed Boundaries
x=0 x' x xe
x=0
x'
xe
x=0
x'
xe
2xf
2xf
2xf
Chap. 6
Table 6.3 Approximating forms for instantaneous linear source functions.13
42
SOURCE TYPE V, VI, VII. Plane Source Slab Reservoir

IV, VIII, IX, X. Slab Source VIII, IX, X. Slab Source in Slab Reservoir V, VI, VII. Plane Source in Slab Reservoir VIII, IX, X. Slab Source in Slab Reservoir
SPACE LIMITS OR BOUNDARY CONDITIONS
APPROXIMATING FORMS EARLY TIMES

(x x)2 exp 2 x (t ) 4 x (t ) 0 1 12 1
x f + ( x x ) x f ( x x ) 1 + erf erf 2 2 x (t ) 2 x (t ) LATE TIMES 1 xe
TIME LIMITS
All x Outside the Source Inside the Source On the Source Boundary
t 2 (20 x ) *
t x2 f
(20 x )
EARLY AND INTERMEDIATE TIMES All Boundary Conditions
Closed Boundaries Constant Pressure Boundaries Mixed Boundaries Closed Boundaries Constant Pressure Boundaries Mixed Boundaries
2 t 5 xe 2 x 2 t 5 xe 2 t 20 x e 2 t 5 xe 2 t 5 xe 2 t 20 x e 2
0 0
2x f xe
0 0
* = nearest distance to the boundary
( ) ( ) ( ) ( ) ( ) ( )
x
2 x
2 2 x x
Chap. 6
x + ( x x ) x f ( x x ) ( y y )2 1 erf f exp + erf 4 (t ) 2 ( ) ( ) 2 y (t ) 2 t 2 t y x x . 2 2 hp n h ( ) n z n t 4 h 1 n z p w z exp cos cos sin 1 + 2 h h n h 2 h h h = 1 n p (6.25) S ( x , x , y , y , z , z w , t ) = 1
43
xf hp z y x hp / 2 h zw
Fig. 6.2 Geometry of the well reservoir system for a partially penetrating vertical fracture in an infinite slab reservoir with impermeable boundaries (Example 6.1).
~ (t )2 x h and using Eq. 6.25 for the source function, S , in Assuming that the production is at a constant rate, q = q f p
Eq. 6.18, we obtain
p (x , y , z , t ) =
q 8cx f h y
x + ( x x ) x f ( x x ) ( y y )2 erf f exp + erf 4 y 2 x 2 x 0

t
4h 1 + h p
n 2 2 z 1 exp h2 n =1 n
nh p nz w nz d sin cos cos h h 2h
(6.26)
Note that if the fracture penetrates the entire thickness of the reservoir (Fig. 6.3); that is, h p = h , then Eq. 6.26 yields
p ( x , y , t ) =
q 8cx f h y
x + ( x x ) x f (x x ) d ( y y )2 erf f . + erf exp 4 2 2 0 y x x

t
(6.27)
xf z y x
Fig. 6.3 A fully penetrating vertical fracture in an infinite slab reservoir with impermeable boundaries.
Chap. 6
44
The fully-penetrating fracture solution given in Eq. 6.27 could also be obtained by constructing the source function as the product of an infinite plane source at y = y in an infinite reservoir (Source I in Tables 6.1 and 6.2) and an infinite slab source of thickness 2 x f at x = x in an infinite reservoir (Source IV) and then using this source function in Eq. 6.18. Figure 6.4 presents examples of transient pressure responses computed from Eq. 6.27. To obtain the results shown in Fig. 6.4, numerical integration has been used to evaluate the right hand side of Eq. 6.18.
1E+3
x = xw = 0 y = yw = 0 xf = 20000 cm q = 736 s.cc/sec k = 0.01 D c = 1.47E-4 atm-1 B = 1.16 r.cc/s.cc h = 914.4 cm = 0.15 = 1.43 cp
PRESSURE
p and d p/dln t, atm.
1E+2
DERIVATIVE
1E+1
1E+0
1E-1 1E+1 1E+2 1E+3 1E+4 1E+5 1E+6 1E+7 1E+8
t, sec.
Fig. 6.4 Transient pressure responses of a uniform-flux fracture computed from Eq. 6.27.
It is also of interest to obtain an early-time approximation for the solution given in Eq. 6.27. Substituting the earlytime approximating forms for the slab sources in an infinite reservoir (approximations given in Table 6.3 for Source Functions IV and VIII), we obtain
S ( x , x , y , y , z , z w , t ) = ( y y )2 exp , 2 y (t ) 4 y (t )
(6.28)
where
0 = 1 2 and 0 = 1 2
for for for
z zw > h p z zw < h p z zw = h p (6.29)
for for for
x x > x f x x < x f . x x = x f
(6.30)
~ (t )2 x h , and substituting the source function given by Eq. 6.28 in Eq. Assuming a constant production rate, q = q f p
6.18, we obtain
2 y y p ( x , y , z , t ) = y t exp y erfc 2 4 y t 2 y t 2cx f h p y q
(6.31)
Chap. 6
45
where erfc(z ) is the complementary error function defined by

erfc (z ) = 1 erf (z ) = 2 exp ( u )du .
2 x
(6.32)
Example 6.2: Transient flow toward a uniform-flux horizontal well of length Lh located at (x , y , z w ) in a closed, homogeneous, rectangular parallelepiped of dimensions 0 x x e , 0 y y e , 0 z h and of initial pressure pi .
The sketch of the horizontal-well-reservoir system considered in this example is shown in Fig. 6.5. If we approximate the horizontal well by a horizontal line source of length Lh , then the resulting source-reservoir system may be visualized as the intersection of three sources: i) an infinite plane source at y = y in an infinite slab reservoir of thickness y e with impermeable boundaries (Source V in Tables 6.1 and 6.2), ii) an infinite slab source of thickness Lh at x = x in an infinite slab reservoir of thickness x e with impermeable boundaries (Source VIII), and iii) an infinite plane source at z = z w in an infinite slab reservoir of thickness h with impermeable boundaries (Source V). Then, by Newmans product method, the appropriate source function can be obtained as follows:
Lh
h zw z
ye y' y x x'
xe
Fig. 6.5 A horizontal well in a closed rectangular parallelepiped (Example 6.2).

S ( x , x , y , y , z , z w , t ) = Lh xe 4x e 1 + L h
1 ye
m 2 2 y (t ) m y my cos exp cos 1 2 + 2 y y ye e e m =1
k exp 1
k =1
k 2 2 (t ) k L h k x kx x sin cos cos 2 x x x 2 xe e e e
(6.33)
n 2 2 z (t ) nz w nz 1 exp + cos 1 2 cos 2 h h h h n =1
~ (t )L , and using Eq. 6.33 for the source function, S , in Eq. Assuming that the production is at a constant rate, q = q h 6.18, we obtain
p ( x , y , z , t ) =
4x 1 + e L h
q cx e y e h 1
m 2 2 y (t ) my my cos exp cos 1 2 + 2 y y ye e e m =1 0 t
k exp
k =1
k 2 2 (t ) kLh kx k x x sin cos cos 2 x x x 2 xe e e e
(6.34)
n 2 2 z (t ) nz w nz cos 1 + 2 exp cos d 2 h h h n =1
Chap. 6
46
Table 6.4 presents the pressure responses for a uniform-flux horizontal well in a closed square computed from Eq. 6.34. We may obtain a short-time approximation for the above solution by using the early-time approximations given in Table 6.3 for Source Functions V and VIII. This yields
p ( x , y , z , t ) =
( y y )2 y + (z z w )2 z , Ei 4c y z Lh 4 t q
(6.35)
where Ei ( x ) is the exponential integral function defined by Eq. 4.33. Eq. 6.35 indicates that the early-time flow is radial in the y z plane around the axis of the horizontal well. This solution corresponds to the time period during which none of the reservoir boundaries influence the pressure response. It is also possible to obtain another approximation for Eq. 6.34 that covers the intermediate-time flow behavior. If we approximate the source function in the x direction (Source Function VIII) by its early- and intermediate-time approximation and the Source Function in the y direction (Source Function V) by its early-time approximation given in Table 6.3, we obtain:
p (x , y , z ,t ) =
( y y )2 exp 4 y cLh h 0 4 y q
n 2 2 z 1 + 2 exp h2 n =1
nz w nz d cos cos h h
x + ( x x ) x f ( x x ) erf f + erf 2 2 x x .
(6.36)
This approximation should correspond to the time period during which the influence of the top and/or bottom boundaries may be evident but the lateral boundaries in the x and y directions do not have an influence on the pressure response. Note that this solution could have been obtained also by assuming a laterally infinite reservoir. In this case, the source function would have been constructed as the product of three source functions: i) an infinite plane source at y = y in an infinite reservoir (Source I in Tables 6.1 and 6.2), ii) an infinite slab source of thickness Lh at x = x in an infinite reservoir (Source IV), and iii) an infinite plane source at z = z w in an infinite slab reservoir of thickness h with impermeable boundaries (Source V).
Chap. 6
Table 6.4 Uniform-flux horizontal well in a closed square
L k kz
47
(2h ) = 10 ,
z w h = 0.5 , rw h = 2 10 3
2kh
2 = 3 x y z , k = 3 k x k y k z , t AD = t x e , x eD = 2 x e k x k Lh
t AD x eD = 2 0.2505 0.2684 0.2801 0.2981 0.3191 0.3309 0.3774 0.4130 0.4695 0.5359 0.5733 0.7202 0.8328 1.0124 1.2315 1.3646 1.9993 2.6274 3.8842 5.7692 7.0254 13.309 19.592 32.159 51.012 63.578 x eD = 4 0.2897 0.3191 0.3416 0.3774 0.4194 0.4430 0.5359 0.6071 0.7196 0.8485 0.9187 1.1704 1.3372 1.5658 1.8078 1.9461 2.5846 3.2131 4.4694 6.3544 7.6115 13.895 20.179 32.745 51.594 64.162
p
x eD = 10 0.3961 0.4695 0.5258 0.6152 0.7196 0.7775 0.9945 1.1450 1.3544 1.5617 1.6636 1.9904 2.1866 2.4380 2.6917 2.8325 3.4732 4.1011 5.3578 7.2428 8.4995 14.783 21.067 33.635 52.479 65.048 x eD = 20 0.5733 0.7196 0.8293 0.9945 1.1704 1.2605 1.5617 1.7482 1.9904 2.2178 2.3268 2.8695 2.8695 3.1254 3.3803 3.5213 4.1619 4.7889 6.0470 7.9318 9.1887 15.472 21.755 34.322 53.171 65.738
1.0E-4 2.0E-4 3.0E-4 5.0E-4 8.0E-4 1.0E-3 2.0E-3 3.0E-3 5.0E-3 8.0E-3 1.0E-2 2.0E-2 3.0E-2 5.0E-2 8.0E-2 1.0E-1 2.0E-1 3.0E-1 5.0E-1 8.0E-1 1.0E+0 2.0E+0 3.0E+0 5.0E+0 8.0E+0 1.0E+1
Chapter 7 THE USE OF SOURCE FUNCTIONS IN THE LAPLACE DOMAIN TO SOLVE UNSTEADY FLOW PROBLEMS
In Chapter 6, we discussed the application of Greens and source functions in the solution of transient flow problems in porous media. The utility of the transient flow solutions is greatly enhanced if their Laplace transforms are available. For example, the application of Duhamels theorem24 to generate variable rate solutions is usually a simple task in the Laplace transform domain. Obtaining the Laplace transforms of the Greens and source functions developed in the time domain, however, is not always an easy task. The problems arise mainly because of the use of the product method solution. For a specific class of functions, Chen et al.25 have presented a technique that may be used to apply the Laplace transformation to the product solution technique. For a more general procedure to develop source function solutions in the Laplace transform domain, however, the product solution technique should be avoided.20 Ozkan and Raghavan18,19 have shown that it is more convenient to develop source function solutions in the Laplace transform domain if the point source solution is used as a building block. Then, other source geometries may be obtained by the superposition (integration) of the point sources along the length, surface, or volume of the source. In this chapter, the derivation of the source function solutions in the Laplace transform domain is discussed.
7.1. Motivations to Derive the Source Function Solutions in the Laplace Transform Domain There are many advantages of developing transient flow solutions in the Laplace transform domain. For example, in the Laplace transform domain, Duhamels theorem24 provides a convenient means of developing transient flow solutions for variable rate production problems by using the solutions for the corresponding constant rate production problem. Duhamels theorem sates that if p and p c denote the pressure drawdown corresponding to the variable production rate q (t ) and constant production rate q c , respectively, then
1 p (M , t ) = qc
q( )
0
p c (M ,t ) d .
(7.1)
Application of the Laplace transformation converts the convolution integral in Eq. 7.1 to an algebraic expression and the Duhamels theorem is given in the Laplace transform domain as follows:
p (M , s ) = s
q (s ) qc
p c (M , s ) .
(7.2)
The simplicity of the expression given in Eq. 7.2 explains our interest in obtaining transient flow solutions in the Laplace transform domain. Another example to explain the convenience of the Laplace domain solutions is for the naturally fractured reservoirs. Common transient flow models of naturally fractured reservoirs lead to the following differential equation in radial coordinates in the Laplace transform domain2
d 2 dr rD D 1 d p f r2 D drD sf (s )p = 0 , f
(7.3)
Chap. 7
THE USE OF SOURCE FUNCTIONS IN THE LAPLACE DOMAIN TO SOLVE UNSTEADY FLOW PROBLEMS
49
where the subscript f indicates the fracture property. In Eq. 7.3, the Laplace transformation is with respect to a dimensionless time defined by
t D = t cl 2
(7.4)
and
rD = r l .
(7.5)
The naturally fractured reservoir function f (s ) is a function of matrix and fracture properties and depends on the model chosen to represent the naturally fractured reservoir.2 The corresponding differential equation for a homogeneous reservoir is obtained by setting f (s ) = 1 and is given by
d 2 dr rD D 1 d p f r2 D drD s p = 0 . f
(7.6)
Note that the general solutions for Eqs. 7.3 and 7.6 are given, respectively, by
p f (s ) = AI 0 sf (s )rD + BK 0 sf (s )rD
and
(7.7)
p(s ) = AI 0 s rD + BK 0 s rD
Then, it may be shown that
s p f [sf (s )] = [sp (s )]s = sf (s )
(7.8)
(7.9)
where the right hand side of Eq. 7.9 indicates the substitution of sf (s ) for s in sp (s ) . This discussion demonstrates that it is possible to derive transient flow solutions for naturally fractured reservoirs by following the same lines as those for the homogeneous reservoirs. Furthermore, if the solution for the corresponding homogeneous reservoir system is known in the Laplace transform domain, then the solution for the naturally fractured reservoir problem may be directly obtained from Eq. 7.9. The above examples explain the advantages of obtaining transient flow solutions in the Laplace transform domain. Obtaining the Laplace transforms of the Greens and source function solutions developed in the time domain by using the methods explained in the previous sections, however, usually pose a difficult problem. The problems arise mainly because of the use of the product method solution. For a specific class of functions, Chen et al.25 have presented a technique that may be used to apply the Laplace transformation to the product solution technique. For a more general procedure to develop source function solutions in the Laplace transform domain, however, the product solution technique should be avoided.20 Ozkan and Raghavan18,19 have shown that it is more convenient to develop source function solutions in the Laplace transform domain if the point source solution is used as a building block. Then, other source geometries may be obtained by the superposition (integration) of the point sources along the length, surface, or volume of the source.
7.2. Point Source Solution in the Laplace Domain Let us consider the flow of a slightly compressible fluid in an infinite, naturally fractured reservoir. We can use the double-porosity model suggested by Warren and Root26 to develop the governing flow equations for naturally
Chap. 7
50
fractured reservoirs. The results, however, will be applicable to the model suggested by de Swaan-O27 by a simple modification. Flow around a point source in an infinite porous medium may be conveniently expressed in spherical coordinates. The differential equations governing flow in a naturally fractured reservoir are given in spherical coordinates by
2 r rD D 1 and
2 p f rD rD
p f p m = + (1 + ) t t D D
(7.10)
(p m p f ) = (1 )
p m . t D
(7.11)
In the above equations, the subscripts f and m indicate the property of the fracture and matrix systems, respectively, and the initial pressure, p i , is assumed to be uniform in the entire system; that is, p fi = p mi = pi . The definitions of the other variables used in Eqs. 7.10 and 7.11 are as follows:
t D = t l 2
(7.12)
where l is a characteristic length in the system,
=kf
{[(Vc )
t f
+ (Vc t )m
t f
]}
]
(7.13) (7.14) (7.15)
= (Vc t ) f
[(Vc )
+ (Vc t )m ,
= l2km k f ,
and
rD =
2 2 2 , xD + yD + zD
(7.16)
where
D = k f kf l
= x , y , or z .
(7.17)
The initial and outer boundary conditions are given, respectively, by
p j (t D = 0,rD > 0) = 0
and
(7.18)
p j (rD ,t D ) = 0 .
(7.19)
~ at t = 0 from The inner boundary condition corresponding to the instantaneous withdrawal of an amount of fluid q a point source is obtained by considering the mass balance on a small sphere. If we require that at any time t = T > 0 , the sum of the flux through the surface of a small sphere around the source location must equal the ~ , instantaneously withdrawn from the sphere at t = 0 , we can write28 volume of the fluid, q
Chap. 7
T
51
lim 4k f l r 2 p f 0 + D rD 0
dt = q ~. rD =
(7.20)
Although the withdrawal of fluids from the sphere is instantaneous, the resulting flow in the porous medium and thus the flux across the surface of the sphere is continuous. Therefore, if q represents the total flux across the surface of the small sphere during the time interval 0 t T , then the mass balance requires that the cumulative production (flux across the surface of the small sphere) at time T be equal to the instantaneous withdrawal volume of fluid from the sphere; that is,
~ = q(t ) dt . q
T
(7.21)
For the condition expressed in Eq. 7.21 to hold for every T 0 , we must have
~ (t ) , q (t ) = q
(7.22)
where (t ) is the Dirac delta function satisfying the properties expressed by Eqs. 7.20 and 7.21. Using the results given by Eqs. 7.21 and 7.22 in Eq. 7.20, we obtain:
p f 4k f 2 l rD 0 + rD lim ~ (t ) . = q rD =
(7.23)
The Laplace transformation of Eqs. 7.10, 7.11, 7.19, and 7.23 yields, respectively,
d r 2 d p f D 2 rD drD drD
1 sf (s )p = 0 , f
(7.24)
where
f (s ) = [s (1 ) + ] [s (1 ) + ] ,
(7.25) (7.26)
p f (rD ) = 0 ,
and d p f 2 lim 4l 3 rD 0 + drD ~ q = . ( ) ( ) V c + V c t f t m rD =
(7.27)
In deriving the above results, we have used the initial condition given by Eq. 7.18 and noted that
L{ (t )} = exp ( st D ) (t ) dt D =
l2
s exp l
0
(t ) dt =
l2
(7.28)
Chap. 7
52
~ (Vc ) + (Vc ) represents the strength of the source for the naturally fractured porous In Eq. 7.27, the term q t f t m
media. The solution of Eqs. 7.24, 7.26, and 7.27 yields the following solution for the pressure distribution in the reservoir ~ (Vc ) + (Vc ) except at the source location (the origin) due to an instantaneous point source of strength q t f t m
acting at t = 0 :
p f =
~ q
exp sf (s ) rD
4l rD
3
(Vc t ) f + (Vc t )m
].
(7.29)
If the source is located at x D , y D ,z D , then by translation, we can write

~ S (R , s ) (Vc ) + (Vc ) , p f = q D t f t m
(7.30)
where
S (R D , s ) = exp sf (s ) R D
] (4l R ) ,
3 D
(7.31)
and
RD =
(x D x D )2 + ( y D y D )2 + (z D z D )2
(7.32)
The instantaneous point source solution for an infinite naturally fractured reservoir represented by the model suggested by Warren and Root26 is also valid for the model suggested by de Swaan-O27 provided that the appropriate f (s ) function is invoked. To obtain the solution for a homogeneous reservoir, f (s ) should be set to unity, V f = 1 , and V m = 0. If we consider continuous withdrawal of fluids from the point source, then by the principle of superposition, we should have
p f =
(Vc t ) f + (Vct )m
2 ~ ( ) S (R , t ) d = l q D D kf 0
tD
~ ( q
0
) S (R D ,t D D ) d D .
(7.33)
The Laplace transformation of Eq. 7.33 yields the following continuous point source solution in an infinite reservoir:
p =
~ exp u R q D 4kl RD
),
(7.34)
where we have dropped the subscript f and defined

u = sf (s ) .
(7.35)
7.3. Line, Surface, and Volumetric Source Solution in the Laplace Domain
The point source solution in the Laplace domain may be used to obtain the source solutions for different source geometries. If we define
Chap. 7
~ , yD yD ,zD z G (x D x D D , s ) = p p q ,
53
(7.36)
where p p represents the appropriate point source solution, then, by the application of the superposition principle, the solution for the withdrawal of fluids from a line, surface, or volume, w , is given by
p =
~ (x q
w
wD
, y wD , z wD ) G (x D x wD , y D y wD , z D z wD ) dM w .
(7.37)
If we assume a uniform-flux distribution in time and over the length, surface, or volume of the source, then we can write
p =
~ q s
G(x D x wD , y D y wD , z D z wD ) dM w .
(7.38)
The following presentation of the source function approach in Laplace domain assumes that the flux distribution is ~=q ~ s . Also, the constant production rate from the length, area, or the volume of the source, , is uniform and q w ~ denoted by q so that q = q w . Our discussions so far only considered sources in infinite reservoirs. These solutions may be easily extended to bounded reservoirs. In the following sections, we present some useful solutions for transient flow problems in bounded porous media. The first group of solutions is for laterally infinite reservoirs bounded by parallel planes in the vertical direction (infinite slab reservoirs). The second and the third groups comprise the solutions for cylindrical and rectangular reservoirs, respectively.
7.4. Solutions for Infinite Slab Reservoirs
In this section, we consider one of the most common reservoir geometries used in pressure transient analysis of wells in porous media. It is assumed that the lateral boundaries of the reservoir are far enough not to influence the solution during the time period of interest. The top and bottom boundaries of the reservoir at z = 0 and z = h are parallel planes and may be of impermeable, constant pressure, or mixed type. Tables 7.1 A 7.1 C present the solutions for the most common well geometries (point source, vertical, fractured, and horizontal wells) in infinite slab reservoirs. Below, we briefly discuss the derivation of these solutions.
Table 7.1 A Laplace Domain Solutions for Wells in Infinite Slab Reservoirs17 Impermeable Boundaries at z D = 0 and h D ~ 2 ~ x D = x D x wD k k x , ~ y D = ( y D y wD ), ~ x f = x f l , Lh = Lh (2l ) , n = u + n 2 2 h D
WELL TYPE Point Source ~ q=q Vertical Well ~h q=q Vertical Well Partial Penetration ~h q=q w
~ q ~ q
PRESSURE DISTRIBUTION, p
z z + 2 K r u K 0 (rD n ) cos n cos n 0 D h h 2klh D s n =1 ~ q h K 0 rD u 2klh D s
4h h w K 0 rD u + 2klh D s
n K
1
n =1
(rD n ) sin n hw cos n

2h
z h
cos n
z h
Chap. 7
~h q
+~ xf
54
Fractured Well ~hx q = 2q f Fractured Well Partial Penetration ~h x q = 2q w f

~ q
2kh D s ~ xf
~2 ~2 K0 xD + yD u d
xf +~ 4h ~2 ~2 h K x + yD u d + w ~ 0 D 2kh D s x f
n sin n 2h cos n h cos n h

1 hw z
n =1
+~ xf
~ xf
2 2 ~ ~ K 0 x D + y D n d
~
Horizontal Well ~L q=q h
+ Lh + Lh z z 2 2 ~ ~ ~2 ~2 K cos n cos n K0 x D + y D n d 0 x D + y D u d + 2 ~ ~ 2kh D s h h L n =1 h Lh
~ q
Table 7.1 B Laplace Domain Solutions for Wells in Infinite Slab Reservoirs17 Constant Pressure Boundaries at z D = 0 and h D ~ 2 ~ x D = x D x wD k k x , ~ y D = ( y D y wD ), ~ x f = x f l , Lh = Lh (2l ) , n = u + n 2 2 h D
WELL TYPE Point Source ~ q=q Vertical Well Partial Penetration ~h q=q w Fractured Well Partial Penetration ~h x q = 2q w f Horizontal Well ~L q=q h
~ q
klh D
~h 2q
K s
n =1
(rD n ) sin n
z h
sin n z h
z h sin n z h
k lh D
2
n K s
1
n =1
(rD n ) sin n hw
2h
z z
sin n
+~ xf
~h 2q
kh D
2
n sin n 2h sin n h sin n h s

1 hw
n =1
xf ~
~2 ~2 K0 xD + yD n d
~ q
kh D
sin n h sin n h K s
z
n =1 ~ Lh
~ + Lh
2 2 ~ +~ xD yD n d
Table 7.1 C Laplace Domain Solutions for Wells in Infinite Slab Reservoirs17 Impermeable Boundary at z D = 0 and Constant Pressure Boundary at z D = h D ~ 2 ~ x D = x D x wD k k x , ~ y D = ( y D y wD ), ~ x f = x f l , Lh = Lh (2l ) , 2 n 1 = u + (2n 1)2 2 h D
WELL TYPE Point Source ~ q=q Vertical Well Partial Penetration ~h q=q w
~ q
klh D
~h 4q
K s
n =1 0
(rD 2n 1 ) cos(2n 1) z cos(2n 1) z

2h
2h
klh D
2
(2n 1) K s
1
n =1
(rD 2n 1 ) sin(2n 1) hw cos(2n 1) z sin(2n 1) z

2h 2h
2h
Chap. 7
~h 4q
55
z
2h
Fractured Well Partial Penetration ~h x q = 2q w f
kh D
2
(2n 1) sin(2n 1) 4h s
1
n =1 +~ x
h w
cos (2n 1)
z
2h
sin(2n 1)
Horizontal Well ~L q=q h

~ q
xf ~
~2 ~2 K0 x D + y D 2 n 1 d
kh D
cos(2n 1) 2h cos(2n 1) 2h K s
n =1 ~ Lh
~ + Lh
2 2 ~ +~ xD yD 2 n 1 d
Let us consider a point source in an infinite slab reservoir with impermeable boundaries at the bottom, z = 0 , and top, z = h . To obtain the point source solution for this case, we use the point source solution in an infinite reservoir given by Eq. 7.34 with the method of images. The result is given by
p =
exp 4kls
n =
~ q
2 2 + zD u rD 1
2 2 2 2 + zD rD 1 + exp u rD + z D 2
2 2 rD + zD 2 ,
(7.39) where
2 2 2 rD = (x D x D ) + (y D y D) ,
(7.40) (7.41) (7.42)
z D1 = z D z wD 2nh D , z D 2 = z D + z wD 2nh D ,
and
hD = h k k z l .
(7.43)
The solution given in Eq. 7.39 is not very convenient for computational purposes. To obtain a computationally convenient form of the solution, we use the summation formula given by28,17
exp a 2 + ( 2n e )2 = 1 2 2 e a + ( 2n e ) n =
+
2 2 K a + 2 K a + n 0 0 e2 n =1
( )
cos n e
(7.44)
and recast Eq. 7.39 as follows:
p =
2 2 K r u + 2 K r u + n 0 D 0 D 2 hD 2klh D s n =1 ~ q
cos n z D cos n z D . h h D D
(7.45)
The point source solutions for infinite slab reservoirs with constant pressure and mixed boundaries at the top and bottom are obtained by following similar lines17 and are given in Tables 7.1 B and 7.1 C, respectively. The point source solutions can be used with Eqs. 7.36 and 7.38 to generate the solutions for the other well geometries given in Tables 7.1 A 7.1 C. For example, to generate the solution for a partially penetrating vertical line-source well of length h w in an infinite slab reservoir with impermeable slab boundaries, we integrate the right hand side of Eq.
Chap. 7
56
7.45 from z w h w 2 to z w + h w 2 with respect to z where z w is the vertical coordinate of the mid-point of the open interval. If h w = h , that is the well penetrates the entire thickness of the slab reservoir, then this procedure yields the solution for a fully penetrating vertical line-source well. The solution for a partially penetrating fracture of height h w and half-length x f is obtained if the point source solution is integrated once with respect to z from
z w h w 2 to z w + h w 2 and then with respect to x from x w x f to x w + x f where x w and z w are the
coordinates of the mid-point of the fracture. Similarly, the solution for a horizontal line source well of length Lh is obtained by integrating the point source solution with respect to x from x w Lh 2 to x w + Lh 2 where x w is the x -coordinate of the mid-point of the horizontal well.
7.5. Solutions for Cylindrical Reservoirs
Solutions for cylindrical reservoirs may also be obtained by starting from the point source solution in the Laplace , , z transform domain. The Laplace-domain solution for a point source located at rD D should satisfy the following 19 diffusion equation in cylindrical coordinates:
2 2 r p + 1 p + p u p = 0 , D 2 2 rD rD rD rD 2 z D
(7.46)
where
2 2 . rD = x D + yD
(7.47)
The point source solution is also required to satisfy the following flux condition at the source location ( rD 0+ , = , z D = z D ):
2k l lim lim D 0 + rD 0 + D
z D + D 2 z D
rD
2
~ q = . dz D rD s =
(7.48)
Assuming that the reservoir is bounded by a cylindrical surface at rD = reD and by the parallel planes at z D = 0 and h D , we should also impose the appropriate physical conditions at these boundaries. We seek a solution of the above problem in the following form:
p = P + Q .
(7.49)
In Eq. 7.49, P is a solution of Eq. 7.46 that satisfies Eq. 7.48 and the boundary conditions at z D = 0 and h D . Then, P may be chosen as one of the point source solutions in an infinite slab reservoir given in Tables 7.5 A 7.5 C, depending on the conditions imposed at the boundaries at z D = 0 and h D . If Q is chosen such that it satisfies the boundary conditions at z D = 0 and h D , its contribution to the flux vanishes at the source location, and P + Q satisfies the appropriate boundary condition at rD = reD , then Eq. 7.49 should yield the point source solution for a cylindrical reservoir with appropriate boundary conditions. As an example, let us consider the case where the boundary conditions of the cylindrical reservoir are given by
( p z )
D z D =0 ,zeD
= 0,
(7.50)
Chap. 7
57
and
( p r )
D rD = reD
=0;
(7.51)
that is, we consider a closed cylindrical reservoir. According to the boundary condition given by Eq. 7.50, we should choose P as the point source solution given in Table 7.1 A (or by Eq. 7.51). Then, using the addition theorem for the Bessel function K 0 (aR D ) given by14
+ ) cos k ( ); I k (arD )K k (arD k = K 0 (aR D ) = + )K k (arD ) cos k ( ); I k (arD k =
for for
rD < rD
,
rD > rD
(7.52)
where
2 2 2 2 rD rD cos ( ) , RD = rD + rD
(7.53)
we can write
P=
+ Ik 2klh D s k =
~ q
u rD K k
) ( u r )cos k ( ) + 2 cos n z
D n =1
cos n
z D hD
hD
+ n 2 2 n 2 2 ( ) + + Ik u r K u r cos k D k D 2 2 hD hD k =
(7.54)
in Eq. 7.54. If we choose Q in Eq. 7.49 as . If rD > rD , we interchange rD and rD for rD < rD
Q=
+ ak I k 2klh D s k =
~ q
( u r )cos k ( ) + 2 cos n z
D n =1
cos n
z D hD
hD
+ n 2 2 ( ) + bk I k u r cos k D 2 hD k =
(7.55)
where a k and bk are constants, then this choice of Q satisfies the boundary condition given by Eq. 7.50 and the contribution of Q to the flux at the source location vanishes. If we also choose the constants a k and bk in Eq. 7.55 as follows, respectively,
ak = I k
( u r )K ( u r ) I ( u r ),
D k eD k eD
(7.56)
and
n 2 2 n 2 2 Kk u+ bk = I k u r r + D eD 2 2 hD hD n 2 2 , u+ Ik r eD 2 hD
(7.57)
Chap. 7
58
then p = P + Q satisfies the impermeable boundary condition at rD = reD given by Eq. 3.134 and thus the point source solution for a closed cylindrical reservoir is given by
p =
K 0 2klh D s ~ q cos n zD hD
( u R ) I ( u r ) I ( u r )K ( u r ) cos k ( ) I ( ur )
+ k D k eD D k D k = k eD
+2
n =1
cos n
z n 2 2 D K0 u R + D 2 hD hD
(7.58)
k =
n 2 2 n 2 2 Ik u r K u r + + D k eD 2 2 hD hD n 2 2 cos k ( ) u+ rD 2 hD n 2 2 u+ Ik r eD 2 hD
The above solution procedure may be extended to the cases where the boundaries are at constant pressure or of mixed type.19 Tables 7.2 A and 7.2 B presents the point source solutions for cylindrical reservoirs for all possible combinations of the boundary conditions. Solutions for other source geometries in cylindrical reservoirs may be obtained by using the point source solutions in Tables 7.2 A and 7.2 B in Eq. 7.37 (or Eq. 7.38). We present two examples below.
Table 7.2 A Laplace Domain Solutions for Point Source Wells in Cylindrical Reservoirs19 Impermeable Boundary at rD = reD
2 2 2 ~ , R 2 = r 2 + r 2 2 r r cos ( ), = u + n 2 2 h 2 , q=q hD D D D D D n D 2 n 1 = u + (2n 1)
CONDITION
p z D
z D = 0 ,hD
+ reD u I r u Kk K R u I k rD u k D cos k ( ) 0 D I r u 2klh D s k = k eD
=0
~ q
) (
) (
) (
+2
cos n h
n =1
zD
D
cos n zD hD
+ I (r )K (r ) z D K 0 (R D n ) I k (rD n ) k D n k eD n cos k ( ) (reD n ) hD Ik k =
z D = 0 ,hD
=0
~ q
k lh D s
p z D
z D =0
n =1
sin n
sin n
+ I (r )K (r ) z D K 0 (R D n ) I k (rD n ) k D n k eD n cos k ( ) (reD n ) hD Ik k =
=0
~ q
klh D
cos(2n 1) 2h s
n =1
z D
D
cos (2n 1)
z D 2h D
z D = hD
=0
+ )K (r ) I (r K 0 (R D 2 n 1 ) I k (rD 2 n 1 ) k D 2 n 1 k eD 2 n 1 cos k ( ) (reD 2 n 1 ) Ik k =
Table 7.2 B Laplace Domain Solutions for Point Source Wells in Cylindrical Reservoirs19 Constant Pressure Boundary at rD = reD
2 2 2 ~ , R 2 = r 2 + r 2 2 r r cos ( ), = u + n 2 2 h 2 , q=q hD D D D D D n D 2 n 1 = u + (2 n 1)
Chap. 7
+ I r u K k reD u I k rD u k D cos k ( ) K 0 R D u 2klh D s I r u k = k eD
59
CONDITION p z D
z D = 0 ,hD
=0
~ q
) (
z D hD
) (
) (
+2
n =1
cos n
zD hD
cos n zD hD
+ I (r )K (r ) K 0 (R D n ) I k (rD n ) k D n k eD n cos k ( ) I k (reD n ) k =
z D = 0 ,hD
=0
~ q
klh D s
p z D
z D =0
n =1
sin n
sin n
+ I (r )K (r ) zD K 0 (R D n ) I k (rD n ) k D n k eD n cos k ( ) hD I k (reD n ) k =
=0
~ q
klh D
cos(2n 1) 2h s
n =1
z D
D
cos (2n 1)
z D 2h D
z D = hD
=0
+ )K (r ) I (r I k (rD 2 n 1 ) k D 2 n 1 k eD 2 n 1 cos k ( ) K 0 (R D 2 n 1 ) I k (reD 2 n 1 ) k =
Example 7.1: A partially penetrating, uniform-flux fracture of height h w and half-length x f in an isotropic and closed cylindrical reservoir. The center of the fracture is at r = 0 , = 0 , z = z w and the fracture tips extend from r = x f , = + to r = x f , = .
) (
The geometry of the fracture-reservoir system considered in this example is shown in Fig. 7.1. The solution for this problem is obtained by first generating a partially penetrating line source and then using this line source solution to , , z w ) is obtained by generate the plane source. The solution for a partially penetrating line source at ( rD integrating the corresponding point source solution given in Table 7.2 A with respect to z from z w h w 2 to
z w + h w 2 and is given by
p =
+
~ h q w K 0 2klh D s 4h
( u R ) I ( u r ) I ( u r )K ( u r ) cos k ( ) I ( ur )
+ k D k eD D k D k = k eD
h w
n sin n 2h cos n h
1 hw
n =1
zD
D
cos n
z wD hD
(7.59)
+ I ( r )K ( r ) I k ( n rD ) k n D k n eD cos k ( ) K 0 ( n R D ) ( n reD ) Ik k =
Chap. 7
60
z =h
II
I
r = xf
= =+
r = xf
xf
re
hw
zw z=0
III IV
Fig. 7.1 Geometry of the partially penetrating fracture in a closed cylindrical reservoir (Example 7.1).
To generate the solution for a partially penetrating plane source that represents the fracture, the partially penetrating line source solution given in Eq. 7.59 is integrated with respect to r from 0 to x f with = + in the third quadrant and with = in the first quadrant. This procedure yields
xf l ~ h q 2 w 2 2rD rD cos ( ) drD p = K 0 u rD + rD 2kh D s 0
xf l
2 2 2rD rD cos ( ) drD K 0 u rD + rD
k =
[cos k ( ) + cos k ( )]
4h
Ik
( u r )K ( u r ) I ( u r )dr I ( ur )
xf l D k eD k D k eD 0
h w
x f l
n sin n 2h cos n h
1 hw
n =1
zD
D
cos n
z wD hD
xf l
2 + 2 2rD rD cos ( ) drD + rD K 0 n rD
2 2 2rD rD cos ( ) drD + rD K 0 n rD xf l
k =
[cos k ( ) + cos k ( )]
( n reD ) I k ( n rD )K k ( n reD ) Ik
)drD I k ( n rD 0
(7.60) It is possible to obtain an alternate representation of the solution given in Eq. 7.60. Using the addition theorem of the Bessel function K 0 (x ) given by Eq. 7.52, the solution in Eq. 7.59 may be written as
p =
~ h + q w Fk k = 2klh D s
h z 4h 1 ( u )cos k ( ) + h n sin n 2h cos n h

w w n =1
D D
cos n
z wD hD , (7.61)
Fk ( n ) cos k ( ) k =
where
Chap. 7
, rD ) F ( , rD Fk ( ) = k ) Fk ( , rD , rD for rD rD for rD rD (7.62)
61
and
(reD )] I k (reD )}. (reD ) I k (b )K k Fk ( , a ,b ) = I k (a ){[K k (b )I k
(7.63)
The integration of the partially penetrating vertical well solution given in Eq. 7.61 with respect to r from 0 to x f (with = + in the third quadrant and with = in the first quadrant) yields the following alternate form of the partially penetrating fracture solution:
xf l ~ h + q w cos k ( ) Fk p = 2kh D s k = 0
( u )dr + cos k ( ) F ( u )dr

D k D 0
xf l
, (7.64)
4h
h w
n sin n 2h cos n h
1 hw
n =1
zD
D
cos n
z wD hD
xf l xf l cos k ( ) Fk ( n )drD + cos k ( ) Fk ( n )drD k = 0 0
where Fk ( , rD , rD )drD = r0 Fk ( )drD D 0 )drD ( ) + F , r , r d r Fk ( , rD , rD D D D k rD 0
for rD x f l . for rD x f l (7.65)
Example 7.2: A uniform-flux, horizontal line-source well of length Lh in an isotropic and closed cylindrical reservoir. The well extends from (r = Lh 2 , = + ) to (r = Lh 2 , = ) and the center of the well is at r = 0 , = 0 , z = zw .
The solution for a horizontal line-source well in a closed cylindrical reservoir is obtained by integrating the corresponding point source solution given in Table 7.2 A with respect to r from 0 to Lh 2 with = + in the third quadrant and with = in the first quadrant. The final form of the solution is given by
Chap. 7
p =
Lh (2 l ) 2 2 2 rD rD cos ( ) drD K 0 u rD + rD 2kh D s 0 ~ q
62
Lh (2 l )
2 2 2rD rD cos ( ) drD K 0 u rD + rD
k =
[cos k ( ) + cos k ( )] cos n h

n =1
Ik
( u r )K ( u r ) ( I ) ( u r )dr I ( ur )
Lh 2 l D k eD k D k eD 0 Lh (2 l )
+2
zD
D
cos n
z wD hD
Lh (2 l )
2 2 2rD rD cos ( ) drD + K 0 n rD + rD
2 2 2rD rD cos ( ) drD K 0 n rD + rD Lh (2 l )
k =
[cos k ( ) + cos k ( )]
( n reD ) I k ( n rD )K k ( n reD ) Ik
)drD I k ( n rD 0
(7.66)
7.7. Solutions for Rectangular Parallelepiped Reservoirs
Solutions for rectangular parallelepiped reservoirs may also be obtained by starting from the point source solution in the Laplace transform domain in an infinite reservoir and using the method of images to generate the effects of the planar boundaries. Although the formal procedure to obtain the solution is fairly easy, the use of the method of images in three directions ( x , y , z ) yields triple infinite Fourier series which may pose computational inconveniences. As an example, the solution for a continuous point source located at x , y , z in a rectangular porous medium occupying the region 0 < x < x e , 0 < y < y e , and 0 < z < h is obtained by applying the method of images to the point source solution given by Eq. 7.34 as follows:19,28
p =
(S 4kls
k = m = n =
~ q
1,1,1
+ S 2 ,1,1 + S1,2 ,1 + S 2 ,2 ,1 + S1,1,2 + S 2 ,1,2 + S1,2 ,2 + S 2 ,2 ,2 ) ,
(7.67) where
exp u (~ x Di 2kx eD )2 + ~ y Dj 2my eD 2 + (~ z Dl 2nh D )2 = 2 2 2 ~ ~ ~ (x Di 2kx eD ) + y Dj 2my eD + (z Dl 2nh D )
S i , j ,l
for i , j , l = 1 or 2 .
(7.68)
and ~ x Ds = x D + ( 1)s x wD ~ y Ds = y D + ( 1)s ywD ~ z Ds = z D + ( 1)s z wD for for for s = 1 or 2 , s = 1 or 2 , s = 1 or 2 .
(7.69) (7.70) (7.71)
Chap. 7
63
It is shown in Ref. 28 that the triple infinite sums in Eq. 7.67 may be reduced to double infinite sums by using the following formula:
cosh u + u+
k = m = n =
S = 2x
eD h D
~ ~ xD cosh u ( y eD y D ) + 2 cos k x eD u sinh u y eD k =1
2k 2
2 x eD
(y eD ~ yD )
2k 2
2 x eD
2k 2
2 x eD
sinh u +
y eD
2n 2 cosh u + (y eD ~ yD ) 2 ~ h z D + 2 cos n D 2 2 hD n 2n 2 n =1 sinh u + y eD u+ 2 2 h h D D
+2
k =1
cos k
~ xD x eD
2 2 hD x eD 2n 2 2k 2 2n2 2k 2 + + u+ sinh u + y eD 2 2 2 2 hD x eD hD x eD cosh u +
2n 2
2k 2
(y eD ~ yD )
(7.72) where
exp u (~ x D 2kx eD )2 + ( ~ y D 2my eD )2 + (~ z D 2nh D )2 . S= 2 2 2 ~ ~ ~ (x D 2kx eD ) + ( y D 2my eD ) + (z D 2nh D )
(7.73)
The resulting continuous point source solution for a closed rectangular reservoir is given by
cosh u ~ x x y D1 + cosh u ~ y D2 + 2 cos k D cos k D 2klx eD s x eD x eD u sinh u y eD k =1 y D1 + cosh n ~ y D2 cosh k ~ y D1 + cosh k ~ y D2 z cosh n ~ z + 2 cos n D cos n D , k sinh k y eD n sinh n y eD hD hD n =1
p =
~ q
(7.74)
+2
cos k x
k =1
xD
eD
cos k
~ ~ x D cosh k ,n y D1 + cosh k ,n y D 2 x eD
k ,n sinh k ,n y eD
where
~ y D1 = y eD y D y D , ~ y D 2 = y eD ( y D + y D ),
2 , n = u + n 2 2 h D
(7.75) (7.76) (7.77)
Chap. 7
2 , k = u + k 2 2 x eD
64
(7.78)
and
2 2 . k ,n = u + k 2 2 x eD + n 2 2 h D
(7.79)
Following the lines similar to those explained above, it is possible to derive the point source solutions in rectangular parallelepiped reservoirs for different combinations of boundary conditions.17,28 These solutions are given in Tables 7.3 A 7.3 C and may be used to derive the solutions for the other source geometries by using Eq. 7.37 (or Eq. 7.38). To demonstrate the derivation of the solutions for the other source geometries in rectangular reservoirs, we consider two examples below.
Table 7.3 A Laplace Domain Solutions for Point Source Wells in Parallelepiped Reservoirs17,28 Impermeable Boundaries at z D = 0 and z D = h D ~, ~ y = y (y + y ) q=q y = y y y , ~
D1 eD D D D2 eD D D 2 2 2 2 n = u + n 2 2 h D , k = u + k 2 2 x eD , k ,n = u + k 2 2 x eD + n 2 2 h D , 2 2 2 2 k 1 = u + (2k 1)2 2 x eD , 2 k 1,n = u + (2k 1)2 2 x eD + n 2 2 h D
AREAL VIEW
~ q
cosh u ~ y D1 + cosh u ~ y D2 x x + 2 cos k D cos k D u sinh u y eD x eD x eD 2klx eD s k =1 cosh k ~ y D1 + cosh k ~ y D2 y D1 + cosh n ~ y D2 z z cosh n ~ + 2 cos n D cos n D k sinh k y eD n sinh n y eD hD hD n =1
+2
k =1
cos k
xD x eD
cos k
~ ~ x D cosh k ,n y D1 + cosh k ,n y D 2 x eD k ,n sinh k ,n y eD
y D1 cosh k ~ y D2 x x cosh k ~ sin k D sin k D k sinh k y eD klx eD h D s k =1 x eD x eD ~ q
+2
n =1
cos n
zD hD
cos n
z D hD
k =1
sin k
xD x eD
sin k
y D1 cosh k ,n ~ y D2 cosh k ,n ~ xD x eD k ,n sinh k ,n y eD
cosh u ~ y D1 + cosh u ~ y D2 x x + 2 cos k D cos k D u sinh u y eD x eD x eD 2klx eD h D s k =1 cosh k ~ y D1 cosh k ~ y D2 y D1 cosh n ~ y D2 z z cosh n ~ + 2 cos n D cos n D k sinh k y eD n sinh n y eD hD hD n =1
~ q
+2
cos k x
k =1
xD
eD
cos k
~ ~ x D cosh k ,n y D1 cosh k ,n y D 2 x eD
k ,n sinh k ,n y eD
Chap. 7
cos (2k 1) x D cos (2k 1) x D klx eD h D s k =1 2 x eD 2 x eD ~ q
65
~ ~ sinh 2 k 1 y D1 + sinh 2 k 1 y D 2 2 k 1 cosh 2 k 1 y eD
+2
n =1
cos n
zD hD
cos n
~ ~ z D sinh 2 k 1,n y D1 + sinh 2 k 1,n y D 2 hD 2 k 1,n cosh 2 k 1,n y eD ~ ~ cosh 2 k 1 y D1 + cosh 2 k 1 y D 2 2 k 1 sinh 2 k 1 y eD
cos (2k 1) x D cos (2k 1) x D 2 x eD 2 x eD klx eD h D s k =1 ~ q
+2
n =1
cos n
zD hD
cos n
~ ~ z D cosh 2 k 1,n y D1 + cosh 2 k 1,n y D 2 hD
2 k 1,n sinh 2 k 1,n y eD
cos (2k 1) x D cos (2k 1) x D 2 x eD 2 x eD klx eD h D s k =1 ~ q
~ ~ cosh 2 k 1 y D1 cosh 2 k 1 y D 2 2 k 1 sinh 2 k 1 y eD
+2
n =1
cos n
zD hD
cos n
~ ~ z D cosh 2 k 1,n y D1 cosh 2 k 1,n y D 2 hD 2 k 1,n sinh 2 k 1,n y eD
Table 7.3 B Laplace Domain Solutions for Point Source Wells in Parallelepiped Reservoirs17,28 Constant Pressure Boundaries at z D = 0 and z D = h D ~, ~ y = y (y + y ) q=q y = y y y , ~
~ ~ z D1 = h D z D z D , z D 2 = h D (z D + z D)
D1 eD D D D2 eD D D 2 2 2 2 n = u + n 2 2 h D , m = u + m 2 2 y eD k = u + k 2 2 x eD , k ,n = u + k 2 2 x eD + n 2 2 h , 2 2 2 k ,m = u + k 2 2 x eD + m 2 2 y eD , 2 k 1 = u + (2k 1)2 2 x eD , 2 2 2 2 2 k 1,n = u + (2k 1)2 2 x eD + n 2 2 h D , 2 k 1,m = u + (2k 1)2 2 x eD + m 2 2 y eD
AREAL VIEW
~ q
cosh u ~ z D1 cosh u ~ z D2 u sinh u h D 2klx eD y eD s z D1 cosh k ~ z D2 x x cosh k ~ + 2 cos k D cos k D x eD x eD k sinh k h D k =1
m =1
+2
cos m y
k =1
yD
eD
cos m
~ ~ yD cosh m z D1 cosh m z D 2 y eD m sinh m h D
+2 ~ 2q
cos k
xD x eD
cos k
~ ~ x D cosh k ,m z D1 + cosh k ,m z D 2 x eD k ,m sinh k ,m h D xD

eD
z z sin n D sin n D klx eD h D s n =1 hD hD
sin k x
k =1
sin k
Chap. 7
y D1 cosh n ~ y D2 z z cosh n ~ sin n D sin n D klx eD h D s hD hD n sinh n y eD n =1 ~ q
66
+2 ~ 2q
cos k x
k =1
xD
eD
cos k
z z x x sin n D sin n D cos (2k 1) D cos (2k 1) D 2 x eD 2 x eD klx eD h D s n =1 hD h D k =1 sinh 2 k 1,n ~ y D1 + sinh 2 k 1,n ~ y D2 2 k 1,n cosh 2 k 1,n y eD
z D1 cosh 2 k 1 ~ z D2 x x cosh 2 k 1 ~ cos (2k 1) D cos (2k 1) D klx eD y eD s 2 k 1 sinh 2 k 1 h D 2 x eD 2 x eD k =1 ~ q
+2
m =1
cos m
yD y eD
cos m
z D1 cosh 2 k 1,m ~ z D2 cosh 2 k 1,m ~ yD y eD 2 k 1,m sinh 2 k 1,m h D ~ ~ cosh 2 k 1 y D1 cosh 2 k 1 y D 2 2 k 1 sinh 2 k 1 y eD
cos (2k 1) x D cos (2k 1) x D klx eD h D s 2 x eD 2 x eD k =1 ~ q
+2
n =1
sin n
zD hD
sin n
~ ~ z D cosh 2 k 1,n y D1 cosh 2 k 1,n y D 2 hD 2 k 1,n sinh 2 k 1,n y eD
Table 7.3 C Laplace Domain Solutions for Point Source Wells in Parallelepiped Reservoirs17,28 Impermeable Boundary at z D = 0 and Constant Pressure Boundary at z D = h D ~, ~ q=q y = y y y , ~ y = y (y + y )
~ ~ z D1 = h D z D z D , z D 2 = h D (z D + z D)
D1
eD
D2
eD
2 2 2 2 2n 1 = u + (2n 1)2 2 h D , 2 k 1 = u + (2k 1)2 2 x eD , k ,2 n 1 = u + k 2 2 x eD + (2n 1)2 2 h D 2 2 2 2 2 k 1,2 n 1 = u + (2k 1)2 2 x eD + (2n 1)2 2 h D , 2 k 1,m = u + (2k 1)2 2 x eD + m 2 2 y eD AREAL VIEW PRESSURE DISTRIBUTION, p
y D1 + cosh 2 n 1 ~ y D2 z z cosh 2 n 1 ~ cos (2n 1) D cos (2n 1) D klx eD h D s 2 n 1 sinh 2 n 1 y eD 2h D 2h D n =1 ~ q
+2
k =1
cos k
xD x eD
cos k
~ ~ x D cosh k ,2 n 1 y D1 + cosh k ,2 n 1 y D 2 x eD k ,2 n 1 sinh k ,2 n 1,m y eD
y D1 cosh 2 n 1 ~ y D2 z z cosh 2 n 1 ~ cos (2n 1) D cos (2n 1) D klx eD h D s 2 n 1 sinh 2 n 1 y eD 2h D 2h D n =1 ~ q
+2
k =1
sin k
xD x eD
sin k
y D1 cosh k ,2 n 1 ~ y D2 cosh k ,2 n 1 ~ xD x eD k ,2 n 1 sinh k ,2 n 1,m y eD
Chap. 7
y D1 cosh 2 n 1 ~ y D2 z z cosh 2 n 1 ~ cos (2n 1) D cos (2n 1) D klx eD h D s 2 n 1 sinh 2 n 1 y eD 2h D 2h D n =1 ~ q
67
+2
cos k x
k =1
xD
eD
cos k
y D1 cosh k ,2 n 1 ~ y D2 cosh k ,2 n 1 ~ xD x eD k ,2 n 1 sinh k ,2 n 1,m y eD
z z cos (2n 1) D cos (2n 1) D klx eD h D s 2h D 2h D n =1 ~ 2q
k =1
cos (2k 1)
x D
2 x eD
cos (2k 1)
~ ~ x D sinh 2 k 1,2 n 1 y D1 + sinh 2 k 1,2 n 1 y D 2 2 x eD
2 k 1,2 n 1 cosh 2 k 1,2 n 1 y eD
z D1 + sinh 2 k 1 ~ z D2 x x sinh 2 k 1 ~ cos (2k 1) D cos (2k 1) D klx eD y eD s 2 k 1 cosh 2 k 1 h D 2 x eD 2 x eD k =1 ~ q
+2
m =1
cos m
yD y eD
cos m
~ ~ y D sinh 2 k 1,m z D1 + sinh 2 k 1,m z D 2 y eD 2 k 1,m cosh 2 k 1,m h D
x x cos (2k 1) D cos (2k 1) D 2 x eD 2 x eD klx eD y eD s k =1 z D1 + sinh 2 k 1,m ~ z D2 y y sinh 2 k 1,m ~ sin m D sin m D y y cosh h m =1 eD eD 2 k 1,m 2 k 1,m D ~ 2q
Example 7.3: A fully penetrating vertical fracture of half-length x f located at x = x w , and y = y w in a closed
rectangular reservoir.
Assuming uniform-flux distribution along the fracture surface, the solution for this problem is obtained by integrating the corresponding point source solution in Table 7.3 A first with respect to z from 0 to h and then with respect to x from x w x f to x w + x f . The result is
p =
cosh u ~ y D1 + cosh u ~ y D2 klx eD h D s u sinh u y eD ~hx q f
2x x y D1 + cosh k ~ y D2 1 x cosh k ~ sin k cos k w cos k + e x f k =1 k xe xe xe k sinh k y eD
(7.80)
xf
where ~ y D1 , ~ y D 2 , and k are given respectively by Eqs. 7.75, 7.76, and 7.78.
Example 7.4: A horizontal well of length Lh in the x -direction located at x = x w , y = y w , and z = z w in a closed rectangular reservoir.
The solution for a horizontal line source well is obtained by integrating the corresponding point source solution in Table 7.3 A with respect to x from x w Lh 2 to x w + Lh 2 and is given by
Chap. 7
p = p 1 + p 2 ,
(7.81)
68
where
p1 =
4xe
cosh u ~ y D1 + cosh u ~ y D2 2klx eD h D s u sinh u y eD ~L q h
y D1 + cosh k ~ y D2 x cosh k ~ + sin k cos k cos k xe Lh k =1 k 2xe xe k sinh k y eD
(7.82)
Lh
xw
and
p 2 =
4xe
z cosh n ~ y D1 + cosh n ~ y D2 z cos n D cos n wD n sinh n y eD 2klx eD h D s n =1 hD hD ~L 2q h
Lh
~ ~ x cosh k ,n y D1 + cosh k ,n y D 2 + sin k cos k cos k Lh k =1 k 2xe xe xe k ,n sinh k ,n y eD
(7.83)
xw
y D1 , ~ y D 2 , n , k , and k ,n are given respectively by Eqs. 7.77 7.79. In Eq. 7.83, ~
7.7. Conversion from Three- to Two-Dimensional Anisotropy
Solutions presented above assume that the reservoir is anisotropic in all three principal directions, x , y , and z with k x , k y , and k z denoting the corresponding permeabilities. In these solutions, an equivalent isotropic permeability, k , has been defined by k = 3 kxk ykz . For some applications, it may be more convenient to define an equivalent horizontal permeability by kh = k x k y (7.85) (7.84)
and replace k in the solutions given above by k h . Note that k takes place in the definition of the dimensionless ~ time t D (Eq. 7.12) in the above solutions. Then, if we define a dimensionless time tD based on k h , then, the ~ relation between tD and t D is given by
~ tD = k h t D k .
(7.86)
Because in the above solutions, the Laplace transformation is with respect to t D , conversion from three- to twodimensional anisotropy requires the use of the following property of the Laplace transforms:
cF (cs ) = L{F (t c )} .
(7.87)
Chap. 7
69
As an example, let us consider the solution for a horizontal well in an infinite slab reservoir. Assuming that the midpoint of the well is the origin ( x wD = 0 , y wD = 0 ) and choosing the half-length of the horizontal well as the characteristic length; that is, l = Lh 2 , we may write the horizontal well solution given in Table 7.1 A as follows:
p =
+1 K0 xD k k x 2kh D s 1
~ q
2 + yD sf (s ) d
+ 2 cos nz D cos nz wD K 0 x D k k x n =1 1
+1
2 + yD
n 2 2 sf (s ) + d 2 hD
(7.88)
In Eq. 7.88, s is the Laplace transform variable with respect to dimensionless time t D based on k and
x D = 2 x k k x Lh ,
(7.89) (7.90) (7.91)
y D = 2 y k k y Lh ,
zD = z h ,
and
h D = 2h k k z Lh .
(7.92)
If we define the following variables based on k h

~ xD = kh k xD , ~ yD = kh k yD , ~ hD = k h k hD ,
(7.93) (7.94) (7.95)
and also note that h D = 2 h k k z Lh = 2 h k x k y k
(L k ) = 2hk
32 h
(Lh k )
(7.96)
then, we may rearrange Eq. 7.88 in terms of the dimensionless variables based on k h as follows:
+1 ~ p (k h ) = K0 xD kh k x 2 (2h Lh )k h (k k h )s 1 ~ (k k ) q h
2 yD +~
k ~ k sf kh kh
s d d
xD kh k x + 2 cos nz D cos nz wD K 0 ~ n =1 1
(7.97)
+1
2 yD +~
k ~ k sf kh kh
n 2 2 s + ~ 2 hD
where
Chap. 7
~ k f kh
~ (k k )s (1 ) + h s = ~ (k k h )s (1 ) +
70
(7.98)
and
= k k h .
If we compare Eqs. 7.88 and 7.97, we may show that
(7.99)
p (k h ) =
k p k (2h Lh )k h kh kh kh D
kh D ~ s = L{p (tD )}, (2h Lh )k h
(7.100)
s as the Laplace transform variable with where we have used the relation given by Eq. 7.87. If we now define ~ ~ respect to tD , we may write the following relation:
p (k h ) =
kh D
(2h
Lh )k h
p (~ s ).
(7.101)
Then, using the relation given by Eqs. 7.101 and Eq. 7.88, we obtain the following horizontal well solution in terms of dimensionless variables based on k h :
p (k h ) =
+1 ~ K0 xD kh k x 2 (2h Lh )k h ~ s 1
~ q
2 ~~ ~ yD s f (s ) d +~
xD kh k x + 2 cos nz D cos nz wD K 0 ~ n =1 1
+1
n 2 2 2 ~~ ~ yD s f (s ) + ~ +~ 2 hD
(7.102)
Chapter 8 COMPUTATIONAL CONSIDERATIONS AND APPLICATIONS OF THE LAPLACE DOMAIN SOLUTIONS TO UNSTEADY FLOW PROBLEMS:
The numerical evaluation of the Laplace domain solutions, especially the ones given in Chapter 7, may be sometimes difficult, inefficient, or even impossible. Alternate computational forms of some of these solutions have been presented in Refs. 18, 19, and 28. Here, we present a summary of the alternate formulas to be used in the computation of the source functions in the Laplace transform domain. Some of these formulas are for computations at early or late times and may be useful to derive asymptotic approximations of the solutions during the corresponding time periods.
8.1. Asymptotic Approximations of the Solutions for Short and Long Times As noted in Section 4.1, the short- and long-time approximations of the solutions correspond to the limiting forms of the solution in the Laplace transform domain as s and s 0 , respectively. We should note that in the solutions developed in Chapter 7, we have defined u = sf (s ) (see Eq. 7.35). From elementary considerations, it is possible to show that the definition of f (s ) given in Eq. 7.25 yields the following limiting forms
s
lim u = lim sf (s ) = s
s
(8.1)
and
s 0
lim u = lim f (s ) = s .
s 0
(8.2)
These limiting forms will be used below in the derivation of the short- and long-time asymptotic approximations. Note that in the following expressions, homogeneous reservoir solutions are obtained by substituting = 1 .
8.2. Computational Considerations The Laplace domain solutions to transient flow problems in porous media may include some components that may make the numerical evaluations difficult or sometimes impossible. In this section, some common computational problems are noted and their potential solutions are addressed. The integral I =
1
i)
K s
a
2 (x D )2 + y D d
This integral arises in the computation of many practical transient pressure solutions and may not be numerically evaluated especially as y D 0 . The following alternate forms of the integral, however, are numerically computable.19
Chap. 8
COMPUTATIONAL CONSIDERATIONS AND APPLICATIONS OF THE LAPLACE DOMAIN SOLUTIONS TO UNSTEADY FLOW PROBLEMS
u (xD a ) 1 2 2 I= K0 + uy D d s u 0
72
u (xD b )
2 2 K0 + uy D d ; x D b ,
(8.3)
u (b xD ) 1 2 2 I= K0 + uy D d s u 0
u (a x D )
2 2 K0 + uy D d ; x D a ,
(8.4)
and
u (xD a ) 1 2 2 I= K0 + uy D d + s u 0
u (b x D )
2 2 K 0 + uy D d ; a x D b .
(8.5)
The integrals in Eqs. 8.3 8.5 may be evaluated by using the standard numerical integration algorithms for y D 0 . For y D = 0 , the polynomial approximations given by Luke29 or the following power series expansion given by Abramowitz and Stegun7 may be used in the computation of the integrals in Eqs. 8.3 8.5:
(x 2 )2k x K 0 ( )d = x ln + 0.5772 K + 2 2 0 k = 0 (k !) (2k + 1) z
( x 2 )2 k 1 + . 2 2 k = 0 (k !) (2k + 1) n =1 n k
(8.6)
For numerical computations and asymptotic evaluations, it may also be useful to note the following relations:19
and
2 2 2 2 K0 +c d = exp ( c ) K 0 +c d , 2 0 z
(8.7)
K
0
2 + c 2 d = exp ( c ) . 2
(8.8)
It can be shown from Eqs. 8.7 and 8.8 that, for practical purposes, when z 20 the right hand sides of Eqs. 8.6 and 8.7 may be approximated by 2 and exp ( c ) 2 , respectively.19,28 As shown in Refs. 18, 19 and 28, it is possible to derive the short- and long-time approximations (that is the limiting forms as s and s 0 , respectively) for the integral I given, respectively, by
lim I =
2 s3 2
exp y D s
(8.9)
where
2 = 1 0 for for for a < xD < b x D = a or x D = b , x D < a or x D > b
(8.10)
Chap. 8
and
73
s 0
lim I =
ba 2s
( ln s + ln 4 2
+ 2) +
2 s
(x D , y D , a ,b ) ,
(8.11)
where = 0.5772K and
(x D , y D , a , b ) =
1 4
{(x
2 2 b ) ln (x D b )2 + y D (x D a ) ln (x D a )2 + y D
]}
. (8.12)
x a y arctan x D b D arctan D 2 y D yD
It is also useful to note the real inversions of Eqs. 8.9 and 8.11 given, respectively, by
2 yD y erfc y D lim I = t D exp D t D 0 2 t D 4 tD 2
(8.13)
and
lim I = 4t D ba ln 2 e + 2 + 2 (x D , y D , a , b ) .
2 2 2 2 exp u + n hD + a y D for y D 0 . +a 2
tD
(8.14)
ii)
The Series S1 =
n =1
cos nz cos nz w u+n

2 2 2 hD
It is possible to note two alternate expressions for the series S 1 that may be convenient for the large and small values of u (that is, for short and long times).28 When u is large,
S1 = hD 2 K
+
n =
2 2 ( z z w 2 n )2 h D + yD
u + a2
+ K0
(z + z w 2 n )
2 2 + yD hD
, 2 exp u + a yD u + a2 2 2 u+a
(8.15)
2 and when u + a 2 << n 2 2 h D ,
Chap. 8
exp u + n 2 2 h 2 + a 2 y D D exp ( ny D h D ) S1 = cos nz cos nz w 2 n h D u + n 2 2 h D + a2 n =1
74
{ln[1 2 exp ( y D h D ) cos (z + z w ) + exp ( 2y D hD )] 4 + ln[1 2 exp ( y D h D ) cos (z z w ) + exp ( 2y D h D )]}

hD
(8.16)
iii)
The Series S 2 =
n =1
sin nz sin nz w u+n

2 2 2 hD
2 2 2 2 exp u + n hD + a y D for y D 0 . +a 2
Alternate computational forms for the series S 2 are given as follows:28 When u is large,
S2 = hD 2 K
+
n =
2 2 ( z z w 2 n )2 h D + yD
u + a 2 K 0
2 2 (z + z w 2n )2 h D + yD
u + a 2 ,
(8.17)
2 and when u + a 2 << n 2 2 h D ,
exp u + n 2 2 h 2 + a 2 y D D exp ( ny D h D ) S2 = sin nz sin nz w 2 n h D + a2 u + n 2 2 h D n =1 .
(8.18)
hD 4
ln
1 2 exp ( y D h D ) cos (z + z w ) + exp ( 2y D h D ) 1 2 exp ( y D h D ) cos (z z w ) + exp ( 2y D h D )
iv)
The Series S 3 =
n =1
cos (2n 1) z cos (2n 1) z 2 2 w u + (2n 1)

2 2
( )+ a
2 4h D
2 exp u + (2n 1)2 2 4h D + a 2 y D for y D 0 .
( )
The following alternate forms for the series S 3 may be convenient for the large and small values of u (that is, for short and long times).28 When u is large,
S3 = hD 2
n =
( 1)
K 0
2 2 ( z z w 2 n )2 h D + yD
u + a 2 + K 0
2 2 ( z + z w 2 n )2 h D + yD
u + a 2 ,
(8.19)
2 and when u + a 2 << (2n 1)2 2 4h D ,
( )
Chap. 8
S3 =
75
cos(2n 1) 2 z cos(2n 1) 2 z
n =1
u + (2n 1)2 2 4h 2 + a 2 y D D exp exp[ (2n 1)y D (2h D )] 2 2 2 (2n 1) (2h D ) + a2 u + (2n 1) 4h D . (z + z w ) + exp ( y D h D ) h 1 + 2 exp[ y D (2h D )] cos 2 + D ln 4 1 2 exp[ y D (2h D )] cos 2 (z + z w ) + exp ( y D h D ) + ln
( ) ( )
(8.20)
(z z w ) + exp( y D hD ) 1 + 2 exp[ y D (2h D )] cos 2 1 2 exp[ y D (2h D )] cos 2 (z z w ) + exp ( y D h D )

1 cos nz D cos nz D K 0 n s n =1 a
v)
The series F =
2 (x D )2 + y D d
2 where n = u + n 2 2 h D
The series F may be written in the following forms by using Eqs. 8.3 8.5:
F= 1
s
n =1
cos nz D cos nz D
n ( x D b ) n ( xD a ) 2 2 2 2 2 2 K0 K0 + n y D d + n y D d 0 0
xD b .
(8.21)
F=
s
n =1
cos nz D cos nz D
n
d
n (a x D )
n (b xD ) 2 2 2 K0 + n yD 0
2 2 2 K0 + n yD
xD a .
(8.22)
and F= 1
s
n =1
cos nz D cos nz D
n
d +
n (b x D )
n ( xD a ) 2 2 2 K0 + n yD 0
2 2 2 K0 + n yD
a xD b .
(8.23)
Computation of the series in Eqs. 8.21 and 8.22 should not pose numerical difficulties. The series in Eq. 8.23, however, converges slowly. Using the relation given in Eq. 8.8, we may write Eq. 8.23 as follows,28
Chap. 8
F (a x D b ) = F1 1
76
s
n =1
cos nz D cos nz D
n
, (8.24)
2 2 2 2 2 2 K 0 + n y D d + K0 + n yD d n (b x D ) n ( xD a )
where
F1 =
s
n =1
cos nz D cos nz D
exp ( n y D ) .
(8.25)
The computation of the series given in Eq. 8.25 is discussed below separately. It is also possible to derive asymptotic approximations for the series F . When s is large (small times), F may be approximated by28
lim F =
hD
4s
n =
2 2 (z D + z D 2n )2 h D + yD
+ K0
(z D z D 2n )
2 hD
2 yD
s exp s y D
) (4s
32
(8.26)
where is given by Eq. 8.10. If s is sufficiently large, then Eq. 8.26 may be further approximated by
lim F =
hD
K0 4s
2 2 )2 h D (z D z D + yD
exp s y D
4s 3 2
).
(8.27)
The inverse Laplace transform of Eq. 8.27 yields
t D 0
lim F =
hD
8
2 (z z )2 h 2 + y 2 t yD D D D D y erfc y D . Ei D exp D 2 t 2 4t D D 4t D 2 (8.28)
For small s (large times), depending on the value of x D , F may be approximated by one of the following expressions28
s 0
lim F =
hD
s
n =1
cos nz D cos nz D n
n ( x D b ) hD
n ( xD a ) hD
2 2 2 2 2 K0 + n y D hD d
2 2 2 2 2 K0 + n y D hD d
xD b ,
(8.29)
Chap. 8
lim F = hD
77
s 0
s
n =1
cos nz D cos nz D n
n (a x D ) hD
n (b xD ) hD
2 2 2 2 2 K0 + n y D hD d
2 2 2 2 2 K0 + n y D hD d
xD a ,
(8.30) and
s 0
lim F = lim F1
s 0
hD
s
n =1
cos nz D cos nz D n a xD b .
2 2 2 2 2 2 + n 2 2 y 2 h 2 d K0 n y h d K + + D D D D 0 n (b x D ) hD n ( x D a ) h D
(8.31) where lim F1 is given below by Eq. 8.37.

s 0
vi)
The series F1 =
s
n =1
cos nz D cos nz D
exp ( n y D
) where n =
2 u + n 2 2 h D
Using the relations given in Eqs. 8.15 and 8.16, the following alternate forms for the series F1 may be obtained, respectively, for the large and small values of s (that is, for short and long times).28 When u is large,
F1 = hD 2s K
+
n =
2 2 (z D z D 2n )2 h D + yD
+ K0
2n ) (z D + z D
2 hD
2 yD
u exp u y D
)(
2s u
(8.32)
2 and when u << n 2 2 h D ,
s 0
lim F1 =
cos nz s
n =1
exp ( y ) exp ( n y h ) n D D D cos nz D n h n D
{ln 1 2 exp( y D hD )cos (z + z w ) + exp( 2 y D hD ) . 4s + ln 1 2 exp ( y D h D ) cos (z z w ) + exp ( 2 y D h D ) }

hD
(8.33)
It is also possible to derive asymptotic approximations for the series F1 . When s is large (small times), F1 may be approximated by28
Chap. 8
lim F1 = hD 2s K
+ 0
78
n =
2 2 (z D + z D 2n )2 h D + yD
+ K0
(z D z D 2n )
2 hD
2 yD
s exp s y D
) (2s
32
(8.34)
If s is sufficiently large, then Eq. 8.34 may be further approximated by lim F1 = hD K0 2s

2 2 )2 h D (z D z D + yD
s exp s y D

) (2s
32
(8.35)
The inverse Laplace transform of Eq. 8.35 yields

lim F1 = hD 2 ( z z )2 h 2 + y 2 t D yD D D D + y erfc y D . Ei D exp + D 2 t 4 4t D D 4t D 2
t D 0
(8.36)
For small s (large times), F1 may be approximated by28

lim F1 =
s 0
{ln 1 2 exp( y D hD )cos (z + z w ) + exp( 2 y D hD ) . 4s + ln 1 2 exp ( y D h D ) cos (z z w ) + exp ( 2 y D h D ) }

hD
(8.37)
vii)
The ratio R1 =
cosh a ( y eD ~ yD ) sinh a y eD
for small values of s (or large values of time)
By elementary considerations, the ratio R1 may be written in the following form:28

R1 = exp a ~ y D + exp a (2 y eD ~ y D ) 1 + exp 2m a y eD . m =1
{ (
]} (
(8.38)
The expression given in Eq. 8.38 provides computational advantages when s is small.
8.3. Example Applications Example 8.10: A fully penetrating, uniform-flux fracture of half-length x f located at x = 0 , y = 0 in an infinite
slab reservoir with closed top and bottom boundaries. The solution for this problem is given in Table 8.5 A. For simplicity, assuming an isotropic reservoir, choosing the ~hx , the solution becomes characteristic length as l = x f and noting that q = 2q f
2kh q
p =
K 2s
1
1
+1
2 ( x D )2 + y D
u d .
(8.39)
Chap. 8
79
We first consider the numerical evaluation of Eq. 8.39 and recognize the right hand of Eq. 8.39 as in the form of the integral I discussed above. We note from Eqs. 8.3 8.5 that Eq. 8.39 may be written in one of the following forms depending on the value of x D .
2kh
q u ( x D +1) 2 2 K0 + uy D 2s u 0
p =
u ( x D 1)
2 2 K0 + uy D
xD 1 ,
(8.40)
2kh
u (1 x D ) 2 2 p = K0 + uy D d q 2s u 0 1
u (1 x D )
2 2 K 0 + uy D d
x D 1 .
(8.41)
and
2kh q u ( x D +1) 2 2 K0 + uy D 2s u 0 1
p =
d +
u (1 x D )
2 2 K0 + uy D
d 1 x D +1 . (8.42)
The numerical evaluation of the integrals in Eqs. 8.40 8.42 for y D 0 should be straightforward by using the standard numerical integration algorithms. For y D = 0 , the polynomial approximations given by Luke29 or the power series expansion given by Eq. 8.6 should be useful. The short- and long-time asymptotic approximations of the fracture solution are also obtained by applying the relations given by Eqs. 8.9 and 8.11, respectively, to the right hand side of Eq. 8.39. This procedure yields for shorttimes
lim 2kh q
p =
4 s3 2
exp y D s ,
(8.43)
or, in real time domain,

2kh q
t D 0
lim
p =
t D exp 2
4 tD
2 yD
y erfc y D , D 2 t 2 D
(8.44)
where is given by Eq. 8.10 with a = 1 and b = +1 . At long-times, the following asymptotic approximation may be used:
lim 2kh q
s 0
p =
1 2s
( ln s + ln 4 2
1 + 2 ) + (x D , y D ,1,+1) , s
(8.45)
or, in real time domain,

2kh q
tD
lim
p =
4t D 1 ln e 2
+ 2 + (x D , y D ,1,+1) ,
(8.46)
where = 0.5772K and (x D , y D ,1,+1) is given by Eq. 8.12.
Chap. 8
80
Example 8.11: A horizontal well of length Lh located at x = 0 , y = 0 , and z = z w in an infinite slab reservoir with closed top and bottom boundaries.
The horizontal well solution for an infinite slab reservoir with impermeable boundaries is given in Table 8.5 A. ~L , the Assuming an isotropic reservoir, choosing the characteristic length as l = Lh 2 and noting that q = q h solution may be written as follows:
2kh q
p =
2kh q
p f + F ,
(8.47)
where 2kh p f
1
(q ) is the fracture solution given by the right hand side of Eq. 8.39,
+1
F is given by
F=
cos n~ z D cos n~ z wD K 0 s n =1 1
2 (x D )2 + y D n d ,
(8.48)
with
2 , n = u + n 2 2 h D
(8.49) (8.50)
~ zD = z h ,
and
~ z wD = z w h .
(8.51)
The computation of the first term in the right hand side of the Eq. 8.47 [ 2kh p f
(q ) ]
is the same as the
computation of the fracture solution given by Eq. 8.39 and has been discussed in Example 8.10 above. The computational form of the second term ( F ) in the right hand side of Eq. 8.47 is given by Eqs. 8.21 8.23. Of particular interest is the case for 1 x D +1 . In this case, from Eqs. 8.24 and 8.25, we have F ( 1 x D +1) = F1 1 s
n =1
cos n~ z D cos n~ z wD
n
, (8.52)
2 2 2 2 2 2 K0 K0 + n yD d + + n yD d n (1 x D ) n ( x D +1)
where
F1 =
s
n =1
exp ( n y D ) .
(8.53)
The computational considerations for the series F1 have been discussed above.
Chap. 8
81
Next, we consider the short- and long-time approximations of the horizontal well solution given by Eq. 8.47. To obtain a short time approximation, we substitute the asymptotic expressions for 2kh p f (q ) and F as s given, respectively, by Eqs. 8.43 and 8.27. This yields lim
2kh
q
p =
hD
K0 4s
2 2 (~ + yD zD ~ z wD )2 h D
s.

(8.54)
where is given by Eq.8.10. The inverse Laplace transform of Eq. 8.54 is given by
2kh q
t D 0
lim
p =
hD
8
2 2 (~ + yD z ~ z wD )2 h D . Ei D 4t D
(8.55)
To obtain the long-time approximation of Eq. 8.47, we substitute the asymptotic expressions for 2kh p f
(q )
and F as s 0 given, respectively, by Eqs. 8.46 and 8.29 8.31. Of particular interest, is the case for 1 x D +1 where we have
lim 2kh q
s 0
p =
1 2s
( ln s + ln 4 2
1 + 2 ) + (x D , y D ,1,+1) s
hD 4s
{ln[1 2 exp( y D
h D ) cos (~ zD + ~ z wD ) + exp ( 2 y D h D )
h zD ~ z wD ) + exp ( 2 y D h D ) } D + ln 1 2 exp ( y D h D ) cos (~ s
n =1
cos n~ z D cos n~ z wD , (8.56) n
2 2 2 2 2 2 2 2 2 2 K0 K0 + n y D h D d + + n y D h D d n (1 x D ) hD n ( x D +1) hD
where = 0.5772K and (x D , y D ,1,+1) is given by Eq. 8.12. The inverse Laplace transform of Eq. 8.56 yields
2kh q
t D
lim
p =
4t D 1 ln 2 e
+ 2 + (x D , y D ,1,+1) h D ) cos (~ zD + ~ z wD ) + exp ( 2 y D h D )
hD 4
{ln[1 2 exp( y D
h zD ~ z wD ) + exp ( 2 y D h D ) } D + ln 1 2 exp ( y D h D ) cos (~
n =1
. (8.57) cos n~ z D cos n~ z wD n
2 2 2 2 2 2 2 2 2 2 K0 K0 d + n y D hD d + + n y D hD n (1 x D ) hD n ( xD +1) hD
Example 8.12: A fully penetrating, uniform-flux fracture of half-length x f in an isotropic and closed cylindrical
reservoir. The center of the fracture is at r = 0 , = 0 and the fracture tips extend from r = x f , = + to
(r = x f , = ) .
Chap. 8
82
The solution for this problem has been obtained in Example 8.6, Eq. 8.163 with h w = h . Choosing the characteristic ~hx , the solution is given by length as l = x , and noting that q = 2q
f f
2kh
1 1 2 2 2rD rD cos ( ) drD p = K 0 u rD + rD q 2s 0
2 2 2rD rD cos ( ) drD + K 0 u rD + rD
.
1
(8.58)
k =
[cos k ( ) + cos k ( )]
Ik
( u r )K ( u r ) I ( u r )dr I ( ur )
D k eD k D D k eD 0
We first note that for the computation of the pressure responses at the center of the fracture ( rD = 0 ), Eq. 8.58 simplifies to
2kh q
p =
1 s
[K (
1 0 0
+ I0 u rD
) ( u r )K ( u r ) I ( u r )]dr .
D 1 eD 1 eD D
(8.59)
It is also possible to find a very good approximation for Eq. 8.58 especially when reD is large. If we assume19
( 1)n K k (
u reD I k
) ( ur )= K ( ur ) I ( ur )
eD 0 eD 0 eD
(8.60)
and use the following relation4

2 2 2rD rD cos ( ) = ( 1)n I n + rD I 0 u rD n =
( u r )I ( u r )cos n( ) ,
D n D
(8.61)
then, Eq. 8.58 may be written as

2kh q
p =
1 1 2 2 2rD rD cos ( ) drD K 0 u rD + rD 2s 0
1 2 2 2rD rD cos ( ) drD + K 0 u rD + rD 0
1 1 K 1 u reD 2 2 2rD rD cos ( ) drD + I 0 u rD + rD 2 s I 1 u reD 0
( (
) )
(8.62)
2 2 2rD rD cos ( ) drD + I 0 u rD + rD 0
Because19
Chap. 8
+ cos
)d = Z ( uR
1
83
cos
2 2 2 rD rD cos ( ) drD + Z u rD + rD
0 1
2 2 2rD rD cos ( ) drD u rD + rD
(8.63)
where
= (x x )2 + ( y y tan ) , R D D D D D
(8.64)
Eq. 3.247 may be recast into the following form:

2kh q
p =
1 2s
+ cos
cos
[K (
0
+I uR 0 D
)K ( u r ) I ( u r )]d . ) ( uR
D 1 eD 1 eD
(8.65)
Although the assumption given in Eq. 8.60 may not be justified by itself, the solution given in Eq. 8.65 is a very good approximation for Eq. 8.58, especially when reD is large. For a fracture at the center of the cylindrical drainage region, Eq. 8.65 simplifies to
2kh q
p =
K 2s
1
1
+1
2 ( x D )2 + y D d + I 0
2 ( x D )2 + y D K ( 1
u reD
) I ( u r ) d .
1 eD
(8.66) It is also possible to obtain short- and long-time approximations for the solution given in Eq. 8.66. For short times, u and the second term in the argument of the integral in Eq. 8.66 becomes negligible compared to the first term. Then, Eq. 8.66 reduces to the solution for an infinite slab reservoir given by Eq. 8.39 of which the short-time approximation has been discussed above in Example 8.10. To obtain a long-time approximation, we evaluate Eq. 8.66 at the limit as s 0 ( u s ). As shown in Section 3.2.3, for small arguments, we may approximate the Bessel functions in Eq. 8.66 by
K 0 (z ) = ln e z 2 , K 1 (z ) = 1 z + (z 2 ) ln e z 2 1 2 , I 0 (z ) = 1 + z 2 4 ,
(8.67)
[(
(8.68) (8.69)
and
I 1 (z ) = z 2 + z 3 16 ,
(8.70)
where = 0.5772K . Then, using Eqs. 8.67 8.70 and neglecting the terms of the order s 3 2 , we may write28
lim K 0
s 0
[ (
s~ rD + I 0
) (
s~ rD K 1
) (
s reD I 1
) (
s reD =
)]
2
2 reD s
+ ln
~ reD 3 r2 + D . 2 ~ rD 4 2reD
(8.71)
If we substitute the right hand side of Eq. 8.71 into Eq. 8.66, we obtain
Chap. 8
2kh q 1 3 1 + (x D , y D ,1,+1) (x D , y D , reD ) + ln reD + + , 2 2 reD s s s s 4 2
84
s 0
lim
p =
(8.72)
where (x D , y D ,1,+1) is given by Eq. 8.12 and

2 2 2 (x D , y D , reD ) = y D 2reD + (x D 1)3 (x D 1)3 12reD .
( ) [
](
(8.73)
The inverse Laplace transform of Eq. 8.72 yields the following long-time approximation for a uniform-flux fracture at the center of a closed square:
lim
2kh
q
p =
2t D
2 reD
+ ln reD
3 4
+ 1 + (x D , y D ,1,+1) + (x D , y D , reD ) .
(8.74)
Example 8.13: A fully penetrating, uniform-flux fracture of half-length x f
in an isotropic and closed
parallelepiped reservoir of dimensions x e y e h . The fracture is parallel to the x axis and centered at xw , yw ,zw .
The solution for this problem has been obtained above in Example 8.8 and, by choosing l = x f , is given by
p Df = + 2kh q 2 x eD
p 1 =
cosh u ( y eD y D y wD ) + cosh u [ y eD ( y D + y wD )] x eD s u sinh u y eD
k sin k x
1
k =1
1
eD
cos k
x wD x eD
cos k
xD x eD
(8.75)
cosh k ( y eD y D y wD ) + cosh k [ y eD ( y D + y wD )] k sinh k y eD
where
2 . k = u + k 2 2 x eD
(8.76)
First, note that the computation of the ratios of the hyperbolic functions in Eqs. 8.75 may be difficult especially when their arguments approach zero or infinity. When s is small (long times), Eq. 8.38 should be useful to compute the ratios of the hyperbolic functions. When s is large (small times), using Eq. 8.38, the solution given in Eq. 8.75 may be written as28
p Df = p D1 + p Db1 + p Db2 ,
where
(8.77)
Chap. 8
p D1 = = 2
85
k sin k x s
1
k =1
1
eD
cos k xD x eD
xD x eD
cos k
x wD exp ( k y D y wD x eD
x eD
x wD +1
cos k
cos k
exp ( k y D y wD x D x eD
k =1 x wD 1
dx D
,
u
K 2s 1
1 k =
+1 +
(x D x wD 2kx eD )2 + ( y D y wD )2
+ K0
(x D + x wD 2kx eD )2 + ( y D y wD )2
u d exp u y D y wD
) (x
eD s
(8.78)
p Db1 =
x eD s u
{exp[
u ( y D + y wD ) + exp u (2 y eD y D y wD )
]
( )
, (8.79)
+ exp u y D y wD + exp u (2 y eD y D y wD
)]}
exp 2m u y eD 1 + m =1
and
p Db 2 = 2
{{exp[ k ( y D + y wD )] + exp[ k (2 y eD y D y wD )]
+ exp ( k y D y wD ) + exp k (2 y eD y D y wD
k s
k =1
1
k
sin k
1 x eD
cos k
xD x eD
cos k
x wD x eD
(8.80)
)]}1 + exp( 2m k y eD )
m =1
The last equality in Eq. 8.79 follows from the relation given by Eq. 8.23. The expression given in Eq. 8.79 may also be written as
p D1 = p Di + p Db 3 ,
(8.81)
where
p Di =
K 2s
1
1
+1
(x D x wD )2 + ( y D y wD )2
u d
(8.82)
and
Chap. 8
p Db3 = +
86
K 2s
1
1
+1
(x D + x wD )2 + ( y D y wD )2
0
u d u
K 2s 1
k =1 1
+ +1
(x D x wD 2kx eD )2 + ( y D y wD )2
u u
+ K0 + K0 + K0
(x D + x wD 2kx eD )2 + ( y D y wD )2 (x D x wD + 2kx eD )2 + ( y D y wD )2 (x D + x wD + 2kx eD )2 + ( y D y wD )2
u d exp u y D y wD
) (x
eD s
(8.83) Therefore, the solution given by Eq. 8.75 may be written as follows for computation at early times (for large values of s ):
p Df = p Di + p Db ,
(8.84)
where p Di is given by Eq. 8.82 and corresponds to the solution for a fractured well in an infinite slab reservoir (see Eq. 8.39 in Example 8.10) and p Db represents the contribution of the lateral boundaries and is given by
p Db = p Db1 + p Db 2 + p Db 3 .
(8.85)
In Eq. 8.85, p Db1 , p Db 2 , and p Db3 are given, respectively, by Eqs. 8.80 8.82. The integrals appearing in Eqs. 8.82 and 8.83 may be evaluated by following the lines outlined by Eqs. 8.3 8.5. It is also possible to derive short- and long-time approximations for the fracture solution in a closed rectangular parallelepiped. The short-time approximation corresponds to the limit of the solution as s . It can be easily shown that the p Db term in Eq. 8.84 becomes negligible compared to the p Di term for which a short-time approximation has been obtained in Example 8.10 (see Eqs. 8.43 and 8.44). To obtain a long-time approximation (small values of s ), the solution given in Eq. 8.75 may be written as follows:28
p Df = H +
k sin k x s
1
k =1
1
eD
cos k
x wD x eD
cos k
xD x eD
cosh k ( y eD y D y wD ) + cosh k [ y eD ( y D + y wD )] k sinh k y eD
(8.86)
where
H=
x eD s
cosh u ( y eD y D y wD ) + cosh u [ y eD ( y D + y wD )]
u sinh u y eD
2 2 1 cos m y D y wD + cos m y D + y wD = + 2 x eD y eD su x eD y eD s m =1 m y eD y eD
(8.87)
and
Chap. 8
2 . m = u + m 2 2 y eD
87
(8.88)
The second equality in Eq. 8.87 results from the formula30
k
k =1
cos kx
2
+a
cosh a ( x )
2a sinh a
1 2a 2
0 x 2 .
(8.89)
For small values of s , replacing u by s and s + a by a , and using30
k =1
cos kx k2
2
6
x
2
x2 4
0 x 2 ,
(8.90)
the term H given by Eq. 8.87 may be approximated by

lim H =
2
x eD y eD s 2
s 0
1 cos m y D y wD + cos m y D + y wD 2 x eD s m =1 m y eD y eD 2 y eD
2y eD 1 y D y wD + y D + y wD ( y D y wD )2 + ( y D + y wD )2 = + + 2 2 x eD y eD s x eD s 3 2 y eD 4 y eD 2
(8.91)
2 The long-time approximation of the second term in Eq. 8.86 is obtained by assuming u << k 2 2 x eD and thus taking the inverse Laplace transform of the resulting expressions, we obtain the following long-time approximation:
lim p Df = +
2t D x eD y eD 2 x eD
2y eD 1 y D y wD + y D + y wD ( y D y wD )2 + ( y D + y wD )2 + 2 x eD y y 3 2 4 eD eD 1
2
k
k =1
sin k
1 x eD
cos k
x wD x eD
cos k
xD x eD
(8.92)
cosh k ( y eD y D y wD ) + cosh k [ y eD y sinh k eD x eD
( y D + y wD )]
Example 8.14: A uniform-flux horizontal well of length Lh in an isotropic and closed parallelepiped reservoir of dimensions x e y e h . The center of the well is at x w , y w , z w and the well is parallel to the x axis.
The solution for this problem has been obtained above in Example 8.9 and, by choosing l = Lh 2 , is given by
2kh q
p = p Df + F1 ,
(8.93)
where p Df is the solution discussed in Example 8.13 and F1 is given by
Chap. 8
F1 = 2 x eD cos n~ z s
n =1
88
cos n~ z wD
cosh n ( y eD y D y wD ) + cosh n [ y eD ( y D + y wD )] n sinh n y eD + 4 s
n =1
k sin k x
1
k =1
.
xD x eD
(8.94)
1
eD
cos k
x wD x eD
cos k
cosh ( y y y ) + cosh [ y ( y + y )] k ,n eD D wD k ,n eD D wD k ,n sinh k ,n y eD
In Eq. 8.94, ~ z D and ~ z wD are given by Eqs. 8.50 and 8.51, respectively,
2 , n = u + n 2 2 h D
(8.95)
and
2 2 . k ,n = u + k 2 2 x eD + n 2 2 h D
(8.96)
The computation and the asymptotic approximations of the p Df term have been discussed above in Example 8.13. To compute the F1 term for long times (small s ), the relation for the ratios of the hyperbolic functions given by Eq. 8.38 should be useful. For computations at short times (large values of s ), following the lines similar to those in Example 8.13, the F1 term in Eq. 8.299 may be written as
F1 = F + Fb ,
where
F= 1 cos n~ z D cos n~ z wD K 0 s n =1 1
(8.97)
+1
(x D x wD )2 + ( y D y wD )2 n d ,
(8.98)
Fb = Fb1 + Fb2 + Fb3 ,

Fb1 = 2 x eD
(8.99)
s
n =1
1
n
{exp[ n ( y D + y wD )] + exp[ n (2 y eD y D y wD )] cos n~ z D cos n~ z wD {
+ exp n (2 y eD y D y wD + exp ( n y D y wD
)]}1 + exp( 2m n y eD )

m =1
(8.100)
) exp( 2m n y eD )
m =1
Chap. 8
Fb 2 = 4
89
{{exp[ k ,n ( y D + y wD )] + exp[ k ,n (2 y eD y D y wD )]
+ exp k ,n (2 y eD y D y wD + exp ( k ,n y D y wD
s
n =1
k =1
1
k ,n k
sin k
1 x eD
cos k
xD hD
cos k
x wD hD
)]}1 + exp( 2m k ,n y eD )

m =1
(8.101)
) exp( 2m k ,n y eD )
m =1
and
Fb3 = 1 cos n~ z s
n =1
cos n~ z wD
K0 1
+1
(x D + x wD )2 + ( y D y wD )2 n d
K
k =1 1
+ +1
(x D x wD 2kx eD )2 + ( y D y wD )2 n .

(8.102)
+ K0 + K0 + K0
(x D + x wD 2kx eD )2 + ( y D y wD )2 n (x D x wD + 2kx eD )2 + ( y D y wD )2 n
(x D + x wD + 2kx eD )2 + ( y D y wD )2 n d
The computational form of the F term in Eq. 8.98 is obtained by applying the relations given by Eqs. 8.21 8.23 and Eq. 8.32. Of particular interest is the case for 1 x D +1 and y D = y wD given by
~ ~ F ( 1 x D +1) = F1 cos nz D cos nz wD K 0 ( )d + K 0 ( )d , s n =1 n n (1 x D + x wD ) n ( x D xwD +1)
(8.103) where
F1 = hD 2s K (~ z
+ 0
n =
~ z wD 2n h D u + K 0 ~ zD + ~ z wD 2n h D u
, 2s u
(8.104)
and
F1 =
n =1
exp ( n y D ) .
(8.105)
Similarly, for 1 x D +1 and y D = y wD , the Fb3 term given in Eq. 8.102 may be written as
Chap. 8
Fb3 = + 1
90
s
n =1 + k =1
1
n
cos n~ z D cos n~ z wD {Ki1 [(x D + x wD + 1) n ] Ki1 [(x D + x wD 1) n ] x D + x wD + 1) n ] Ki1 [(2kx eD x D + x wD 1) n ]
{Ki [(2kx
1
eD
+ Ki1 [(2kx eD + x D + x wD + 1) n ] Ki1 [(2kx eD + x D + x wD 1) n ]}}
+ Ki1 [(2kx eD x D x wD + 1) n ] Ki1 [(2kx eD x D x wD 1) n ]
+ Ki1 [(2kx eD + x D x wD + 1) n ] Ki1 [(2kx eD + x D x wD 1) n ]
(8.106)
where
Ki1 (x ) = K 0 ( )d .
(8.107)
The short- and long-time approximations of the p Df term in Eq. 8.93 have been discussed in Example 8.13. A small-time approximation for F1 given by Eq. 8.97 is obtained by using u = s and noting that as s ,
F + Fb F . Then, substituting the short-time approximations for p Df and F given, respectively, by Eqs. 8.43
and 8.27 into Eq. 8.93, the following short-time approximation is obtained:28 lim 2kh q
p =
hD
2 + ( y D y wD )2 s , K 0 (~ zD ~ z wD )2 h D 4s
(8.108)
where is given by Eq. 8.10. The inverse Laplace transform of Eq. 8.108 yields
2kh q
t 0
lim
p =
hD
8
2 (~ + ( y D y wD )2 z ~ z wD )2 h D . Ei D 4 t D
(8.109)
The long-time approximation of Eq. 8.93 is obtained by substituting the long-time approximations of p Df and F1 . The long time-approximation of p Df is obtained in Example 8.13 (see Eq. 8.86 8.91). The long-time approximation of F1 is obtained by evaluating the right hand side of Eq. 8.94 as s 0 ( u 0 ) and is given by
F1 = 2h D x eD z n cos n~ 1
n =1
cos n~ z wD
cosh n ( y eD y D y wD ) + cosh n [ y eD ( y D + y wD )] n sinh n y eD + 4 s
n =1
k sin k x
1
k =1
,
xD x eD
(8.110)
1
eD
cos k
x wD x eD
cos k
cosh ( y y y ) + cosh [ y ( y + y )] k ,n eD D wD k ,n eD D wD k ,n sinh k ,n y eD
where
Chap. 8
n = n h D ,
91
(8.111)
and
k ,n = k x eD + n h D .
Thus, the long-time approximation Eq. 8.93 is given by
2kh q
(8.112)
p = p Df + F1 ,
(8.113)
where p Df and F1 are given, respectively, by Eqs. 8.92 and 8.110. For computational purposes, however, F1 may be replaced by
F1 = F + Fb1 + Fb 2 + Fb3 .
(8.114)
In Eq. 8.114, F , Fb1 , Fb 2 , and Fb3 are given, respectively, by

F= cos n~ z D cos n~ z wD K 0 n n =1
1
+1
(x D x wD )2 + ( y D y wD )2 d ,
(8.115)
Fb1 =
2h D x eD
z n cos n~ 1
n =1
{exp[ n ( y D + y wD )] + exp[ n (2 y eD y D y wD )] cos n~ z wD {
+ exp n (2 y eD y D y wD + exp ( n y D y wD

)]}1 + exp( 2mn y eD )

m =1
(8.116)
) exp( 2mn y eD )
m =1
Fb 2 = 4
z cos n~
n =1
cos n~ z wD
{{exp[ k ,n ( y D + y wD )] + exp[ k ,n (2 y eD y D y wD )]
+ exp k ,n (2 y eD y D y wD + exp ( k ,n y D y wD
k =1
1
k ,n k
sin k
1 x eD
cos k
xD hD
cos k
x wD hD
)]}1 + exp( 2mk ,n y eD )

m =1
(8.117)
) exp( 2mk ,n y eD )
m =1
and
Chap. 8
Fb3 = 1 cos n~ z 2
n =1
92
cos n~ z wD
K0 1
+1
(x D + x wD )2 + ( y D y wD )2 n d
K
k =1 1
+ +1
(x D x wD 2kx eD )2 + ( y D y wD )2 n .

(8.118)
+ K0 + K0 + K0
(x D + x wD 2kx eD )2 + ( y D y wD )2 n (x D x wD + 2kx eD )2 + ( y D y wD )2 n
(x D + x wD + 2kx eD )2 + ( y D y wD )2 n d
In the computation of the integrals and the trigonometric series, the relations given by Eqs. 8.3 8.5 and 8.21 8.23 are useful.
References: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. Al-Hussainy, Ramey, H. J., Jr., and Crawford, P. B.: The Flow of Real Gases Through Porous Media, Jour. Pet. Tech. (May 1966) 624-636. Raghavan, R.: Well Test Analysis, Prentice Hall Petroleum Engineering Series, Englewood Cliffs, New Jersey (1993) 28-31, 336-435. Fancher, G. H., Lewis, J. A., and Barnes, K. B.: Some Physical Characteristics of Oil Sands, Min. Ind. Expt. Station, Pennsylvania State College Bull. (1933) 12 322-333. Forcheimer, P.: Wasserbewegung durch Boden, Z. Ver. Deutch (1901) 45 1782-1788. Watson, G. N.: A Treatise on the Theory of Bessel Functions, The University Press, Cambridge (1944). Churchill, R. V.: Operational Mathematics, McGraw-Hill Book Company, New York (1972) 2. Abramowitz, M. and Stegun, I. A.: Handbook of Mathematical Functions, Dover Publications, Inc., New York (1972) 1020-1029. Stehfest, H.: Numerical Inversion of Laplace Transforms, Communications ACM, 13 No. 1 (1970) 47-49. Crump, K. S.: Numerical Inversion of Laplace Transforms Using a Fourier Series Approximation, J. ACM 233 (1976) 89-96. van Everdingen, A. F. and Hurst, W.: The Application of the LaPlace Transformation to Flow Problems in Reservoirs, Trans. AIME 198 (1953) 171-176. Agarwal, R. G., Al-Hussainy, R., and Ramey, H. J., Jr.: An Investigation of Wellbore Storage and Skin Effect in Unsteady Liquid Flow I. Analytical Treatment, SPEJ (Sept. 1970) 279-290. Chen, C. C. and Raghavan, R.: An Approach to Handle Discontinuities by the Stehfest Algorithm, SPEJ (Dec. 1996) 363. Gringarten, A. C. and Ramey, H. J. Jr.: The Use of Source and Greens Functions in Solving Unsteady-Flow Problems in Reservoirs, SPEJ (Oct. 1973) 285-296. Carslaw, H. S. and Jaeger, J. C.: Conduction of Heat in Solids, Oxford University Press, Oxford (1959) 353386. Stakgold, I.: Greens Functions and Boundary Value Problems, John Wiley & Sons, New York (1979) 86-104. Lord Kelvin (Sir William Thomson): Mathematical and Physical Papers, Cambridge at the University Press (1884) Vol. II, 41. Ozkan, E. and Raghavan, R.: New Solutions for Well-Test-Analysis Problems: Part 1 Analytical Considerations, SPEFE (Sept. 1991) 359-368. Ozkan, E. and Raghavan, R.: New Solutions for Well-Test-Analysis Problems: Part 2 Computational Considerations and Applications, SPEFE (Sept. 1991) 369-378. Raghavan, R. and Ozkan, E.: A Method for Computing Unsteady Flows in Porous Media, Longman Scientific & Technical (1994). Raghavan, R.: The Method of Sources and Sinks A Perspective, SPE Memorial Series, Volume 1, Richardson, TX (1995) 135-143. Nisle, R. G.: The Effect of Partial Penetration on Pressure Build-Up in Oil Wells, Trans., AIME (1958) Vol. 213, 85-90. Newman, A. B.: Heating and Cooling Rectangular and Cylindrical Solids, Ind. and Eng. Chem. (1936) Vol. 28, 545. Ozkan, E., Sarica, C., Haciislamoglu, M., and Raghavan, R.: Effect of Conductivity on Horizontal Well Pressure Behaviour, SPE Advanced Technology Series, Vol. 3, No: 1, March 1995, 85-94. Duhamel, J. M. C.: Mmoire sur la mthode gnrale relative au mouvement de la chaleur dans les corps solides polong dans les milieux dont la temprature varie avec le temps, J. de Ec. Polyt. (Paris) 14 (1833) 2077. Chen, H.-Y., Poston, S. W., and Raghavan, R.: An Application of the Product-Solution Principle for Instantaneous Source and Greens Functions, SPEFE (June 1991) 161-168. Warren, J. E. and Root, P. J.: The Behavior of Naturally Fractured Reservoirs, SPEJ (Sept. 1963) 245-255; Trans., AIME, 228. De Swaan-O., A.: Analytical Solutions for Determining Naturally Fractured Reservoir Properties by Well Testing, SPEJ (June 1976) 117-122; Trans., AIME, 261. Ozkan, E.: Performance of Horizontal Wells, Ph.D. dissertation, U. of Tulsa, Tulsa, OK (1988). Luke, Y. L.: Integrals of Bessel Functions, McGraw Hill Book Co., Inc., New York City (1962). Gradshteyn, I. S. and Ryzhik, I. M.: Table of Integrals, Series, and Products, Academic Press, Inc., Orlando (1980).
25. 26. 27. 28. 29. 30.

Applied Mathematics of Fluid Flow in Porous Media

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applied Mathematics of Fluid Flow in Porous Media

Uploaded by

Copyright:

Available Formats

APPLIED MATHEMATICS OF FLUID FLOW IN POROUS MEDIA

Erdal Ozkan, Ph.D.

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2005

Chapter 2 EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

Then, the time rate of change of mass within S is

EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

and the total system compressibility, c t , is given by

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

Then, Eq. 2.19 can be written as

k k j , where k may be chosen arbitrarily (to satisfy

k x k y k z ), Eq. 2.21 may be transformed into the diffusion equation

where is the diffusivity constant defined by

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

Here, the pseudopressure is defined by1

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

EQUATIONS OF TRANSIENT FLUID FLOW IN POROUS MEDIA

2.3. Assumptions and Limits

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

Chapter 3 BESSELS EQUATION AND BESSEL FUNCTIONS

BESSEL'S EQUATION AND BESSEL'S FUNCTIONS

In Eq. 3.5, (x ) is the gamma function defined by

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

BESSEL'S EQUATION AND BESSEL'S FUNCTIONS

Figure 3.1 shows these functions graphically.

BESSEL FUNCTION I0(z), K0(z), I1(z), K1(z)

Fig. 3.1 Bessel Functions of order zero and one.

where = 0.5772K , and

Also, for large arguments, the following relations may be useful:

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

BESSEL'S EQUATION AND BESSEL'S FUNCTIONS

(3.21) (3.22) (3.23) (3.24)

(3.25) (3.26) (3.27) (3.28)

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

Chapter 4 LAPLACE TRANSFORMATION

where H (t t 0 ) is Heavisides unit function defined by

for t < t 0 for t > t 0

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

k N 2 (2k )! . [(N 2) k ]! k!(k-1)!(i-k )!(2k-i )!

(4t )]} t 2 (4t )]} (2t )

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

4.2. Example Applications

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

From Eqs. 4.23 and 4.25, we obtain

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

FINITE WELLBORE RADIUS

1E-1 1E-1 1E+0 1E+1 1E+2 1E+3 1E+4 1E+5

we obtain the line source solution as follows:

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

The Laplace transforms of Eqs. 4.37 4.39 yield, respectively,

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

Using the outer boundary condition given by Eq. 4.42, we obtain

Using the inner boundary condition given by Eq. 4.41 yields

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

In Eq. 4.51, 1 , 2 , etc. are the roots of

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

(4.59) (4.60) (4.61)

The general solution of Eq. 4.59 is

Applied Mathematics of Fluid Flow in Porous Media, Erdal Ozkan, 2003

The use of the condition in Eq. 4.56 requires that C1 = 0 . Therefore,

Using Eq. 4.63 in Eq. 4.62 yields