Well Test Analysis and

Interpretation

TYBTech Petroleum Engineering

 MODULE II
Solutions to Diffusivity
Equation for Well Test
Analysis
 Diffusivity Equation
• Diffusivity equation was derived by combining the following fundamental laws:
• Law of Conservation of Mass (Continuity Equation)
• Darcy’s Law (Transport Equation)
• Equation of State
• Diffusivity equation is given by:

Where: p = pressure, psia

r = radius, ft
Ø = porosity
μ = viscosity, cp
Ct = Total system compressibility, 1/psia
k = permeability, md
t = time in hours
Assumptions:
Diffusivity
Equation
 Diffusivity Equation
• For non-ideal gas, diffusivity equation is given by:

Where z = gas deviation factor

For simultaneous flow of oil, gas and water,

Where = total system compressibility given by:

And total mobility

 There are four solutions to the
diffusivity equation commonly
employed in well testing:
1. Solution for a bounded cylindrical
reservoir
Solutions to
2. Solution for an infinite reservoir
the Diffusivity with a well considered to be a line
Equation source with zero wellbore radius
3. Pseudo-steady state solution
4. Solution that includes wellbore
storage for a well in an infinite
reservoir
 • This solution requires two boundary
conditions and one initial condition
specified
1. Well produces at a constant rate,
1. Bounded qB
Cylindrical 2. The well with wellbore radius rw, is
Reservoir centered in a cylindrical reservoir of
radius re and that there is no flow
across the outer boundary
3. Before the production begins, the
reservoir is at uniform pressure pi
 • The solution is given as follows:

Bounded
Cylindrical
Reservoir • Where, and
Solution
• And, are the roots of

• J1 and Y1 are Bessel functions

 • Assume that:
1. Well produces at a constant rate qB
2. The well has zero radius
3. The reservoir is at uniform pressure,
pi
2. Infinite 4. The well drains an infinite area (i.e. p
Cylindrical ->pi as r -> r infinite)
Reservoir • Under these conditions, the solution is given
with Line as follows:
Source Well
Where,

And p (psi) is the pressure at radius r (ft) from

the well at time t (hours)
 • Analysis of these solutions show that
Ei function solution is an accurate
approximation to the more exact
2. Infinite solution for the time
Cylindrical
Reservoir with • For time less than , the zero well size
Line Source limits the accuracy of the equation
Well • At times greater than , the reservoir’s
boundaries begin to affect pressure
distribution in the reservoir, therefore
reservoir is no more infinite acting
 • For further simplification of the solution to
the flow equation we can assume that:
2. Infinite
• For x < 0.02, Ei(-x) = ln (1.781x)
Cylindrical
• For 0.02<x<10.9, Ei function table can
Reservoir
be used for calculating the value for the
with Line function
Source Well • For x > 10.9, Ei(-x) can be considered
zero for application in well testing
 • For wells which have formation damage or
have undergone stimulation, the solution
needs a modification
2. Infinite
Cylindrical • Introduction of pressure drop due to skin
can be useful for better estimate of flowing
Reservoir pressures
with Line
Source Well
Therefore,
 • Combining the solution to diffusivity equation with the equation
for pressure drop due to skin,
+
=]
2. Infinite • For r = rw, the argument of Ei function becomes sufficiently small
Cylindrical after a short time such that we can use the logarithmic
Reservoir approximation,
with Line ]
Source Well • It is convenient to express skin factor s as follows:
-------------------------------- (A)
Thus the drawdown is given as:
]
 • Equation A provides some insights into physical significance
2. Infinite of the sign of the skin factor
Cylindrical • IF ks < k, s will be positive, greater the contrast between the
Reservoir values, deeper will be formation damage, larger the
with Line formation damage
Source Well • IF ks > k, well will be stimulated, s will be negative
• IF ks = k, the well is neither damaged nor stimulated and s =
0
 • A well and the reservoir have the following characteristics.
The well is producing only oil at a constant rate of 20 STB/D.
Data describing the well and formation are:
• µ = 0.72 cp
• K = 0.1 md
• Ct = 1.5 x 10^-5/psi
• Pi = 3000 psi
PROBLEM • re = 3000 ft
• rw = 0.5 ft
• Bo = 1.475 RB/STB
• h = 150 ft
• Ф = 0.23
• S=0
• Calculate the reservoir pressure at a radius of 1 ft after 3
hours of production, then calculate the pressure at radii 10
and 100 ft after 3 hours pf production
PROBLEM
PROBLEM
PROBLEM
 • Limiting form of diffusivity equation and describes pressure
behavior with time for a well centered in a cylindrical reservoir
of radius re
• Limiting form is that the solution is only valid for large times, so
that summation involving exponentials and Bassel function is
3. Pseudo- negligible after this time (t > )
steady ]------(B)
state
• Note that during this period, differentiating above equation, w
solution get

• Pore volume of liquid filled reservoir is given by: Vp =

• Then,
 • Another form of equation B is useful which makes use of
average reservoir pressure by replacing initial reservoir
pressure
• The volumetric average reservoir pressure can be found from
3. Pseudo- the material balance equation
steady • The pressure decrease resulting from removal of qB RB/D of
state fluid for t hours is
solution =
• Substituting in Equation B
]
• Thus,
 • Introducing skin factor for wells that are damaged or
stimulated
3. Pseudo-
steady
state ---------------(C)
solution
AND

--------------------------------(D)
 • Further we define average permeability as:

3. Pseudo- • Thus average permeability Kj is lower than the true bulk

steady permeability k in damaged wells
state • These quantities are equal in case where skin factor is zero
solution
• Productivity index (PI) of an oil well is defined as:
 • A well produces 100 STB/D of oil at measure flowing
bottomhole pressure of 1500 psi. A recent pressure survey
showed that the average reservoir pressure is 2000 psi. Logs
indicate net sand thickness of 10 ft. The well drains an area
with drainage radius re of 1000 ft, the wellbore radius is 0.25
ft. Fluid samples indicate that at current reservoir pressure
PROBLEM oil viscosity is 0.5 cp and formation volume factor is 1.5
RB/STB.
1. Estimate the PI for the tested well
2. Estimate formation permeability from these data
3. Core data indicates an effective permeability to oil of 50 md. Does this
imply that the well is either damaged or stimulated? What is the
apparent skin factor?
 • Equations C and D describe cylindrical reservoir with well
centered in it
Flow
Equations • A similar equation that includes the reservoir geometry is
given by:
for
Generalized
Reservoir Where:
Geometry A = drainage area, sq ft
= shape factor for a specific drainage area shape and well
location, dimensionless
 Table for
shape factor
for various
geometries of
the reservoir
 • For a given reservoir geometry, the maximum time a
reservoir is infinite acting can be determined using the entry
in the column “Use infinite system solution……”
Flow
Equations • Since , this means time in hours is calculated from
for
Generalized • Time required for PSS equation to be accurate within 1% can
Reservoir be found from the entry in the column “Less than 1% error
Geometry for >”

• Finally time required for the PSS solution to be exact is found

from the entry in the column “Exact for >”
 • For each of the following reservoir geometries, calculate the
time in hours for which
• Reservoir is infinite acting
• The PSS is exact
• The PSS solution is accurate within 1%
1. Well centered in a circular drainage area
2. Well centered in square drainage area
PROBLEM 3. Well centered in one quadrant of square drainage area
• In each case, A = 17.42 x 10^6 sq ft (40 acres)
• Ф = 0.2
• rw = 0.3 ft
• B = 1.2 RB/STB
• µ = 1 cp
• k = 100 md
• Ct = 10^-5/psia
4. Radial Flow
in Infinite
Reservoir with
Wellbore
Storage
 • Consider a shut in oil well at uniform and unchanging
4. Radial pressure
Flow in • Reservoir pressure will support the column of liquid inside
Infinite the well to some height
Reservoir • If the well opened to flow, the first drop of oil produced will
with be from the stored fluid inside the well
Wellbore
• The initial flow rate from the reservoir will be zero
Storage
• With increasing flow time, at a constant surface production
rate the downhole rate will approach the surface rate
 • Consider a well with liquid/gas interface in the wellbore as
shown in the diagram
• From the mass balance, the rate of liquid in is in RB/D
• The rate of liquid out is qB in RB/D
4. Radial • And the rate of fluid accumulation in the well is:
Flow in
Infinite
Reservoir • Assuming a constant wellbore area Awb and constant oil
with formation volume factor B, the balance becomes
Wellbore -------------------------------------- [1]
Storage • For a well with surface pressure pt,
---------------------------------------------------[2]
• Where ρ is the density of the fluid in the wellbore (lbm/ft3)
• And = lbf/lbm then,
--------------------------------------------------[3]
 Thus, -------------------[4]
4. Radial
Flow in Defining a wellbore storage constant as:
Infinite -------------------------------------------------------[5]
Reservoir Then,
with ------------------------------------------[6]
Wellbore
For zero or unchanging surface pressure, pt, (not necessarily a
Storage valid assumption)
------------------------------------------------[7]
 • Introducing dimensionless variables here, let qi be the
surface rate at t = 0 and introducing dimensionless time and
dimensionless pressure
4. Radial
Flow in ----------------------------------------[8]
Infinite ---------------------------------------------------[9]
Reservoir Substituting and simplifying we get,
with -----------------------------------------[10]
Wellbore
Defining dimensionless wellbore storage constant as,
Storage
------------------------------------------------[11]
Then, ----------------------------------[12]
----------------------------------------------[13]
4. Radial
Flow in • The analytical solution to Equation 13 is presented in next
Infinite slide
Reservoir • From this figure, the values of (and thus ) can be determined
with with given values of , and s
Wellbore
• Two properties of this plot should be discussed here:
Storage
4. Radial
Flow in
Infinite
Reservoir
with
Wellbore
Storage
 1. Presence of Unit Slope Line:
• At earliest times, for a given value of and for all values of
4. Radial skin, a unit slope line exists
Flow in • Remains until all production comes from the wellbore and
Infinite none from the reservoir
Reservoir • For = 0, Equation 13 becomes:
with • OR =
Wellbore
• Integrating from = 0 to and :
Storage
= OR
+ =
• Thus as long as = 0, a graph of v/s will have a slope of unity
 2. End of Wellbore Storage Distortion
• Once wellbore storage is over,
4. Radial • The solutions to the diffusivity equation would become the
same as if there had never been any wellbore storage i.e. the
Flow in same as when = 0
Infinite
• Note that the solutions for finite and = 0 become identical
Reservoir
after sufficient elapsed time
with
Wellbore • Useful empirical observation: End of wellbore storage
distortion occurs approximately one and a half log cycles
Storage after the disappearance of unit slope line
• Another useful observation is that dimensionless time at
which wellbore storage distortion ceases is given by:
 • Radius of investigation is the distance a pressure transient
travels into the formation
Radius of • The equation to calculate the ROI is given by:
Investigatio
n • Uses of ROI concept:
• To find extent of damage
• For well test design (How long should the test be?)
• To determine the length of time before stabilized (PSS) flow is seen
 Radius of
• A company wishes to run a flow test on an exploratory well for sufficiently
Investigatio long to ensure that the well will drain a cylinder of more than 1000 ft radius.
n Preliminary well and fluid analysis suggests that k = 100 md, ф = 0.2, Ct =
2x10^-5/psi, µ = 0.5 cp. What length flow test appears advisable? What flow
PROBLEM rate do you suggest?
 • Ei function solution applicable only to describe pressure distribution in an
infinite acting reservoir caused by the production of a single well
Principle of • Well producing at a constant rate at time t = 0
Superpositi • Application of superposition can remove some of these restrictions
on • Principle of superposition is given as follows:
• The total pressure drop at any point in a reservoir is the sum of the pressure
drops at the point caused by flow in each of the wells in the reservoir
 1.
Application
to Multiple
Flowing
Wells in
Same
Reservoir
 • Consider three wells, Well A, B and C
• These wells start production at the same time from an infinite reservoir
• Application of principle of superposition tells us that:
1.
Application
to Multiple
Flowing
• In terms of Ei function solutions and logarithmic approximations:
Wells in
Same =
Reservoir

• Where, , , = Rate at which well A, B and C produce respectively

2. Pressure
Behavior in
Bounded
Cylindrical
Reservoir
 • Consider a well located at a distance ‘L’ from a no-flow boundary (sealing
fault)
• Mathematically, this problem is identical to the problem of a well at a
distance ‘2L’ from an “image” well
• A line equidistant between two wells can be shown to be a no flow
boundary
2. Pressure
• Thus this is an example of two wells in an infinite reservoir
Behavior in
• In terms of Ei function solutions and logarithmic approximations:
Bounded
Cylindrical =
Reservoir

• The imaging technique can be extended to model:

1. Pressure distribution between two well intersecting at 90 deg
2. Pressure behavior of a well with two parallel boundaries
3. Pressure behavior for multiple wells surrounded by rectagular
shaped reservoir
3. Variable
Rate
Producing
Well
 • Consider a well which produces at a rate from time 0 to
3. Variable • At , the rate is changed to and at time , the rate is changed to
Rate • The problem to solve is at time t > , what is the pressure at the sandface of
Producing the well?
Well • We can apply superposition, but in this case each well that contributes to
the total pressure drawdown will be at the same location in the reservoir
• The well simply will be turned on at different times
 • First contribution is from a well producing at a rate starting at time
t=0

3. Variable
• Starting at time , the new rate is
Rate
Producing • Introducing a well 2 producing at rate - at time
Well • The total time elapsed since this well started producing is (t - )

• Similarly contribution from well 3 will be:

 • The total drawdown for a well with 2 rate changes is given by:

3. Variable = + +
Rate =
Producing
Well
 • Horner reported an approximation to avoid the use of superposition
in case of variable rate history well
• With Horner’s approximation, one can replace sequence of Ei
functions for variable rates with single Ei function with single
producing time and single producing rate
Horner’s
• Single rate is most recent non-zero rate,
Approximati
• The single producing time is found by dividing the cumulative
on production by the most recent rate
• Thus the pseudo-producing time is calculated as:

• The pressure behavior inside the well can be modeled using:

 • Basis for Approximation:
1. If we use single rate, that is most recent rate, such maintained for
a significant duration of time can determine the pressure
distribution near the wellbore and out to the distance traveled by
pressure transient
2. Given single rate to use, it is intuitive that we choose effective
production time such that product of rate and producing time
Horner’s results in correct cumulative production. This way material balance
Approximati will be maintained accurately
on • When is the approximation correct
1. If the most recent rate is maintained significantly long duration of
time, Horner’s approximation is sufficiently accurate
2. For a new well that undergoes rapid rate changes, it is advisable
to maintain last constant rate for at least twice as long as the
previous rate
3. In case of any doubt, one should revert back to application of
superposition to model the production history of a well
 • Following completion, a well is produced for a short duration of
time and shut in for a build up test. The production history was as
follows:

Production Time (hours) Total Production (STB)

25 52
PROBLEM
12 0
26 46
72 68

1. Calculate the pseudo-producing time

2. Is Horner’s approximation adequate for this well?