Scribd
Upload
60 views

Interference Test: Tom Aage Jelmert Ntnu Department of Petroleum Engineering and Applied Geophysics

1) An interference test involves producing from one well and measuring pressure responses in one or more observation wells to determine communication between wells. 2) Type curve analysis is commonly used to analyze interference test data by creating dimensionless variables and matching the field data curve to a standard type curve curve. 3) Key parameters like permeability and storage capacity can be calculated from the matched points on the type curve using the equations relating the dimensionless groups.

Uploaded by

Muhammad Aamer ishfaq
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
60 views

Interference Test: Tom Aage Jelmert Ntnu Department of Petroleum Engineering and Applied Geophysics

1) An interference test involves producing from one well and measuring pressure responses in one or more observation wells to determine communication between wells. 2) Type curve analysis is commonly used to analyze interference test data by creating dimensionless variables and matching the field data curve to a standard type curve curve. 3) Key parameters like permeability and storage capacity can be calculated from the matched points on the type curve using the equations relating the dimensionless groups.

Uploaded by

Muhammad Aamer ishfaq
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Tom Aage Jelmert

NTNU
Department of Petroleum Engineering
and Applied Geophysics

INTERFERENCE TEST
An interference test involves more than one well. It is conducted by producing at least
one well and by measuring the pressure response in at least one observation well. The
observation well has a pressure gage to record pressure. The flow rate is zero.

Objectives:
o Determine whether two or more wells are in pressure communication.
o When communication exists, provide estimates of the average
permeability, k, and the average storage capacity, φct.
o Investigate the directional properties of the permeability (anisotropy).
In the latter case more than one observation well is necessary.

For the purpose of this discussion we will use a simplified model. Let us assume that
we have two wells, one observation and one production well in an infinite reservoir.
The observation well does not produce. The two wells are separated by a distance r.

The effect of the producing well at the observation well is given by the line-source
solution. It is usually assumed that a possible skin around the production well has
no effect on the pressure measured in the observation well. Hence:

qµ B  φµ ct r 2 
∆p = pi − p ( r ) = − Ei  − 
4π kh  4kt 

where p(r) denotes the pressure at position r at time t. There is a difference between
the effect of a line-source and cylindrical for small values of time. The line-source
solution is valid when:

kt
> 25
ϕµ ct r 2

Due to the fact that the distance r is large, it may take a long time before the time
criterion for the applicability of the logarithmic approximation to the Ei solution is
valid. It may not be economical to run the test for such a long time period. Hence
semilog analysis may not be possible.

A convenient way to analyze such tests is type curve analysis. A type curve is a log-
log plot of the mathematical solution of a test in dimensionless form.

1
In the following we will define three dimensionless groups, dimensionless
radius, time and pressure. The dimensionless groups are defined such that each
dimensionless parameter is proportional to the corresponding real one. Hence

pD = α∆p
tD = βt

Taking the logarithm on both sides yields:

log pD = log ∆p + log α


and

log tD = log t + log β

The important consequence of the above equations is that a log pD vs log tD and a log
∆p vs. log t plot are exactly identical in shape, but shifted with respect to each other
when plotted on the same log-log scale.

The translation factors, α and β, are proportional to the well and reservoir
parameters. These can be determined by the type curve matching for a given reservoir.
With known α and β, their defining equations provide two equations that can be
used to compute two unknowns.

The line-source solution may be written in dimensionless form as follows:

 
1  r  2
1  1 
pD = − Ei  −  = − D
Ei − 
2  4t 'D  2  4t D 
 rD2 
 

The conventional definitions of dimensionless variables are:

2πkh
pD = ( pi − p (r ))
qµB

r
rD =
rw

kt
tD =
φ µ ct rw2

Tom Aage Jelmert INTERFERENCE TEST 2


Dimensionless variables, however, are not unique. One may have several definitions
for a group. Let us define a new dimensionless time as follows:

kt
t D′ =
φµ ct r 2

Multiplication and division with rw on the right hand side yields.

kt rw2
t D′ =
φµ ct rw 2 r2

tD
t ′D =
rD2

tD
Hence the group may be regarded as a dimensionless time with basis in r.
rD2

tD
The type curve for the Ei-solution is usually presented as a log pD vs. log graph.
rD2
Then the line-source solution (for any distance) will show up as a single curve.

A schematic of the matched curves is shown below.

Figure 1: Type curve matching

Procedure:
a) Locate a log-log plot of the line source-solution. The dimensionless graph
is referred to as the type curve.
b) Plot the field curve, ∆p = pi - p(r) vs. t, on log-log transparent paper. Use the
same scale as for the type curve.

Tom Aage Jelmert INTERFERENCE TEST 3


c) Slide the field curve horizontally and vertically on top of the type curve until a
match is found. The grid axes must be parallel while sliding.
d) Pick a convenient match point. Mark both graphs, use a needle.
e) Calculate the permeability from the pressure match:

p Dm 2πkh
=
∆p m qµB
f) Calculate the storage capacity, φc, from the time match:

 tD 
 2
 rD  m k
=
tm φµ ct r 2

Index m denotes match-point. The variables denoted by index m are known quantities.
Hence one unknown may be determined from each of the above equations.

Note that the above equations derive from the definitions of the dimensionless
variables. This is because these define the translation factors, α and β. There is
a difference between the response of a cylindrical and a line-source well for small
values of time. Hence it may not be possible to match the field curve to the entire field
curve. In such cases, disregard the few earliest points.

Tom Aage Jelmert INTERFERENCE TEST 4

You might also like