PETE 324
Lecture 21
Rosalind Archer
Analysis of Drawdown Tests
21.2
Infinite-acting radial flow in a
homogenous reservoir is governed by:
p wf
qB
kt
p i 162.6
3.22 0.8686s
log
2
kh
c t rw
This is of the form:
p wf p1hr m log(t )
21.3
Analysis of Drawdown Tests
where :
qB
m 162.6
kh
k
p1hr p i m log
3.22 0.8686s
2
c t rw
This implies the pressure data should form
a straight line on a semilog plot.
21.4
Analysis of Drawdown Tests
Analysis procedure
Plot all data!
Fit a straight line to the data, remembering this
solution does not account for radial flow which
may distort the early time data. If the pressure
derivative is also plotted it should be at a
constant level during infinite-acting radial flow
(m=2.303p)
Use the slope of the semilog straight line (m) to
determine permeability (k)
Use the value of p(1hr) [read from the semilog
straight line] to determine skin (s).
21.5
Analysis of Drawdown Tests
If the data departs from the semilog straight line in late
time in indicates the presence of boundaries:
21.6
Analysis of Drawdown Tests
Time can be related to radial distance in a
well test via the concept of the radius of
investigation:
rinv 2.434 10
kt
c t
This can be used to estimate the distance
to the reservoir boundary based on the
time at which the pressure data departs
from the semilog straight line.
21.7
Analysis of Buildup Tests
Pressure buildup tests involve recording
pressure data while a well is shut in after a
period of flow.
21.8
Analysis of Buildup Tests
The pressure response for a buildup can
be constructed by using superposition and
the drawdown solution.
An fake injection well (located at the same
location as the production well) is turned on at
t = tp. Its rate is -q.
t p t
qB
p ws p wf (t 0) 162.6
log(t p ) log
kh
t
k
3.22 0.8686s
log
2
c t rw
21.9
Analysis of Buildup Tests
This implies a plot of pws versus (tp+t)/t
should be a straight line. (Horner plot)
21.10
Analysis of Buildup Tests
Analysis procedure
Plot all data (p versus (tp+t)/t)
Fit straight line to middle time region (early
time is distorted by wellbore storage).
k
Find slope (m) and extrapolation
of straight line
to pws(1hr)
qB
k 162.6
mh
p (1hr ) p (t 0)
k
p
wf
s 1.151 ws
log
3.22
log
2
m
t p 1
c t rw
21.11
Analysis of Buildup Tests
Average reservoir pressure is estimated
using p* (extrapolated from the straight line
on the Horner plot) using the MatthewBrons-Hazebrok (MBH) technique (not
covered).
Remember t =t-tp i.e. time since the well
was shut-in.
21.12
Analysis of Buildup Tests
In practice the production time (tp) may be
unknown. Rather than assuming a value
for tp it is ignored. This is strictly valid if the
reservoir has been produced to
pseudosteady state conditions.
Miller-Dyes-Hutchinson (MDH) analyzed
this situation.
21.13
Analysis of Buildup Tests
qB
p ws p wf (t 0) 162.6
log t
kh
log
k
3.22 0.8686s
2
c t rw
21.14
Analysis of Buildup Tests
qB
k 162.6
mh
p ws (1hr ) p wf (t 0)
k
s 1.151
log
3.22
2
m
c t rw
21.15
Variable-Rate Tests
The pressure response of a test with a
sequence of variable flow rates can also be
constructed using superposition.
p i p wf
n
162.6
(q j q j1 ) log( t t j1 )
kh j1
q n B
k
162.6
3.22 0.8686s
log
2
kh
c t rw
21.16
Variable-Rate Tests
It is convenient to normalise both sides of
this equation by the current flowrate:
p i p wf
n (q j q j1 )
162.6
log(t t j1 )
qn
kh j1
qn
B
k
3.22 0.8686s
162.6
log
2
kh
c t rw
21.17
Variable-Rate Tests
This equation is a straight line (OdehJones plot):
y mx b
where :
n (q q
j
j1 )
x
log(t t j )
qn
j1
(p i p wf )
y
qn
B
m 162.6
kh
k
b m log
3.22 0.8686s
2
c t rw
21.18
Wellbore Storage
Definition of wellbore storage coefficient:
Fluid filled wellbore
Cs c wb Vwb
Rising or falling liquid level
144
A wb
Cs
5.615 (g / g c )
Wellbore Storage
21.19
During wellbore storage dominated flow the
following pressure-time relationships hold:
Drawdown
p wf
qB
pi
t
24Cs
Buildup
p ws
qB
p wf (t 0)
t
24Cs
21.20
Wellbore Storage
Dimensionless wellbore storage coefficient
5.615C
CD
2
2c t hrw
21.21
Type Curve: Pressure
21.22
Type Curve: Press. Derivative
Last Words
21.23
Be consistent!
whatever is characterized as wellbore storage,
radial flow etc on one plot e.g. Cartesian must
be consistent with the presentation of the data
on all other plots e.g. semilog/Horner.