PRACTICAL RESERVOIR MONITORING

Well Test Interpretation Methodology

Semi-Log Analysis

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Drawdown Semi-Log Analysis: the MDH Plot (Miller-Dyes-Hutchinson)

In drawdown analysis, the log approximation to the Exponential Integral gives:

162.6q   k 3.22750.86859S 
pDd 
kh 
log t  log Ctrw²  
 

which can be written as: pDd mlog t  b

On the MDH plot, one can solve for m and b by reading the coordinates of two
points:
t = 0, pDd = pi, and
t = 1 hr, pDd = p1hr.

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Drawdown Pressure Profile: the MDH Plot

Because the pressure change is proportional to the logarithm of elapsed time when
IARF is reached, a graph of P vs Log t will yield a straight line of slope m.
The effects of wellbore storage and skin are superimposed onto the ‘ideal response’
as shown below.

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Drawdown Semi-Log Analysis (cont’d)

162.6q
The solution is then: kh
m
 pi p1hr 
and S 1.1513 log k 3.2275
 m  Ctrw²  

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Build-up Analysis

In practice, it is not often possible to conduct drawdown analysis. This is because

drawdown analysis applies to a constant flow rate, a condition which is difficult to
maintain during well tests.
To remedy this shortcoming, it is more practical to analyze build-up periods by
resorting to the the principle of superposition of states.
Modern well testing now offers multiple possibilities to analyze drawdown (‘flow’)
periods by measuring the flow rates downhole during testing. For the interpretation,
the principle of superposition is generalized into a technique called the pressure-flow
convolution.

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
The Principle of Superposition of States

Because of the linearity of the pressure response equation, the response during a buid-
up period is equal to the sum of the responses of two drawdown periods:
- Flow rate ‘q’ from time t = 0, and
- Flow rate ‘-q’ from time t = tp (drawdown production time).

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Single Flow Period Superposition for Build-up Analysis

Considering a single flow period of duration tp:

pBu pi pwf pDd t  pDd tpt 

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Build-up Semi-Log Analysis: the Horner Plot

For a single flow period, the superposition

function is the Horner time:

tp  t
t
On a semi-log plot, the extrapolated pressure
is the static reservoir pressure, provided that
- The reservoir has not entered
depletion regime during the drawdown.
- No late-time effects will affect the
buildup after the end of the buildup
(this is impossible to ascertain without
testing longer).
 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Build-up Semi-Log Analysis: the Horner Plot (cont’d)

162.6q
On the Horner plot, the solution is again: kh
m
p1hr  pwf
and S  1.1513  log k   3.2275 
 m  Ctrw²  

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Generalized Superposition for Build-up Analysis

When the well has been submitted to a series of flow periods prior to build-up, one
must consider a ‘generalized superposition function’ as follows:

Sn t     qi  qi 
i  N t  1  ln t  ti  
  
i 1  qN t 

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Multi-Rate Build-up Analysis

When the pressures are plotted versus

Sn(t), the solution is identical to the
case of a single flow period (Horner
plot).

On a semi-log plot, the extrapolated

pressure is the static reservoir pressure,
with the same restrictions as apply to
the Horner plot.

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology

Pressure Derivative
Log-Log Analysis

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
The Pressure Derivative

Modern well testing advances (1983) have culminated with the introduction of the
Pressure Derivative PD’ as an indispensable complement to plotting pressures versus
time. By definition:
dp dp
p'   t
dLn t  dt
The Pressure Derivative is the slope of the semi-log plot as shown below.

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Dimensionless Variables

In order to solve the diffusivity equation in typical situations applicable to all possible
values of the physical parameters, one uses dimensionless variables defined as
follows:
• Dimensionless distance: rD r in which rw is the wellbore radius.
rw
• Dimensionless pressure: pD2 kh (pi p) in which p i is the initial
pressure. q

• Dimensionless time: tD  k Δt in which t is the elapsed time.

φμCtrw²

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Homogeneous Reservoir with Wellbore Storage and Skin

Because the skin just adds to the pressure drop in the wellbore, the dimensionless
skin S just adds to the PD function in the solution of the diffusivity equation for
IARF:
pD 1 Ln tD 0.809072S 
2

In physical terms:
162.6q   k 3.22750.86859S 
p t   pi log(t )log Ctrw² 
kh    

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Homogeneous Reservoir with Wellbore Storage and Skin (cont’d)

The IARF solution for a well with wellbore storage and skin has been expressed as:

pD 1 Ln tD 0.809072S 

2
In log-log analysis, it is preferrable to re-write the pressure response as:

  
pD 1 Ln tD 0.80907LnCDe2S
2 CD

in which CD is the ‘dimensionless wellbore storage constant’:

C D C
2Ctrw²h
 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Type Curves

By plotting the theoretical pressure

response PD versus tD/CD, (instead of
vs tD), one obtains a way of
characterising in a unique way the
IARF solution (for a well with
wellbore storage and skin for
example).
One thus defines an array of ‘type
curves’, each curve corresponding to a
value of the sensitivity parameter
CDe**2S.
The inclusion of the pressure
derivative on this plot was a major
breakthrough in well test
interpretation.  Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Attributes of the Log-Log Plot: Early Time Behaviour

At early times, the pressure response is dominated by the wellbore effect. The
solution of the diffusivity equation is:

pD tD
CD
This plots as a ‘unit slope’ on a graph of pD vs tD/CD.
Then
dpD dpD tD
pD'
dLn t 
CD
D
tD
dt
  pD
D CD

and the derivative matches the pressure response on a unit slope.

This particularity of early time behaviour is one of the most conspicuous features of a
log-log plot in well test interpretation.
 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Attributes of the Log-Log Plot: IARF

The solution of the diffusivity equation for IARF is:

pD 1 Ln tD 0.809072S 

2
dpD 1
pD' 
Then
dLn tD 2
When IARF is reached, the pressure derivative levels off to a plateau on the log-log
plot. The corresponding value of PD is 0.5. Again, this characteristic leveling off of
PD’ upon reaching IARF is one of the most conspicuous features of the log-log plot in
well test interpretation.

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology

Type Curve Matching

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Data Set and Type Curve Array

The data collected during a well test are in the form of couples (pressure-time). These
are initially presented as a log-log plot of pressure variations vs elapsed time, with the
computation of the pressure derivative.
Type-curve matching has for objective the superposition of the data set over the array
of type curves corresponding to the model chosen, and the extraction of the test target
parameters.
This will be done by
- shifting the data horizontally (time match).
- shifting the data vertically (pressure match).
- finding the matching type curve (and its derivative) with its characteristic CDe**2S.

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Data Set and Array of Type-Curves

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Matched Data Set

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Pressure Match: Extracting kh

From the expression of dimensionless pressure

pD kh p
141.2q
one defines the pressure match Mp

pD
Mp  kh
p 141.2q
Mp is read as the value of pD matching a specific value of p. Then

kh141.2qMp

 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Time Match: Extracting C

From the expressions of dimensionless time and ‘wellbore storage constant’:

tD  0.000295kh t
CD C
one defines the time match Mt

 t  D

M  C  0.000295kh
t
D
t C
Mt is read as the value of tD/CD matching a specific value of t. Then

C  0.000295kh
Mt
 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Skin Match: Extracting S

One reads the value of Ms on the matching type curve:

MS CDe2S
Then

 
S  1 Ln MS
2 CD
with CD calculated from its dimensionless expression:

CD 0.8936C
Ctrw²h
 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Type-Curve Match Example: Data Set
TCMATCH.WTD (Field Data)

10000

1000
Pressure change, psi

100

10


0.001 0.01 0.1 1 10 100 1000
Gamma Experts
Petroleum Engineering
Equivalent time, hrs
September 2002 Yves Chauvel
 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Type-Curve Match Example: Unmatched Overlay
TCMATCH.WTD (Drawdown type curve, Radial equivalent time)
Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100

10
1000
Dimensionless pressure

1
Pressure change, psi

100

0.1
10

0.01
1
0.001 0.01 0.1 1 10 100 1000
Equivalent time, hr

0.001


0.001 0.01 0.1 1 10 100 1000 10000 100000 Gamma Experts
Petroleum Engineering
Dimensionless time
September 2002 Yves Chauvel
 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Type-Curve Match Example: Matched in Pressures
TCMATCH.WTD (Drawdown type curve, Radial equivalent time)
Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100

1000

10
Pressure change, psi

100
Dimensionless pressure

10

0.1

1
0.001 0.01 0.1 1 10 100 1000
Equivalent time, hr
0.01

0.001


0.001 0.01 0.1 1 10 100 1000 10000 100000
Gamma Experts
Petroleum Engineering
Dimensionless time
September 2002 Yves Chauvel
 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Type-Curve Match Example: Matched in Both Times and Pressures
TCMATCH.WTD (Drawdown type curve, Radial equivalent time)
Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100

1000

10
Pressure change, psi

100
Dimensionless pressure

10

0.1

1
0.001 0.01 0.1 1 10 100 1000
Equivalent time, hr
0.01

0.001
0.001 0.01 0.1 1 10

Dimensionless time
100 1000 10000 100000
 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel

 PRACTICAL RESERVOIR MONITORING
Well Test Interpretation Methodology
Type-Curve Match Example: Extraction of Time, Pressure and Skin Match
TCMATCH.WTD (Drawdown type curve, Radial equivalent time)
Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100
CDe2s=7x109
pD=10 p=262 1000
psi

10
Pressure change, psi

100
Dimensionless pressure

10

0.1

1
0.001 0.01 0.1 1 10 100 1000
Equivalent time, hr
0.01 teq=0.0546 hr

0.001
0.001 0.01 0.1

tD/CD=1
1 10

Dimensionless time
100 1000 10000 100000
 Gamma Experts
Petroleum Engineering

September 2002 Yves Chauvel