Chapter 5 Well Testing (III)
Weibo Sui
Ph.D, Associate Professor
College of Petroleum Engineering, CUPB
Type Curve
• What are type curves?
– Type curves are graphic plots of theoretical solutions to flo
w equations under specific initial and boundary conditions
of the interpretation model representing a reservoir-well s
ystem.
– Most common type curves are presented in dimensionless
pressure (pD) versus dimensionless time (tD) .
Infinite reservoir, line source well (Transient)
The dimensionless groups are defined as
kh
pD pi p
141.2qB
r
rD
rw
re
reD
rw
0.0002637 kt
tD
ct rw2
0.894C
CD
ct hrw2
Agarwal Type Curve
Agarwal et al. (1970)
Type Curve Matching
• Log-log type curve analysis make use of the dimensionless vari
ables. Since dimensionless pressure and time are linear functi
ons of actual pressure and time, we can calculate kh and φh:
kh kh
pD pi p log p log pD log
141.2qB 141.2qB
0.0002637 kt 0.0002637k
tD log t log t D log
ct rw2 ct rw2
p
kh 141.2qB D
p M
0.0002637 kh 1
hct
rw2 tD
t M
Gringarten Type Curve
Bourdet’s Derivative Plot (1983)
• The derivative plot provides a simultaneous presentation of lo
g ∆p vs. log ∆t and log dp/d(lnt) vs. log ∆t.
– Pressure change log-log plot (Gringarten)
– Pressure change derivative log-log plot
d p d p
p t (Drawdown test)
d ln t d t
Bourdet’s Derivative Plot (1983)
Wellbore Storage Effect (WBS)
During WBS, the pressure change is a linear function of time since
the beginning of the transient:
p qB qB
qB 24C p t log p log t log
t 24C 24C
Which means the unit slope. For the pressure change derivative:
d p
p t
d t
p qB
Substitute into above p, we find that
t 24C
qB
p d t p
24C
Infinite Acting Radial Flow (IARF)
70.6qB kt
p r, t pi ln 2
kh 1688 ct r
d p 70.6qB 141.2qB
p 0.5
d ln t kh kh
which means p is a constant value.
In dimensionless pressure and dimensionless time form, we have
kh
pD pi p
141.2qB
0.0002637k
tD
ct rw2
Therefore, the dimensionless pressure change derivative is
dpD dp
pD t D D 0.5
d ln tD dtD
Combined Gringarten and Bourdet Plot
The Pressure Derivative
To compute the pressure derivative, calculate the pressure change:
p pi pwf (t ) for drawdown data, and
p pws pwf (t p ) for buildup data, where
pi Initial reservoir pressure
pwf Bottom hole flowing pressure
pws Bottom hole shutin pressure
t Elapsed time since start of drawdown or buildup transient
t p Duration of production time prior to shutting in for buildup
Compute tp as the cumulative production prior to the buildup test
divided by the last rate before shut-in.
13
Pressure Derivative
For drawdown analysis, the pressure derivative, in its simplest form, is
computed as the derivative of p with respect to the natural logarithm (ln) of
the elapsed time (ti)
dp / d ln( t ) [ p (t i 1 ) p (t i 1 )] /[ ln (t i 1 ) ln(t i 1 )] for the elapsed time Δt t i t 0
where t 0 Starting time for the transient data
For buildup analysis, the preferred pressure derivative calculation is
d p / d [ p(ti 1 ) p(ti 1 )] / [ i 1 i 1 ] for the elapsed time Δt ti t0
where
superposition time and
i ln[t p ti / (t p ti )]
Flow Regimes
• Wellbore Storage
• Radial Flow
• Spherical
• Linear
• Bilinear
15
Radial Flow Regimes for Vertical Wells
Top of Bottom
zone of zone
Partial Radial Flow Complete Radial Flow
Sealing Boundary
Fracture
Fracture
Boundary
Actual Well Image Well
Pseudoradial Flow to Fracture
Hemiradial Flow to Well
Near Sealing Boundary
Radial Flow Regimes for Horizontal Wells
Radial flow
Pseudoradial flow
Hemiradial flow
17
Spherical Flow Regimes
Spherical Flow to Partially Hemispherical Flow to
Completed Zone Partially Completed Zone
18
Linear Flow Regimes
Fracture
Fracture
Boundary
Fracture Linear Flow Linear Flow to Fracture Bilinear Flow
Linear Flow to Linear Flow to Well
Horizontal Well in Elongated
Reservoir
Flow Regime Identification
e
at
st
r
Radial ea
dy
Li n
ea
Sp
st
he
do
ri c
al
eu
Ps
Radial Radial
ge e ar
a Li n
t or ar
s ne
re Li
b o ear
l Bilin
el
W
FRID Tool
Flow Region Identification
• Wellbore Storage (WBS) - Estimate Cs, wellbore
storage coefficient (bbls/psi)
• Middle time region (MTR) - calculate skin, k & p*
2001/01/02-0000 : OIL
• Late Time region (LTR) - boundaries, kh variations
(pi or p* for depleted reservoir)
DP + DERIVATIVE (PSI/STB/D)
e
at
st
r
Radial nea
y
i
ad
L
te
Sp
10 -2
he
os
ric
ud
a l LTR
e
Ps
MTR
Radial Radial
WBS
10 -3
e ar
ag
10r-2 10 -1 0i
10L
ne
10 1 10 2
o ar
st
Delta-T (hr)
neWell Test Diagnostic Flow Regions
re Li
b o ear
l Bilin
el
W
Time Region Identification
2001/01/02-0000 : OIL
DP + DERIVATIVE (PSI/STB/D)
10 -2
LTR
MTR
WBS
10 -3
10 -2 10 -1 10 0 10 1 10 2
Delta-T (hr)
Well Test Diagnostic Flow Regions
22
Early and Middle Time Analysis
Drawdown Analysis
• C, k, s
– cartesian plot (C)
– semilog analysis (k, s)
– log-log derivative (C, k, s)
24
Drawdown Analysis
102/01/01-1200 : N/A
4000. 4500. 5000. 5500. 6000.
pressure PSI
0. 10. 20. 30.
rates STB/D
0. 500. 1500.
0. 10. 20. 30.
Time (hours)
Cartesian Drawdown Analysis – C
2002/01/01-1200 : OIL
20.
15.
Dp mC
DP (PSI)
Zoom
10.
Dt
5.
0.
0. 5. 10. 15. 20.
Delta-T (hr)
Cartesian Plot
qB qB qB
p t mC C in bbl/psi
24C 24C 24mC
Semi-log Drawdown Analysis – s
2002/01/01-0000 : OIL
4995.
4990.
P PSI
4985.
Semilog Slope, m
4980.
SLOPE
4975.
0.0100 HR 0.10 HR 1.0 HR 10. HR
-3000. -2000. -1000. 0. 1000.
Superposition(T)
Semilog Analysis
162.6qB
k
mh
pi pwf k
s 1.151 1hr
log 3.23
m ct rw
2
2002/01/01-0000 : OIL
Log-Log Drawdown Analysis – C
ENDWBS
DP + DERIVATIVE (PSI/STB/D)
10 -2
pressure derivative, m'
STABIL
10 -3
UNIT SLP
10 -3 10 -2 10 -1 10 0 10 1
Delta-T (hr)
Log-log Derivative Analysis
d (p ) p 2 p1
d (t ) t 2 t1
(Dt2, Dp2)
qB
log Dp C
d (p )
(Dt1, Dp1) 24
d (t )
2002/01/01-0000 : OIL
Log-Log Drawdown Analysis – k, s
ENDWBS
DP + DERIVATIVE (PSI/STB/D)
10 -2
pressure derivative, m'
STABIL
10 -3
UNIT SLP
10 -3 10 -2 10 -1 10 0 10 1
Delta-T (hr)
Log-log Derivative Analysis
70.6qB
k , m is the level of the derivative, m=m/2.303
mh
pi pwf kt
s 1.151 log 3.23 for t in IARF
m ct rw
2
PBU Analysis
(straight line methods)
• C, k, s, p*
– cartesian plot (C)
– semilog analysis (k, s, p*)
– - log-log derivative (C, k, s, p*)
PBU Analysis
2002/01/01-1200 : OIL
4995.
pressure PSI
4985.
4975.
0. 10. 20. 30.
rates STB/D
800.
400.
0.
0. 10. 20. 30.
Time (hours)
Data Plot
Cartesian PBU Analysis – C
2002/01/01-1200 : OIL
20.
15.
Dp mC
DP (PSI)
Zoom
10.
Zoom Dt
5.
0.
0. 5. 10. 15. 20.
Delta-T (hr)
Cartesian Plot
qB
C in bbl/psi
24mC
Horner PBU Analysis – k, p*
p*
Horner slope, m
4995.
4990.
P PSI
4985.
4980.
10 0 10 1 102 103
(Tp + dT)/dT
Semilog (Horner) Analysis
162.6qB t p t
pws pi log
kh t
162.6qB
k
mh 33
Horner PBU Analysis – s
p*
Horner slope, m
4995.
4990.
P PSI
4985.
4980.
10 0 10 1 102 103
(Tp + dT)/dT
Semilog (Horner) Analysis
pwst 1hr pwf t 0 k
s 1.151 log 3.23
m ct rw
2
34
2002/01/01-1200 : OIL
Log-Log PBU Analysis – C, k, s
ENDWBS
DP + DERIVATIVE (PSI/STB/D)
10 -2
UNIT SLP STABIL
10 -3
10 -3 10 -2 10 -1 10 0 10 1
Delta-T (hr)
Log-log Derivative Analysis
35
2002/01/01-0000 : OIL
Log-Log PBU Analysis – C
DP + DERIVATIVE (PSI/STB/D) ENDWBS
10 -2
pressure derivative, m'
STABIL
10 -3
UNIT SLP
10 -3 10 -2 10 -1 10 0 10 1
Delta-T (hr)
Log-log Derivative Analysis
d (p ) p 2 p1
(Dt2, Dp2) d (t ) t 2 t 1
log Dp qB
C
d ( p )
(Dt1, Dp1) 24
d ( t )
36
2002/01/01-0000 : OIL
Log-Log PBU Analysis – k, s
ENDWBS
DP + DERIVATIVE (PSI/STB/D)
10 -2
pressure derivative, m'
STABIL
10 -3
UNIT SLP
10 -3 10 -2 10 -1 10 0 10 1
Delta-T (hr)
Log-log Derivative Analysis
70.6qB
k , m is the level of the derivative
mh
p k t p t
s 1.151 ws
log 3.23 log log t
m cr 2 t
w p
for pws , t in IARF, where pws pws pwf ( t p )
37
p, p’, psi Finite Conductivity Hydraulic Fracture
Linear Flow (1/2 slope)
Bilinear Flow (1/4 slope)
Time, hrs
Finite Conductivity Hydraulic Fracture
Finite Conductivity Hydraulic Fracture
e
at
st
r
Radial nea
dy
Li
ea
Sp
st
he
do
r ic
al
eu
Dp, Dp’, psi
Ps
Radial Radial
e r
ag i nea
r L
s to ea
r
e Li n Linear Flow (1/2 slope)
bor ear
l Bilin
el
W
Bilinear Flow (1/4 slope)
Time, hrs
Finite Conductivity Hydraulic Fracture
Late Time Analysis
Late-Time Analysis: Outer Boundary
Dp, Dp’, psi
Complete Radial Flow
Time, hrs
Drawdown
Closed Boundary – PSS Flow
In a closed system, after the pressure wave has reached the outer boundary
p
const
t
Based on the material balance and compressibility definition,
1 dV p 1 0.234qBdt
ct
V p dpwf Vp dpwf
where V p is the pore volume of the reservoir, Vp Ah .
ct is in psi 1 , V p is in ft 3 , q is in STB/d, B is in bbl/STB, t is in hr, pwf is in psi.
We can do the following things according to this relationship:
(1) Calculate V p by the slope of the straight line pwf vs. t
dpwf 0.234qB
slope m Vp
dt ctV p
(2) The initial oil in place (N p , STB) is given by
1
N V p So
5.615
Closed Boundary – PSS Flow
(3) By integration, the above equation can be written as
pwf 0.234qB t
pi wfdp
V p ct 0
dt
0.234qB
pwf t
V p ct
d pwf 0.234qB
pwf t
d ln t V p ct
vs. log t is a straight line with unit slope (m=1).
Thus we know log pwf
Dp, Dp’, psi
Complete Radial Flow
Time, hrs
Drawdown
Rectangular Drainage Area
Boundary Models – Single Sealing Fault
Characteristic flow regimes:
• Radial flow
• Hemi-radial flow
Boundary Models – Single Sealing Fault
The pressure drop can be simulated using the image well method,
pi pwf p1 p2
162.6qB kt 70.6qB 948 ct 2 L 2
lg 2
Ei
kh 1688 ct rw kh kt
For the short beginning of the drawdown test, the pressure has not reached
the fault, the pressure disturbance caused by the image well cannot be felt,
948 ct 2 L 2
(t is very small, 2L rw , thus Ei 0)
kt
162.6qB kt
pwf pi lg
2
kh 1688 c r
t w
t
pi m lg
1688rw2
m lg t pi m lg
1688rw2
Boundary Models – Single Sealing Fault
948 ct 2 L 2
After the fault has been felt by the well, i.e. Ei 0，
kt
the p2 cannot be omitted any longer.
948 ct 2 L
2
If 0.01, the pressure drop equation can be approximated as
kt
t t
pwf pi m lg m lg
1688 2 L
2
1688rw2
2
t
pi m lg
1688 rw 2 L
t
pi 2m lg
1688rw 2 L
2m lg t pi 2m lg
1688 rw 2 L
Compared with the early-time radial flow equation, we can find that the slope
of the straight line in the semi-log plot doubled when the fault is felt.
Semi-log Plot
Boundary Models – Single Sealing Fault
Consider the cross point (t = t x ) of the radial flow and hemi-radial
flow straight line, we can solve the distance to the sealing fault,
tx tx
pi m lg pi 2m lg
1688rw2
1688rw 2 L
Solved for L,
0.000148kt x
L
ct
At the same time, if we draw the log-log derivative plot, we find that
the derivative level doubled when the fault is felt.
Boundary Models – Single Sealing Fault
Dimensionless Pressure Derivative Plot
Dimensionless Pressure Derivative Plot
Intersecting Fault
θ = 60⁰
A 60 deg. angle between the intersecting faults is equivalent to the
active well plus 5 image wells for a total of 6 effective wells. The
derivative level will rise to 6 times the level of the initial radial flow
response.
Intersecting Fault
q = 30o
Time Region Analysis
• Early-time analysis
– wellbore storage
– skin factor
• Middle-time analysis
– reservoir model (IARF, hydraulic fractured, natural fractured r
eservoir)
– reservoir properties (permeability etc.)
• Late-time analysis
– outer boundary
Well Testing Analysis Procedures
• Data plots
– prepare log-log plots of pressure change and pressure change deri
vative vs. elapsed time during the test.
– prepare special plots of the data (semi-log plot etc.)
• Qualitative type-curve analysis
– identify the appropriate reservoir model
– identify any characteristic flow regime that can be analyzed
with special analysis techniques
• Semi-log or specialized analysis
– estimate formation properties
• Quantitative type-curve analysis
– confirm or complement specialized analysis results
Cutting-Edge Well Testing Technique
• Reservoir and well model
– Various boundaries
• Constant pressure, closed system, faults, composite reservoir, leak
y fault, incomplete boundary
– Naturally fractured reservoir, multilayer reservoir
– Hydraulic fractured well, partially penetration well, horizon
tal well
– Multiphase flow
Cutting-Edge Well Testing Technique
• Pressure-Transient-Analysis Software
– Initialization
– Test design
– Loading/editing data
– Diagnostic tools
– Modeling capability
• Model selection, parameter estimation, numerical model
– Optimization
– Report generation
Permanent Downhole Gauge
In-Class Exercise
Please use log-log pressure change derivative plot and semi-log p
lot to solve problem 1 and 2. Beside determining the parameters
required in the problem, please determine the early-time, middl
e-time and late-time flow regime.
1. PTT Ch2, Problem 16
2. PTT Ch2, Problem 37
In-Class Exercise
3. The following table gives measured data for a buildup test for a finite-ac
ting well. Before shut-in for buildup, the well pressure was declining line
arly at 0.431 psi/hr. Use this information to determine the following para
meters.
(1) Reservoir pore volume, Vp
(2) Permeability-thickness product, kh
Reservoir Parameters
qBo, RB/D 333.3
ct, psi-1 8×10-6
μ, cp 2
m*, psi/hr -0.431
t p t p
(hours) (psia) (hours) (psia)
1 3138.65 15 3165.94
2 3146.75 16 3166.30
3 3151.31 17 3166.62
4 3154.42 18 3166.89
5 3156.73 19 3167.13
6 3158.54 20 3167.33
7 3160.01 21 3167.50
8 3161.22 22 3167.64
9 3162.24 23 3167.75
10 3163.11 24 3167.85
11 3163.85 25 3167.92
12 3164.49 26 3167.97
13 3165.04 27 3168.00
14 3165.52 28 3168.02
In-Class Exercise
4. A well is opened to flow at 150 STB/day for 24 hours. The flow rate is the
n increased to 360 STB/day and lasted for another 24 hours. The well flo
w rate is then reduced to 310 STB/day for 16 hours. Calculate the pressu
re drop in a shut-in well 700 ft away from the well given:
=15% h 20ft k 100md o 2cp
Bo 1.2bbl/stb rw 0.25ft pi 3000psi ct 12 106 psi 1
