minimun pore volume in well test
DESCRIPTION
Minimun Pore Volume in Well TestTRANSCRIPT
1
SPE Distinguished Lecturer Program
Primary funding is provided by
The SPE Foundation through member donations
and a contribution from Offshore Europe
The Society is grateful to those companies that allow their
professionals to serve as lecturers
Additional support provided by AIME
Society of Petroleum Engineers
Distinguished Lecturer Programwww.spe.org/dl
The Determination of Minimum Tested Volume and
Future Well Production from the Deconvolution of
Well Test Pressure Transients
Tim WhittleBg Group
Society of Petroleum Engineers
Distinguished Lecturer Programwww.spe.org/dl
3
Well Test Objectives
• Fluid Characterisation (PVT)
• Well Performance (Flow)
• Reservoir Description (Model)
• Reservoir Deliverability (Flow)
• Flow Assurance (Facilities)
• Clean Up (Production)
4
Types of Well Test
Fluid
Characterisation
Well
Flow
Reservoir
Description
Reservoir
Flow
Exploration
Appraisal
Extended
(EWT)
Production
Primary objectives depend on the type of test
5
Wireline Formation Tests
Objective WT WFT
Fluid
Small Volume
Large Volume
Well Flow
Reservoir Flow
Reservoir Description
Versus Depth
Formation
Boundaries
WFT and WT are not equivalent
6Pressure, p and Flowrate, q
pwf
qtest
Flow rate, q
Pre
ssure
, p
00
Operating Point
wfr
test
pp
qPI
Productivity Index
AOFP
PI
pr
SurfaceFlow, qtest
Reservoir
Well
Bottom hole
prpwf
Well Performance
7
How to Improve Well Performance?
Flow rate, q
Pre
ssure
, p
00
pr
Outflow 1 to Outflow 2 Change in completion (tubing, choke, artificial lift…)
Inflow 1 to Inflow 2 Change in well/reservoir (perfs, acid, frac, well type…)
Need to understand Inflow to see if improvement is possible…
∆q ∆q ∆q
8
Reservoir Deliverability
t1
Flow rate, q
Pre
ssure
, p
00
pr
t2
t3
t3> t2 > t1
• Reservoir constrained
– Complex boundaries
(e.g. channel sands)
– Low permeability
2000
4000
Pre
ssure
[psi
a]
TH
P [
ps
ia]
0
20
40
Gas
Rate
Pro
du
cti
on
[M
Ms
cf/
D]
07-Nov-2008 09-Nov-2008 11-Nov-2008 13-Nov-2008 15-Nov-2008
Pressure [psia], Gas Rate [MMscf/D] vs Time [ToD]
9
-1000
0
1000
2000
3000
4000
5000
6000
22-Apr 23-Apr 24-Apr 25-Apr
Pre
ssu
re (
psia
)
Elapsed time (Date)
0
1000
2000
3000
4000
5000
6000
7000
8000
Oil
Ra
te (
ST
B/D
)
Pressure History
t1
Flow rate, q
Pre
ssu
re, p
00
pi
t2
t3
t3> t2 > t1
• Depletion
• Hopefully not seen in a
well test!
Reservoir Deliverability
pi
Depletion
10
Pressure Transient Analysis
Flow, q, Pressure, p, and Time, t
Pre
ss
ure
, p
Time, t
q(t)
Ra
te
0
pi
Δp
Δt
Δq
p(t)Steady state
Pseudo-steady state
Transient
11
Log-log Diagnostic Plot
Early Time Middle Time Late TimeNear WellStorage
Skin
Fractures
Partial Completion
ReservoirHomogeneous
2-Porosity
Multi-layer
BoundariesNo-Flow
Constant Pressure
Permeability thickness,
kh, and skin, S
Stabilisation
Infinite Acting
Radial Flow
½ Slope
Linear Flow
Channel
Unit Slope
Well Storage
Unit Slope
Depletion
Assuming single constant rate drawdown...
Δp’ = dp/d(ln t) = t dp/dt
Deri
vati
ve, Δ
p’
Elapsed Time, Δt (hrs)
Pre
ssu
re C
han
ge, Δ
p (
psi)
0.01 0.1 1 10010 10000.1
1
10
100
12
Pressure Transient Derivative Response
0.01 0.1 1 10 100 seconds1 10 100 hours
1 10 100 days
Spherical
Radial
WellboreStorage
Horizontal/Fractured Well
ReservoirBoundaries
WFTPT
Time (k = 750 mD)
13
Scale
While
Drilling
WirelineWell Test
Pressure Test Sampling
Volumes 1-10 cc 5-50 cc 10-100 l 1-10000 m3
x-factor 1 5 10000 106-109
Times 1-5 min 1-15 min 1-5 hr 12hrs – 12days
x-factor 1 1-3 60 720-20000
(Mini-frac)
14
Scale
Radius of investigation:
k/µ = 10 mD/cp
Øct = 0.15x10-5 1/psi
h = 75 ft
While
Drilling
Wireline
Well TestPressure Test
Sampling/
Mini DST
Flow Time 5 s 10 s 15 min 12 hr
Flow Volume 5 cc 10 cc 3000 cc 40x106 cc (250 bbl)
Shut Time 30 s 3 min 5 min 24 hr
Δp/Δt (psi/min) 0.18 0.003 0.06 0.018
Theoretical ri (ft) 5 17 23 300
Practical ri* (ft) 2 4 15 250
ct
tkri
* Assuming a gauge resolution/noise of 0.03 psi
(Mini-frac)
15
Example – Low Permeability – Two Wells
kh=16 mDft
kh=2.5 mDft
kh=6 mDft
Derivative describes heterogeneity in time/space
16
Data Acquisition: Well Test Sequence of Events
Pre
ss
ure
Time0
Ra
te
Pre
ss
ure
TimeR
ate
0
Δq
Ideal Case Actual Case
Pre
ss
ure
Elapsed Time
Deri
vati
ve
, Δ
p’
Log-log Plot
Entire Test
In general, only shut-ins give sufficiently high quality pressure transients
Pre
ss
ure
Elapsed Time
Deri
vati
ve
, Δ
p’
Log-log Plot
Only Build-up
?
18
Deconvolution by Iteration using superposition
Pre
ss
ure
Time
Ra
te
0
Δq
Ra
te
Time0
Pre
ss
ure
Time0
Ra
te
+
Non-linear Least Squares Minimisation
tmax
tmax
tmax
Iterations
2
19
265
hrs
4750
4950
5150
0 40 80 120 160 200 240Time [hr]
0
10
20
Ga
s R
ate
, M
Ms
cf/
dP
res
su
re, p
sia
48 hrs 48 hrs
Example
20
1E-3 0.01 0.1 1 10 100
Time (hrs)
1E+8
1E+9
1E+10
Ga
s P
ote
nti
al a
nd
De
riv
ati
ve
(p
si2
/cp
)
1000
1E+11
48 hrs 265 hrs
Example - DST
Deconvolved Data
Build-up Data
Longer duration of deconvolved data larger radius of investigation?
21
Pressure Transient Analysis Workflow
Pressures
P v tDeconvolve
Model Select
Simulate
Model
Catalogue
Model
Parameters
Fit
Rates
q v t
OK?N
Another
Model?Done
YN
Y
Diagnose
With Deconvolution
Minimum
Tested
Volume
SPE 116575
22SPE 116575
Minimum Tested Pore Volume
0.001 0.01 0.1 1 10 100
Time (hrs)
1
10
100
Pre
ss
ure
ch
an
ge
an
d D
eri
va
tiv
e (
ps
i)
1000
1000
Deconvolved Data
Build-up Data
Unit
Slope
(pss)
maxp
maxt
max
max)1(
p
tq
c
SSTOIP
t
wtested
'
max
max)1(
pnm
tq
c
SGIIP
t
wtested
23
Same Principle as Reservoir Limits Test (MBH)
Flow, q, Pressure, p, and Time, t
Pre
ss
ure
, p
Time, t
q(t)
Rate
0
pi
p(t)
Pseudo-steady state
Transient
Minimum
End of Test?
25
Example 1 - Gas
Input:Sw = 0.15
ct = 6.62E-5 1/psi
q = 40.4 MMscf/d
pbar = 8135.32 psia
μbar = 0.032 cp
zbar = 1.247
z
p
pm
tq
c
SGIIP
t
wtested
2)1('
max
max
Δtmax= 93.6 hrs
Δm(p)’max = 2.30E8 psi**2/cp
247.1032.0
32.81352
830.2
24/6.934.40
5625.6
)15.01(
EEGIIPtested
bscfMMscf 53.33534
26
Example 2 - Oil
Input:Sw = 0.129
ct = 9.44E-6 1/psi
q = 2380 stb/d
Δtmax= 304 hrs
Δm(p)’max = 20.6 – 63.1 psi
6.20
24/3042380
644.9
)129.01(
ESTOIPtested
MMstbstb 135843,39,139
max
max)1(
p
tq
c
SSTOIP
t
wtested
1.63
24/3042380
644.9
)129.01(
ESTOIPtested
MMstbstb 1.44261,053,44
Max Min
Uncertainty in deconvolution uncertainty in connected volume
27
Example 3 - Gas
z
p
pm
tq
c
SGIIP
t
wtested
2)1('
max
max
Input:Sw = 0.1
ct = 0.00131 1/psi
q = 10.7 MMscf/d
pbar = 865.2 psia
μbar = 0.0128 cp
zbar = 0.873
Δtmax= 136 hrs
Δm(p)’max = 2.37E5 psi**2/cp
873.00128.0
2.8652
537.2
24/1367.10
331.1
)1.01(
EEGIIPtested
bscfMMscf 2.27200,27
29
Example 4 - Oil
Input:Sw = 0.15
ct = 1.5E-5 1/psi
q = 1220 stb/d
Δtmax= 136 hrs
Δm(p)’max = 93.5 – 530 psi
5.93
24/941220
55.1
)15.01(
ESTOIPtested
MMstbstb 89.2000,894,2
max
max)1(
p
tq
c
SSTOIP
t
wtested
530
24/941220
55.1
)15.01(
ESTOIPtested
MMstbstb 51.0564,510
Max Min
Uncertainty in deconvolution uncertainty in connected volume
30
Example 5 – Oil Design
Input:Sw = 0.15
ct = 1E-5 1/psi
q = 5000 stb/d
Δtmax= 60 hrs
Δm(p)’max = 105 psi
105
24/605000
63
)15.01(
ESTOIPtested
MMstbstb 7.3333708571
max
max)1(
p
tq
c
SSTOIP
t
wtested
Design Input:k = 90 mD
h = 7 m
ø = 0.11
rw = 0.3 ft
μ = 0.5 cp
pi = 5300 psia
(rinv = 5250 ft)
No Boundaries
31
Example 5 – Oil Design
Input:Sw = 0.15
ct = 1E-5 1/psi
q = 5000 stb/d
Δtmax= 60 hrs
Δm(p)’max = 683 psi
683
24/605000
63
)15.01(
ESTOIPtested
MMstbstb 2.55182138
max
max)1(
p
tq
c
SSTOIP
t
wtested
Design Input:k = 90 mD
h = 7 m
ø = 0.11
rw = 0.3 ft
μ = 0.5 cp
pi = 5300 psia
d1 = 500 ft
d2 = 1000 ft
Channel Boundaries : Significantly reduces tested volumes
32
Elapsed Time, Δt (hrs)
Pre
ssu
re C
han
ge, Δ
p (
psi)
0.01 0.1 1 10010 10000.1
1
10
100
Coefficient of Reservoir Complexity (CRC)
Stabilisation Infinite Acting Radial
Flow
Unit Slope
PSS
Applies to deconvolved data
De
riv
ati
ve,
Δp’
maxp
intp
int
max
p
pCRC
CRC is similar to Dietz Shape Factor, CA
(inversely proportional?)
34
Comparison of CRC with Dietz Shape Factor, CA
(Tom Street – May 2009)
Coefficient of Reservoir Complexity (ref. SPE 116575) vs. Dietz Shape
Factor
y = -0.3612x + 11.852
R2 = 0.6994
y = -2.274Ln(x) + 11.743
R2 = 0.8496
0
2
4
6
8
10
12
14
16
18
0 5 10 15 20 25 30 35
Dietz Shape Factor
CR
C
35
Pressure Transient Analysis Workflow
Pressures
P v tDeconvolve
Model Select
Simulate
Model
Catalogue
Model
Parameters
Fit
Rates
q v t
OK?N
Another
Model?Done
YN
Y
Diagnose
Production
Forecast
SPE 122299
SPE 122299
36
Extrapolation methods for Production Forecast
10
100
1000
10000
0.001 0.01 0.1 1 10 100 1000 10000
Elapsed time, dt (hrs)
Pre
ss
ure
Ch
an
ge
an
d d
eri
va
tiv
e (
ps
i)
Unit slope
Worst
case
-1 unit slope
Best case
Most likely
Extrapolate with Different Cases
Knowing
STOIIP/GIIP
37
Example 6: Gas – Prediction from DST
-1000
0
1000
2000
3000
4000
5000
6000
01-Jan 02-Jan 03-Jan 04-Jan 05-Jan 06-Jan
Pre
ssu
re (
psia
)
Elapsed time (Date)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Ga
s R
ate
(M
Mscf/D
)
38
1
10
100
1000
10000
100000
0.0000001 0.00001 0.001 0.1 10 1000
nm
(p)
Cha
ng
e a
nd
Deri
va
tive
(p
si)
Elapsed time (yrs)
Log-Log Deconvolution - Flow Period 15
Example 6: Gas – Prediction from DST
GIIP = 150 bcf
Unit Slope
PSS
Unit Slope
WBS
Deconvolved pressure derivative extrapolation defines dynamic response
Extrapolation
2
21 years
1 1 year
?
Deconvolution
39
Production Forecast
(pwf = 1500 psi)
0
2
4
6
8
10
12
14
16
18
0 5 10 15
Time (years)
Cu
mu
lati
ve
Ga
s (
bc
f)
Ra
te (
MM
sc
f/d
)
Case 1 Rate "
Case 1 Cum
Case 2 Cum
Case 2 Rate50%
Example 6: Gas – Prediction from DST
40
Example 7: Sensitivity to Initial Pressure
0
1000
2000
3000
4000
5000
6000
0 10 20 30 40 50
Time (days)
Pre
ssu
re (
psia
)
1.00E+06
1.00E+07
1.00E+08
1.00E+09
1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02 1.00E+03
Elapsed Time
Pseu
do
-pre
ssu
re C
han
ge a
nd
Deri
vati
ve
pi = 5495
pi = 5490
pi = 54860
20
40
60
80
100
120
140
5484 5486 5488 5490 5492 5494 5496 5498 5500 5502
Initial Reservoir Pressure (psia)
Ga
s V
olu
me
(b
sc
f)
Min Tested Volume
Recovery after ten years
(pwf = 1500 psia)
41
2850
2860
2870
2880
2890
2900
2910
2920
2930
0 10 20 30 40 50 60 70 80 90 100 110 120 130
Pre
ssu
re (
psia
)
Elapsed time (hrs)
0
50
100
150
200
250
Ga
s R
ate
(M
Mscf/
D)
Pressure History
Example 7: Gas – Prediction from Initial Production Test
42
0.01 0.1 1 10 100
Elapsed Time (hrs)
1E+5
1E+6
1E+7
Ps
eu
do
-pre
ss
ure
Ch
an
ge
an
d D
eri
va
tiv
e (
ps
i2/c
p)
Observed Extrapolated
1E+8
1000
Worst
Most
Likely
Best
Example 7: Gas – Prediction from Initial Production Test
43
0
2
4
6
8
10
12
14
0 0.2 0.4 0.6 0.8 1Time (years)
Cu
mu
lati
ve p
rod
ucti
on
(b
cf)
Best
Most likely
Worst
Actual
Example 7: Gas – Prediction from Initial Production Test
44
-3000
-2000
-1000
0
1000
2000
3000
4000
5000
0 1 2 3 4 5 6 7 8 9
Pre
ssu
re (
psia
)
Elapsed time (yrs)
0
50
100
150
200
250
Me
asu
red
Ga
s R
ate
(M
Mscf/
D)
Prediction at day 142Prediction at day 379
Prediction at day 507
Example 8: Gas – Prediction from Permanent Gauge Data
45
0.1 1 10 100 1000 10000
Elapsed Time (hrs)
1E+6
1E+7
1E+8
Pse
ud
o-p
ressure
ch
an
ge
an
d D
eriva
tive
(p
si2
/cp
)
Day 142Day 379Day 507
1E+9
100000
Constrained
46
0
50
100
150
200
250
0.00 2.00 4.00 6.00 8.00 10.00
Producing Time (years)
Cu
mu
lati
ve G
as (
bcf)
Measured
Predicted at day 142
Predicted day 379
Prediction at day 507
Predicted at day 142
Constrained to GIIP
47
Limitations
• Deconvolution assumes single phase flow in the reservoir and therefore cannot be used to predict e.g. water breakthrough.
• Deconvolution currently only works for single wells; i.e. it does not take into account the influence of nearby producers and injectors.
(These limitations do not prevent the use of deconvolution but need to be considered when examining results).
48
Conclusions
• With the availability of robust deconvolution, it is possible to
extract important information from well test data quickly and
easily prior to any further analysis or models.
• Uncertainty in the deconvolution carries through to uncertainty
in results.
• The deconvolved derivative provides the signature of the
dynamic behaviour of a well which can be extrapolated to
predict future well production.
• The late time derivative response defines the long term well
and reservoir performance.
• Permanent downhole pressure gauges allow continuous
updating of the deconvolution which reduces the uncertainty in
future well performance.
49
Summary
Tested volumes and future well production can be
estimated from pressure transient data prior to
building complex models.
Use the rate normalized log-log derivative plot to
compare the response between build-ups and
between wells…
Derivative Comparison – Oil and Water
50
1E-4 1E-3 0.01 0.1 1 10 100 1000 10000 1E+5 1E+6
Time [hr]
1E-3
0.01
0.1
1
10
100
1000
Pre
ssure
[psi]
RubyJo #4 DST #1_standard_tmw.ks3 - Diagnosis (ref)
RubyJo #4 DST #2_standard_tmw.ks3 - Diagnosis
16-29a-15_tmw.ks3 - Diagnostic
17-12-4A_Working_File_tmw.ks3 - Main BU
20-6-3-DST1_TMW.ks3 - Diagnostic
DST1aCompleteSimplified_tmw.ks3 - Diagnostic
Guara-1 DST-1 Analysis-3_tmw.ks3 - Diagnostic
Jorbaer_DST3update_tmw.ks3 - Diagnostic
Peebs #1 (core hole) DST #2_tmw.ks3 - Diagnosis
RJS-628A_BG_TW_AllRates.ks3 - PostFrac PP
Compare files: Log-Log plot (dp and dp' normalized [psi] vs dt)
Derivative Comparison - Gas
51
1E-4 1E-3 0.01 0.1 1 10 100 1000 10000 1E+5 1E+6
Time [hr]
1E+6
1E+7
1E+8
1E+9
1E+10
1E+11
1E+12
Gas p
ote
ntia
l [psi2
/cp]
A15_July 02 2010_tmw.ks3 - No Partial Completion
Bounty_DST1a2010.ks3 - GC Main (ref)
BUpMoran27-6_tmw.ks3 - Horizontal DP
BUpOdenHeirs_tmw.ks3 - Analysis 4
ca48Canal1_tmw_new.ks3 - Channel
DST1c.ks3 - Analysis 1
Hasdrubal A1_v2.ks3 - 1-P Closed
HBH-4DST_tmw_2.ks3 - Partial Completion + Increasing h
Horseshoe-1 Interpretation_TMW.ks3 - 3 Zones
PA_v17.ks3 - Analysis 14
Endeavour_NR_v11_tmw.ks3 - homogeneous
DAP-3_CR_July2010data_5sec data_1stSept2010.ks3 - Analysis 1
Compare files: Log-Log plot (dm(p) and dm(p)' normalized [psi2/cp] vs dt)