Fundamentals of Modern

Production Decline Analysis

Shahab Gerami, Ph.D

Outline

Field Data & Its Importance

Short-term data (Well testing data)

Long-term Data (Production data)

Direct solution and Inverse Solution Methods

An Introduction to Well Performance Analysis

Flow regimes,

Basic well model

Simplifying assumptions

Diffusivity equation for slightly compressible oil,

Solution to diffusivity equation,

Constant rate solution
Constant pressure solution

Application of the Solution (Type curves)

Treatment of the Diffusivity Equation for Gas reservoirs

Production Analysis

Traditional (Empirical) Decline Curves

Production Analysis Using Type Curves (Fetkovich)

Modern Production Analysis Concepts

Material balance time
Conversion of constant pressure solution to constant rate solution
Material balance pseudo-time

Modern Production Analysis Methods

Blasingame
Agarwal

A Systematic and Comprehensive Approach

Analysis of Production Data

Reservoir
Information

Predictive
Models
(forward
solution)

Production
Analysis
Models
(backward
solution)
(i) Well test
models
(ii) Material

Production Forecast

balance models
(iii) Decline
curve analysis

Field Data
(i) Well test data
(ii)

Production data

Economic Study
and Decision
Making for the
Field Development
33

Roles of Reservoir Analysis Models

Reservoir
Information

Production
Analysis Models

Field Data
(i) Well test data
(ii) Production data

Production Data
Electronic data measurements
(Flowing pressure & Flow rate)

More sophisticated analysis methods

( Take into account both pwf & q)

Direct versus Inverse Solutions

Direct solution (Convolution)
Input

system

Output (?)

Inverse solution
(Deconvolution)
Input

System (?)

Example of a simple system

Output

Actual measurement compared

to the system

Inverse Solution Compared to Actual System

Inverse solution can be used for the identification of system characteristics
Inverse solution can result in grossly erroneous answers
Whereas the mathematics is correct, the utility of the results derived from
this mathematically process is questionable.

Characteristic of Inverse Solution

Non-unique solution (the inverse solution has its
limitation)
A good looking history match is not a good enough
answer

Input-System-Response
Input Perturbation

Reservoir
Mechanism

Output Response

Well test interpretation is essentially an inverse problem and in general is better

suited to analytical solution.

Model Input

Mathematical
Model

Model Output

The objective of well test analysis is to describe an unknown system S ( well +

reservoir) by indirect measurement ( O a pressure response to I a change of
rate).

The unknown system (S) may be a type curve and its derivative to
describe the reservoir characteristics.

Development of Hydraulic Diffusivity Equation for Flow of

a Slightly Compressible Oil and Its Solution Subjected to
Different Boundary Conditions

Physical model
Simplifying assumptions
Mathematical model

Choosing an appropriate element

Governing equation

Initial and Boundary conditions

Infinite acting

10

Constant rate production

Constant pressure production

Finite acting

Mass balance
Momentum balance (Darcys law)
Equation of state

Constant rate production

Constant pressure production

Solutions
Laplace space solutions

Time domain solutions

Simplified solutions

Applications (Drawdown (single rate & multi rate), Reservoir limit test, Build up,
Superposition (time & space), ),

10

Physical Model
Reservoir Engineering Model
Works 95+ percent of the time...
Why? Pressure and volume averaging
of reservoir properties.
When does it not work? High contrast
in reservoir properties.

Actual Reservoir Model

Complex geology.
Complex fluid behavior.
Poor lateral (and vertical) continuity.

11

Simplifying Assumptions

rw

12

Mathematical Model-Governing Equation

Mass balance

o A v r r o A v r

V
o

o V t
t

t t

Momentum balance (Darcys law)

k p
v gr
r
Equation of state
o ob exp co ( p pb )

13

13

Hydraulic Diffusivity Equation

1 p 1 p
r

r r r t
k

ct
Hydraulic diffusivity equation determines the velocity at which
pressure waves propagate in the reservoir. The more the
permeability the faster the pressure wave will propagate.

14

14

15

16

17

18

A reservoir model is the superposition of

reservoir, inner, and outer boundary conditions

Equation
p Model-Governing
ct p
1Mathematical
r

r r r
k t
Initial Condition : p pi ,
Well
production
Constant rate

Constant
19
pressure

t 0, r rw

Flow
regime
Finite
acting
(Bounded)
Finite
acting

Inner Boundary
Condition

rw

qBo

2 rw hk
rw

p wf

Outer Boundary
conditions

r
p

0
r re

0
r re

19

Dimensionless Hydraulic Diffusivity

Equation
1 p ct p
r

r r r
k t

2 p D 1 p D p D

2
rD rD t D
rD
20

van Everdingen- Hurst Constant Terminal Rate

Solution Bounded Cylindrical Reservoir
(exact solution)
n2 t D

e
J 12 n reD
2t D
p wD t D 2 n reD 0.75 2 2 2
2

reD

J
n 1
n
1
n eD
1 n

J 1 n reD Y 1 n J 1 n Y 1 n reD 0

Approximate Solutions
1. Infinite cylindrical reservoir with line-source well
2. Bounded cylindrical reservoir, pseudo steady-state flow

21

Solutions- Laplace Domain (Sabet,

.1991)
Infinite-acting reservoir

Constant rate
solution

p wD t D

Boundary dominated flow- approximate late time

p wD t D

22

1
n t D 0.80908
2

2t D
n reD 0.75
2
reD

Reservoir-Limits Test
(Estimation of Reservoir Pore Volume)
pwf

141.2qB
pi
kh

pwf
t

0.0005274k

re
0.75
t n

2
ct re

rw

0.07447 qBo
ct re2

V p re2 h

p wf

0.234qBo

t
ctV p

Deliverability Equation & Well Productivity

Index
p pwf

141.2qB

kh

q
J

p pwf

re
0.75
n
rw

kh

re
0.75
141.2 B n
rw

The late-time behavior for all reservoir sizes: is an exponential decline

Transient flow is independent of reservoir size, all reservoirs would follow the
same curve at early time (transient flow) and would only deviate at late times,
when the reservoir boundary is felt.
The higher the reservoir size, the longer the transient flow

Depletion Above the Bubble-point Pressure

Constant pressure solution

qD t D

For

4 t

1
DA

exp
4A
4A

ln
ln
2
2

C A rw

C
r
A w

t DA t pss

It may be used for reservoir limit test

4 kh p i pwf
4 t DA
ln q
ln
4A

4A
ln
B ln

2
2

C
r

C
r
A w
A w

Solutions- Laplace Domain (Sabet,

.1991)
Infinite-acting reservoir

Constant rate
solution

K 0 rD S
S S K1 S

pD S

Bounded reservoir

pD S

K r

S I 0 rD S I1 rDe S K 0 rD S
S S K1 S I1 rDe S K1 rDe S I1 S
1

De

Infinite-acting reservoir

Constant pressure
solution

Bounded reservoir
qw D S

28

K1 rD S

S
K
S
0

qD S

K1 rDe S I1 S I1 rDe S K1 S
S K1 rDe S I 0 S K 0 S I1 rDe S

Bessel Differential Equation

Modified Bessel Differential Equation

Properties of Bessel function

d
I 0 rD S S I 1 rD S
drD

d
K 0 rD S S K1 rD S
drD

Consideration of Complexibilities
accounting for the skin factor
Hawkins (1956) suggested that the permeability in the skin zone, i.e., skin, is uniform and the
pressure drop across the zone can be approximated by Darcys equation. Hawkins proposed
:the following approach

Consideration of Complexibilities- wellbore storage

In q sf B

Out qB
Accum.

d 24 wbVwb
dt

q sf B qB 24Vwb

d
dt

Vwb

We can write

1 d
dp

B C st
Define
Assume

d wb d wb dp wb
dp

wb c wb wb
dt
dp dt
dt

q sf q

24c wbVwb
B

wb

C c wbVwb

wb R

q sf q

qsf

dp w

dt

24C wb dp w
B
dt

CD

0.8936C 0.8936c wbVwb

ct hrw2
ct hrw2

Consideration of Complexibilities- Limited Flow Entry

Partial Penetration Skin

Consideration of Complexibilities- Two phase flow?

34

Components of Well Test Models

Direction (Vertical, Horizontal)
Well

Storage (Constant, Changing)

Completion (Damaged, Fractured and Acidized)
Homogeneous
Heterogeneous

Reservoir

Composite
Multilayer
Dual porosity

Flow boundaries (No flow, Constant pressure, infinite)

Boundaries
35

Geometrical boundaries (Circular, Rectangular)

Flow Regimes
Common geometric shapes
Different flow patterns may appear at different times in a single
test

Flow regimes follow sequence within model

Reservoir Flow Geometry

Radial flow

Linear flow

Spherical flow

37

Hemispherical flow

Dimensionless Variables
qB 948ct r 2
p pi 70.6
Ei

kh
kt

r
w

rD

r
rw

kh pi p
1

Ei
141.2qB
2
0.0002637 kt

4
2

ct rw

kh pi p
0.0002637 kt
pD
tD
141.2qB
ct rw2

1
rD2
pD Ei
2 4t D

Dimensionless Variables
Radial Flow With WBS And Skin
kh pi p
pD
141.2qB

0.0002637 kt
tD
ct rw2
r
rD
rw

khps
s
141.2qB

0.8936C
CD
ct hrw2

Type Curves
A type curve is a graphical representation of the theoretical solutions to
flow equations.

Type Curve Analysis

1.

Identify wellbore storage and middle time regions on type curve.

2.

Identify pressure response for a well with high, zero, or negative

skin.

3.

Calculate equivalent time.

4.

Calculate wellbore storage coefficient, permeability, and skin factor

from type curve match.

Gringarten Type Curve

Constant rate production
Vertical well
Infinite-acting homogeneous reservoir
Single-phase, slightly compressible liquid
Skin factor can be modeled with an apparent radius
Constant wellbore storage coefficient

Surface Rate

Rate

Bottomhole
Rate
Time

q qsf B
dpw

dt
24Vwb cwb

Pressure Type Curve

CDe2s=100

D
p

10
10
1
0.1
0.01

CDe2s=1060

CDe2s=0.01

Derivative Analysis

Derivative: the slope of the semi-log plot of pressure versus time.

dp
d ln t

Perhaps one of the major advantages in using the pressure derivative

in conjunction with pressure is the identification of the flow regime.
wellbore storage,
skin,
closed outer boundary,
vertically fractured well and others).

Derivative Analysis: Transient Radial Flow Regime

162.6qBo
pi pwf t pwf
kh

log t log

k
3.23 0.87 S
2
ct rw

dpwf

162.6qBo

d log t
kh

162.6qBo
dpwf

0 log t log
log
kh
d log t

Derivative Analysis: P.S.S Radial Flow Regime

141.2qB
pi p wf p wf
kh

0.0005274 k

re
0.75 S
t n

ct re
r

1
d log t dt
t

dpwf

dpwf

0.1715qBo
2.3026
t t
d log t
dt
hct re2

0.1715qBo
dpwf
1 log t log
log
hc r 2
d
log
t

t e

Derivative Type Curve

D
D
tp'

10
10
1
0.1
0.01

CDe2s=1060
CDe2s=100

CDe2s=0.01

Pressure And Derivative Type Curves

D
D
D
p,tp'

10
10
1
0.1
0.01

Time Regions On The Type Curve

100

WBS

Transition

Radial Flow

pD

10

pD=0.5

Horizontal Derivative

Unit Slope Line

0.1

Early Time
Region
0.01
0.01

0.1

Middle Time
Region
10

100

tD/CD

1000

10000

100000

Estimating Skin Factor

100

High Skin

pD, tDpD'

10

No Skin
1

0.1

Negative Skin

0.01
0.01

0.1

10

100

tD/CD

1000

10000

100000

Type Curve Matching

Plot field data on log-log scale
Align horizontal part of field data and type curve derivative
Align unit slope part of field data and type curve
Select value of CDe2s that best matches field data

TCMATCH.WTD (Field Data)

10000

Pressure change, psi

1000

100

10

1
0.001

0.01

0.1

Equivalent time, hrs

10

100

1000

TCMATCH.WTD (Drawdown type curve, Radial equivalent time)

Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100

10

0.1

Pressure change, psi

Dimensionless pressure

1000

100

10

0.01
1
0.001

0.01

0.1

10

100

1000

Equivalent time, hr
0.001
0.001

0.01

0.1

10

Dimensionless time

100

1000

10000

100000

TCMATCH.WTD (Drawdown type curve, Radial equivalent time)

Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100
1000

Pressure change, psi

Dimensionless pressure

10
100

10

0.1
1
0.001

0.01

0.1

10

100

1000

Equivalent time, hr
0.01

0.001
0.001

0.01

0.1

10

Dimensionless time

100

1000

10000

100000

TCMATCH.WTD (Drawdown type curve, Radial equivalent time)

Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100
1000

Pressure change, psi

Dimensionless pressure

10
100

10

0.1
1
0.001

0.01

0.1

10

100

1000

Equivalent time, hr
0.01

0.001
0.001

0.01

0.1

10

Dimensionless time

100

1000

10000

100000

TCMATCH.WTD (Drawdown type curve, Radial equivalent time)

Radial flow, Single porosity, Infinite-acting: Varying CDe2s
100

CDe2s=7x109

1000

Pressure change, psi

Dimensionless pressure

10
100

10

0.1
1
0.001

0.01

0.1

10

100

1000

Equivalent time, hr
0.01

0.001
0.001

0.01

0.1

10

Dimensionless time

100

1000

10000

100000

Interpreting Type Curve Match

Calculate k from the pressure match point ratio p/pD
Calculate CD from the time match point ratio teq/tD
Calculate s from the matching stem value CDe2s

The Diagnostic Plot

1. Identify time regions.
2. Identify flow regimes.
3. List factors that affect pressure response in early time.
4. List boundaries that affect pressure response in late time.

Pressure change, derivative, psi

Time Regions on the Diagnostic Plot

Early-time
region

Elapsed time, hrs

Middletime
region

Late-time
region

Volumetric Behavior
Wellbore Storage

qBt
p
24C
Pseudosteady-State Flow
pi pwf

0.0744qBt 141.2qB

2
kh
ct hre

re 3
s
ln

rw 4

General Form

p mV t bV

Derivative

mV t bV
p
t
t
mV t
t
t

Pressure change, derivative

Volumetric Behavior

1
Time

Radial Flow
General Form

Derivative

p m log t b
p
m log t b
m
t
t

t
t
2.303

Flow Periods for Vertically Fractured Well

Linear Flow
Hydraulic Fracture

4.064qB kt

khL f ct

General Form

p mLt1 2 bL

12

p
mL t 1 2 bL
1
t
t
mL t 1 2
t
t
2
Pressure change, derivative

Derivative

Time

Bilinear Flow
Hydraulic Fracture
44.1qB 1
p
wk f
h

12

c k
t

General Form
14

p mB t 1 4 bB
Derivative

p
mB t 1 4 bB 1
t
t
mB t 1 4
t
t
4

Spherical Flow
pi pwf

4krp

ct rp2
1
kt

General Form

p bS mS t 1 2

Derivative

bS mS t 1 2
p
1
t
t
mS t 1 2
t
t
2

Flow Regimes on the Diagnostic Plot

Wellbore
storage

Radial
flow
Spherical flow

Recharge?

Challenges To Deal with In Gas Reservoirs

1 k p
p

r c
r r r
t

p
p
c

2
r
r
2

1 p c p

r r
k t

Assumptions are usually made for liquid flow:

Viscosity is independent of pressure.
The pressure gradient is small and therefore is negligible.
The liquid compressibility is small and constant, so that the
product.
2

p 1 p c p

2
r r
k t
r

69

Forchheimer Equation

Challenges To Deal with In Gas Reservoirs

compressibility

1. Gas has a non-linear inflow relationship flowing

wellbore pressure is not proportional to well flow rate
2. Gas properties significantly changes with pressure

reservoir pressure

For gas wells, compressibility (and viscosity) can not

be considered constants at low reservoir pressure
71

Gas Flow Governing Equations

Treatment of Gas Flow Equations in

a Porous Medium
g
1 r g v gr


r
r
t

Al-Hussainy and Ramey (1966)

Pseudo-pressure Transformation
(1) Variable compressibility factor
(2) Variable viscosity

Fraim and Wattenbarger (1987)

Pseudo-time Transformation
(1) Variable compressibility
(2) Variable viscosity

p
p 2
dp
Z
p

dt
t a p t
ct
t

1 i cti
r

r r r
k t a

74

Concept of Pseudo-Time
Pseudo-time corrects for gas compressibility non-linearity:

dt
ta ct i
0 c t
t

Evaluated at average reservoir pressure

Using pseudo-time requires you to know how the reservoir pressure varies with time
Since this is usually not known, the process is iterative:
1. Estimate Gi (gas in place)
2. Calculate pR from pi and cumulative production
3. Calculate ta, and plot data using ta
4. Calculate Gi from type-curves
5. Use new Gi to re-calculate ta (steps 1-3)
6. Repeat steps 4-5 until convergence

History of Production Analysis

1920's Purely Empirical Approaches
1940's Empirical Approaches, but some physics is creeping in (Arps, etc.)
1960's Decline Type Curves (Fetkovich, etc.)
1980's Variable Rates/Pressures (Blasingame, etc.)
2000+ Fully Integrated Production Analysis

76

The objective of Production Analysis:

A reliable forecast
Modern decline analysis uses production rates and flowing pressures
Reserve evaluation tool that can provide reliable estimates of
recoverable reserve and fluid in-place
Practical and economic alternative to well testing for evaluation of skin
and permeability
Uses pressure transient theory ( mathematics is similar to the well test
but the focus is different- long term data not short term transient data)

77

Analysis of Production Data

Arps
Fetkovich
Blasingame
Agarwal-Gardner
Flowing Material Balance

Some Methods Yield Recoverable Reserves, while Others Give

Hydrocarbons-in-place.
78

Thank you for your attention . . .

79

Production Model
(3) Deliverability Equation
wf

q w p sc T re 3

p
ln
Tsc kh rw 4

80

Estimation of kf
pscTq
p i p
2 khTsc

Slope

kh

n t a

n t a

t a
n 2 0.80908
rw

pscTq

2 khTsc

pscTq

2 Tsc Slope
81

Well test data vs. Production Data

Well Test Data

Production Data

Constant rate assumption

Constant pressure assumption

Smooth/ Minimal disturbances

Noisy/ Disturbances

Measured at bottom hole

conditions

Measured at surface conditions

Frequent measurements

Average over time periods

Controlled test

Missing/ Inconsistent data

The Volumetric Method

Does not incorporate production performance data

Depends on parameters that are not well known( areal extent)

The Material Balance Method

Uses real production data

Expansion of reservoir fluid= Production
Reservoir pressure declines over the time

Traditional Decline Curve Analysis Method

t
Uses real production data
Empirical relation
Gives expected ultimate recovery