SPE 80513
A New Method for Determination of Gas Reservoir Parameters with
Rate Decline Data
M. Li, L.T. Sun, and S.L. Li, Southwest Petroleum Institute
Copyright 2003, Society of Petroleum Engineers Inc.
This paper was prepared for presentation at the SPE Asia Pacific Oil and Gas Conference and
Exhibition held in Jakarta, Indonesia, 15–17 April 2003.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
For the boundary-dominated flow in a gas reservoir producing
against constant wellbore pressure, Fraim proved that
normalized time VS. rate data traces Arps exponential
decline curve. However, it is tiresome to match type-curve
when normalized time is introduced. To obtain normalized
time, G, a value for original gas in place, is assumed and is
then used to calculate normalized time and to match the
curve with the Arps’ base b curve closest to the exponential
curve. A new value of G is worked out. This process will
be repeated several times until convergence is
obtained for the exponential decline curve.
Technique developed in this paper provides a rapid and
accurate way to analyze rate decline data. Normalized time
VS. rate data follows the exponential decline curve,
indicating non-linear regression technique can be applied to
tackle the problem. Regression technique is used to obtain
the exponential decline parameters Di and qi based on which
reservoir parameters deteremined with Fraim’s technique can
be easily derived from formulas presented in the paper. The
put technique matches Arps' exponential curve automatically
and determines reservoir parameters without type-curve
matching . The validation and procedure of the present
technique with rate decline data from Fraim’s literature is also
described in the paper.
Introduction
Fetkovich1,2 introduced the idea of log-log type-curve analysis
to single-well analysis for both transient- and boundary-
dominated flow periods. Boundary-dominated flow for
constant pressure production is similar to pseudosteady-state
flow for constant-rate production His type-curve analysis was
intended to be a rapid way to estimate performance when a
well is producing against a constant bottomhole
pressure(BHP). From his type-curves, future performance can
be forecast along with estimates of oil and in place and
ultimate recovery.
For boundary-dominated flow, Fetcovich showed the Arps3
family of curves with b as a parameter. The b=0 case is for
exponential decline of a liquid reserve Fetkovich used the
curve of b between 0 and 1 for matching solution-gas-drive
depletion and gas reservoir depletion. Matching and
extrapolating these curves is equivalent to using the harmonic
and hyperbolic declines with the usual semilog decline curves.
He showed cases where these curves were useful.
Fraim4 concerned with the long-term boundary-dominated
flow of real gases in closed reservoirs The purpose of his
paper was to improve the use of Fetkovich’s type curves for
gas well analysis. A normalized time was introduced that
applied to the viscosity and compressibility at the average
pressure. Fraim proved that normalized time VS. rate data
traces Arps exponential decline curve. The normalized time
was used for boundary-dominated flow analysis. Only Darcy
flow was taken into account
This paper presents a rapid and accurate method to analyze
rate decline date on the basis of Fraim’s work. Non-linear
regression technique was applied to obtain the exponential
decline parameters Di and qi based on which reservoir
parameters determined with Fraim’s work can be easily
derived from formulas presented in the paper. The put
technique matches Arps' exponential curve automatically and
determines reservoir parameters without type-curve matching .
The validation and procedure of the present technique with
rate decline data from Fraim’s literature is also described in
the paper.
Rate Decline Data Theory
Fraim4 showed that the depletion of a closed real gas reservoir
could be expressed as an exponential decline if a normalized
time was used. The normalized time was defined as
t µ ( p )c g ( p ) ……………(1)
tn = ∫
o (µcg )i
dt
The exponential decline expression is shown as follows
 2 J g ( p / z)i 
q = qi exp − .t n  ……….……(2)
 G ( µc g ) 
2 SPE 80513

The expression for Jg is defined as qi = exp(a) ……......……..………………..(11)

1.987 × 10−5 k g hTsc …..…..….…………...(3)

Jg = Determination of Gas Reservoir Parameters
1  2.2458 A  We use the pseudo-steady equation for a real gas in terms of pp
pscT ln  
2  C Arw2  in the form of a productivity index equation as5,6

Defining dementionless normalized time and q = J g ( p p − p pwf ) ………………….…..(12)

dementionless producing rate, Fraim presents type curve
method for determining G, kgh and kg/φ.
Replacing pp with ppi, we obtain gas producing rate at initial
Regression Analysis condition
Gas Producing rate traces Arps exponential decline
expression(see Eq.2), Thus we have qi = J g ( p pi − p pwf ) ………….…..……….(13)

q = qi ⋅ exp( − Di t n ) ……………….……...……..(4)
After determining decline parameters Di and qi in Eq.4, the
original-gas-in- place G can be derived from Eq.4 and Eq.5
Where Di is defined as

2 J g ( p / z) i …………….………..…..……..(5)
Di = qi 2( p / z ) i
G= ………...… .…..(14)
Di (µc g )i ( p pi − p pwf )
G ( µC g ) i

From Eq.4 we obtain

Replacing p with pi in Eq.12, we have the pseudosteady-
ln q = ln qi − Di t n ……………………..…….(6)
state equation for a real gas at initial condition. From the
We set equation we have

Y = ln q a = ln qi
q i p sc T 1 2 ln(2.2458 A / c A rw2 )
kgh = .…...….(15)
X = tn b = − Di 1.987 × 10 − 5 Tsc ( Ppi − Ppwf )

Then Eq.6 can be expressed as

Original gas in place G can be expressed as

Y = a + b ⋅ x ……………………..….……(7)
 p T ………..…...…...(16)
The estimates of a, b in above equation can be determined by G =   ⋅ sc ⋅ ( A − πrw2 )hϕS gi
using regression analysis. The estimates are  z i pscT

n
 n  n  Substituting G in Eq.16 into Eq.5, we have
∑x y
i =1
i i −  ∑ xi  ∑ yi 
 i =1  i =1  ………..……………..(8)
b= 2
n
 n  A 2.2458 A
n∑ xi2 −  ∑ xi  ( − rw2 ) ln( )
i =1  i =1  kg π C A rw2 …..………...(17)
= Di S gi
1 bn n
ϕ 0.00633 / 4( µc g ) i
a = ∑ y i − ∑ xi …….………….……………(9)
n i =1 n i =1
Where Finally we have Eq.15, 16 and 17 from which parameters
obtained with Fraim type curve matching can be worked out
xi = (t n ) i yi = ln(q ) i without typ curve matching. Notice that Eq.15, 16 and 17 are
deduced without taking water and rock compressibility into
account. Replacing Cg with Ct, intergral expression for tn
Once we obtain the estimates of a,b in Eq.7, we can have Di should be modified as follows4
and qi in Eq.4
t ( µct ) i dt
tn = ∫
[ ]
……...…(18)
Di = −b …………………..……………………..(10)
0
µ ( p )c ( p ) 1 − ( p i − p ) c f
SPE 80513 3

Example rw= effective well radius, ft

The example data(Tab.1) is from Fraim’s literature case 1. The s=saturation
results(data vs rate) calculated with a numerical simulator are t=time, days
shown in Tab.2. Fraim used these data to test the validation of tn = normalized time, days
his normalized-time transformation.
As same as Fraim technique, we need to estimate G and use T = reservoir temperature, °R
it to calculate normalized time. To obtain a relative accurate Tsc= temperature at standard conditions, °F
G, we use actual time to replace normalized time in Eq.4. at ϕ =porosity, fraction
first iteration Subscripts
Iteration 1 (Real Time) g = gas
Di=8.913×10-4, qi=58.184, substituting it into Eq.14, we i = initial condition
obtain G =87.63 Bscf sc = standard condition
Iteration 2 (Normalized Time) t =total
Di=1.402 × 10-4, qi=64.61, substituting it into Eq.14, we w = water
obtain G =60.53 Bscf wf = wellbore flow condition
Iteration 3 (Normalized Time)
References
Di=1.130×10-4, qi=63.983, substituting it into Eq.14, we
1. Fetkovich, M.J.:”Decline Curve Analysis Using Type-
obtain G =61.5Bscf Curves,”JPT(June 1980) 1065-77.
From Eq.15 we have kgh=60.80(md.m) 2. Fetkovich,M.J.:”Decline-Curve Analysis Using Type-
From Eq.17 we have kg/φ=32.50(md.m Curves-Case Histories,”SPEFE(Dec. 1987) 637-56.
Comparison is made in Tab.3. From Tab.3 we can see that 3. Arps,J.J.:”Analysis of Decline Curves,” Trans.,
the results in this paper and Fraim’s work equals nearly real AIME(1945)160, 228-47.
value. however, the present method do not need any type 4. Fraim, M.L and Wattenbarger, R.A.:”Gas Reservoir
curve match Decline-curve Analysis Using Type Curves with Real Gas
Psudopressure and Normalized Time,” SPEFE(Dec.
Conclusion 1987) 671-82.
1. This paper presents a new method to calculate gas 5. Al-Hussainy.r.,Ramey,H.J.Jr. and Crawford, P.B:”The
reservoir parameters without type curve matching by flow of Real Gases Through Porous Media,”JPT(May
using regression technique. 1966) 624-36;Trains., AIME,237.
2. The validation test shows in the paper that Gas Reservoir 6. Russell,D.G. and Prats, M.”The Practical Aspects of
parameters such as G, kgh and kg/ϕ can be worked out Interlayers Cross-flow,”JPT(June 1962)589-94
with the method presented in this paper.
Si Metric Conversion Factors
Acknowledgment
We thank the support of CNPC to its key Laboratory for cp × 1.0* E-03 = Pa.s
Special Gas Reservoir Development we are working
ft × 3.048* E-01 = m
Nomenclature ft2 × 9.290 304* E-02 = m2
A = area，ft2 ft3 × 2.831 685 E-02 = m3
b = decline-curve paprmeter psi × 6.894 757 E+00 = kPa
cf = rock compressibilty , 1/psia
cg = gas compressibity, 1/psia *conversion factor is exact
ct = total compressibility, 1/psia
cw= water compressibility, 1/psia
c A = Deitz shape factor
G = original gas in place，Mscf
Gp= cumulative gas production，Mscf
H = formation thickness, ft
Jg = gas productivity index, Mscf/D-psia2-cp
K = formation permibility,md
p = pressure, psia
p = average reservoir pressure, psia
2
Pp = real gas pseudopressure, psia /cp
psc= pressure at standard condition, psia
pwf= wellbore flowing pressure, psia
qg =surface flow rate, Mscf/d
qiii=initial gas rate, Mscf/d
4 SPE 80513

TABLE.1-RESERVOIR DATA TABLE.3 COMPARSION

Shape Case 1
Parameters Result this paper Result from Fraim’s real value
h, md 200
G (Bscf) 61.50 61.50 17.40
k, md 10 Kgh (md.m) 60.80 60.80 60.90
T, °F 200 kg/ϕ (md) 32.50 32.50 32.0

φ, fraction 0.30
cf, 1/psia 3.0×10-6
C
w,
1/psia 3.0×10-6
Sgi, fraction 0.99
rw, ft 0.25
s 0
A, ft2 3.848×107
pi, psia 5,000
pwf, psia 3,000
G, Bscf 61.51
Gas Gravity 0.601

TABLE 2- CASE DATA FROM FRAIM

DATA
Producing rate Time
(Mscf/D) (days)
30,400 0.052
27,300 0.180
22,200 0.729
20,100 17.7
18,100 83.2
16,400 174
14,700 83.2
13,000 520
9,650 668
7,990 880
6,330 1,150
4,750 1,490
3,250 1,950
1,800 2,690
921 3,550
472 4,410