See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/272697082

WELL DELIVERABILITY PREDICTIONS OF GAS FLOW IN GAS-CONDENSATE

RESERVOIRS, MODELLING NEAR-CRITICAL WELLBORE PROBLEM OF TWO
PHASE FLOW IN 1 -DIMENSION

Article  in  Brazilian Journal of Petroleum and Gas · December 2012

DOI: 10.5419/bjpg2012-0013

CITATIONS READS

4 1,054

4 authors, including:

Oyinkepreye Orodu Favour Makinde

Covenant University Ota Ogun State, Nigeria Robert Gordon University
81 PUBLICATIONS   333 CITATIONS    6 PUBLICATIONS   47 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evaluation of the Rheological Parameters of Local Bentonite Mud treated with Nanoparticles and its Impact on Drilling Performance View project

Holistic Wellbore Sidetrack Evaluation for Reservoir Management View project

All content following this page was uploaded by Oyinkepreye Orodu on 29 August 2016.

The user has requested enhancement of the downloaded file.

 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

WELL DELIVERABILITY PREDICTIONS OF GAS FLOW IN GAS-CONDENSATE

RESERVOIRS, MODELLING NEAR-CRITICAL WELLBORE PROBLEM OF TWO
PHASE FLOW IN 1 -DIMENSION

a
Orodu, O. D. 1; a Ako, C. T.; a Makinde, F. A.; a Owarume, M. O.

a
Covenant University, Department of Petroleum Engineering, Ogun State, Nigeria

ABSTRACT
Production of gases from gas-condensate reservoirs are known to bear certain challenges largely due to
the formation of retrograde condensates that hinder gas flow. The drop out of this liquid creates flow
regions that are characterized by the liquid saturation as it affects the mobility of the two phase flow,
thereby preventing the effective modeling of well productivity. In this study, a predictive model based on
an analytical approach is developed to predict gas flow in gas condensate reservoirs. This study compares
the estimated gas flow from the developed model for gas-condensate reservoirs to the flow of an existing
model for gas reservoirs. This study observes the effects of liquid drop-out on productivity at low
pressures and the condensate unloading pressure, which is comparable to that of commercial software.

KEYWORDS
well deliverability modeling; wellbore problems; two- phase flow; 1-Dimension

1
To whom all correspondence should be addressed.
Address: Department of Petroleum Engineering, School of Engineering, College of Science & Technology, Covenant University,
Km. 10 Idiroko Road, Ota, Ogun State, Nigeria
Telephone: +234 (0) 7061132990 | E-mail: [email protected], [email protected]
doi:10.5419/bjpg2012-0013

159
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

1. INTRODUCTION reservoir pressure below saturation pressure of

fluid-in-place. Later, Fussell (1973) modified the
Reservoirs bearing gas condensates are version of 1-D radial model developed by Roebuck
becoming more common as developments are et al. (1969) to study long term well performance.
encountering greater depths and higher pressures Despite the modifications, the condensate
and temperatures. Accuracy in engineering accumulations in the producing region remained
computations for gas-condensate systems such as greater than the ones obtained experimentally for
precision in well testing, in estimating reserves, in the constant volume depletion process. Cable et al.
sizing surface facilities, and in predicting (2000) adopted the use of special core analysis
productivity trends depends upon basic data for near-well relative permeability to model
understanding of phase and flow relationships. productivity in a full-field model. Their study
When comparing dry gas reservoirs with gas- considers the importance of liquid recovery and
condensate reservoirs, one can observe the change in yield, composition gradients, and
existence of many special factors that affect the reduction in well deliverability caused by
performance of gas-condensate reservoirs during condensate blockage.
the exploitation process.
Because the most important and complex
At the time of discovery, a typical gas- phenomena associated with condensate banking
condensate reservoir pressure might be above or and productivity reduction is relative permeability,
close to the critical pressure. At this time, there there have been many investigations of gas
exists only single-phase gas. However, as the condensate relative permeability. Hinchman and
production is carried out, decline in isothermal Barree (1985) showed that the productivity above
pressure occurs. Also, and as the bottom-hole the dew point pressure is controlled by the
pressure in a flowing well falls below the dew point reservoir’s permeability and thickness, in addition
of the fluid, a liquid-hydrocarbon phase is formed. to the viscosity of the gas. Below the dew point,
The formation of a retrograde condensate results the degree of productivity reduction is controlled
in a buildup of a liquid phase around the wellbore, by the critical condensate saturation and the shape
leading to a decrease in the effective permeability of the gas and condensate relative permeability
to gas into the wellbore. The productivity loss curves, as well as the choice between imbibitions
associated with condensate buildup can be and drainage relative permeability curves. Whitson
substantial. Afidick et al. (1994) and Barnum et al. et al. (1999) showed that relative permeability
(1995) listed several instances in which well effects in gas-condensate reservoirs can be
productivities have been reported to decline by a classified into three categories: near well steady-
factor of two to four as a result of condensate state gas/oil flow; bulk of reservoir, beyond well
accumulation. Barnum et al. (1995) found that vicinity, where liquid mobility is zero/ negligible;
production loss is severe for low productivity and water encroachment, where gas and/or
reservoirs, such as those with a Kh less than retrograde condensate are trapped.
1000md-ft. They reported that the critical
condensate saturation range from 10-30% and can Bauget et al. (2005) developed a novel
decrease the productivity by a factor up to five due approach for calculating representative field
to condensate accumulation near the well bore. relative permeability. This method is based on a
physical model that takes into account the various
Fevang (1995), Ali et al. (1997), and Gringarten mechanisms of the process: bubble nucleation
and Al-Lamki (2000) showed that, when reservoir (pre-existing bubbles model), phase transfer
pressure around a well drops below the dew point (volumetric transfer function), and gas
pressure, retrograde condensation occurs and displacement (bubble flow). In the model, the
three regions are created with different liquid researchers identified a few invariant parameters
saturations. which were not sensitive to depletion rate and
were specific to the rock/fluid system. These
Various models have been developed to study invariant parameters were determined by history
deliverability. Among these models is the steady- matching one experiment at a given depletion rate.
state flow concept by O’Dell and Miller (1965),
with pessimistic production rates at average Jamiolahmady et al. (2006) used a large data

160
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

bank of gas/condensate relative permeability to number of reservoir and well parameters.

develop a general correlation accounting for the Condensate blockage is deemed important if the
combined effect of coupling and inertia as a pressure drop from the reservoir to the wellbore
function of fractional flow. The parameters of the represents a significant percentage of the total
new correlation were either universal, applicable to pressure drop from reservoir to delivery point (e.g.
all types of rocks, or could be determined from a surface separator) at the time and after a well
commonly measured petrophysical data. And, goes on decline. Reservoirs with low-to-moderate
Bozorgzadeh and Gringarten (2007) showed that permeability (<10–50md) are often considered
well deliverability depends mainly on the gas problem wells and it becomes critical to address
relative permeabilities at both the end point and the condensate blockage issue properly. Wells with
near wellbore saturation, as well as on the high kh products (>5–10,000 md-ft) are typically not
reservoir permeability. affected by reservoir pressure drop because the
well’s deliverability is constrained almost entirely
This work studies well deliverability by the tubing. In this case, condensate blockage is a
(productivity) predictions of gas flow in a gas- non-issue. In terms of reservoir well performance,
condensate reservoir modeling near critical well- the near-well behavior, determined by the near-
bore problems in 1-D. Since gas condensate is a well relative permeability functions, is the
mixture of gas and liquid, this work will focus on dominant factor.
finding the optimum way to improve gas flow
(reducing condensate build up) around the well The diffusivity equation (Eq. 1) is solved based
bore. on dimensionless groups with respect to certain
conditions and assumptions which are elaborated
and stated below.

2. METHODOLOGY
The typical chemical composition of a gas- 1   p  Ct p
r   (1)
condensate mixture is dominated by volatile r r  r  k t
components such as methane, and a rather ‘small’
amount of heavy hydrocarbon components An expression for total compressibility (Eq. 2)
(<15mol-%). Despite its composition, these heavier with respect to the pressure derivative of oil
components make up a considerably larger formation volume factor and oil saturation was
percentage of the liquid phase, retrograde proposed. It was derived based on the combination
condensate, formed during pressure decrease continuity equation for the rate of in and out a
below an upper dew point (McCain Jr., 1989). This control volume made up of a porous medium for
study is based on a fluid characterized by C7+ of oil and gas respectively (see Appendix). This is
13.35mol-% and molecular weight of 144.77g/mol.; incorporated into the diffusivity equation.
GOR, 3414.6 scf/sepp bbl; 51.2OAPI, and reservoir
fluid molecular weight of 40.22g/mol.  So Bo  t 
 So     Ct (2)
 Bo  o 
For practically any retrograde condensate
reservoir, the condensate saturation is, throughout where
most of the reservoir, so low that its mobility is
much less than gas mobility. For practical purposes   1  So
it can be considered immobile. Nevertheless, this Bo    and So 
p  Bo  p
gas-dominated flow behavior is not at all correct in
the vicinity of gas-condensate wells, where Where r represents the radius; p, pressure; φ,
condensate saturations often reach high values porosity; µ, viscosity; k, permeability; Ct, total
(>50%), and oil permeability may exceed gas compressibility; t, time; So, oil saturation; Bo, oil
permeability (krg/kro < 1). formation volume factor; λt, total mobility, and λo,
Condensate blockage near the wellbore may oil mobility.
reduce gas well deliverability appreciably, though The reservoir is categorized pseudo-steady in
the severity of the obstruction depends on a nature having a constant production. The wellbore

161
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

is classified as finite. The flow in the reservoir is Substituting the initial conditions, Eq. 4 into Eq.
also considered with a no-flow outer boundary. 8:
Physically, a no-flow outer boundary could be
sealing faults or pinch-outs. For the mathematical 1 d  dpD 
interpretation of these conditions, certain  rD   UPD (10)
rD drD  drD 
mathematical relations such as Laplace and Bessel
functions are put in use. Also, to make the Where Eq. 10 has the form of the modified
simplification easier, dimensionless groups are Bessel equation,
used. From the diffusivity equation, the partial
differential equation is given by: 1 d  dv 
 x   v  0 (11)
x dx  dx 
1   pD  pD
 rD  (3)
rD rD  rD  tD This represents the general solution for:

where rD, pD and tD are dimensionless radius, V  AI o (  x)  BKo (  x) (12)

dimensionless pressure and dimensionless time,
respectively. Therefore, the general solution of Eq. 10 gives:

At initial conditions, PD  AI o ( U rD )  BKo ( U rD ) (13)

pD (rD , tD  0)  0 (4) Where “A” and “B” are arbitrary constants

determined by the boundary conditions.
The inner boundary condition is constant rate
production, Taking the Laplace transform of the boundary
condition, Eq. 6 gives:
 pD 
 rD   1 (5)
 rD ( rD 1)  dPD 
  0 (14)
 drD  reD 
The outer boundary condition is for a no-flow
boundary, Differentiating Eq. 13, one can obtain:

 pD   dPD 
  0 (6)    A U I1 ( U rD )  B U K1 ( U rD ) (15)
 rD reD  drD 

Taking the Laplace transform of the partial At rD = reD, this gives,

differential equation, Eq. 3;
0  A U I1 ( U reD )  B U K1 ( U reD ) (16)
 1   pD
 
  pD 
L     L   (7) Writing “A” in Eq. 16 in terms of “B”, gives:
 rD rD  rD
 
  tD 

This converts the partial differential equation B U K1 ( U reD ) K1 ( U reD )

A  B (17)
into ordinary differential equations, A U I1 ( U reD ) I1 ( U reD )

1 d  dpD  Using the other conditions to solve for the

 rD   UPD  PD (rD , tD  0) (8) constants “A” and “B”.
rD drD  drD 
The inner boundary in Laplace space is:
Where;
 dpD  1
 p   rD   (18)
L    UPD  P(t  0) (9)
 t   drD  rD 1 U

PD is the Laplace transform of pD. Substituting Eq. 18 into Eq. 15:

162
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

1 k ko kg 1 1
  A U I1 ( U reD )  B U K1 ( U reD ) (19)   ;  (26a,b)
U  o g Ct So Co  S g Cg  C f

Substituting Eq. 17 into Eq. 19 and solving for B, Therefore,

B
1 
 I1  U reD  
 tD 
0.0002637t  k g k g 
  
1 

(20) (27)
U 2  K1
3
 
U reD I1 ( U )  I1  U reD  K1 ( U ) 

 rw
2 
  g  g  So Co  S g Cg  C f 

One can now solve for “A”, by substituting the re

reD  (28)
relationship for the constant “B” given in Eq. 20 rw
into Eq. 17:
Where rw is well radius; re, reservoir radius; ko,
1
A 3 
 K1  U reD  
 relative permeability to oil; kg, relative permeability
(21)
U 2  K1  
U reD I1 ( U )  I1  U reD  K1 ( U ) 

to gas; µo, oil viscosity; µg, gas viscosity; So, oil
saturation; Sg, gas saturation; Cf, formation
Substituting Eq. 21 and Eq. 20 into Eq. 13: compressibility; Co, oil compressibility; Cg, gas
compressibility and others are as defined for

PD  
  
1  K1 U reD I o ( U rD )  I1  
U reD K o ( U rD ) 

Equations 1 to 2.
(22)
3

U 2  K1 U reD I1 ( U )  I1  
U reD K1 ( U ) 
  ko kg 
pwD 
h
141.2qB
 pi  pwf     
 g 
(29)
This is the dimensionless pressure solution in  o

terms of U and rD. Using the Cauchey’s inversion

theorem (O’Neil, 1987), the solution becomes: Substituting Eq.29, 28 and 25 into Eq. 24:

 reD2 ln rD  3reD  4reD ln reD  2reD  1

4 4 2
 rD22
PD  rD , tD   2  t  
 reD  1  4 D   reD2  1 4  reD2  1
2

(23)
(30)
2 r
2
 r ln rD
2
3reD4  4reD4 ln reD  2reD2  1
  tD   
D eD

reD  1  4   r  1 4  reD2  1
2 2 2
eD

For the pressure at wellbore, PwD, reD>>1, Eq. 23

becomes: Where pwf is bottom hole flowing pressure; pi,
initial reservoir pressure; q, flow rate; h, reservoir
2 1  r ln1 3r
2 4
4r 4 ln r 1 height,
1 2t others
and 3 are as previously defined.
PwD  tD   2   tD   eD 2  eD4  eD 4 eD  2  4  2D   ln reD
 reD   4   reD  4reD 4reD 2reD 4rThis eD equation
reD  4
(Eq. 30) shall be used to predict
(24)
gas production performance in comparison with
 reD ln1 3reD 4reD ln reD
2 4 4
1 1 2tD 3
         ln r existing solution.
  reD2  4reD 2reD2 4reD4  reD2  4
4 eD
4reD4

In order to convert to field units, the

dimensionless groups have to be properly 3. RESULTS AND DISCUSSION
expressed.
The objective of this study is built around the
0.0002637t k 1 accurate prediction of the deliverability of gas in a
tD    (25)
 rw2  Ct gas condensate reservoir. It takes into
consideration the various limiting factors to the
Since this experiment considers two phases (gas proper production from such reservoirs. In this
and oil): context, condensate blockage represents the main
problem experienced in these reservoirs, and is
generally known to alter the deliverability as a

163
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

Table 1. Reservoir and well parameters.

Reservoir Pressure 4903 psia

o
Reservoir Temperature 200 F
Porosity 0.20
Saturation Pressure 4440 psia
Reservoir Radius 800 ft
Reservoir Wellbore Radius 0.369 ft
Reservoir Depth 10000 ft
Depth of Mid Perforations 8000 ft
Well head Pressure 1000 psia
o
Well head Temperature 40 F

Table 2. Reservoir fluid PVT data.

Pressure Gas Density z-Factor Gas FVF Relative Gas Viscosity Gas Compr. Gas Oil
3 -1
psia g/cm cu ft/SCF Vol. cp psia Saturation Saturation

4440 0.429 0.818 0.00344 1.000 - - 1.00000 0.00000

4415 0.249 0.890 0.00375 1.002 0.0298 3.54E-05 0.80000 0.20000
4295 0.243 0.882 0.00382 1.012 0.0291 4.13E-05 0.74375 0.25625
3858 0.222 0.860 0.00415 1.066 0.0265 5.13E-05 0.68750 0.31250
3453 0.200 0.846 0.00456 1.141 0.0243 6.11E-05 0.63125 0.36875
3048 0.175 0.841 0.00514 1.241 0.022 6.49E-05 0.57500 0.42500
2241 0.126 0.848 0.00705 1.503 0.0183 7.92E-05 0.51875 0.48125
1433 0.076 0.883 0.00721 2.752 0.0155 8.02E-05 0.46250 0.53750
1044 0.055 0.907 0.00750 3.271 0.0146 8.76E-05 0.40625 0.59375

result of reduced permeability. The method 3.1 Computation – Darcyan and Non-
employed involves an analytical approach. It may Darcyan (Models)
not prove to be as detailed or as thorough as the
method of numerical approximations which, Using the derived/developed correlation below
however, is more complex in nature. for a simplified 2-phase Non-Darcy unsteady state
(from Eq. 30), computations where made to obtain
The approach used to validate this research is as the flow rate of the gas phase, as seen in Table 3.
follows: obtaining accurate field data (preferably a
variety of them), using the obtained data to make Likewise, the established correlation for Darcy
computations with the developing correlations, steady state flow of natural gas is used to predict
produced gas.
using the obtained data to make computations
with established correlations either similar or
closely related in nature, making comparisons with
1424q ZT  re 
obtained results, and making suitable arguments to pe2  pwf
2
  ln  s  (31)
back them up.
kh  rw 

Darcy’s relationship for steady state flow of gas

The field data used is shown in Table 1. Table 2
(Eq. 31) is used to predict this result. Darcy’s model
is obtained from a PVT analysis, these data have
cannot be looked at as an exact comparison to the
aided the analysis of this work.
developed correlation for the 2-phase unsteady
state used in this study. However, considering the

164
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

Table 3. Predicted gas flow rate. point pressure; this is actually the near wellbore
region, as explained in the next sub section. All
P (initial) Pwf Non-Darcyan Darcyan other parameters required to compute flow rate in
psia psia q(MSCF/D) q(MSCF/D) Eqn. 30 are as listed in Tables 1 and 2.
4440 4440 0 0
4440 4415 68.115 51.084 3.2 Comparison of correlation (models)
4440 4295 387.826 649.207
Comparison of the two IPR results can be
4440 3858 1432.871 1196.944 observed in Figure 1. The entire study is centered
4440 3453 2211.488 1756.095 on Regions 1 and 2. Region 1 is characterized by
4440 3048 2766.994 2375.172 condensate saturation above critical condensate
4440 2241 3186.897 3651.326 saturation (CCS) and, hence, both gas and liquid
4440 1433 4261.179 4786.304 phases are mobile. Region 1 is the main source of
4440 1044 4626.343 5163.696 deliverability loss in a gas-condensate well. Gas
relative permeability is reduced drastically in this
region due to condensate buildup. The reduction in
relative permeability to gas reaches its highest
deliverability of gas, this relationship can be used levels in this region. Even though condensate
with the allowance of certain discrepancies. buildup starts from region 2, the liquid phase is
The application of Eq. 30 utilizes PVT data (Table immobile. The two-phase flow in region 1 is the
2) and special core analysis data to carry out the main cause of gas relative permeability reduction.
following comparisons. Corey’s relationship Region 2 is the intermediate zone where
permeability model is used to fit measured core condensate dropout begins and defines a region of
data, ko/kg ratio is obtained from net accumulation of condensate. The condensate
saturation is below the critical value (Scc) and
kg / ko   1  1 (  g / o ) where µg and VroCCE effectively only gas is flowing in this region because
 VroCCE 
oil mobility is reduced or zero.
are extracted from PVT data and VroCCE is a constant
composition expansion oil relative volume. The One can assume that the results obtained from
outer reservoir radius is at the point where the Darcy correlation are applicable to gas
condensate saturation is above critical condensate reservoirs that have very similar conditions to that
saturation. The pressure at this point is the dew

Figure 1. IPR of the developed correlation for Non-Darcyan and Darcyan models.

165
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

of the condensate reservoir. During the early phase From the results obtained, it is safe to say that
of production from region 2, one can observe that production must be carried out at a high flowing
the flow rate shown by the model developed in the pressure (far above the dew point), irrespective of
present study is higher than the one shown by the corresponding flow rate. This is to prevent or
Darcy’s correlation (Figure 1). This can be prolong the formation of condensates that would
attributed to the fact that at this stage, the gas further hinder production.
phase still has considerable mobility and the flow
rate is still higher than that of the liquid phase. The
difference in flow rate takes place because
condensates are still building up. But, as the NOMENCLATURE
production pressure declines, the more the
condensates build and leading to a reduction in the µ = viscosity
gas mobility. This leads to a situation where the µg = gas viscosity
flow rate of the gas shown under the developed µo = oil viscosity
correlation is lower than the flow rate shown by Bo = oil formation volume factor
the Darcy correlation. That configures the CCS = critical condensate saturation
phenomenon observed around region 1. The Cf = formation compressibility
pressure point where the flow rate of gas starts to Cg = gas compressibility
reduce represents the condensate unloading Co = oil compressibility
pressure. The condensate unloading pressure for Ct = total compressibility
the above scenario is about 2,250 psi. h = reservoir height
k = absolute permeability
The present study may contain shortcomings. kg = relative permeability to gas
Among the possible inaccuracies, the major points ko = relative permeability to oil
to consider are: p = pressure
pD = dimensionless pressure
 The developed correlation is modeled with the pi = initial reservoir pressure
assumption that the fluid flow is Darcy in pwD = dimensionless pressure at wellbore
nature, where as in the actual sense, it is a non- pwf = bottom hole flowing pressure
Darcy flow; q = flow rate
 An analytical approach was considered in the r = radius
development of this correlation. This cannot be rD = dimensionless radius
said to be a very detailed analysis because this re = reservoir radius
approach deals with many assumptions and rw = well radius
does not vary sufficient parameters; Sg = gas saturation
 This method does not support a case scenario So = oil saturation
where the composition of the mixture is t = time
constantly changing with time. This can be seen, VroCCE = constant composition expansion oil relative
for example, when gas injection is carried out; volume
 The model strictly considers a two-phase flow tD = dimensionless time
(excluding water), which is observed in very rare λo = oil mobility
cases. A better model would include water λt = total mobility
effects. φ = porosity

4. CONCLUSIONS 5. REFERENCES
This study aimed to developing a method Afidick, D. ; Kaczorowski, N. J.; Srivinas, B.
suitable for analyzing the delivery patterns of gas in Production performance of a retrograde gas
a gas-condensate field using an empirical approach. reservoir: A case study of the Arun field. SPE paper
It provides a simple deduction on a possible 28749-MS, 1994. http://dx.doi.org/10.2118/28749-MS
solution to a gas condensate challenge.

166
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

Ali, J. K.; Mc Gauley, P. J.;Wilson, C. J. Mc Cain Jr, W. D. The Properties of Petroleum

Experimental studies and modeling of gas Fluids, 2nd Ed., PennWell Publishing Company, p.
condensate flow near the wellbore. SPE paper 155, 1989.
39053-MS, 1997. http://dx.doi.org/10.2118/39053-MS
O’Dell, H. G.; Miller, R. N. Successfully cycling a
Barnum, R. S.; Brinkman, F. P.; Richardson, T. low-permeability, high-yield gas condensate
W.; Spillette, A. G. Gas condensate reservoir reservoir. SPE paper 1495-PA, p.41-47, 1967.
behavior: Productivity and recovery reduction due http://dx.doi.org/10.2118/1495-PA
to condensation. SPE paper 30767-MS, 1995.
http://dx.doi.org/10.2118/30767-MS O’Neil, P.V. Advanced Engineering Mathematics.
nd
2 Ed., Wadworth Publishing Co., Belmont,
Bauget, F.; Egermann, P.; Lenormand, R. A new California, 1987.
model to obtain representative field relative
permeability for reservoirs produced under Roebuck Jr., I. F.; Ford, W. T.; Henderson, G. E.;
solution-gas drive. SPE Journal of Reservoir Douglas Jr., J. The compositional reservoir
Evaluation and Engineering, v.8(4), p. 348-356, simulator: case iii – The radial geometry.
2005. http://dx.doi.org/10.2118/84543-PA Unpublished paper, available from Core
Laboratories, Inc., Dallas, Texas, 1969.
Bozorgzadeh, M.; Gringarten, A.C. Estimating
productivity controlling parameters in Whitson, C. H.; Fevang, O.; Saevareid, A. Gas
gas/condensate wells from transient pressure Condensate Relative Permeability for Well
data”, SPE Reservoir Evaluation & Engineering, Calculations. SPE paper 56476, 1999.
v.10, p. 100-111, 2007.
http://dx.doi.org/10.2118/94018-PA

Cable A. S.; Mott R. E.; Mike S. X-Ray in-situ APPENDIX

saturation in gas condensate relative permeability
studies. SCA 2000 - 39 AEA Technology PLC, In determining the gas deliverability in a gas
Winfrith Technology Center, Dorchester, Dorset, condensate reservoir, some equations have to be
DT2 8ZE, UK, 2000. derived, taking into consideration fundamental
principles of oil and gas flow. This is a more
Fevang, O.; Whitson, C. H.; Trondheim, U. complex approach, considering that it involves two
Modeling gas condensate deliverability. SPE paper phases which do not present a constant
30714-PA, SPE Reservoir Engineering, v. 11(4), p. 1- compositional volume throughout the reservoir.
16, 1995.
Starting from the basic diffusivity equation,
Fussell, D. D. Single-well performance for gas which considers the principle of conservation of
condensate reservoirs. Journal of Petroleum mass, an equation of motion, and an equation of
Technology, v. 25(7), p. 860-870, 1973. state, one can attain the diffusivity equation. For a
http://dx.doi.org/10.2118/4072-PA single-phase flow, the diffusivity equation is given
by:
Gringarten, A. C.; Al-Lamki, A.; Daungkaew, S.;
Mott, R.; Whittle, T. M. Well test Analysis in gas 1   p  Ct p
Ccondensate reservoirs. SPE paper 62920, 2000. r   (A1)
r r  r  k t
Hinchman, S. B.; Barree, R. D. Productivity loss
in gas condensate reservoirs. SPE paper 14203-MS, For the multiphase flow, one must consider the
1985. http://dx.doi.org/10.2118/14203-MS continuity equation for each phase. The equation
includes a unit control volume containing oil and
Jamiolahmady, M; Danesh, A.; Tehrani, D. H.; gas saturations So and Sg.
Sohrabi, M. Variations of gas/condensate relative
permeability and production rate at near-wellbore The oil mass balance of a system is represented
conditions: A general correlation. SPE Journal of by:
Reservoir Evaluation and Engineering, v. 9(6), p.
688-697, 2006. http://dx.doi.org/10.2118/83960-PA

167
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

mo formation volume factors and solubility are

M o  (A2) functions of pressure:
t

Radial flow coordinates leads to: 1   p  o Bo   So  p B 

 1 So p 
r      o o   So
r r  r  ko p  Bo  t ko  Bo p t
 p
1 M o m
 r o ; M o  o uo ; mo  o so (A3)
r r t (A9)
1   p  o Bo   So  p B  1 So p
   1  p 
r  
For oil flow in a radial system,    o o   So   
r r  r  ko p  Bo  t ko  Bo p t
 p  Bo  t 

1 o uo  So
 r  os (A4) Let,
r r t Bo

  1 
Bo   
p  Bo 
ko P os ko P
ko P uo o   o 
uo   ; o r Bo o r (A5)
o r So
So 
p
Based on the on Darcy’s equation of motion in a
  1   1 Bo
    Bo    2
control volume for the oil phase, 1

p  Bo  t Bo p
1   1  ko  P  So
    r  (A6)
r r  Bo  o  r t Bo Therefore,

The equation above is simplified by assuming 1   p  o Bo  So So Bo  p

constant oil permeability, viscosity and formation r     2  (A10)
r r  r  ko  Bo Bo  t
volume factor.

1  P  B    So
r  o o  (A7)
r r r  ko  t Bo 1   p  o  So Bo  p
 r     So   (A11)
r r  r  ko  Bo  t
Where mo is mass of oil per volume; mg, mass of
gas per unit volume and ρgs, represents gas density Expanding Eq. A8:
at standard condition; M o , mass flux; ρg, gas
density; ρos, oil density at standard condition; ρg, oil 1   p   g Bg   S g  p
density; u, flow velocity; µo, oil viscosity; µg, gas  r     
viscosity; ko, oil permeability; kg, gas permeability;
r r  r  k g p  Bg  t
Bo, oil formation volume factor; Bg, oil formation
volume factor; So, oil saturation; Sg, gas saturation; 1   p   g Bg  S g S g Bg  p
r     2  (A12)
φ, porosity; p, pressure; r, radius, and t, time. r r  r  kg  B g Bg  t
The continuity equation can be written for each
phase considering that the rate of mass of the Oil and gas mobility are given by,
control volume less the rate of mass out is given by
ko
the following similarity of gas equation: o  (A13)
o
1   p   g Bg   S g 
 r      (A8) kg
r r  r  k g t  Bg  g  (A14)
g
Expanding the partial derivatives, with respect
to time, using chain rule and observing that the Substituting Eq. A13 and A14 into A11 and A12:

168
 BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 6 n. 4 | p. 159-169 | 2012 | ISSN 1982-0593

1   p  1 So Bo  p Defining total compressibility, Ct, and obtaining

 r     So   (A15) expressions for oil and gas compressibility in terms
r r  r  o  Bo  t of formation volume factor:

Ct  SoCo  S g Cg (A22a)

1   p  Bg  S g S g Bg  p
r     2  1
(A16) Co   Bo
r r  r  g  B g Bg  t Bo
(A22b)

Now equating oil and gas relationships to derive 1

a single equation to describe multiphase flow: Cg   Bg (A22c)
Bg
1 So Bo  p Bg  S g S g Bg  p
  So      2  (A17) Substituting Eq. A22b and A22c into Eq. A22a:
o  Bo  t o  B g Bg  t
So Bo S g Bg
Ct   
g  So Bo   S g Bg  Bo Bg
(A23)
 So     S g   (A18)
o  Bo   Bg 
Therefore,
S g  So  1 (A19a)
 So Bo  t 
 So     Ct (A24)
Differentiating Eq. A19a with respect to  Bo  o 
pressure gives:
The above equation, Eq. A24, may be
S g  So  0 (A19b) substituted into Eq. 1, giving the diffusivity
equation to calculate compressibility.
Total mobility in this case is λt
1   p  Ct p
o  g  t (A20) r   (A25)
r r  r  t t
Substituting Eq. A19a and A20 into A24

 So Bo  
 t  o    So Bo S g Bg
 So      1    (A21)
 Bo    o    Bo Bg

169

View publication stats