PVT MODELLING FOR A GAS CONDENSATE RESERVOIR AND

DETERMINATION OF CONDENSATE LOSSES

E. PIRAMOON
NATIONAL IRANIAN SOUTH OILFIELDS CO. (NISOC), PETROLEUM ENGINEERING DEPARTMENT

ABSTRACT
The exact determination of the precise reservoir fluid properties (PVT) in addition to the
positive role in evaluation of correct performance and volume of fluid in place calculation for
any reservoir, has got a doubly important deciding value for gas condensate reservoirs. This
is particularly important because of the liquid drop out phenomenon due to reservoir
pressure drop as a result of fluid production. These liquid drop outs are immobile and
irreducible due to the low degree of saturation and are considered as losses.
Among the other factors which is important in gas condensate reservoirs and is strongly
dependent on fluid properties, is gas recycling to prevent condensate losses. Gas recycling
stops or reduces the reservoir pressure drop which increases the condensate recovery.
Therefore without a thorough knowledge of reservoir PVT fluid properties, it will not be
possible to assess the degree of liquid drop out or to design recycling projects.
Presently in many countries of the world specially Iran, giant gas condensate reservoirs with
average to high liquid-gas ratios exist, which primarily due to natural depletion, have had
large liquid drop outs and losses (sometimes up to 30 mstb/d).
Proper reservoir fluid properties modelling, provides a better fluid characterization in
reservoir simulation.

INTRODUCTION
The reservoir in question has a giant dome gas which possesses retrograde behavior and a thin
oil bearing column. Several gas samples at different wells and times, before and after the start
of production, have been collected and complete PVT analysis on some of them performed.
In this survey all the experimental tests were analyzed with the use of a phase behavior
package. Proper oil and gas samples were selected and analyzed with an integrated PR
EOS modelling. Finally a single model to characterize both dome gas and oil column
was generated. The results of this modelling can be applied in compositional and
black oil reservoir simulators. Utilization of a reservoir fluid model, past and future
performance of reservoir will quantify the condensate losses.

EXPERIMENTAL PROPERTIES OF RESERVOIR FLUIDS

Totally seven gas samples and three oil samples were analyzed with almost complete tests for
some samples, and thus reservoir fluid properties were characterized. Table-1 indicates the
reservoir dome gas composition utilizing all seven samples.Table-2 indicates the reservoir oil
column composition.
As shown in Table-1, the C1 mole percent in dome gas samples varies from 80.96 to
83.89 and that of C5+ (important in liquid drop out) varies from 3.89 to 5.24.
Based on experimental measurements the volume of liquid drop out of dome gas as a
percentage of hydrocarbon pore volume for G2, G4, G6 and G7 samples are shown in Fig.-1.
The high liquid drop out in sample G2 is due to deep producing interval and variation of
dome gas composition with depth which will be covered later. The amount of experimental
producible condensate for C3+, C4+, C5+ cuts of dome gas samples (G4, G6 and G7) are
illustrated in Fig.-2.
Complete experimental analyses were performed on oil samples. The temperature for O3
sample was 208 F, which was different from the other two samples, temperature (180 F).
Figures 3 to 5 illustrate experimental values of Bo, Rs and Do for oil samples.

VARIATION OF COMPOSITION AND DEW POINT PRESSURE OF DOME

GAS WITH DEPTH
Dew point pressure measured in all dome gas samples from different producing intervals are
equal to reservoir static pressure at sampling depths. This indicates that the reservoir dome
gas at initial conditions at each depth has been saturated which entails compositional variation
of dome gas with depth. The existence of a thin oil column at lower depths of the reservoir is
another proof of the saturation state of initial dome gas. Table-4 indicates the dome gas
composition at three depths, i.e., near the reservoir crest, near the initial GOC and at datum
depth (2/3 of gas column thickness).

MODELLING OF RESERVOIR FLUID PROPERTIES

At first, dome gas and oil samples were analyzed separately utilizing a phase behavior
package and the proper samples were selected according to conformance between
computational and experimental data and validity of experiments.
Proper dome gas and oil column samples were analyzed simultaneously under different
scenarios and finally the most proper dome gas sample, G4, and oil column sample, O2,
resulted in the best integrated unique EOS. In the integrated modelling of reservoir oil and
dome gas the heavy unknown components (C7+) were split into three sub-groups, the first
two lighter sub-groups were jointly considered for dome gas and oil column fluids. The
third heavy sub-group was assigned only for oil.
The properties of these three sub-groups were determined with regression operations using
PR EOS on experimental data of the selected oil and gas samples in an integrated analysis.
The properties which were utilized as variable parameters for the proper matching of
experimental and modelling results are as follows; critical temperature and pressure, volume
shift, acentric factors, BIC of the heavy unknown components and experimental data weight
factors. Table-3 illustrates the finalized properties of unknown components. In PVT
modelling the normal and iso butane and pentane components were lumped.

THE COMPARISON OF MODELLING AND EXPERIMENTAL RESULTS

Figures 6 and 7 illustrate the modelling and experimental results for some of the dome gas
properties. Figures 8 and 9 also illustrate the modelling and experimental result of important
oil properties. As observed, there is a close matching of experimental and modelling results.

2
DETERMINATION OF CONDENSATE LOSSES
As stated before at initial conditions, the reservoir dome gas was at dew point pressure and
the start of production was simultaneous with liquid drop out in dome gas with the resultant
losses. Using the fluid properties model, the condensate drop out of the associated produced
gas was calculated utilizing CVD test data . The produced gas in CVD experiments at
different pressures, was separated under the separator prevailing conditions and then the
ratio of liquid to gas (LGR) was determined. Fig.-10 illustrates the amount of LGR as a
percentage of initial condensate in place (OCIP) with respect to pressure drop.
Reservoir performance history indicates a linear variation of dome gas pressure in terms of
cumulative gas production and prediction of reservoir pressure was accordingly performed.
Fig.-11 shows the reservoir performance with coordinates indicating pressure drop with
respect to percentage of dome gas cumulative production (%OGIP). Utilizing Fig.-10 and 11
the variation of LGR with respect to cumulative gas production is obtained, (Fig.-12).
The area under the curve in Fig.-12 indicates the amount of producible condensate (Cp) at
each pressure stage (Fig.-12 must be converted to LGR vs. Gp). Condensate losses are
calculated utilizing the following equation:

Cond. Loss = (LGRi*Gpc-Cpc) + (LGRi-LGRc)*Grc

Cond. Loss: stb

Gpc : current cumulative gas production, mmscf
LGRi : initial liquid to gas ratio, stb/mmscf
Cpc : current cumulative condensate production, stb
LGRc: current liquid to gas ratio, stb/mmscf
Grc : current remaining gas in place, mmscf
Condensate losses as a percentage of initial condensate in place is plotted in Fig.-13 with
respect to percentage cumulative gas production.

CONCLUSION
1- Gas condensate reservoirs are very sensitive to fluid characteristics due to liquid drop out
phenomenon and gas recycling projects. Therefore PVT modelling should be carefully
performed.
2- PVT modelling for gas condensate reservoirs having an oil bearing column, should be
performed in an integrated manner.
3- PVT fluid modelling and analysis of reservoir performance history of a gas condensate
reservoir, will result in precise calculation of condensate losses.

NOMENCLATURE
Bo = oil formation volume factor
Rs = solution gas to oil ratio
Do = oil density
GOC = gas oil contact level
LGR = liquid -gas ratio
OGIP = original gas in place
OCIP = original condensate in place

3
 REFERENCES
1 - Whitson, C. H ; "Effect of physical properties estimation on equation of state
prediction, SPE paper 11200, 1982.

Table - 1 : Composition of Dome Gas Wells Samples (%mole)

Comp. / Wells G1 G2 G3 G4 G5 G6 G7
C1 83.89 80.96 82.98 82.1 83.113 83.12 83.02
C2 4.72 6.04 5.93 5.78 5.994 5.98 5.97
C3 2.92 3.12 2.88 2.87 2.871 2.77 2.93
IC4 0.59 0.54 0.54 0.56 0.516 0.47 0.51
NC4 1.2 1.78 1.2 1.23 1.182 1.04 1.12
IC5 0.52 0.67 0.47 0.52 0.456 0.37 0.45
NC5 0.61 0.59 0.53 0.6 0.54 0.44 0.53
C6 0.72 0.81 0.72 0.72 0.612 0.63 0.67
C7+ 2.68 3.17 2.44 3.1 2.37 2.73 2.2
CO2 2.14 2.32 2.21 2.44 2.206 2.22 2.45
N2 0 0 0.1 0.08 0.14 0.14 0.15
TOTAL 100 100 100 100 100 100 100
C7+, Mw 137 141 135 132 134 131 127.3
C7+, SG 0.782 0.797 0.78 0.773 0.779 0.772 0.779
Sampling Date B.P. B.P. B.P. B.P. B.P. B.P. A.P.

Table -2 : Composition of Oil Wells Samples

10
Comp. / Wells O1 O2 O3
G7
C1 56.54 53.23 52.509 8 G6
Liquid Drop (% HPV)

C2 6.67 6.45 6.242 G4

G2
C3 4.1 3.63 4.237
6
IC4 0.8 0.89 0.885
NC4 2.14 2.3 2.213
0.94 0.98 1.124
4
IC5
NC5 1.16 1.26 1.271
C6 1.71 2.28 2.289 2
C7+ 25.85 28.98 27.558
CO2 0 0 1.671
0
TOTAL 100 100 100
0 1000 2000 3000 4000 5000
C7+, Mw 236 240 236 Press. (Psia)
C7+, SG 0.8655 0.8731 0.8646
Fig. -1 :Experimental Liquid Drop for Some
Sampling Date B.P. B.P. B.P.
of Gas Samples at T=180 F

5 1.8

C3+
G4
L i q u i d C o n te n t (G P M )

4 C4+
C5+ G6 1.6

G7G4
O i l FV F (rb/s tb)

3
G6 1.4
G7
2 G4
G6 1.2
1 G7 O1 @ 185 F
O2 @ 185 F
1 O3 @ 208 F
0
0 1000 2000 3000 4000 5000
Press. (psia) 0.8
0 1000 2000 3000 4000 5000 6000
Fig. -2 :Experimental Liquid Content for Press. (Psia)
Some of Gas Samples at T=180 F 4 Fig. -3 :Experimental Oil FVF for Oil Samples
 Table - 3 : Splitting Components Properties

User Comp. Tc (R) Pc (psia) Vc (cuft/lbmole) Mw T boil (R) %mole in Gas %mole in Oil

GM1 1044.5 381.1 7.96 115.69 717.2 2.8837 10.396

GM2 1396.7 282.9 12.74 208.72 985.2 0.2163 10.426
GO1 1480.3 101.7 28.68 438.41 1458.4 0 8.157

1500 0.9

1200 O1 @ 185 F
0.82 O2 @ 185 F

O il De nsity (lb/cuft)
O3 @ 208 F
Rs (scf/stb)

900
0.74

600
O1 @ 185 F 0.66
O2 @ 185 F
300 O3 @ 208 F
0.58

0
0 1000 2000 3000 4000 5000 6000 0.5
Press. (Psia) 0 1000 2000 3000 4000 5000 6000
6 100 Press. (Psia)
Fig. -4 :Experimental Solution Gas Oil Ratio for
Oil Samples Model
Fig. -5 :Experimental Oil Density for Oil Samples
5 Model
C u m . G a s P r o d .( % IG i n

80
Li qu i d Drop (%H PV)

Test Test
4
60
ce l l )

3
40
2
20
1

0
0
0 1000 2000 3000 4000 5000
0 1000 2000 3000 4000 5000
1.8 1200 Press. (Psia)
Press. (Psia)
Fig. -6 :Comparison of Liquid Drop at T=180 F 1000
Fig. -7 :Comparison of Cumulative Gas Prod. at T=180 F
1.6
O i l FVF (rb/stb)

800
Rs (scf/stb)

1.4
600
Model
1.2
400 Test
Model
Test
1 Table - 4 : Composition of Dome Gas at Three Depths with oil
200 Composition after Modelling

Components 0
Dome Gas (%mole) Oil Column
0.8
0 1000 2000 3000 4000 5000 6000
0 1000 2000 3000 4000 5000 6000
Near GOC Datum Near Crest Pre ss. (Psia)
(%mole)
Press. (Psia)
CO2 2.3981 2.433 2.4409 Fig. - 9 :Comparison
0 of Solution Gas Ratio at
Fig.
C1
-8 :Comparison of80.65
Oil FVF at T=185 F81.933 82.3161 T=185 F
53.23
C2 5.7169 5.7691 5.779 6.44
C3 2.9174 2.8777 2.8653 3.63
C4 1.9213 1.8121 1.7834 3.19
C5 1.3038 1.1511 1.1133 2.24

5
 C6 0.9086 0.752 0.7142 2.28
GM1 3.9039 3.053 2.8222 10.396
GM2 0.2806 0.2195 0.1656 10.426
GO1 0 0 0 8.157
TOTAL 100 100 100 100

40 3000

2500
LGRi -LG Rc (S TB /MMS C F)

32

2000

Press. Drop (psi)

24

1500
16
1000

8
500

0 0
0 600 1200 1800 2400 3000 0 20 40 60 80 100
Press. Drop (Psi) Gas Prod. (%O GIP)
50
40 Fig.-10 :Liquid Gas Ratio Drop Vs. Pressure Drop C on de n s ate Los s (%O C IP) Fig. -11 :Pressure Drop Vs. Cumulative Gas Prod.

40
LG Ri -LG Rc (S TB /MMS C F)

32

30
24

20
16

10
8

0 0 20 40 60 80 100

0 20 40 60 80 100
Gas Production (%O GIP)
Gas Cum. Prod. (%O GIP) Fig. -13 :Condensate Loss Vs. Cumulative Gas
Fig. -12 :Liquid Gas Ratio Drop vs. Pressure Drop Production