SPE 91483 Evaluation of a Depleted Gas-Condensate Reservoir for Gas Storage

K. Aminian,SPE, West Virginia University, Alan Brannon, SPE, Dominion Transmission, and S. Ameri,SPE, West Virginia University

Copyright 2004, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2004 SPE Eastern Regional Meeting held in Charleston, West Virginia, U.S.A., 1517 September 2004. This paper was selected for presentation by an SPE Program Committee following review of information contained in a proposal 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 a proposal of not more than 300 words; illustrations may not be copied. The proposal 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 A depleted gas condensate reservoir in the Appalachian Basin was considered for Gas Storage. Upon completion of the primary production, a retrograde condensate is present in the reservoir. When depleted gas condensate reservoirs are used for gas storage, gas injection will pressurize the reservoir and will re-vaporize the condensate. As a result, the produced gas during withdrawal cycle will contain heavy hydrocarbons which need to be removed by processing units so the gas does not form liquid phases by retrograde condensation in pipeline. Peng-Robinson Equation of State (PR-EOS) was utilized to predict the liquid yield during storage cycle. The first step in applying PR-EOS was to simulate the primary depletion and check against the reported primary condensate yields. Upon successful history matching the primary production, the remaining retrograde and residue gas were mixed with pipeline gas to pressurize the reservoir to its original pressure. The fluids were assumed to be completely mixed. Since the volume percent liquid in the reservoir at the end of primary depletion is 5-10%, gravity drainage of the hydrocarbon liquid may be minimized. However, it is possible that the liquid phase can potentially be wetting the pore walls. This may interfere with the mixing and revaporization process. PR-EOS was then used to predict the liquid yield for several separator pressures. In addition, several cases without complete mixing were considered to study the impact of mixing on liquid yield. The results were then utilized for design of surface facilities. Introduction Depleted gas or oil reservoirs after primary production are used for gas storage. After refurbishing the facilities, natural gas is injected when demand for natural gas is low (summer time) and withdrawn when demand for natural gas is high (cold winter months). For dry gas fields, the withdrawn gas is

normally dehydrated and sent to distribution systems. When depleted gas-condensate reservoirs are used for storage, the withdrawn gas will have a different composition than injected gas. This is due to the fact that retrograde liquid condenses out in the gas-condensate reservoir during the depletion process. This liquid is non-recoverable due to low liquid saturation in the reservoir. To prevent loss of retrograde liquid in gas-condensate reservoirs, gas cycling may be employed to displace rich gas phase, strip the liquids followed by recompression for gas injection in the reservoir to maintain its pressure. Alternatively, a depleted gas-condensate can be pressurized by gas injection to re-vaporize the retrograde liquid. Niether of these approches are usually economically feasible due to high cost of gas and compression reqirements. Therefore, most depleted gas-condensate reservoirs contain retrograde liquid at the conclusion of the primary production. Consequently, the gas withdrawn from a gas-condensate storage reservoir contains heavy hydrocarbons as a results of retrograde liquid evaporation. To prevent retrograde condensation in the gathering and tranmission distribution systems, the withdrawn gas must be processed to remove these heavy hydrocarbons. The proper design of surface and processing facilities is essential in reliable operation of the gas-condensate storage fields. Significant volume of liquid can be produced from the withdrawn gas, even with relatively low liquid yield, as results of high gas rates associated with storage operation. Background The volumetric performance of gas condensate reservoirs is is an essential requirement for optimum design and operation of equipment involved in production, transportation, and processing facilities. The performance of gas-condensate reservoir is strongly composition-dependent therefore, it is necessary to predict reservoir fluid compositional changes under varying pressures and depletion processes. Normally, a PVT study is performed on the reservoir fluid to define the fluid properties at reservoir conditions and at surface separation conditions. For gas condensate, laboratory studies generally conducted are Conatant Composition Expansion (CCE) and Constant Volume Depletion (CVD) tests. CVD tests are considered to be representative of the primary depletion and retrograde condensation behavior of gas condensate reservoirs. Data generally obtained from this test are dew pint pressure, gas produced, liquid dropout and gas deviation factor at various pre-determined pressure steps ranging from discovery pressure to separator pressure.

www.petroman.ir

SPE 91483

Detailed PVT studies are costly, tedious and time consuming. Consequently, an Equation of State (EOS) is used to match the reported experimental data. Once matched, the EOS can be used to predict the fluid properties over a wide range of pressure and temperature conditions. The EOS approach is most commonly used for natural gas systems due to its applicability at high pressures for both liquid and vapor phases. Probably the most successful EOS for natural gas property calculation is the one proposed by Peng and Robinson (PR) 1. The primary input data to the EOS is the composition of the reservoir fluid in terms of mole percent. It naturally follows that if the input data set is incomplete or inaccurate, the subsequent matching with EOS will be problematical. The PVT study includes the compositional analysis of the reservoir fluid in terms of mole percent. The compositional analysis usually ends up with the undefined plus fractions, commonly known as the C7+ fractions, which contain an indefinite number of components with a carbon number higher than six. The molecular weight and specific gravity of the plus fraction may be the only measured data available. A number of papers2-5 report comparison of PR-EOS and laboratory PVT results for gas condensate. Most of these reports emphasize the C7+ charcterization as the key element in attaining agreement between EOS and laboratory results. Numerous authors have published papers about how to characterize the plus fraction.6-9 A general procedure consists of splitting the plus fraction into certain carbon numbers and regrouping the split fractions into certain pseudocomponents. Methodology A depleted gas condensate reservoir in the Appalachian Basin was considered for Gas Storage. This reef reservoir is an excellent candidate for storage due to high deliverability and compactness. The reservoir is near depletion and retrograde liquid which has condensed out during primary production is believed to be present in the reservoir. One of the key parameters for storage facility design and operation is the liquid yield during the storage withdrwal cycles. A PR-EOS phase behavior package10 was utilized to simulate retrograde condensation in the reservoir and to predict the liquid yield during storage withdrwal cycles. The following steps were followed in this study: 1. Comparison with Laboratory PVT Stuty The results of a laboratory PVT study which had been conducted on a re-constituted original reservoir fluid were available. The laboratory measurements provided dew point pressure, gas produced, liquid dropout, gas deviation factor, and compositional information at various pre-determined pressure steps. However, the extended compositional analysis of the original reservoir fluid was not available. A method proposed by Ahmed et al7 for characterizing C7+ was utilized to estimate extended compositional analysis for the original reservoir fluid. The extended compositional analysis was then used as input for PR-EOS phase behavior package to simulate the laboratory study.

2.

Comparison with Primary Production History Data on gas and liquid production at various measured reservoir pressures was available. The dry gas and condensate production data were utilized to determine cumulative wet gas production and liquid yield at various reservoir pressures. The measured reservoir pressures were used in conjuction with PREOS phase behavior package to simulate cumulative wet gas production and liquid yield at various pressures. The results were then used for comparison against field data. Material Balance Study A material balance study was conducted using 2phase deviation factors and cumulative wet gas production to determine the Initial Gas-in-Place (IGIP). The results were then utilized to estimate residue gas and liquid quantities at the conversion pressure. Simulation of Storage In order to predict the composition of the reservoir fluid during storage cycles, the residue gas and liquid reservoir pressure were then completely mixed with pipeline gas in a proportion to return the reservoir pressure to discovery pressure. Finally, the performance of the reservoir for 3 cycles of storage operations was simulated assuming 50% of gas in reservoir as top gas.

3.

4.

Results and Discussion To evaluate the accuracy of the extended compositional analysis, the predicted retrograde dewpoint pressure was compared against the laboratory measured value. The initial extended compositional analysis did not generate a close match. Therefore, extended compositional analysis was modified to obtain a close match with laboratory measured retrograde dewpoint pressure. Upon obtaining a close match with dew point pressure, a number of indicators were utilized for comparison against laboratory results. They included cumulative gas production, produced gas gravity, molecular weight of C7+ in the produced gas, 2-phase z-factors, and the volume percent of condensate in the reservoir at various reservoir pressures. Figures 1 illustrates the various comparisons. The results of simulation studies do not completely agree with the laboratory results but they were within reasonable ranges. The key indicator for purposes of storage performance simulations is the volume percent of condensate in the reservoir. It is important to note that even though the predicted liquid volume percents are higher than those measured in the laboratory study, the volume percent of liquid in the reservoir at time of conversion to storage is relatively close to the laboratory measured value. This is because the simulated results indicate more liquid evaporation at lower pressures (below 500 psig). Figure 2 illustrates the comparison of predicted and field data on average liquid yields. Again, they predicted values are within reasonable ranges. It should be noted that accurate separator conditions were not available and they were

www.petroman.ir

SPE 91483

estimated. These assumptions obviously influence the results of the predictions. Thie similarity between the predicted and measured field liquid yields further confirms that PR-EOS phase behavior package can provide reliable results for predicting retrograde condensation in the reservoir. The production history and predicted 2-phase z-factors by the simulator as well as two phase z-factors from laboratory study were utilized to generate material balance plots (P/z vs. Cumulative Wet Gas Production). Figure 3 illustrates this plot for both sets of 2-phase z-factors. Same values for IGIP were obtained using both sets of 2-phase z-factors. It should be noted that even though the two set of z-factors were different, their similar trend versus pressure resulted in the same values of IGIP. This is mailnly due to relatively small liquid volumes in the reservoir. Figure 4 summarizes the predicted liquid yields for 3 storage cycles. In prediction of liquid yields, it was assumed the injected gas was completely mixed with the residue gas and liquid. The full mixing assumption is based on the fact the the reservoir is highly porous and permeable and the liquid is completely immobile due to low saturation. The reevaporation of the liquid is mainly caused by pressure increase. The injection well configurations are also such that the injected gas will contact the entire reservoir and therefore the risdue gas and injected gas will be fully mixed. However, it is important to evaluate the results of first storage withdrawal cycle to verify the degree of mixing and accuracy of the predicted liquid yields. Conclusions The following conclusions were reached in this study: 1. PR-EOS can be used to predict the phase behavior of gas condensate reservoir accurately. Characterization of C7+ is critical in obtaining reliable predictions by PR-EOS.

Referneces 1. Peng, D.Y.and Robinson, D. B.: A New TwoConstant Equation of State, Ind. Eng. Chem. Fundamentals (1976), 15, No. 1, 54. 2. Katz D., Firoozabadi A.: Predicting Phase Behavior of Condensate/Crude-oil Systems Using Methane Interaction Coefficients, JPT, Nov. 1978, 1949. Coats, K.H.:Simulation of Gas Condensate Reservoir Performance, JPT, October 1985, 1870. Whitson, C.H. and Torp, S.B.: Evaluating ConstantVolume Depletion Data, JPT, March 1983, 610. Kenyon, D.E. and Behie, G.A.: Third SPE Comparative Solution Project: Gas Cycling of Retrograde Condensate Reservoirs, JPT, Dec. 1987. Whitson, C.: Characterizing Hydrocarbon Plus Fractions, SPEJ, 23, 1983, 683. Ahmed, T., Cady, G., and Story, A.: A Generalized Correlation for Characterizing the Hydrocarbon Heavy Fractions, paper SPE 14266 presented at the 1985 SPE Annual Technical Conference and Exhibition held in Las Vegas, September 22-25, 1985. Montel, F. and Gouel, P.L.: A New Lumping Scheme of Analytical Data for Composition Studies, paper SPE 13119 presented at the 1984 SPE Annual Technical Conference and Exhibition held in Houston, September 16-19. Coats, K.H. and Smart, G.T.: Application of a Regression-Based EOS PVT Program to Laboratory Data, SPE Reservoir Engineering, May 1986, 277. Aminian, K.: Phase-Equilibrium and Reservoir Depletion Calculations on PC Using the PengRobinson Equation of State, SPE Computer Applications, May-June 1989, 20.

3.

4.

5.

6.

7.

8.

9.

2.

10. 3. The Initial Gas in Place (IGIP) is not sensitive to two-phase z-factors used in material balance equation. PR-EOS can be used to predict the phase behavior of a condensate reservoir after conversion to storage if level of mixing can be estimated.

4.

5.

The prediction of liquid yields during storage cycles is critical for design of surface and processing facilities.

www.petroman.ir

SPE 91483

100

0.85 0.84

90 0.83 80

Laboratory Results

0.82 0.81

70 0.8 % Gas In Place Produced 60 Produced Gas Gravity 0.79 0.78 0.77 0.76 0.75 30 0.74 20 0.73 0.72 10 0.71 0 0 300 600 900 1200 1500 Pressure 1800 2100 2400 2700 3000 0.7 0 500 1000

Laboratory Results

50

Simulation Results

40

Simulation Results

1500 Pressure

2000

2500

3000

135

0.95 130 0.9

Laboratory Results
125

0.85

Laboratory Results
2-Phase Z-Factor 0.8 M.W. of C7+ 120

0.75

115

0.7

0.65

110

Simulation Results
0.6 105 0.55

Simulation Results

0.5 0 500 1000 1500 Pressure 2000 2500 3000

100 0 300 600 900 1200 1500 Pressure 1800 2100 2400 2700 3000

Simulation Results
6

Vol % o Retrograde Liquid in Reservoir

Laboratory Results
1

0 0 500 1000 1500 Pressure 2000 2500 3000

Figure 1: Comparison of EOS Predictions and Laboratory PVT Study Results

www.petroman.ir

SPE 91483

60

50

40

Yield, Bbl/MMCF

Simulation Results
30

20

Field Data

10

0 500

1000

1500

2000

2500

Pressure

Figure 2: Comparison of EOS-Predicted and Field Liquid Yields

4000

EOS 2-Phase z-Factors

P/z-2 Phase

2000

1000

Laboratory 2-Phase z-Factors

0 0 2 4 6 8

Cumulative Wet Gas Production, BCF

Figure 3: Material Balance Plot

www.petroman.ir

SPE 91483

16.00 14.00 12.00

First Cycle

Second Cycle

Bbls/MMCF

10.00 8.00 6.00 4.00 2.00 0.00 1400

1600

1800

2000

2200

2400

2600

2800

Reservoir Pressure, Psia

Figure 4: Predicted Storage Liquid Yields

www.petroman.ir