The Effect of Phase Data on Liquids Recovery During Cycling of a Gas Condensate Reservoir
- Del D. Fussell
- Document ID
- Society of Petroleum Engineers
- Society of Petroleum Engineers Journal
- Publication Date
- April 1972
- Document Type
- Journal Paper
- 96 - 102
- 1972. Society of Petroleum Engineers
- 5.4.3 Gas Cycling, 5.2.1 Phase Behavior and PVT Measurements, 4.6 Natural Gas, 5.3.2 Multiphase Flow, 4.1.5 Processing Equipment, 5.2 Reservoir Fluid Dynamics, 5.8.8 Gas-condensate reservoirs, 5.5 Reservoir Simulation, 5.1 Reservoir Characterisation, 5.2.2 Fluid Modeling, Equations of State
- 0 in the last 30 days
- 213 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
A linear, one-dimensional compositional model was employed to study the effect of phase data on the recovery of the pentanes and heavier fraction during gas cycling of a sour gas condensate reservoir. The results show that the phase data used and the method of handling the effects of composition on phase data can affect the computed recovery performance.
The prediction of performance for reservoirs containing volatile fluids historically has been very difficult. The development of compositional reservoir simulators has given engineers a new tool to attack this problem. Several recent papers, have discussed the application papers, have discussed the application of this type of simulator to the prediction of reservoir performance.
Huang showed that the predicted reservoir performance for a three-component system (methane, performance for a three-component system (methane, n-butane and n-decane) was not sensitive to small errors in phase viscosity and density data, but was sensitive to the accuracy of the methane K-value data. He recommended developing phase data based on an experimental depletion curve for the original fluid system.
Experimental data for the original reservoir fluid system would be expected to be the minimum information required for reservoir simulation during gas cycling. Large composition changes occur in a reservoir during injection of a dry gas (usually plant residue gas) with a greatly different composition than the reservoir gas phase. The predicted performance, then, is expected to be sensitive to performance, then, is expected to be sensitive to the method selected for describing the phase equilibria of the volatile fluid, particularly for pressures just below the saturation pressure. pressures just below the saturation pressure. In this study, gas cycling calculations were performed on a gas-condensate reservoir where performed on a gas-condensate reservoir where experimental phase behavior data were only available for the original reservoir fluid. The objectives are to show:
1. that differences in laboratory-determined depletion curves and differences in techniques employed to calculate compositional dependence of K-values based upon these curves will affect the predicted recovery of the pentanes and heavier predicted recovery of the pentanes and heavier components (C5+) during gas cycling of a gascondensate reservoir;
2. why experimental data necessary to obtain the compositional dependence of the K-values should be obtained; and
3. the use of the Redlich-Kwong equation of state to calculate necessary fluid properties as required by the compositional simulator.
DEVELOPMENT OF PHASE DATA
The predicted simulation for cycling of a gascondensate or a volatile-oil reservoir should be based upon experimental phase data that include compositional effects upon K-values. The procedure for obtaining the necessary data has been outlined by Jacoby and Yarborough. This procedure can be used to obtain data for three or four mixtures of different compositions. These mixture compositions and volumetric data can be used to test available correlations or to generate K-values for each of the mixtures as described by Jacoby and Yarborough. If K-values are generated for the different mixtures, a correlation that uses convergence pressure or another parameter can be developed to handle the compositional dependence. Sufficient data on mixtures of cycling gas and retrograde condensate were not available to generate K-values, which is often the case for a reservoir study, but the available data and the techniques utilized to obtain K-values for the compositional model are discussed.
Experimental measurements were made for three different samples of the gas-condensate reservoir fluid used in this study. Data on two of the samples were obtained at Laboratory A, while data on the third sample were measured at Laboratory B.
|File Size||653 KB||Number of Pages||7|