Determination of Recovery and Relative Permeability for Gas Condensate Reservoirs

Abstract

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractCoreflood experiments on gas condensate flow behavior were conducted for a Libyan gas condensate reservoir. The objectives were to investigate the effects of rock and fluid characteristics on critical condensate saturation (CCS), gas and condensate relative permeability's, hydrocarbon recovery and trapping by water injection, and incremental recovery by subsequent blowdown and vaporization by dry gas injection.The water/gas relative permeability data were generated using implicit historical matching simulator which uses a reverse history matching technique to generate a full suite of relative permeability curves over the range of interest. The results of the tests on the reduction in gas permeability due to retrograde condensate accumulation demonstrated that, in general, the effective permeability to gas decreased significantly at pressures below the dew point pressure due to the condensate trapping effect for all the stacks investigated in this study. Permeability continues to drop at very low pressures even though theoretically condensate should be revaporizing. The recoveries of the liquid condensate during the vaporization by dry gas injection at pressures below the dew point pressure were generally high. The recovery ranged from 66% to 70% for the high permeability core stack and 86% to 98% for the lower permeability core stack. In theory, it is possible for all of the condensate to be revaporized into the methane gas stream. However, in practice, the recovery is usually less than 100% due to macroscopic sweep efficiency limitations and mass transfer limitations within the rock matrix.Poor mobility ratio and viscous fingering during the methane injection characterize the re-vaporization of gas condensate by methane gas. This results in early breakthrough of the methane gas and very high gas/liquid ratios during the test. The volume of methane gas injection required to recover a significant amount of the condensate liquid in the core is a function of temperature, pressure, composition of the gas, heterogeneity and especially the permeability of the core sample.

Prediction of Condensate-to-Gas Ratio for Retrograde Gas Condensate Reservoirs Using Artificial Neural Network with Particle Swarm Optimization

“…An increase in the molecular weight leads to an increase in the CGR value. This finding is in agreement with several previously published research. − ,− In addition, impact of dewpoint pressure on the CGR supports the validity of the empirical correlations available in the literature. − ,− Increase in dewpoint pressure results in an increase in the value of CGR. It should be also noted here that as the reservoir temperature increases, the magnitude of the CGR ...…”

“…In addition to the data generated from the experimental part, some data from the literature were also used in developing the PSO-ANN model for CGR prediction. A part of the data used in this study is presented in Appendix A, and the rest can be tracked in the open literature. ,,,,− ,− …”

“…As previously mentioned, experimental works, numerical modeling, empirical equations, and PVT studies of gas condensate reservoirs have proved that CGR is a function of temperature, pressure, and composition of fluid mixture in the reservoir. To take into account the effect of mixture composition, some of the published research studies used molecular weight of C 7+ . − In addition, there are some studies in the open literature that considered molecular weight of C 30+ in their computations. − ,− In order to select the best inputs for the ANN method, the following sets of input variables were examined as shown in Table : 1) T , P d , M w C30+ ; 2) T , P d , M w C7+ ; 3) T , P d , mixture composition, and 4) T , P d , M w of the mixture.…”

“…A set of 295 experimental data was collected to develop and examine the proposed PSO-ANN model. The data series includes 45 data from South Pars gas-condensate reservoir and 250 data from the literature. ,,,,− ,− The ranges of the parameters used to develop the PSO-ANN are given in Table .…”