PARTICLE SWARM OPTIMIZATION OF SOLUTION GAS OIL RATIO

Abstract
Reservoir fluid properties are very important in reservoir engineering computations such as material balance calculation, well test analysis, reserves estimate and numerical reservoir simulations. Ideally, these properties should be obtained from laboratory pressure-volume-temperature (PVT) analysis. Quite often, however, these measurements are either not available, or very costly to obtain. In such cases, empirically derived correlations are used to estimate the needed properties, all computation therefore, will depend on the accuracy of the correlations used for estimating the fluid properties. Hence in this study, Standing’s correlation for estimating the solution gas-oil ratio was optimized using a particle swarm optimization (PSO) algorithm to minimize the error associated in estimating solution gas-oil ratio from correlation at various depletion pressure. The optimized correlation was taken as a function of bubble point pressure, API gravity, gas gravity and reservoir temperature. PVT data from differential liberation test was used to validate this study’s correlation and the result obtained shows that the optimized correlation for this study matches closely with the experimental values, also the newly optimized correlation was validated with other models and the results gave the least average relative error of 3.34 and a correlation coefficient of 0.998 after 216th successive iterations by the particle swarm optimization algorithm. Okotie, S. | Department of Petroleum Engineering, Federal University of Petroleum Resources (FUPRE), Effurun, Delta State, Nigeria