Thesis (M.Eng)-- Chulalongkorn University, 2008

Permeability which has the highest level of heterogeneity is one of the most important parameters which affect reservoir performance such as production profile and recovery processes. To study such effect, reservoirs with different degrees of reservoir heterogeneity were created and studied. In this thesis, Sequential Gaussian Simulation (SGS) is used to generate maps of the reservoir. Moreover, sensitivity analysis of spatial continuity and SGS is performed to assess uncertainties by varying range, nugget and random seed. When the reservoir model is generated, Dykstra-Parsons coefficient (VDP) is computed to measure the degree of heterogeneity. Then, reservoir simulation is performed to study the effect of heterogeneity on reservoir performance for reservoirs with different permeability distributions that have been generated. The effect of heterogeneity on performance prediction for solution gas drive reservoir is quantified. The uncertainties associated with the results obtained from the heterogeneity are assessed. The oil recovery factors (RF) at abandonment are compared. As VDP increases, RF slightly decreases. When considered the time to abandonment, reservoir with the highest VDP takes the longest time to produce oil. In general, the more heterogeneity, the longer time it takes to recover the fluid.