Non-Isothermal Compositional Simulation Study for Determining an Optimum EOR Strategy for a Middle-East Offshore Heavy-Oil Reservoir with Compositional Variations with Depth

Abstract

The objective of this study was to find an optimum EOR strategy for a Middle-East offshore heavy-oil reservoir that exhibits reservoir-fluids-properties variations with depth using non-isothermal compositional simulations that honor the fluids-compositional variations with depth. The observed compositional variations are such that the oil density changes from 20 °API in the crest to 11 °API in the deep part and the live-oil viscosity increases from 14 cP in the crest part to 449 cP in the deep part of the reservoir. Because the concept of thermodynamic equilibrium is not valid for the reservoir with compositional variations, we used the theory of irreversible thermodynamics to develop a compositional PVT model that captures the observed compositional and oil-properties variations with depth. Next, the PVT model was tuned against CO2 and hydrocarbon-gas swelling and MMP tests. Subsequently, we developed a compositional reservoir dynamic model that uses the single compositional PVT model and can simulate the degree of CO2/hydrocarbon-gas miscibility in oil. Then, we performed a dynamic IOR/EOR screening that includes water injection, hydrocarbon-gas injection, CO2 injection, water-alternating-CO2 injection, polymer injection, polymer-alternating-CO2 injection (PAG-CO2), and simultaneous polymer and CO2 injection (SPCO2). For simultaneous polymer and CO2 injection, polymer was injected at the top while CO2 was injected at the bottom. The simulation runs of these scenarios were elucidated in detail. The developed compositional PVT model successfully reproduces the observed fluids-properties and compositional variations with depth. In this way, the calculated fluids properties are continuous with depth because there is only a single PVT model for a single PVT region. The performances of different EOR scenarios were compared with each other. The simulated incremental oil recovery increases in the sequence of water injection, hydrocarbon-gas injection, WAG-CO2 injection, CO2 injection, polymer (22 cP) injection, PAG-CO2, and SPCO2. The reason for higher incremental recoveries with combined CO2-polymer scenarios is that both the macroscopic sweep (with polymer) and microscopic displacement efficiency (with CO2 and polymer) remain high. Although the CO2 injection pressure is lower than the MMP, the condensing- and vaporizing-gas drives are very efficient to the remaining oil saturation to low values (< 0.10). The other advantage of SPCO2 injection is that the intermediate and deep layers are well contacted and swept by the injected fluids. At the crest scale, combined CO2-polymer scenarios can increase the do-nothing recovery by 85–119%.