Comparison of different models for predicting drainage relative permeability using pore scale numerical simulation of supercritical carbon dioxide and brine flow.

Abstract

In today's technologically growing and cost optimisation era, Digital Rock Physics (DRP) is becoming a potential alternative tool to the high cost and time-consuming method of Special Core Analyses (SCAL), for the estimation of reservoir fluid properties. The key objective of this study is to compare different models for predicting drainage relative permeability using pore scale numerical simulation of supercritical carbon dioxide and brine flow, the former being the non-wetting phase and the latter being the wetting phase. The simulations are done on an established computational fluid dynamics (CFD) simulator: ANSYS-CFX. From the simulations, we obtain the saturation values. Using these values and four common models found in the literature, we collaborate the J-function prediction model and capillary pressure values to calculate the relative permeability for the wetting and non-wetting phases. In this study, it is concluded that the Brooks-Corey-Burdine model produces the best relative permeability curves. It is demonstrated that the pore size distribution index (λ) has a positive correlation with the wetting-phase relative permeability and an inverse relation with the non-wetting phase relative permeability. Noteworthy to mention that during the study, it was found that when the J-function is substituted into the relative permeability equations, the predicted relative permeabilities of wetting and non-wetting phase outcomes appears to be more representative, yet simple to compute. The robust methodology described in this paper to evaluate the simulation results with various established models is generic and could be useful in current day oil field practices to serve as a cost-effective alternative to SCAL experiments and form key inputs for typical oil field development planning.