Predictions of relative permeability for low permeability reservoirs and its scale effect

Nowadays, reservoirs with a low or extremely low permeability are of crucial importance for additional oil/gas recovery. Better understandings of multiphase displacement at pore scale in such unconventional reservoirs are necessary. Since the microscale interfacial transport plays a very important role in such processes, researchers were devoted to discover phenomenon at smaller and smaller resolution, and even further to nanoscale. However, this is extremely consuming on both experimental techniques and computation resources. Therefore, we are wondering whether there exists a cut-off resolution or not, below which the absolute and relative permeability are changeless. In this study, the Navier-Stokes equation for fluid flow is directly solved through porous media. For two-phase flow, the volume-of-fluid method is applied. The numerical code is written within the framework of OpenFOAM, which is a widely used open source CFD solver recently. The method is validated through several testing cases, which indicates the accuracy of capturing the interface action, wetting phenomenon, and permeability. After validations, we perform a series of modeling of single-phase and two-phase flow simulations in low permeability reservoirs, with special focus on the scale effect. The rock samples are taken from the low-permeability Changqing oil field in China. We reproduce the digital rock structures at various scales by using the micro-CT scanned data. The results indicate that the scanning scale has great impact on the predicted permeability, and the reservoir permeability increases with resolution due to the increasing of micro pore connectivity. However, this tendency is found to decline when reaching a higher resolution.