Two-phase flow in heterogeneous porous media: A multiscale digital model approach

Abstract

Pore-scale fluid flow simulation on digital rocks from X-ray CT experiments can predict the physical properties and analyze the distribution characteristics of the fluids, which is critical for many problems in environmental remediation and geo-materials. It is straightforward to numerically simulate the multiphase flow on the digital rocks from CT scanning if they are homogeneous. However, it is of great difficulty to obtain large and high-resolution digital models of heterogeneous samples containing multiscale features only by CT scanning since the CT machines strike a balance between the field of view and resolution. In addition, the current methods of multiphase flow modeling face challenges in dealing with large samples and heavy computations. To address such issues, a unified and integrative modeling approach for constructing multiscale digital images and simulating multiphase flow in complex and heterogeneous pore systems was presented. Firstly, we developed a hybrid modeling method for producing multiscale digital models to represent the heterogeneous pore systems on which two-phase flow simulations are computed. The results show that the constructed models built by the hybrid method contain more tiny pores than the ones from regular CT imaging and exhibit better connectivity and permeability. The pore/throat size distributions, porosity, and permeability of the digital models produced by the hybrid method are closer to the experimental data than the ones from the single-scale models. Moreover, the non-uniform meshes for complex pore systems and parallel computing on a supercomputing cluster are combined to improve the computational efficiency of the two-phase flow simulation. The distribution characteristics of fluids in pore systems at the end of two-phase flow were also analyzed, which improves our understanding of two-phase flow.