LAUSR

Abstract It is of great significance to analyze the distribution characteristics of the non-wetting phase during the multiphase flow and identify the positions of the oil ganglia since it can guide us to design the schemes for improving oil recovery. In this paper, the two-phase (brine and oil) flow experiment and simulation are performed on a water-wet sandstone. A comprehensive analysis of the distribution of the non-wetting (oil) phase in 3D pore space and individual pores are presented. To calculate the pore occupancy of the non-wetting ganglia in every single pore from 3D CT experimental data, we proposed a technique, called the Axis-aligned Bounding Box (AABB) algorithm, to determine which oil blobs exist in a unique pore. The pore-by-pore distribution characteristics of the residual phase are then analyzed based on the experimental and numerical data. The results reveal the large oil clusters are separated into small blobs during the imbibition process, which results in increasing the number of non-wetting ganglia, dropping its connectivity, and increasing the smoothness. Moreover, the larger the volume of one pore, the larger the volume and saturation of residual oil ganglia in the pore are. Furthermore, our analysis based on the experimental and numerical results indicates that most of the non-wetting ganglia reside in the median-size pores. Hence, the finding provides us a piece of significant evidence to better target the pore-spaces for the purpose of increasing the recovery.