Multiple-point Statistics to Generate Pore Space Images

Multiple-point statistics (MPS) on a two-dimensional (2D) thin section image are used to generate a three-dimensional (3D) pore space image with an assumption of isotropy for orthogonal planes. The method gives images that preserve typical patterns of the void space seen in thin section. Using only single and two-point statistics in the reconstruction often underestimates the void connectivity, especially for low porosity materials; however, multiple-point statistics method significantly improves the void connectivity. The method is tested on sandstone and carbonate samples. Permeability is predicted directly on the 3D images using the lattice Boltzmann method (LBM). The numerically estimated results are in good agreement with experimentally measured permeability. Furthermore, the method provides an important input for the creation of geologically realistic networks for pore-scale modeling to predict multiphase flow properties.