ABSTRACT Predicting the physical properties of porous materials is difficult due to the complexity of the underlying microstructure that determines physics at the macroscopic scales of interest to the practitioner. The study of the relation between macroscopic physics and the microstructure can be largely facilitated if simplified and physically relevant representations of the microstructure are available. Herein, it is to reduce the structural complexity of a 3D digital image of an actual sample to the only rules describing the association of pore throat sizes and shapes to adjacent pores. Such rules, shown by a variety of independent approaches to govern the main part of flow and transport are measured by image analysis and expressed in the form of a pore network model. Relying on a random model of porous microstructures, the approach can be applied to isotropic porous materials whose observation by imagery is limited to 2D due to the small spatial resolution of the porosity.
Read full abstract