A hybrid process-based and stochastic reconstruction method of porous media

Abstract

In the present study we have developed a hybrid method that combines statistical with process-based reconstruction methods to recreate the three dimensional morphology of a given porous medium. The method starts with the generation of unconsolidated porous media, made from a process-based method such as the random packing of equal or unequal sized hard spheres and proceeds with a statistical method, the simulated annealing algorithm that matches the original image statistics obtained from two-dimensional SEM micrographs. We have determined the Darcy permeability in the reconstructed 3D domains and compared it with the Kozeny–Carman formula and experimental data obtained from in-house measurements of a sintered porous ceramic and from reported values for the anode cermet of a solid oxide fuel cell. The hybrid method simulation results are in very good agreement with the experimental permeability values and slightly better than those obtained by using simulated annealing alone.