Simulation of Random Packing of Polydisperse Particles

Porous media is of great interest to many engineering and science applications such as fluid flow and transport through natural subsurface systems, and gas-liquid flow through packed beds in petroleum and chemical applications. Micro-structure characterization of porous media is critical to predict or understand macroscopic properties such as permeability, pressure-saturation relations, or residual saturation of non- wetting fluids in multiphase porous media systems. In this presentation, an optimization based model that simulates random packing of polydisperse particles to mimic porous media systems is presented. The model generates systems with properties match pre-specified ones. The model incorporates physically based properties of random packing of glass bead systems obtained from 3D competed tomography images. Properties include the distribution of particle contacts (i.e., coordination number) and particle size distribution. Pore-structure characterization of the packings including tests for isotropy, heterogeneity, and randomness is presented. The model can be used to investigate the impact of variation in packing arrangement and pore-structure on macroscopic properties.