Two Methods For Pore Network Of Porous Media

Shengyan Gao,Jay N. Meegoda,Liming Hu
DOI: https://doi.org/10.1002/nag.1134
IF: 4
2012-01-01
International Journal for Numerical and Analytical Methods in Geomechanics
Abstract:Two approaches of generating pore networks of porous media are presented to capture the pore fabric. The first methodology extracted pore structure from a computer simulated packing of spheres. The modified Delaunay tessellation was used to describe the porous media, and modified NelderMead method in conjunction with three pore-merging algorithms was used to generate the pore size and coordination number distributions of the randomly packed spheres. The Biconical Abscissa Asymmetric CONcentric bond was used to describe the connection between two adjacent voids. This algorithm was validated by predicting pore structure of a cubic array of spheres of equal radius with known pore sizes, throat sizes and coordination number distributions. The predicted distributions of pore structure agreed well with the measured. Then, the algorithm was used to predict pore structure and permeability of randomly packed spherical particles, and predicted permeability values were compared with published experimental data. The results showed that the predicted permeability values were in good agreement with those measured, confirming the proposed algorithm can capture the main flow paths of packed beds. The second methodology generated an equivalent pore network of porous media, of which the centers of voids were located in a regular lattice with constant pore center distance. However, this network allowed for matching both main geometrical and topological characteristics of the porous media. A comparison of the two approaches suggested that the second approach can also be used as a predictive tool to quantitatively study the microscopic properties of flow through porous media. Copyright (c) 2012 John Wiley & Sons, Ltd.
What problem does this paper attempt to address?