Permeability estimation of porous media by using an improved capillary bundle model based on micro-CT derived pore geometries

The purpose of this work is to develop a permeability estimation method for porous media. This method is based on an improved capillary bundle model by introducing some pore geometries. We firstly carried out micro-CT scans to extract the 3D digital model of porous media. Then we applied a maximum ball extraction method to the digital model to obtain the topological and geometrical pore parameters such as the pore radius, the throat radius and length and the average coordination number. We also applied a random walker method to calculate the tortuosity factors of porous media. We improved the capillary bundle model by introducing the pore geometries and tortuosity factors. Finally, we calculated the absolute permeabilities of four kinds of porous media formed of glass beads and compared the results with experiments and several other models to verify the improved model. We found that the calculated permeabilities using this improved capillary bundle model show better agreement with the measured permeabilities than the other methods.