Application of A* algorithm for microstructure and transport properties characterization from 3D rock images

Thorough comprehension of flow behavior in underground porous media is fundamental for several applications such as oil and gas production, Underground Gas Storage, CO2 storage, and Enhanced Geothermal Systems. Macroscale petrophysical parameters, as well as hydraulic parameters, are strongly linked to the microstructure of the rock. In this paper, we present a methodology for the geometric analysis and characterization of the pore structure of 3D binary images of rocks. The geometric analysis is based on the A* pathfinding algorithm extended to 3D domains and on the measurement of the pore radius along the identified paths. The analysis is carried out for the main flow directions to obtain a tensorial representation of tortuosity, effective porosity, and representative pore radius, to provide permeability estimation and effective characterization of anisotropy. Moreover, the approach provides the analysis of pore size distribution and constriction. The methodology was applied to synthetic but realistic rock samples, generated through the QSGS algorithm. Two case studies, representative of an isotropic and an anisotropic porous media, are presented. Validation was carried out through comparison with FVM hydrodynamic modeling. Analysis of the results shows that the presented geometric approach can provide thorough and reliable characterization of the porous media.