Mesoscopic modeling approach and application based on rock thin slices and nanoindentation

Abstract

The current rock mesomechanics modeling methods have limitations such as imprecise characterization degrees and unreasonable parameter settings. A method for optimizing rock mesomechanical parameters was proposed based on the framework of the polycrystalline discrete element method. The modeling strategy was to characterize the geometric irregularity of mineral grains using the 3D Voronoi tessellation and distinguishing the inter‐ and intra‐granular characteristics by particle growth method. The optimization idea includes determining the mineral type and size distribution through quantitative analysis of thin sections and determining mineral strength through nanoindentation experiments. The method was applied to investigate the effect of mineral dissolution on rock mechanical properties. The results indicated that digital images could quantitatively identify the mineral type, corresponding size distribution, and dissolution degree. Deterministic parameters obtained through nanoindentation experiments reduce the uncertainty of parameter calibration, thus improving the model's reliability. The application results indicated that the rock is more prone to deformation and damage with increased dissolution. The microcracks expand and connect with the original defects to form macroscopic cracks. The clustering of microcracks is more likely to occur between two adjacent relatively large local defects and further interconnect with the original local defects to form macroscopic destructive cracks.