Numerical Simulation Study of Brittle Rock Materials from Micro to Macro Scales Using Digital Image Processing and Parallel Computing

Abstract

The multi-scale, high-resolution and accurate structural modeling of rocks is a powerful means to reveal the complex failure mechanisms of rocks and evaluate rock engineering safety. Due to the non-uniformity and opacity of rocks, describing their internal microstructure, mesostructure and macro joints accurately, and how to model their progressive fracture process, is a significant challenge. This paper aims to build a numerical method that can take into account real spatial structures of rocks and be applied to the study of crack propagation and failure in different scales of rocks. By combining the failure process analysis (RFPA) simulator with digital image processing technology, large-scale finite element models of multi-scale rocks, considering microstructure, mesostructure, and macro joints, were created to study mechanical and fracture behaviors on a cloud computing platform. The Windows-Linux interactive method was used for digital image processing and parallel computing. The simulation results show that the combination of a parallel RFPA solver and digital image modeling technology can achieve high-resolution structural modeling and high-efficiency calculation. In microscopic simulations, the influence of shale fractures and mineral spatial distribution on the fracture formation process can be revealed. In the mesostructure simulation, it can be seen that the spatial distribution of minerals has an impact on the splitting mode of the Brazilian splitting model. In the simulation of a joined rock mass, the progressive failure process can be effectively simulated. According to the results, it seems that the finite element parallel computing simulation method based on digital images can simulate the multi-scale failure process of brittle materials from micro to macro scales. Primarily, efficient parallel computing based on a cloud platform allows for the multi-scale, high-resolution and realistic modeling and analysis of rock materials.