This paper summarizes the use of the cellular automaton model in the modeling of rock failure processes. Three cellular automaton models, i.e. the physical cellular automaton (PCA), lattice cellular automaton (LCA), and elasto-plastic cellular automaton (EPCA), developed by the authors are presented and the associated applications described. The paper concentrates on the theory, basic ideas and application of the elasto-plastic cellular automaton which is an integration of elasto-plastic theory, cellular automaton theory, statistical theory, and rock mechanics.. In this study, a series of numerical tests, including rock failure processes under uniaxial compression, failure processes of rocks under cyclic loading and the simulation of Class I and Class II complete stress-strain curves, were conducted to reproduce some of the typical experimental phenomena. It is found that the EPCA model has the ability to simulate the initiation, propagation and coalescence of micro-cracks of rocks during the failure processes. The EPCA was also used to study the evolution of stress and deformations, as well as failure and permeability evolution in the Excavation Disturbed Zone (EDZ) of crystalline rocks. The results highlight the strong impact of fractures, and plastic deformation on the behavior of the EDZ, especially the evolution of permeability around the drift.
The mechanism of rock fracturing is of great interest in a wide variety of geotechnical applications, including the stability of land slopes, hydropower station construction, deep tunneling and the long-term disposal of radioactive wastes. Many of the difficulties that arise in geotechnical engineering are the result of the complexity of rock materials and the environment in which the rock exists. Rock is a = heterogeneous materials= and always contains various planes of weakness such as joints, faults and other geological structures, which have varying degrees of influence on the overall stability of the rock mass. Thus, research into rock failure and damage is one of the most difficult subjects, and much work has been devoted to the study of rock failure mechanisms using experimental, theoretical and numerical methods.
With the development of computer science over the last decades, numerical modeling has been widely used to investigate rock mass stability problems, the design of excavation support, and to plan and modify construction procedures.
In fact, in the past decades, many numerical methods and models have been developed. The combinations of statistical theory with numerical models, such as the lattice model (Cundall, 1988), the bonded particle model (Potyondy, 2004), and the RFPA model (Tang, 2000) based on the FEM have been found to be appropriate to simulate the progressive failure of brittle and heterogeneous materials such as rocks. Bažant and Oh (1993) considered the softening effect of the material and developed a random particle model. Zhong and Chang (1999) proposed a micro model to simulate the tensile fracture process of brittle material. Xing et al. (1999) developed a 3D beam-particle model for simulating meso-mechanical behavior of rock materials. Schlangen and Mier (1992) modeled the brittle failure process of concrete-like materials by using a simplified grid model.
