LAUSR

Abstract The combined finite-discrete element method (FDEM) is extensively employed to model rocks and rock-like materials, in which calibration against the results from uniaxial/triaxial compression tests, Brazilian tests and shear tests have been widely carried out. However, since different element sizes and loading rates were used, it is difficult to assess the numerical results of these studies if the effects of the element size and loading rate are ignored. This paper discusses the effect of the element size, loading/unloading rate and unloading mode on the cracking processes in laboratory-scale (uniaxial compression tests and Brazilian tests) and field-scale (circle tunnel excavations). The results indicate that the element size and loading rate should not be less than 27–28 meshes in a diameter zone and larger than 0.5 m/s, respectively, in laboratory-scale. In field-scale models, four different tunnel diameters (3, 4, 5 and 6 m) are used in tunnel excavation simulations. The results reveal that the mesh number around the tunnel wall should not be less than 120, meanwhile the element size should not be longer than the length of fracture process zone (lFPZ). The unloading rate for the field-scale model is influenced by the model size and rock density. It cannot be determined directly but a new method is proposed to determine the unloading rate based on crack development curve. To compare the fracture patterns well to those observed from the tunnel simulations, the unloading mode of an exponential formula is apposite.