Sensitivity analysis of fracture energies for the combined finite-discrete element method (FDEM)

2021 
Abstract The combined finite-discrete element method (FDEM) has been widely used in the numerical study of rock mechanics and geotechnical engineering. Selection of the correct parameter values is a prerequisite to ensure the accuracy of the simulation results. However, current calibration procedures are relatively cumbersome, and the parameters are strongly dependent on the element size. To eliminate the dependence of parameter values on the element size, this paper proposes a new constitutive model based on the stress–strain relationship. In addition, uniaxial compression and direct tension simulations are used to investigate the sensitivities of type II and type I fracture energies, i.e., GII and GI, including the influence of the GI (or GII) value on the simulation results of direct tension (or uniaxial compression) and the influence of different parameters (such as Young’s modulus E, Poisson's ratio ν, cohesion c, internal friction angle φi, tensile strength ft and element size h) on the selection of the values of GII and GI. The following results are obtained. (1) As GI or GII increases, the strength of the rock sample increases, and the plastic characteristics are more obvious, but the failure mode tends to be stable. (2) As the E value increases, the value of GI or GII decreases as a power function. As the value of ft or c increases, the value of GI or GII increases as a power function. However, other parameters, including ν, φi and h, have no effect on the selection of the values of GII and GI. (3) Actual uniaxial compression and triaxial compression simulations prove that the calibrated GII and GI values are reasonable. In this study, only two fracture energies, GII and GI, must be calibrated; other macroparameters can be obtained through laboratory tests, and other microparameters can be obtained from previous empirical or theoretical values, which greatly improves the calibration efficiency of FDEM input parameters.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    55
    References
    3
    Citations
    NaN
    KQI
    []