The combined finite-discrete element method (FDEM) has been extensively used for rock fracturing simulation. The zero-thickness intrinsic cohesive elements are commonly implemented in FDEM and pre-inserted in the rock model between adjacent finite elements prior to simulation. However, because of the different constitutive laws for cohesive and finite elements, the rock model domain may deform like a discontinuum in the elastic stage (abbreviated as dFDEM). This could cause unrealistic material deformation and also reduce computational efficiency. Here, we propose a novel node binding algorithm to ensure the continuum behavior of materials prior to fracture onset (abbreviated as cFDEM), which can also automatically achieve the explicit separation of fracture surfaces without using complex node splitting algorithm. We validate the robustness of the proposed cFDEM and demonstrate its advantage compared to dFDEM. The work provides a novel perspective for rock fracturing simulation in FDEM.
The combined finite-discrete element method (FDEM) (Munjiza, 1992), which merges FEM-based analysis of continua with DEM-based contact processing for discontinua, provides an effective solution to simulate the fracturing behavior in rocks. Generally, the FDEM is realized using the intrinsic cohesive zone model (ICZM) with a traction-separation law, in which the modeling domain is first discretized into a series of finite elements, and then cohesive elements are inserted into the common boundaries between adjacent finite elements (Munjiza, 2004).
Because finite elements and intrinsic cohesive elements use different types of constitutive laws, they deform at different rates even in the elastic deformation stage, and thus may cause discontinuous strains across adjacent finite elements and make the originally continuous model domain behave like a discontinuum before fracture onset (for convenience, it is referred to as dFDEM hereafter. Since the inherent stiffness difference between the intrinsic cohesive elements and the solid finite elements, the dFDEM usually yields a smaller overall material elastic modulus than the true value, i.e., stiffness reduction, or the so-called artificial compliance problem (Xu et al., 2022).
