CONSTRUCTION METHOD OF PARTICLE FLOW MODEL FOR ROCK WITH PRIMARY HIDDEN MICRO-FISSURES AND CALIBRATION METHOD OF MICRO-PARAMETERS

Discrete element simulation is one of the best ways to study the failure mechanism of the rock with fracture. The quality of the numerical simulation depends on the accuracy of the particle flow model and the rationality of the micro-parameters. However, the existing researchers usually take the particle flow model with pre-set regular fractures as the research object and pay more attention to the detailed analysis of the simulated phenomenon. The progress of the discrete element model building and the micro-parameters determination are ignored. Therefore, the methods to build the particle flow model of rock with hidden micro-fissures and determine its micro-parameters are established based on previous studies, CT scanning and image digitization technologies. The cryptocrystalline basalt with primarily hidden micro-fissures is taken as an example to apply and verify the methods. The following are found. CT scanning combined with image digitization technology can make the particle flow model of fractured rock more accurate. The micro-parameters of the parallel bond model get through the relationships between the micro-parameters. The mechanical parameters obtained by simulation should be adjusted due to the interaction of parameters. Besides, to determine the micro-parameters used to simulate, the response between micro-parameters and the mechanical parameters obtained by simulation at different dip angles must be analyzed. For the cryptocrystalline basalt with primarily hidden micro-fissures, the combination of "parallel bond model+smooth joint model" can be better applied to describe the mechanical properties of the contact of its particle flow model. The accuracy of the particle flow model and the rationality of the micro-parameters are greatly enhanced by the proposed methods. In addition, this research also can provide reference for the micro-parameters determination of particle flow model with multiple constitutive models.