Rock-Discontinuity-Fragmentation (R-Dis-Frag): A Computer Package for the Characterization of Large-Scale Fractured Rock Masses in Real-Time

Abstract

ABSTRACT: Rock block geometry (size, frequency, and shape) is an important characteristic in many rock engineering studies including rock mass classification, discrete element method, surface and underground stability analysis, drilling, blasting design and hydro-geological analyses. There are several polyhedral modelers available, but the limited computational capacity is one of the main issues when dealing with the characterization of real scale fractured rock mass up to millions of fractures at a real engineering time scale. This paper presents a novel approach for a voxel-based polyhedral modeler, optimized to exploit the computing parallelism of graphics processing units (GPUs). In this approach, there is no limitation for the complexity of rock blocks, fracture shapes, and excavation surfaces with complex morphologies. The computational time for the geometrical characterization of rock blocks for a 100-meters cubical domain with 2000,000 fractures is less than 3 minutes. 1. INTRODUCTION One of the most important sources of uncertainty for the evaluation of fractured rock mass behavior is the "block geometric uncertainty" which plays a critical role in the design of underground and surface opening. According to the literature review, the most significant discrete fracture network parameters when analyzing block geometry uncertainties are orientation, size, topology, volumetric intensity, and 3D mechanical connectivity (Rodriguez and Sitar, 2008; Wang et al, 2013; Zheng et al., 2015; Li et al., 2018, Hekmatnejad et al., 2021). If fractures in the rock mass cannot be fully described in 3D, uncertainties appear in the block formation and block geometry (size, shape, location), which in turn create another kind of uncertainty – that of driving force and resistance force; it also affects the amount of required support in tunneling application and the probability of hanging up occurrences at draw points in cave mining. When considering a discrete fracture network model, the most fundamental and important step is the identification of complex formed blocks, their geometry, and even their stability states (block stability analysis) for instability analysis around the surface and underground excavations (see Figure 1).