A discrete fracture network based approach to defining in situ, primary and secondary fragmentation distributions for the Cadia East panel cave project

Abstract

One of the most critical elements in pre-caving assessment process is determining the likely distribution of rock mass fragmentation, with the impact of poor or unexpected fragmentation upon cave operations being significant. For a number of years it has been recognised that discrete fracture network (DFN) tools could assist in the fragmentation assessment, primarily through providing a better description of the pre-caving in situ fragmentation distribution. To date the evaluation of primary and secondary fragmentation has been mostly carried out using alternative methods based on engineering principles and practical experience. More recently, synthetic testing of relatively small DFN models has been proposed to assess fragmentation mechanisms. However, it is argued that neither of these approaches can fully capture the broader heterogeneity of the rock mass, drawing on only a limited portion of the rock mass characterisation data. Recent work has demonstrated the sensitivity of rock mass fragmentation to the volumetric fracture intensity property P32 and the importance of determining the critical intensity value at which the transition from intact massive rock mass to kinematically mobile rock mass occurs. To address these issues, the authors have developed an approach that has at its core the development of a full scale DFN model description of fracture orientation, size and intensity built up from all available geotechnical data. The model fully accounts for a spatially variable description of the fracture intensity distribution. Primary fragmentation analysis is undertaken by using a DFN based rule approach, which draws from an explicit numerical simulation of fracture mechanisms to characterise stress induced fracturing within a given rock mass. Direct modelling of secondary fragmentation related to mining-induced stress and comminution of caved material in the broken ore column poses significant challenges as large-scale discrete modelling (including fracturing) of processes requiring a centimetre-scale mesh discretisation becomes computationally prohibitive. To obviate this problem, this paper introduces a method to assess secondary fragmentation based on a probabilistic analysis by combining the DFN derived primary block volume distribution with micro-defect intensity data to derive a probability of block degradation during caving.