ABSTRACT
Deep shafts provide critical support services for underground mines and therefore robust engineering assessment and design processes are vital for their successful, cost-effective construction and reliable operation. With the trend towards increasingly challenging and complex shaft projects in terms of diameter, depth, ground conditions and demands for optimisation, there is an ever-increasing need for improvements to the engineering processes. Conventionally, geotechnical assessments for shafts have widely employed empirical methods such as the Raise Bore Rock Quality (QR) system proposed by McCracken & Stacey (1989). While these methods have provided a useful tool to aid design, they have typically been based on a limited database of project experience and so their results need to be interpreted in the context of local experience and engineering judgement. To advance the engineering process and achieve optimised designs, a more quantified approach that considers the site-specific conditions is required. This paper describes an approach based on Discrete Fracture Network (DFN) modelling integrated with discontinuum-based numerical modelling. In this approach, features observed in drill core and televiewer logs are used to derive a stochastically representative ‘synthetic rock mass’ model, which then can be exploited to assess potential rock mass behavior. The DFN model developed was incorporated into a full-fledged 3D discontinuum model to assess the rock mass instability, joint-controlled wedges and stress-induced spalling potential. This combined approach allows us to quantify the likelihood of instability in different rock strata, potential rock wedge sizes, factor of safety of different wedges, potential rock mass deformation and potential major failure mechanisms which are a combination of major instability factors. Ultimately, this approach allows engineers to optimise possible solutions for shaft excavation and support design.
