Open Collections

Abstract

Mines are being developed at greater depths to meet the growing global demand for critical minerals (e.g., copper and gold). These mines require extensive networks of mining tunnels to access, extract, and transport the ore. As the experience base for the commonly applied mining methods (e.g., panel and block caves) comes from shallower depths, where the geological conditions differ, there are critical gaps in the engineering approaches for assessing tunnel stability and ground support, leading to potentially unsafe and unreliable designs. In particular, the engineering approaches to address the potential for spalling and strainbursting mechanisms are limited. Spalling is a gradual process, whereas strainbursting is sudden, energetic, and a type of ‘rockburst’. A methodology is proposed for recording, databasing, and performing data-driven assessments of rockbursting in deep mines to correctly identify the damage mechanisms, the triggers, and important controlling factors. Novel databasing techniques and new indices are introduced, which are suited for large-scale mines where the damage from rockbursts occurring within a short temporal window can be widespread. Data driven analyses are introduced to track rockburst damage, cross-reference the data against other data sets, distinguish rockburst damage mechanisms, and identify key contributing factors. Using two-dimensional bonded block models, a methodology is proposed to simulate spalling and strainbursting explicitly. While existing numerical tools are reasonably adept at forecasting spalling, there is a critical gap in forecasting strainbursting. Therefore, a new contact constitutive model was developed, and the role of mechanical damping was explored in detail as it relates to the peak and post-peak behaviour of brittle rock masses. Conceptual simulations were developed to validate that the approach could correctly capture fracturing mechanisms, variable rupture energies, and post-peak behaviours with spalling and strainbursting mechanisms emerging ‘naturally’ in the simulations. A detailed case study of the Deep Mill Level Zone mine in Indonesia was used to validate the data driven and numerical assessment approaches. The value of the novel approaches is demonstrated in ground support design assessments, preventative support maintenance planning, and understanding key contributing factors that influence susceptibility to strainbursting. These approaches will contribute to improved safety and reliability of mining tunnel designs.