Challenges in multi-scale hard rock behaviour evaluation at deep underground excavations

Abstract

As a consequence of rapid growing trend of resource extraction in world, depth of excavations for resource exploitation increases. Eventually excavations faces with transition from low stress to high stress condition. In this paper, comprehensive aspects on rock behaviour at deep underground excavation were investigated. The state of art of rock behaviour at micro- meso- and macro-scale were discussed and relevant challenges along with achieved knowledge, experiences, and research results were presented. At micro-scale, research results revealed that, apart from chemical bonding, rock behaviour significantly influenced by deficiencies such as; particle-crystal boundaries, heterogeneity, pores and micro-cracks, which reduces the rock strength 2-3 order of magnitude. Granite SEM images proves the deficiencies between crystals, micro-cracks and pores at each crystal, and weakness and foliation of mica components. When stresses applied on specimen, new tensile cracks nucleated and initiated from the edge of existing micro-cracks, and rate of crack propagation depends on the differential stress level. At meso-scale, true triaxial testing makes it possible to apply different stress paths in the ranges of ground in situ stresses, concentrated stresses and even dynamic loads. Careful assessment of the full stress–strain curves of the true triaxial test results of granite and conventional triaxial test results of Marble shows that rock mechanical properties such as magnitude of linear elasticity, ductility domain, peak strength value, ranges of brittleness, and residual strength level significantly differs with changing confining stresses. The rock stress – strain behaviour variation were categorised to four distinct stages consisting; 1) Elastic-stable micro-cracking, 2) Stable - unstable micro-cracking, 3) Unstable micro-cracking-brittle failure, and 4) Brittle failure-residual strength. The ranges of rock behaviour at each stage with different confining stresses were illustrated, which could be used as input for mechanical parameters in design analysis. At macro-scale, counteraction between ‘Rock Mass Composition (RMC)’, ‘Active Stress Condition (ASC)’, and ‘Excavation Method, Size and Orientation (EMSO)’ to estimate the ‘Rock Mass Behaviour (RMB)’ were discussed and presented as a verbal equation. To reduce the sudden failure risk, a micro-seismic monitoring system were designed and implemented for perdition and warning of failure and evacuation in timely manner. To verify the presented approaches, rock mass behaviour and failure mechanisms were illustrated in a deep gold mine in Western Australia. To manage the ground behaviour; considering the static and dynamic loading and interlocked nature of rock masses at deep underground excavations, the ratio of “Ground energy demand” to “support energy absorption capacity” is mostly used for stability evaluation. Finally, it should be noted that, the geomechanics at general and deep underground geomechanics specifically is a developing field due to incapability to achieve proper ground characteristics, huge number of variables and their coupled interactions, and incompetence in analysis them properly. Therefore, the results from current analysis should not be taken as granted and always solid engineering judgement must involve in interpretation and design. It is also hoped that future development in sophisticated ground exploration technologies along with advances in computation science will assist geomechanics engineers to mature their knowledge of rock mass behaviour and safe and economic design in engineering activities.