Health monitoring of ground support system through point cloud processing: enhanced deformation analytics phase

Purpose In underground mining, geostatic pressure results in drift surface deformation. To ensure the stability and safety of the mining environment, a so-called ground support systems are commonly used. Today, there are various types of ground support systems, such as rockbolts, steel sets, mesh and shotcrete, cable bolts, ground anchors, etc. Ground support systems based on rockbolts are used to provide the tensioning support to the compressive strength of the rocks. Surface deformation occurring in the drift brings changes in the rockbolt protruding from the surface. Monitoring and predicting the health of the ground support would assist the decision process for rehabilitation of the ground support in the drift to maintain safe working conditions. Design/methodology/approach In this article, a methodology has been proposed aimed at the quantification of deformation of underground mining drifts through point cloud processing (PCD) and other data sources to support health monitoring, development of decision support tools and integration with the mine Trigger, Action, Response Plans (TARPs). Findings Quantification and comparison of rockbolt protruding length is indicative of the surface deformation, which can be visualised to access regions requiring rehabilitation. Research limitations/implications The results of the developed algorithms are dependent on the quality of the data governed by human process during data collection. Recommendations have been made to the industries to improve the practices. Practical implications The results were used to augment the PCD to create interactive visualisations on virtual reality (VR) and augmented reality (AR) systems for off-site or on-site inspections. Social implications The presented work improves the safety of the working environment in deep underground mines by providing quantifiable results and useable visualisations. Originality/value The proposed methodology processes point cloud data, collected in underground mining drifts, in a campaign-based manner. PCD processing focuses on (1) comparison of PCD quality, (2) computation of deformation while compensating for imperfect registration and (3) comparison of rockbolt regions using Wasserstein distance.