An inrush event is a sudden and uncontrolled inflow of material into an underground mine excavation, typically from an extraction-level drawpoint when in a cave mine setting. Inrushes can be dry, albeit these are rare, or in most cases wet creating a more mobile flow material (also referred to as mud rushes or wet muck spills) (Figure 1). Inrushes pose a major risk to mine safety, assets, and production by causing equipment damage, ore handling difficulties, production delays, unplanned dilution and, in extreme cases, injuries and fatalities.
Research into inrush causative factors and triggers has been underserved, despite caving operations increasingly being impacted by the hazard. Related risks will continue to increase as operating caves mature, extend to greater depths (with greater column heights), and/or break through to surface or into an overlying older cave. ICaRN is working to address related knowledge gaps and develop improved understanding, forecasting and hazard mitigation solutions in collaboration with our industry partners.
Identification of an inrush hazard level generally relies on classification, examining the percentage of fines present in the muck and its water saturation through drawpoint sampling. Once identified, various mitigation measures are taken as part of the inrush hazard management plan, the most effective being the utilization of remote equipment (e.g., semi-autonomous loaders). Automation has significantly improved operational safety when dealing with inrush hazards. However, mines can face severe interruptions in daily production due to equipment damage, the lower capacity of remote equipment and reduced drawpoint availability due to inrush events. Prediction of these interruptions is challenging due to uncertainties associated with the spatial distribution, frequency, and severity of inrush events.
One objective of ICaRN’s research is to identify the significant factors influencing inrush susceptibility (event likelihood) and severity (event volume and runout distance). In this project, the research questions are being addressed utilizing new developments and techniques in statistical analyses and machine learning applied to databases compiled by our partner mines (Figure 2). The knowledge gained from historical inrush events will be used to develop tools to predict the spatial and temporal pattern of inrush events and provide estimates of inrush severity. An example of a wet muck spill susceptibility map developed by Varian (2022) for the Deep Ore Zone (DOZ) mine is shown in Figure 3.
In this project, special focus has been placed on how inrush susceptibility and severity are affected by draw strategies. For example, Figure 4 shows how the probability of a high-volume wet muck spill event changes at different ranges of two draw-related variables: draw rate () and differential draw index (DI); DI is a new parameter proposed in this project for quantification of differential draw.
Incorporating effective draw-related variables into an inrush susceptibility/severity assessment tool gives the tool the capability of significantly outperforming the previously developed tools, by being able to guide prediction of the temporal pattern of inrush events and guide risk-informed draw optimization.
