Research on data-driven coal mine environmental safety risk assessment system

Abstract

Coal mining operations are one of the most dangerous activities in the world, how to control coal mine environmental safety accidents has become a new challenge, there is an urgent need for an objective, accurate, comprehensive, and dynamic assessment system to assess the coal mine environmental safety risk. The paper constructs a data-driven coal mine environmental safety risk assessment system to guide coal mines under different environmental conditions to achieve environmental safety risk assessment: Constructing coal mine environmental safety risk assessment indicator system and threshold rules; Obtaining multi-source heterogeneous data such as expert experience data, objective monitoring data, and subjective reported data through expert prior knowledge, advanced sensors, and on-site technology; Eliminating data heterogeneity through fuzzy language transformation, range standardization, and threshold rule transformation based on data characteristics; Integrating multi source heterogeneous data through FAHP, CRITIC, Euclidean distance function, Grey cluster model based on end-point mixed triangular whitenization weight function, Grey clustering level − risk conversion value model (GCL-RCV model), Real-time risk status weighting model, and Linear weighting model. The accuracy, real-time performance, and effectiveness of the assessment system have been verified through case study. The system has achieved dynamic real-time assessment of coal mine environmental safety risk, identified and controlled critical risk, and has good industrial application prospect. It is of great significance for risk control in the mining process of underground coal mine and beneficial for the sustainable development of underground coal mine.