A data-driven quantitative model for predicting floor groundwater inrush risk under deep and thick coal seam mining

Abstract

Abstract With the increase of coal mining depth, water hazards in the coal mine floor occur frequently. The coal production process is faced with complex water inrush mechanism and variable water inrush main control factors, and the uncertainties among the factors make the prediction of floor water inrush more difficult. In this paper, Tangjiahui Coal Mine, a Northwest China typical coalfield, in the Inner Mongolia Autonomous Region is taken as the research object. The prediction index system including aquifer capacity, aquiclude capacity, and geological structure is selected, with seven prediction factors being considered. Secondly, the analytic hierarchy process and entropy weight method are used to calculate the subjective and objective weights. On this basis, two models of comprehensive weight based on AHP-EW improved by game theory and improved variable weight of floor water inrush risk based on the foundation of comprehensive weight are constructed. The predicted results are displayed by using the powerful spatial management and information processing functions of GIS, and the performance of the two models is discussed and compared. By comparing the prediction results with the in-situ water inrush points, it is found that these positions are in the relatively hazardous areas of floor water inrush, which proves that the prediction model has high accuracy. Finally, the prevention measures of floor water inrush are put forward according to the risk zoning results. The research results can provide a scientific theoretical basis for mine water disaster prediction, and it is also conducive to the sustainable utilization of groundwater resources.