Water-inrush assessment of coal 12 floor using a GIS-based bayesian network for Fangezhuang Coal Mine with collapse column

Abstract：In order to evaluate the risk of coal mine water inrush of the collapse column, taking Fangezhuang coal mine for example, on the basis of depth analysis of Fangezhuang coal mine water inrush factors, the K-Means algorithm was used to identify and discrete the data of water inrush factors. The combination of ARCGIS and BN was used to establish water inrush risk assessment model of 12 coal floor. By learning of the parameters and reasoning of water inrush in BN, hazardous areas of water inrush was conducted division, and the effects of the model was analyzed. As it turned out, hazardous areas of water inrush are located in the southeast and northwest. The northwest part belongs to typical water inrush areas of collapse column, and the highest risk areas are basically consistent with abnormal development zone of collapse column in the northwest;to a certain extent,widely distributed collapse pits in the northwest increase potential possibility of water inrush. The southeast dangerous area, typical water inrush area of fault, concentrated in surrounding areas of two major fault lines between FS-F8 and F4-FIO-F11-F12. With large possibility of water inrush above areas, it should take some appropriate measures to prevent water inrush. The fitting efficiency of evaluation can reach 80%. So using the method of GIS-BN is feasible and accurate to evaluate water inrush risk in coal mine.