Prediction Of Coal And Gas Outburst Risk Based On RES Theory

As a geological disaster,complex coal and gas outburst often occur in underground coal mines,posing a serious threat to the safety of mine production in China.According to statistics,among the about 5,300 coal mining enterprises in China,there are 840 high-gas mines and 72 mines with coal and gas outburst characteristics.From the perspective of the whole process of coal and gas outburst,the prominent influencing factors are complex and diverse.The single evaluation index has been unable to determine the complex coal mining conditions,and the selection of prominent risk prediction index has become more applicable and flexible.The prediction of outburst danger can not only reduce the accident rate,but also have great significance to the maintenance of mine safety production order.Based on the RES theory,the cause mechanism and prediction method of the coal and gas outburst,and the disadvantages of the BP neural network,the method is optimized by SSA to obtain the interaction matrix coding value,the mass index,and the model is verified and applied in coal mine.There are mainly the following contents:(1)Summarize the research content and direction of coal and gas at home and abroad,summarize the mechanism hypothesis of coal and gas accidents,and determine five quantitative indexes through grey correlation analysis methods and references,summarize the prediction methods and characteristics,and establish prediction model based on RES theory.(2)In the interaction matrix coding,SSA optimizes the BP neural network,and the optimized SSA-BP network is better and the results are more accurate;obtain the membership of each influence index,realize the combination of membership and weight,and establish the cloud model of coal and gas outburst risk prediction.(3)Verify the accuracy of the model in Sershishan coal mine in Hunan Province.By mining the effects of the plastic deformation on 1 and 3,including the surface by Flac3D,the prediction index parameters in the critical value method and prediction model.