Study On Surface Movement Of Coal Seam Group In Loess Mountain Area

The loess-covered mining area is an important coal production area in my country.There are relatively more researches on surface movement of mining subsidence in the loess mining area in China,and few studies on surface movement in this area abroad.The surface movement of mining subsidence in the loess mining area is more complicated,while the mining surface damage of coal seam groups is more serious,and the law of subsidence is extremely complicated,which seriously restricts the sustainable development of the economy.Therefore,the research on the surface movement of coal seam mining in the loess mountain area has certain practical significance for ensuring the high efficiency and high yield of the loess mining area.First of all,the thesis starts from the terrain analysis and divides the loess mountainous terrain into several typical characteristics of near-level,unidirectional slope(positive slope,negative slope),combined slope(convex,concave),and uses numerical software to build numerical models of different terrains.Draw the surface movement curve under different terrain conditions,and analyze the surface movement characteristics of the loess mountainous area with the measured data.Secondly,using the grey correlation analysis method,the main influencing factors of surface movement in the loess mountainous area are analyzed,and the sensitivity of each main influencing factor to surface movement is obtained.Finally,collect the measured data of multiple different working faces in the loess mining area,establish a neural network model,learn and train part of the measured data,and predict the surface movement of the Sandaogou mining area,compare the predicted value with the measured value,and obtain the predicted value and The average relative error of the measured value is 8.02%.The paper achieved the following results:(1)With the help of numerical software simulation,it can be concluded that the mining of the coal seam on a one-way slope,the amount of subsidence slipped on the slope of the top of the slope and the amount of mining subsidence form "superposition",and "offset" is formed at the bottom of the slope.During mining,the degree of "superimposition" and"offset" on the surface increases,and the center of the subsidence basin gradually moves to the top of the slope.In combined slopes,the ground movements of coal seam groups vary greatly,which indicates that the surface movements near the slope change point are more complicated.(2)Comprehensive evaluation of the main influencing factors of surface movement in coal seam mining through gray correlation analysis and analytic hierarchy process,the evaluation results obtained by the two methods are consistent.The analysis shows that the sensitivity of the main influencing factors of surface movement in the mining of coal seams in the loess mountainous area,from large to small,are:mining width,surface slope direction,mining depth,and mining height.(3)Using neural network to predict mining surface movement of coal seam group in loess mountain area,the results show that:for complex nonlinear systems,the fitting ability of neural network has a certain limit,and it cannot be completely fitted.The expected output value and measured value of the network There are certain errors,but most of the prediction results are relatively close to the measured values.The average relative error between the predicted results and the measured values is 8.02%after analysis,which can meet the needs of engineering applications.