Study On Mining Subsidence Prediction Model And Parameters Inversion In Mountain Area

There are 1/3 of coal mines in mountain areas in our country,and the coal that is stored in the western mountain areas is becoming the focus of resource mining.The influence of underground coal mining on the surface in mountain areas is much greater than that of the plain areas,and there is some more discontinuous damage on surfaces,such as cracks,collapse pits and landslide,etc.In this paper,the general characteristics of surface movement and deformation in mountain areas were analyzed based on the observed data from observation stations of Jialequan 1208?1210 working face and Xiqu22101 working face in Taiyuan,Shanxi province;the Logistic time function was proposed to describe the dynamic laws of the subsidence of surface points with time.Based on the theory of probability integral method and some knowledge of soil mechanics,the mechanism of slope sliding under the influence of mining in mountain area and the relationship between the predicted slip amount(slip subsidence?W,slip horizontal movement?U)and topography were analyzed,and it was verified by two working faces as examples.The improved fruit flies algorithm was used to invert the prediction parameters related to Xiqu 22101 working face.At the same time,the parameters of the subsidence curves of each point varying with time on five observation lines on two working faces were fitted by Logistic time function,and the law of the obtained parameters was also analyzed.The main results achieved are as follows:(1)The calculation coordinate system of any shape working face based on probability integral prediction model was established automatically,so that mining subsidence prediction can be carried out more quickly and accurately.(2)The mathematical characteristics of Logistic time function and its parameters significance were analyzed,and the applicability of using Logistic time function to describe the variation regulation of the surface point subsidences with time is shown.(3)According to the observed data of surface observation stations,the variation law of surface movement and deformation curves in mountain areas,especially the distribution characteristics of extreme points on curves,were analyzed.The parameters of the surface moving angles and the dynamic parameters of the moving deformation and the probability integral method prediction parameters were obtained,and the laws of the surface subsidence changing with time were analyzed.(4)Based on the combination theory of the probability integral method and the slip mechanism of the mountain slope,the expected formulas of the slip subsidence and slip horizontal movement were derived,and the mechanism of the slip of the mining slope under the influence of the tensile-type ground crack was analyzed by the related knowledge of the soil mechanics.An example verification was carried out based on the Jialequan 1208?1210 working face.The estimated values of subsidence and horizontal movement which is considering the slip of the slope were compared with the measured value,and the mean square deviation is obviously reduced,indicating the applicability of the prediction model of the mining subsidence in the mountain area in consideration of the slip of the slope.(5)On the basis of the measured data of Xiqu 22101 working face,the prediction parameters of the probability integral method were inverted by using the improved fruit flies algorithm,the measured value was taken as a reference,and the estimated values obtained by using the optimized parameters and the parameters calculated by the empirical formula were compared,the mean square deviation of the fitting subsidence was reduced from 248 mm to 122 mm,and the mean square deviation of the horizontal movement was reduced from 397 mm to 386 mm.(6)By analyzing the time-dependent variation of the subsidence value of each observation line,the following results are obtained:The fitting R~2 of the Logistic time function for the time-varying subsidence curve for each point is more than 97%.The curves of coefficient K and r with the distance between the surface point and the open-off cut in Logistic time function can be fitted by cubic polynomial,and the fitting R~2 is more than 80%.The curve of coefficient c with the distance between the surface point and the open-off cut can be fitted by normal distribution function,and the fitting R~2 is above 98%.It is shown that the Logistic time function is applicable to describe the dynamic subsidence of surface points in mountain areas.