Research On Mining Subsidence Law Analysis And Prediction Methods Based On DInSAR Technology

Huainan area is rich in coal resources,and while a large number of coal mining drives economic development,it also has a bad impact on the local ecological environment and the production or life of residents,so it is extremely necessary to strengthen the monitoring of mining subsidence.By obtaining the subsidence data to analyze the subsidence law of the mining area,it can not only provide continuous surface deformation map and forecast disaster,but also provide reliable reference data for other mining areas with similar geological and mining conditions,such as coal mining method,maximum surface subsidence and subsidence basin range,single point subsidence trend and so on.In view of the limitation of traditional measurement and the disadvantage of consuming a lot of manpower and material resources,DInSAR technology is applied to monitor the subsidence of 1613(1)working face in Huainan Mining Area.Combining the complementary advantages of the two data,the subsidence law and subsidence prediction of the study area are analyzed and studied.The main research results are as follows:(1)The interference processing of 22 image data is completed to obtain 21 interference images,and the timing analysis of the interference results of 17 of them is carried out to roughly confirm that the monitoring results of DInSAR are basically consistent with the mining time and subsidence range.(2)The time of horizontal observation and DInSAR monitoring is unified,the LOS deformation is decomposed into east-west,north-south and vertical deformation,and the two kinds of data are assimilated.Based on this,the mining subsidence parameters such as inclination,curvature,advance influence Angle and advance influence distance are solved,and the mining subsidence law is obtained.(3)The probability integral parameter inversion model combining DInSAR and multi-universe parallel quantum genetic algorithm is constructed.The model is constructed based on probability integral prediction model and LOS projection principle,and the feasibility of the model is verified by simulation experiments.On the premise of reconstructing survey lines and points based on leveling data,the model has achieved good results in inverting the probability integral parameters of the study area.(4)A fuzzy time series prediction model of surface subsidence based on improved Wolf pack algorithm is constructed.Combined with the characteristic that the time series need the same time interval,the vertical subsidence data monitored by DInSAR is applied to predict the subsidence amount of the maximum subsidence point in the stable period,and a good effect is achieved.Figure[30]Table[12]Reference[81]...