Research On Mining Area Subsidence Monitoring And Dynamic Prediction Based On SBAS-InSAR

The surface deformation caused by underground mining is time-dependent and highly nonlinear,and it will cause gradual damage to the surface structure during the underground mining process.Traditional monitoring methods have certain shortcomings.The main feature of differential synthetic aperture radar measurement technology(InSAR)is that it has high spatial resolution and high accuracy in a wide coverage area.Due to its unique advantages,this technology is widely used in ground deformation monitoring.However,in coal mining areas,large-scale subsidence of the surface will occur in a short period of time,resulting in inaccurate InSAR results and limiting its use in monitoring mining subsidence.Based on this,this paper proposes a data processing method that combines the SBAS-InSAR technology with the probability integration method used to predict mining subsidence to overcome these shortcomings.Combined with the improved knothe time function,a dynamic prediction model for the settlement of the mining area was established,and the dynamic change process of the surface settlement of the mining area was obtained.Mainly conducted the following research:(1)First,use SBAS-InSAR technology to process 41 scenes sentienl-1A data covering a coal mine in northern Shaanxi,including image registration,filtering,unwrapping,and extraction of high-coherence points for regression analysis to separate deformation phase and error Phase,and through SVD decomposition and least square method to extract the time series cumulative settlement from the LOS to the mining area,obtain the settlement information of the stable boundary point on the edge of the subsidence basin in the mining area,and select the two GPS observation stations that coincide with the direction and inclination The accuracy of the boundary points was verified.It is found that the maximum residual error of the strike boundary point B12 is 3mm,and the maximum residual error of the tendency boundary point B14 is 6mm,both of which are within the tolerance range,indicating that the results of the boundary point of the subsidence basin are trustworthy.(2)Because the static probability integration method cannot predict the dynamic mining subsidence of the mining area over time,this paper proposes a dynamic time probability integration prediction model that combines the improved power exponent knothe time function model with the static probability integration method.It is often difficult to obtain the model parameters of the static probability integration method.In this paper,the static probability integration model is combined with the boundary point deformation information obtained from the InSAR side view direction to establish the fitness function,and then the differential evolution gray wolf optimization algorithm is used to optimize the probability of the mining area.The integral method predicts the parameters to be calculated.According to the LOS deformation of the boundary point time series obtained by InSAR,the redundant observation equation is established,and the model parameters of the improved knothe time function are obtained by the least square method.Thus,the InSAR-ITPIM dynamic probability integral prediction model is established to predict the surface subsidence during the mining process,and the predicted results are compared with the actual GPS results for verification.(3)Combining the geological structure and rock layer information of the mining area,establish a Moore-Coulomb model to solve the unbalanced force through FLAC3D software,and select the excavation distance according to the stoping distance.Thus,the mining subsidence process of the working face is dynamically simulated,and the dynamic surface subsidence information of the working face is obtained.The research results show that the predicted result of InSAR-ITPIM dynamic settlement is better than the actual GPS settlement.The maximum average error reaches 0.111m,and the maximum root mean square error reaches 0.135m.The error is within the tolerance range,which has certain practical application value.The settlement law of FLAC3D numerical simulation is consistent with that of InSAR-ITPIM,which shows that FLAC3D numerical simulation has certain reference value in dynamic settlement simulation of mining subsidence,and can provide theoretical basis for post-mining treatment and surface subsidence simulation of mining area.