Research On Multi-scale Analysis And Deformation Law Of Mining Area Surface Subsidence Based On Wavelet Transform

In the mining area subsidence monitoring work,due to the influence of natural conditions,data processing methods and other factors,the use of unmanned aerial vehicle photogrammetry to monitor the mining area surface deformation will lead to the lack of information expression integrity and low accuracy of the occurrence of surface subsidence,which will affect the research and analysis of the mining area surface deformation law and subsidence prediction.In this thesis,the subsidence monitoring data of a mine in Inner Mongolia obtained based on the UAV photogrammetry method are studied to improve the integrity and accuracy of the expression of the surface subsidence information,and the analysis of the surface deformation law is carried out.The research shows that the application of terrain analysis method and wavelet multi-scale analysis method to the processing of the mining area surface subsidence data can improve the integrity and accuracy of the expression of the surface subsidence information,After the subsidence prediction and deformation law analysis,it can provide guidance for people to accurately understand the areal subsidence basin,optimize the prediction model,and provide guidance for the protection of surface buildings in the subsidence area of the mining area and similar mining areas and the restoration of the ecological environment of the mining area.The main research results of this thesis are as follows:(1)The DEM of the two phases of ground point clouds is constructed by local polynomial interpolation,and the subsidence basins with different resolutions are generated by superposition.Then,the slope,surface roughness and topographic relief of the subsidence basins with different resolutions are extracted,statistically analyzed and their variation rules are analyzed.It is concluded that the statistical indicators of topographic factors that have the best correlation with the resolution of the subsidence basins are the mean slope,the maximum roughness and the mean topographic relief;Through comprehensive analysis of the change trend of the statistical index of the topographic factors with the best correlation with the resolution,it is concluded that the best resolution of the subsidence basin is 4m.(2)The wavelet multi-scale decomposition and threshold de-noising of the 4m resolution subsidence basin show that the root mean square error of the initial subsidence basin is 117 mm,the abnormal subsidence information of the subsidence basin after wavelet processing has been effectively eliminated,the surface fragmentation degree and dispersion degree have been greatly weakened,and the precision is the best when the number of decomposition layers is 4,the root mean square error is 83 mm,which is 29% higher than the original subsidence basin.(3)The comparative analysis between the measured surface subsidence and the predicted subsidence by probability integral shows that the measured surface subsidence will not occur in the flat bottom area as obvious as the predicted subsidence in the fully mined area.The main reasons are that the coal seams with equal thickness are selected when predicting,while the actual coal seams are of unequal thickness;The terrain change where the subsidence area is located will affect the surface subsidence to a certain extent,and the probability integral model cannot take this factor into account;The overlying strata usually have geological structures such as faults of different sizes that are difficult to consider in probability integral prediction.(4)The comparison and analysis of the measured and predicted subsidence deformation,such as inclination and horizontal deformation,show that the measured values are scattered in the area where the predicted strike direction inclination and horizontal deformation are zero,and they are quite different from the predicted values;In a large number of areas where the horizontal deformation is expected to be zero,the measured horizontal deformation that can cause Class IV damage to surface buildings has occurred.The main reasons for this phenomenon are: changes in the thickness and dip angle of coal seams,topographic conditions,and the structure of rock strata that can be and are difficult to be detected.