Mining Subsidence Modeling Based On Airborne LiDAR Point Cloud In Yushen Mining Area

Yushen mining area is an important coal production base in China,the extensive surface subsidence here caused by coal mining is becoming more and more serious.In the subsidence monitoring of coal mine areas,conventional geodesy,InS AR and other remote sensing methods have certain limitations.As an important technical means to collect large-scale and continuously distributed three-dimensional geospatial information,unmanned aerial vehicle LiDAR can quickly acquire the surface subsidence characteristics of mining areas through the superposition of multi-period data.However,the subsidence model based on the existing mainstream point cloud filtering and interpolation algorithms often contains significant noises,which can limit the practical application of this technology in mining areas.The surface of a mining working face in Yushen mining area was taken as the experimental area.In view of its topographic fluctuation and low vegetation coverage geographical environment,unmanned aerial vehicle LiDAR was used to obtain four sets of ground point cloud data in two phases,and error analysis and model improvement were carried out in point cloud filtering,interpolation,superposition and subsidence modeling.The main research contents and results are as follows:(1)The applicabilities of several main point cloud interpolation algorithms in the geographical environment of Yushen mining area were compared and analyzed.Through the sample point verification of interpolating DEM,combined with the error statistics results,the interpolation effects of different algorithms were analyzed from interpolation accuracy and efficiency.The results show that Authorized Digital Elevation Model has the highest interpolation accuracy and the fast processing speed.(2)The effects of multiple point cloud filtering algorithms on terrain modeling in the geographical environment of Yushen mining area were studied.The filtering result of each algorithm was interpolated by the above optimal algorithm to construct DEM,and the applicabilities of different algorithms in the geographical environment of the research area were analyzed by combining model error,algorithm operating efficiency and algorithm principle.The results show that the accuracy of DEM generated by different filtering algorithms is obviously different,and the effect of Progressive Triangulated Irregular Network Densification Filtering is the best.(3)The error sources and characteristics of the initial subsidence DEM of mining area were systematically analyzed.The digital elevation models generated by the above optimal filtering and interpolation algorithm were superimposed to obtain the initial subsidence DEM of the mining area and error analysis was carried out.The results show that the initial subsidence DEM errors are mainly derived from the small space migration between the point cloud data from different periods,the water area and other scanning conditions in subsidence area,the noise caused by non-ground points after filtering,and point cloud interpolation error.(4)Based on the morphology and error characteristics of surface subsidence DEM in mining area,a wavelet threshold denoising method with reference to subsidence boundary was proposed.The results were verified from three aspects:visualization of denoising results,comparison with measured data and analysis of main section of subsidence.The results show that the smoothness of subsidence DEM is significantly improved by denoising,and after removing all kinds of gross error points,the root-mean-square errors of two groups of denoising results at measured points are all less than 0.05m,and the error points within 10mm increased to 15%.(5)According to the distribution characteristics of subsidence DEM,the hyperbolic function model was used to fit the main section and three-dimensional morphology of subsidence DEM.The results show that the fitting correlation coefficient of the model is higher than 0.99,and the fitting results are reliable.Based on the analysis of slope deformation on the edge of subsidence DEM and the error analysis in the stable region of subsidence DEM,an extraction method of subsidence boundary based on slope analysis of subsidence DEM was proposed,and the calculation and statistics of surface subsidence area and volume are further realized.Based on the subsidence curve of the main section of strike and inclination,the characteristics of surface tilt and curvature were obtained.The results of this paper can provide a feasible technical approach for airborne LiDAR technology to be used in the efficient and accurate monitoring of coal mining subsidence in western mining areas.