机载激光点云构建西部矿区开采沉陷模型研究

针对榆神矿区地形起伏而植被覆盖度较低的条件,采用无人机激光雷达(LiDAR)对采煤沉陷盆地进行地面扫描和建模,按现有的点云滤波和插值算法所构建的沉陷模型精度不足的问题,该文提出基于格网高程排序的类地面点云提取算法,通过改变格网大小和点云提取比例来优化点云滤波效果,改进了地表沉陷模型的精度。在此基础上,选用概率积分法和GaussAmp函数对沉陷模型进行最小二乘拟合,结果表明GaussAmp函数对于试验区地表动态沉陷模型的拟合精度更高。

Model research on mining subsidence of western mining area based on airborne laser point cloud

Aiming at the conditions of undulating terrain and low vegetation coverage in Yushen Mining Area,this paper uses unmanned aerial vehicle(UAV)light detection and ranging(LiDAR)technology to scan and model the coal mining subsidence basin,and finds the accuracy of the subsidence model constructed by the existing point cloud filter and interpolation insufficient.For this reason,a ground-like point cloud extraction algorithm based on grid elevation sorting was proposed.The grid filtering effect was optimized by changing the grid size and point cloud extraction ratio,and the accuracy of the surface subsidence model was improved.On this basis,the probability integration method and the GaussAmp function were used to perform a least squares fit on the subsidence model.The results showed that the GaussAmp function had a higher accuracy for the dynamic subsidence model of the ground surface in the experimental area.