~高强度煤炭开采导致地表被破坏、生态环境逐渐恶化，违背了矿山绿色开采的原则，极大地制约了矿区可持续发展。因此，加强矿区因煤炭资源开采引起的地表形变监测对于矿区生态环境修复和灾害防治具有十分重要的意义。无人机LiDAR（Light Detection and Ranging，LiDAR）能够快速采集大范围三维地理信息，凭借其高效、快速、经济的优势，在资源保护、线路巡查、测绘地理信息等方面得到广泛应用。但由于LiDAR点云滤波算法自适应性较差，难以稳健地获取地面点云，导致生成的数字高程模型（Digital Elevation Model，DEM）精度难以满足开采沉陷监测需求，其在矿山开采沉陷监测中还存在着一定的局限性。本文面向“榆神矿区浅埋深群煤开采岩移观测的研究及应用项目”企业合作项目拟解决的关键问题，以凉水井煤矿某工作面为研究对象，将无人机LiDAR技术应用于矿山开采沉陷监测，研究内容如下：
（1）从点云误差、点云剖面检验和目视判别等三个方面对比分析了当前主流的点云滤波算法。针对榆神矿区多低矮沙生植被、多风沙丘陵等地形特征，改进了渐进三角网加密滤波算法，使改进算法的点云I、II类误差达到0.76%和2.01%，减少了原算法通常会将陡坡误分为非地面点的现象，有效地改善了点云滤波算法的精度。
（2）探究DEM格网尺寸和点云密度对DEM精度的影响规律。通过实验分析，获取了本研究区内及相似地形下点云插值生成DEM时，格网尺寸选择0.1m，地面点云密度控制在每平方米90~100个点会具有较好的精度。
（3）根据地面高精度DEM，获取无人机LiDAR开采沉陷面域监测成果，并通过水准观测数据进行精度验证，分析其误差规律；针对研究区4-3煤层上覆4-2煤层的情况进行多煤层开采沉陷的数值模拟，分析下沉结果，探究多煤层重复采动下沉规律。结果表明：基于无人机LiDAR获取的DEM，其均方根误差（Root Mean Square Error，RMSE）为0.264m；通过DEM叠加做差获取研究区地表下沉量，由于系统误差的消除，地表下沉量的RMSE提高到0.085m，说明该技术在浅埋大采高煤层的开采沉陷监测中具有一定的适用性；根据数值模拟结果推断：4-2煤层虽然在开采完毕后经历相当长时间塌陷，但仍可能存在一定的裂隙和空洞，上覆岩层再短期内很难达到稳定状态，4-3煤层的开采对4-2煤层上覆岩层造成进一步扰动，引起4-3煤层开采过程中地表平均下沉值大于平均采高，说明多煤层开采对上覆岩层造成的破坏进一步加剧，下沉量更大、破坏性更强，需加强多煤层开采过程中对地表建、构筑物的防护。
 

~High-intensity coal mining leads to the destruction of the surface and the gradual deterioration of the ecological environment, which violates the principle of green mining and greatly restricts the sustainable development of the mining area. Therefore, it is of great significance to strengthen the monitoring of surface deformation caused by coal mining in mining areas for ecological environment restoration and disaster prevention in mining areas. Unmanned Aerial Vehicle (UAV) Light Detection and Ranging (LiDAR) can quickly collect geographic information in a wide range of three-dimensional space. With its advantages of high efficiency, rapidity and economy, it is widely used in resource protection, communication line inspection, surveying and mapping geographic information and so on. Because of the poor adaptability of LiDAR point cloud filtering algorithm, it is difficult to obtain the ground point cloud robustly, which leads to the poor accuracy of the generated ground Digital Elevation Model (DEM), and it still has certain limitations in mining subsidence monitoring. Combined with the " Research and Application Project of Rock Movement Observation in Shallow Buried Deep Group Mining in Yushen Mining Area ", taking a working face of Liangshuijing Coal Mine as the research area, UAV LiDAR technology is applied to mining Subsidence monitoring, the research contents are as follows:
(1) The current mainstream point cloud filtering algorithms are compared and analyzed from three aspects: point cloud error, point cloud profile inspection and visual discrimination. Aiming at the terrain features such as low sandy vegetation and sandy hills in Yushen mining area, the progressive triangulation encryption filtering algorithm is improved, and the errors of point cloud I and II after improvement are 0.76% and 2.01% respectively, which reduces the phenomenon that steep slopes are usually mistakenly divided into non-ground points in the original algorithm and effectively improves the accuracy of point cloud filtering algorithm.
(2) The influence of DEM grid size and point cloud density on DEM accuracy is studied. Through experimental analysis, it is believed that in this study area and similar terrain， the optimal grid size of DEM generated by point cloud interpolation is 0.1m, and the ground point cloud density should be controlled at 90 ~ 100 points per square meter.
(3) According to the high-precision DEM on the ground, the UAV LiDAR area monitoring results are calculated and the accuracy is verified by the leveling data, and the error rule is analyzed; for the situation of the 4-2 coal seam overlying the 4-3 coal seam of the working face in the study area, the numerical simulation of the multi-coal seam mining subsidence is carried out, the subsidence results are analyzed, and the subsidence law of the multi-coal seam repeated mining is explored. The results show that the RMSE of the single-point measurement by UAV LiDAR is 0.264m; The surface subsidence in the study area is obtained by DEM stacking and difference. Due to the elimination of systematic errors, the RMSE of the surface subsidence is increased to 0.085m. It indicates that this technology has certain applicability in mining subsidence monitoring of shallow buried coal seam and large mining height coal seam; The numerical simulation results show that although the 4-2 coal seam collapsed for a long time after mining, there are still a lot of cracks and cavities. It is difficult for the overlying strata to reach a stable state in a short period of time, the mining of the 4-3 coal seam caused further disturbance to the overlying strata of the 4-2 coal seam, causing the average surface subsidence value to be greater than the average mining height during the mining process of the 4-3 coal seam, indicating that the damage to the overlying strata caused by multi-coal seam mining is further intensified, the subsidence is larger, and the destructibility is stronger, and it is necessary to strengthen the protection of the surface during multi-coal seam mining.
 

[1] 杨昊睿,宁树正,丁恋, 等. 新时期我国煤炭产业现状及对策研究[J]. 中国煤炭地质, 2021, 33(S1): 44-48.

[2] 袁亮. 煤炭精准开采科学构想[J]. 煤炭学报, 2017, 42(01): 1-7.

[3] 孙杰, 陈美英, 唐朝苗, 等. 我国煤炭资源勘查现状跟踪研究[J]. 中国煤炭地质, 2017(11): 5-12.

[4] 李凤明. 我国采煤沉陷区治理技术现状及发展趋势[J]. 煤矿开采, 2011, 016(003): 8-10.

[5] 孙华. 浅析我国采煤沉陷区的分布与治理情况[J]. 华北自然资源, 2020(06): 114-115.

[6] 李树志. 我国采煤沉陷区治理实践与对策分析[J]. 煤炭科学技术, 2019, 047(001): 36-43.

[7] 胡炳南, 郭文砚. 我国采煤沉陷区现状、综合治理模式及治理建议[J]. 煤矿开采, 2018, 023(002): 1-4.

[8] 陈延涛. 凉水井煤矿浅埋煤层综采工作面顶板运动及矿压显现规律研究[D]. 北京：中国矿业大学, 2016.

[9] Nong Z, Dong-jiang P, Yi-ming Z, et al. Evaluation of relative mining intensity in western China based on interval analytic hierarchy process[J]. Electronic Journal of Geotechnical Engineering, 2014, 19: 2941-2953.

[10] Wu Z, Xia T, Nie J, et al. The shallow strata structure and soil water content in a coal mining subsidence area detected by GPR and borehole data[J]. Environmental Earth Sciences, 2020, 79(22): 1-13.

[11] Jian W , Peng X , Hui X C . Coal mining GPS subsidence monitoring technology and its application[J]. Mining Science and Technology (China), 2011, 21(004): 463-467.

[12] 刘晓磊. 西部煤炭绿色开发地质保障技术研究现状与发展趋势[J]. 内蒙古煤炭经济, 2020(15): 181-182.

[13] 芦家欣. 基于机载LiDAR点云的榆神矿区采煤沉陷建模[D]. 西安：西安科技大学, 2020.

[14] 汪云甲. 矿区生态扰动监测研究进展与展望[J]. 测绘学报, 2017, 46(10): 1705-1716.

[15] Du Y , Lei Z , Feng G , et al. On the Accuracy of Topographic Residuals Retrieved by MTInSAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, PP(99): 1-13.

[16] Casu F, Manzo M, Lanari R. A quantitative assessment of the SBAS algorithm performance for surface deformation retrieval from DInSAR data[J]. Remote Sensing of Environment, 2006, 102(3-4): 195-210.

[17] Zhi W L , Yang Z F , Jian J Z , et al. Retrieving three-dimensional displacement fields of mining areas from a single InSAR pair[J]. Journal of Geodesy, 2015, 89(1): 17-32.

[18] 牛峰明. 机载LiDAR矿区沉陷信息提取方法研究[D]. 焦作: 河南理工大学, 2016.

[19] Ren H , Zhao Y , Xiao W , et al. A review of UAV monitoring in mining areas: current status and future perspectives[J]. 国际煤炭科学技术学报：英文版, 2019, 006(003): P.320-333.

[20] 冯增文. 基于多期DEM的地质灾害与环境动态监测[D]. 北京: 中国地质大学(北京), 2015.

[21] 张永庭,徐友宁,梁伟, 等. 基于无人机载LiDAR的采煤沉陷监测技术方法——以宁东煤矿基地马连台煤矿为例[J]. 地质通报, 2018, 37(12): 2270-2277.

[22] 袁豹. 基于平面约束的机载LiDAR航带平差方法研究[J]. 测绘工程, 2015, 24(10): 27-30.

[23] 陈金磊,许文学. 机载LiDAR航带拼接方法分析[J]. 测绘通报, 2019(S2): 112-116.

[24] Guo Q , Li W , Yu H , et al. Effects of Topographic Variability and Lidar Sampling Density on Several DEM Interpolation Methods[J]. Photogrammetric Engineering & Remote Sensing, 2010, 76(6): 701-712.

[25] 张靖. 基于克里金算法的点云数据插值研究[D]. 西安: 长安大学, 2014.

[26] 李朋飞,张晓晨,严露, 等. 复杂地形中机载LiDAR点云构建DEM的插值算法对比[J].农业工程学报, 2021, 37(15): 146-153.

[27] 高原,朱娅男,陈传法, 等. 高精度DEM建模的加权径向基函数插值方法[J/OL]. 武汉大学学报(信息科学版): 1-9[2022-04-10].

[28] Vosselman G. Slope based filtering of laser altimetry data[J]. International archives of photogrammetry and remote sensing, 2000, 33(B3/2; PART 3): 935-942.

[29] Wang M, Tseng Y H. Automatic segmentation of LiDAR data into coplanar point clusters using an octree-based split-and-merge algorithm[J]. Photogrammetric Engineering & Remote Sensing, 2010, 76(4): 407-420.

[30] 靳生洪,杨鸿海,王莲玉. 基于格网化LIDAR点云数据坡度滤波方法的研究[J]. 测绘与空间地理信息, 2013, 36(06): 154-156.

[31] 何培培,万幼川,黄桂平, 等. 城区LiDAR点云自适应坡度的滤波算法[J]. 遥感信息, 2017, 32(05): 62-68.

[32] 赵明君,刘超,高翔, 等. 基于多尺度网格的点云自适应坡度滤波算法[J]. 合肥工业大学学报(自然科学版), 2021, 44(02): 266-272.

[33] Susaki J. Adaptive slope filtering of airborne LiDAR data in urban areas for digital terrain model (DTM) generation[J]. Remote Sensing, 2012, 4(6): 1804-1819.

[34] 汪文琪,李宗春,付永健, 等. 一种多尺度自适应点云坡度滤波算法[J/OL]. 武汉大学学报(信息科学版): 1-8[2021-10-12].

[35] Kilian J, Haala N, Englich M. Capture and evaluation of airborne laser scanner data[J]. International Archives of Photogrammetry and Remote Sensing, 1996, 31: 383-388.

[36] Zhang K, Chen S C, Whitman D, et al. A progressive morphological filter for removing nonground measurements from airborne LIDAR data[J]. IEEE transactions on geoscience and remote sensing, 2003, 41(4): 872-882.

[37] Chen Q. Improvement of the Edge‐based Morphological (EM) method for lidar data filtering[J]. International Journal of Remote Sensing, 2009, 30(4): 1069-1074.

[38] 李峰,崔希民,袁德宝, 等. 改进坡度的LiDAR点云形态学滤波算法[J]. 大地测量与地球动力学, 2012, 32(05): 128-132.

[39] Hui Z , Hu Y , Yao Y , et al. An Improved Morphological Algorithm for Filtering Airborne LiDAR Point Cloud Based on Multi-Level Kriging Interpolation[J]. Remote Sensing, 2016, 8(5): 1-16.

[40] 惠振阳, 程朋根, 官云兰, 等. 机载LiDAR点云滤波综述[J]. 激光与光电子学进展, 2018, 55(006): 1-9.

[41] 赖旭东,万幼川. 基于平坦度的激光雷达强度图像的滤波算法[J]. 中国激光, 2005(10): 23-27.

[42] Goepfert J, Soergel U, Brzank A. Integration of intensity information and echo distribution in the filtering process of LIDAR data in vegetated areas[J]. Proceedings of the Silve Laser, 2008.

[43] 李伟. 基于回波次数的点云强度滤波方法在处理LiDAR数据中的研究与应用[J]. 经纬天地, 2019(03): 61-65.

[44] 解全波,田茂义,冯成凯, 等. 结合多波束点云强度和高程信息滤波算法研究[J]. 海洋测绘, 2021, 41(01): 65-69.

[45] 张继贤, 林祥国, 梁欣廉. 点云信息提取研究进展和展望[J]. 测绘学报, 2017, 46(10): 10.

[46] 林祥国, 张继贤, 宁晓刚, 等. 融合点,对象,关键点等3种基元的点云滤波方法[J]. 测绘学报, 2016, 45(11): 10.

[47] Yang B , Huang R , Dong Z , et al. Two-step adaptive extraction method for ground points and breaklines from lidar point clouds[J]. Isprs Journal of Photogrammetry & Remote Sensing, 2016, 119(sep.): 373-389.

[48] 胡海瑛,惠振阳,李娜. 基于多基元特征向量融合的机载LiDAR点云分类[J]. 中国激光, 2020, 47(08): 237-247.

[49] Farabet, Clement, Couprie, et al. Learning Hierarchical Features for Scene Labeling[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 35(8): 1915-1929.

[50] 钱鸣高,缪协兴,许家林. 岩层控制中的关键层理论研究[J]. 煤炭学报, 1996(03): 2-7.

[51] 吴立新,王金庄. 连续大面积开采托板控制岩层变形模式的研究[J]. 煤炭学报, 1994(03): 233-242.

[52] 麻凤海,丁彧. 大倾角多煤层开采地表移动规律的数值模拟研究[J].中国矿业, 2009, 18(06): 71-73.

[53] 张杰,侯忠杰. 浅埋煤层非坚硬顶板强制放顶实验研究[J]. 煤田地质与勘探, 2005(02): 15-17.

[54] 甘智慧,尚慧,杜荣军, 等. 基于FLAC~(3D)和DEM数据的缓倾斜煤层开采沉陷分析[J]. 煤田地质与勘探, 2021, 49(03): 158-166.

[55] 王永国,王明,许蓬. 巴彦高勒煤矿3-1煤层顶板垮落裂缝带发育特征[J]. 煤田地质与勘探, 2019, 47(S1): 37-42.

[56] 王永国,王明,刘增平, 等. 巴彦高勒煤矿多相变沉积条件下煤层顶板含水层富水性特征[J]. 煤田地质与勘探, 2019, 47(S1): 56-61.

[57] 孙学阳,张齐,李成, 等. 陕北某矿双煤层开采对覆岩影响的模拟对比[J]. 煤田地质与勘探, 2020, 48(04): 183-189.

[58] 杜子龙,仪梦婷. 基于FLAC3D的深部开采沉陷预计数值模拟[J]. 测绘与空间地理信息, 2021, 44(07): 182-185.

[59] Zhang W, Qi J, Wan P, et al. An easy-to-use airborne LiDAR data filtering method based on cloth simulation[J]. Remote sensing, 2016, 8(6): 501.

[60] 李卉,李德仁,黄先锋, 等.一种渐进加密三角网LIDAR点云滤波的改进算法[J]. 测绘科学, 2009, 34(03): 39-40+216.

[61] 柳红凯,徐昌荣,徐晓. 基于渐进加密三角网机载LIDAR点云滤波改进算法研究[J]. 赣州：江西理工大学学报, 2016, 37(03): 50-55+60.

[62] 刘洋,习晓环,王成, 等. 一种改进的渐进加密三角网点云滤波算法[J]. 测绘科学, 2020, 45(05): 106-111+125.

[63] 胡孟琦. 基于表面拟合的多尺度区域生长点云滤波方法[D]. 武汉: 武汉大学,2020.

[64] 李一凡,刘成洲,崔腾飞, 等. 煤矿开采沉陷预计过程的FLAC~(3D)数值模拟研究[J].北京测绘, 2018, 32(10): 1156-1160.

[65] 闫功力,孟凡超. 测绘新技术在矿山测量中的应用[J]. 北京测绘, 2012(04): 100-102.

[66] 杜伟. 倾斜岩层巷道围岩变形破坏过程的Flac~(3D)数值模拟研究[D]. 绍兴: 绍兴文理学院, 2019.

[67] 刘剑,郝兵元,黄辉. 大采宽工作面开采沉陷规律的FLAC~(3D)模拟研究[J]. 煤炭技术, 2014, 33(11): 132-134.

[68] 张天军,羽玥,张磊, 等. 矿山开采沉陷的FLAC~(3D)数值模拟分析[J].煤炭技术, 2018, 37(02): 11-14.

[69] 陈晓祥,谢文兵. 采矿过程数值模拟模型左右边界的确定[J]. 煤炭科学技术, 2007(04): 96-99+92.

[70] 南英华. 安家岭井工1矿开采沉陷数值模拟岩体力学参数确定[D]. 北京: 中国地质大学(北京), 2012.