潘麗娟1，王冬梅2，任旭華1，石一凡2

　　(1.河海大學水利水電學院，江蘇 南京 210098;

　　2.江蘇省水利科學研究院，江蘇 南京 210017)

　　摘 要：針對傳統外業測量水庫庫容估算方法易受地形、氣象條件影響，存在危險性高、效率低等問題，以侖山水庫為研究對象，采用機載LiDAR技術估算水庫面積及水量變化。通過實地數據采集，利用KD-樹算法剔除點云中的粗差點，并采用漸進加密不規則三角網(PTIND)濾波分離出地面點，完成LiDAR點云數據的預處理;根據點云數據建立精細化數字高程模型(DEM)，提取不同水位的水庫水面面積;最終利用積分和傳統棱臺體積估算方法對比分析不同水位水量差值變化。結果表明：與高程實測值相比，DEM高程反演值滿足高程精度要求，基于機載LiDAR的水量估算受庫底坡度變化的影響，估算值低于傳統估算方法的估算值，估算結果更為精確，可為監測水庫水量變化提供科學依據。

　　關鍵詞： LiDAR 技術;水量變化;典型水庫;估算;機載;無人機;DEM ;點云

　　Estimation of water storage variations in typical reservoirs based on airborne LiDAR technology

　　PAN Lijuan1，WANG Dongmei2，REN Xuhua1，SHI Yifan2

　　(1.College of Water Conservancy & Hydropower Engineering，Hohai University，Nanjing 210098，China;

　　2. Jiangsu Hydraulic Research Institute，Nanjing 210017，China)

　　Abstract：To overcome the deficiency of traditional reservoir monitoring methods due to terrain and climate, this paper applies airborne LiDAR technology to Lunshan reservoir and develops Digital Elevation Models (DEMs) to calculate the changes of reservoir area and water storage. Based on the collected data, the KD-Tree algorithm and Progressive Triangulated Irregular Network Densification (PTIND) are applied to eliminate laser data noise and isolate the ground points. A refined DEM is developed based on pre-processed point cloud data to extract water surface area at different water levels of the reservoir. The variations of water storage at different water levels are estimated and compared by using integral calculus and conventional algorithm. The results show that the inversion values of the DEM meet the accuracy requirements compared with the measured elevation. The estimation of water storage based on airborne LiDAR is impacted by the change of reservoir bottom slope. Compared to traditional observation method, the LiDAR-estimated value is much lower, thus the estimation result is more accurate. This research can provide scientific reference for monitoring the variation of reservoir storage.

　　Key words: LiDAR technology;water storage variations;typical reservoir;estimation;airborne;drone;DEM;point cloud