为了提高点云数据处理的精度和三维精细化建模的效率，利用三维激光扫描技术获取变电站的点云数据，提出一种电力设备分类建模的思路。首先，采用“V”字型拼接标靶组配合扫描仪获取变电站点云数据；其次，根据设备结构特点分为线性设备、规则设备和复杂设备3种类型，线性设备通过分离点云、提取关键点和拟合多段线等进行批量自动化建模，规则设备通过构建标准模型库提高模型复用率，最终完成变电站精细化模型构建。结果表明合理的标靶布局（“V”字型）能够有效地提高点云数据的拼接精度和效率；所提出的三维精细化建模思路，线性设备建模效率提升80%以上，规则设备建模效率提升30%左右，整体上提升了变电站三维精细化建模的效率，这为实现电网运检智能化提供数据基础。

In order to improve the accuracy of point cloud data processing and the efficiency of 3D fine modeling, we used 3D laser scanning technology to obtain point cloud data from substations and proposed a power equipment classification modeling method. Firstly, we used a V-shaped splicing target group in conjunction with a scanner to obtain substation point cloud data. Then, according to the structural characteristics of equipment, we divided it into linear equipment, regular equipment, and complex equipment three types. Linear equipment performs batch automated modeling by separating point clouds, extracting key points, and fitting polylines. Regular equipment improves model reuse rate by building a standard model library. Finally, we completed the refined model construction of substations. The results show that a reasonable target layout(V-shaped) can effectively improve the accuracy and efficiency of point cloud data stitching. The proposed 3D refined modeling method improves the efficiency of linear equipment modeling by over 80%, while regular equipment modeling efficiency by about 30%.Overall, it improves the efficiency of 3D refined modeling in substations, providing a data foundation for achieving intelligent power grid operation and inspection.