Robust Parsing For Elementes Of Electric Power Transmission Corridor From Point Clouds

Affected by the distribution of resources and the development of clean energy,the state has accelerated the construction of high-voltage,extra-high voltage and ultra-high voltage transmission lines which is the key role of power transmission,so the automatic analysis of the spatial relationship of power elements in corridor channel is a necessary means to improve the efficiency and quality of power operation and maintenance.In recent years,a variety of research on transmission corridor,such as three-dimensional modeling,risk detection and so on,have emerged gradually.The realization basis of these research comes from the analysis of relevant power elements.It can be seen that the robust analysis of the central line tower elements in corridor space is of great significance and engineering application value.This paper focuses on the segmentation and identification of tower and power line elements in the transmission corridor.Firstly,a small number of outliers in airborne LiDAR scanning data are eliminated by filtering method,and the number of point clouds is reduced by using down sampling to improve the efficiency of the subsequent algorithm.Secondly,spatial grid is used to reduce the influence of terrain meandering on the analytical algorithm.Next,the transmission corridor environment is divided into tower environment and power line environment for element analysis.Based on the analysis of the spatial structure characteristics of transmission corridor,the pre extraction of tower elements is realized based on the point cloud height difference characteristics and continuity characteristics;the plane center coordinates of tower are determined by hierarchical slicing and k-means clustering,and the optimization model is constructed by combining the single transverse length of tower.The results of applying the optimization model to the pre extraction of tower elements are as follows:first,the optimization identification of tower elements is realized.Based on the extraction results of tower elements,the transmission corridor with broken line distribution is divided into multiple linear single file data by region segmentation,which reduces the difficulty of data processing.Then,by analyzing the spatial structure characteristics of the power line point cloud,a reverse search algorithm is proposed to lock the grid space where the power line is located,which effectively avoids the influence of the ground feature points on the extraction accuracy,and then the FD-DBSCAN algorithm with fixed search direction is proposed,which can efficiently and accurately identify a single power line from the power line set.The accuracy and robustness of the algorithm for power element analysis in different terrain and different tower types are verified by experimental simulation of multiple sets of transmission corridor data in different environments.