Research On Automatic Analysis Of Hidden Danger Of Unmanned Aerial Vehicle Transmission Line Inspection Tree Barrier

With the development of China's economy,the scale of the power grid has gradually expanded,and high-level,long-distance transmission lines are also increasing.However,China's vast territory and complex terrain make it very difficult to manage transmission lines.Hidden danger of tree barriers is one of the main reasons for the malfunction of current transmission lines.At present,the collection methods of transmission line tree barrier information can be divided into three types according to the operation mode,including artificial tree-based hidden barrier information collection,drone-based hidden barrier information collection and lidar-based hidden barrier information collection.However,the traditional manual power grid inspection management model inevitably has many problems such as poor manual inspection effect,low work efficiency,high labor cost,etc.,which has been unable to adapt to the current society's high efficiency inspection,high workload and fine efficiency of the power grid.Job requirements.With the emergence of drones,more and more technicians and scholars at home and abroad began to study how to detect the distance between trees and transmission lines based on the data collected by drones,so as to further analyze the hidden dangers of tree barriers,but they still exist Many problems,such as low detection accuracy and unstable detection results.At present,there are few studies on detecting hidden obstacles based on point cloud data,and compared with the UAV image tree barrier measurement technology,its cost is expensive,the operation is complicated,and the hidden danger of equipment safety is greater,which requires a professional team to complete.In view of the current industry pain points that rely on manual processing of massive image data and low accuracy in the patrol inspection of unmanned aircraft power barriers,this article obtains a 500 kv "Chu Sui AC" line about 20 km based on the drone intelligent flight software developed by the research team The image data of the power line corridor is extracted from the point cloud data of the power line corridor by techniques such as air-three solution and dense matching,and the automatic classification of the massive point cloud of the transmission line is studied according to the characteristics of the power line scene;the dichotomy and RANSAC algorithm are further used based on the result of the classification Research the technology of automatic power line fitting and other technologies to realize the analysis of hidden dangers of UAV power tree barriers,and apply the researched analysis algorithm of hidden dangers of electric power barriers to the Kunming,Yunnan 500 kv ultra-high voltage AC line for verification and analysis.The research content of this article mainly includes the following three aspects:(1)UAV power line corridor image acquisition and processing.Reverse technology based on binocular recognition,using "U" type flight mode,by setting parameters such as flight altitude,heading overlap and side overlap,the power lines on both sides of the tower are photographed from different angles,so that the power lines on both sides of the tower can be Displayed in different photos,multiple image data with stereo pairs are obtained.The point cloud data of the power line corridor was obtained from the drone images with stereo pairs through technologies such as space three solution and dense matching.(2)Point cloud classification and power line fitting.Accurate classification of massive point cloud data is the premise and basis for the analysis of hidden dangers of tree barriers.In order to solve the existing point cloud classification method without the phenomenon of point cloud misdivision and omission in complex terrain,complex features and discontinuous surface,this paper is based on the characteristics of pole towers,power lines and vegetation point clouds to study the classification of point cloud and obtain the power line point cloud.Based on dichotomy and RANSAC algorithm,the power line is accurately fitted and modeled.(3)Analysis of hidden dangers of power transmission lines.Based on the extracted power line data,use the section distance to calculate and analyze the distance between the transmission line and the features below it,and automatically compare the distance with the specified safety distance,marking the distance between the transmission line and the feature at the standard safety distance Within the area;after making a preliminary judgment of the tree barrier information according to the distance,the visual verification technology is used to further check the analysis results of the tree barrier hidden dangers,improve the accuracy of the analysis results of the existing tree barrier hidden danger methods,thereby reducing field work Check efficiency.In this paper,the proposed automatic analysis method for hidden troubles in the inspection of unmanned aerial vehicle transmission lines is verified in 500 Kv “Chu Sui Intersection” with the distance of 20 km,and the results are analyzed.The process is shown in detail,and the overall scene is evaluated and analyzed for accuracy.