Extraction And Reconstruction Of Multi Split Power Lines From Airborne Laser Data

For the inspection work of power corridor system,it is necessary to rely on employees on the ground to inspect the scene in person,sometimes with the help of helicopters carrying optical cameras.However,using these methods will consume a lot of manpower resources,material resources and financial resources.Li DAR(Light Detection and Ranging)can obtain the spatial position information of ground objects accurately,and the development of unmanned aerial vehicle reduces the cost of flying inspections.Therefore,Li DAR has become the most cost-effective working equipment for power corridor inspection by virtue of its many advantages.In order to improve the efficiency of power inspection and reduce the consumption of required resources,airborne Li DAR is considered to be a good choice to scan the target scene.First of all,the experiment uses the airborne Li DAR system to obtain the 3d point cloud data of the power line corridor,and according to the property of LIDAR data,the characteristics of different ground objects in the corridor are mathematically defined.Then the effective multiclass relevance vector machine algorithm is used to classify the point cloud,and the defined characteristic data is taken as the input,and the output data is the power-line point cloud information.Finally,the curve fitting of the point cloud extracted from the power line is carried out.The fitting results can provide a mathematical basis for power line fault diagnosis(such as wind analysis,snow and ice cover,etc.),and have good engineering practice value.Aiming at the problem of accurate reconstruction of high-voltage power lines,a new high precision 3D reconstruction method for multi-split power lines(six splits,eight splits)in Li DAR point cloud data is proposed in this paper.Firstly,according to the algorithm proposed in this paper,the integral split power line point cloud is fitted on the XOY plane and the catenary on the XOZ plane respectively,and the curve equation obtained is taken as the center line,then the coordinates of each point relative to the center line are calculated,K-means clustering method is used to cluster the relative coordinates to extract the sub-wire points contained in the split power line.Finally,each sub-conductor point cloud is fitted on XOY and XOZ planes with the same method to get the final equation of the curve.The experimental results show that the reconstruction algorithm proposed in this paper has good robustness and can reconstruct multi-split power lines more accurately.