A novel line position recognition method in transmission line patrolling with UAV using machine learning algorithms

摘要

Overhead transmission line is the main approach for energy transporting in power systems. To ensure the safety of power system, regular patrolling is required to be carried out. Current UAV patrolling method needs UAV to be controlled by skilled staff on the ground, which brings about low efficiency and high cost. To achieve autonomous line patrolling, magnetic sensors can be placed on the UAV and magnetic field data can be used to determine the relative position between UAV and transmission lines. To realize real-time patrolling, the speed of calculation and line position recognition is expected to be fast, so machine learning algorithm is considered as a good option. Neural network and SVM methods are exploited to achieve line position recognition and the methods are verified in a DC transmission system and a satisfying accuracy is obtained on the randomly picked test set. This method can achieve UAV line patrolling autonomously, which is safer, suitable for various situations and has a broad future prospect.