Automatic Identification and Defect Diagnosis of Transmission Line Insulators Based on YOLOv3 Network

Abstract

Insulator equipment is an important part of the transmission line of the power grid. It can play a good insulation role among the conductor, the crossbar and the tower. Whether the insulator can work normally directly affects the stable operation of the power grid. To this end, for the transmission line insulator images acquired by drones or robots, an online recognition and defect diagnosis model of transmission line insulators based on YOLOv3 network is proposed. By training YOLOv3 network, the characteristics of various insulators under complex backgrounds are learned and accurately recognized, and combined with particle filter algorithm for defect diagnosis of insulators in various states. The simulation results of the transmission line inspection image show that the proposed automatic insulator identification and defect diagnosis method can quickly and accurately identify the insulator from the transmission line inspection image, and diagnose whether the insulator is damaged and the position of the defect, which is beneficial to improve the transmission line Intelligent inspection level.