Self-Explosion Defect Detection Method of Glass Insulator Based on YOLOv4

As an important part of transmission lines, insulators play an important role in ensuring the safe operation of power systems. The faults caused by the self-explosion of insulators account for a large proportion of the transmission line faults. If they are not checked in time and remedial measures are taken, serious consequences may occur. In response to this problem, this paper proposes a detection method for self-explosion of glass insulators based on YOLOv4. YOLOv4 is a single-stage target detection algorithm, which consists of an input terminal, a reference network, a Neck network and a Head output terminal. The YOLOv4 algorithm extracts the features of the inputted glass insulator self-explosion image data, so as to realize the identification and location of the self-explosive insulator. The average accuracy of the best model can reach 89.00%. The method proposed in this paper can meet the real-time detection requirements for self-explosive insulators in the field inspection in terms of speed and accuracy, and has a good application prospect.

I. Introduction

Today, smart grids have become a new trend in the development of the world’s power grids[1]. The development of smart grids in my country has gradually formed a basic pattern of “West-to-East power transmission” with UHV as the backbone power grid, which not only has long transmission distances but also has harsh environments along the way. As an important part of transmission lines, insulators play an important role in ensuring the safe operation of power systems. The faults caused by the self-explosion of insulators account for a large proportion of the transmission line faults. If they are not checked in time and emedial measures are taken, serious consequences may occur[2]. Therefore, timely discovery of the self-explosion defect of insulators is of great significance to ensure the safety of transmission lines and improve the reliability of the power grid[3].