Insulator Faults Detection in Aerial Images from High-Voltage Transmission Lines Based on Deep Learning Model

Chuanyang Liu,Huajie Xu,Jingjing Liu,Zuo Sun,Yiquan Wu

doi:10.3390/app11104647

Abstract

Insulator fault detection is one of the essential tasks for high-voltage transmission lines’ intelligent inspection. In this study, a modified model based on You Only Look Once (YOLO) is proposed for detecting insulator faults in aerial images with a complex background. Firstly, aerial images with one fault or multiple faults are collected in diverse scenes, and then a novel dataset is established. Secondly, to increase feature reuse and propagation in the low-resolution feature layers, a Cross Stage Partial Dense YOLO (CSPD-YOLO) model is proposed based on YOLO-v3 and the Cross Stage Partial Network. The feature pyramid network and improved loss function are adopted to the CSPD-YOLO model, improving the accuracy of insulator fault detection. Finally, the proposed CSPD-YOLO model and compared models are trained and tested on the established dataset. The average precision of CSPD-YOLO model is 4.9% and 1.8% higher than that of YOLO-v3 and YOLO-v4, and the running time of CSPD-YOLO (0.011 s) model is slightly longer than that of YOLO-v3 (0.01 s) and YOLO-v4 (0.01 s). Compared with the excellent object detection models YOLO-v3 and YOLO-v4, the experimental results and analysis demonstrate that the proposed CSPD-YOLO model performs better in insulator fault detection from high-voltage transmission lines with a complex background.

Highlights

To solve the above issues, inspired by the work of [38] and You Only Look Once (YOLO) models, this paper proposed a Cross Stage Partial Dense
Compared with most other object detection models, YOLO-v3 has the advantages of faster detection speed and higher detection accuracy, but it still has several challenges when directly applied for insulator detection in aerial images from transmission lines
In order to solve the above issues, on the basis of YOLO-v3 and DenseNet, a Cross Stage Partial Dense YOLOv3 (CSPD-YOLO) model is proposed in this paper

Summary

Introduction

As one of the most common and fault-prone components, the insulator plays an important role in mechanical support and electrical insulation during the operation of power grids [1,2], as shown, where the insulator is encircled by a red rectangular box. The occurrence of insulator faults is diverse and random, and in the event of an insulator fault, it will harm the safety and stable operation of the entire transmission line, and even cause huge economic losses to power grids. In order to ensure the normal operation of power grids, insulator fault detection has become one of primary tasks for transmission lines’ intelligent inspection [4]. The complexity and variability of application scenarios have brought huge challenges to the automatic identification of insulator faults

Methods

Results

Conclusion