基于深度学习的输电线路缺陷检测研究综述

A Review of Transmission Line Defect Detection Research Based on Deep Learning

输电线路是电力系统的重要组成部分,对电力系统的正常运行起到了至关重要的作用。传统的人工巡检已经无法满足电力系统对于输电线路安全可靠运行的要求。随着科技的不断发展,基于深度学习的目标检测算法凭借良好的检测性能被越来越多领域所应用,依靠深度学习的目标检测算法在极大程度上提高了输电线路缺陷检测的检测速度和准确性。首先对输电线路缺陷深度学习目标检测算法的发展历程进行介绍;其次针对输电线路缺陷检测的研究难点进行分析;最后对输电线路缺陷检测的改进思路进行了整理,主要从小目标检测的算法优化、复杂背景下的算法优化、边缘网络轻量化的算法优化三部分进行介绍;并对输电线路缺陷检测进行总结与展望。

Transmission lines are an important part of the power system and play a vital role in the normal operation of the power system. The traditional manual inspection has been unable to meet the requirements of the power system for the safe and reliable operation of transmission lines. With the continuous development of science and technology, the target detection algorithm based on deep learning has been applied in more and more fields by virtue of its good detection performance, and the target detection algorithm relying on deep learning has greatly improved the detection speed and accuracy of transmission line defect detection. Firstly, the development process of deep learning object detection algorithm for transmission line defects is introduced;secondly, the research difficulties of transmission line defect detection are analyzed;finally, the improvement ideas of transmission line defect detection are organized, mainly in three parts: algorithm optimization from small target detection, algorithm optimization in complex context, and algorithm optimization for edge network lightweighting;and the transmission line defect detection is summarized and prospected.