基于计算机视觉的架空输电线路机器人巡检技术综述  

Survey of Inspection Technology of Overhead Transmission Line Robot Based on Computer Vision

作　　者：李振宇 郭锐 赖秋频 杨军[3] 雍民 王亮[3] 傅思遥[3] LI Zhenyu;GUO Rui;LAI Qiupin;YANG Jun;YONG Min;WANG Liang;FU Siyao(Shandong Electric Power Company Research Institute of State Grid of China,Jinan 250003,China;Shandong Luneng Intelligent Technology Co.,Ltd.,Jinan 250014,China;School of Electrical Engineering WuhanUniversity,Wuhan 430072,China;Licheng District Power Supply Company of Jinan City of State Grid of China,Jinan 250100,China)

机构地区：[1]国网山东省电力公司电力科学研究院,山东济南250003 [2]山东鲁能智能技术有限公司,山东济南250014 [3]武汉大学电气工程学院,湖北武汉430072 [4]国网济南市历城区供电公司,山东济南250100

出　　处：《中国电力》2018年第11期139-146,共8页Electric Power

基　　金：国家自然科学基金项目(51277135);国家电网公司总部科技项目(架空输电线路全程巡检机器人关键技术研究及应用;520600160004)~~

摘　　要：架空输电线路是电网安全运行和电能可靠输送的关键,对其进行定期巡检至关重要。计算机视觉系统具有集成度高、交互性好、自动化程度高、处理速度快等特点,能够在架空输电线路巡检过程中各类设备的准确识别和故障判断中发挥重要作用,因此计算机视觉系统在架空输电线路机器人巡检与故障诊断中具有广泛的应用前景。从导线的识别、杆塔识别以及绝缘子串的识别等架空输电线路关键设备方面综述了机器人计算机视觉巡检技术及其研究现状,分析了多种相应的图像处理方法,在此基础上对基于机器学习的架空输电线路机器人视觉巡检关键技术进行了总结与展望。Overhead transmission lines are the key to safe operation of power grid and reliable transmission of electric power, and itis very important to carry out regular inspection. The computer vision system has the characteristics of high integration, goodinteractivity, high degree of automation and fast processing speed, it can play an important role in the accurate identification and faultjudgment of all kinds of equipment during the inspection process of overhead transmission lines, therefore, the computer visionsystem has a wide range of applications in the overhead transmission line robot inspection and fault diagnosis. In this paper, the robotcomputer vision inspection technology and its research status are summarized from the aspects of wire identification, toweridentification and insulator string identification, the corresponding image processing methods are analyzed, on this basis, theoverhead transmission line robot visual inspection key technology has been summarized and foreseen.

关 键 词：计算机视觉 架空输电线路 机器人巡检技术 机器学习 图像处理 设备识别 

分 类 号：TM755[电气工程—电力系统及自动化]