中国电力 ›› 2018, Vol. 51 ›› Issue (11): 139-146.doi: 10.11930/j.issn.1004-9649.201801094

• 信息与通信 • 上一篇    下一篇

基于计算机视觉的架空输电线路机器人巡检技术综述

李振宇1,2, 郭锐1,2, 赖秋频3, 杨军3, 雍民4, 王亮3, 傅思遥3   

  1. 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
  • 收稿日期:2018-01-09 修回日期:2018-03-26 出版日期:2018-11-05 发布日期:2018-11-16
  • 作者简介:李振宇(1990-),男,工程师,从事图像处理与模式识别研究,E-mail:846472189@qq.com
  • 基金资助:
    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)。

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

LI Zhenyu1,2, GUO Rui1,2, LAI Qiupin3, YANG Jun3, YONG Min4, WANG Liang3, FU Siyao3   

  1. 1. Shandong Electric Power Company Research Institute of State Grid of China, Jinan 250003, China;
    2. Shandong Luneng Intelligent Technology Co., Ltd., Jinan 250014, China;
    3. School of Electrical Engineering Wuhan University, Wuhan 430072, China;
    4. Licheng District Power Supply Company of Jinan City of State Grid of China, Jinan 250100, China
  • Received:2018-01-09 Revised:2018-03-26 Online:2018-11-05 Published:2018-11-16
  • Supported by:
    This work is supported by National Natural Science Foundation Project (No.51277135); and Science and Technology Project of SGCC (Research and Application of Key Technologies for Inspection Robots Overhead Transmission Line, No.52600160004).

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。

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

Abstract: Overhead transmission lines are the key to safe operation of power grid and reliable transmission of electric power, and it is very important to carry out regular inspection. The computer vision system has the characteristics of high integration, good interactivity, high degree of automation and fast processing speed, it can play an important role in the accurate identification and fault judgment of all kinds of equipment during the inspection process of overhead transmission lines, therefore, the computer vision system has a wide range of applications in the overhead transmission line robot inspection and fault diagnosis. In this paper, the robot computer vision inspection technology and its research status are summarized from the aspects of wire identification, tower identification and insulator string identification, the corresponding image processing methods are analyzed, on this basis, the overhead transmission line robot visual inspection key technology has been summarized and foreseen.

Key words: computer vision, overhead transmission line, robot inspection technology, machine learning, image processing, device identification

中图分类号: 

  • TM755