一种基于视觉识别定位的变电站电容器检修机器人的设计  

Design of a Substation Capacitor Maintenance Robot Based on Visual Recognition and Localization

作　　者：严宇 刘良燏 廖剑斌 阳应伟 龚杰 樊绍胜[4] 李方 YAN Yu;LIU Liangyu;LIAO Jianbin;YANG Yingwei;GONG Jie;FAN Shaosheng;LI Fang(State Grid Hunan Electric Power Company Extra High Voltage Substation Company,Changsha 410004,China;Substation Intelligent Operation and Inspection Laboratory of State Grid Hunan Electric Power Co.,Ltd.,Changsha 410004,China;School of Power and Mechanical Engineering,Wuhan University,Wuhan 430072,China;School of Electrical and Information Engineering,Changsha University of Science and Technology,Changsha 410114,China;Guangdong Kekaida Intelligent Robot Co.,Ltd.,Foshan 528300,China)

机构地区：[1]国网湖南超高压变电公司,长沙410004 [2]国网湖南超高压变电公司变电智能运检实验室,长沙410004 [3]武汉大学动力与机械学院,武汉430072 [4]长沙理工大学电气与信息工程学院,长沙410114 [5]广东科凯达智能机器人有限公司,广东佛山528300

出　　处：《电力学报》2024年第4期348-356,共9页Journal of Electric Power

摘　　要：为安全高效地实现对变电站内电容器的电容检测,同时完成电容器绝缘子串末端接线柱的螺栓紧固,保证电容器的正常运行,进而维护电网运行安全,设计了一种基于视觉识别定位的变电站电容器检修机器人。结合变电站的整体结构与机器人的作业环境,对机器人作业部分的结构进行了针对性设计,选用合适硬件并构建了控制系统。使用YOLOv5目标识别网络对电容器末端绝缘子串中心进行了检测与三维定位,实现了检测精度与识别速度间的平衡,对绝缘子整体识别率达到95%以上,绝缘子中心定位精度达到2.1 mm。基于识别定位所得位置,移动作业末端可完成电容器电容检测并实现对电容器上绝缘子串末端螺栓紧固作业,完成变电站电容器的检修工作。在根据变电站电容器设置参数构建试验场地,进行了电容器电容检测试验与电容器绝缘子串末端螺栓紧固试验,验证了基于视觉识别定位的电容器检修机器人能有效识别目标并完成检修工作。In order to realize capacitor safety and high efficiency detection of capacitors in substation,complete the bolting of terminal terminals of capacitor insulator string,ensure the normal operation of capacitors,and then maintain the safety of power grid operation,a transformer substation capacitor maintenance robot based on visual recognition and positioning is designed.Combined with the overall structure of the substation and the working environment of the robot,the structure of the working part of the robot is designed,appropriate hardware is used and the control system is built.The YOLOv5 target recognition network is used to detect and position the center of the insulator string at the end of the capacitor,achieving a balance between detection accuracy and identification speed.The overall identification rate of the insulator reached more than 95%,and the positioning accuracy of the insulator center reached 2.1 mm.Based on the position obtained by identification and positioning,the working end is moved to complete the capacitor capacitance detection and realize the function of bolting the end of the insulator string on the capacitor,so as to achieve the overhaul of the capacitor on the substation.An experiment site is constructed according to the capacitor placement parameters in substation,and the capacitor capacitance detection experiment and the capacitor insulator string end bolting experiment are carried out,which verifies that the capacitor overhaul robot based on visual recognition and positioning can effectively identify the target and complete the overhaul work.

关 键 词：变电站 电力电容器 检修机器人 视觉识别 电容检测 螺栓紧固 

分 类 号：TP242.2[自动化与计算机技术—检测技术与自动化装置]