基于无人机图像识别的输电线路绝缘子污秽在线监测  

Drone Photographing and Image Recognition-based Online Monitoring of Insulator Contamination in Transmission Lines

作　　者：熊向阳 XIONG Xiangyang(State Grid Gansu Provincial Electric Power Company,Tianshui Power Supply Company,Tianshui 741000,China)

机构地区：[1]国网甘肃省电力公司天水供电公司,甘肃天水741000

出　　处：《电工技术》2024年第14期79-81,共3页Electric Engineering

摘　　要：传统的输电线路绝缘子污秽在线监测特征提取采用面积覆盖的方式,导致监测对比度较低,进而增加最终监测结果的相对误差。为此,提出了基于无人机图像识别的输电线路绝缘子污秽在线监测方法,并进行了相应的设计与实践。根据当前测定要求,先进行图像采集,采用多点位的方式,以提升监测的对比度,并实现积污特征多点位提取。在此基础上,设计无人机图像识别线路绝缘子污秽在线监测模型,采用动态跟踪标定的形式实现在线监测。测试结果表明:针对选定的5条输电线路绝缘子污秽监测,对比传统HHT绝缘子污秽在线监测方法、传统4G通信绝缘子污秽在线监测方法,所设计的无人机图像识别绝缘子污秽在线监测方法的相对误差最低,被控制在2.5以下,说明在无人机图像识别技术的辅助下,设计的在线监测方法更加有效,针对性强,应用误差可控,在一定程度上提高了监测的精准度和稳定性。The conventional online monitoring feature extraction of insulator contamination on transmission lines adopts an area coverage method,which leads to low monitoring contrast and increased relative error of the final results.Therefore a method for online monitoring of insulator contamination in transmission lines based on drone photographing and image recognition was proposed,and corresponding design and practice were carried out.According to the current measurement requirements,image acquisition is carried out first,using a multi-point approach to improve the contrast of monitoring and achieve multi-point extraction of pollution features.On this basis,an image recognition model for online monitoring of transmission line insulator contaminations is designed,and dynamic tracking calibration is adopted to achieve online monitoring.The test results show that for the selected 5 transmission line insulator contamination monitoring,compared with the groups of conventional methods such as HHT-based and 4G-based monitoring,the designed method ultimately obtained a monitoring relative error of less than 2.5.This indicates that with the assistance of unmanned aerial vehicle image recognition technology,the designed online monitoring method is more effective,targeted,and has controllable application errors,which to some extent improves the accuracy and stability of monitoring.

关 键 词：无人机技术 输电线路 图像识别 绝缘子 污秽识别 在线监测 

分 类 号：TN911.73[电子电信—通信与信息系统]