基于边-云协同的输电线路综合在线监测系统  

Integrated On-line Monitoring System of Transmission Line Based on Boundary-cloud Cooperation

作　　者：叶保璇 王康坚 余盛达 易婷婷 黄廷城 Ye Baoxuan;Wang Kangjian;Yu Shengda;Yi Tingting;Huang Tingcheng(Wenchang Power Supply Bureau,Hainan Power Grid Co.,Ltd.,Wenchang,Hainan 570100,China;Guangzhou Power Electrical Technology Co.,Ltd.,Guangzhou 510670,China)

机构地区：[1]海南电网有限责任公司文昌供电局,海南文昌570100 [2]广州市奔流电力科技有限公司,广州510670

出　　处：《机电工程技术》2020年第11期73-75,共3页Mechanical & Electrical Engineering Technology

基　　金：海南电网有限责任公司科技项目(编号:070500KK52190001)。

摘　　要：保障输电线路安全运行是电网运维的重要目标,传统的多种输电线路在线监测系统监测功能单一,数据处理和智能识别任务位于后台,导致数据通讯和存储压力大。目前的输电线路边缘端识别产品,其图像智能识别速度慢、准确率不高。提出基于边-云协同的输电线路综合在线监测系统,通过深度学习技术在边缘端实现一次识别,并将所得疑似缺陷图片传至云端进行二次识别,解决大量数据传输困难的问题。同时传感器采集的多类型数据可送至后台进行全面分析,综合判断线路周围环境。结合边缘端和云端的双重识别和综合数据分析,大大提升了检测准确率,保证输电线路稳定运行。To ensure the safe operation of transmission lines is an important goal of power grid operation and maintenance.The traditional online monitoring systems of various transmission lines have single monitoring function,and the tasks of data processing and intelligent identification are located in the background,which leads to the pressure of data communication and storage.The speed of intelligent image recognition is slow and the accuracy is not high.An integrated on-line monitoring system of transmission line based on edge cloud collaboration was proposed.Through deep learning technology,the primary identification was realized at the edge,and the suspected defect images were sent to the cloud for secondary identification,which solved the problem of large amount of data transmission difficulties.The multi type data collected by the sensor can be sent to the background for comprehensive analysis and comprehensive judgment of the surrounding environment of the line.Combined with the dual identification of edge and cloud and comprehensive data analysis,the detection accuracy is greatly improved and the stable operation of transmission lines is ensured.

关 键 词：边云协同 输电线路 深度学习 综合监测 

分 类 号：TM75[电气工程—电力系统及自动化]