基于CNNs和红外热成像的电气检修智能保护头盔  

Intelligent protective helmet for electrical maintenance based on CNNs and infrared thermal imaging

作　　者：刘佳[1] 王兴媛 王帆 赵进全[2] LIU Jia;WANG Xingyuan;WANG Fan;ZHAO Jinquan(China Electric Power Research Institute,Beijing 100192,China;School of Electrical Engineering,Xi'an Jiaotong University,Xi'an 710049,China)

机构地区：[1]中国电力科学研究院,北京100192 [2]西安交通大学电气工程学院,陕西西安710049

出　　处：《电子设计工程》2022年第22期78-83,88,共7页Electronic Design Engineering

基　　金：国家电网公司总部管理科技项目资助(5600-201926164A-0-0-00)。

摘　　要：针对电气检修过程中检测效率低下、容易误判电气故障、现场作业监控不周全等诸多问题,结合运动识别检测算法及热成像技术,开发了一种电气检修人员智能保护头盔,实现了对人员作业状态的连续监测和对设备发热故障的检测与预警。该文利用标准发热源和工业测温仪,通过模拟实际工作状态对头盔各个功能进行测试。测试结果表明,智能保护头盔可在1 m距离上准确识别直径1 cm大小的发热源;可对30~120℃的物体进行自动测量,测量误差小于±1℃;可实现对人员静止、步行、跑步等五类型运动进行检测,识别准确率高于95%,为电气检修过程中的故障检测与预警提供了一种有效工具。Aiming at many problems such as low detection efficiency,easy misjudgment of electrical faults,incomplete monitoring of field operations in the process of electrical maintenance,this article combines motion recognition detection algorithms and thermal imaging technology to develop an intelligent protective helmet for electrical maintenance personnel to achieve Continuous monitoring of operating status and detection and early warning of equipment heating faults.This article uses standard heating sources and industrial thermometers to test the various functions of the helmet by simulating actual working conditions.The test results show that the smart protective helmet can accurately identify heat sources with a diameter of 1 cm at a distance of 1 m;it can automatically measure objects between 30℃ and 120℃,and the measurement error is less than±1℃;it can achieve static,walking,and Five types of sports including running are detected,and the recognition accuracy rate is higher than 95%.It provides an effective tool for fault detection and early warning in the process of electrical maintenance.

关 键 词：智能保护头盔 电气检测 主动预警 异常温度检测 运动检测 

分 类 号：TN919.5[电子电信—通信与信息系统]