###
DOI:
电力大数据:2023,26(06):-
←前一篇   |   后一篇→
本文二维码信息
基于边缘AI的配网非定点作业现场实时智能安全监护成套装置研制
许 逵1, 李鑫卓1, 张 历1, 张俊杰1, 杨 宁2
(1.贵州电网有限责任公司电力科学研究院;2.武汉光谷信息技术股份有限公司)
The Development of Real-time Intelligent Safety Supervision Kit Based on Edge AI for Non-fixed-location Operations in Distribution Network
Xu Kui1, Ll Xin Zhuo1, Zhang Li1, Zhang Jun Jie1, Yang Ning2
(1.Electric Power Research Institute of Guizhou Power Grid Co., Ltd;2.Wuhan Optics Valley Information Technology Co., Ltd)
摘要
图/表
参考文献
相似文献
本文已被:浏览 135次   下载 1009
投稿时间:2023-06-16    修订日期:2023-09-04
中文摘要: 为满足带电作业、高处作业等配电网复杂作业场景的安全监护要求,解决配网非定点作业人工监护效率低、现场终端能力单一协同不足、通信安全较难保障等实际问题,文章提出了一种具有强边缘智能、主动监护、精准定位与辨识、多层通讯安全防护等特点的智能安全监护成套装置。首先,该装置基于自研物联网模组搭配可扩展高性能边缘计算模组,实现了视频监控设备的主动控制、作业人员活动的跟踪监测、监控视频的实时就地智能分析与违规行为预警。其次,装置通过配套定位设备,实现了作业地点和作业范围的动态划定以及作业人员的精准定位。最后,装置以硬件加密为基础,实现了可信通讯设备的自组网络和通讯内容的多层加密。实验结果表明,该装置在典型场景下的平均检测精度可以达到90%以上,可以极大提高配网户外非定点作业场景下的安全监护效率,提升作业现场安全的实时监管能力。
Abstract:In order to meet the safety supervision requirements of complex operation scenarios in the distribution network such as live operations and high-altitude operations, and solve the practical problems of low efficiency of manual supervision, insufficient coordination of on-site terminals, and difficult to secure on-site wireless communication, this paper proposes a dedicated real-time intelligent safety supervision kit with strong edge intelligence, high-accuracy positioning and predicting, pro-active monitoring, and high-level communication security. Firstly of all, based on IoT modules and scalable high-performance edge computing modules, the kit realize pro-active control of surveillance terminals, the tracking and analyzing operator activities in real-time, and the early warning of violation behavior. Secondly, the kit achieves dynamic delimitation of the on-site operation location and range, as well as precise positioning of personnel. Finally, with hardware encryption, the kit realizes encrypted communication for Ad-hoc local and remote network of trusted devices. Experimental results show that the average detection accuracy of this device in typical scenarios can reach more than 90%, which can significantly improve the safety supervision efficiency in non-fixed-location operation scenarios and enhance the real-time supervision ability.
文章编号:     中图分类号:    文献标志码:
基金项目:中国南方电网科技项目(066600KK52210050)。
引用文本: