配网设备局部放电边缘侧精简特征提取与短报文传输技术研究  

Research on edge-side simplified feature set extraction and short message transmission mechanism for partial discharge of distribution network equipment

作　　者：何楠 刘弘景 刘宏亮 刘可文 任明[2] HE Nan;LIU Hongjing;LIU Hongliang;LIU Kewen;REN Ming(State Grid Beijing Electric Power Research Institute,Beijing 100075,China;State Key Laboratory of Electrical Insulation and Power Equipment,Xi’an Jiaotong University,Xi’an 710049,China)

机构地区：[1]国网北京市电力公司电力科学研究院,北京100075 [2]电工材料电气绝缘全国重点实验室,西安交通大学,陕西西安710049

出　　处：《电工电能新技术》2024年第1期95-104,共10页Advanced Technology of Electrical Engineering and Energy

基　　金：国网北京市电力公司科技项目(520223200065)。

摘　　要：电力设备内部潜伏性缺陷和故障可通过对局部放电的监测发现。目前随着电力设备尤其是配网设备数量的激增,海量的局部放电监测数据加剧了系统计算、传输和存储的负担。因此,本文基于无线传感网络架构提出了一种分布式感知网络,选用了远距离无线电作为主要传输手段,并构建了一种短报文传输机制以提升数据传输效率;其次,基于典型开关柜局部放电数据选取了原始特征集合,并采用变异数分析方法构建了一个8维度的精简特征集,通过边缘计算的方式实现数据特征提取与精简;最后,依次从识别准确率、传输密度、节点功耗以及运算量四个方面评估了网络的监测效能。本文解决了边缘侧和传输侧的关键问题,对于泛在电力物联网的建设提供了一定的技术参考。Detection of partial discharge(PD)can effectively reflect latent defects and faults in power equipment,which is one of the important means of power grid monitoring.With the surge in the number of power equipment,especially in the distribution network equipment,massive PD monitoring data intensifies the burden of calculation,transmission and storage in the system.Therefore,this paper proposes a distributed sensing network based on wire-less sensor network,selects LoRa as the main transmission means,and constructs a short message transmission mechanism to improve data transmission efficiency.Secondly,the original feature set is selected based on the typical partial discharge data of the switchgear,and an 8-dimension simplified feature set is constructed by using the method of variance analysis and edge computing to achieve data simplification and idle computing power utilization.Finally,the monitoring performance of the network is evaluated from four aspects:recognition accuracy,transmission density,node power consumption and computation.This paper solves the key problems in the edge side and the transmission side,and provides a certain technical reference for the construction of ubiquitous electric Internet of Things(IoT).

关 键 词：局部放电 分布式感知网络 边缘计算 短报文传输 

分 类 号：TM855[电气工程—高电压与绝缘技术]