一二次融合电力环网箱接地定位算法设计  

Design of grounding location algorithm of primary and secondary fusion power netting trunk

作　　者：陈昊 李思焱 葛明明 CHEN Hao;LI Siyan;GE Mingming(State Grid Wuhu Power Supply Company,Wuhu 241000,China)

机构地区：[1]国网芜湖供电公司,安徽芜湖241000

出　　处：《电子设计工程》2023年第18期127-131,共5页Electronic Design Engineering

基　　金：国网公司科技项目(JL71-15-042)。

摘　　要：针对配电网接地故障可能对电力可靠供应造成严重威胁的现状,开展了一二次融合电力环网箱接地定位算法的设计研究。在分析了一二次融合电力环网箱结构与优点的基础上,采用经验模态分解(EMD)算法来处理故障信号,并获得了多个内涵模态分量(IMF)。同时将IMF分量作为卷积神经网络(CNN)的输入,利用CNN自动学习进行故障信号与位置的关联分析,进而实现对故障的精准定位。通过对实际电网数据集进行的测试结果表明,EMD算法可实现对不同频率故障特征的提取,以减少后续神经网络模型的训练时长;且CNN模型具有深层的网络结构,能够提高故障定位的准确度,并保障电能的高可靠供应。In view of the current situation that the distribution network grounding fault may pose a serious threat to the reliable power supply,this paper carries out the design and research of the primary and secondary fusion power ring network box grounding location algorithm.Based on the analysis of the structure and advantages of the primary and secondary fusion power ring cage,the Empirical Mode Decomposition(EMD)algorithm is used to process the fault signal to obtain multiple Intrinsic Mode Functions(IMF),and the IMF components are used as the input of Convolution Neural Network(CNN).The correlation analysis between the fault signal and the fault location is realized through CNN automatic learning to realize the accurate location of the fault.The test results on the actual power grid data set show that EMD algorithm can extract fault features with different frequencies and reduce the training time of subsequent neural network model.CNN model has a deep network structure,which can improve the accuracy of fault location and ensure the high and reliable supply of power.

关 键 词：电力环网箱 故障定位 神经网络 电网接地 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TN99[自动化与计算机技术—控制科学与工程]