基于电压波形分割-聚类的配电网弧光接地辨识方法研究  

Identification Method of Arc Grounding in Distribution Network Based on Voltage Waveform Segmentation and Clustering

作　　者：李琼 吴义辉 龙国华 邹阳 范瑞祥 LI Qiong;WU Yihui;LONG Guohua;ZOU Yang;FAN Ruixiang(Electric Power Research Institute of State Grid Jiangxi Electric Power Co.,Ltd.,Nanchang 330096,China;School of Information Engineering,Nanchang Hangkong University,Nanchang 330063,China)

机构地区：[1]国网江西省电力有限公司电力科学研究院,南昌330096 [2]南昌航空大学信息工程学院,南昌330063

出　　处：《电瓷避雷器》2023年第6期11-19,共9页Insulators and Surge Arresters

基　　金：国网江西省电力有限公司项目(编号:52182023000X)。

摘　　要：本研究提出了一种基于波形子序列分割和联合三相电压-零序电压波形特征提取聚类的单相弧光接地故障辨识方法,用于解决传统基于阈值判断方法在配电网中检测弧光接地故障时受到电流的限制。该方法首先对录波器电压波形故障段进行分割,建立三相电压和零序电压的波形子序列。然后通过特征指标分析和主成分分析方法对波形特征进行全面描述和降维处理。最后采用基于波形子序列的K-means聚类算法对降维后的特征分布进行故障辨识。实验结果表明,该方法在试验样本中具有较高的准确采集率,达到96.78%,对于配电网故障监测和诊断具有重要的实际应用价值。The author proposes a single-phase arc grounding fault identification method based on waveform subsequence segmentation and joint three-phase voltage-zero-sequence voltage waveform feature extraction clustering,which is used to solve the traditional threshold-based judgment method for detecting arc-flash ground faults in distribution networks.is limited by current.In this method,the fault segment of the voltage waveform of the oscilloscope is segmented first,and the waveform subsequences of the three-phase voltage and the zero-sequence voltage are established.Then,the waveform features are fully described and dimensionally reduced by means of characteristic index analysis and principal component analysis.Finally,the K-means clustering algorithm based on waveform subsequences is used to identify faults on the feature distribution after dimensionality reduction.The experimental results show that the method has a high accuracy acquisition rate of 96.78% in the test samples,and has important practical application value for distribution network fault monitoring and diagnosis.

关 键 词：电压波形分割 波形子序列 特征分析 K-MEANS聚类 弧光接地故障辨识 

分 类 号：TM862[电气工程—高电压与绝缘技术]