Wen Qingfeng, W. Xiaoguang, Yue Xiangying, Liang Zehui, L. Longji, Xi Xiaoguang, Ma Yuyan, Zhang Chi
{"title":"Lightning Overvoltage Identification in Distribution Line based on Relevance Vector Machine","authors":"Wen Qingfeng, W. Xiaoguang, Yue Xiangying, Liang Zehui, L. Longji, Xi Xiaoguang, Ma Yuyan, Zhang Chi","doi":"10.1109/CICED.2018.8592419","DOIUrl":null,"url":null,"abstract":"Realizing lightning overvoltage online identification is of great significance to improving the practical of the lightning online monitoring device in the failure analysis. Therefore, a method based on relevance vector machine (RVM) is attempted to be applied to lightning overvoltage identification. The Hilbert-Huang transform method was introduced to analyze the waveform characters of induced lightning, shield failure and back flashover. Characteristic quantity was generated and input into the RVM to construct the intelligent lightning overvoltage identification machine. The method integrated the feature information of lightning overvoltage, and outputted the probabilities of various lightning overvoltage. The PSCAD simulation results demonstrate that, the training time is low and the recognition rate is high. The identification method proposed can be well applied in identification of lightning overvoltage.","PeriodicalId":142885,"journal":{"name":"2018 China International Conference on Electricity Distribution (CICED)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 China International Conference on Electricity Distribution (CICED)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CICED.2018.8592419","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Realizing lightning overvoltage online identification is of great significance to improving the practical of the lightning online monitoring device in the failure analysis. Therefore, a method based on relevance vector machine (RVM) is attempted to be applied to lightning overvoltage identification. The Hilbert-Huang transform method was introduced to analyze the waveform characters of induced lightning, shield failure and back flashover. Characteristic quantity was generated and input into the RVM to construct the intelligent lightning overvoltage identification machine. The method integrated the feature information of lightning overvoltage, and outputted the probabilities of various lightning overvoltage. The PSCAD simulation results demonstrate that, the training time is low and the recognition rate is high. The identification method proposed can be well applied in identification of lightning overvoltage.