{"title":"Forecasting of Dissolved Gases in Power Transformer Oil Based on DOG -LSSVM Regression and Artificial Bee Colony","authors":"Yiyi Zhang, Liuliang Zhao, Jiake Fang, Jian Jiao, Changyi Liao, Xin Li","doi":"10.1109/POWERCON.2018.8601863","DOIUrl":null,"url":null,"abstract":"In order to accurately forecast the development trend of transformers and the trend in the early stage of faults, this paper provides a method to predict the dissolved gas content in transformer oil with DOG wavelet kernel function and artificial bee colony algorithm (ABC) based on least squares vector machine regression (LSSVR). This paper first optimizes the parameters of LSSVR by ABC, then constructs the LSSVR model with the DOG, finally evaluates the prediction performance based on the measurement of the average absolute error percentage (MAPE) and the square correlation coefficient $(r^{2})$ to prove the accuracy and effectiveness of this method.","PeriodicalId":260947,"journal":{"name":"2018 International Conference on Power System Technology (POWERCON)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 International Conference on Power System Technology (POWERCON)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/POWERCON.2018.8601863","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
In order to accurately forecast the development trend of transformers and the trend in the early stage of faults, this paper provides a method to predict the dissolved gas content in transformer oil with DOG wavelet kernel function and artificial bee colony algorithm (ABC) based on least squares vector machine regression (LSSVR). This paper first optimizes the parameters of LSSVR by ABC, then constructs the LSSVR model with the DOG, finally evaluates the prediction performance based on the measurement of the average absolute error percentage (MAPE) and the square correlation coefficient $(r^{2})$ to prove the accuracy and effectiveness of this method.