{"title":"Composite System Reliability Analysis using Deep Learning enhanced by Transfer Learning","authors":"Dogan Urgun, C. Singh","doi":"10.1109/PMAPS47429.2020.9183474","DOIUrl":null,"url":null,"abstract":"This paper proposes a new algorithm for evaluation of power systems reliability based on Artificial Intelligence. This algorithm proposes an efficient technique to gather training samples and training Convolutional Neural Networks (CNN) for computing power system reliability indices considering changes in system parameters. It is shown that the computational efficiency gained by machine learning can be increased even further by reducing the time required for collecting training samples and applying transfer learning. Three different modifications of IEEE Reliability Test System (IEEE-RTS) are used to show the performance of proposed method during changes in system. The results of case studies show that CNNs together with the proposed algorithm provide a good classification accuracy while reducing computation time.","PeriodicalId":126918,"journal":{"name":"2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PMAPS47429.2020.9183474","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
This paper proposes a new algorithm for evaluation of power systems reliability based on Artificial Intelligence. This algorithm proposes an efficient technique to gather training samples and training Convolutional Neural Networks (CNN) for computing power system reliability indices considering changes in system parameters. It is shown that the computational efficiency gained by machine learning can be increased even further by reducing the time required for collecting training samples and applying transfer learning. Three different modifications of IEEE Reliability Test System (IEEE-RTS) are used to show the performance of proposed method during changes in system. The results of case studies show that CNNs together with the proposed algorithm provide a good classification accuracy while reducing computation time.