S. Shukla, M. Manohar, C. Patel, T. Tailor, Arun Rathore
{"title":"A Combined S-transform and Ensemble of DT based Protection scheme for six-phase transmission line","authors":"S. Shukla, M. Manohar, C. Patel, T. Tailor, Arun Rathore","doi":"10.1109/ICPEE54198.2023.10060048","DOIUrl":null,"url":null,"abstract":"The current transmission infrastructure must be able to transmit more electricity due to the alarming increase in energy consumption. Instead of relying on the constraints of three-phase double-circuit lines to meet the growing power needs, a new generation of six-phase transmission systems has emerged as a potential solution. The protection challenges with six-phase transmission systems are exceptionally challenging, however, because of the enormous number of probable flaws. Given the advantages of six-phase systems over three-phase systems in meeting rising power demands and the dearth of existing protection schemes for six-phase lines, there is room for the development of a consistent protection scheme for a six-phase system that can satisfactorily perform the proposed protection tasks under a wide range of operating contingencies. To that end, this research proposes a novel protection strategy based on an ensemble of decision trees and the S-transform. To train the fault detector/classifier, the estimated frequency and phase of the current signals in real time have been utilised as input characteristics, and the S-transform has been used to do so. With a one-cycle moving window, it is feasible to estimate the input characteristics in a recursive fashion. To ensure the anticipated scheme is applicable to a six-phase system, it has been subjected to extensive simulations of a variety of fault conditions with a broad range of fault characteristics. The proposed relaying approach can reliably detect and classify issues in a single cycle across all test cases.","PeriodicalId":250652,"journal":{"name":"2023 International Conference on Power Electronics and Energy (ICPEE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 International Conference on Power Electronics and Energy (ICPEE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICPEE54198.2023.10060048","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
The current transmission infrastructure must be able to transmit more electricity due to the alarming increase in energy consumption. Instead of relying on the constraints of three-phase double-circuit lines to meet the growing power needs, a new generation of six-phase transmission systems has emerged as a potential solution. The protection challenges with six-phase transmission systems are exceptionally challenging, however, because of the enormous number of probable flaws. Given the advantages of six-phase systems over three-phase systems in meeting rising power demands and the dearth of existing protection schemes for six-phase lines, there is room for the development of a consistent protection scheme for a six-phase system that can satisfactorily perform the proposed protection tasks under a wide range of operating contingencies. To that end, this research proposes a novel protection strategy based on an ensemble of decision trees and the S-transform. To train the fault detector/classifier, the estimated frequency and phase of the current signals in real time have been utilised as input characteristics, and the S-transform has been used to do so. With a one-cycle moving window, it is feasible to estimate the input characteristics in a recursive fashion. To ensure the anticipated scheme is applicable to a six-phase system, it has been subjected to extensive simulations of a variety of fault conditions with a broad range of fault characteristics. The proposed relaying approach can reliably detect and classify issues in a single cycle across all test cases.