Identification of FDNEs Via Frequency-Partitioned Loewner Matrix Fitting Technique

カテゴリ: 研究会(論文単位)

論文No: HV18094

グループ名: 【B】電力・エネルギー部門 高電圧研究会

発行日: 2018/08/02

タイトル(英語): Identification of FDNEs Via Frequency-Partitioned Loewner Matrix Fitting Technique

著者名: Medina Edgar(CINVESTAV Guadalajara),Morales Jesus(Polytechnique Montreal),Ramirez Abner(CINVESTAV Guadalajara)

著者名(英語): Edgar Medina(CINVESTAV Guadalajara),Jesus Morales(Polytechnique Montreal),Abner Ramirez(CINVESTAV Guadalajara)

キーワード: Frequency response|Loewner Matrix|Singular Value Decomposition|Rational approximation|Frequency response|Loewner Matrix|Singular Value Decomposition|Rational approximation

要約(日本語): This paper presents a novel approach to identify frequency-dependent network equivalents (FDNEs) based on the Loewner Matrix (LM) fitting technique. The proposed method consists of partitioning the fitting frequency range into a user-defined number of sections. Then, the system is locally identified by applying LM to each section. Finally, a global rational model is obtained as a combination of each sub-model. The proposed technique presents a considerably improved performance compared to traditional LM technique.

要約(英語): This paper presents a novel approach to identify frequency-dependent network equivalents (FDNEs) based on the Loewner Matrix (LM) fitting technique. The proposed method consists of partitioning the fitting frequency range into a user-defined number of sections. Then, the system is locally identified by applying LM to each section. Finally, a global rational model is obtained as a combination of each sub-model. The proposed technique presents a considerably improved performance compared to traditional LM technique.

原稿種別: 英語

PDFファイルサイズ: 970 Kバイト