Estimation of Core Losses in Three-Phase Dry-Type Transformers Using Adaptive-Network Based Fuzzy Inference Systems (ANFIS)

Kul S., Yıldız B., TEZCAN S. S.

Electric Power Components and Systems, cilt.50, sa.16-17, ss.1006-1013, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 50 Sayı: 16-17
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1080/15325008.2022.2144550
  • Dergi Adı: Electric Power Components and Systems
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Business Source Elite, Business Source Premier, Communication Abstracts, Compendex, Environment Index, INSPEC, Metadex, DIALNET, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.1006-1013
  • Anahtar Kelimeler: core loss estimation, dry-type transformer, adaptive neuro-fuzzy inference system
  • Gazi Üniversitesi Adresli: Evet

Özet

© 2022 Taylor & Francis Group, LLC.This study presents the core loss estimation of three-phase dry-type transformers using adaptive network-based fuzzy interface systems. Accurate estimation of losses is very important during the design stage. The ANSYS/Maxwell and ANFIS programs are used to estimate core losses based on time-dependent analysis of magnetic field distributions and FEA parametric analysis, respectively. For this, we used 804 result data obtained by finite element (FEA) parametric analysis for a given number of primary turns (Formula presented.) excitation voltage (Formula presented.) and load resistance (Formula presented.) For validation, 20 of the 804 data were randomly selected from the data used in the training and testing processes. The remaining 784 data were used for training. The error obtained by the validation test is 1.9978. As a result, it has been shown that the parameters given at the design stage can be accurately estimated with ANFIS, and parameter estimation can be made for any input and output value.