Hysteresis and loss prediction for high-permeability grain-oriented electrical Steel by Material Characterization

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

For loss prediction of a transformer it is required to model the loss of its core, constructed out of high-permeability grain-oriented electrical steel (HiB GOES). This work predicts the magnetic loss in the rolling direction (RD) of a sheet of H105-30 HiB GOES for flux densities up to 1.9 T, and frequencies up to 300 Hz. Material characterization parameters, obtained by statistical loss separation for sinusoidal excitation, are applied in a hysteresis model to predict dynamic behavior, from which the loss is determined. This dynamic behavior is solely determined by material characterization. The maximum error for the predicted loss is 4.74%, the RMS error is 2.22% (\hat{B}\gt0.5T, f=50 Hz).

Original languageEnglish
Title of host publication2019 19th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering (ISEF)
PublisherInstitute of Electrical and Electronics Engineers
Number of pages2
ISBN (Electronic)9781728115603
DOIs
Publication statusPublished - 20 May 2020
Event19th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering, (ISEF2019) - Prouvé Congress Center, Nancy, France
Duration: 29 Aug 201931 Aug 2019
https://isef2019.sciencesconf.org/

Conference

Conference19th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering, (ISEF2019)
Abbreviated titleISEF2019
CountryFrance
CityNancy
Period29/08/1931/08/19
Internet address

Keywords

  • Epstein frame
  • High-permeability grain-oriented electrical steel
  • iron loss prediction
  • separation of losses

Fingerprint

Dive into the research topics of 'Hysteresis and loss prediction for high-permeability grain-oriented electrical Steel by Material Characterization'. Together they form a unique fingerprint.

Cite this