Magnetic Properties Identification by Using a Bi-Objective Optimal Multi-Fidelity Neural Network

verfasst von: Marco Baldan, Paolo Di Barba, Bernard Nacke
Abstract: In order to identify the magnetic properties of magnetic steel, the synergy between the data arising from the experimental activity, an FE model, and the use of a multi-fidelity surrogate could relieve the burden of the total cost. A neural network, with as many outputs as fidelity levels, is adopted in quality of metamodel to describe the forward problem [forward neural network (FNN)]. FNN is trained using multiple losses aiming at getting a robust surrogate that is poorly sensitive to the chosen norm. This makes it bi-objective optimal since several error metrics are simultaneously minimized. In addition, a conjugate, inverse net (INNCJ) is built, which is a ready-to-use tool for inverse properties identification, since no optimization runs are required. Its performances are compared to those obtained with a transfer learning-based approach (INNTR) and a single-fidelity inverse neural network (INNSF). Finally, a real $B - H$ curve identification task has been solved, thereby validating the conjugate inverse net.
Organisationseinheit(en): Institut für Elektroprozesstechnik
Externe Organisation(en): Università degli Studi di Pavia
Typ: Artikel
Journal: IEEE transactions on magnetics
Band: 57
ISSN: 0018-9464
Publikationsdatum: 17.05.2021
Publikationsstatus: Veröffentlicht
Peer-reviewed: Ja
ASJC Scopus Sachgebiete: Elektronische, optische und magnetische Materialien, Elektrotechnik und Elektronik
Elektronische Version(en): https://doi.org/10.1109/TMAG.2021.3068705 (Zugang: Geschlossen)

BibTeX

@article{8e1540c5b6804423a139fa218725a65d,
title = "Magnetic Properties Identification by Using a Bi-Objective Optimal Multi-Fidelity Neural Network",
abstract = "In order to identify the magnetic properties of magnetic steel, the synergy between the data arising from the experimental activity, an FE model, and the use of a multi-fidelity surrogate could relieve the burden of the total cost. A neural network, with as many outputs as fidelity levels, is adopted in quality of metamodel to describe the forward problem [forward neural network (FNN)]. FNN is trained using multiple losses aiming at getting a robust surrogate that is poorly sensitive to the chosen norm. This makes it bi-objective optimal since several error metrics are simultaneously minimized. In addition, a conjugate, inverse net (INNCJ) is built, which is a ready-to-use tool for inverse properties identification, since no optimization runs are required. Its performances are compared to those obtained with a transfer learning-based approach (INNTR) and a single-fidelity inverse neural network (INNSF). Finally, a real $B - H$ curve identification task has been solved, thereby validating the conjugate inverse net.",
keywords = "Bi-objective optimality, inverse problem, magnetic permeability, multi-fidelity, neural network",
author = "Marco Baldan and Barba, {Paolo Di} and Bernard Nacke",
year = "2021",
month = may,
day = "17",
doi = "10.1109/TMAG.2021.3068705",
language = "English",
volume = "57",
journal = "IEEE transactions on magnetics",
issn = "0018-9464",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",
}