Solving 1D non-linear magneto quasi-static Maxwell's equations using neural networks

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autoren

Organisationseinheiten

Externe Organisationen

  • Politecnico di Milano
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)204-217
Seitenumfang14
FachzeitschriftIET Science, Measurement and Technology
Jahrgang15
Ausgabenummer2
Frühes Online-Datum5 Feb. 2021
PublikationsstatusVeröffentlicht - 17 Feb. 2021

Abstract

Electromagnetics (EM) can be described, together with the constitutive laws, by four PDEs, called Maxwell's equations. “Quasi-static” approximations emerge from neglecting particular couplings of electric and magnetic field related quantities. In case of slowly time varying fields, if inductive and resistive effects have to be considered, whereas capacitive effects can be neglected, the magneto quasi-static (MQS) approximation applies. The solution of the MQS Maxwell's equations, traditionally obtained with finite differences and elements methods, is crucial in modelling EM devices. In this paper, the applicability of an unsupervised deep learning model is studied in order to solve MQS Maxwell's equations, in both frequency and time domain. In this framework, a straightforward way to model hysteretic and anhysteretic non-linearity is shown. The introduced technique is used for the field analysis in the place of the classical finite elements in two applications: on the one hand, the B–H curve inverse determination of AISI 4140, on the other, the simulation of an induction heating process. Finally, since many of the commercial FEM packages do not allow modelling hysteresis, it is shown how the present approach could be further adopted for the inverse magnetic properties identification of new magnetic flux concentrators for induction applications.

ASJC Scopus Sachgebiete

Zitieren

Solving 1D non-linear magneto quasi-static Maxwell's equations using neural networks. / Baldan, Marco; Baldan, Giacomo; Nacke, Bernard.
in: IET Science, Measurement and Technology, Jahrgang 15, Nr. 2, 17.02.2021, S. 204-217.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Baldan M, Baldan G, Nacke B. Solving 1D non-linear magneto quasi-static Maxwell's equations using neural networks. IET Science, Measurement and Technology. 2021 Feb 17;15(2):204-217. Epub 2021 Feb 5. doi: 10.1049/smt2.12022
Download
@article{45bf827a6e1e4170b045ac41548e8c5c,
title = "Solving 1D non-linear magneto quasi-static Maxwell's equations using neural networks",
abstract = "Electromagnetics (EM) can be described, together with the constitutive laws, by four PDEs, called Maxwell's equations. “Quasi-static” approximations emerge from neglecting particular couplings of electric and magnetic field related quantities. In case of slowly time varying fields, if inductive and resistive effects have to be considered, whereas capacitive effects can be neglected, the magneto quasi-static (MQS) approximation applies. The solution of the MQS Maxwell's equations, traditionally obtained with finite differences and elements methods, is crucial in modelling EM devices. In this paper, the applicability of an unsupervised deep learning model is studied in order to solve MQS Maxwell's equations, in both frequency and time domain. In this framework, a straightforward way to model hysteretic and anhysteretic non-linearity is shown. The introduced technique is used for the field analysis in the place of the classical finite elements in two applications: on the one hand, the B–H curve inverse determination of AISI 4140, on the other, the simulation of an induction heating process. Finally, since many of the commercial FEM packages do not allow modelling hysteresis, it is shown how the present approach could be further adopted for the inverse magnetic properties identification of new magnetic flux concentrators for induction applications.",
author = "Marco Baldan and Giacomo Baldan and Bernard Nacke",
note = "Funding information: Authors would like to thank the Leibniz Universit{\"a}t Hannover for the support in bearing the publication costs.",
year = "2021",
month = feb,
day = "17",
doi = "10.1049/smt2.12022",
language = "English",
volume = "15",
pages = "204--217",
journal = "IET Science, Measurement and Technology",
issn = "1751-8822",
publisher = "Institution of Engineering and Technology",
number = "2",

}

Download

TY - JOUR

T1 - Solving 1D non-linear magneto quasi-static Maxwell's equations using neural networks

AU - Baldan, Marco

AU - Baldan, Giacomo

AU - Nacke, Bernard

N1 - Funding information: Authors would like to thank the Leibniz Universität Hannover for the support in bearing the publication costs.

PY - 2021/2/17

Y1 - 2021/2/17

N2 - Electromagnetics (EM) can be described, together with the constitutive laws, by four PDEs, called Maxwell's equations. “Quasi-static” approximations emerge from neglecting particular couplings of electric and magnetic field related quantities. In case of slowly time varying fields, if inductive and resistive effects have to be considered, whereas capacitive effects can be neglected, the magneto quasi-static (MQS) approximation applies. The solution of the MQS Maxwell's equations, traditionally obtained with finite differences and elements methods, is crucial in modelling EM devices. In this paper, the applicability of an unsupervised deep learning model is studied in order to solve MQS Maxwell's equations, in both frequency and time domain. In this framework, a straightforward way to model hysteretic and anhysteretic non-linearity is shown. The introduced technique is used for the field analysis in the place of the classical finite elements in two applications: on the one hand, the B–H curve inverse determination of AISI 4140, on the other, the simulation of an induction heating process. Finally, since many of the commercial FEM packages do not allow modelling hysteresis, it is shown how the present approach could be further adopted for the inverse magnetic properties identification of new magnetic flux concentrators for induction applications.

AB - Electromagnetics (EM) can be described, together with the constitutive laws, by four PDEs, called Maxwell's equations. “Quasi-static” approximations emerge from neglecting particular couplings of electric and magnetic field related quantities. In case of slowly time varying fields, if inductive and resistive effects have to be considered, whereas capacitive effects can be neglected, the magneto quasi-static (MQS) approximation applies. The solution of the MQS Maxwell's equations, traditionally obtained with finite differences and elements methods, is crucial in modelling EM devices. In this paper, the applicability of an unsupervised deep learning model is studied in order to solve MQS Maxwell's equations, in both frequency and time domain. In this framework, a straightforward way to model hysteretic and anhysteretic non-linearity is shown. The introduced technique is used for the field analysis in the place of the classical finite elements in two applications: on the one hand, the B–H curve inverse determination of AISI 4140, on the other, the simulation of an induction heating process. Finally, since many of the commercial FEM packages do not allow modelling hysteresis, it is shown how the present approach could be further adopted for the inverse magnetic properties identification of new magnetic flux concentrators for induction applications.

UR - http://www.scopus.com/inward/record.url?scp=85100572981&partnerID=8YFLogxK

U2 - 10.1049/smt2.12022

DO - 10.1049/smt2.12022

M3 - Article

AN - SCOPUS:85100572981

VL - 15

SP - 204

EP - 217

JO - IET Science, Measurement and Technology

JF - IET Science, Measurement and Technology

SN - 1751-8822

IS - 2

ER -

Von denselben Autoren