A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear Materials

This work explores the use of Physics-Informed Neural Networks (PINNs) and a newly proposed approach, called the STacked Adaptive Residual PINN (STAR-PINN), to solve magnetic diffusion problems in the magneto quasi static regime. The study covers both one- and two-dimensional domains. The key advant...

Full description

Saved in:
Bibliographic Details
Main Authors: Shayan Dodge, Sami Barmada, Alessandro Formisano
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11122441/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849736734515920896
author Shayan Dodge
Sami Barmada
Alessandro Formisano
author_facet Shayan Dodge
Sami Barmada
Alessandro Formisano
author_sort Shayan Dodge
collection DOAJ
description This work explores the use of Physics-Informed Neural Networks (PINNs) and a newly proposed approach, called the STacked Adaptive Residual PINN (STAR-PINN), to solve magnetic diffusion problems in the magneto quasi static regime. The study covers both one- and two-dimensional domains. The key advantage of this new architecture is the ability to refine predictions through multiple lightweight PINN blocks to achieve accurate results with lower computational cost and less architectural complexity than more advanced neural networks like Recurrent Neural Networks or Convolutional Neural Networks. The simplicity and efficiency of STAR-PINN make it a promising solution for tackling large-scale and nonlinear challenges in computational electromagnetics.
format Article
id doaj-art-5450f7727c0a45de9bca46f73fade13c
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-5450f7727c0a45de9bca46f73fade13c2025-08-20T03:07:11ZengIEEEIEEE Access2169-35362025-01-011314138014139410.1109/ACCESS.2025.359786911122441A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear MaterialsShayan Dodge0Sami Barmada1https://orcid.org/0000-0003-1414-1114Alessandro Formisano2https://orcid.org/0000-0002-7007-5759DESTeC, University of Pisa, Pisa, ItalyDESTeC, University of Pisa, Pisa, ItalyDipartimento di Ingegneria, Università degli Studi della Campania “Luigi Vanvitelli,”, Aversa, ItalyThis work explores the use of Physics-Informed Neural Networks (PINNs) and a newly proposed approach, called the STacked Adaptive Residual PINN (STAR-PINN), to solve magnetic diffusion problems in the magneto quasi static regime. The study covers both one- and two-dimensional domains. The key advantage of this new architecture is the ability to refine predictions through multiple lightweight PINN blocks to achieve accurate results with lower computational cost and less architectural complexity than more advanced neural networks like Recurrent Neural Networks or Convolutional Neural Networks. The simplicity and efficiency of STAR-PINN make it a promising solution for tackling large-scale and nonlinear challenges in computational electromagnetics.https://ieeexplore.ieee.org/document/11122441/Physics-informed neural network (PINN)magneto-quasi-static (MQS)time-domain electromagneticsnonlinear magnetic materialsstacked neural networksresidual network
spellingShingle Shayan Dodge
Sami Barmada
Alessandro Formisano
A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear Materials
IEEE Access
Physics-informed neural network (PINN)
magneto-quasi-static (MQS)
time-domain electromagnetics
nonlinear magnetic materials
stacked neural networks
residual network
title A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear Materials
title_full A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear Materials
title_fullStr A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear Materials
title_full_unstemmed A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear Materials
title_short A STacked Adaptive Residual PINN (STAR-PINN) Approach to 2D Time-Domain Magnetic Diffusion in Nonlinear Materials
title_sort stacked adaptive residual pinn star pinn approach to 2d time domain magnetic diffusion in nonlinear materials
topic Physics-informed neural network (PINN)
magneto-quasi-static (MQS)
time-domain electromagnetics
nonlinear magnetic materials
stacked neural networks
residual network
url https://ieeexplore.ieee.org/document/11122441/
work_keys_str_mv AT shayandodge astackedadaptiveresidualpinnstarpinnapproachto2dtimedomainmagneticdiffusioninnonlinearmaterials
AT samibarmada astackedadaptiveresidualpinnstarpinnapproachto2dtimedomainmagneticdiffusioninnonlinearmaterials
AT alessandroformisano astackedadaptiveresidualpinnstarpinnapproachto2dtimedomainmagneticdiffusioninnonlinearmaterials
AT shayandodge stackedadaptiveresidualpinnstarpinnapproachto2dtimedomainmagneticdiffusioninnonlinearmaterials
AT samibarmada stackedadaptiveresidualpinnstarpinnapproachto2dtimedomainmagneticdiffusioninnonlinearmaterials
AT alessandroformisano stackedadaptiveresidualpinnstarpinnapproachto2dtimedomainmagneticdiffusioninnonlinearmaterials