EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Operator-aware physics-informed neural networks for forward and inverse problems in solid mechanics

Amir Ghorbani Ghezeljehmeidan, Willem Dirk van Driel, Justin Dauwels

Machine Learning: Science and Technology · 01.08.2026 · 7 (4) · 045013

DOI: 10.1088/2632-2153/ae8071 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Machine Learning: Science and Technology, online erschienen am 08.07.2026. Verfasst von Amir Ghorbani Ghezeljehmeidan und 2 weiteren Autor:innen. Die Arbeit nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NL ministry EZK.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Physics-informed neural networks (PINNs) on complex domains are limited by input representations that encode geometry but do not reflect the physics of the governing PDE. We propose an operator-aware PINN for solid mechanics problems that embeds precomputed eigenmodes of the problem’s own discrete operator as geometry and physics-aware features within a weak-form variational formulation. The displacement field is represented by a neural mapping enriched by operator-aligned features rather than being constrained to a finite element trial space. A hybrid chain-rule formulation propagates spatial derivatives through both coordinate and eigenmode branches, yielding strain fields consistent with operator structure while preserving neural expressivity. The approach maintains the stability of variational formulations while mitigating the spectral bias and geometry-encoding limitations of classical PINNs. For inverse elasticity, the same eigenmodes act as virtual test functions in a weak-form residual, mitigating the stiffness-collapse pathology. The framework provides an instance-based alternative to differentiable finite element method without labeled training data. It achieves competitive accuracy and connects spectral operator theory with physics-informed neural computation.

Förderung

  • NL ministry EZK

Zitieren

APA

Ghezeljehmeidan, A. G., Driel, W. D. V., & Dauwels, J. (2026). Operator-aware physics-informed neural networks for forward and inverse problems in solid mechanics. Machine Learning: Science and Technology, 7(4), 045013. https://doi.org/10.1088/2632-2153/ae8071

Vancouver

Ghezeljehmeidan AG, Driel WDV, Dauwels J. Operator-aware physics-informed neural networks for forward and inverse problems in solid mechanics. Machine Learning: Science and Technology. 2026;7(4):045013. doi:10.1088/2632-2153/ae8071

BibTeX
@article{Ghezeljehmeidan2026,
  author = {Ghezeljehmeidan, Amir Ghorbani and Driel, Willem Dirk van and Dauwels, Justin},
  title = {Operator-aware physics-informed neural networks for forward and inverse problems in solid mechanics},
  journal = {Machine Learning: Science and Technology},
  year = {2026},
  volume = {7},
  number = {4},
  pages = {045013},
  doi = {10.1088/2632-2153/ae8071}
}
Kostenlos starten →