Physics-informed two-tier neural network for non-linear model order reduction
Advanced Modeling and Simulation in Engineering Sciences · 14.11.2024 · 11 (1)
DOI: 10.1186/s40323-024-00273-3 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advanced Modeling and Simulation in Engineering Sciences, online erschienen am 14.11.2024. Verfasst von Yankun Hong und 2 weiteren Autor:innen. Beteiligte Einrichtung: Centre for Analysis, Scientific Computing and Applications, Eindhoven University of Technology, Eindhoven, 5600MB The Netherlands. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 86 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: H2020 European Research Council.
Abstract (Originalsprache)
In recent years, machine learning (ML) has had a great impact in the area of non-intrusive, non-linear model order reduction (MOR). However, the offline training phase often still entails high computational costs since it requires numerous, expensive, full-order solutions as the training data. Furthermore, in state-of-the-art methods, neural networks trained by a small amount of training data cannot be expected to generalize sufficiently well, and the training phase often ignores the underlying physical information when it is applied with MOR. Moreover, state-of-the-art MOR techniques that ensure an efficient online stage, such as hyper reduction techniques, are either intrusive or entail high offline computational costs. To resolve these challenges, inspired by recent developments in physics-informed and physics-reinforced neural networks, we propose a non-intrusive, physics-informed, two-tier deep network (TTDN) method. The proposed network, in which the first tier achieves the regression of the unknown quantity of interest and the second tier rebuilds the physical constitutive law between the unknown quantities of interest and derived quantities, is trained using pretraining and semi-supervised learning strategies. To illustrate the efficiency of the proposed approach, we perform numerical experiments on challenging non-linear and non-affine problems, including multi-scale mechanics problems.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Centre for Analysis, Scientific Computing and Applications, Eindhoven University of Technology, Eindhoven, 5600MB The Netherlands.
ORCID-Profile: Yankun Hong
Förderung
- H2020 European Research Council
Zitieren
APA
Hong, Y., Bansal, H., & Veroy, K. (2024). Physics-informed two-tier neural network for non-linear model order reduction. Advanced Modeling and Simulation in Engineering Sciences, 11(1). https://doi.org/10.1186/s40323-024-00273-3
Vancouver
Hong Y, Bansal H, Veroy K. Physics-informed two-tier neural network for non-linear model order reduction. Advanced Modeling and Simulation in Engineering Sciences. 2024;11(1). doi:10.1186/s40323-024-00273-3
BibTeX
@article{Hong2024,
author = {Hong, Yankun and Bansal, Harshit and Veroy, Karen},
title = {Physics-informed two-tier neural network for non-linear model order reduction},
journal = {Advanced Modeling and Simulation in Engineering Sciences},
year = {2024},
volume = {11},
number = {1},
doi = {10.1186/s40323-024-00273-3}
}
EUVIMED