Detail publikačního výsledku

Physics-Informed Neural Networks for Multiscale Large Deformation Analysis of Metamaterials

LI, H.; KHODAEI, Z.; KOTOUL, M.; ALIABADI, M.

Originální název

Physics-Informed Neural Networks for Multiscale Large Deformation Analysis of Metamaterials

Anglický název

Physics-Informed Neural Networks for Multiscale Large Deformation Analysis of Metamaterials

Druh

Stať ve sborníku v databázi WoS či Scopus

Originální abstrakt

Physics-informed neural networks (PINNs) have recently emerged as a promising alternative to traditional numerical methods for solving solid mechanics problems. In this work, we propose a novel PINN architecture designed for homogenisation problems of metamaterials under large deformation. The architecture incorporates periodic functions to ensure exactly imposed boundary conditions and employs an energy-based loss for efficient training. Three representative metamaterial structures—octet truss, gyroid, and spindoid—are selected as case studies. The results demonstrate that the proposed PINN achieves accuracy comparable to finite element analysis (FEA), while offering improved computational efficiency for high-volume-fraction structures. Beyond accuracy and speed, the meshfree nature and flexibility of PINNs provide clear advantages, highlighting their potential as a scalable tool for modelling complex materials.

Anglický abstrakt

Physics-informed neural networks (PINNs) have recently emerged as a promising alternative to traditional numerical methods for solving solid mechanics problems. In this work, we propose a novel PINN architecture designed for homogenisation problems of metamaterials under large deformation. The architecture incorporates periodic functions to ensure exactly imposed boundary conditions and employs an energy-based loss for efficient training. Three representative metamaterial structures—octet truss, gyroid, and spindoid—are selected as case studies. The results demonstrate that the proposed PINN achieves accuracy comparable to finite element analysis (FEA), while offering improved computational efficiency for high-volume-fraction structures. Beyond accuracy and speed, the meshfree nature and flexibility of PINNs provide clear advantages, highlighting their potential as a scalable tool for modelling complex materials.

Klíčová slova

Physics informed neural network; Homogenisation; Metamaterials; Multiscale Analysis

Klíčová slova v angličtině

Physics informed neural network; Homogenisation; Metamaterials; Multiscale Analysis

Autoři

LI, H.; KHODAEI, Z.; KOTOUL, M.; ALIABADI, M.

Vydáno

16.02.2026

Nakladatel

Elsevier BV

Kniha

Procedia Structural Integrity

Periodikum

Procedia Structural Integrity

Svazek

80

Stát

Nizozemsko

Strany od

23

Strany do

30

Strany počet

8

URL

https://www.sciencedirect.com/science/article/pii/S2452321626001058

Plný text v Digitální knihovně

http://hdl.handle.net/11012/527829

BibTex

@inproceedings{BUT201477,
  author="{} and  {} and Michal {Kotoul} and  {}",
  title="Physics-Informed Neural Networks for Multiscale Large Deformation Analysis of Metamaterials",
  booktitle="Procedia Structural Integrity",
  year="2026",
  journal="Procedia Structural Integrity",
  volume="80",
  pages="23--30",
  publisher="Elsevier BV",
  doi="10.1016/j.prostr.2026.02.003",
  url="https://www.sciencedirect.com/science/article/pii/S2452321626001058"
}