Foundation model for composite microstructures: Reconstruction, stiffness, and nonlinear behavior prediction
Materials & Design · 2025-09 · 257 · 114397
DOI: 10.1016/j.matdes.2025.114397 ↗Kurz gefasst
Fachartikel in der Zeitschrift Materials & Design, 2025-09. Verfasst von Ting-Ju Wei und 1 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Artificial Intelligence Research Center, National Science and Technology Council.
Abstract (Originalsprache)
We present the Material Masked Autoencoder (MMAE), a self-supervised Vision Transformer pretrained on a large corpus of short-fiber composite images via masked image reconstruction. The pretrained MMAE learns latent representations that capture essential microstructural features and are broadly transferable across tasks. We demonstrate two key applications: (i) predicting homogenized stiffness components through fine-tuning on limited data, and (ii) inferring physically interpretable parameters by coupling MMAE with an interaction-based material network (IMN), thereby enabling extrapolation of nonlinear stress-strain responses. These results highlight the promise of microstructure foundation models and lay the groundwork for future extensions to more complex systems, such as 3D composites and experimental datasets.
Förderung
- Artificial Intelligence Research Center
- National Science and Technology Council
Zitieren
APA
Wei, T., & Chen, C. (2025). Foundation model for composite microstructures: Reconstruction, stiffness, and nonlinear behavior prediction. Materials & Design, 257, 114397. https://doi.org/10.1016/j.matdes.2025.114397
Vancouver
Wei T, Chen C. Foundation model for composite microstructures: Reconstruction, stiffness, and nonlinear behavior prediction. Materials & Design. 2025;257:114397. doi:10.1016/j.matdes.2025.114397
BibTeX
@article{Wei2025,
author = {Wei, Ting-Ju and Chen, Chuin-Shan},
title = {Foundation model for composite microstructures: Reconstruction, stiffness, and nonlinear behavior prediction},
journal = {Materials & Design},
year = {2025},
volume = {257},
pages = {114397},
doi = {10.1016/j.matdes.2025.114397}
}
EUVIMED