Physics-Informed Neural-Network-Based Generation of Composite Representative Volume Elements with Non-Uniform Distribution and High-Volume Fractions
Tianlu Zheng, Chaocan Cai, Fan Yang, Rongguo Wang, Wenbo Liu · 2025
Read Physics-Informed Neural-Network-Based Generation of Composite Representative Volume Elements with Non-Uniform Distribution and High-Volume Fractions by Tianlu Zheng, Chaocan Cai, Fan Yang, Rongguo Wang, Wenbo Liu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12787896.txt.
Description
Tianlu Zheng and four co-authors grow fake composite samples inside a neural net in a 2025 paper. Engineers test fiber composites on digital blocks called volume elements, whose fiber layout mirrors the real stuff. Learning-based makers need huge training sets, and packing fibers past a set crowding hits a wall.
Physics rules are baked into the net's loss instead: fibers may not overlap, the fiber share is set, and edge rules hold. With no training pile at all, the net packs fibers past four-fifths of the space, with clumps where real cloth clumps and chance at wide scales. Bend tests on the blocks follow.
Catalog record
| CREATOR | Tianlu Zheng, Chaocan Cai, Fan Yang, Rongguo Wang, Wenbo Liu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |