A New Method to Predict Damage to Composite Structures Using Convolutional Neural Networks
Laurent Mezeix, Ainhoa Soldevila Rivas, Antonin Relandeau, Christophe Bouvet · 2023
Read A New Method to Predict Damage to Composite Structures Using Convolutional Neural Networks by Laurent Mezeix, Ainhoa Soldevila Rivas, Antonin Relandeau, Christophe Bouvet (2023) 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-10672642.txt.
Description
Laurent Mezeix, Ainhoa Soldevila Rivas, Antonin Relandeau, and Christophe Bouvet use neural networks to predict impact damage in composites in a paper from 2023. Aircraft parts made of fiber composites must tolerate hits, but full simulations are slow and costly. The team gathered damage data from past studies and from their own impact simulations. A data expansion step grew the set from under two hundred cases to several thousand.
They trained image-based networks to predict dent sizes and hidden layer splits. A blend of two network designs beat a single one. Length predictions landed within a fraction of a millimeter. The authors say such models could speed composite design. Human checks would still confirm safety-critical parts.
Catalog record
| CREATOR | Laurent Mezeix, Ainhoa Soldevila Rivas, Antonin Relandeau, Christophe Bouvet |
|---|---|
| YEAR | 2023 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |