| About this Abstract |
| Meeting |
MS&T24: Materials Science & Technology
|
| Symposium
|
Integrated Computational Materials Engineering for Physics-Based Machine Learning Models
|
| Presentation Title |
Developing Data-Driven Strength Models Incorporating Temperature and Strain-Rate Dependence |
| Author(s) |
Nicole K. Aragon, David Montes de Oca Zapiain, Corbett C. Battaile, Hojun Lim |
| On-Site Speaker (Planned) |
Nicole K. Aragon |
| Abstract Scope |
β-tin exhibits complex deformation with a significant strength-dependence on temperature and strain rate, which conventional strength models struggle to characterize accurately. To address this, we trained data-driven models on a set of compression tests and split-Hopkinson pressure bar tests at various strain rates and temperatures using genetic programming to perform symbolic regression. The accuracy and robustness of the data-driven models were evaluated by comparing model predictions to experimental results for data not included in the model training set. In this presentation, the accuracy of the strength predictions from the developed models will be compared to conventional strength models. Finally, to further validate their performance and exemplify their robust nature, the new strength models will be demonstrated using finite element simulations at different temperatures and strain-rates. |