| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Artificial Intelligence Applications in Integrated Computational Materials Engineering
|
| Presentation Title |
Data-Driven Modeling of Dislocations for Multi-Scale Simulations |
| Author(s) |
Nithin Mathew |
| On-Site Speaker (Planned) |
Nithin Mathew |
| Abstract Scope |
Multi-scale modeling of deformation has played a key role in predicting the strength, degradation, and failure of polycrystalline materials. Nucleation, propagation, and interaction of dislocations are key mechanisms for understanding this behavior and these have been primarily investigated through the development of mechanistic and phenomenological models. Advent of exascale computing and advanced machine learning algorithms opens up new venues of data-driven multi-scale models where physics-informed machine learning models, in conjunction with physically-motivated descriptors, can effectively bridge between different computational scales with minimal constitutive and/or phenomenological assumptions. In this talk I will discuss data-driven upscaling applied to modeling of dislocations, primarily focusing on atomistic simulations and discrete dislocation dynamics. Use of uncertainty quantification in conjunction with well-grounded physical constructs enables efficient exploration of the parameter space to learn the underlying physics of dislocation motion and interaction. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Machine Learning, Modeling and Simulation |