| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Bridging Scale Gaps in Multiscale Materials Modeling in the Age of Artificial Intelligence
|
| Presentation Title |
A machine learning-assisted dislocation density-based crystal plasticity model for fcc aluminum |
| Author(s) |
Karel M. Ziminsky, Cheng Sun |
| On-Site Speaker (Planned) |
Karel M. Ziminsky |
| Abstract Scope |
This research seeks to utilize machine learning-assisted methods with dislocation dynamics simulations to provide a physical basis for a mesoscale crystal plasticity model. While multiple methods exist for analyzing deformation behavior, these are limited to narrowly defined scales due to either computational expense or phenomenological constraints. This necessitates providing a means of reducing the computation cost while retaining sufficient connection to dislocation dynamics modeling. Machine learning methods will be used to predict and update parameters previously treated as constants used in a dislocation density-based formulation for calculating slip rate within a crystal plasticity finite element model. This will provide a more accurate and flexible model than a traditional empirical crystal plasticity model while being faster than having to rerun traditional dislocation dynamics simulations at identical timescales. This model will be validated with the tensile properties of aluminum. Future work will consist of model applications for understanding creep behavior. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Machine Learning, Modeling and Simulation |