About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Frontiers in Solidification X
|
| Presentation Title |
Decoding Solidification: AI-Driven Insights from Atomic-Scale Interface Characteristics to Microstructure Evolution |
| Author(s) |
Sepideh Kavousi, Mohsen Asle Zaeem |
| On-Site Speaker (Planned) |
Sepideh Kavousi |
| Abstract Scope |
Multiscale computational modeling has become a powerful tool for understanding microstructural evolution during solidification, offering atomic-to-mesoscale insights into interface dynamics, dendrite morphology, and solute segregation. However, the complexity and volume of data generated by these simulations present challenges in interpretation and workflow efficiency. To address this, we apply advanced data analytics and machine learning (ML) tools to extract, accelerate, and generalize the underlying physical relationships. In this presentation, we share our recent developments in integrating ML with molecular dynamics and phase-field modeling to establish a scalable, physics-informed framework for predictive solidification science. Key topics include:
• New insights into process–structure relationships;
• Uncertainty quantification and propagation across modeling hierarchies;
• Defect-induced modulation in interfacial anisotropy and its influence on microstructure evolution;
• Accelerated development of interatomic potentials tailored for high-temperature conditions. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Solidification, Machine Learning, Modeling and Simulation |