About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Artificial Intelligence Applications in Integrated Computational Materials Engineering (AI-ICME)
|
| Presentation Title |
Quantifying Relationship Between Creep and Microstructure of Metals Using Symbolic Regression |
| Author(s) |
Rasika Jayarathna, Benjamin Rhoads, Samrat Choudhury |
| On-Site Speaker (Planned) |
Benjamin Rhoads |
| Abstract Scope |
Creep models provide a rapid quantification of creep in metals and can serve as a powerful alternative to experimentation, which can take years to test materials up to rupture. However, typically used creep models often require many assumptions and simplification. Additionally, many essential elements of the microstructure are not typically considered in these models, causing additional uncertainty. In this study, we use modern machine learning tools such as symbolic regression trained on experimentally measured creep data of pure metals to extract equations governing the creep as a function of stress, temperature, and microstructure features such as grain size, and are thus better able to generalize across different metals. This approach demonstrates the capability of explainable artificial intelligence tool symbolic regression to determine physical principles underlying experimental data to aid in the design and optimization of materials at high temperature and stress over long periods of time. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Mechanical Properties, Machine Learning, Modeling and Simulation |