About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Advances in Multi-Principal Element Alloys III: Mechanical Behavior
|
Presentation Title |
D-22: Physics Informed Machine Learning Model for Predicting Mixing Enthalpy of Multi Principal Element Alloys |
Author(s) |
Cailey Ruderman, Samuel Vogin, Christopher Tandoc, Berk Soykan, Migual Ferrer, Yong-Jie Hu |
On-Site Speaker (Planned) |
Cailey Ruderman, Samuel Vogin |
Abstract Scope |
In the realm of materials science, refractory high entropy alloys (RHEAs) have garnered significant attention due to their exceptional mechanical and thermal properties. The enthalpy of mixing, a crucial thermodynamic parameter, governs the stability and phase behavior of these alloys, thereby influencing their performance in various applications. However, accurately predicting the enthalpy of mixing for RHEAs remains a challenging task due to the complexity of their composition and structure. In this study, we propose a machine learning model designed to predict the enthalpy of mixing for RHEAs with high accuracy and efficiency. Leveraging a diverse dataset of over 200 ab-initio density functional theory (DFT) Calculations. Our model employs advanced regression techniques combined with physics informed descriptors to produce robust high throughput predictions of mixing enthalpy. The present work also includes methods to identify compositions not currently in the dataset, that have high leverage to maximize model improvement with the least amount of additional data. The proposed model not only offers a valuable tool for materials design and discovery but also contributes to a deeper understanding of the thermodynamic behavior of refractory high entropy alloys. |
Proceedings Inclusion? |
Planned: |