| About this Abstract |
| Meeting |
2024 TMS Annual Meeting & Exhibition
|
| Symposium
|
Computational Discovery and Design of Materials
|
| Presentation Title |
Accelerating Property Predictions in NiTi Shape Memory Alloys with Machine Learning and DFT |
| Author(s) |
Mehran Bahramyan, James G. Carton, Dermot Brabazon |
| On-Site Speaker (Planned) |
Dermot Brabazon |
| Abstract Scope |
This study focuses on developing a machine learning model to predict properties of NiTi shape memory alloy when a third element is present. The model utilizes density functional theory (DFT) calculations as input features and has been trained to predict important properties, such as transformation temperatures, elastic constants, and phase stability. The primary objective is to enhance the prediction process, which is typically time-consuming and computationally demanding when relying solely on DFT calculations for each property of interest. By leveraging machine learning, this model offers a more efficient alternative, enabling rapid property predictions without extensive DFT simulations. This approach holds significant promise for accelerating materials research and alloy design in the field of shape memory alloys. The accuracy and reliability of the machine learning model has been verified by comparing its results to known DFT and/or experimental outcomes, ensuring its suitability for predicting properties in NiTi alloys. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Machine Learning, Modeling and Simulation |