About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
AI/Data Informatics: Computational Model Development, Verification, Validation, and Uncertainty Quantification
|
Presentation Title |
A Deep Learning Framework for Designing BCC Refractory Multi-principal Element Alloys with Optimized Strength |
Author(s) |
Ali K. Shargh, Jing Luo, Christopher D. Stiles, Jaafar A. El-Awady |
On-Site Speaker (Planned) |
Ali K. Shargh |
Abstract Scope |
Refractory multi-principal element alloys (RMPEAs) are defined as alloys composed of five or more unique elements. It is believed that BCC microstructures promote the mechanical properties of RMPEAs. Nevertheless, the design space that needs to be explored for the desired microstructure and mechanical properties is extremely large. Alternatively, deep learning models can be used to explore the design space quickly. We present a deep learning framework that searches the RMPEAs composition space for alloys with BCC phase and optimized strength. Our framework is trained on a large dataset containing thousands of RMPEAs with several elements and atomic percentages. The RMPEAs are labeled with their constituent phases at different temperatures and their corresponding strength. The constituent phases of the RMPEAs are obtained from CALPHAD calculations, while the strength is calculated using our theoretical based framework. The predictions are validated against high throughput experiments. |
Proceedings Inclusion? |
Planned: |
Keywords |
High-Entropy Alloys, Machine Learning, Mechanical Properties |