About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Advances in Multi-Principal Element Alloys V: Mechanical Behavior
|
| Presentation Title |
Exploiting Process-Structure-Property Relationships for the Accelerated Development of High Entropy Alloys |
| Author(s) |
Raymundo Arroyave |
| On-Site Speaker (Planned) |
Raymundo Arroyave |
| Abstract Scope |
Recent years have witnessed remarkable progress in the application of Artificial Intelligence and Machine Learning techniques to accelerate materials development. Among these, Bayesian Optimization (BO) has gained prominence in alloy design due to its data efficiency, flexibility, and principled approach to decision-making. However, most BO-driven campaigns have remained microstructure-agnostic, focusing primarily on optimizing chemistry and processing parameters with respect to properties, while neglecting the critical role of microstructure. In this talk, I will present one of the first demonstrations of a microstructure-aware BO framework for alloy development. Specifically, I will discuss our recent efforts to navigate the chemistry–processing design space in pursuit of optimal recrystallized microstructures. These microstructures are evaluated over a range of strain rates to construct a multidimensional Pareto front balancing multiple mechanical properties. This work highlights the importance of integrating microstructural descriptors into the optimization loop and represents a significant step toward microstructure-informed materials design. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
High-Entropy Alloys, Machine Learning, ICME |