About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Mechanical Behavior at the Nanoscale VII
|
Presentation Title |
High Throughput Multi-Objective Optimization of FCC Complex Concentrated Alloys for Extreme Conditions |
Author(s) |
Raymundo Arroyave, Mrinalini Mulukutla, Danial Khatamsaz, Daniel Salas, Trevor Hastings, Daniel Lewis, Nicole Person, Wenle Xu, James David Paramore, Brady Butler, Douglas Allaire, Ibrahim Karaman, Raymundo Arroyave |
On-Site Speaker (Planned) |
Raymundo Arroyave |
Abstract Scope |
Exploring the vast compositional space of FCC Complex Concentrated Alloys (CCA) using traditional materials discovery methods is a daunting task. In this presentation, we discuss a framework that merges iterative multi-constraint multi-objective Bayesian optimization with CALPHAD-based phase stability forecasts and machine learning modeling. This system has been employed to efficiently navigate the compositional domain of FCC CCAs, incorporating three or more elements from Co, Cr, Fe, Ni, V, and Al, to identify alloys with optimal mechanical properties for extreme conditions, specifically high-strain rates. In this work, the objective space is investigated using a combination of small-scale tensile experiments and high-strain rate nanoindentation. In less than nine months, we generated and characterized five iterations (80 alloys), which represent just 0.15% of the total 53,124 alloy possibilities. Yet, this small sample was sufficient to define the Pareto front of the target mechanical properties, showcasing the framework's superior efficiency compared to conventional techniques. |
Proceedings Inclusion? |
Planned: |
Keywords |
ICME, Machine Learning, High-Entropy Alloys |