About this Abstract |
Meeting |
MS&T24: Materials Science & Technology
|
Symposium
|
Additive Manufacturing Modeling, Simulation, and Machine Learning: Microstructure, Mechanics, and Process
|
Presentation Title |
Machine Learning Informed Inverse Design of an Additively Manufacturable Al Alloy Strengthened by Both Eutectic and Nanoprecipitates |
Author(s) |
Zhaoxuan Ge, S. Mohadeseh Taheri-Mousavi |
On-Site Speaker (Planned) |
Zhaoxuan Ge |
Abstract Scope |
Two major strengthening mechanisms exist in Al alloys: eutectic strengthening and precipitation strengthening. However, the challenge is to design an alloy that is synergically strengthened with both mechanisms. Here, we leverage machine learning and CALPHAD-based ICME techniques to explore a high-dimensional design space and find potential design pathways. Our model alloy is an additively manufacturable Al-Ni-Er-Zr multi-component alloy. Inverse design via Bayesian optimization will be applied with objective functions constructed from the (1) eutectic phase development, (2) nanoprecipitation, and (3) combined features together with printability. Multiple scenarios will be discussed. Our design concepts can be applied to various Al model systems to be used in high-strength applications. |