About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
AI/Data Informatics: Computational Model Development, Verification, Validation, and Uncertainty Quantification
|
Presentation Title |
A Machine Learning Approach to Optimize T5 Heat Treatment Conditions for Al-Si Alloys |
Author(s) |
Dongwon Shin, Tomas Grejtak, Sun Yong Kwon, James Allen Haynes |
On-Site Speaker (Planned) |
Dongwon Shin |
Abstract Scope |
We introduce a materials data analytics workflow for predicting the optimal T5 heat treatment conditions for Al-Si alloys. Samples collected from a step mold are utilized to create a comprehensive training dataset. We have systematically varied T5 heat treatment conditions, i.e., temperature and duration, and measured the hardness and electrical conductivities. Through the use of a trained surrogate model, we successfully identify T5 heat treatment conditions that lead to both increased hardness and electrical conductivities for a commercial Al-Si alloy. Subsequent experimental validation is performed to confirm the accuracy and effectiveness of the predicted optimum T5 conditions. We also discuss the impact of the present study to rapidly optimize T5 condition of Al-alloys for automotive high-pressure die casting applications. This research was supported by the Vehicle Technologies Office, the U.S. Department of Energy. |
Proceedings Inclusion? |
Planned: |