About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Phase Stability, Phase Transformations, and Reactive Phase Formation in Electronic Materials XXV
|
| Presentation Title |
Physically Interpreted Machine Learning Predictions of Yield Strength in 7xxx Aluminum Alloys |
| Author(s) |
Chien-Shuo Huang, Yu-Chen Liu |
| On-Site Speaker (Planned) |
Chien-Shuo Huang |
| Abstract Scope |
In this work, we introduce a machine learning model designed to estimate the yield strength of 7xxx series aluminum alloys by analyzing relationships between composition, processing parameters, and mechanical performance. Rather than relying solely on statistical correlations, we assessed the model’s physical validity by comparing its outputs with results from the CALPHAD simulation. Further interpretation was performed through classical precipitation-strengthening theory, focusing on strengthening behavior during both under-aging and over-aging stages. Our results demonstrate that the machine learning model produces strong predictions that not only exceed the accuracy of conventional simulations but also align with metallurgical principles governing phase transformation and hardening. This integrated approach reveals the model’s capability to bridge data science and materials physics, offering a robust foundation for future alloy design and optimization grounded in both predictive power and physical relevance. |
| Proceedings Inclusion? |
Planned: |