About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Circular Metallurgy: Design, Technology, Application
|
| Presentation Title |
Optimisation of aluminum alloy grade system to facilitate mixing and recycling |
| Author(s) |
Dmitry G. Eskin, Tanu Tiwari |
| On-Site Speaker (Planned) |
Dmitry G. Eskin |
| Abstract Scope |
Aluminum alloys have experienced extensive development due to favorable properties and high recycling efficiency. However, proprietary alloy development led to excessive grade proliferation with minimal compositional differences (<1%) within alloy series. Nowadays it hinders recycling, creating clear challenges such as sorting issues, internal scrap mixing and uncontrolled composition of mixed-alloy scrap, complicating production of compositions that can deliver the range of desired properties.
To address these challenges, we used a data-driven optimisation framework to streamline aluminum alloy recycling. By collecting data on chemical composition and properties, we apply machine-learning techniques to identify alloys with balanced performance. An alloy mixing algorithm computes mixing ratios of alloys within sub-clusters to target the optimised compositions. It enables over 50% reduction in grades within all aluminum alloy series and supports more efficient, sorted recycling without loss of properties. Designed for recycling both within and between alloy series, it is validated using metallurgical reasoning. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Aluminum, Machine Learning, Recycling and Secondary Recovery |