About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Artificial Intelligence Applications in Integrated Computational Materials Engineering
|
Presentation Title |
Sustainable Aluminum Alloy Design via Computer Vision |
Author(s) |
Fatih G. Sen, Saikrishna Sanniboina, Aaditya Lakshmanan, Heath Murphy, Paul Nolan, Sazol Das |
On-Site Speaker (Planned) |
Fatih G. Sen |
Abstract Scope |
Aluminum has been increasingly the material of choice with circular solutions for packaging, automotive and aerospace. To improve the circularity of aluminum, increasingly larger quantity and variety of scrap has been utilized. This introduces various microstructural features, which significantly impact product performance. It is of higher importance to construct process-structure-property relationships (PSP) to design sustainable aluminum alloys that satisfy the performance requirements. Here, we have employed computer vision methods to build a deep neural network (DNN) using a large set of scanning electron microscope (SEM) images with energy dispersive spectroscopy (EDS) data that can identify phases on given microscopy images. We carried out inverse design approaches to discover aluminum alloy microstructures that can meet the target properties with generative artificial intelligence (A.I.) methods. Using the physics-informed machine learning (ML) models, manufacturing route of alloys and mechanical properties are predicted. Overall, our A.I. based framework accelerates sustainable aluminum alloy design. |
Proceedings Inclusion? |
Planned: |
Keywords |
Aluminum, Sustainability, Computational Materials Science & Engineering |