About this Abstract |
Meeting |
12th International Conference on Magnesium Alloys and their Applications (Mg2021)
|
Symposium
|
12th International Conference on Magnesium Alloys and their Applications (Mg 2021)
|
Presentation Title |
Towards the Future of Alloy Design Using Artificial Intelligence: The KASSANDRA Method |
Author(s) |
Luis Angel Villegas Armenta, Konstantinos Korgiopoulos, Christina Katsari, Mihriban O. Pekguleryuz |
On-Site Speaker (Planned) |
Luis Angel Villegas Armenta |
Abstract Scope |
The design of novel lightweight magnesium alloys is an essential tool to face future challenges; from reducing vehicle greenhouse emissions to improve the performance of biodegradable implants, being able to quickly improve specific magnesium properties is of paramount importance to succeed. However, the traditional methodology that is used to develop novel magnesium alloys consists of several trial-and-error steps, which can take several years and large amounts of investment until an optimized composition is defined. Artificial intelligence plays a vital role in the Materials 4.0 revolution; it can significantly accelerate the development of novel materials at a lower cost. Hence, our research focuses on the development of novel Mg alloys assisted by machine learning. In this presentation, we introduce the development of a Mg-Sn-Zn-Ca alloy tailored for 3D printing. Using our proprietary machine learning method named KASSANDRA we were capable of identify all the existing design spaces for a given compositional range. Then, the proposed alloys are synthesized through permanent mold casting and tested using low-cost techniques. Our preliminary results demonstrate that it is possible to narrow down the most promising design spaces to focus our attention during the testing phase, hence reducing the need for multiple iterations to obtain a complex optimal composition. |
Proceedings Inclusion? |
Planned: At-meeting proceedings |