About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Environmental Degradation of Multiple Principal Component Materials
|
Presentation Title |
Accelerating the Design and Discovery of Tribocorrosion-resistant Metals by Interfacing Multiphysics Modeling With Machine Learning and Genetic Algorithms |
Author(s) |
Yucong Gu, Wenjun Cai, Lin Li |
On-Site Speaker (Planned) |
Lin Li |
Abstract Scope |
Tribocorrosion, a complex phenomenon involving simultaneous mechanical and corrosion attacks on metal surfaces, presents significant challenges in maintaining the performance of alloys under complex service conditions. In this study, a novel approach is introduced, combining machine learning (ML) and genetic algorithm (GA)-based optimization, to design aluminum-based alloys specifically tailored for tribocorrosion conditions. By employing a multi-physics finite element analysis (FEA) model and identifying key material parameters, the ML model, utilizing ensemble artificial neural networks, accurately predicts tribocorroded surface profiles and total material loss. The GA optimization, in conjunction with the developed ML model, efficiently identifies the optimal material properties, encompassing both mechanical and electrochemical factors, that yield superior tribocorrosion resistance within the explored space. The effectiveness of the proposed design framework is confirmed through high-fidelity FEA simulations. This data-driven approach can be customized to guide the development of tribocorrosion-resistant materials across diverse applications, extending beyond the realm of aluminum alloys. |
Proceedings Inclusion? |
Planned: |
Keywords |
Environmental Effects, Computational Materials Science & Engineering, Modeling and Simulation |