About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Advances in Magnetism and Magnetic Materials
|
Presentation Title |
Discovery of Novel Magnetic Fe-Co-Ni Alloys by Machine Learning and Combinatorial Experiments |
Author(s) |
Raju V. Ramanujan, Shakti P Padhy, Li Ping Tan, Varun Chaudhary |
On-Site Speaker (Planned) |
Raju V. Ramanujan |
Abstract Scope |
A combination of machine learning (ML) and combinatorial experiments was deployed for the accelerated development of superior magnetic materials relevant to electrical systems. A database of Fe-Co-Ni-based alloys, incorporating multiple property data, was curated from the literature. A multi-output ML model was developed to predict the properties of Fe-Co-Ni alloys. Multi-objective Bayesian optimization identified promising alloy compositions with superior property combinations, e.g., Fe65.8Co28.7Ni5.5 and Fe61.5Co23.1Ni15.4. These compositions were experimentally validated; the results for arc melted samples were within 8 to 10% of the predicted values. Combinatorial experiments were employed to explore the Fe-Co-Ni composition space; promising compositions such as Fe12Co33.1Ni54.9 and Fe36.5Co55.1Ni8.4 were pinpointed. Our results demonstrate the potential of our accelerated approach to expediting materials discovery. This work is supported by the AME Programmatic Fund by the Agency for Science, Technology and Research, Singapore under Grant No. A1898b0043, and Production Area of Advance (AoA) at Chalmers University of Technology. |
Proceedings Inclusion? |
Planned: |
Keywords |
Magnetic Materials, Machine Learning, Computational Materials Science & Engineering |