This $192,372 Project Grant from the National Science Foundation Directorate for Mathematical and Physical Sciences under the Mathematical and Physical Sciences program (CFDA 47.049) supports the development of new methods to systematically explore and predict material microstructures using artificial intelligence techniques. The awardee, George Mason University, will adapt leading data science and machine learning methods to discover a practical representation of microstructure state space that balances retaining predictive information while achieving sufficient dimensionality and generality for a flexible materials database. Specifically, the research aims to define physically-motivated metrics to evaluate microstructure similarity at local and global scales, leverage local metrics with manifold learning to construct coordinate representations for mapping microstructure windows and distributions, and create a proof-of-concept microstructure database. The funding also supports cross-training undergraduate and graduate students in mathematics and materials science between institutions. The two-year project period began September 1, 2022.
Generated 1/6/24, 11:59 PM