Scientific Machine Learning in Advanced Manufacturing
Thursday, March 27, 2025
12:00 pm - 1:00 pm
Predicting grain formation during alloy solidification through simulations can be computationally expensive. This talk explores GrainNN, a reduced-order model for epitaxial grain growth in additive manufacturing conditions. GrainNN is a sequence-to-sequence, long-short-term-memory neural network that evolves the dynamics of manually crafted features. Results have demonstrated the faster performance of GrainNN compared to phase field simulations. The model enables predictive simulations and uncertainty quantification of grain microstructure that were not previously possible.
About the Speaker
George Biros, Ph.D., is a professor of mechanical engineering and chair in simulation-based engineering sciences at the University of Texas-Austin Oden Institute for Computational Engineering and Sciences. He earned his Ph.D. in Computational Science and Engineering from Carnegie Mellon University. Along with collaborators, Biros has received the Association for Computing Machinery Gordon Bell Prize twice.
- Location
- Online
- Cost
- Free