ME Talk

Physics-informed neural networks to solve PDEs

Venue: July 18, 2022, 4:00 PM, A. R. Auditorium, ME @ IISc
Speaker: Dr. Sukumar. N, Civil & Environmental Engineering, University of California, Davis

ABSTRACT

Main emphasis will be on introducing physics-informed neural networks (PINNs) and a new approach to exactly impose boundary conditions in PINNs to solve PDEs. First, I will complete what was leftover from the previous lecture – will present numerical results on how virtual element technology can be used to increase the critical time step in linear elastodynamic simulations on poorly-shaped tetrahedral meshes

Submit a Comment

Your email address will not be published. Required fields are marked *