Seminar "Gravity wave propagation by physics informed neural networks: nonlinear Schrödinger equations and potential flow."

Speaker: Norbert Hoffmann (Hamburg University of Technology / Imperial College London) 

Title: Gravity wave propagation by physics informed neural networks: nonlinear Schrödinger equations and potential flow.

Abstract:

The propagation of gravity waves is a widespread phenomenon. In this talk I will present some of our recent work on using physics informed neural networks to model and simulate nonlinear gravity waves. The underlying physics will be included on two modelling levels: the nonlinear Schrödinger equation for narrow-banded processes, and potential flow for more general cases. It shows that physics informed neural networks, and some extensions of them, turn out as very handy ways to simulate and predict, but also to take into account measured or synthetic data.