Physics-informed neural networks for fluid flows
Frank van Ruiten

Supervisors TU Delft: Deepesh Toshniwal and Matthias Moller

start of the project: February 2020

The Master project has been finished in November 2022 by the completion of the Masters Thesis and a final presentation has been given.

For working address etc. we refer to our alumnipage.

Summary of the master project:
Machine learning is increasingly being used to successfully perform a variety of scientific and engineering tasks. The reason is simple: its huge potential for automating and accelerating tasks that are difficult to handle by conventional techniques. In the field of Numerical Analysis, an interesting research question is: how can learning-based methods be used effectively for numerically solving equations that describe physical processes? This project will investigate this question by focusing on fluid flows, and within the context of the recently introduced physics-informed neural networks (Raissi et al., 2019).

Contact information: Kees Vuik

Back to the home page or the Master students page of Kees Vuik