Dear all,
This Thursday (the 25th) at 15h00 we will have the next speaker in our seminar on Machine Learning and Uncertainty Quantification for Scientific Computing. Jurriaan Buist will talk about closure modelling for pipe flows using neural networks. You can find the zoom link and the abstract below.
Best regards,
Wouter Edeling
Join Zoom Meeting https://cwi-nl.zoom.us/j/86539157750?pwd=aUZld01lY0pFQXZBM20vcDgxa3hFQT09
Meeting ID: 865 3915 7750 Passcode: 573238
25 Mar. 2021 15h00: Jurriaan Buist (CWI): Energy conservation for the one-dimensional two-fluid model for two-phase pipe flow
The one-dimensional two-fluid model (TFM) is a simplified model for multiphase flow in pipes. It is derived from a spatial averaging process, which introduces a closure problem concerning the wall and interface friction terms, similar to the closure problem in turbulence. To tackle this closure problem, we have approximated the friction terms by neural networks trained on data from DNS simulations.
Besides the closure problem, the two-fluid model has a long-standing stability issue: it is only conditionally well-posed. In order to tackle this issue, we analyze the underlying structure of the TFM in terms of the energy behavior. We show the new result that energy is an inherent 'secondary' conserved property of the mass and momentum conservation equations of the model. Furthermore, we develop a new spatial discretization that exactly conserves this energy in simulations.
The importance of structure preservation, and the core of our analysis, is not limited to the TFM. Neural networks that approximate physical systems can also be designed to preserve the underlying structure of a PDE. In this way, physics-informed machine learning can yield more physical results.