Machine Learning for Hydrodynamic Stability

Speaker:  David Silvester, University of Manchester.

Time:       11:00-12:00, Oct. 16^th

Location: 1A200, SIST

Host:   Qifeng Liao

Abstract:

A novel machine-learning strategy for investigating the stability of fluid flow problems is described in this talk. The computational procedure is demonstrably robust and does not require parameter tuning. The essential feature of the strategy is that the computational solution of the Navier--Stokes equations is a reliable proxy for laboratory experiments investigating sensitivity to flow parameters. The applicability of our bifurcation detection strategy is demonstrated by an investigation of two classical examples of flow instability.  The codes used to generate the numerical results are available online. 

Bio:

David Silvester is best known for his work on finite element methods, fast iterative solvers for fluid dynamics, and uncertainty quantification. He has more than 65 refereed publications on topics such as iterative solution of Stokes and Navier–Stokes systems, preconditioning, and error estimation in finite element methods. Silvester's books include Finite Elements and Fast Iterative Solvers: With Applications in Incompressible Fluid Dynamics and Essential Partial Differential Equations (co-authored with David F. Griffiths and John W. Dold).David Silvester has served as the elected President of the UK and Republic of Ireland section of the Society for Industrial and Applied Mathematics (2009–2011). He became a Fellow of the Society for Industrial and Applied Mathematics in 2023.