MechE Colloquium: Beyond classical linear instability analysis- nonnormality and nonlinearity

Event details
Date | 07.10.2025 |
Hour | 12:00 › 13:00 |
Speaker | Prof. François Gallaire, Institute of Mechanical Engineering, School of Engineering, EPFL |
Location | Online |
Category | Conferences - Seminars |
Event Language | English |
Abstract: Hydrodynamic instabilities are ubiquitous in natural flows and across engineering. They are classically analyzed by a linear instability analysis, aiming at the determination of the dominant eigenvalues and eigenmodes, deemed to capture the post-bifurcation dynamics and the response to external disturbances. This approach remains insufficient in many flow situations, either because the dynamics is not correctly captured by a small subset of eigenmodes (nonnormality) or because the amplitude of external disturbances is determinant (nonlinearity). The lack of predictive power of classical linear instability analysis is evidenced by challenges in deriving reduced order models that faithfully capture the observed behavior, as well as - more generally - our limited understanding of the physical processes mediating the transition from laminar to turbulent flows and the patterns that may emerge during transition.
I will illustrate our recently developed theoretical tools to model flow such situations, where classical linear stability approaches are insufficient. I will specifically demonstrate the rigorous derivation of amplitude equations in the presence of nonnormality, nonlinearity and stochastic forcing. These amplitude equations allow us to describe, control and physically understand mechanisms underlying the dynamics of many seemingly simple, yet technologically relevant and often mundane flow configurations. After illustrating the concepts in the context of intricate patterns of cold water droplets dripping from the condensation film forming under a Turkish bath ceiling and the surprising nonlinear damping of sloshing motions in a glass of liquid covered by foam, I will demonstrate the power of these novel predictive theoretical tools to flow configurations of interest in a broad range of engineering applications like shear layers, jets, and wakes and characterize their nonlinear stochastic response.
These theoretical models complement simulation and modern data-driven ML approaches, providing mechanistic understanding of mechanisms underlying both important and also simply beautiful flow phenomena.
Biography: François Gallaire graduated from Ecole Polytechnique in 1998 and earned a Master’s Degree in Liquids Physics from Université Pierre et Marie Curie in 1999. He completed his Ph.D. in 2002 at LadHyX, on swirling jet instabilities and vortex breakdown. After six years at CNRS in the mathematics department in Nice, he joined EPFL in 2009, founding the Fluid Mechanics and Instabilities Lab (LFMI). He became an associate professor in 2016 and served as Mechanical Engineering Section director for 5 years. In 2019, he was named a fellow of the American Physical Society. His research focuses on the fundamental description of hydrodynamic instabilities, free interface phenomena, and microfluidics. He is associate editor in Physics Review Fluids.
I will illustrate our recently developed theoretical tools to model flow such situations, where classical linear stability approaches are insufficient. I will specifically demonstrate the rigorous derivation of amplitude equations in the presence of nonnormality, nonlinearity and stochastic forcing. These amplitude equations allow us to describe, control and physically understand mechanisms underlying the dynamics of many seemingly simple, yet technologically relevant and often mundane flow configurations. After illustrating the concepts in the context of intricate patterns of cold water droplets dripping from the condensation film forming under a Turkish bath ceiling and the surprising nonlinear damping of sloshing motions in a glass of liquid covered by foam, I will demonstrate the power of these novel predictive theoretical tools to flow configurations of interest in a broad range of engineering applications like shear layers, jets, and wakes and characterize their nonlinear stochastic response.
These theoretical models complement simulation and modern data-driven ML approaches, providing mechanistic understanding of mechanisms underlying both important and also simply beautiful flow phenomena.
Biography: François Gallaire graduated from Ecole Polytechnique in 1998 and earned a Master’s Degree in Liquids Physics from Université Pierre et Marie Curie in 1999. He completed his Ph.D. in 2002 at LadHyX, on swirling jet instabilities and vortex breakdown. After six years at CNRS in the mathematics department in Nice, he joined EPFL in 2009, founding the Fluid Mechanics and Instabilities Lab (LFMI). He became an associate professor in 2016 and served as Mechanical Engineering Section director for 5 years. In 2019, he was named a fellow of the American Physical Society. His research focuses on the fundamental description of hydrodynamic instabilities, free interface phenomena, and microfluidics. He is associate editor in Physics Review Fluids.
Practical information
- General public
- Free