Newsletter 2024.11 Index
Theme : "Mechanical Engineering Congress, 2024 Japan (MECJ-24)"
|
Modeling of the Reynolds stress transport equation – use of generative AI
 |
| Shinnosuke OBI Keio University |
Abstract
The Reynolds-Averaged Navier-Stokes (RANS) framework has long been the standard tool for engineering computational fluid dynamics (CFD). Most RANS turbulence models rely on the eddy viscosity concept, which omits various physical phenomena inherent to the original Reynolds stress transport equations. Directly modeling the transport equation has been a significant topic within the turbulent flow research community and warrants increased attention, even though complex problems can often be addressed by Large Eddy Simulation (LES) or other advanced CFD techniques. This article presents the author's personal perspective on the future direction of transport equation modeling, diverging from the traditional approach that has focused on the pressure-strain rate tensor for decades. Particular attention is paid to the correlation between the instantaneous pressure gradient and velocity, which directly represents the interaction of vortices, an essential process in turbulent fluid motion. Furthermore, the author emphasizes the need for innovative approaches that incorporate a more comprehensive understanding of turbulent structures and their dynamic interactions, suggesting that advancements in computational power and data-driven methodologies will play a crucial role in the evolution of turbulence modeling. This extended view aims to inspire new research avenues that could lead to more accurate and reliable predictive models in engineering applications.
Key words
Reynolds stress, Transport equations, Velocity-pressure correlations