A systems approach to turbulence and interactions: towards high Reynolds number
This workshop is partly a followup of previous ones, in D’Alembert, Paris, in May 2015 and May 2016. Transport of passive scalar by an isotropic velocity field has been extensively studied. What happens when the velocity flow is rendered anisotropic by various effects, as mean shear and mean stratification, is not well known; triadic closures, with quantitative comparisons to high resolution DNS’s, are well suited for a systems approach to such anisotropic turbulence. In addition, it is informative to compare the passive scalar and the active one, such as the density, temperature or concentration fluctuation with a feedback to fluctuating velocity via buoyancy effcts. Of course, a broader range of models, theories, applications, DNS, LES and physical experiments will be addressed as follows:

Extension of spectral theory towards anisotropic shear flows, with passive and active (buoyant here) scalar field. Application to practical (simpler) models for engineering and environment,

Towards unsteady and inhomogeneous flows. Contribution to a mean flow theory. Wallbounded turbulent flows.

And beyond: improved LES models, flows with solid particles.
About the generic term A systems approach to, we have now a well advanced project in mind, far beyond both the domain of classical hydrodynamic stability and the domain of RANS modelling. With respect to the first theme, the base flow for stability is replaced by a mean flow (Reynolds decomposition), which is not known a priori, and the three couplings are investigated, meantofluctuating, as in Rapid Distortion Theory, feedback from the fluctuatingtomean via generalized Reynolds stresses, and fluctuatingtofluctuating, via elaborate nonlinear models. Particularly the last interaction allows us to investigate very high Reynolds numbers, outside the scope of hydrodynamic stability theory. With respect to RANS modelling, of course one recovers the Reynolds decomposition and the three (meantofluctuating, uctuatingtomean, and fluctuatingtof uctuating) interactions, but the statistical approach is performed scalebyscale, including for instance twopoint spectral models, and gives acces to dominant modes and leading structures of the turbulent flow.
Dates
du lundi 15 mai 2017 au mardi 16 mai 2017
Lieu
École centrale de Marseille, M2P2, domaine universitaire de Château Gombert.
Organisateurs
 Claude Cambon, LMFA
 Pierre Sagaut, M2P2