Advanced Multiscale CFD and Turbulence Modelling targeting HPC
CECAM-IT-SIMUL, Politecnico de Torino
The scientific areas are twofold, Computational Fluid Dynamics (CFD) at various scales and High Perfomance Computing (HPC). Flows in equilibrium (macro-scale) will be first looked at, focusing on turbulence modelling, followed by flows under non-equilibrum (micro-scale) requiring extra modelling than the simple Navier-Stokes-Fourier equations.
Considered as advanced turbulence modelling, Large-eddy simulation (LES) is a technique which relies on filtering the Navier-Stokes equations, that was first introduced by (Smagorinsky, 1963). It has become much more popular recently because of the computing power increase, even if still rarely used by industry, for instance. Vast computing power is required, because modelling is only carried out at the sub-grid scale level.
Another advanced turbulence modelling technique relying on the Reynolds averaged Navier-Stokes (RANS) equations, and which is able to catch anisotropy in the flows, is based on second order moment closure. The method used by this School will be the SSG model (Speziale, Sarkar and Gatski, 1991), and will be applied to turbomachinery.
Rarefied gas flows and their requirements in terms of modelling will be introduced. The Regularised 26 Moment Method (Gu and Emerson, 2009) which has proven to be able to properly model the flow when the traditional Navier-Stokes-Fourier equations fail (Knudsen number between 0.1 and 1, e.g. corresponding to early/mid-transition regime) will be used.
The School will insist on both interlinked aspects, CFD and HPC, showing that advanced multiscale modelling (turbulence or rarefied gas for 3-D problems) is impossible without HPC. The path for developing/optimising a a massively parallel CFD+HPC software will also be introduced (Fournier et al., 2011).
Charles Moulinec (UKRI - STFC Daresbury Laboratory) - Organiser
Juan Uribe (EDF R&D UK Research Centre) - Organiser
stefano rolfo (STFC) - Organiser