Simulations of turbulent boundary layers at high Reynolds numbers

Three-dimensional stereoscopic (Anaglyph 3D image to be viewed with red-cyan glasses) view of a spatially developing turbulent boundary layer.
The study of simplified canonical flows allows for deducing important properties of the physics. Therefore, a number of canonical flow cases have emerged as standard model problems to study wall-bounded turbulence.

In technical as well as geophysical applications, the flow of a fluid around solid bodies is ubiquitous. Although in real cases, the geometry of the immersed object is usually complex and the flow is dominated by pressure gradients in various directions, the study of simplified canonical flows allows for deducing important properties of the physics. For this reason, a number of canonical flow cases have emerged as standard model problems to study wall-bounded turbulence, e.g. channels and pipes. However, the simulation of open, spatially developing flows has only recently become the focus of extensive research interest, mainly due to the very long numerical domains, which in turn cause considerable computational cost. At KTH both large-scale numerical simulations (with up to 10 billion grid points) of turbulent boundary layers and experiments in the MTL wind tunnel of exactly the same physical setup were performed. This showed, for the first time, that excellent agreement with experiments could be obtained, indicating that the numerical techniques are suitable for accurate simulation. This further strengthens the notion of a numerical experiment.

References: 

• P. Schlatter and R. Örlü¸ (2010) "Assessment of direct numerical simulation data of turbulent boundary layers." J. Fluid Mech. 659. 116-126
• P. Schlatter and R. Örlü¸(2012) "Turbulent boundary layers at moderate Reynolds numbers: inflow length and tripping effects." J. Fluid Mech. 710. 5-34