Since it’s discovery a century ago by Ludvig Prandtl, the boundary layer has been a controversial subject of research. It is primarily characterized by the hypothesis that the layer thickness is small when compared to the characteristic length scale of the entire flow which leads to a strong simpliflication of the Navier-Stokes equations. The resulting simpler parabolic equation still contains, however, two non-linear terms that make theoretical approaches extremely difficult.
When the Reynolds number is increased to a high enough value, the boundary layer becomes turbulent due to these nonlinear terms. Turbulence being 3-dimensionnal, multiscale, and unsteady, Extracting Reynolds-averaged equation from the boundary layer equation is presently the simplest and most economical way to model it in practical applications. However, this does not give an adequate solution as a correct modeling of theturbulent term of this equation is still a real challenge. This is a key issue in the perspective of building a reliable numerical model able to predict the flow in many configurations of high practical interest: aircraft, car, boat, energy production, engines…
As was shown very early by hot wire anemometry , the intensity of turbulence is very high and anisotropic over most of the boundary layer thickness, and is maximum near to the wall. From an experimental perspective, this is a challenge that is yet to be solved. The near-wall-turbulence peaks are found to be beyond the present spatial resolution, particularly when increasing the Reynolds number. Moreover, many researchers have shown that near-wall turbulence is not a just random process that can be characterized statistically. A whole zoology of coherent structures have been put in evidence since the nineteen-fifties.
Based on advanced optical metrology, Professor Michel Stanislas at Ecole Centrale de Lille in France, made, in the last 30 years, a personal contribution to this field of research. Hedescribes the tracks he followed in trying to understand the physics of near-wall turbulence in a synthesis article published in Physical Review Fluids (Vol. 2,No. 10) DOI: 10.1103/PhysRevFluids.2.100506.
In this contribution, the author first addresses the canonical flat plate boundary layer problem, firstly very close to the wall and then in the outer region when the Reynolds number is high enough to initiate an outer turbulent peak. With the help of Stereoscopic Particle Image Velocimetry, the author was able to characterize in detail the coherent structure organization. In a second step, he considers the case of a turbulent boundary layer undergoing a mild adverse pressure gradient, which is the most challenging case for modelling, and shows that a strong instability due to the pressure gradient can be hidden in the near wall turbulence..
The study contributes to advancing optical metrology for the purpose of better understanding of turbulence physics and shows thatccessing a large amount of accurate space and time-resolved velocity data in turbulent flows opens the way for a deeper understanding of fine physical details.
Currently RANS models cannot predict the outer turbulence peak which appears and develops in the canonical flat-plate boundary layers, channel and pipe flows where the Reynolds number increases. The author’s results provide a better understanding of the coherent structure organization in this region. However, the link with the very near wall turbulence is yet to be established. This brings in a challenging field of research if the objective to come up with a fully predictive turbulence model is to be met.
The findings of Michel Stanislas evidence the great potential of a joint experimental-numerical method. The insights provided by the current optical metrology paves way for significant improvements in turbulence modeling in the near future.
The author considers that the near wall region is the Achille heel of any turbulence model (RANS, LES, DES, Lattice-Boltzmann…). He would like to encourage young researchers who like sophisticated measurement techniques (based on lasers) and advanced mathematical tools (such as POD and low order dynamical systems) to dive into near wall turbulence as there is still a lot to discover in this field.
Laser light sheet visualization of the coherent structures in a turbulent boundary layer. The boundary layer thickness is 30 cm, the external velocity is 3 m/s. The field of view is about 1.2 m long. The repetition rate is 4 im/s.
Michel Stanislas. Near wall turbulence: An experimental view. Physical Review Fluids 2, 100506 (2017).
Go To Physical Review Fluids