Detailseite
Projekt Druckansicht

Numerische Untersuchungen zu großskaligen Bewegungen in turbulenten Platten-Grenzschichten mit Druckgradient

Fachliche Zuordnung Strömungsmechanik
Förderung Förderung von 2020 bis 2024
Projektkennung Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 445495622
 
Erstellungsjahr 2024

Zusammenfassung der Projektergebnisse

The present work was performed in the context of the investigation of turbulent superstructures (DFG-SPP 1881). The initial objective of identifying these as a linear instability mechanism of the turbulent mean flow profile and comparing them with structures from the Direct Numerical Simulation (DNS) had to be modified. The first task was to find the most suitable methods for detecting turbulent superstructures in the DNS data. A positive pressure gradient was successfully used to obtain large structures already at lower Reynolds numbers. The detection work then proved that the high- and low-speed streaks of the u-velocity field play an all-determining role. This was demonstrated by studying instantaneous convection velocities of all flow variables (velocity, pressure, density and temperature). At the same time, these represent an objective, i.e. Galilei-invariant structure, which can also be found in the Lagrangian view of the flow field. For this purpose, however, one must deal with the regions of minimal stretching in the flow field, keyword LAMS (Lagrangian Areas of Minimal Stretching). These are located between the usually considered regions of high stretching in the FTLE scalar field. The LAMS analyses carried out in this work provide a new, but overall consistent view of material transport in a turbulent flat-plate boundary layer. Accordingly, coherent fluid regions sink from the laminar external flow through the boundary layer into the region of maximal vortex activity at y+≈ 20, where they are then dissolved by the turbulent mixing process. A similar process begins in the viscous sublayer, from which cohesive regions rise up into this mixing zone. The idea of interpreting turbulent superstructures as an instability of the mean flow has proven to be correct, as a corresponding publication from China has just shown, in which this was achieved with a resolvent analysis. However, the idea that eigenmodes can then also be compared directly with the turbulent fluctuations must be revised. This would require an initial spectral distribution from the simulation, which would then have to be prescribed to the analysis. The results of coupling DNS with inlet boundary conditions from the experiment demonstrate the applicability of the approach for future DNS studies where specific upstream or far-field boundary conditions are required. They could help to reduce the high initialisation costs associated with conventional inlet boundary conditions.

Projektbezogene Publikationen (Auswahl)

 
 

Zusatzinformationen

Textvergrößerung und Kontrastanpassung