LES of shock wave / turbulent boundary layer interaction

eferred to as “higher-Reynolds-number experiment” (figure 1.4a). By
matching directly the experimental parameters, the prediction quality
of the employed sub-grid-scale model can be assessed without further
assumptions. Given a successful validation, the computational results
provide a reliable source for further analysis.
While the reference experiment is done for the entire compressiondecompression
ramp configuration, in the simulation we split this configuration
into two parts, namely, the compression- and the decompressioncorner
part. The reason for splitting the problem into two parts is
twofold. First, the estimated computational cost of well-resolved LES
for the full configuration is rather large and leads to rather long computational
times. Second, the compression-corner interaction is sufficiently
complex to justify a separate investigation. Preliminary compressioncorner
results based on the analysis of limited statistical data were given
by Loginov et al. (2004c, 2006b).
In section 2 we provide the problem formulation and give a brief summary
of the simulation method which is essentially the same as in Stolz
et al. (2001a). A simulation for the turbulent boundary layer along a flat
plate was used to provide inflow data for the compression ramp. Results
for this precursor simulation are summarized in section 3. The main
subject of this paper is the analysis of the compression-corner flow in
section 4. A discussion of results of the successive decompression corner
simulation is given in section 5. The last section 6 gives a summary and
final conclusions.
15