ENGINEERING AND CFD

Shock-wave/boundary-layer interactions (SBLIs) play an important part in many engineering applications. They are common in internal and external aerodynamic flows. A prime example is the transonic flow over an airfoil. The flow is accelerated over the airfoil and a supersonic pocket forms which is terminated by a shock-wave. The shock implies a sudden, almost discontinuous change in all flow quantities. This imposes large unfavourable loads on the airfoil and increases drag. When the shock is strong enough, a recirculation bubble can form at the impingement point and further complicate the situation. Depending on a number of factors the position of the shock begins to oscillate. The onset of this effect, called buffeting, has to be prevented in order for regular flight to continue.

Numerical treatment of such SBLIs is difficult as the important flow features place competing demands on the applied numerical algorithms. As the flow also critically depends on the correct treatment of near wall turbulence, the use of turbulence-modelling techniques that would alleviate the computational burden is not feasible. This necessitates the use of direct numerical simulation (DNS) of the Navier-Stokes equation for the treatment of shock-wave/boundary-layer interactions. As a DNS has to fully resolve all scales of motion present within the flow, the presence of small-turbulence within the boundary-layer leads to very high computational demands which cannot be tackled without the use of high-performance computing methods. Using the HPC infrastructure provided by the HLRS it then possible to perform a detailed direct numerical simulation of a transonic shock-wave/boundary-layer interaction.

The detailed numerical database created this way is then used to study the phenomenon of shock-buffeting and especially its possible origins in detail. Analysis from conventional spectral and time series analysis is combined with various modal decompositions of the flow to identify possible origins of the shock-buffeting.

This information can be used to construct mechanisms of flow control which can be used to inhibit the onset of shock-oscillations in practical applications and as such extend the range of possible operating conditions for effected machinery.

Research Team & Contact Information:
Jens Brouwer, Julius Reiss, Jörn Sesterhenn

Prof. Dr. Sc. techn. habil. Jörn Sesterhenn 
Technische Universität Berlin
Department of Computational Fluid Dynamics
Müller-Breslau-Str. 8, D-10623 Berlin (Germany)
E-mail: Joern.Sesterhenn (at) TU-Berlin.DE