Optimization of the fluidic control of the separation in a transonic channel flow

Flow control of transonic shock wave / boundary layer interactions is investigated in the context of transonic air inlets. A shock wave / turbulent boundary layer configuration with separation zone is considered in the transonic wind tunnel S8. Simulations using Reynolds Averaged Navier Stokes (RANS) and Zonal Detached Eddy Simulation (ZDES, an hybrid method between RANS and Large Eddy Simulation, LES) modeling are carried out to compute the mean flow and its fluctuations in the interaction region.The main separation occurring in the middle of the test section is controlled by fluidic Vortex Generators (VGs). Using RANS modeling, ten VGs are positioned along the span, upstream of the interaction. A first Kriging algorithm is then used to optimize the pitch and skew angles of the VGs. An optimal configuration found is thereafter defined and a second Kriging algorithm is used to enhance the efficiency of the control of the corner flow by two more VGs, which position and angles are varied. The goal of these optimizations is to minimize the total pressure losses downstream of the interaction. The optimal configuration leads to slightly reverse jets with respect to the main flow. A comparison of the vorticity patterns generated by aligned or reverse jets is therefore conducted.This control set-up and a more standard fluidic control (with jets oriented downstream) are adapted to the transonic wind tunnel test section. These control set-ups, as well as the clean case, are precisely described using notably static and unsteady pressure measurement and tomographic Particle Image Velocimetry (PIV). These measurements allow to precisely quantify the efficiency of the two different control set-ups and confirm the interest of slightly reverse jets in order to improve the efficiency of the control of corner flow separation.