RANS CFD Applied to 2D Canonical Shock Wave Turbulent Boundary Layer Interaction

In high-speed flight, Shock Wave Turbulent Boundary Layer Interaction (SWTBLI) is a common occurrence in a wide range of external and internal flow problems affecting aircraft, missiles, rockets, and other projectiles. The range of flows in which SWTBLI play a significant, if not dominant role spans transonic to hypersonic and sea level to high altitude flows. The fundamental physics of SWTBLI are often best examined in canonical situations. A Reynolds Averaged Navier Stokes (RANS) Computational Fluid Dynamics (CFD) method prediction tool for simulating high speed SWBLI flow fields for 2D canonical configurations was enhanced in this research. The 2D canonical configurations represented in the research comprised: 1) oblique impinging shockwave onto flat plate boundary layer and 2) compression corner. RANS CFD method simulations are extensively used for engineering prediction of SWTBLI in high speedflows. However, in the presence of strong shockwaves significant disagreement with experimental data is typically observed as all pertinent flow physics are not suitably captured by standard RANS simulations. Non capture of the shock unsteady motion is known to be one of the key flow physics not accounted for by the turbulence models used in RANS methods and therefore a significant reason for their poor performance in strong SWTBLIs. Methods allowing for the effects of shock unsteadiness and variable turbulent Prandtl number PrT have therefore been integrated into the turbulence models of the RANS CFD method developed in this research in an attempt to capture more of the pertinent flow physics associated at SWTBLI regions and so improve the calculations. The commercial software ANSYS Fluent tool was used to develop the RANS CFD method high speed SWTBLI flow field prediction capability presented in this paper. The aerothermodynamics data specifically used to evaluate the CFD method SWTBLI predictive capability for canonical configurations comprised experimental measurements obtained at SWTBLI regions for surface heat flux and pressure and Schlieren images displaying captured flow field shockwave topology.Overall, results obtained from the modified turbulence models show a substantial improvement in prediction of key aerothermodynamic parameters as compared to results obtained using the standard turbulence closure models. A case in point included significant reduction in overprediction of the critical design peak surface heat flux parameter situated at the reattachment shock region that plagues many models. For the oblique impinging shockwave configuration test case undertaken for example, the computed reattachment shock region peak surface heat flux error margin achieved against experiment by the standard models was demonstrated to be as high as 400% approximately. This error margin was reduced to as low as 15% by the modified models.