Investigation of Low-High Fidelity Turbulent Models in Scramjet Engines

In the present work, we aim to investigate the predictive capabilities of low and high-fidelity turbulence models and understand how they perform in complex conditions in a scramjet engine. In this study, we use the DLR experiments [1] to present a comparative performance assessment of low to high fidelity turbulence models. The DLR case is interesting since it exhibits turbulence due to wall effects as well as due to shear layer. Results show relatively good agreement when compared against the validation data set from the experiment. DHRL was able to resolve more turbulence structures compared to all other models. HRL showed better prediction of the momentum deficit in the wake of the strut than all other models. Additionally, results for reacting case show that choice of turbulence models affects the intensity and lift-off distance of the flame where RANS and DHRL show the most rigorous flame and a smooth peak in temperature profile near the base of the strut at the centerline while IDDES and HRL show no combustion at the same location. Comparison against computational results shows an underprediction of the flame lift-off distance.