Performance of turbulence models to predict supersonic boundary layer flows

Abstract

Abstract The aim of this paper is to analyze the compressibility effect and to evaluate the performance of the turbulence models to predict the near-wall compressible flows. For this reason the k - e models without any compressibility correction and the k - ω 2 of the Wilcox-Rubesin model, derived for compressible flows are chosen. Another interest for this paper is the anisotropy phenomenon, which characterizes the wall-bounded flows. In order to seek the behaviour of this feature, the second order model is selected. The calculation has been performed for the supersonic flat-plate boundary layer. The results presented for this simple test case show that all the tested models gave approximately good results. The second order model and the Wilcox-Rubesin model perform better than the other first order models. The incompressible Chien and Launder-Sharma models underestimate the turbulent kinetic energy production. To improve the prediction of this quantity for compressible flows, we choose the Chien model and introduce the correction suggested by Nichols. The results present better agreement with the available data. Moreover, to emphasize the anisotropy given by the second order model, we use the compressibility corrections suggested by Vandromme. The comparison with the first calculations obtained by the non-modified model show some ameliorations in the results.