Parameter uncertainty quantification for a transition model by predicting 3-D transitional flows at hypersonic speeds
摘要
The Reynolds-averaged Navier-Stokes (RANS)-based transition model is widely used in practical applications. However, its implementation introduces several parameters with significant uncertainties, which can affect the accuracy of the simulations. The ensemble Kalman filtering (EnKF) method was used to investigate the parameter uncertainties of a compressible three-equation k-ω-γ transition model for hypersonic transitional flows. The heat flux and transition location were chosen as the observation variables from the experimental results. After the hypersonic flow past a 2-D plate with the 1st Tollmien-Schlichting (T-S) mode and 2nd Mack mode was predicted, C3 was found to be the most sensitive parameter, which corresponds to the Mack mode. The standard deviation of C3 was 11.88% of its mean value. Five parameters were analyzed by predicting the hypersonic flow past a blunted cone at various angles of attack, additionally including the crossflow mode. The results show that the most sensitive parameters are C2 and C7, which correspond to the T-S mode and crossflow mode, respectively. When the angle of attack increases, the uncertainty of the crossflow mode decreases.