Abstract To gain a better knowledge of the bimodal phenomenon in intermittent corner separation, a hybrid RANS/LES turbulence model is applied to simulate the unsteady flow in a linear cascade.… Click to show full abstract
Abstract To gain a better knowledge of the bimodal phenomenon in intermittent corner separation, a hybrid RANS/LES turbulence model is applied to simulate the unsteady flow in a linear cascade. After the validation of the Stress Blended Eddy Simulation (SBES) model with the detailed experimental data, the flow mechanisms of intermittent coherent vortical structures and bimodal phenomenon are clarified according to unsteady numerical results. An indicator is proposed to evaluate the degree of the bimodal phenomenon. Firstly, the intermittency in corner separation is found, which arises from the formation of coherent vortical structures. The collision of the boundary layer and backflow near the trailing edge gives rise to large-scale coherent vortices around the separation surface. Then, according to the histogram of velocity probability density functions (PDF), the bimodal phenomenon found near the separation surface indicates that there are two non-periodic switching flow modes. Combination of vortical structures and velocity distributions manifests that the intermittency of coherent vortical structure is responsible for the switching flow modes around the separation surface. Thirdly, three characteristic parameters are defined to describe the features of two flow modes, corresponding to two Gaussian distributions that PDF decomposed into. The indicator Ra, which measures the difference between two flow modes, is used to evaluate the degree of the bimodal phenomenon. A larger Ra represents the higher probability of bimodal phenomenon occurring in flow fields, which can be used to quantitatively determine the position and fluctuation of the corner separation surface.
               
Click one of the above tabs to view related content.