Three-dimensional (3D) corner separation in a linear highly loaded compressor cascade is studied by using delayed detached-eddy simulation (DDES) method. This paper studies the flow mechanism of corner separation, including vortical structures and turbulence characteristics. The vortical structures are analyzed and the distributions of Reynolds stresses and turbulent anisotropy are also discussed in detail. The results show that there exist different kinds of vortical structures, such as horseshoe vortex, passage vortex, wake shedding vortex, and “corner vortex.” Before the corner separation forms, the passage vortex becomes the main secondary vortex and obviously enhances the corner separation. At approximate 35% chord position, the corner vortex begins to form, enlarges rapidly, and dominates the secondary flow in the cascade. The corner vortex is a compound vortex with its vortex core composed of multiple vortices. Streamwise normal Reynolds stress contributes greatest to the turbulence fluctuation in the corner region. The turbulence develops from two-dimensional (2D) turbulence in the near-wall region to one-component type turbulence in the corner region. The turbulence tends to be more anisotropic when the flow is close to the endwall within the corner separation region.

Tip leakage vortex (TLV) has a large impact on compressor performance and should be accurately predicted by computational fluid dynamics (CFD) methods. New approaches of turbulence modeling, such as delayed detached eddy simulation (DDES), have been proposed, the computational resources of which can be reduced much more than for large eddy simulation (LES). In this paper, the numerical simulations of the rotor in a low-speed large-scale axial compressor based on DDES and unsteady Reynolds-averaged Navier–Stokes (URANS) are performed, thus improving our understanding of the TLV dynamic mechanisms and discrepancy of these two methods. We compared the influence of different time steps in the URANS simulation. The widely used large time-step makes the unsteadiness extremely weak. The small time-step shows a better result close to DDES. The time-step scale is related to the URANS unsteadiness and should be carefully selected. In the time-averaged flow, the TLV in DDES dissipates faster, which has a more similar structure to the experiment. Then, the time-averaged and instantaneous results are compared to divide the TLV into three parts. URANS cannot give the loss of stability and evolution details of TLV. The fluctuation velocity spectra show that the amplitude of high frequencies becomes obvious downstream from the TLV, where it becomes unstable. Last, the anisotropy of the Reynolds stress of these two methods is analyzed through the Lumley triangle to see the distinction between the methods and obtain the Reynolds stress. The results indicate that the TLV latter part in DDES is anisotropic, while in URANS it is isotropic.

Three-dimensional corner separation is a common phenomenon that significantly affects compressor performance. Turbulence model is still a weakness for RANS method on predicting corner separation flow accurately. In the present study, numerical study of corner separation in a linear highly loaded prescribed velocity distribution (PVD) compressor cascade has been investigated using seven frequently used turbulence models. The seven turbulence models include Spalart–Allmaras model, standard k–ɛ model, realizable k–ɛ model, standard k–ω model, shear stress transport k–ω model, v2–f model and Reynolds stress model. The results of these turbulence models have been compared and analyzed in detail with available experimental data. It is found the standard k–ɛ model, realizable k–ɛ model, v2–f model and Reynolds stress model can provide reasonable results for predicting three dimensional corner separation in the compressor cascade. The Spalart–Allmaras model, standard k–ω model and shear stress transport k–ω model overestimate corner separation region at incidence of 0°. The turbulence characteristics are discussed and turbulence anisotropy is observed to be stronger in the corner separating region