Effects of Turbulent Boundary Conditions on the Prediction of the Secondary Flow Field in a Linear Compressor Cascade
For most technical applications, simulations of the Reynolds-Averaged-Navier-Stokes equations has become a standard analysis tool, since it brings a good compromise between computational accuracy and costs. However, turbulence models have to be implemented to close the system of differential equations. To study the effects of turbulent boundary conditions on the prediction of the secondary flow field in a linear compressor cascade with tip clearance, the state of the art RANS solver TRACE in conjunction with Wilcox’ k-ω-turbulence model is used. Besides a stagnation point anomaly prevention, no turbomachinery specific modifications of the turbulence model are applied. Transition is not considered. The current investigations focus on the influence of the imposed turbulent inlet quantities (k0, ω0) on the development of the wall-bounded flow in the cascade. The turbulent kinetic energy k is basically described as a function of the turbulence intensity level measured in an equivalent experimental setup. For the reconstruction of turbulent fluctuations beyond measuring accessibility in the vicinity of the wall, an analytical approach is proposed and validated with DNS data of turbulent flat plate and fully developed channel flows. To identify the influence of different dissipation rates ω on the characteristics of the secondary flow, the free stream turbulent length scale LtFS is varied in four steps ranging from 0.2 up to 5 millimeters. Additionally, the effects of different span-wise length scale distributions across the inlet flow boundary layer are considered.