Abstract

The aim of this work is to predict the boundary layer transition and the heat transfer on a highly loaded transonic turbine cascade using Large Eddy Simulations (LESs) with prescribed inlet synthetic turbulence. The numerical simulations were performed for the flow in a linear turbine cascade tested at the von Karman Institute for Fluid Dynamic (MUR test case). For the numerical case, two operating conditions with two different levels of free-stream turbulence intensity are evaluated. For the lower turbulence level case (Tu = 0.8%, MUR132) a laminar inflow is used for the LES simulations whereas for the higher one (Tu = 6%, MUR237) the inlet turbulence is prescribed by using the Synthetic Eddy Method (SEM) of Jarrin. The first part of this work deals with the LES setup. The standard Smagorinsky model was used as closure model. A value of the Smagorinsky constant CS = 0.05 was chosen whereas the turbulent viscosity was reduced in the region closest to the wall by changing the definition of the Smagorinsky length scale. To handle the strong fluctuations in the flow field the cell fluxes are computed using the WENO-P scheme. In the second part, precursor RANS and LES simulations are used to set the optimal values of the SEM parameters and to guarantee the correct level of turbulence at the blade leading edge. The turbulence decay of the synthetic turbulence is compared with the one of the RANS κ–ωSST model. Finally, a comparison between experimental and numerical results is done and the ability of LES to predict the boundary layer transition and the heat transfer on the blade surface is evaluated for the two different inflow conditions.

This content is only available via PDF.
You do not currently have access to this content.