No abstract

Free-stream turbulence preceding high-pressure turbine blades has a crucial impact on blade fields including the heat transfer on the wall. Many parameters characterize this turbulence; its intensity, length scales and physical spectrum are addressed in the study of various operating points of the LS89 configuration. Usually, operating points where weak turbulence is injected are well predicted for all fields by Direct Numerical Simulations (DNS) and Large Eddy Simulations (LES). The MUR235 operating point however, with an experimentally injected turbulence level of 6%, remains incorrectly predicted when imposing the experimental values in the simulations. Such difficulties raise many questions amongst which mesh size and turbulent kinetic energy spectrum are of specific importance for LES. Going away from synthetic turbulence injection by imposing a physical energy spectrum can help improving the prediction of heat transfer. From the present study, it seems that turbulent spots developing in a pre-transition region for higher levels of turbulence on the suction side are important features to capture for proper predictions. In parallel, typical structures of boundary layers such as streamwise oriented vortices have been observed and their existence conditions the heat transfer field on the blade wall. From this specific study, all of these physical processes are seen to be highly dependent on the turbulent specification and turbulent transition observed for the MUR235 case. Depending on these inflow specifications, a transitional boundary layer may be encountered upstream of the shock thus modifying the heat transfer profile.

Summary Uncertainty quantification (UQ) is receiving more and more attention for engineering applications particularly from robust optimization. Indeed, running a computer experiment only provides a limited knowledge in terms of uncertainty and variability of the input parameters. These experiments are often computationally expensive, and surrogate models can be constructed to address this issue. The outcome of an uncertainty quantification study is, in this case, directly correlated to the surrogate's quality. Thus, attention must be devoted to the design of experiments to retrieve as much information as possible. This work presents 2 new strategies for parameter space resampling to improve a Gaussian process surrogate model. These techniques indeed show an improvement of the predictive quality of the model with high‐dimensional analytical input functions. Finally, the methods are successfully applied to a turbine blade large‐eddy simulation application: the aerothermal flow around the LS89 blade cascade.

Some possible future High Fidelity CFD codes for LES simulation of turbomachinery are compared on several test cases increasing in complexity, starting from a very simple inviscid Vortex Convection to a multistage axial experimental compressor. Simulations were performed between 2013 and 2016 by major Safran partners (Cenaero, Cerfacs, CORIA and Onera) and various numerical methods compared: Finite Volume, Discontinuous Galerkin, Spectral Differences. Comparison to analytical results, to experimental data or to RANS simulations are performed to check and measure accuracy. CPU efficiency versus accuracy are also presented. It clearly appears that the level of maturity could be different between codes and numerical approaches. In the end, advantages and disadvantages of every codes obtained during this project are presented.