Influence of synthetic inlet turbulence on the prediction of low Mach number flows

•LES of turbulent flows are performed with four synthetic turbulence generators.•Synthetic turbulence generators are applied to both non reactive and reactive flows.•The sensitivity of the results to the generator chosen is analyzed in details. Time-resolved numerical simulations of fluid flows, such as Large Eddy Simulations (LES), have the capability of simulating the unsteady dynamics of large scale energetic structures. However, they are known to be intrinsically sensitive to inflow conditions the modeling of which may become a crucial ingredient of the computational model. The present work reports LES of both reactive and non-reactive turbulent channel flows. The flow configuration and associated conditions correspond to those associated with a reference experimental database that has been gathered at the French aerospace Laboratory of Onera. The focus of our study is placed on the influence of synthetic inlet turbulence in this experimental geometry, i.e., the principal aim is to investigate the sensitivity of the flow dynamics and mixing to inflow conditions. The analysis undoubtedly confirms that, even with properly set mean velocity and turbulence kinetic energy profiles as available from experimental data, both non-reactive and reactive flow fields still remain very sensitive to the choice of the synthetic turbulence model. This sensitivity is illustrated for four distinct turbulent inflows obtained from white noise (WN), digital filter (DF) by Klein et al. (2003), random flow generator (RFG) by Smirnov et al. (2001), and synthetic eddy model (SEM) by Jarrin et al. (2009). Finally the results obtained for reactive flow conditions clearly emphasize the influence of the retained model on the chemical reaction rate statistics. This conclusion confirms how relevant are the developments devoted to synthetic turbulence for the computational investigation of turbulent combustion.