Current Status of Jet Noise Predictions Using Large-Eddy Simulation

TLDR: A survey of the current applications of large-eddy simulation for the prediction of noise from single stream turbulent jets is given in this paper, with special attention paid to relationship between numerical and/or modeling choices and the prediction accuracy.

Abstract: A survey of the current applications of large-eddy simulation for the prediction of noise from single stream turbulent jets is given. After summarizing the numerical techniques used, the data predicted by the simulations are given at conditions from subsonic, heated jets to supersonic, unheated jets. Mach numbers between 0.3 and 2.0 are considered. Following the data presentation, an analysis of the trends exhibited by the data is given, with special attention paid to relationship between numerical and/or modeling choices and the prediction accuracy. The data support the conclusion that the most limiting factor in current large-eddy simulations is the thickness of the initial shear layer, which is commonly one order of magnitude thicker than what is found experimentally. There is also a large amount of uncertainty regarding the influence of the subgrid scale model on the predictions. The influence of inflow conditions is discussed in depth. Uncertainties in the inflow conditions currently prohibit the simulations from reliably predicting the potential core length. The centerline evolution of the mean and fluctuating axial velocity is strongly coupled to the resolution of the initial shear layers, but can be made to agree within experimental uncertainty when sufficiently thin initial shear layers are used. The maximum achieved Strouhal number of the sound in the acoustic far field is 1.5-3.0, depending on flow condition; this limit is due to numerical resources. A listing of some of the open questions and future directions concerning jet noise predictions using large-eddy simulation concludes the survey.