Comparison of RANS-based methods for the prediction of noise emitted by subsonic turbulent jets

Abstract

In this paper a comparison between four RANS-based jet noise prediction methods is conducted. We aim to evaluate the differences between methods derived from two mathematical frameworks: the Lighthill equation and the linearized Euler equations. A numerical investigation is shown for different geometries and jet operating conditions: (a) single stream jets with Mach number ranging from 0.5 to 1.0 and temperature ratio of 1.0 and 2.0; (b) single stream jets with the effect of a chevron nozzle at Mach number of 0.9 and temperature ratio of 0.86. The necessary information concerning the statistical properties of the turbulent sources was gathered from a RANS CFD simulation using a k–e turbulence model. Sound pressure level plots are shown for an observer at 90◦ related to the jet axis. Experimental data are used to corroborate the numerical predictions. Predictions based on both mathematical frameworks showed good agreement with experiments, however, methods based on the Lighthill equation present a simpler formulation and rely on less empirical coefficients.