Correlation analysis of flow and sound in non-isothermal subsonic jets based on large eddy simulations

The sound radiation mechanism is investigated in non-isothermal subsonic jets at acoustic Mach number 0.9 and at temperature ratios of 0.86, 1.0, and 2.7, using causality methods on the results predicted by large eddy simulations. Turbulence signals in the jet flows are rearranged according to the acoustic analogies, such as the streamwise and radial shear signals, ux′ and ur′, the self-signal, ui′uj′, the entropy fluctuation, s′, and the vortex norm, |ω|. The ray-tracing method is applied to locate the acoustically correlated regions, and the two-point correlation is employed to analyze the acoustic correlation. The results show that the cancelation between the s′ and ux′, and between the s′ and ur′, vary with temperature ratio, and, thus, may affect the radiation directionality. The density fluctuation, ρ′, suppresses the acoustic correlations of the shear- and self-signals in the cold jet, respectively, but strengthens them in the hot jet; thus, the ρ′ enhances the cancelation both in cold and in hot jets. The intense negative fluctuation of the ux′ is found to be associated with the cancelation mechanism. It induces the peaks of ρ′ and s′ by disturbing the average fields. The jet temperature changes the averaged density and entropy so as to change the cancelation mechanism. Two large-scale acoustically correlated parts of the coherent structures are visualized by the zonal correlation of the ρ′. Their acoustic correlations cancel each other in the hot jets, but enhance each other in the cold jets. The acoustically correlated regions represented by the zonal correlations of the s′ and |ω| are in small scale and gather around the end of the potential core region.