UC Irvine

This work presents advances towards a linear surface-based model for the jet noise source in simple and complex jets of relevance to the propulsion of high-speed aircraft. The model would be informed by low-cost, Reynolds-Averaged Navier-Stokes (RANS) computations of the flow field. It would be a practical alternative to experiments or high-fidelity computations, such as large eddy simulations (LES), both of which are resource-intensive and entail long turnaround times.

This study encompasses one single-stream and two triple-stream jets. The connection between the vortical field and its pressure signature on the edge of the jet is investigated. These regions are represented by the surface of peak Reynolds stress (OSPS) and the “radiator surface” respectively. The radiator surface is the location where the jet noise source model would be prescribed for the computation of its noise propagation.

The single-stream jet is an isothermal round jet at Mach 0.9 calculated by large eddy simulation (LES), which enables the computation of two-point space-time correlations throughout the jet and its near-acoustic field. It is observed that the nature of the space-time correlations is different for axial velocity fluctuations and pressure fluctuations. Velocity-based correlations appear to capture localized turbulent events, whereas pressure-based correlations appear dominated by the interaction of large eddies with the surrounding potential flow.

Of the triple-stream jets, one is coaxial and the other has an eccentric tertiary flow that yields nose suppression in preferred directions. They are computed by LES and RANS and used to verify key modeling assumptions. In particular, the analysis of velocity scales obtained by two-point correlations validates the RANS-based models for the convective velocity Uc of the noise-generating turbulence, which is a crucial factor in noise generation.

The radiator surface is located near the boundary between the rotational and irrotational fields and is defined as the surface on which the Uc distribution equals that on the OSPS. The edge of mean vorticity is nearly coincident with the radiator surface, which suggests a straightforward RANS-based criterion for locating this surface. Additionally, it is found that the edge of the jet is affected by sparse vortices that peel off from the main flow and travel along the radiator surface. Those vortices represent the last remnants of the vortical field and cause a layer of negative pressure skewness at the edge of the jet.

Axial and azimuthal turbulence scales are examined on the OSPS and the radiator surface of the three jets, and compared with scales extracted from RANS. Simple relationships are inferred that may aid the development of rapid predictive models.