On the Contribution of Higher Azimuthal Modes to the Near- and Far-Field of Jet Mixing Noise

Kurzfassung

The prediction of jet mixing noise is studied using a stochastic realization of the Tam & Auriault source model. The acoustical sources are generated by means of the Random Particle-Mesh Method (RPM), which utilizes turbulence statistics as provided by solu- tions to the Reynolds Averaged Navier-Stokes (RANS) equations. The generated stochas- tic sound sources closely realize the two-point cross-correlation function used in the jet noise model to prescribe the fine-scale sound source. The RPM code is coupled with the DLR CAA solver PIANO. The azimuthal-modal decomposed linearized Euler equations are applied as governing equations. With this approach, it is possible to evaluate jet noise spectra at any position in the near-field. Based on an azimuthal decomposition, 3-D sound radiation from the jet can be reproduced at the computational price of a few axisymmetric 2-D computations. Furthermore, it will be shown, that we are able to verify the imple- mented methodology with the results published for the genuine model. The spectra are correctly predicted in terms of sound pressure levels, Mach scaling exponent and spectral shape. A Strouhal number range of up to St = 10 can be covered using the first six az- imuthal mode components of the broadband source. To reach higher Strouhal numbers more azimuthal modes have to be adopted. The presented results reveal the importance of individual azimuthal contributions to the total spectra. To evaluate the spectra in the far-field, the generated near-field noise is extrapolated with a modal Ffowcs-Williams & Hawkings (FWH) method. For the static single stream jet (Ma = 0.9) two different kinds of extrapolation were used - a simplified extrapolation and the modal FWH method. With this computational case, it was possible to predict a jet noise spectrum in the range of St = 0.01 . . . 20. To investigate the effect of different nozzle configurations on sound gener- ation, different nozzle configurations, i.e. dual-stream nozzles with and without nozzle lip treatments are simulated. Good agreement with experimental data for the noise reduction potential of nozzle lip treatments is found.