We model far-field sound generated from turbulent jets using input-output analysis. We investigate small perturbations about Reynolds-averaged Navier-Stokes (RANS) solutions of Mj = 1.5 ideally expanded axisymmetric turbulent jets. We consider both isothermal and heated jets with jet-to-ambient temperature ratios Tj/T∞ = 1 and Tj/T∞ = 1.74, respectively. The dynamics of small perturbations are governed by the linearized Navier-Stokes (LNS) equations, accounting for the effects of eddy viscosity. We extend the system state to the acoustic far-field using the linear Ffowcs Williams-Hawkings (FW-H) equation. In this way, our hybrid input-output/FH-W method efficiently connects input uctuations embedded in the jet turbulence to pressure outputs in the far-field. By projecting acoustic far-field data extracted from a high-fidelity large eddy simulation (LES) database onto orthonormal sets of output modes, we demonstrate that a significant portion of the LES acoustic energy may be recovered using only a few output modes. From the resulting acoustic fields, we examine the role of optimal and sub-optimal input-output modes in noise generation mechanisms of high-speed turbulent jets. Finally, we repeat input-output analysis over a wide range of frequencies, and find that our analysis indicates a broadening of the far-field acoustic spectra as the radiation angle increases.
|Original language||English (US)|
|Title of host publication||23rd AIAA/CEAS Aeroacoustics Conference, 2017|
|Publisher||American Institute of Aeronautics and Astronautics Inc, AIAA|
|State||Published - 2017|
|Event||23rd AIAA/CEAS Aeroacoustics Conference, 2017 - Denver, United States|
Duration: Jun 5 2017 → Jun 9 2017
|Name||23rd AIAA/CEAS Aeroacoustics Conference, 2017|
|Other||23rd AIAA/CEAS Aeroacoustics Conference, 2017|
|Period||6/5/17 → 6/9/17|
Bibliographical noteFunding Information:
This work was partially supported by the Amelia Earhart Fellowship. High-fidelity simulations were made available by grants of computer time at the ERDC and AFRL supercomputing centers as well as at the Argonne National Laboratory. The authors also gratefully acknowledge discussions with M. Jovanovi?, S. Lele, T. Colonius, P. Jordan, P. Schmid, and M. Juniper.