Airframe noise predictions using the Ffowcs Williams-Hawkings equation

Abstract

This paper considers potential sources of error when using the Ffowcs Williams-Hawkings equation to make predictions of airframe noise, which entails a relatively low-speed, uniform incoming flow encountering geometry of varying complexity. Numerical simulations are used to investigate several model problems where Ffowcs Williams-Hawkings integration surfaces are placed on solid surfaces as well as in the flow. Comparisons with the pressure obtained directly from the simulations reveal that when solid surfaces are used, the acoustic calculations can produce erroneous results in upstream directions and when scattering bodies block the line of sight from observers to the source. Using solid surface input data implies ignoring all volumetric source effects, which include noise generation as well as flow effects. Nonuniform flow alone, such as is found in a steady boundary layer, was not found to be a significant source of error, so the amplitude and phase changes induced by turbulent eddies in massively separated flow regions is speculated to be the primary cause of the error.

Numerical Simulation of Grid-Generated Turbulence Interaction with a NACA0012 Airfoil

“…In fact, the differences of the OASPL computed from the present simulation is within 4 dB of the experimental measurements. Lockard [66] performed a study comparing the solid and permeable FW-H formulation in PowerFLOW and demonstrated that there exists some differences in capturing the far field noise spectra from the two formulations, and hence could be of interest in the prospective study. In addition to the far field acoustics, the near-field aeroacoustic characteristics in the form of surface pressure fluctuation spectra is obtained and compared with the experiments, which ensures consistency between the near-field and far field prediction from the present simulation study.…”