Trailing edge noise estimation by tomographic Particle Image Velocimetry

Abstract

The feasibility of estimating broadband trailing edge noise with high-speed tomographic Particle Image Velocimetry (PIV) measurements is studied. A thin plate terminating in a sharp trailing edge provides a generic test case for turbulent boundary layer trailing edge interaction noise.Far-field noise is linked to the wavenumber–frequency spectrum of the surface pressure fluctuations in proximity of the trailing edge through diffraction theory. High-speed tomographic PIV measurements return volumetric and time-resolved information about all velocity components for the resolved spatio-temporal scales and can therefore provide the required statistical quantities. For the turbulent boundary layer interacting with the trailing edge, these statistics include the auto-spectral density, spanwise correlation length, and convection velocity of the unsteady surface pressure, which are thus estimated.Acoustic phased array measurements in an anechoic environment provide a reference for comparison. Over the resolved frequency band, PIV based noise estimation results compare favorably with the reference measurements. Especially at lower frequencies, where existing, empirical models for the unsteady surface pressure spectrum are not accurate, tomographic PIV can offer an alternative approach to complex and intrusive model instrumentation for assessing the relevant statistical quantities.