SEPARATING AERODYNAMIC FROM ACOUSTIC-INDUCED VELOCITY ON A PORTION OF AN ACOUSTIC LINER THROUGH MODAL DECOMPOSITION METHODS

The separation of acoustic-induced velocity from the turbulent velocity fluctuations is tackled on a numerical database representing a segment of an acoustic liner subjected to a grazing acoustic wave and a turbulent grazing flow. The evaluation of the acoustic velocity in the current test case has practical implications for the liner’s impedance calculation and sound absorption properties. The separation of the acoustic induced velocity from the turbulent fluctuation is provided by modal decomposition methods: proper orthogonal decomposition (POD), spectral proper orthogonal decomposition (SPOD) and the recently developed canonical correlation decomposition (CCD). For all the decomposition methods, the capability to decompose the acoustic and the aerodynamic component is affected the amplitude of the acoustic waves with respect to the background turbulence. The CCD and SPOD are found to outperform POD when the acoustic amplitude is low compared with the background turbulence. For SPOD, the acoustic forcing frequency needs to be known a-priori or easily identifiable in the spectrum to correctly filter out the acoustic induced velocity. POD and CCD have the advantage of automatically capture non-linear effects due to the vortex shedding which are associated with high order modes.