Numerical Prediction of Sound Generated from Flows with a Low Mach Number

Abstract

Prediction and reduction of sound generated from a flow are becoming increasingly important in various fields of engineering. This is mainly due to the fact that the magnitude of such sound rapidly increases with increasing flow velocity. In fact it grows with 5th to 8th power of the flow velocity.This paper describes methods to predict sound that is generated from a relatively low-speed flow often times encountered in mechanical engineering, such as flows around propeller fans, high-speed trains and automobiles, and demonstrates their capability at present. It is essentially the unsteady motions of vortices in flow that generate sound. Prediction of sound is thus possible if source fluctuations in flow are adequately computed. Furthermore, because the feedback effects of sound onto the source flow field are in general negligible for such sound that is generated from low-speed flows, it is possible to predict sound by separate computations of the source flow field and resulting acousitical field. Source fluctuations in the flow are computed by a large-eddy simulation (LES) with Dynamic Smagorinsky Model (DSM) and they are fed to the following acoustical computation as input data. Applications of this method will be presented for various classes of flow-induced sound, including bluff body sound, sound resulting from rotor-stator interactions in turbomachinery and sound generated from a transitional/turbulent boundary layer on an aerofoil. Finally, direct Navier-Stokes computations of sound will be briefly described and their possible applications to the prediction of sound generated from low-speed flows will be discussed with some examples.