Low-Dimensional Signatures of the Sound Production Mechanisms in Subsonic Jets: Towards their Identification and Control

Abstract

Neareld pressure and turbulence velocity data from measurements performed in subsonic jets with Mach numbers of 0.3 and 0.85 are analysed in the context of feedback control of jet noise. The low-dimensional building-blocks of the pressure and velocity elds are extracted, by means of joint Proper-Orthogonal, Fourier- azimuthal Decompositions, and studied with a view to understanding their respective roles in the production of jet noise. These building blocks exhibit many of the characteristics commonly associated with instability waves in forced jets, and it is thus possible to identify their pertinent properties where the production of sound is concerned, properties which will serve as input for an eventual feedback controller. for a reduced order description of complex dynamical systems. At both Syracuse and Poitiers these techniques have recently been applied to free jets with the focus, traditionally turned towards the dynamic of the turbulence, now shifted to consider the mechanism by which this turbulence generates sound. This is a highly complex problem, the science of which, although now over fty years old, has struggled to come up with reliable modelling and reduction strategies. The crux of the problem is to be found in the fact that a subsonic jet constitutes an acoustic source with an efcienc y the order of 0.01 . So, we are faced with a problem where the dynamic we must understand and learn to control is a random function of space and time and has an efcienc y of approximately zero! All however is perhaps not lost. We know that free jets present a certain underlying determinism - the coherent structures, the Kelvin Helmholtz instability, the structured azimuthal modes. These phenomena underlie the global turbulence characteristics of the o w, and have been shown to be important in the sound generation process. If their dynamic, and that of the mechanism by which they produce sound can be understood and modelled, then feedback control of this component of the jet noise may be feasible. The experiments described in this paper aim to do just that. Work has been focused on ascertaining the relationship between the deterministic component of the o w and its sound