Acoustic source analysis of a supersonic rectangular chevron jet

Abstract

In this thesis, acoustic source terms corresponding to Goldstein's generalized acoustic analogy are computed from a high-fidelity simulation of a supersonic jet issuing from a rectangular nozzle with chevrons. The simulation data are validated against experimental measurements from a flow configuration involving a nozzle of precisely the same geometry. A statistical description of the simulated flow field is established in detail including an in-depth look at first, second, and fourth order statistics. This thesis investigates the theoretical underpinning of reduced-order acoustic source models by testing the assumptions of quasi-normality and statistical axisymmetry. First, the quasi-normality hypothesis is tested using Millionshchikov's identity. This identity allows fourth order acoustic source statistics to be expressed in terms of second order statistics. This is a simplifying assumption upon which most of the models rely. In addition, local statistical axisymmetry is tested using basic quadratic forms of the fourth order correlation terms to determine if local fluctuations in the transverse directions of the flow are equivalent. It is found that the flow field is not quasi-normal in the axial direction but is however quasi-normal in the transverse directions. Our analysis also shows that the flow is locally statistically axisymmetric close to the edges of the flow field but not near the center. Previously used acoustic source models are fit to the fourth order correlation statistics. Specifically, this thesis performs a detailed analysis of four different models: the Gaussian model, moving-frame model, fixed-frame model, and modified-distance model. These ii models were previously used to describe correlation data from axisymmetric jets. In this thesis, we assess the accuracy of these models in the context of highly complex nozzle shapes. The latter three models are found to be similar in accuracy, while the Gaussian model is found to be a poorer fit. The thesis concludes with an analysis of the large scale turbulent structures in the flow field. It is observed that there is noticeable large scale coherence near the edges of the flow. Therefore, since large scale coherence is a primary mechanism of sound generation, it is believed that the large scale turbulence significantly contributes to the sound generated from this complex flow field.

Similarity scaling of jet noise sources for low-order jet noise modelling based on the Goldstein generalised acoustic analogy

“…This is true both for the experiment and eddy-resolving (e.g. Large Eddy Simulation (LES)) computational modelling [22][23][24][25] and leads to unwanted assumptions and artificial calibration parameters based on the farfield noise measurements, which implicitly involves some (unwanted) scaling of the far-field propagation effects and makes the entire low-order jet noise prediction scheme less robust.…”