RANS-Based mixing noise predictions of installed asymmetric jet flows

Abstract

As the world’s dependency on air travel and transportation increases, the public’s concern over noise pollution around airports grows. Engine manufacturers must meet increasingly stringent noise certification targets while designing new products. Traditional noise analysis techniques often assume that jets are axisymmetric which is not true for realistic geometries. Therefore, a more detailed analysis is needed to see how to model complex asymmetric jets. A method has been previously developed at the University of Southampton, called Light hill’s Acoustic Analogy with Ray Tracing (LRT), to predict the jet mixing noise for civil aircraft engines. In this thesis, the LRT method is used to analyse the change in the jet mixing noise that different geometric features cause. By studying how the mixing noise scales and how the source distribution changes with geometry, this can be used to inform decisions for future commercial aircraft nozzle designs. The main contributions of this work are as follows: RANS has been shown to predict the flow field of the isolated axisymmetric and asymmetric jets that have been studied. This allows the LRT acoustic model to predict the change in the isolated jet mixing noise to within 0.5dB as the geometry changes. The azimuthal variation in the noise is accounted for in the ray tracing calculation. LRT now includes the ability to predict the reflected mixing noise from solid surfaces. RANS prediction for static installed cases showed good agreement with experimental data underneath the wing, leading to accurate prediction above St = 2 of the high-frequency noise. Despite over-predicting the turbulence levels downstream of the wing trailing edge, this has minimal impact on the far-field mixing noise. It is hoped that this work will be used in the future to help understand the jet-surface interaction that dominates installed jet noise below St = 1.

More information

Identifiers

ORCID for Matthew Wellman: orcid.org/0000-0003-1984-2710

Catalogue record

Export record