Skip to main navigationSkip to main content
The University of Southampton

Stochastic methods

(Funded by EPSRC (First Grant))
The overall aim of this research project is to develop stochastic-based computational aero-acoustics methods. A particular objective is to devise efficient computational methods to investigate broadband fan noise and use them to assess the influence of various parameters on interaction noise. A synthetic turbulence model has been developed and combined with an aero-acoustic propagation model in order to predict broadband fan noise.

A synthetic turbulence model has been developed that is able to reproduce three-dimensional isotropic unsteady turbulent velocity fields by filtering white noise. The filter is expressed in terms either of the correlation function or the energy spectrum function. In contrast with previous filter-based models, non-Gaussian filters such as Liepmann and von Karman energy spectra can be considered. While Gaussian filters provide better performances from a computational point of view,  an improvement in accuracy is expected from Liepmann and von Karman filters.

The turbulence generator model has been combined with an aero-acoustic propagation model by introducing the turbulent velocity field as a source in the linearized Euler equations. The aero-acoustic model consists of a parallel, multi-block finite difference code in the time domain for the linearized Euler equations where the turbulence is included as a boundary condition on the flat plate. Turbulence is generated with a purely Lagrangian approach where vortex-particles are launched upstream of the aerofoil with a random strength and convected downstream by the mean flow.

Acoustic pressure field around a flat plate interacting with stochastically generated turbulence.

Privacy Settings