Computational Techniques for Turbulence Generated Noise
Doctoral thesis, 2004

Computational techniques for the noise generated by high Mach number subsonic jets have been investigated. The main focus has been on the hybrid noise prediction method SNGR (Stochastic Noise Generation and Radiation), which is based on unsteady source modeling for the inhomogeneous linearized Euler equations (ILEE). The unsteady source model developments include a time filtering technique and the use of a convective operator to evolve the source field in the inhomogeneous mean flow of a jet. The unsteady source modeling also includes anisotropy in terms of Reynolds stresses as well as length scales. Inhomogeneous linearized Euler equations in conservative formulation are derived and it is shown that when a proper source field is specified, the proposed ILEE accurately predict the sound generation and propagation in inhomogeneous flows. The proposed SNGR method has been applied to three high Mach number subsonic jets. It is found that, when properly calibrated, the results from the method in terms of sound pressure level directivity are in good agreement with measurements. The spectral content in the emitted sound however, shows some discrepancies, especially at low frequencies. The model is found to accurately predict the increased sound emission of an increased jet exit Mach number, but that a heated jet could not be properly evaluated due to numerical instabilities. An anisotropic model of the two-point velocity correlation tensor for homogeneous turbulence is proposed whose functional form is determined by six scalar correlation functions. The model is still under development but is believed to enable more accurate statistical Lighthill's analogy based noise predictions from anisotropic turbulence.

Author

Mattias Billson

Chalmers, Department of Thermo and Fluid Dynamics

Subject Categories

Mechanical Engineering

Physical Sciences

ISBN

91-7291-423-8

Doktorsavhandlingar vid Chalmers tekniska högskola. Ny serie: 2105

More information

Created

10/8/2017