Predicting sound power radiated from engine bays using decomposition of the vector field of sound intensity
Konferensbidrag (offentliggjort, men ej förlagsutgivet), 2011
Encapsulation of fans, engines and machinery is a standard noise-control treatment. The encapsulations have generally large apertures to allow for airflow cooling. They form partial enclosures with coupling between the complex interior wave field and exterior radiation in an unbounded domain. A typical encapsulation is an engine bay where geometry, position and acoustical properties of screens and absorbers are used to control the radiated noise. The mid- and high-frequency range is often of interest, i.e.~frequencies for which the wavelength is on the order of or smaller than geometrical details. A method suitable for this kind of problem is the intensity potential appraoch.
The intensity potential approach is based on the Helmholtz decomposition of the vector field of time-averaged sound intensity into its irrotational and rotational components. The local power balance in a lossless medium is expressed in terms of the irrotational component only, and results in the Poisson equation for a scalar intensity potential of this component only. It captures the power flow from sources to the surroundings by considering the part of the intensity field that is related to flow of energy. A strength of the method is that no additional assumptions have to be made apart from what is behind the local power balance; it follows directly from the Helmholtz decomposition of the active intensity. Further, it does not violate the geometrical divergence of energy in free-field as the classical equation of acoustical diffusion. It is therefore suitable for cases where a partial enclosed volume has strong coupling to an exterior unbounded domain.
The physical and theoretical background of the method is presented together with evaluations by comparison to experimental data. The comparison to experimental data shows that the method is useful in practical applications. The approach provides a robust prediction tool that has its strength at high frequencies for cases with complex geometry and multiple broadband sources.