Noise is an unwanted by-product of many engineering activities and as such it is considered to be a major environmental problem. In this proposal we will be concerned with noise produced by flows, which is also known as aeroacoustics. The most well-known example is airplane noise as produced by the jet engine but also by the boundary-layer flow on the fuselage. Other examples of applications where noise is generated by a flow, are: the wake of a high speed train, a combustion flame, the blades of a wind turbine and at smaller scale the ventilator noise in personal computers. In some applications noise may also generate vibrations that can damage the equipment.
Fifty years ago, Lighthill has developed a theory which identifies the acoustic sources in a flow. It turns out that the most important sound source is related to the non-linear terms in the equations of motion. These terms are in particular important in a turbulent flow. However, turbulence is characterized by time and space fluctuations in the velocity and this results in a complexity, which prevents analytical computation or even direct measurement of the acoustic source term. Only the so-called far-field sound radiated by the source can be measured. For a long time it was not possible to calculate the sound source term numerically. However, with modern computer facilities turbulent flows can be simulated in detail and from the results the acoustic source terms can be calculated.
In the present research of the proposer, simulations of turbulent flows have been carried out in order to identify and compute acoustic sources in a turbulent flow. In this proposal these simulations are to be extended to turbulent flows at low Mach numbers and for a large range of Reynolds numbers. At the same time it is the aim to perform experiments for exactly the same geometry as used in the simulations. This will allow a thorough validation of the simulations and their underlying approximations. After such validation has proved that the simulation technique is realistic, it becomes possible to address the several open issues in computational aeroacoustics such as, e.g. Reynolds number effects, effects of source convection and sound refraction by temperature gradients. The ultimate goal is to develop numerical tools by which one can predict and compute sound emissions of complex turbulent flows.
Keywords: Aeroacoustics, Turbulence,
Computational Fluid Dynamics (CFD), Direct Numerical Simulation (DNS),
Large-Eddy Simulation (LES).