Research project: Beamforming applications in aeroacoustics
We are working on novel microphone array techniques specifically designed to accurately measure and characterise noise generated by aircraft.
We are working on novel microphone array techniques specifically designed to accurately measure and characterise noise generated by aircraft.
Air traffic has been growing intensively around the world, and has become the de facto modality for international civil transport. Therefore a much larger population is now constantly exposed to aircraft noise, driving research institutes around the world to work on decreasing aircraft noise impact.
In this research project, we aim to develop specialised microphone array designs and evaluate a range of signal processing algorithms with the aim of measuring airfoil noise. We will attempt to optimise the array geometry based on the acoustic properties of a typical airfoil in a wind tunnel, with the aim of optimising the inversion process and achieving a better result in separating and quantifying both leading edge and trailing edge noise.
This work is being sponsored under the "Science Withour Borders" programme of the National Council for Scientific and Technological Development (CNPq) of the Federal Government of Brazil.
Why Airfoil Noise?
Major advances in engine noise reduction have already been obtained in the past 50 years, and airframe noise - generated by the interaction of a turbulent air flow with aircraft structures such as the landing gears and the fuselage - has been identified as the next relevant component in the overall noise emission levels for modern aircraft.
One very important case study in aeroacoustics is the noise generated by an airfoil in a turbulent air flow; this study is relevant for a diverse range of noise sources such as aircraft wings, turbofan engines and wind turbines. The main relevant noise mechanisms in this case are the interaction of the incoming turbulent flow with the airfoil leading edge and the interaction of the boundary layer around the airfoil with its trailing edge.
These broadband noise sources are very closely spaced in typical airfoils, making it difficult for an experimenter to evaluate each source separately. Besides, leading edge noise typically dominates the overall noise emissions, thus masking the trailing edge noise and rendering it “invisible” to a typical microphone array.
Why Designing a Specialised Array?
Microphone arrays can be used to separate the acoustic contribution of multiple sources, effectively creating a “spatial filter”. The effectiveness of such filter is however limited in terms of spatial resolution and dynamic range; the classical beamforming algorithm, for example, will have its spatial resolution dictated by the array length and the frequency of operation, and its dynamic range dictated by its spatial windowing function and the background noise levels.
A range of aeroacoustic-specific array signal processing algorithms have been successfully developed in the past 15 years, such as CLEAN-SC and DAMAS. These algorithms, however, often need the cross-spectral matrix (CSM) as an input, which is a usual intermediate step when computing the classical beamforming algorithm. Thus, it appears logical to assume that an optimised array design will improve the performance of both classical and modern algorithms.
The novel contribution of this work is the analysis of the acoustic characteristics of airfoil noise radiation in wind tunnels in the near field, and the design of an optimised microphone array around these characteristics. We aim to achieve a microphone array design methodology that allows the evaluation of leading edge noise and trailing edge noise separately for a wide range of frequencies.
Fabio Casagrande Hirono, Filippo Fazi