Aircraft noise is an important topic of research, as researchers have been seeking ways to reduce it. The current thesis aims to advance the state of the art of airfoil trailingedge (TE) noise reduction, by experimentally testing numerically optimized TE designs studied by the author, and obtained from semi-empirical models prediction. Different configurations of a non-flat plate, NACA-0012 airfoil, and flat-plate trailing-edge serrations are investigated. A wind tunnel test section that provides the required environment is designed and created. The test section walls have been acoustically treated to simulate an acoustically far-field environment with forwarding flight. The two sides of the wind tunnel test section are fitted with anechoic chambers and lined with acoustic transparency tensioned cloth screens which act as an interface between the test section and the anechoic chambers to provide a smooth flow surface while eliminating the need for a jet catcher and reducing interference effects. Its aeroacoustic performance is measured. Results show that background noise is comparable with other aeroacoustic wind tunnels worldwide. A straight TE airfoil and flat plate were tested as a benchmark, and then compared with serrated trailing-edge geometries. Results show that the serration geometry is effective in reducing noise and that noise radiated from the TE is at least 6 dB higher than the background noise, satisfying the requirements for aeroacoustic measurements.
The ability of the trailing-edge serrations to reduce TE noise is examined through numerical optimization study. Three different serration geometries are optimized for the overall noise from 0.1 kHz to 10 kHz. The noise spectra was initially modeled using semi-empirical models by Howe, for a semi-infinite flat plate, at zero angle of attack and at low Mach numbers. The single-size sawtooth optimization study found the optimum performing TE sawtooth geometry (in terms of noise reduction), which was confirmed experimentally. The measurements confirmed that numerical optimization predicted a much larger noise reduction compared to measured values. Comparison of single-size sawtooth, slit and sinusoidal TE designs show that, while the three geometries reduced noise from a straight TE, sawtooth serrations result in larger noise reduction than singlesize slit and sinusoidal serrations. Further, it was found that the single-size sawtooth TE that produces the overall least noise has the largest feasible amplitude and smallest feasible width.
Noise radiation from optimized trailing-edges with four different sawtooth serrations have been investigated. The noise spectra was modeled using semi-empirical models by Lyu, for a semi-infinite flat plate, at zero angle of attack and at low Mach numbers. It has been shown that the sawtooth TE serration design can have a significant effect on the level of noise reduction. The optimum single-size of sawtooth tends to maximum amplitude hopt = hmax, and minimum wavelength, λopt = λmin. Moreover, the greatest noise reduction of up to 12.89 dB over a wider range of frequencies was found with one of the configurations. Numerical and experimental results are in agreement that optimized serrated trailing-edge configurations can yield less TE noise compared to the traditional straight trailing-edge configuration. The optimization studies underestimated the measurements by about 2 to 12 dB, but the overall SPL over the frequency spectrum for the different serration geometries is very suggestive of good agreement comparing with the previous study [28]. The results are promising and further investigations to upgrade the models could potentially reduce or eliminate the underestimation.