Airfoil noise is an important part for both aircraft and wind turbine noise. In this thesis, broadband trailing edge noise and leading edge noise is investigated by a high-order computational aeroacoustics (CAA) method. Numerical instability suppression methods are proposed for the linearised governing equations to facilitate a stable and accurate numerical result. The proposed methods are developed to minimize the side effect to the solution field. Validation simulations are performed to examine the performance of the proposed methods. It is demonstrated that the proposed methods can suppress the numerical instability and obtain an accurately solved acoustic field. A numerical method is developed for the broadband trailing edge noise study. Based on the connection between boun...[ Read more ]

Airfoil noise is an important part for both aircraft and wind turbine noise. In this thesis, broadband trailing edge noise and leading edge noise is investigated by a high-order computational aeroacoustics (CAA) method. Numerical instability suppression methods are proposed for the linearised governing equations to facilitate a stable and accurate numerical result. The proposed methods are developed to minimize the side effect to the solution field. Validation simulations are performed to examine the performance of the proposed methods. It is demonstrated that the proposed methods can suppress the numerical instability and obtain an accurately solved acoustic field. A numerical method is developed for the broadband trailing edge noise study. Based on the connection between boundary layer turbulence and wall pressure fluctuations, the broadband sound sources are modelled using turbulence statistics. Calculated numerical results are validated through analytical method and experimental measurements. The effects of Mach number, angle of attack, airfoil thickness, and airfoil camber are studied using the proposed method. The leading edge noise of a heaving airfoil is studied in this thesis by using the synthesised turbulence with the linearised governing equations. The unsteady mean flow solver is validated through experimental measurements, and the CAA solver is validated with an analytical theory. The inviscid mean flow assumption in the leading edge noise study is examined for both stationary and heaving airfoils. It is found that the inviscid mean flow leads to a higher leading edge noise prediction at high reduced frequencies, and the difference is larger under a lower mean flow velocity. However, an over-predicted result is obtained for the heaving case due to an additional noise source. The effect of the heaving motion on the leading edge noise is studied using isotropic and anisotropic turbulence. The effect of various heaving parameters on the radiated noise is also investigated.