Numerical Investigations On Aero-acoustics Of Wind Turbine Blade Based On Improved BPM Model

With the large-scale development of wind power, wind farms are now more and more close to residential areas, and then wind turbine noise problem becomes more serious. In order to ensure the development of wind power, also to reduce the noise pollution, carrying out the noise researches of wind turbines, is very important in the field of engineering and academic. Airfoil self-noise is the base of blade and wind turbine aero-acoustics, is also the main research object of this paper.The basic knowledge of acoustics is introduced at first, and also the Brooks-Pope-Marcolini(BPM) semi-empirical self-noise prediction model, the compiling of the prediction program. The BPM model program is verified by calculation cases of airfoils of NACA0012, S822 and FX63-137, also the deficiencies existing in the model are pointed out. Aimed at the deficiencies, an improved scheme is proposed, which joins panel method and boundary layer (BL) equations to get BL parameters, improving the BL calculation module of the original BPM model. Through this, an improved BPM model considering airfoil geometric parameters is available, which is verified in this paper. To further understand the characteristics of the self-noise, studies of self-noise under random wind conditions are carried out, quantifying the impacts for low and high frequency noise from the atmospheric turbulence intensity (TI) and the random angle of attack (AOA) of airfoil.Research results show that the BPM model is well applicable of noise prediction for some airfoils, except the airfoils with large camber and/or thickness, which is due to the lack of consideration of impacts for BL from the airfoil geometric parameters. To solve the above problem, adopting the panel method and solution of BL equacion to obtain the BL parameters, replacing the BL module of BPM model, thus an improved BPM model is got, which gives enough consideration for airfoil geometric parameters. The uncertainty study on self-noise under random wind conditions shows that the atmospheric TI, uncertainty of AOA do not significantly affect the average sound pressure level(SPL) of each frequency, and the low-frequency noises are sensitive to AOA changes, while the high frequency noises are more sensitive to atmospheric TI. When the wind speed obeys a normal distribution, the aero-acoustic SPLs still obey normal distributions in different frequencies. However, when the AOA obeys a normal distribution, the aero-acoustic SPLs in different frequencies tend to obey beta distributions, instead of normal distributions.