| The wind turbine noise simulation analysis model established in this study could better predict noise at a horizontal distance longer than twice the rotor radius from the turbine tower with an error of not more than ±2.6dB(A). Ka(f) curve and directivity in this model were further modified based on a brief review of existing modification process of Chinese wind turbine noise simulation model. On the basis of the model established above, a user-friendly wind turbine noise simulation analysis platform to predict noise of various wind turbine types and sound field under different running conditions was developed through MATLAB program language. Relationship between aerodynamic noise and hub height, rotor radius, blade chord length as well as blade rotation speed was respectively studied through the platform built, and suggestions to optimize these parameters were offered.Different sound adjustment methods were designed to regulate the original wind turbine noise samples according to the low frequency and the amplitude modulation of the wind turbine noise. Afterwards, the subjective annoyance of noise samples before and after adjustment was evaluated through paired comparison test. Results showed that the methods adding natural wind sound, pink noise on specific frequency band or traffic noise were all failed to improve the subjective feelings on the wind turbine.According to the subjective annoyance masking method for wind turbine noise proposed by the study group, which is based on sound adjustment technology, acoustic signal processing techniques and methods, were explored in order to improve the subjective feelings on the wind turbine. In this exploratory study, wind turbine noise was expected to be modulated into the noise sound like natural wind noise, by means of adding kinds of narrowband pink noise with particular time-frequency domain characteristics. |
PDF Full Text Request