Development of a general sound source model for wind farm application

Abstract

With the confidentiality of turbine blade geometry, wind farm developers and/or end-users have difficulties to optimize their wind turbine operations and wind farm layouts with respect to noise as existing sound source models rely on detailed blade geometry of wind turbines. This paper presents a new methodology of modeling wind turbine noise that does not need to access the details of blade geometry. The new methodology consists of two steps: First, the candidate wind turbine is approached by up/down scaling from a known wind turbine (for example the National Renewable Energy Laboratory 5 MW wind turbine) according to its operational conditions and limited data provided from the manufacturer; then a sound source model (for example Amiet's model) is applied to calculate the sound pressure level at needed positions. The developed methodology is validated against wind turbine noise measurements for a 4 MW wind turbine with unknown blade geometry; good agreement was found, with an averaged difference of 1.2 dB for various wind conditions under neutral, stable and unstable atmospheric thermostability. By combining with a sound propagation model, the new methodology can be used for designing and controlling wind farms.