Far-Field Noise Prediction of Wind Turbines at Different Receivers and Wind Speeds: A Computational Study

Abstract

Listen

Translate article

Far-field noise propagation from wind turbines propel development of wind farms to an issue for public acceptance. Airstream contains pressure fluctuations as a result of instability, giving a regular eddy pattern or an irregular turbulent motion which are responsible for the sound produced by wind turbine blades. Aeroacoustic noise emanated from a wind turbine is mainly generated by the interactions of tip and trailing edge of wind turbine blades with the mechanics in wake region such as inflow turbulence structures, boundary layer separation and vortex shedding. Hence, there is a strong necessity for an analytical investigation for noise reducing design and development of the technology in order to further expand wind farms. The objectives of this study are to analyze the far-field aeroacoustics of wind turbines with the purpose of predicting far-field sound pressure levels at different receivers and monitoring total acoustic power captured within wind turbine performance for various wind speeds. Blades are modeled based on NREL S825 airfoil since it has high maximum lift and low profile drag. With the purpose of predicting far-field noise, the Ffowcs Williams-Hawkings (FW-H) acoustics model is the preferred method in order to compute the far-field sound signal which is released from near-field flow. As the key attribute of the research, detached eddy simulation (DES) provides accurate results for the desired simulation since it is a hybrid modeling approach that combines features of Reynolds-averaged Navier-Stokes (RANS) simulation in boundary layers and irrotational flow regions, and large-eddy simulation (LES) in unsteady separation regions. In addition, SST K-Omega detached eddy turbulence model is used due to its good compromise between robustness, computational cost and accuracy. Aerodynamic and aeroacoustic analysis of a wind turbine is performed using a three-dimensional model and a commercial CFD Software, STAR-CCM+. In order to predict far-field sound pressure levels and acoustic powers on different locations, five point receivers are defined downstream of the wind turbine model. Receivers are placed one diameter, D, over the wind turbine rotor blades with 1D, 2D, 5D, 10D and 15D away from the wind turbine that represent receivers 1 to 5. Higher acoustic powers are delivered at closer receivers. It means that acoustic power fades out with larger distances. It is observed that there is a fractional variation of 61%, 17%, 6% and 3% as compared to the receiver 1 for receivers 2, 3, 4 and 5 respectively. Moreover, the results show that variation in total acoustic power is non-linear and higher acoustic powers will be captured for higher velocities. This comparison is done between wind speeds of 10m/s and 15m/s.