Propagation of wind turbine noise: measurements and model evaluation

Abstract

Sound propagation from wind turbines through the atmosphere has previously been analytically examined through numerous studies, but the amount of comparisons to measurements is limited. This paper applies three different sound propagation models; The ISO 9613-2 model, the Nord2000 ray tracing model and the WindSTAR-Pro model, to evaluate the noise at receivers in the near- and far-field of a wind turbine. The evaluation is performed by accounting for different meteorological and terrain parameters. For validation purposes, the study further uses two different data sets from measurement campaigns; Loudspeaker noise and wind turbine noise measurements. The validation evidently shows that especially the WindSTAR-Pro model and the Nord2000 model provide good agreement to the measurements for some of the validation cases, while other results are heavily influenced by the turbine wake, signal to noise ratio in the measurements or directivity of the noise source.

An inter-model comparison of parabolic equation methods for sound propagation from wind turbines

“…The significant interferences observed in this study are especially expected to be caused by the single point source representation of the wind turbine. When applied to wind turbine noise propagation, the use of distributed uncorrelated point sources along the rotor plane has previously shown good agreement with wind turbine noise measurements (Nyborg et al, 2022). The reason is that the distributed point sources average out the interference effects from the ground at the lower boundary.…”

“…In this way, the source can be approximated as a point source and the source intensity is well known through microphone measurements 2 m away. The experimental method will briefly be described, and further details about the data selection and the background noise correction method can be found in Nyborg et al (2022).…”

“…The present study is further supported by comparisons to two experimental campaigns. Previously, comparisons between the GFPE model and flat terrain (over water) measurements have been done (Bolin et al, 2014), while the GTPE model has been evaluated and compared to noise propagation measurements by Cao et al (2022) and Nyborg et al (2022). The measurement campaigns included in this paper consider the measured propagation of noise from a loudspeaker attached to the nacelle of a wind turbine in relatively flat terrain (Nyborg et al, 2022), and the measured noise of a wind farm in undulating terrain (Conrady et al, 2020).…”

Optimization of wind farm operation with a noise constraint

“…In previous work with the WindSTAR model 36 point sources have been used (Cao et al, 2022). The computational time of running 36 individual computations for each octave band frequency is however excessive for optimization purposes, and the use of the three distributed point sources has previously shown good comparison with field measurements (Nyborg et al, 2022). Furthermore, the use of three point sources to represent the wind turbine rotor has been compared to one and twenty-one point sources by .…”

“…Both sound propagation models are coupled with the Topfarm optimization framework (Pedersen et al, 2021;Réthoré et al, 2014) and the PyWake framework (Pedersen et al, 2019), which is used for the wind farm modelling. The WindSTAR model has previously been validated against field measurements of sound propagation (Nyborg et al, 2022;Cao et al, 2022) and compared to other sound propagation models in order to further verify it. These studies showed overall good results from the WindSTAR model.…”