Perception vs. Quantification of Wind Turbine Noise

Abstract

The number of wind farms in the world, and in Norway, is increasing. As a consequence, the issue of noise annoyance and possible negative health effects caused by wind turbine noise, has become more prevalent. There are two different measures that are commonly used to describe wind turbine noise: sound emergence, which is the difference between the total sound pressure level and the background noise, and the total sound pressure level. Both are given in A-weighted decibel values.

Some assume that the sound emergence better correlates to noise annoyance than the total sound level. There is no evidence for this assumption. Therefore the aim of this thesis was to better understand sound emergence through the use of a listening test created with GUIDE in MATLAB. To obtain sound samples to use in the listening test, monaural and binaural recordings of wind turbine noise and background noise had to be acquired. 34 test subjects took the listening test, which consisted of 51 sound samples, including six control sounds.

The main objectives of this thesis was to find if the unpleasantness of wind turbine noise at different emergences is level dependent and if spatial information is relevant in the rating of unpleasantness. Audibility of the wind turbine for the lowest sound emergences and sound pressure levels was also investigated.

From the test responses it was found that the unpleasantness related to sound emergence is level dependent. However, this dependency decreases with descending emergence. The responses to the control sounds showed that wind turbine noise is experienced more unpleasant than background noise. A two-sided Wilcoxon rank-sum test showed that spatial information is not relevant in the rating of unpleasantness. The audibility of wind turbine noise seem to depend more on sound pressure level than on sound emergence.