Benchmark characterisation and automated detection of wind farm noise amplitude modulation

Phuc D Nguyen,Kristy L Hansen,Bastien Lechat,Peter Catcheside,Branko Zajamsek,Colin H Hansen

doi:10.1016/j.apacoust.2021.108286

Abstract

Amplitude modulation (AM) is a characteristic feature of wind farm noise and has the potential to contribute to annoyance and sleep disturbance. Detection, quantification and characterisation of AM is relevant for regulatory bodies that seek to reduce adverse impacts of wind farm noise and for researchers and wind farm developers that aim to understand and account for this phenomenon. We here present an approach to detect and characterise AM in a comprehensive and long-term wind farm noise data set using human scoring. We established benchmark AM characteristics, which are important for validation and calibration of results obtained using automated methods. We further proposed an advanced AM detection method, which has a predictive power close to the practical limit set by human scoring. Human-based approaches should be considered as benchmark methods for characterising and detecting unique noise features.

Highlights

Amplitude modulation (AM) of wind farm noise (WFN) is a unique feature known to contribute to annoyance (Ioannidou et al, 2016; Lee et al, 2011; Schäffer et al, 2016) and possibly sleep disturbance (Bakker et al, 2012; Liebich et al, 2020; Micic et al, 2018)
Ratings greater than three were classified as AM, and all other samples were classified as no AM
This study demonstrates that random forest-based AM detection is a good approach for AM classification, and substantially outperforms traditional AM detection methods to achieve classification performance close to that of humans

Summary

Introduction

Amplitude modulation (AM) of wind farm noise (WFN) is a unique feature known to contribute to annoyance (Ioannidou et al, 2016; Lee et al, 2011; Schäffer et al, 2016) and possibly sleep disturbance (Bakker et al, 2012; Liebich et al, 2020; Micic et al, 2018). AM in the context of WFN is defined as a periodic variation in sound pressure level (SPL) at the blade-pass frequency (Bass et al, 2016; Hansen et al, 2017), typically between 0.4 and 2. Is typically most prominent during the evening and night-time when environmental conditions tend to be more favourable for AM (Conrady et al, 2020; Hansen et al, 2019; Larsson and Öhlund, 2012). AM is a highly variable phenomenon, depending on meteorological conditions (Conrady et al, 2020; Larsson and Öhlund, 2014; Paulraj and Välisuo, 2017), distance from the wind farm and wind farm operating conditions (Hansen et al., 2019), making AM challenging to detect. AM is commonly detected using simple engineering methods (Hansen et al, 2017) using specific noise features (single predictor).

Objectives

Methods

Results

Discussion

Conclusion