Wavelet-based denoising for wind turbine blades damage detections using audio signals

Zikang Chen; Junhong Zhang; Weihao Zhuang; Fu Teng; Jing Wang; Qi Xi

doi:10.1117/12.3004547

25 September 2023 Wavelet-based denoising for wind turbine blades damage detections using audio signals

Zikang Chen, Junhong Zhang, Weihao Zhuang, Fu Teng, Jing Wang, Qi Xi

Proceedings Volume 12788, Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023); 127882E (2023) https://doi.org/10.1117/12.3004547
Event: Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023), 2023, Kuala Lumpur, Malaysia

Abstract

Currently, the detection of wind turbine blade damage mainly relies on regular plan-based maintenance and manual inspections. In this study, a method for extracting audio features and detecting damage in wind turbine blades with wavelet denoising is proposed. This method first uses wavelet denoising to process the original audio signal, the denoised audio is then split into frames with Hamming windowing function. After that, multi-scale features are extracted in both time and frequency domains. Principal component analysis is used to reduce the dimensionality of the features, and clustering canters are obtained through K-means clustering analysis. Finally, Gaussian distribution outlier detection is used to detect audio signals from damaged blades. Experimental results using lab-generated audio data show that the proposed method has high accuracy and strong robustness in detecting wind turbine blade damage.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Zikang Chen, Junhong Zhang, Weihao Zhuang, Fu Teng, Jing Wang, and Qi Xi "Wavelet-based denoising for wind turbine blades damage detections using audio signals", Proc. SPIE 12788, Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023), 127882E (25 September 2023); https://doi.org/10.1117/12.3004547

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available