Temperature prediction of wind turbine bearing based on multichannel parallel deep residual neural network

Ke-jia Zhuang; Cong Ma; Li Zou; Xin-yu Yang; Zheng-kun Xie; Jun Hu

doi:10.1117/12.3014857

4 March 2024 Temperature prediction of wind turbine bearing based on multichannel parallel deep residual neural network

Ke-jia Zhuang, Cong Ma, Li Zou, Xin-yu Yang, Zheng-kun Xie, Jun Hu

Proceedings Volume 12981, Ninth International Symposium on Sensors, Mechatronics, and Automation System (ISSMAS 2023); 129815N (2024) https://doi.org/10.1117/12.3014857
Event: 9th International Symposium on Sensors, Mechatronics, and Automation (ISSMAS 2023), 2023, Nanjing, China

Abstract

Temperature, as an important indicator in wind turbine condition monitoring, can reflect the working state of bearing promptly and intuitively. To enhance feature extraction capability and temperature prediction accuracy, a multi-channel parallel deep residual neural network is proposed. The model consists of a parallel network composed of Deep Separable Convolutional Neural Network (DSCNN) and Bidirectional Long Short-Term Memory (BiLSTM), enabling the model to extract spatial and temporal features effectively. Additionally, the residual connections are employed to avoid the gradient vanishing problem that may occur with increased model depth, significantly improving feature extraction capability and temperature prediction performance. To validate the effectiveness of the proposed model, a comparative experiment was conducted, and compared with four comparison models, the method described in this paper achieves higher model prediction accuracy.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Ke-jia Zhuang, Cong Ma, Li Zou, Xin-yu Yang, Zheng-kun Xie, and Jun Hu "Temperature prediction of wind turbine bearing based on multichannel parallel deep residual neural network", Proc. SPIE 12981, Ninth International Symposium on Sensors, Mechatronics, and Automation System (ISSMAS 2023), 129815N (4 March 2024); https://doi.org/10.1117/12.3014857

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