Rolling bearing fault diagnosis based on multiscale residual shrinkage network

Le Hong; Chuanbo Wen

doi:10.1117/12.3037506

27 September 2024 Rolling bearing fault diagnosis based on multiscale residual shrinkage network

Le Hong, Chuanbo Wen

Proceedings Volume 13275, Sixth International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2024); 1327526 (2024) https://doi.org/10.1117/12.3037506
Event: 6th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2024), 2024, Wuhan, China

Abstract

In actual industrial environments, noise interference poses a challenge for fault diagnosis models to effectively extract precise fault features from vibration signals. Accordingly, this paper proposes a fault diagnosis method based on Multiscale Residual Shrinkage Network (MsRSN). Firstly, a wide convolution kernel is used in the first layer to extract the short-time features in the signal and initially suppress the high-frequency noise, and then three branches of Residual Shrinkage Network are designed to extract the multi-scale features, and adaptive feature fusion is carried out on the extracted multi-scale features by Efficient Channel Attention. The soft thresholding function is introduced into residual module and the attention mechanism is constructed for adaptive noise reduction. The standard convolution with 256 channels in the residual module is replaced with Depthwise separable convolution, which reduces the number of parameters and speeds up training. Next, Meta-ACON activation function is added to the adaptively fused features to adaptively activate the model output of the neurons to enhance the generalisation performance of the model, finally the classification is performed by softmax. Experiments show that the multiscale residual contraction network can still accurately identify fault categories under strong noise interference and has good generalisation ability.

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

Citation Download Citation

Le Hong and Chuanbo Wen "Rolling bearing fault diagnosis based on multiscale residual shrinkage network", Proc. SPIE 13275, Sixth International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2024), 1327526 (27 September 2024); https://doi.org/10.1117/12.3037506

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

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Convolution

Shrinkage

Feature extraction

Interference (communication)

Signal to noise ratio

Vibration

Performance modeling

Show All Keywords

Keywords/Phrases

Search In:

Publication Years