Presentation Details
  1. Home
  2. Presentation List
  3. Presentation Details
Bearing Fault Diagnosis Method Based on Cascaded Stochastic Resonance and FSResNet Network
ID：63 View Protection：ATTENDEE Updated Time：2025-11-10 11:32:31 Hits：161 Oral Presentation

Start Time：2025-11-23 09:50（Asia/Shanghai）

Duration：20min

Session：S2 Parallel Session 2 » S2-2Parallel Session 2-23 AM

No files

Abstract
As a key component of rotating machinery, rolling bearings play a critical role in supporting loads, reducing friction, and enabling power transmission. However, the harsh operating environment and feature aliasing among different faults make accurate classification of fault states challenging. In this paper, a rolling bearing fault classification method based on stochastic resonance and an enhanced residual network is proposed. First, a two-stage cascaded stochastic resonance model is constructed to suppress high-frequency noise and enhance weak impulse components. Second, a novel feature—Root mean square-Kurtosis Energy—is proposed, which integrates signal energy distribution and impact intensity through a nonlinear interaction mechanism, thereby improving separability between fault modes. Finally, an attention mechanism is incorporated into the skip connection module of the ResNet-V2 network, enabling spatial mapping and local enhancement of features to strengthen the network's response to key fault information and improve discriminative capability. Laboratory results demonstrate that the proposed method achieves high fault recognition accuracy under complex working conditions and can effectively distinguish fault categories.
 
Keywords
Rolling bearings,stochastic resonance,root mean square-kurtosis energy,FSResNet,fault classification
Speaker
Xiyu Pan
postgraduate China Jiliang University

Submission Author
Xiyu Pan China Jiliang University
Zhou Shao China Jiliang University
Yiding Wu China Jiliang University
Zuozhou Pan China Jiliang University
Submit Comment
Verify Code Change Another
All Comments
Important Date

  • Conference Date

    Nov 21

    2025

    to

    Nov 23

    2025

  • Oct 20 2025

    Draft paper submission deadline

  • Dec 08 2025

    Registration deadline

Sponsored By
IEEE Instrumentation and Measurement Society
South China University of Technology
Organized By
South China University of Technology
Contact Information
Login to view full contact details
Previous Conferences