A Hierarchical Multi-Scale Fusion Network with Diversity-Promoting MoE for Bearing Fault Diagnosis
ID:187 View Protection:ATTENDEE Updated Time:2026-10-19 18:10:29 Hits:57 Poster Presentation

Start Time:Pending(Asia/Shanghai)

Duration:Pending

Session:No Session »

No files

Abstract
Bearing fault diagnosis is essential for rotating machinery safety but remains challenging due to complex time–frequency patterns, limited samples, and imbalanced fault types. Existing models struggle to exploit multi-scale features and coordinate expert predictions effectively. To address this, we propose a hierarchical multi-scale expert fusion network that integrates multiscale convolutional feature extraction with a dual-path Mixture-of-Experts decision mechanism. Its core innovation is a masked symmetric Kullback–Leibler divergence regularization at the decision level, which promotes expert predictions and mitigates homogenization under small-sample and imbalanced conditions. The effectiveness of the proposed method is verified on multiple benchmark datasets, where it achieves 99.36% accuracy on the SEU dataset.
Keywords
Rotating machinery,Feature learning,Intelligent fault diagnosis,Multi-scale feature extraction
Speaker
Yiming Guo
Mr Beijing University of Civil Engineering and Architecture

Submission Author
Yiming Guo Beijing University of Civil Engineering and Architecture
Zhang Jinrui Beijing University of Civil Engineering and Architecture
Kehui Zhu Beijing University of Civil Engineering and Architecture
Xinming Li Beijing University of Civil Engineering and Architecture
Zhongheng Liu Beijing University of Civil Engineering and Architecture
衍学 王 北京建筑大学机电与车辆工程学院
Submit Comment
Verify Code Change Another
All Comments
Important Date
  • Conference Date

    Nov 06

    2026

    to

    Nov 08

    2026

  • Oct 15 2026

    Draft paper submission deadline

  • Nov 10 2026

    Registration deadline

Sponsored By
IEEE Instrumentation and Measurement Society
Organized By
Sichuan University