Article Details

Abstract

To address the challenge in current rolling bearing fault diagnosis where single vibration signal representations struggle to stably extract mechanism-related features, this paper proposes a Physics-Based Guided Bi-Branch Network (PBINet) to enhance fault feature recognition performance during fault diagnosis. The proposed method takes the raw time-domain signal and a transformed envelope spectrum as inputs, uses two convolutional branches to extract impact features and band-structure features, respectively, and introduces a gated fusion mechanism to achieve channel-adaptive weighting, thereby completing complementary feature fusion and discriminative learning. Experiments conducted on the CWRU and uOttawa datasets show that the proposed method achieves diagnostic accuracies of 99.2% and 100%, respectively. The results indicate that the proposed method can realize high-accuracy fault diagnosis under complex operating conditions.

Keywords

rolling bearing; fault diagnosis; feature fusion; bi-branch network

References

1. [1] Lv Fan, Xu Kai, Sun Xiangyu, etc Research on Transfer Learning Bearing Fault Diagnosis Based on Physical Mechanism Data Generation [J/OL]. Machine Tool and Hydraulic (2025-12-25) [2026-01-04] https://link.cnki.net/urlid/44.1259.th.20251225.0918.006.
2. [2] Zeng Xinling, Longjiang, Wei You, etc Fault diagnosis method for rolling bearings based on VMD-CWT and Swin Transformer [J]. Mechanical Manufacturing and Automation, 2025, 54 (6): 18-23, 34
3. [3] WANG Y X，RICHARD M，XIANG J W，et al. Research on variational mode decomposition and its application in detecting rub-impact fault of the rotor system [J].Mechanical Systems and Signal Processing，2015，60/61：243-251.
4. [4] MENG L J，XIANG J W，WANG Y X，et al.A hybrid fault diagnosis method using morphological filter–translation invariant wavelet and improved ensemble empirical mode decomposition [J].Mechanical Systems and Signal Processing，2015，50/51：101-115.
5. [5] LI X F，XIE F Q，PAN X L，et al. Intelligent fault diagnosis for a wind turbine gearbox via multisensor information fusion with adaptive weights [J].Energy，2025，335：138352.
6. [6] Wu Yuechao, Zheng Bo Fault diagnosis method for rolling bearings based on MLP Mixer BiGRU fusion network [J]. Science and Technology Innovation and Application, 2025, 15 (36): 157-160
7. [7] SMITH W A，ROBERT B R. Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study [J].Mechanical Systems and Signal Processing，2015，64/65：100-131.
8. [8] MERT S，DUMOND P，BOUCHARD M.University of Ottawa constant load and speed rolling-element bearing vibration and acoustic fault signature datasets [J/OL].Data in Brief，2023，49：109327（2023-06-21）.https://doi.org/10.1016/j.dib.2023.109327.
9. [9] Xue Bin, Li Yingshun, Guo Zhannan, etc Bearing fault diagnosis based on multi-scale ResNet feature fusion [J]. Control Engineering, 2025, 32 (9): 1652-1658
10. [10] Gao Kai, Li Xunhao, Hu Lin, etc Lane changing intention prediction based on multi head attention CNN-LSTM [J]. Journal of Mechanical Engineering, 2022, 58 (22): 369-378

Funding

2024 Zhejiang Industrial Vocational and Technical College High level Teaching Quality Management Project; 2024 Zhejiang Industrial Vocational and Technical College High level Teaching Construction and Cultivation Project

About the authors

Han Shijie (1997-), male, Han ethnicity, from Shaoxing, Zhejiang Province, China, teaching assistant, master's student, research direction: fault diagnosis, artificial intelligence.