Academic Journal of Engineering and Technology Science, 2025, 8(5); doi: 10.25236/AJETS.2025.080510.
Hongyu Zhou
College of International Education, Changchun University of Technology, Changchun, China, 130000
Aiming at the challenges of extracting weak bearing faults under strong background noise and determining the optimal component in singular spectrum decomposition (SSD) for rolling bearings, this study proposes a singular spectrum decomposition method based on kurtosis optimization. Firstly, the bearing vibration signal is decomposed using SSD to obtain multiple singular spectral components (SSC). Then, based on the sensitivity of bearing faults to kurtosis, the kurtosis value of each component is calculated, and the optimal component is selected at the point of maximum kurtosis. Finally, the selected component undergoes envelope demodulation to obtain the characteristic frequency and complete fault diagnosis. Experimental results demonstrate that the proposed method effectively extracts bearing faults under strong background noise. It also provides a reliable tool for early fault diagnosis of rolling bearings in industrial equipment, offering significant practical engineering value in reducing enterprise operational costs and ensuring the safe and stable operation of critical equipment.
Vibration Signal; Singular Spectrum Decomposition (SSD); Kurtosis; Fault Diagnosis; Rolling Bearings
Hongyu Zhou. Research on Rolling Bearing Fault Feature Extraction Based on Kurtosis Optimized Singular Spectrum Decomposition. Academic Journal of Engineering and Technology Science (2025), Vol. 8, Issue 5: 70-76. https://doi.org/10.25236/AJETS.2025.080510.
[1] K. DRAGOMIRETSKIY and D. Zosso, “Variational Mode Decomposition,”. IEEE Transactions on Signal Processing, 2014, 62(3) 531-544.
[2] ZHOU Jie, WANG Yunyi, Chen Chuanhai, et al. Improved MED-SSD based composite fault diagnosis method for gearbox[J]. Journal of Jilin University, 2022, 52(2) 450-457.
[3] YANG Wenxian, PENG Zhike, WEI Kexiang, et al. Superiorities of variational mode decomposition over empirical mode decomposition particularly in time-frequency feature Extraction and wind turbine condition monitoring [J]. IET Renewable Power Generation, 2016, 11(4):443-452.
[4] ZHAO Bo LI He. Noise Reduction Method of Vibration Signal Combining EMD and LSF[J]. Journal of Vibration, Measurement & Diagnosis, 2022, 42(3): 606-610.
[5] WANG Zhenya Wang, WU Xing, LIU Tao, et al. Research on Fault Characterization of Spindle Bearing by Singular Spectral Decomposition and Mutual Information Extraction[J]. Journal of Vibration and Shock. 2023, 42(15): 23-30+47.
[6] LI Hua, LIU Tao, WU Xing, et al. Application of Singular Value Ratio-based SVD in Bearing Fault Diagnosis [J]. Journal of Mechanical Engineering, 2021, 57(21): 138-149.