Advertisement

Prediction of Remaining Life of Rolling Bearing Based on Optimized EEMD

  • Tong Wu
  • Caixia Gao
  • Ziyi Fu
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 528)

Abstract

Aiming at the problem that the early vibration signal has a weak decay characteristic in the prediction of the remaining life of the rolling bearing, a method for optimizing the bearing residual life prediction based on the optimized ensemble empirical mode decomposition (EEMD) is proposed. First, the eigenmode decomposition of the vibration signal is performed. The effect depends on two important parameters: the average number of times and the size of the added noise. Therefore, white noise criteria are added to the set of empirical mode decomposition. Then, the decomposed intrinsic mode function (IMF) is filtered with the gray correlation degree of the envelope spectrum to filter out IMF components with decay characteristics and reconstruct signals. Finally, Multi-feature parameter vector of the reconstructed signal, its redundancy is removed by principal component analysis (PCA), and then input neural network to predict bearing residual life. Experiments show that the proposed method has higher prediction accuracy and stability.

Keywords

Life prediction Ensemble empirical mode decomposition Nuclear principal component analysis Rolling bearing 

References

  1. 1.
    Y. Jia, Robust control with decoupling performance for steering and traction of 4WS vehicles under velocity-varying motion. IEEE Trans. Control Syst. Technol. 8(3), 554–569 (2000)Google Scholar
  2. 2.
    Y. Jia, Alternative proofs for improved LMI representations for the analysis and the design of continuous-time systems with polytopic type uncertainty: a predictive approach. IEEE Trans. Autom. Control 48(8), 1413–1416 (2003)Google Scholar
  3. 3.
    F. Wang, X. Chen, C. Liu et al., Reliability assessment and life prediction of rolling bearings based on KPCA and WPHM. Vib. Test. Diagn. 37(3), 476–483 (2017)Google Scholar
  4. 4.
    Z. Shen, X. Chen, Z. He et al., Residual life prediction of rolling bearing based on relative features and multivariable support vector machines. Chin. J. Mech. Eng. 49(2), 183–189 (2013)Google Scholar
  5. 5.
    C. Lue, G. Tang, S. Yanyang et al., Application of adaptive EEMD method in ECG signal processing. Data Acquis. Process. 26(13), 361–368 (2011)Google Scholar
  6. 6.
    J. Liu, J. Cheng, Y. Liu, Life prediction of rolling bearing based on LCD and GMM. Mod. Manuf. Eng. 7(7), 120–124 (2016)Google Scholar
  7. 7.
    Y. Lei, Mechanical fault diagnosis based on improved Hilbert-Huang transform. Chin. J. Mech. Eng. 47(5), 71–77 (2011)Google Scholar
  8. 8.
    Z. Zhang, X. Shi, Q. Shi et al., Fault feature extraction of rolling bearing based on improved EMD and spectral kurtosis. J. Vib. Meas. Diagn. 33(3), 478–482 (2013)Google Scholar
  9. 9.
    Wang Y, Jiang Yicheng, S. Kang et al., Roller bearing fault location and performance degradation degree diagnosis method based on optimized set EMD. J. Instrum. Instrum. 1(7), 1834–1840 (2013)Google Scholar
  10. 10.
    H. Qiu, J. Lee, J.Lin, Wavelet filter-based weak signature detection method and its application on roller bearing prognostics. J. Sound Vib. 1066–1090 (2006)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.School of Electrical Engineering and AutomationHenan Polytechnic UniversityJiaozuoChina

Personalised recommendations