Condition Monitoring Using Periodic Time-Varying AR Models

  • Andrew C. McCormick
  • Asoke K. Nandi
Part of the Applied and Numerical Harmonic Analysis book series (ANHA)


Model based approaches to vibration monitoring can provide a means of detecting machine faults even if data is only available from the machine in its normal condition. Previous work has assumed stationarity in the vibration signals and exploited only time-invariant models. The cyclostationary nature of rotating machine vibrations can be exploited by using periodic time-varying autoregressive models to model the signal better than time-invariant models, and this can improve the fault detection performance. Experimental data collected from a small rotating machine set subjected to several bearing faults was used to compare time-varying and time-invariant model based systems.


Fault Detection Vibration Signal Rolling Element Rolling Element Bearing Outer Race Fault 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    H. Akaike. A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19:716–723, 1974.MathSciNetCrossRefMATHGoogle Scholar
  2. [2]
    D. C. Baillie and J. Mathew. A comparison of autoregressive modeling techniques for fault diagnosis of rolling element bearings. Mechanical Systems and Signal Processing, 10(1):1–17, 1996.CrossRefGoogle Scholar
  3. [3]
    J. R. Dickie and A. K. Nandi. AR modelling of skewed signals using third-order cumulants. IEE Proceedings-Part VIS, 142(2):78–86, 1995.CrossRefGoogle Scholar
  4. [4]
    W. A. Gardner. Exploitation of spectral redundancy in cyclostationary signals. IEEE Signal Processing Magazine, 8(2):14–36, 1991.CrossRefGoogle Scholar
  5. [5]
    A. C. McCormick and A. K. Nandi. Real time classification of rotating shaft loading conditions using artificial neural networks. IEEE Transactions on Neural Networks, 8(3):748–757, 1997.CrossRefGoogle Scholar
  6. [6]
    A. K. Nandi and J. A. Chambers. New lattice realisation of the predictive least-squares order selection criterion. IEE Proceedings-F, 138(6):545–550, 1991.Google Scholar
  7. [7]
    M. Pagano. Periodic and multiple aut or egression. Annals of Statistics, 6:1310–1317, 1978.MathSciNetCrossRefMATHGoogle Scholar
  8. [8]
    G. D. Zivanovic and W. A. Gardner. Degrees of cyclostationarity and their application to signal detection and estimation. Signal Processing, 22:287–297, 1991.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 1998

Authors and Affiliations

  • Andrew C. McCormick
    • 1
  • Asoke K. Nandi
    • 1
  1. 1.Department of Electronic and Electrical EngineeringUniversity of StrathclydeGlasgowUK

Personalised recommendations