0
TECHNICAL PAPERS

HMM-Based Fault Detection and Diagnosis Scheme for Rolling Element Bearings

[+] Author and Article Information
Hasan Ocak

Department of Electrical Engineering and Computer Science,  Case Western Reserve University, Cleveland, OHhxo9@cwru.edu

Kenneth A. Loparo

Department of Electrical Engineering and Computer Science,  Case Western Reserve University, Cleveland, OHkal4@cwru.edu

J. Vib. Acoust 127(4), 299-306 (Sep 23, 2004) (8 pages) doi:10.1115/1.1924636 History: Received May 10, 2002; Revised September 23, 2004

In this paper, we introduce a new bearing fault detection and diagnosis scheme based on hidden Markov modeling (HMM) of vibration signals. Features extracted from amplitude demodulated vibration signals from both normal and faulty bearings were used to train HMMs to represent various bearing conditions. The features were based on the reflection coefficients of the polynomial transfer function of an autoregressive model of the vibration signals. Faults can be detected online by monitoring the probabilities of the pretrained HMM for the normal case given the features extracted from the vibration signals. The new technique also allows for diagnosis of the type of bearing fault by selecting the HMM with the highest probability. The new scheme was also adapted to diagnose multiple bearing faults. In this adapted scheme, features were based on the selected node energies of a wavelet packet decomposition of the vibration signal. For each fault, a different set of nodes, which correlates with the fault, is chosen. Both schemes were tested with experimental data collected from an accelerometer measuring the vibration from the drive-end ball bearing of an induction motor (Reliance Electric 2 HP IQPreAlert) driven mechanical system and have proven to be very accurate.

Copyright © 2005 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Four-state left-to-right hidden Markov model

Grahic Jump Location
Figure 2

Actual and model estimated vibration signals

Grahic Jump Location
Figure 3

Order of the model versus signal-to-noise ratio of the predicted data

Grahic Jump Location
Figure 4

Vibration feature extraction

Grahic Jump Location
Figure 5

HMM-based fault detection

Grahic Jump Location
Figure 6

Diagnosis stage of a HMM-based fault detection/diagnosis scheme

Grahic Jump Location
Figure 7

Modified diagnosis stage of a HMM-based fault detection/diagnosis scheme

Grahic Jump Location
Figure 8

Wavelet packet decomposition (WPD) of a vibration signal

Grahic Jump Location
Figure 10

Probabilities of normal and faulty data given the HMM for normal condition

Grahic Jump Location
Figure 11

HMM probabilities for an inner race fault

Grahic Jump Location
Figure 12

HMM probabilities for an outer race fault

Grahic Jump Location
Figure 13

HMM probabilities for a ball fault

Grahic Jump Location
Figure 14

Probabilities of faulty data given the HMM for the inner race fault

Grahic Jump Location
Figure 15

Probabilities of faulty data given the HMM for the outer race fault

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In