0
Research Papers

Phase Space Feature Based on Independent Component Analysis for Machine Health Diagnosis

[+] Author and Article Information
Qingbo He1

Department of Precision Machinery and Precision Instrumentation,  University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of Chinaqbhe@ustc.edu.cn

Ruxu Du

 Institute of Precision Engineering, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SARrdu@mae.cuhk.edu.hk

Fanrang Kong

Department of Precision Machinery and Precision Instrumentation,  University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of Chinakongfr@ustc.edu.cn

1

Corresponding author.

J. Vib. Acoust 134(2), 021014 (Jan 19, 2012) (11 pages) doi:10.1115/1.4005006 History: Received December 06, 2010; Revised July 07, 2011; Published January 19, 2012; Online January 19, 2012

This paper proposes a new feature extraction method based on Independent Component Analysis (ICA) and reconstructed phase space. The ICA-based phase space feature unifies the system dynamics embedded in vibration signal and higher-order statistics expressed in phase spectrum and hence, is effective for machine health diagnosis. The new feature extraction is done in three steps: first, the Phase Space Reconstruction (PSR) is performed to reconstruct a phase space with the dimension covering dynamic structure information; second, the ICA bases are trained by a number of constructed phase points; and finally, the new feature is quantitatively calculated by evaluating the correlation property of transformed coefficients based on ICA bases. The presented feature contains plentiful phase information with the training pattern, which is often under evaluated when using existing methods. It has excellent pattern representation property and can be applied for signal classification and assessment. Experiments in an automobile transmission gearbox validate the effectiveness of the new method.

FIGURES IN THIS ARTICLE
<>
Copyright © 2012 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Procedure of ICA-based phase space feature extraction

Grahic Jump Location
Figure 2

Principle demonstration of embedding dimension determination for defect-induced vibration

Grahic Jump Location
Figure 3

The automobile transmission gearbox: (a) structure of the gearbox and (b) gearbox setup

Grahic Jump Location
Figure 4

Typical vibration signals of the gearbox in three conditions: (a), (c), (e) are waveforms and (b), (d), (f) are spectra for normal, slight fault and severe fault condition, respectively

Grahic Jump Location
Figure 5

ICA-based phase space feature values of samples in different conditions by different models

Grahic Jump Location
Figure 6

Effect of feature models on feature calculation

Grahic Jump Location
Figure 7

Clustering result of samples by different feature models: (a) the proposed three-model combined feature, (b) PCA feature, and (c) KPCA feature

Grahic Jump Location
Figure 8

Between-class scatter values for different feature models

Grahic Jump Location
Figure 9

Between-class scatter and within-class scatter values for the proposed ICA-based phase space feature as well as the PCA and KPCA features

Grahic Jump Location
Figure 10

ICA-based phase space feature values of samples in different conditions by different models and their combination

Grahic Jump Location
Figure 11

Comparison of different features for evaluating samples in different gear conditions: (a) proposed ICA-based phase space feature, (b) kurtosis, (c) subband energy, and (d) ApEn

Grahic Jump Location
Figure 12

Effect of different features on gear health degradation assessment: (a) proposed ICA-based phase space feature, (b) kurtosis, (c) subband energy, and (d) ApEn

Grahic Jump Location
Figure 13

Representation of samples by an independent feature of three-model features

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