Abstract
This paper proposes novel multi-layer neural networks based on Independent Component Analysis for feature extraction of fault modes. By the use of ICA, invariable features embedded in multi-channel vibration measurements under different operating conditions (rotating speed and/or load) can be captured together. Thus, stable MLP classifiers insensitive to the variation of operation conditions are constructed. The successful results achieved by selected experiments indicate great potential of ICA in health condition monitoring of rotating machines.
Similar content being viewed by others
References
Alexander, Y., David, M. J. and Robert, P. W., 1999. Robust Machine Fault Detection with Independent Component Analysis and Support Vector Data Description. Pattern Recognition Group, Dept. of Applied Physics, Delft University of Technology. Delft, The Netherlands.
Alexander, Y., Amir, L. and Robert, P. W., 2002. Blind Separation of Rotating Machine Sources: Bilinear Forms and Convolutive Mixtures. Neurocomputing-Special Issue on ICA/BSS, p. 123–134.
Bell, A. and Sejnowski, T., 1995. An information-maximization approach to blind separation and blind deconvolution.Neural Computation, 7: 1129–1159.
Chakra, Ch. and Allan, J., 2000. Sparse PCA Extracting Multi-Scale Structure From Data. Department of Computer Science, University of Toronto, 6 King's College Road Toronto, ON M5S 3H5, Canada.
Chitroub, S., Houacine, A. and Sansal, B., 2001. Compound PCA-ICA Neural Network Model for Enhancement and Feature Extraction of Multi-Frequency Polarimetric SAR Imagery. Signal Processing Laboratory, Electronics Institute, University of Sciences and Technology of Houari Boumedienne.
Gelle, G., Colas, M. and Delaunay, G., 2000. Blind sources separation applied to rotating machines monitoring by acoustical and vibrations analysis.Mechanical Systems and Signal Processing,14(3): 427–442.
Govindan, A. and Deng, G., 1998. Independent Component Analysis applied to Electrogram Classification during Atrial Fibrillation. School of Electronic Engineering, La University, Melbourne Vic-3083, Australia.
Hansen, L. K., 2000. Blind Separation of Noisy Image Mixtures, Based on a Workshop Held after the 1999 International Conference on Artificial Neural Networks (ICANN), Kluwer Academic Publishers.
Lee, S. J., Jung, S. B., Kwon, J. W. and Hong, S. H., 1999. Face Detection and Recognition using PCA. IEEE TENCON.
Lee, T. W., 2000. Nonlinear Approaches to Independent Component Analysis. The Salk institute, Institute for Neural Computation, University of California, San Diego, USA.
Luis, B. Almeida., 1999. ICA of Linear and Nonlinear Mixtures based on Mutual Information. IST and INESC-ID, Lisbon, Portugal.
Lyon, R.H., 1987. Machinery Noise and Diagnostics. Butterworths, Boston, p. 342–345.
Porrill, J., Stone, J. V., Berwick, J., Mayhew, J. and Coffey, P., 2000. Analysis of Optical Imaging Data Using Weak Models and ICA. Based on a Workshop Held after the 1999 International Conference on Artificial Neural Networks (ICANN), Kluwer Academic Publishers.
Vigario, R., Sarela, J. and Oja, E., 2000. Searching for Independence in Electromagnetic Brain Waves, Based on a Workshop Held after the 1999 International Conference on Artificial Neural Networks (ICANN), Kluwer Academic Publishers.
Xu, M., 1998. Manual for Machinery Fault Diagnosis. Xi'an Jiaotong University Press, Xi'an City (in Chinese).
Author information
Authors and Affiliations
Additional information
Project supported by the National Natural Science Foundation (No. 50205025) and the Natural Science Foundation of Zhejiang Province (No. 5001004).
Rights and permissions
About this article
Cite this article
Wei-dong, J., Shi-xi, Y. & Zhao-tong, W. Extracting invariable fault features of rotating machines with multi-ICA networks. J. Zhejiang Univ. Sci. A 4, 595–601 (2003). https://doi.org/10.1631/jzus.2003.0595
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1631/jzus.2003.0595
Key words
- Independent Component Analysis (ICA)
- Mutual Information (MI)
- Principal Component Analysis (PCA)
- Multi-Layer Perceptron (MLP)
- Residual Total Correlation (RTC)