PCA-Based Feature Selection Scheme for Machine Defect Classification
- 22 November 2004
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 53 (6), 1517-1525
- https://doi.org/10.1109/tim.2004.834070
Abstract
The sensitivity of various features that are characteristic of a machine defect may vary considerably under different operating conditions. Hence it is critical to devise a systematic feature selection scheme that provides guidance on choosing the most representative features for defect classification. This paper presents a feature selection scheme based on the principal component analysis (PCA) method. The effectiveness of the scheme was verified experimentally on a bearing test bed, using both supervised and unsupervised defect classification approaches. The objective of the study was to identify the severity level of bearing defects, where no a priori knowledge on the defect conditions was available. The proposed scheme has shown to provide more accurate defect classification with fewer feature inputs than using all features initially considered relevant. The result confirms its utility as an effective tool for machine health assessment.Keywords
This publication has 18 references indexed in Scilit:
- A hybrid neural network-feature-based manufacturability analysis of mould reinforced plastic partsProceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 2001
- Prognosis of remaining bearing life using neural networksProceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, 2000
- A virtual instrumentation system for integrated bearing condition monitoringIEEE Transactions on Instrumentation and Measurement, 2000
- Fatigue Failure Progression in Ball BearingsJournal of Tribology, 2000
- Prediction of Machine Deterioration Using Vibration Based Fault Trends and Recurrent Neural NetworksJournal of Vibration and Acoustics, 1999
- Adaptive prognostics for rolling element bearing conditionMechanical Systems and Signal Processing, 1999
- New Statistical Moments for Diagnostics of Rolling Element BearingsJournal of Manufacturing Science and Engineering, 1997
- Monitoring and diagnosis of rolling element bearings using artificial neural networksIEEE Transactions on Industrial Electronics, 1993
- Transform representation of the spectra of acoustic speech segments with applications. I. General approach and application to speech recognitionIEEE Transactions on Speech and Audio Processing, 1993
- Low-dimensional procedure for the characterization of human facesJournal of the Optical Society of America A, 1987