PCA-Based Feature Selection Scheme for Machine Defect Classification

TLDR

The sensitivity of defect‑characteristic features varies greatly under different operating conditions. The study aims to develop a PCA‑based feature selection scheme that identifies bearing defect severity without prior defect knowledge. The scheme uses PCA to select features and was validated on a bearing test bed with both supervised and unsupervised classification methods. The scheme achieved more accurate defect classification with fewer features and confirmed its effectiveness as a machine health assessment tool.

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.