Bearing Fault Diagnosis of Induction Motor Using Thermal Imaging

Abstract

Temperature is one of the most important indicators of health of rotating machines. To provide automatic fault diagnosis using temperature information it is meaningful to extract optimal features which can be used to identify the potential fault. An automatic bearing fault diagnosis approach using non-invasive contactless thermal infrared imaging is proposed for early fault diagnosis of an induction motor. The authors analyse thermal images of four different bearing conditions in a three phase induction motor: outer race defected bearing, inner race defected bearing, lack of lubrication and healthy bearing. As a first step thermal images of induction motor were acquired then preprocessed using 2D-DWT to decomposed the thermal images. This is followed extracting relevant features and selecting the strongest feature using Mahanabolis distance critera. Finally the selected features are given to a SVM classifier for classification of bearing condition. The experimental results indicate decent classification performance of the proposed technique for bearing fault detection.