A two-step fault diagnosis framework for rolling element bearings with imbalanced data

Abstract

Rolling element bearings constitute the key parts on rotating machinery and their fault diagnosis are of great importance. In this paper, a novel Two-Step fault diagnosis framework is proposed to diagnose the status of rolling element bearings with imbalanced data. The Wavelet Packet Transform (WPT) is used to determine the feature vectors. 16-dimensional wavelet packet node energies were extracted from the original datasets as the feature vectors prepared to input to the classifiers. Next, our proposed framework consists of two steps for the fault diagnosis, where Step One makes use of Weighted Extreme Learning Machine (weighted ELM) in an effort to classify the normal or abnormal categories, and Step Two further diagnoses the underlying anomaly in details. The effectiveness of our proposed approach is testified on the raw data collected from the rolling element bearing experiments conducted in our Institute, and the empirical results showed that our approach is really fast and can achieve the diagnosis accuracies more than 95%.