Fault Detection of Rolling Element Bearing Based on Principal Component Analysis

The Principal Component Analysis (PCA) has been widely used to detect and diagnose the faults of industry processes. However, this technique is rarely applied to the fault detection and diagnosis of rolling element bearing. The main reason is that PCA is a kind of multivariate statistical technology, but bearing vibration signal is one dimensional time series. A new method of fault detection based on PCA for rolling element bearing is proposed in this paper. Firstly, the vibration signal is mapped into a high dimensional space. Then, PCA is applied in this space. The proposed method is tested with experimental data collected from drive end ball bearing of a 2 hp Reliance Electric motor driven mechanical system. The simulation results show the PCA-based method of bearing fault detection is effective and is superior to the traditional PCA-based approach.