Rolling bearing fault diagnosis based on EEMD sample entropy and PNN

Abstract

A fault diagnosis method for rolling bearing based on ensemble empirical mode decomposition (EEMD) sample entropy and probabilistic neural network (PNN) is proposed for non-steady and non-linear signals. First, the rolling bearing signals are decomposed into intrinsic mode function (IMF) using EEMD. Then, the kurtosis of each component is calculated. Five components with large kurtosis are selected and the sample entropy is extracted to form the feature vectors. Finally, the feature vectors are input to the PNN for fault diagnosis. The method is used to classify the type of the rolling bearing fault. The results show that the accuracy of fault diagnosis of the proposed method is 100%, which proves the effectiveness of the proposed method.

Keywords

• fault diagnosis
• neural nets
• rolling bearings
• vibrational signal processing
• hilbert transforms
• probability
• ensemble empirical mode decomposition sample entropy
• probabilistic neural network
• kurtosis model
• rolling bearing fault diagnosis method
• nonlinear signal
• nonsteady signal
• imf model