Fault Diagnosis Method of Rolling Bearing Based on BP Neural Network

A fault diagnosis method of rolling bearing based on BP neural network and time domain parameters of vibration signal was proposed to realize fast fault diagnosis. The input vectors of the BP neural network were skewness, kurtosis, peak and margin of vibration signal. The structure of the neural network was determined with simulation research. Gradient descending method was used to train the parameters of BP neural network. Experiment results of fault diagnosis showed that with this method fast diagnosis of rolling bearing faults could be realized effectively.