Fault diagnosis of ball bearing based on energy feature and research of intelligent classification method

Abstract

Roll bearing is the most important part of mechanical equipment, which has the merits of low friction resistance, convenient installment and easily realized lubrication. So it is commonly used in rotated machine. Whether bearing is normal or not affects straightly the mechanical working state, any fault or invalidation took place in the running of the machine will bring serious sequence and great economic loss. So it is necessary to check the state of the bearing and diagnose the fault. In this paper, the energy features of the vibration signal of the roll bearing are extracted by Wavelet Packet, then the method of RBF Neural Network is presented to diagnose the faults, the faults can realize intelligent classification by this method. Simulation is used to assist in the roll bearing faults diagnosis. The simulation results obtained indicated that the method of energy features extracted by Wavelet Packet is efficient and the method of intelligent classification can identify the faults well. © 2012 Springer-Verlag GmbH.