The Fault Diagnosis Of Gear Based On Ensemble Empirical Mode Decomposition And Radial Basis Function Neural Network

Now the mechanical equipment is movi ng to autom ated production, large-scale structure, efficient operation. The relations hip between equipment become more and more closely. It has a high dem and on the reliability. A part of mechanical equipment failure may stop the entire production process. Directly or indir ectly economic losses will increase exponentially. Mechanical equ ipment of high reliability, high safety, low failure rate is more important to the modernization of industrial production. Gear transmission has the advantages of compact structure, constant transmission ratio and large transmission torque. It is the most widely used application of transmission equipment. At the sam e time, it is the fault prone parts in mechanical production equipment. Therefore, in order to prevent mechanical equipment to stop running, prolong the using time, to carry out condition monitoring and diagnosis, detect of fault in gear system, have very important meaning for safe production and improve economic benefit.Research presents the method of gear fault diagnosis based on RBF neural network and ensemble em pirical mode decomposition. Analysed the mechanism and characteristics of gear vibration. Elaborated the analysis method in the time domain and frequency domain. Introduced the fault di agnosis method based on wavelet packet decomposition and ensemble empirical mode decomposition. Introduced the application of neural network in fault diagnosis field.Experiments conducted at faul t simulation test bed. Thr ough vibration signal, to verify the m ethod of gear fault diagnosis based on the RBF neural network and the ensemble empirical m ode decomposition. The experimental results show this m ethod has a good effect and a extensive prospect of application.