Research On Fault Diagnosis System Of EMU Bearing Based On EEMD And Wavelet Packet

With the rapid development of China's high-speed railway,its transport capacity has gradually met the needs of national economic and social development,and accompanying EMU vehicle safety operation has also become an important research direction.Rolling bearing is the core part of EMU,and the research of bearing fault diagnosis system is imperative.Based on this,the fault diagnosis and analysis method of rolling bearing was researched and fault diagnosis system for rolling bearing of EMU was designed and developed to realize the fault diagnosis and analysis of the bearing without dismantling the axle box bearing in order to ensure the safe operation of the high speed EMU.Specific research content has the following aspects:Rolling bearing structure,fault formation mechanism and analysis method of bearing fault was researched,according to the comparation of the advantages and disadvantages of different fault diagnosis method,vibration analysis was selected to diagnose the fault of the bearing due to the vibration analysis method has the advantages of cheap sensors,simple setting and accurate test results,and the vibration frequency of each part of the bearing was calculated.The various of signal processing methods in vibration analysis were researched,commonly used signal processing methods were simulated and the simulation results showed that the finer frequency components are decomposed compared with the wavelet transform,and the EEMD is better to solve the modal aliasing phenomenon than the EMD.However,the wavelet packet is related to the optimal selection of wavelet bases and the IMFs of EEMD are not belong to signal itself 100 persent,thus both of them have their own limitations in signal processing.Based on this,a fault diagnosis method of rolling bearing based on EEMD and wavelet packet analysis was proposed.The fault diagnosis method of rolling bearing based on EEMD and wavelet packet analysis was researched.Firstly,the signal was denoised by wavelet packet,and then EEMD was used to decompose the denoised signal.Finally,the cross-correlation coefficients were calculated and the envelope spectrum of the IMFs component which correlation coefficient meet the relevant conditions were analyzed.In order to prove the validity of the diagnosis method,bearing fault data was collected from railway administration.The experimental results showed that the method could overcome the limitation of the traditional method,and could effectively extract the fault characteristic information of the bearing and accurately determine the specific location of the bearing failure.The software development of fault diagnosis system for rolling bearing of EMU was reaearched and a programmable fault pattern recognition method was proposed.Based on this method,a fault diagnosis system for rolling bearing based on Visual Basic and MATLAB was designed and developed.Visual Basic has rich interface design resources and MATLAB has strong engineering computing power,the combination of interface development and engineering computing is achieved with the support of strong technical background.It is convenient for software operators to use because the software can not only analyze and diagnose the fault of the bearing vibration signal,but also include the modules of "query record","experience value maintenance","software management and maintenance" and so on.