Research And Realization On Fault Diagnosis Algorithm For Rolling Bearings Of Train Running Gear

With the continuous development of urban rail transit,which led to the urban rail transit safety has been received more and more attention.Rolling bearings are a key component of the train running gear,and their good working condition is the key to the safe operation of the train.Therefore,it is of great practical significance to carry out fault diagnosis research on rolling bearings of train running gear.Due to the complex operating environment of trains,and rolling bearings are usually integrated into other mechanical equipment,the vibration signal collected contains a lot of interference information such as impulse noise and cyclostation noise in addition to the vibration signal of the bearing itself,which brings great difficulties to rolling bearings fault diagnosis.Based on the above situation,this thesis takes rolling bearing as the research object,and takes the fault diagnosis of rolling bearing under strong noise interference as the research objective.Based on the resonance demodulation technology,the research is carried out from two aspects of vibration signal noise reduction and optimal fault demodulation frequency band selection.The main research contents of this thesis are as follows:(1)Introduced the basic structure and main failure modes of rolling bearings.The vibration signal model of three typical faulty rolling bearings are established based on the vibration mechanism of rolling bearings.Through the analysis of the time-frequency domain characteristics of the models,it is concluded that the vibration signals of bearing faults have resonance modulation.(2)The noise in the signal will influence the effect of bearing fault diagnosis.A noise reduction algorithm based on CEEMD is proposed,which improved the noise reduction algorithm based on EMD,in order to reduce the modal mixing and reconstruction error.The simulation results show that the noise reduction algorithm based on CEEMD can highlight the high-frequency resonance components in the signal and retain most of the features of the original signal,which provides a good condition for further extraction of fault features.(3)In view of the large number of shock vibrations and cyclostation vibrations which is not related to bearing failures in the bearing vibration signal of the train running gear,the FK and Protrugram are susceptible to interference,and the correct demodulation frequency band cannot be selected.In this thesis,starting from the difference between disturbance vibration and bearing fault vibration envelope spectrum characteristics,referring to Protrugram’s frequency band division method,an optimal demodulation frequency band selection algorithm based on envelope spectrum fault feature recognition is proposed.The simulation results show that the frequency band selection algorithm proposed in this thesis can give the correct frequency band for bearing fault demodulation under the interference of shock vibration and cyclostation vibration,which embodies strong engineering application value.(4)Design a comprehensive vibration test bench to simulate the actual vibration signal generation,transmission and storage process of the bearing of the train running gear,and collect the bearing vibration data of different types of failures.Combining the noise reduction algorithm and demodulation frequency band selection algorithm proposed in this thesis with resonance demodulation technology,applied to the analysis of measured data,the bearing was successfully diagnosed,indicating that the rolling bearing fault diagnosis algorithm proposed in this thesis is useful in actual project.