Study On Fault Diagnosis System Of High-speed Train Bearings

In recent years,the High-speed Train is developing quickly,which is because of its fast,efficient,large transport capacity and low energy consumption,it also makes the safe operation of High-speed Railway has received more and more attention.As one of the important running components of the High-speed Train,the high-speed train bearings’ operating conditions directly affect the safe and efficient operation of the train,once the bearing has broken it will cause huge losses,so the research of its’ fault diagnosis has great significance.In this paper,the fault types of high-speed train bearings are summarized after field investigation,then analyze the characteristics of the vibration signals of fault bearings.Introducing the calculation methods of rolling bearing fault characteristic frequency and the basic bearing fault diagnosis methods in detail,such as time domain analysis,frequency domain analysis and time-frequency analysis of the vibration signals.Based on the above methods,nine fault diagnosis models based on neural network are established.By using the bearing test data of West Reserve University to test these diagnosis models,to find better diagnostic methods for the establishment of the follow-up diagnostic system.In the test samples,the diagnostic accuracy are all more than 80%,which provides a solid theoretical basis for the establishment of subsequent fault diagnosis system.A bearing fault diagnosis system is designed,it is composed of software and hardware,the software part is using the GUI of MATLAB software to design the off-line fault diagnosis system of high-speed train bearings,which is established to realize the purpose of off-line bearing fault diagnosis,it can realize the functions of login interface designing,data importing,time domain index statistics,waveform analysis,fault diagnosis and fault report exporting,it is used to simplify the diagnosis procedure and facilitate human-computer interaction.In hardware part,the small bearing testing machine is the main body,then through selecting the sensors,loading system,motor,signal acquisition equipment to build the hardware platform,which can achieve the purpose of collecting vibration signals.This paper uses the tapered roller bearing to set up high-speed train bearing fault simulation experiments.First,setting the outer ring,inner ring and rolling body peeling fault,and the cage fracture fault,which is used to prepare for the collecting of signals.Secondly,designing the experimental hardware platform and overall scheme,using this hardware platform to collect different signals of normal and fault bearings,the bearings should be tested in different speed and loading conditions.Finally,using the system to analysis the signals and diagnosis,good diagnostic results verify the reliability of the system.