Research On Real-time Fault Detection System Of Train Wheel Bearings

The state of the wheel bearing is directly related to the safety of the train.The fault diagnosis technology of the bearing can not only be used to monitor the operation of the bearing,but also can further predict the bearing operating life.It can improve the operation and management level of the train and maintenance efficiency.In view of the above,real-time bearing fault detection system is deeply studied in this paper.First,the relationship between bearing noise and vibration and the characteristics of the sound signal when the bearing is running are studied,and it is proved that the mechanism,feasibility and advantages of bearing failure can be found by using run and noise.At the same time,the structure and common failure form of the bearings are studied.Acoustic diagnosis is taken as the main detection method and the temperature diagnosis is the auxiliary detection method.By comparing the common processing methods of fault signals,the signal processing methods suitable for the detection system of train wheel bearing failure are selected.The bearing fault detection system is built according to the factory site inspection conditions.The arm cabinet is specifically designed for integrated hardware conditions such as load sound sensors,temperature sensors,industrial computer and other hardware.The software for the detection system of bearing failure is developed in NI’s LabVIEW.Finally,the field test was carried out in the running and test rig of the factory.The operation stability and reliability of the system were verified by the normal bearing,inner ring fault bearing and rolling fault bearing.The results show that EMD noise reduction of can effectively remove the interference part of the sound signal to improve its signal-noise ratio.The time domain diagnostic module can initially diagnose whether the bearing is defective.The resonant demodulation based on wavelet packet decomposition and EEMD-Hilbert Spectrum can further extract the fault frequency of the train bearing.The bearing fault diagnosis system is built on the basis of virtual instrument,with strong stability,high reliability,high running speed,strong scalability,friendly human-computer interface and so on.In addition,the further research and popularization of the fault detection system will establish a solid foundation for the digital management of train bearing quality.