Rail Vehicle Bogie Motor Bearing Fault Diagnosis System Development

At present, high-speed railway and city rail transportation are in rapid development stage, as directly carrier for carrying passengers, the safety of railway vehicle directly influences the life safety of the passengers. Therefore, it is necessary for rail vehicles to diagnose the fault to prevent or avoid the unnecessary losses because of the vehicle fault. Motor bearings of bogie is one of the key components of railway vehicles and it plays a significant role in rail vehicles. Because of its complex structure, fault diagnosis of motor bearings of bogie become pivotal and difficult. How to diagnose it efficiently, fast and accurately is an important question that worthy to study.Based on the outstanding achievements of predecessors and 863 project requirements, wavelet and improved neural network model of fault diagnosis is used in rail vehicle motor, the specific target of diagnosis is motor bearing. Rail vehicle motor bearing fault diagnosis system is designed in this paper and four operating modes(normal, outer ring fault, inner fault and ball fault) of motor bearing are analyzed. After noise reduction in the signal, using "wavelet packet-energy" to extract the characteristic vector and input them to the neural network for fault identification. In order to verify the improvements of this algorithm, it shows contrast of the traditional neural network diagnostic results and improved neural network diagnostic results. At the same time, motor bearing fault diagnosis experiments are done in the lab. Experimental results show that the method is effective for motor bearing fault detection and identification.Finally, rail vehicle motor bearing fault diagnosis system has been realized in this paper. First, notes the various function modules, shows the operation interface in the system, realizes the fault diagnosis on the motor bearing, and then ensure the safe operation of rail vehicle. So it has good practicability.