A Bearing Condition Monitoring And Fault Diagnosis System For Locomotive Stationary Test

According to Chinese Railway Standard,traction load test must be performed on depot-repaired locomotive before entry into service.The rolling element bearing is one of the most important units in locomotive,and bearing failure would probably bring great suffering to railway transportation and safety.Therefore,a novel bearing condition monitoring and fault diagnosis system for locomotive stationary test is developed. Using the on-line temperature and vibration signal,the working condition of bearings in the running gear of the locomotive is examined.The contributions and conclusions are made as follows:The bearing faults are of the properties of variety and uncertainty.The incipient fault of bearings only results in the characteristic change of their vibration signal.When the faults develop,the level of their vibration will increase.As serious faults emerge,the temperature will rise.In view of this situation,the compound sensor is introduced into bearing fault diagnosis.An efficient fault testing procedure that considers the on-line temperature and vibration signal is developed.The higher order spectra are introduced into signal analysis of bearings.This enables the non-stationary and non-Gaussian signals to be thoroughly analyzed. Furthermore,the 11/2-dimension spectrum is employed to exact the quadratic phase couple caused by initial faults of bearings.The 11/2-dimension spectrum can suppress Gaussian noise effectively,but the effect is not obvious on non-Gaussian noise.Therefore a novel signal analysis procedure is presented.The wavelet transform is employed to reduce noise and 11/2-dimension spectrum is used to extract the features than can recognize bearing condition. Experiments are made to verify the theory and results.Based on the investigations mentioned above,an intelligent diagnostic system is developed.The system is successful applied to the bearings of the running gear of a locomotive.The effectiveness of proposed system is verified by stationary test.