Research On Condition Monitoring And Fault Diagnosis Of High Speed Traction Motor

In recent years,with the rapid development of high-speed railway technology,the function of high-speed railway is also increasingly perfect,but the equipment is more complicated,therefore putting forward higher requirements to ensure the safety and reliability of high-speed railway,ensuring the safe and stable operation of electric locomotive in the high-speed railway has become the key and core of the development of high speed railway system.High speed traction motor is one of the key equipments of electric locomotive,and its working condition is related to the safe operation of the electric locomotive.Therefore,carrying out technology research on condition monitoring and fault diagnosis of high-speed rail traction motor can be able to identify whether the high speed traction motor is abnormal in whole or in part,to find the fault and its causes early,and forecast the fault development trend.The condition monitoring and fault diagnosis of high speed traction motor have also been discussed in this paper,the main research work is as follows:(1)The composition and common fault type of high speed traction motor are introduced,the fault mechanism of stator turn to turn short circuit fault and rotor broken bar fault are analyzed,the relationship between stator turn to turn short circuit fault,rotor broken bar fault and fault characteristic frequency are established.(2)Taking CRH2 high-speed train group used MT205 traction motor as the research object,the 2D and 3D models of high speed traction motor named MT205 are established and the stator turn to turn short circuit fault and the rotor broken bar fault are modeled and simulated by using the finite element analysis software called Ansoft Maxwell.The method of 8 layer wavelet packet decomposition and reconstruction forthe simulation signal of the stator current is put forward by using the Coif5 wavelet base function,combining the frequency band energy calculation,the fault characteristic frequency band can be extracted effectively.(3)The 30 sets of data of different fault frequency bands are extracted as training and testing samples of least squares support vector machines(SVM)diagnosis method,Compared with the neural network diagnostic method in the same environment,it is proved that the least squares support vector machines(SVM)diagnosis method has better diagnosis.The improved method of least squares support vector machines based on particle swarm optimization is proposed,and its effectiveness is verified.(4)In the LabVIEW environment,a platform for condition monitoring and fault diagnosis of high speed traction motors is built.Using MATLAB and LABVIEW hybrid programming to design the stator current wavelet transform interface and optimize the SVM fault diagnosis interface.The three kinds of simulation signals are introduced into the test to realize the function of analysis and diagnosis.At the same time,the data inquiry and print report function are realized.