Ac Traction Motor Fault Diagnosis Research

This dissertation focuses on the study of detection method of asynchronous motor used by train based on wavelet analysisBased on the introduction of the element theory of wavelet, the time frequency and the phases-frequency of wavelet were studied. The general process of fault signal by using wavelet analysis and the Mallat arithmetic about wavelet were introduced.A advanced method for real-time harmonics cancellation based on frequency shifting principle is deduced, for the fact that the scope of the fault signal is very faint compared with the circuit of the motor. the simulating study shows that the way is efficient to reduce the interfere of the strong signal and emphasis the fault signal. meantime, the method improves the ability of acquiring the fault signal of asynchronous motor. the simulating study has been done in the field of the application of wavelet to signal decomposition.The fault signal of asynchronous motor takes on the characteristic of singularity and non-placidity. the use of Fourier analysis has the limit to deal with these signal. however, wavelet analysis have advantage over Fourier analysis and is suit for the process of these signal. In the paper, wavelet analysis is used to detect the fault singularity signal of asynchronous motor. Through the simulating experiments, the faults ofasynchronous motor are studied .such as: the circuit short of single phases and windings of stator and the break of the squirrel cage.The main characteristic of asynchronous motor used by electric train in china is introduced, the harmonic influence for the detection of fault signal produced by the SPWM converter is studied deeply. The asynchronous motor model which focuses on the electric locomotive of model DJ120 is build by using MATLAB. the simulating study are finished .such as: the circuit short of single phases, the squirrel cage. The characteristic signal are acquired by the simulation, moreover, the characteristic signals were analyzed by the program of wavelet. The result shows that the wavelet analysis can acquire the feature component of motor fault accurately.A fault diagnosis system was designed for diagnosing the squirrel cage based on wavelet analysis.