Research Of Induction Motor Bearing Fault Diagnosis System Based On EMD

In engineering, environmental impact, inherent motor asymmetry, noise and other factors will lead to some"false"characteristics of the induction motor manifestation, which may confuse characteristics of the initial issuance of the fault of the motor bearing, resulting in miscarriage of justice.How to ensure a high sensitivity and high reliability, and realize the initial issuance fault detection of the induction motor bearing has become the key issues and difficult problems to be resolved.Traditional signal processing methods of induction motor bearing fault diagnosis are based on the fourier transform, but fourier transform only analysis not linear, steady signal.So it is essential to choose appropriate signal processing methods that are suitable for nonstationary signals to extract fault feature.Therefore, in this paper, EMD will be introduced in to induction motor bearing fault diagnosis, which suitable for analysis of non-linear, non-stationary signal sequence and has high SNR. This paper's aim is to extract fault feature of asynchronous motor bearing system by using EMD and hilbert-marginal spectrum ,the main work is as follows:1. Introduced basic knowledge of the induction motor fault diagnosis and rolling bearing fault diagnosis. Studied the mechanism of rolling bearing vibration, forms of rolling bearing failure and different types of vibration, given formulas of the fault characteristic frequency ,shows the feasibility of diagnosis to rolling bearing .2. Studied the theory of EMD, explored the effects endpoint of EMD, made to improve the technology of signal extension.Prepared the entire procedure of EMD and hilbert–marginal spectrum by Matlab .Simulated signal analysis in order to verify the validity of the method.3. Did the experiment of induction motor bearing failure, extracted the vibration signals of induction motor bearing, did EMD and hilbert–marginal spectrum to the vibration signals collected data.The results show that this method effectively diagnosed the early weak failuresof induction motor bearing. And quoted the experimental data of others, did EMD and hilbert–marginal spectrum to the vibration signals collected data, get the better diagnostic results.