Research On Motor Bearing Fault Diagnosis Based On External Magnetic Field

Induction motors are widely used in the fields of national defense,military,transportation,production and life due to their excellent characteristics such as simple structure,cheap price and reliable operation.The bearing as a vital part of the induction motor,once the fault is very serious.Therefore,the research of bearing fault diagnosis is of great significance to ensure the smooth operation of induction motor.Due to the bearing fault diagnosis based on external magnetic field has the advantages of non-invasive and convenient detection,it has gradually become a hot topic in this field.In this paper,a new air gap change model is proposed considering the actual situation that the bearing fault will cause the air gap eccentricity.In view of the actual situation that the bearing fault will cause the air gap change and the torque fluctuation at the same time,the formula of the external magnetic flux density of the motor under the bearing fault is derived by combining the two theoretical models.Compared with the air gap change model and the torque fluctuation model,the new theoretical model increases the rotor frequency and has more frequency components.Due to the complex frequency components of external magnetic signals and the difficulty in fault feature frequency identification,a variational empirical modal decomposition(VMD)and Teager-Kaiser energy operator(TKEO)based on parameter optimization for bearing fault detection is proposed in this paper.’Firstly,VMD is used to decompose the magnetic density signal.Considering that the VMD method needs to specify the number of decomposition in advance and the number of decomposition directly affects the decomposition result,this paper proposes a method to calculate the mean value of instantaneous frequency for parameter optimization.Then the correlation coefficient of the decomposition result of VMD is calculated,so an effective modal function is selected for signal reconstruction.In order to solve the problem that there is no single fault characteristic frequency in the magnetic density signal,TKEO processing is carried out.TKEO demodulates the fault characteristic frequency and many modulation side frequencies containing the characteristic frequency,which provides a sufficient basis for the bearing fault diagnosis.Finally,a bearing fault test platform is built and the magnetic density signals of healthy bearing and fault bearing are collected.The VMD-TKEO method of parameter optimization is used to process the magnetic density signal,the bearing fault detection is realized,the correctness of the improved external magnetic density theoretical model is verified,and the effectiveness of the proposed method is also verified.By comparison with EMD,EEMD,EWT and TKEO methods,the results show the effectiveness and superiority of the proposed method.