Research On The Feature For Motor Bearing Fault Based On The External Magnetic Field

Induction motors are widely used in power transmission field because of its simple structure, reliable performance and low cost. Its ability to work directly affects the operational efficiency and safety. Bearing fault has the highest probability, so this paper mainly studies the feature and diagnosis method of induction motor bearing fault.Currently, motor fault diagnosis based on the external flux signature analysis has been widely studied. The feature of external magnetic flux density of induction motor when it is healthy or it has static or dynamic eccentricity or with bearing fault is derived in this paper. The components of principal slot harmonics in spectrum can be used to estimate the motor speed, and bearing fault characteristic frequency exists in the vicinity of PSH. In order to verify the theoretical analysis, the electromagnetic finite element software Ansoft and electromechanical system design software Simplorer is used to simulate the bearing fault in this paper, and the characteristic frequency of bearing fault can be observed from the spectral of the simulation data of motor external magnetic flux density. Then test the actual induction motor with bearing outer raceway fault, the data acquisition of motor external magnetic flux density is completed by dual-axis magnetic sensor HMC6042, A/D conversion chip AD7606with high-precision and DSP, and the weak characteristic frequency of bearing fault can be observed from the spectrum of external radial flux density signal. Simulation and experiments show the feasibility of bearing fault diagnosis based on the external magnetic field signature.Due to the impact of the higher harmonics of the stator current and background noise, the characteristic component in the external magnetic flux density signal is very weak. It is difficult to extract the characteristics of weak periodic signals under the colored background noise by using the traditional spectrum analysis and wavelet decomposition methods, and the severity of the bearing fault is difficult to judge from the amplitude of the frequency. Therefore, this paper uses the Volterra series to identify the bearing fault system and analyze the characteristics of time-domain kernels, and the features of nonlinear output frequency response functions, they can characterize the severity of bearing fault.