Demagnetization Faults Robust Diagnosis Of Permanent Magnet Synchronous Motor Based On Adaptive Sliding Mode Observer

Demagnetization faults diagnosis of permanent magnet synchronous motor(PMSM) is always the key technical problems focused in actual engineering and one of the hot issues of scientific researches. At present,the methods of demagnetization faults diagnosis are more focused on the spectrum analysis, harmonic analysis, the finite element method,analytical method and so on, but few of researches use real-time observation of demagnetization magnetic flux to deal with fault diagnosis. Therefore, it’s significant to use real-time observation of magnetic flux to study the demagnetization faults diagnosis of PMSM.This paper treats the PMSM as the research object, based on its running features and the consideration of its parameter disturbance, to detect the demagnetization faults. The main researches are the following:By searching and analyzing a large number of research papers both at home and abroad, we compare and conclude the demagnetization faults diagnosis of PMSM, and summarize problems that worth further research. We point out that the research of demagnetization faults of PMSM based on internal parameters disturbance of motor is of great practical and theoretical significance.Three-phase voltage equation of PMSM stator is given for modeling problems of demagnetization faults. Voltage equation changes into two-phase magnetic field oriented coordinates which is relatively static with rotor through conversion of coordinates. Then the flux equation is added into voltage equation. Finally according to the figure of PMSM permanent magnet flux variation, the demagnetization voltage equation under magnetic field oriented coordinates is obtained.Real-time detection method for demagnetization faults based on an adaptive sliding mode observer is presented for multiple parameters change of PMSM. The demagnetization faults are first modeled for the PMSM under magnetic field oriented coordinates. Then with regard to the problem that soft faults of internal multiple parameters change, an adaptive sliding mode observer is designed. Finally, The adaptiveestimation algorithm of permanent magnet flux is given and the demagnetization faults diagnosis of PMSM is realized.Real-time detection method for demagnetization faults based on a cascade adaptive sliding mode observer is presented for time-varying change of stator resistance. Firstly, with regard to the problem that the traditional adaptive observer fails to precisely detect the demagnetization faults due to the time-varying change of the stator resistance, a method combining the modified adaptive with the sliding mode variable structure control is presented and then a cascade adaptive sliding mode observer is designed. Finally, the adaptive estimation algorithm of the stator resistance is given and an estimation equation for permanent magnet flux is established.