| With the development of permanent magnet materials and power electronic technology, the Permanent Magnet Synchronous Motor(PMSM) is more widely applied in the modern industry field. In order to meet the needs of high-performance high-precision control applications, the conventional control strategies of PMSM are required to use the related motor parameters. Since motor running temperature, flux saturation and other factors that may lead to changes in motor parameters, this paper mainly focuses on the online identification of PMSM parameters.Firstly, the mathematical model of PMSM is established. The vector control system based on space vector pulse width modulation(SVPWM) is studied. The methods of off-line identification of stator resistance and stator inductance are discussed. The availability of control system and off-line parameters identification are proved by simulation analyses. The basic platform for online parameters identification is provided.Secondly, the recursive algorithms and model reference adaptive system(MRAS) are applied to the online parameters identification of PMSM. The recursive algorithms of recursive least square(RLS) and extended Kalman filter(EKF) are studied primarily. The MRAS algorithm includes two kinds of different adaptive rules which are based on Lyapunov stability and Popov hyperstability. Identification speed, precision, dynamic characteristics and trackability of those algorithms are analyzed in detail.Finally, compare the characters of recursive algorithms and MRAS to get an improved algorithm. The improved algorithm adds RLS in MRAS identification process and combines the advantages of RLS and MRAS. This algorithm identification speed is fast, the trackability can be adjusted by the forgetting factor and the identification precision is high. It can increase MRAS identification speed, reduce the dynamic oscillation and enhance the trackability obviously while keeping identification precision in a certain range. |
PDF Full Text Request