基于两段卡尔曼滤波的永磁同步电机磁链在线辨识

摘要

针对永磁同步电机的磁链辨识问题,采用两段卡尔曼滤波算法对磁链参数进行在线估计.获取准确的永磁体磁链参数对于永磁同步电机的控制有十分重要的意义.基于经典卡尔曼滤波(KF)的磁链辨识方法能够快速准确地在线辨识永磁体磁链,但存在运算量大、硬件成本高的问题.为此,将两段式卡尔曼滤波算法用于永磁体磁链辨识中.首先以交直轴的电流和磁链为状态变量构建四阶KF算法,并在此基础上设计两段卡尔曼滤波算法,最终利用2个二阶KF进行磁链的在线辨识.实验结果表明,该方法能获得准确的永磁体磁链参数,而且相比于经典卡尔曼滤波减少了23.8％的运算时间.%A method of online flux identification of permanent magnetic synchronous motor (PMSM) based on two-stage Kalman filter was proposed.Acquiring accurate permanent magnet flux parameter is very important for the control of PMSM.Flux identification method based on traditional Kalman filter (KF) could identify permanent magnet flux rapidly and accurately,but KF consumes great calculation and high-cost hardware.To solve this problem,utilizes two-stage Kalman filter to identified flux parameter.Firstly,a four-order KF was constructed with the quadrature-direct axis current and flux being the state variables,and then two-stage Kalman filter was designed on this basis,so that two two-order KF could be used to accomplish the flux identification process.In the end,experimental results show that the proposed method could acquire accurate permanent magnet flux,and reduces 23.8％ calculation time compared with traditional KF.