基于自适应扩展Kalman滤波器的永磁同步电机无速度传感器控制

Speed sensorless control of PMSM based on adaptive extended Kalman filter

通过引入一种基于Sage-Husa噪声估计器的自适应扩展Kalman滤波器,给出了一种永磁同步电机无速度传感器控制方案.选取定子固定坐标系下的电机模型,首先得到了基于扩展Kalman滤波器的永磁同步电机转速估计方程.在此基础上,结合Sage-Husa噪声估计器,得到了基于自适应扩展Kalman滤波器的永磁同步电机转速估计方程.仿真结果表明,基于自适应扩展Kalman滤波器的方法,不仅可以准确地估计出电机的转速和转子位置,而且可以自适应确定扩展Kalman滤波器的一个关键参数——系统噪声协方差矩阵.与传统的扩展Kalman滤波器方法相比,本方法具有更好的实用性.

By introducing an adaptive extended Kalman filter （AEKF） based on Sage-Husa noise estimator, a speed sensorless control scheme for permanent magnet synchronous motor （PMSM） is proposed. Firstly, relying on the PMSM model in the fixed stator coordinate, the estimation equation of motor speed based on extended Kalman filter （EKF） is obtained. Then, by combining with the Sage-Husa noise estimator, the estimation equation of motor speed based on adaptive EKF（AEKF） is obtained. Simulation results indicate that the scheme based on AEKF can not only estimate the motor speed and position accurately, but also adaptively obtain one of the key parameters of EKF, i. e. , the covariance matrix of system noise. Compared with the traditional methods based on EKF, the proposed method is more practicable.