基于多新息扩展卡尔曼滤波的直接转矩控制研究

Research on Direct Torque Control Based on Multi-Innovation Extended Kalman Filter

为解决传统扩展卡尔曼滤波(EKF)算法在非线性系统观测模型中因误差引起的系统状态估计精度下降等问题,以永磁同步电动机为例对直接转矩控制系统进行研究。将多新息辨识理论引入扩展卡尔曼滤波中,采用多新息扩展卡尔曼滤波(MI-EFK)算法,将其应用于无传感器直接转矩控制系统中,分别构建了EKF和MI-EKF观测算法仿真模型,并对其进行对比验证。研究结果表明,基于MI-EKF观测算法更有效,改进后的观测器波动范围缩小,其中转速波动范围缩小为原来的22%左右,转矩脉动范围缩小为原来的14%左右。

In order to solve the problem that the accuracy of a system state estimation is re duced, which is caused by the error of the traditional extended Kalman filter（EKF） algorithm in the observation model of the nonlinear system, a direct torque control system is studied by taking the permanent magnet synchronous motor as an example. The multi innovation identification theory is introduced into the extended Kalman filter. The multi innovation extended Kalman filter（MI EKF） algorithm is applied to the sensorless direct torque control system. The simulation models of EKF and MI EKF observation algorithms are constructed respectively. Moreover, the models are corn pared and verified. The research results show that the MI EKF observation algorithm is more effec tive, and the improved observer fluctuation range is reduced. The range of the rotational speed fluc tuation is reduced to 22%, and the torque ripple range is reduced to 14%.