采用扩展卡尔曼滤波的永磁同步电机转矩控制

Direct Torque Control of PMSM Based on Extended Kalman Filter

针对现有的永磁同步电机观测器精度低、参数易发散的问题,提出了一种采用衰减记忆卡尔曼滤波器的永磁电机直接转矩控制器。通过建立的永磁同步电机旋转坐标系下的状态模型,选取旋转坐标系下的电流、转子位置和转速作为状态变量,实现了电机转矩和磁链的衰减记忆卡尔曼滤波器估计系统。通过引入衰减因子降低过去量测值对系统的影响,改善了衰减记忆卡尔曼参数易发散的问题。仿真和实验结果表明该方法能够准确观测到电机转矩和磁链,在负载转矩突变时具有较强的鲁棒性和可靠性。

Aiming at the problems of low precision and easily divergent parameters of existing PMSM observers,a direct torque controller for PMSM with MAEKF was proposed.By establishing the state model in the rotating coordinate system of the PMSM,the current,rotor position and speed in the rotating coordinate system were selected as the state variables to realize the MAEKF estimation system of the motor torque and flux linkage.By increasing the attenuation factor to reduce the impact of past measurements on the system,the problem that the Kalman parameters of the MAEKF were easily divergent is improved.Simulation and experimental results show that the method can accurately observe the motor torque and flux linkage,and has strong robustness and reliability when the load torque changes suddenly.