基于混合自适应扩展卡尔曼滤波的永磁同步直线电机等效机械参数辨识策略

Equivalent Mechanical Parameter Identification of Permanent Magnet Synchronous Linear Motor Based on Hybrid Adaptive Extended Kalman Filter

针对机械参数辨识精度受推力系数波动影响而降低的问题,提出等效机械参数模型,将推力系数与待辨识参数相结合,并推导该模型在控制器调谐和前馈补偿上与传统模型的等效性。为了进一步实现等效机械参数在线辨识,构建基于混合自适应扩展卡尔曼滤波的辨识器。混合自适应扩展卡尔曼滤波在系统噪声矩阵和渐消因子中加入自适应机制,能够估计系统噪声,并在负载突变时快速收敛。通过对系统可观性和矩阵正定性的推导,证明算法的稳定性。最后,在一台永磁同步直线电机上进行实验,验证所提算法的有效性及其在PWM周期为32 kHz和10 kHz时在线运行的可行性。

Aiming at the problem that the identification accuracy of mechanical parameters suffers from the fluctuation of thrust coefficient,an equivalent mechanical parameter model is proposed.In this model,the thrust coefficient is combined with the parameters to be identified.In addition,the equivalence between the proposed model and the traditional model in controller tuning and feedforward compensation is deduced.Furthermore,an identifier based on hybrid adaptive extended Kalman filter is constructed to identify the equivalent mechanical parameters online.The hybrid adaptive extended Kalman filter adds adaptive mechanism to the system noise matrix and the fading factor,which can estimate the system noise and converge fast with sudden loads.Moreover,the stability of the proposed algorithm is proved by the derivation of the system observability and positive definite property of the matrix.Eventually,the experiments are carried out on a permanent magnet synchronous linear motor,the effectiveness of the proposed algorithm is verified,and its online feasibility in 32 kHz and 10 kHz PWM cycle is demonstrated.