永磁同步电机参数自动辨识方法研究

Research on parameter automatic identification of PMSM

针对不同电机参数上存在差异只适用于某种特定参数的交流伺服驱动器的特点,提出了一种基于Smith预估补偿和RBF神经网络与PID控制器相结合的永磁同步电机自动辨识控制算法。该方法利用了Smith预估补偿能克服纯滞后和RBF能处理非线性问题、在线自学习整定PID参数的优点,在调速模块的伺服控制系统中更加有效。通过系统建模和仿真实验结果表明,该算法能够有效辨识定子电阻、电感,并得到较为理想的实验结果。

According tothe characteristics of different parameters of AC servo drive motor difference is only applicable to a specific parameter, a permanent magnet synchronous motor （PMSM） parameter automatic identificationcontrol algorithm which based on Smith predictive compensation and RBF neural network combined with PID controller is introduced in the study. The method uses a Smith predictor to overcome the pure lag and RBF can deal with nonlinear problems, online learning setting PID parameter advantages, is more effective in PMSM parameter automatic identification control system. Finally, the paper establishesthesimulationmodeofthecontrol algorithm,experimental results show that, the algorithm can effectively identify the stator resistance, inductance, and get ideal experimental results.