永磁同步电机参数辨识及MTPA和前馈控制研究

Parameter Identification and MTPA and Feed-Forward Control Study ofPermanent Magnet Synchronous Motor

针对传统控制算法在电机参数变化时电流分配偏离最优解、稳定性差等问题,对参数在线辨识进行了研究,并将辨识结果用于电压前馈补偿以及MTPA控制算法中,实现交直轴完全解耦,提高MTPA控制效率。使用遗忘因子可变的递推最小二乘算法,通过对遗忘因子进行函数调整来实现辨识准确性和辨识速度的平衡。仿真和实验结果表明,相对于传统PI控制和MTPA控制,该算法不仅能够有效提高参数动态辨识效果,还能提高电流跟随精度,减小震荡,具有更好的控制精度和动态性能。

Aiming at the problems of traditional control algorithms,such as current allocation deviating from the optimal solution and poor stability when the motor parameters change,the on-line identification of the parameters is investigated,and the identification results are used in voltage feed-forward compensation and MTPA control algorithms to realize the complete decoupling of the alternating and straight axes and to improve the efficiency of MTPA control.The recursive least squares algorithm with variable forgetting factor is used,and the balance between identification accuracy and identification speed is achieved by functionally adjusting the forgetting factor.Simulation and experimental results show that,compared with the traditional PI control and MTPA control,this algorithm can not only effectively improve the parameter dynamic discrimination effect,but also improve the current following accuracy,reduce the oscillation,and have better control accuracy and dynamic performance.