PMSM位置伺服系统的复合自抗扰控制

Neural network based active disturbance rejection control of PMSM position servo system

PMSM位置伺服系统是强耦合、参数时变的非线性系统。基于自抗扰控制把系统的内扰动和外扰动通过ESO扩张成总扰动,然后对非线性实现实时精确补偿。自抗扰控制器中的非线性控制率参数的选择直接影响控制效果,设计了神经网络对非线性控制率的参数进行在线整定。同时引入了遗传算法对神经网络的重要参数进行优化。数值仿真表明该方法具有更强鲁棒性和控制精度。

PMSM position servo system is a strongly coupled and time-varying nonlinear system.Based on ADRC,the internal and external disturbances of the system are expanded into total disturbances by ESO,and then the real-time accurate compensation for nonlinearity is realized.The selection of the parameters of the nonlinear control rate in ADRC directly affects the control effect.A neural network is designed to tune the parameters of the nonlinear control rate online.At the same time,genetic algorithm is introduced to optimize the important parameters of the neural network.The numerical simulation shows that the method has stronger robustness and control accuracy.