电动加载系统分数阶迭代学习复合控制

针对电动加载系统存在多余力矩扰动的问题，提出一种以位置闭环和力矩闭环为反馈控制、迭代学习控制为补偿控制的复合控制策略。为提高加载系统动态性能及降低建立模型的复杂性，驱动永磁同步电机采用直接转矩控制方式，建立加载系统频域模型。在位置闭环和力矩闭环采用分数阶PIλDμ控制器代替常规PID控制器，迭代学习补偿控制采用分数阶迭代学习控制器，利用分数阶微积分的信息记忆特性提高控制系统的动态性能和鲁棒性，通过理论分析给出分数阶PD型迭代学习控制器的收敛条件。对正弦和梯形波载荷进行力矩加载实验及多余力矩抑制实验，验证了该控制方法的有效性。 

Composite control for electric dynamic loading system based on fractional order iterative learning

To solve the problem of surplus torque, a composite torque controller is proposed for electric dynamic loading system where the position and torque closed-loops are applied for feedback control and the iterative learning control is applied for feedforward compensation. To improve the dynamic response and reduce the complexity of the system model, direct torque control is used for the permanent magnet synchronous motors. The mathematical model of the electric dynamic loading system is established in frequency domain. The traditional PID and iterative learning control are replaced by fractional order PI^λD^μ for position and torque closed-loops. Fractional order iterative learning control is used for feedforward compensation. The dynamic response and robustness of the controller are enhanced owing to the information memory characteristics of fractional order calculus. The convergence conditions of the fractional order PD-type iterative learning control are presented through theoretical analysis. Experiments on the loading torque with sinusoidal and trapezoidal wave torque and testing of the surplus torque suppression were performed. The experimental results prove that the controllers are effective in achieving a better torque tracking and restraining the surplus torque for the electric dynamic loading system.