基于ELM的永磁直线同步电机位移跟踪动态面反步滑模控制

针对永磁直线同步电机(PMLSM)易受到参数摄动、负载扰动等不确定因素的影响,进而影响其位移跟踪控制精度的问题,提出一种基于非线性干扰观测器(NDO)和极限学习机(ELM)的动态面反步滑模控制方法.首先,通过构造NDO对系统模型中的非匹配不确定项进行动态观测,并将反步控制、动态面控制与滑模控制相结合,完成PMLSM位移跟踪控制器的设计,在提高系统抗干扰能力的同时,避免常规反步控制中的“微分爆炸”问题;其次,采用ELM神经网络对系统模型中的匹配不确定项进行逼近估计,并将输出的估计值引入设计的动态面反步滑模控制器中进行补偿;再次,采用人工鱼群-蛙跳混合算法对所设计控制器的主要参数进行优化设计,提高系统的收敛速度和稳定精度;最后,将所提出控制方法与其他控制方法进行仿真对比,仿真结果表明了所提出方法的有效性.

Dynamic surface backstepping sliding mode control for the displacement tracking of permanent magnet linear synchronous motor based on extreme learning machine

For the problem of that the permanent magnet linear synchronous motor(PMLSM) is prone to be affected by uncertain factors such as parameter perturbation, load disturbance, etc., which affects its displacement tracking control accuracy, a dynamic surface backstepping sliding mode control method is proposed based on the nonlinear disturbance observer(NDO) and the extreme learning machine(ELM). Firstly, the NDO is developed to dynamically observe the mismatched uncertainty in the system model, and the displacement tracking controllers for the PMLSM are presented by combining backstepping control with dynamic surface control and sliding mode control, which improves the anti-interference ability of the system, and avoids the ``differential explosion'' problem during using the conventional backstepping control. Then, ELM neural networks are used to approximate the matched uncertainties in the system model, and the estimated values of the outputs are introduced into the designed dynamic surface backstepping sliding mode controllers for compensation. Furthermore, the artificial fish-frog jump hybrid algorithm is adopted to optimize the main parameters of the designed controllers, which improves the convergence speed and stability accuracy of the system. Finally, the proposed control method is compared with other control methods, and the simulation results verify the effectiveness of the proposed control method.