Abstract：Magnetic levitation (maglev), as the only way of the next generation of trains, its suspension control algorithm has important academic significance and practical application value. An adaptive sliding mode control method based on RBF neural network compensation is proposed to solve the problem of precise control of suspension height of the suspension ball with fixed height levitation instrument as the object of study. The mathematical model of suspension ball is improved and sliding mode variable structure controller is designed. Due to the inherent discontinuous switching characteristics of sliding mode controller, the chattering phenomenon of the system is caused. Using the constant speed reaching law can suppress the chattering phenomenon, thus improving the dynamic response performance of the system. Finally, the universal approximation property of RBF neural network is used to reach the on-line approximation of the state variable term which can not be accurately modeled and detected in the system is used to compensate the control system and the stability analysis is carried out by using the Lyapunov criterion at the same time. The simulation results show that the adaptive sliding mode controller based on RBF neural network compensation is superior to the classical PID controller in response time, anti-interference and overshoot.
杨杰, 黄晨, 石恒. 径向基函数神经网络补偿的悬浮球悬浮高度自适应滑模控制[J]. 电气技术, 2020, 21(2): 26-30.
Yang Jie, Huang Chen, Shi Heng. Adaptive sliding mode control for suspension height of the suspended ball based on rbf neural network compensation. Electrical Engineering, 2020, 21(2): 26-30.
2020, Vol. 21 
