摘要
针对单电磁导向系统参数变化及外部扰动对悬浮气隙高度产生的影响,提出了RBF神经网络自适应滑模控制方法.采用RBF神经网络并利用其学习功能,对直线电梯单电磁悬装置不确定参数进行自适应补偿,取代了常规滑模控制切换部分,并且消除了系统高频抖振现象.通过比例微分并行控制提高了RBF神经网络参数的收敛性,改善了局部极小现象的发生,增强了系统的鲁棒性,并采用Lyapunov稳定性理论证明了系统的稳定性.Matlab仿真显示该方法具有良好的跟踪性和鲁棒性.
For the influence of parameter variation and extra disturbance of single electromagnetic guiding system on the maglev air-gap altitude, a RBF neural network method based on adaptive sliding mode control was proposed. With the RBF neural network and its learning function, the adaptive compensation of uncertain parameters for the single electromagnetic maglev actuator of linear elevator was performed, which could replace the switching part of conventional sliding mode control and eliminate the high-frequency chattering phenomenon of the system. The convergence of parameters for the RBF neural network gets enhanced through the proportional and differential parallel control, the occurrence of local minimum phenomenon is minished, and the robustness of the system is improved. In addition, the stability of the system is proved with Lyapunov stability theory. The Matlab simulated results show that the proposed method exhibits good tracking performance and robustness.
出处
《沈阳工业大学学报》
EI
CAS
北大核心
2013年第3期251-256,共6页
Journal of Shenyang University of Technology
基金
辽宁省教育厅科学技术研究资助项目(L2010404)
