不确定磁悬浮轴承模型的自适应PID控制研究  

Study on the Control of the Adaptive PID of the Model of Uncertain Magnetic Suspension Bearings

作　　者：李莹[1] 王西伟[1] 谢翔[1] 顾冬梅[1] 董长海 LI Ying, WANG Xing - wei, XIE Xiang, GU Dong - mei, DONG Chang - hai(School of Electrical and hfformation Engineering, Anhui University of Science and Technology, Huainan Anhui 232001, Chin)

机构地区：[1]安徽理工大学电气与信息工程学院,安徽淮南232001

出　　处：《安徽理工大学学报（自然科学版）》2018年第2期40-43,共4页Journal of Anhui University of Science and Technology:Natural Science

摘　　要：针对磁悬浮系统具有开环不稳定、非线性以及传统PID控制器由于固定参数无法达到很好的控制效果问题,提出一种RBF神经网络前馈逆补偿-模糊RBF神经网络反馈控制方法,将模糊控制和RBF神经网络融合到PID控制器参数调整中,满足磁悬浮系统静态和动态性能要求。实验结果表明,采用常规PID控制有大约4%的超调,改进后的控制基本没有,且响应时间提前0.3s,具有更好的适应性和鲁棒性,可以更有效地控制磁悬浮系统。Aiming at the magnetic levitation system＇s open-loop instability,nonlinearity and difficulty in achieving good control for the traditional PID controller due to the fixed parameters,a RBF neural network inverse compensation was put forward,which was a control method of adaptive fuzzy RBF feedback control and integrated fuzzy control and RBF neural network into the PID controller parameter adjustment to satisfy the static and dynamic performance requirements in magnetic levitation system. The experimental results show that the conventional PID control had about 4% overshoot,the improved control overshoot was almost zero,and the response time was 0. 3 s ahead; therefore with better adaptability and robustness,maglev system could be controlled more effectively.

关 键 词：PID控制 RBF神经网络 模糊RBF神经网络 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]