计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (2): 223-226.DOI: 10.3778/j.issn.1002-8331.2011.02.067

• 工程与应用 • 上一篇    下一篇

基于SVM逆系统的非线性系统广义预测控制

李超峰,卢建刚,孙优贤   

  1. 浙江大学 工业控制技术国家重点实验室,杭州 310027
  • 收稿日期:2009-08-21 修回日期:2009-10-13 出版日期:2011-01-11 发布日期:2011-01-11
  • 通讯作者: 李超峰

GPC algorithm of nonlinear systems based on support vector machine inverse control

LI Chaofeng,LU Jiangang,SUN Youxian   

  1. State Key Laboratory of Industrial Control Technology,Zhejiang University,Hangzhou 310027,China
  • Received:2009-08-21 Revised:2009-10-13 Online:2011-01-11 Published:2011-01-11
  • Contact: LI Chaofeng

摘要: 针对工业过程中普遍存在的非线性被控对象,提出了一种基于支持向量机(SVM)逆系统的广义预测控制算法。该方法根据广义预测控制基于预测模型的特点,将基于支持向量机系统辨识的方法应用于逆系统构建和广义预测控制。该方法利用SVM强大的非线性映射能力离线辨识被控非线性系统的α阶逆模型,并将辨识出的逆模型连接在原被控统之前形成一个α阶纯延时伪线性系统。然后采用广义预测控制(GPC)算法实现对构造出的伪线性系统的预测控制。仿真实验表明了该算法的有效性和优越性。

关键词: 支持向量机, 非线性系统, α阶逆系统, 广义预测控制

Abstract: To deal with the control problems of non-linear system,a new Generalized Predictive Control(GPC) algorithm based on Support Vector Machine(SVM) inverse control is proposed.The SVM is applied to identify the αth-order inverse model of the controlled plant,which is cascaded before the original system to creat a pesudo-linear sub-system.And then for the simply pesudo-linear sub-system the generalized predictive control algorithm can be used.Simulation results of a typical non-linear system show that the presented method can accurately model the inverse dynamic system,and has the characteristics of great tracking accuracy,better robustness stability,and anti-disturbance ability.

Key words: Support Vector Machine(SVM), nonlinear system, αth-order inverse system, generalized predictive control

中图分类号: