基于RBF网络非线性系统逆控制的一种设计方案  

Design of Inverse Control of Nonlinear System Based on RBF Neural Network

作　　者：张绍德[1] 李坤[1] 张世峰[1] 

机构地区：[1]安徽工业大学电气信息学院,马鞍山243002

出　　处：《系统仿真学报》2006年第9期2688-2690,共3页Journal of System Simulation

基　　金：安徽省"十五"攻关项目资助(01012053);安徽省教育厅自然科学基金资助项目(2004KJ059)

摘　　要：基于逆动力学控制的思想,提出一种RBF神经网络逆控制与PID控制相结合的在线自学习控制方案。辨识器采用RBF神经网络结构和最近邻聚类算法,实现了对系统逆动力学模型的动态辨识。并将辨识模型作为控制器模型,与被控对象串联,构成一个动态伪线性对象,从而使非线性对象的控制问题转换为线性对象的控制问题。仿真实验证明该控制策略不仅能使系统具有良好的动态跟踪性能和抗干扰能力,而且具有较强的鲁棒性。Based on the though t of inverse system control, a method of on-line self-learning control strategy was proposed, which combines inverse control based on RBF neural network with PID control. The system identifier based on RBF neural network which applies nearest neighbor clustering algorithm realizes the identification of the inverse dynamic system model. The model of controller which is the copy of identifier and the plant controlled are in series, which forms a dynamic pseudo linear system. Consequently, the control problem of non-linear plant is converted into that of linear plant. With the help of simulations, the control strategy based on RBFNN inverse controller can not only improve dynamic track performance and resistance to disturbance of system, but also possess excellent robustness.

关 键 词：RBF神经网络 直接逆控制 在线自学习 最近邻聚类算法 

分 类 号：TP273.5[自动化与计算机技术—检测技术与自动化装置]