基于LS-SVM的网络化控制系统自适应预测控制  

Approach of Adaptive Prediction Control on Networked Control Systems Based on Least-squares Support Vector Machines

作　　者：李春茂[1] 肖建[1] 张玥 

机构地区：[1]西南交通大学电气工程学院,四川成都610031

出　　处：《系统仿真学报》2007年第15期3494-3498,3502,共6页Journal of System Simulation

基　　金：国家自然科学基金(60674057);教育部博士点基金资助项目(20040613013);四川省应用基础研究基金(05JY029-006-04)

摘　　要：针对网络化控制系统(NCS)中的随机时变时延,提出了一种用最小二乘支持向量机(LS-SVM)预估网络时延的方法。先将网络时延建模为非线性时间序列,再用径向基函数(RBF)作为LS-SVM的核函数,建立了网络控制系统的时延预测模型,然后用该模型预估的时延作为控制器的参数,对网络化控制系统的时延进行补偿和预测控制。仿真结果表明提出的时延预测方法,对网络控制系统的随机时变时延有较高的预测精度,根据该时延设计的控制器能使系统的输出很好地跟踪期望的输出。A time-delay estimate algorithm for networked control systems based on least-square support vecotr machines was proposed, Modeling the time delay of networked control systems as a nonlinear time series, the future time delay could be predicted by least-square support vector machines with the radial basis function kernel. An adaptive predictive control algorithm was proposed to compensate and predicte for the time delay of networked control systems. Simulation results show that the proposed method has good performance in time delay predictation of networked control systems, and the plant output can trace desired output effectively.

关 键 词：网络化控制系统 最小二乘 支持向量机 时延估计 自适应控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]