Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (3): 192-194.DOI: 10.3778/j.issn.1002-8331.2010.03.059

• 工程与应用 • Previous Articles     Next Articles

Generalized predictive control using LS-SVM error compensation

ZHAI Yong-jie,LI Hai-li,WANG Dong-feng,HAN Pu   

  1. College of Control Science and Engineering,North China Electric Power University,Baoding,Hebei 071003,China
  • Received:2008-10-28 Revised:2009-01-13 Online:2010-01-21 Published:2010-01-21
  • Contact: ZHAI Yong-jie


翟永杰,李海丽,王东风,韩 璞   

  1. 华北电力大学 控制科学与工程学院,河北 保定 071003
  • 通讯作者: 翟永杰

Abstract: Learning the multi-step forecast optimization strategy from Dynamic Matrix Control(DMC) and Model Algorithmic Control(MAC),Generalized Predictive Control(GPC) has a strong ability to overcome load disturbance,random noise and delay change,and the selected model has less parameters,so it is easy to control.However,according to research,GPC has some limitations in the problem of model mismatch.LS-SVM is developed based on Support Vector Machines,and has sound functions in regression and classification.On the basis of conscientiously studying the Least Squares Support Vector Machine(LS-SVM) principle,the GPC based on LS-SVM error compensation is proposed,and is simulated on two models.From the comparison with the conventional GPC,it proves that the algorithm had better performances in control.

Key words: Generalized Predictive Control(GPC), Least Squares Support Vector Machine(LS-SVM), error compensation

摘要: 广义预测控制(Generalized Predictive Control,GPC)汲取了DMC(Dynamic Matrix Control)、MAC(Model Algorithmic Control)中的多步预测优化策略,抗负载扰动、随机噪声、时延变化等能力强,且选取模型参数少,利于控制。然而,据研究发现GPC对模型失配问题有一定的局限性。最小二乘支持向量机(Least Squares Support Vector Machine,LS-SVM)是在支持向量机的研究基础上发展而来的,具有良好的回归、分类功能。在认真学习LS-SVM原理的基础上,提出了基于LS-SVM误差补偿的广义预测控制,并选择两个模型进行了仿真实验。通过与常规GPC的比较,表明了该算法具有更优的控制性能。

关键词: 广义预测控制, 最小二乘支持向量机, 误差补偿

CLC Number: