ReviewofSVM-basedControlandOnlineTrainingAlgorithm

代码说明：

  支持向量机以其模型结构简单、较好的推广能力和全局最优解等特点已经被用来进行智能 控制的研究，主要包括采用支持向量机回归的非线性时间序列的建模与预测、系统辨识等建模方 面的研究以及优化控制、学习控制和预测控制等方面的研究以及采用支持向量机的故障诊断的研 究。由于现有SVMR基于二次规划的优化方法不适合控制过程的在线训练，因此出现了对SVMR 在线训练算法的研究。分析了国内外这些研究内容的最新研究进展，旨在探讨归纳支持向量机在控 制领域研究的主要成果和存在的问题，以便为进一步的研究提供一定的支持与帮助(Abstract:The support vector machines,which characterizes a simple model structure,good generalization,global optimal solution etc.,has been applied in the intelligent control.The current research includes SVMR modeling and prediction of nonlinear times serial,system identification,optimal control,learning control,predict control etc.,and SVM-based fault diagnosis.Since SVMR based on the quadratic programming unfits for the online training of control,the current research includes the online training of SVMR.The latest development of those contents in the domestic and overseas research were analyzed,and the main achievements and problems in SVMR control were summed up in order to provide some support and help for the further research. )

文件列表：

支持向量机控制与在线学习方法研究的进展.kdh,566479,2009-02-13