动态加权最小二乘支持向量机.pdfVIP

第21 卷第10 期 　控　制　与　决　策 　2006 年10 月 Vol. 21 No. 10 　Control and Decision　 　　　Oct. 2006 : 1001-0920(2006) 10-1129-05 范玉刚, 李　平, 宋执环 ( , , 310027) : 提出一种基于动态加权最小二乘支持向量机( - ) 的时间序列预 方法. 动态加权 - 能够跟踪 LS SVM LS SVM 时变非线性系统的动态特性, 适合于系统辨识和时间序列预 ; 同时采用鲁棒方法确定权系数, 以减小噪声的影响. 将动态加权LS-SVM 算法应用于工业PT A 氧化过程中的4-C A 浓度预 , 结果显示, 动态加权LS-SVM 预 精度 高, 能够有效减小噪声的影响. : 最小二乘支持向量机; 时间序列预报; PTA 氧化过程 : T P274: A Dynamic Weighted Least Squares Support Vector Machines FA N Yu-gang , L I P ing , SON G Zhi-huan ( National Lab of Industrial Control T echnology Institute of Industrial Process Control Zhejiang University, , , Hangzhou 310027, China. Correspondent : FAN Yu-gang, E-mail: ygfan@ iipc.zju.edu.cn) Abstract: A time series forecasting method based on dynamic w eighted least squares support vector machine(LS- ) . - SVM is proposed Dynamic w eighted LS SVM is suitable for system recognition and time series prediction because the algorithm can track the dynamics of nonlinear time-varying systems. The w eights are determined by a robust method in order to reduce the effect of the noise data. Dynamic weighted LS-SVM is applied to predict the concen- tration of 4-Carboxybenzaldchydc (4-C A in purified terephthalic acid PTA oxidation process Results indicate) ( ) . that the proposed method reduces the effect of outliers and yields high accuracy. Key words: Least squares support vector; Time series prediction; PTA oxidation process 1 引　　言 , SVM . [4] (SVM) LS-SVM , Suykens , LS-SVM, LS- , SVM .