摘要
针对全局模型难以精确描述复杂工业过程的问题,提出一种基于模糊支持向量机(F_SVMs)的多模型(F_SVMs MM)建模方法。用模糊支持向量分类算法(F_SVC)对输入数据进行预处理,得到多模型模糊隶属度;用模糊支持回归算法(F_SVR)建立多模型(MM)估计器。应用该方法对pH中和滴定过程进行建模,仿真结果表明,F_SVMs MM跟踪性能好、泛化能力强,比USOCPN方法和标准支持向量机(SVMs)方法具有更好的性能和推广能力。
A new multiple model (MM) approach to model complex industrial process is proposed by using fuzzy support vector machines (F_SVMs). A fuzzy support vector classifier (F_SVC) algorithm is used to pretreat the input data set and obtain the fuzzy memberships. A fuzzy support vector regression (F_SVR) algorithm is used to construct MM estimator. By applying the proposed approach to a pH neutralization titration process experiment, F_SVMs MM not only provides satisfactory approximation and generalization property, but also achieves superior performance to USOCPN multiple modeling method and global modeling method based on standard SVMs.
出处
《控制与决策》
EI
CSCD
北大核心
2003年第6期646-650,共5页
Control and Decision
基金
国家"十五"863重大项目基金资助课题(2001AA413130)。
