基于模糊聚类方法的T-S模糊系统建模  

T-S Fuzzy System Modeling Based on Fuzzy Clustering Method

作　　者：李林峰[1] 孙长银[1] 

机构地区：[1]河海大学电气工程学院,南京210098

出　　处：《三峡大学学报（自然科学版）》2006年第2期147-150,共4页Journal of China Three Gorges University：Natural Sciences

基　　金：国家自然科学基金重大项目(50595412);中国博士后基金项目(2004036125)

摘　　要：提出了用一个聚类验证准则设计模糊c均值聚类算法,这个聚类验证准则是用来确定模糊c均值算法中合适的聚类数.针对T-S模糊模型,由模糊c均值聚类算法确定其逻辑前件参数,进而采用最小二乘算法确定模糊推理规则的后件参数.最后,应用本文建模方法对一个非线性实例进行仿真计算,并与其它方法进行了比较,结果表明本文方法是有效的.This paper develops a new cluster validity criterion designed for the fuzzy c-mean clustering algorithm. The proposed cluster validity criterion is used to determine the appropriate number of clusters in FCM. And then import T-S fuzzy model, the premise parameters are determined by fuzzy c-means clustering algorithm. Furthermore a least square algorithm is provided to determine the consequent part of each rule. A practical application is used to illustrate the proposed modeling approach; and its performance is compared with those of other approaches. The new approach is proved to be effective.

关 键 词：T—S模糊模型 模糊C均值聚类算法(FCM) 最小二乘算法 

分 类 号：O159[理学—数学]