基于数据简化及卡尔曼滤波方法的T-S模糊模型辨识  

Data Reduction and Kalman Filter Method for T-S Fuzzy Model Identification

作　　者：刘小雍 方华京 LIU Xiao-yong;FANG Hua-jing(College of Engineering and Technology,Zunyi Normal College,Zunyi 563002,China;School of Automation,Huazhong University of Science&Technology,Wuhan 430074,China)

机构地区：[1]遵义师范学院工学院,遵义563002 [2]华中科技大学自动化学院,武汉430074

出　　处：《科学技术与工程》2018年第9期254-259,共6页Science Technology and Engineering

基　　金：国家自然科学基金项目(61473127);贵州省教育厅青年项目(黔教合KY字[2016]254);遵义师范学院博士项目基金(遵师BS[2015]04号);贵州省千层次创新人才项目(遵市科合人才[2017]19号)

摘　　要：T-S模糊辨识方法已成为模糊集理论和应用中的重要研究。仅在获取被辨识系统输入-输出数据情况下,提出一种能自动建立T-S(Takagi-Sugeno)模糊模型的有效方法。在结构辨识阶段,首先利用GK(Gustafson-Kessel)聚类算法对原始数据集进行聚类划分为若干个子集;再引入重叠因子对每个子集的大小做进一步简化;同时获取各个聚类中心及宽度。对于参数辨识,应用卡尔曼滤波方法求解后件参数值。最后,通过两个实验研究,论证所提出方法的有效性和优越性。An efficient method for automatically constructing a T-S(Takagi-Sugeno)fuzzy model,where only the input-output data of the identified system are available,is presented.In the structure identification step,a clustering method based on the GK(Gustafson-Kessel)algorithm is applied to partition the original data sets into several subsets and the size of each subset is again repartitioned and reduced by the introduced overlap factor,where the clustering centers and spreads corresponding to each subset are also obtained respectively.In the parameter identification step,the Kalman filter algorithm is applied to solve consequent parameter values of the T-S fuzzy model from input-output data.Finally,the effectiveness and superiority of the proposed approach is demonstrated using two experiments.

关 键 词：T-S模糊建模 数据简化 卡尔曼滤波 重叠因子 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]