计算机工程与科学2012，Vol.34Issue(4)：82-87,6.DOI:10.3969/j.issn.1007-130X.2012.04.017
基于UKF的自适应模糊推理神经网络
An Adaptive Network Based Fuzzy Inference System Based on UKF
摘要
Abstract
Much of the current research interest in neuro-fuzzy hybrid systems is focused on how to generate an optimal number of fuzzy rules in a neuro-fuzzy system and investigate the automated methods of adding and pruning fuzzy rules. To deal with this problem, an adaptive network based fuzzy inference system (ANFIS) based on UKF is presented. Firstly, fuzzy rules and their parameters of ANFIS-RR are obtained by subtractive clustering. Secondly, the parameters are learned by linear least square and the back propagation algorithm. Thirdly, the non-linear dynamical system expression of fuzzy networks is analyzed, and LLS and UKF are used to learn linear and non-linear parameters respectively. Then, a method of error descending rate is used as the fuzzy rule pruning strategy, so that the rule which plays an unimportant role in the system is deleted. Finally, by typical experiments of function approximation and system identification indicate that fuzzy networks obtained by the proposed algorithm has a more tightened structure and better generalization than other algorithms.关键词
UKF/自适应模糊推理神经网络/规则约简/系统辨识/函数逼近Key words
unscented Kalman filter/adaptive network based fuzzy inference system (ANFIS)/rule reduction/ system identification/function approximation分类
信息技术与安全科学引用本文复制引用
徐小来,朱华勇,贺中武,王伟,牛轶峰..基于UKF的自适应模糊推理神经网络[J].计算机工程与科学,2012,34(4):82-87,6.基金项目
中国博士后科学基金资助项目(201150M1562) （201150M1562）
中国博士后特别资助项目(201104765) （201104765）
