首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于T-S模型的模糊神经网络
引用本文:孙增圻,徐红兵.基于T-S模型的模糊神经网络[J].清华大学学报(自然科学版),1997(3).
作者姓名:孙增圻  徐红兵
作者单位:清华大学计算机科学与技术系(孙增圻),智能技术与系统国家重点实验室(徐红兵)
摘    要:一种基于Takagi-Sugeno模型的模糊神经网络由前件网络和后件网络两部分组成。前件网络用来匹配模糊规则的前件,它相当于每条规则的适用度。后件网络用来实现模糊规则的后件。总的输出为各模糊规则后件的加权和,加权系数为各条规则的适用度。所提出的模糊神经网络具有局部逼近功能,且具有神经网络和模糊逻辑两者的优点。它既可以容易地表示模糊和定性的知识,又具有较好的学习能力。给出了调整规则后件参数及前件隶属度函数参数的学习算法,举例说明了它的逼近性能。

关 键 词:模糊逻辑  神经网络  T-S模型  函数逼近

Fuzzy neural network based on T S model
Sun Zengqi,Xu Hongbin.Fuzzy neural network based on T S model[J].Journal of Tsinghua University(Science and Technology),1997(3).
Authors:Sun Zengqi  Xu Hongbin
Institution:Sun Zengqi,Xu Hongbin Department of Computer Science and Technology,Tsinghua University, State Key Laboratory of Intelligent Technology and Systems,Beijing 100084
Abstract:The network based on Takagi Sugeno model is comprised of two parts:premise network and consequent network. The premise network is to match the premise of a fuzzy rule. The output of the premise network corresponds to the fitness value of a fuzzy rule. The consequent network computes the consequence of a fuzzy rule. The total output is equal to the weighting sum with weighting coefficients being equal to the fitness value of fuzzy rules. The proposed network has the ability of local mapping, which shows advantages of both neural network and fuzzy logic. The network can express the fuzzy and qualitative knowledge easily . It also has good learning capacity. The learning algirithms for tuning consequent parameters and membership parameters in premise are derived . An example is given to show the approximation abilities of the fuzzy neural network .
Keywords:fuzzy logic  neural network  Takagi Sugeno (T S) model  function approximation
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号