【作者】 荣莉莉； 王众托；

【Author】 Rong Lili Wang Zhongtuo(Institute of Systems Engineering, Dalian University of Technology, 116024)

【机构】 大连理工大学系统工程研究所；

【摘要】 本文提出了一种提高多输入模糊推理系统精度的算法。首先,建立一个多输入单输出的模糊神经网络(MISO-FNN),从样本数据中提取描述系统关系的模糊规则。然后,利用模糊神经网络进行学习,在调整隶属函数的参数同时,对规则进行动态筛选,以使所得到的规则高精度地反映样本数据的特性。通过一个仿真实例验证了该算法的有效性。更多还原

【Abstract】 In this paper, an algorithm of improving the accuracy of the multi-inputs fuzzy inference system is proposed. A multi-input single-output fuzzy-neural network (FNN) model is used to realize the algorithm. Through this FNN, the fuzzy rules are extracted from numerical samples directly. The Back-Propagation learning method is used to improve the accuracy of the fuzzy inference system. The method to determine the fuzzy rules dynamically in the learning process is also presented. The fuzzy rules obtained can describe the characteristic of the samples with high accuracy.更多还原