基于高阶模糊推理系统的曲线拟合研究  

Study on curve fitting based on high-order fuzzy reference system

作　　者：刘燕[1] 谭冰[2] 赵静[3] 

机构地区：[1]大连工业大学信息科学与工程学院,辽宁大连116034 [2]南阳师范学院数学与统计学院,河南南阳473061 [3]广西民族大学数学与计算机科学学院,广西南宁530006

出　　处：《南阳师范学院学报》2011年第12期5-8,共4页Journal of Nanyang Normal University

基　　金：河南省自然科学基金资助项目(102300410184);河南省教育厅自然科学基金项目(2009B110017)

摘　　要：利用高阶模糊推理系统进行曲线拟合.首先构建高阶模糊推理系统模型,然后基于梯度下降的学习算法对高阶模糊推理系统模型的权值进行改进,最后进行曲线拟合,并与基于最小二乘的曲线拟合结果进行比较.结果表明,高阶模糊推理系统比基于最小二乘的曲线拟合具有更强的非线性处理能力,更适用于曲线拟合.This paper is mainly focus on curve fitting using a high-order fuzzy reference system. A model of highorder fuzzy reference system is firstly constructed, then an improved learning algorithm founded on gradient descent is used to update the weights of the high-order fuzzy reference system. An example about curve fitting based on high-order fuzzy reference system is given in this paper, the results show that high-order fuzzy reference system has higher capability on nonlinear problem than curve fitting based on least squares, then is more suitable for curve fitting.

关 键 词：模糊推理系统 曲线拟合 梯度下降 Pi—Sigma 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] O233[自动化与计算机技术—控制科学与工程]