基于泛函网络的非线性回归预测模型及学习算法  

Non-linear regression forecast model based on functional networks and learning algorithm

作　　者：何登旭[1] 李艳芳[1] 刘向虎[2] 周永权[1] 

机构地区：[1]广西民族大学数学与计算机科学学院,南宁530006 [2]运城学院应用数学系,山西运城044000

出　　处：《计算机工程与应用》2008年第24期74-77,共4页Computer Engineering and Applications

基　　金：国家自然科学基金 No.60461001;国家民委科学基金(No.05GX06) ;广西省自然科学基金No.桂科自 0728054;广西民族大学研究生教育创新计划项目(No.gxun-chx0749)~~

摘　　要：在非线性回归预测中,预测函数的拟合是其难点和关键,直接影响预测精度。当系统非线性较强时,传统方法不易于处理,拟合和预测结果不理想。泛函网络是最近提出的一种对神经网络的有效推广,在处理非线性问题时有一定的优势。为此提出了基于泛函网络的非线性回归预测模型和相应的学习算法。并分别就一元非线性回归预测和多元非线性回归预测给出了相应的实例。计算机仿真结果表明,泛函网络预测模型拟合度和预测精度都明显高于某些传统的方法,有较好的理论和应用价值。Fitting of forecast function is very difficult and important in non-linear regression forecast problems.The accuracy is directly affected by the fitting of forecast ＂function.Non-linear model in the traditional method is difficult to solve the system whose non-linear is stronger,and the result of fitting and forecast is not ideal.Function network is a recently introduced extension of neural networks.It has certain advantages solving non-linear problems.Non-linear regression forecast model and learning algo- rithm based on functional networks are proposed in this article.Some examples about one-dimensional non-linear regression fore- cast and multi-dimensional non-linear regression forecast are pi＇ovided.The simulation results demonstrate that forecast model based on functional networks whose accuracy of fitting and forecasting is more than some traditional methods has some value about theory and application.

关 键 词：泛函网络 非线性回归 预测 学习算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]