基于Matlab的BP神经网络结构与函数逼近能力的关系分析

人工神经网络是一种非线性动态数学模型,广泛应用于非线性系统建模、系统辨识、函数逼近等方面。介绍BP网络的结构和学习过程,并介绍利用Matlab人工神经网络工具箱设计BP网络的步骤,在此基础上设计了BP网络以验证其函数逼近能力,仿真结果说明了BP网络具有很强的函数逼近能力。并分析BP网络结构和函数逼近能力的关系,得出网络的结构直接影响网络对函数的逼近能力和效果。

Analysis of Relation between the Structure of BP Feed - forward Neural Network and Precision of Function Proximate Based on Matlab

Artificial Neural Network is a nonlinear dynamic mathematic model and can be applied to model nonlinear system,class identification and function approximation.In this paper,the structure and learning process of BP feed-forward neural network are presented.The paper introduces the approach to design a BP feed-forward neural network using Matlab/neural network toolbox,based on the network verify the ability of function approximation.The result shows that BP feed-forward neural network has a strong ability to approximate function.The paper also analyzes the relation between the structure of BP feed-forward neural network and precision of function approximate,educing that the structure BP feed forward neural network directly affects the approximate ability and effect.