工矿自动化2013，Vol.39Issue(2)：47-50,4.DOI:10.7526/j.issn.1671-251X.2013.02.012
基于径向基神经网络的电涡流传感器输出特性拟合研究
Research of output characteristic fitting of eddy-current sensor based on radial-basis function neural network
摘要
Abstract
In view of problem that eddy-current sensor cannot reflect measured physical quantity accurately caused by higher nonlinear of output characteristic parameter, the paper proposed a scheme of using RBF neural network to fit output characteristic parameter of eddy-current sensor. The scheme uses newrb function to create RBF neural network, and takes measured physical quantity as input matrix and output of eddy-current sensor as output matrix to train the RBF neural network, so as to obtain low root-mean-square error and smooth output characteristic fitting curve of eddy-current sensor. The simulation result showed that RBF neural network can effectively realize fitting of output characteristic of eddy-current sensor by selecting proper creating function and expanding coefficient.关键词
电涡流传感器/输出特性拟合/径向基神经网络/newrb函数/扩展系数Key words
eddy-current sensor/ output characteristic fitting/ RBF neural network/ newrb function/ expanding coefficient分类
矿业与冶金引用本文复制引用
尤文坚,梁兵,李荫军..基于径向基神经网络的电涡流传感器输出特性拟合研究[J].工矿自动化,2013,39(2):47-50,4.基金项目
广西省教育厅自然科学科研项目(200911LX547). （200911LX547）
