首页 | 本学科首页   官方微博 | 高级检索  
     

一种新型BP神经网络的生成算法
引用本文:赵春雨,韩学军,丛恒斌,闻邦椿.一种新型BP神经网络的生成算法[J].材料与冶金学报,1999(2).
作者姓名:赵春雨  韩学军  丛恒斌  闻邦椿
作者单位:东北大学黄金学院(赵春雨,韩学军),东北大学机械工程学院(丛恒斌,闻邦椿)
摘    要:提出了一种新型的神经网络的训练方法,包括输入层节点的自动寻找技术和隐含层节点的自动生成技术.为了验证该算法的正确性,利用该算法对同向回转观机驱动振动系统进行了神经网络逼近分析,计算机模拟结果表明该算法可以准确地确定输入层和隐含层节点的数目.

关 键 词:神经网络  输入层节点  隐含层节点  振动系统

A Novel Training Method for BP Neural Network
Zhao Chunyu, Han Xuejun, Cong Heagbin, Wen Bangchun.A Novel Training Method for BP Neural Network[J].Journal of Materials and Metallurgy,1999(2).
Authors:Zhao Chunyu  Han Xuejun  Cong Heagbin  Wen Bangchun
Affiliation:Zhao Chunyu; Han Xuejun; Cong Heagbin; Wen Bangchun
Abstract:A novel training method for BP neural network is proposed in this paper,which involves automatic finding the nodes of input layer and automatic generation of hidden layers.In order to verify the eHiciency of the proposed algorithm, a computer simulation is carried out, that a BP neural network is used to study the dynamics of vibrating machine with twin-motor rotating in the same direction. The results of computer simulation show that the proposed methed can effectively find the nodes of input layer and generate the nodes of hidden layers.
Keywords:BP neural network  input layer  output layer  vibrating machine
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号