红外和神经网络无损分析VC 银翘片中二组分含量
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:


Two-component Nondestructive Analysis of VC Yinqiao Tablets with NIR And Bp Neural Network
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    应用近红外光谱和BP 神经网络对VC 银翘片进行非破坏定量分析,讨论了神经网络 输入点数的选择和影响神经网络各参数的确定。采用了贝叶斯规范化的网络训练方法,解决了BP 神经网络稳定性较差的问题,提高了网络的预测精度。实验及数据处理结果表明,应用本文方法对VC 银翘片进行非破坏定量分析方法是可行和有效的。

    Abstract:

    The application of NIR and artificial neural network for Vc Yinqiao tablet nondestructive quantitative analysis were investigated. The choice of ANN inputting node number and ANN’s parameters affecting network were discussed , used Bayesian Regularization mode to train Bp Neural Network.With this mode ,the question of Bp Neural Network's weak stalility were resolved ,improved the forecasting precision of Neural Network. The experiment and the result of data process shows the application of Bp Neural Network In the nondestructive quantitative Analysis of Vc Yinqiao Tablets is effective and viable.

    参考文献
    相似文献
    引证文献
引用本文

白英奎,申铉国,丁 东.红外和神经网络无损分析VC 银翘片中二组分含量[J].激光与红外,2004,34(5):354~356
. Two-component Nondestructive Analysis of VC Yinqiao Tablets with NIR And Bp Neural Network[J]. LASER & INFRARED,2004,34(5):354~356

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2005-05-27
  • 出版日期: 2004-10-15