J4

• 化学 • 上一篇    下一篇

人工神经网络-近红外光谱法定量分析Cofrel粉末药品

刘名扬1,2, 赵景红2, 张晓明3, 任玉林3   

  1. 1. 清华大学 化学系, 北京 100084; 2. 辽宁出入境检验检疫局 技术中心, 辽宁省 大连 116001; 3. 吉林大学 化学学院, 长春 130012
  • 收稿日期:2006-05-26 修回日期:1900-01-01 出版日期:2007-03-26 发布日期:2007-03-26
  • 通讯作者: 刘名扬

Quantitative Analysis of Cofrel Medicines in Powder by Artificial Neural Network Methodnear Infrared Spectroscopy

LIU Mingyang1,2, ZHAO Jinghong2, ZHANG Xiaoming3, REN Yulin3   

  1. 1. Department of Chemistry, Tsinghua University, Beijing 100084, China; 2. Centre of Technique, Liaoning Entryexit Insection and Quarantine Bureau, Dalian 116001, Liaoning Province, China; 3. College of Chemistry, Jilin University, Changchun 130012, China
  • Received:2006-05-26 Revised:1900-01-01 Online:2007-03-26 Published:2007-03-26
  • Contact: LIU Mingyang

摘要: 用人工神经网络(ANN)解析Cofrel药品的近红外(NIR)漫反射光谱, 实现了对其有效成分磷酸苯丙哌啉的精确快速定量测定. 设计了最好的ANN模型. 分别讨论了输入层和隐含层节点数对测定结果的影响. 用HPLC法的测定结果作为标准, 磷酸苯丙哌啉浓度预 测值的相对误差RE (%)<4.0%, 可以应用于Cofrel药品实际生产中的质量控制.

关键词: 人工神经网络, 近红外漫反射光谱, 定量分析, 科福乐

Abstract: Artificial neural network (ANN) arithmetic was used to resolve near infrared (NIR) reflectance spectra of Cofrel medicine in powder. The contents of benproperine phosphate, which is effective pharmaceutical component in Cofrel medicine, has been accurately fast quantitatively predicted. The best model of ANN was designed. The effects of input nodes and hidden nodes on the results were discussed respectively. Comparison of the results with those of HPLC indicates the relative error of benproperine phosphate is less than 4.0%. The analytical result could be applied to the qualitative control of Cofrel medicines in practical manufacture.

Key words: artificial neural network, near infrared reflectance spectra, quantitative analysis, Cofrel

中图分类号: 

  • O657.1