J4

• 化学 • 上一篇    下一篇

人工神经网络-近红外光谱法非破坏定量分析Norvasc药片

刘名扬1,2, 赵景红1, 孟 昱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. 辉瑞制药有限公司, 辽宁省 大连 116600
  • 收稿日期:2006-11-27 修回日期:1900-01-01 出版日期:2007-09-26 发布日期:2007-09-26
  • 通讯作者: 刘名扬

Nondestructive Quantitative Analysis of Norvasc in Tablets by Artificial Neural Network Methodnear Infrared Spectroscopy

LIU Mingyang1,2, ZHAO Jinghong1, MENG Yu3   

  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. Prefz Pharmaceutical Co., Dalian 116600, Liaoning Province, China
  • Received:2006-11-27 Revised:1900-01-01 Online:2007-09-26 Published:2007-09-26
  • Contact: LIU Mingyang

摘要: 用人工神经网络(ANN)解析Norvasc药片的近红 外(NIR)漫反射光谱, 实现了对其中有效成分苯磺酸氨氯地平精确的非破坏定量测定. 设计了最佳的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. 以HPLC法的测定结果为标准, 苯磺酸氨氯地平浓度预测值的相对误差RE(%)<4.2%.

关键词: 人工神经网络, 近红外漫反射光谱, 非破坏定量分析, Norvasc

Abstract: Artificial neural network (ANN) algorithm was used to analyse the near infrared (NIR) reflectance spectrum of Norvasc in tablet medicine. The content of amlodipine besylate, which was pharmaceutically available in Norvasc medicine, was accurately nondestructively quantitatively determined. The best model of ANN was designed. The effects of input nodes and hidden nodes on the results were discussed, respectively. Compared the results with those of HPLC, the relative error (RE%) of mlodipine besylate was less than 4.2%. 

Key words: artificial neural network, near infrared reflectance spectra, nondestructive quantitative analysis, Norvasc

中图分类号: 

  • O657.33