期刊导航

论文摘要

人工神经网络在虎杖提取过程质量控制的研究

Study on Fast Quality Control in Extracting Process of Rhizoma Polygoni Cuspidati with ANN

作者:汪学楷(四川大学 化学工程学院);杨阳(四川大学 化学工程学院);武佳(四川大学 化学工程学院);李晖(四川大学 化学工程学院)

Author:Wang Xuekai(School of Chem. Eng., Sichuan Univ.);Yang Yang(School of Chem. Eng., Sichuan Univ.);Wu Jia(School of Chem. Eng., Sichuan Univ.);Li Hui(School of Chem. Eng., Sichuan Univ.)

收稿日期:2011-11-09          年卷(期)页码:2012,44(4):176-179

期刊名称:工程科学与技术

Journal Name:Advanced Engineering Sciences

关键字:虎杖苷;白藜芦醇;人工神经网络;遗传算法;偏最小二乘法

Key words:polydatin;resveratrol;artificial neural network(ANN);genetic algorithm;partial least square(PLS)

基金项目:

中文摘要

针对人工神经网络收敛速度慢及容易陷入局部最优解的缺点,结合遗传算法全局搜索的特点,提出了一种基于遗传算法和人工神经网络建立近红外透射光谱快速测定虎杖提取液中虎杖苷和白藜芦醇含量的方法。结果表明,模型效果良好,方法简便、准确可靠。与偏最小二乘法(PLS)相比较,该方法具有更高的预测能力,适用性更强,有望应用于药物提取过程的中间控制。

英文摘要

In view of the shortcomings of artificial neural network, such as slow convergence, easily falling into local optimum, based on ANN and GA, a method was put forward to establish a model which could achieve fast analysis of Polydatin and Resveratrol in the extracting process of Rhizoma Polygoni Cuspidati by near infrared spectroscopy(NIR) . The method was proved to be convenient, rapid, accurate and reliable. It has the better precision of prediction and applicability compared with PLS, and provides a promising application for pharmaceutical production process and quality control.

关闭

Copyright © 2020四川大学期刊社 版权所有.

地址:成都市一环路南一段24号

邮编:610065