基于近红外光谱技术的甘草提取过程最优建模方法研究  

Research on optimal modeling strategy for licorice extraction process based on near-infrared spectroscopy technology

作　　者：王海霞[1,2] 所同川[1,2] 余河水[1,2] 李正[1,2,3] WANG Hai-xia SUO Tong-chuan YU He-shui LI Zheng(College of Pharmaceutical Engineering of Traditional Chinese Medicine, Tianjin University of Traditional Chinese Medicine, Tianjin 300193, China Modern Industrial Technology Research Institute of Traditional Chinese Medicine, Tianjin University of Traditional Chinese Medicine, Tianjin 300193, China Tianjin Key Laboratory of Modern Chinese Medicine, TianjiTt University of Traditional Chinese Medicine, Tianjin 300193, China)

机构地区：[1]天津中医药大学中药制药工程学院,天津300193 [2]天津中医药大学现代中药产业技术研究院,天津300193 [3]天津市现代中药省部共建国家重点实验室(培育),天津300193

出　　处：《中国中药杂志》2016年第19期3537-3542,共6页China Journal of Chinese Materia Medica

基　　金：天津市科技创新体系及条件平台建设计划项目(14TXZYJC00440);天津市科技创新体系及条件平台建设计划项目(15PTCYSY00030)

摘　　要：通过实时采集正常操作条件下及发生异常工况时的甘草提取液的动态近红外光谱数据,结合主成分分析法(PCA)、偏最小二乘回归法(PLSR)和平行因子-偏最小二乘回归联用法(PARAFAC-PLSR)建立3种甘草提取过程的实时监测模型,并分析各种模型的特点。结果表明,基于3种方法建立的模型均能在一定程度上预测异常加热工况的发生,但同时也存在一定误判。其中,PCA方法建立的模型出错率最高,在60 min之前就出现3次"故障"误判,不适用于该过程的分析应用。而PLSR和PARAFAC-PLSR模型基本效果相似,校正集相关系数分别高达0.934 2,0.928 1,验证集相关系数也分别达到了0.856 7,0.828 3;并且这2种方法建立的预测模型误判率较低,首次成功预测的故障均发生于75 min。此外,PLSR和PARAFAC-PLSR模型均能在一定程度上预测出系统状态的走势。说明基于动态近红外光谱动态数据建立的PLSR和PARAFAC-PLSR模型均具有良好的在线监测和预测功能,为中药提取过程动态监测方法的优化选择提供了参考依据。The manufacture of traditional Chinese medicine（ TCM） products is always accompanied by processing complex raw materials and real-time monitoring of the manufacturing process. In this study,we investigated different modeling strategies for the extraction process of licorice. Near-infrared spectra associate with the extraction time was used to detemine the states of the extraction processes. Three modeling approaches,i. e.,principal component analysis（ PCA）,partial least squares regression（ PLSR） and parallel factor analysis-PLSR（ PARAFAC-PLSR）,were adopted for the prediction of the real-time status of the process. The overall results indicated that PCA,PLSR and PARAFAC-PLSR can effectively detect the errors in the extraction procedure and predict the process trajectories,which has important significance for the monitoring and controlling of the extraction processes.

关 键 词：近红外光谱 甘草 平行因子法 质量控制 过程分析 动态监测 

分 类 号：R284.2[医药卫生—中药学] O657.33[医药卫生—中医学]