【作者】 刘冰； 陈晓辉； 史新元； 孙立新； 乔延江； 毕开顺；

【Author】 LIU Bing~1,CHEN Xiao-hui~1,SHI Xin-yuan~2,SUN Li-xin~1,QIAO Yan-jiang~2,BI Kai-shun~(1*) (1.Shen yang pharmaceutical University,Shen yang 110016,China;2.School of Chinese Pharmacy,Beijing University of Chinese Medicine,Beijing 100102,China)

【机构】 沈阳药科大学； 北京中医药大学中药学院；

【摘要】 目的:利用近红外光谱分析技术,建立乳块消糖衣片包厚度快速、无损的测定方法。方法:采用Kemard-stone法对训练集样本和预测集样本进行分类,比较不同的建模方法,并对光谱的预处理方法、建模波段、主因子数的选择进行了详细的讨论。结果:采用偏最小二乘回归方法(PLSR)建立包衣厚度的近红外光谱定量分析校正模型。所建模型的相关系数R=0.9961,校正误差均方根(RMSEC)为0.0149,预测误差均方根(RMSEP)为0.0194。结论:预测结果表明,本文所建方法快速、无损、可靠,可推广应用于中药生产包衣过程的在线检测。更多还原

【Abstract】 OBJECTIVE:Near Infrared Spectroscopy analytical technique is used for establishing a rapid and nondestructive method to determine coating thickness of Ruikuaixiao tablets.METHODS:Training and testing sets are partitioned by Kernard-stone algorithm.Different calibration methods are compared,spectral pretreating methods,wavenumber selection and factors of NIRS are discussed in detail.RESULTS:Partial least squares regress(PLSR) was used for building the quantitive calibration model.Correlation coefficients(R),root mean square error of calibration(RM SEC),and root mean square error of prediction(RMSEP) obtained by PLSR model were 0.9961,0.0149,and 0.0194 respectively.CONCLUSION:The predictive result shows that the method of this study is rapid,non-destructive and credible,which could be applied to the on-line monitor of Chinese medicine tablets’coating process.更多还原