【摘要】 采用HPLC为对照分析法测定消渴丸中格列本脲的含量,应用偏最小二乘法(PLS)建立NIR光谱与格列本脲成分HPLC测定值之间的多元校正模型,以校正集的交互验证均方根误差(RMSECV)及相关系数(R2)为优化参数,选择最佳预处理方法和波长范围及最佳主成分数,并对未知样本进行含量预测.校正集最佳主成分数为5,R2为0.9986,RMSECV为0.0359.对预测集样品格列本脲的预测均方根误差(RMSEP)为0.00397.该方法预测精度高,具有方便快捷的优点,可作为消渴丸质量检测和在线质量监控的方法予以推广.更多还原

【Abstract】 The high performance liquid chromatography was used as the reference method to determine the content of glibenclamide in Xiaoke pills,and the multivariate calibration model based on PLS algorithm was developed to correlate the spectra and the corresponding values were determined by the reference method.The pretreatment method,wavelength range and optimal component were optimized according to root mean square error of cross-validation(RMSECV) and the corelation coefficient(R2) in calibration set.The principal component,R2 and RMSECV in calibration set were 5,0.998 6 and 0.035 9,respectively.The root mean square error of prediction(RMSEP) in prediction set was 0.003 97.This new method is so fast,accurate and easy-operating,that it could be generalized as an instrument of online quality control of the Xiaoke pills production.更多还原