论文部分内容阅读
目的采用近红外光谱技术,结合不同变量筛选方法对金银花提取过程中绿原酸量进行快速测定。方法采用组合间隔偏最小二乘法(SIPLS)、竞争自适应抽样方法(CARS)、变量投影重要性(VIP)、连续投影算法(SPA)4种不同变量筛选方法,以HPLC测定值作参比,建立金银花中绿原酸定量模型并进行比较,优选出最佳变量筛选方法。结果经SIPLS方法所建绿原酸模型预测能力最好,预测集决定系数(Rpre2)和预测均方根误差(RMSEP)分别为0.990 3和2.316%。结论近红外光谱法结合SIPLS变量筛选方法建立的绿原酸定量模型性能良好,满足中药提取过程实时监测分析的精度要求,可用于中药提取过程的快速分析。
Objective To determine the amount of chlorogenic acid during the extraction of honeysuckle by near-infrared spectroscopy and different screening methods. Methods The methods of SIPLS, CARS, VIP and continuous SPA (SPA) were used in this study. The results of HPLC were compared with those of other methods. The establishment of a quantitative model of chlorogenic acid in honeysuckle and compared to optimize the screening method for optimal variables. Results The chlorogenic acid model built by SIPLS method had the best prediction ability. The predicted set of prediction coefficient (Rpre2) and predicted root mean square error (RMSEP) were 0.990 3 and 2.316% respectively. Conclusion The quantitative model of chlorogenic acid established by near-infrared spectroscopy combined with SIPLS variable screening method has good performance and meets the requirement of real-time monitoring and analysis of the extraction process of traditional Chinese medicine. It can be used for rapid analysis of the extraction process of traditional Chinese medicine.