论文部分内容阅读
以丹参醇提过程为研究载体,采用离线方式采集过程样品的近红外光谱,并以HPLC测定值为参考值,采用偏最小二乘法建立丹参酮ⅡA的定量校正模型,对丹参提取过程进行监测。由于原药材批次间的质量差异性,会危及原有模型稳健性,基于此,该文首次将简单区间计算法(simple interval calculation,SIC)应用于中药提取过程近红外定量模型更新,且与random selection(RS)更新方法进行比较。基于SIC算法的更新结果:RMSECV,RMSEP,RPD分别为0.006 8 g·L-1,0.005 4 g·L-1,3.14;基于RS方法的更新结果:RMSECV,RMSEP,RPD分别为0.006 4 g·L-1,0.006 8 g·L-1,2.50。结果表明SIC比RS更优,为今后中药提取过程的质量控制与监测提供了研究基础。
Taking the extraction process of Salvia miltiorrhiza as the research carrier, the near-infrared spectrum of the sample was collected by off-line method. The quantitative analysis model of tanshinone ⅡA was established by partial least squares method and the extraction process of Salvia miltiorrhiza was monitored. Due to the quality difference between the original lot of medicinal materials, the robustness of the original model would be endangered. Based on this, the article first applied the simple interval calculation (SIC) to update the near-infrared quantitative model of traditional Chinese medicine extraction process, random selection (RS) update method to compare. Based on the SIC algorithm, the results of RMSECV, RMSEP and RPD were 0.006 8 g · L-1 and 1.005 4 g · L-1, 3.14 respectively. The updated results based on the RS method: RMSECV, RMSEP and RPD were 0.006 4 g · L-1,0.006 8 g · L-1,2.50. The results showed that SIC was better than RS, which provided a research basis for the quality control and monitoring of the extraction process of Chinese medicine in the future.