[关键词]
[摘要]
目的 通过多指标组合赋权的正交试验综合评分和反向传播(back propagation,BP)神经网络预测综合得分对比优化筛选刷格痛风颗粒的水提工艺参数。方法 采用正交试验设计,以料液比、提取时间、提取次数为考察因素,没食子酸、盐酸黄柏碱、绿原酸、咖啡酸、紫花前胡苷、盐酸小檗碱、芦荟大黄素、大黄酸含量,干浸膏得率和指纹图谱相似度为评价指标,采用G1-熵权法进行权重赋值计算综合评分。对比正交试验结果和BP神经网络建模仿真优化结果,寻找刷格痛风颗粒的最佳水提工艺参数。结果 基于2种分析方法对比发现,正交试验分析得到最佳提取工艺的综合评分低于BP神经网络建模结果,故最终确定刷格痛风颗粒的最佳提取工艺参数为料液比1:12,提取2次,每次3 h。结论 优选出的刷格痛风颗粒最佳水提工艺稳定可行,正交试验结合BP神经网络进行方法寻优高效可靠,为复方制剂的开发和现代化研究提供了新的思路和参考。
[Key word]
[Abstract]
Objective The water extraction technology parameters of Shuage Tongfeng Granules (刷格痛风颗粒, STG) were optimized by multi-index comprehensive score of orthogonal test and BP neural network. Methods Orthogonal test design was adopted using the solid-liquid ratio, extraction time and extraction times as investigation factors, and the contents of gallic acid, phellodendrine chloride, chlorogenic acid, caffeic acid, nodakenin, berberine hydrochloride, aloe-emodin, rhein, dried extraction yield and fingerprint similarity were used as evaluation indexes. And the comprehensive scores were obtained by G1-entropy weight method. The optimal water extraction parameters of STG were obtained by comparing orthogonal test analysis and BP neural network modeling simulation optimization. Results Based on the comparison of the two analytical methods, it was found that the comprehensive score of the optimal water extraction parameters obtained by orthogonal test analysis was slightly lower than that obtained by BP neural network modeling. The optimal water extraction parameters were determined as follows:the solid-liquid ratio was 1:12, and extracted twice for 3 h each time. Conclusion The optimized water extraction parameters of STG is stable and feasible, and the orthogonal test combined with BP neural network is efficient and reliable, which provides new ideas and references for the development and modernization of compound preparation of Chinese medicine.
[中图分类号]
R283.6
[基金项目]
国家重点研发计划课题(2019YFC1712503);四川省科技厅区域创新合作项目(2020YFQ0007);西南民族大学研究生创新型科研项目(CX2021SZ85)