| 摘要: |
| 目的 运用正交试验和人工神经网络(BP-ANN)模型相结合的方法对通痹凝胶贴剂提取工艺进行优化。方法 以总生物碱含量、龙胆苦苷含量、总固体得率为指标,运用层次分析法(AHP法)结合CRITIC法确定3个指标的权重系数;采用人工神经网络模型对正交试验得到的提取工艺进行仿真优化,最终确定贴剂最佳提取工艺。结果 总生物碱、龙胆苦苷、总固体得率的综合权重分别为0.526、0.344、0.130;正交试验优选通痹凝胶贴剂提取工艺为药材加12倍量60%乙醇提取2次,每次2 h;人工神经网络模型仿真优化的提取工艺为药材加10.8倍量67%乙醇提取2次,每次1.7 h。结论 正交试验法与人工神经网络模型优化的提取工艺具有提取效率高、能耗低的优势,可为中药提取工艺的研究提供新的方法。 |
| 关键词: 通痹凝胶贴剂 正交试验法 反向传播人工神经网络 |
| DOI:10.13506/j.cnki.jpr.2019.04.004 |
| 基金项目:广东省省级科技计划项目(No.2014A020212463) |
|
|
|
| Extraction process optimization for Tongbi Gel-Patch by orthogonal test and back propagation artificial neural network |
| MA Jianchun1,2,LI Yong3 |
| (1.Department of Pharmacy,The First Affiliated Hospital of Guangdong Pharmaceutical University,Guangzhou 510080,China;2.Department of Clinical Pharmacy,Guangdong Pharmaceutical University,Guangzhou 510006,China;3.Insititute of Materia Medica,Guangdong Provincial Key Laboratory of Advanced Drug Delivery,Guangdong Pharmaceutical University,Guangzhou 510006,China) |
| Abstract: |
| Objective To optimize the extraction process of Tongbi Gel-Patch by Orthogonal Test and Back Propagation Artificial Neural Network(BP-ANN).Methods With the content of Gentiopicroside ,Aconitine and total solid yield as indexes,the weight coefficient of three of which were determined by AHP and CRITIC method. the optimum extraction process was optimized by orthogonal test and BP-ANN.Results The weight coefficient of three indexes was 0.526, 0.344 and 0.130.The extraction process was optimized by orthogonal test as following: Tongbi Gel-Patch was added with 12 times 60% ethanol,decocted for 120 min for 2times Optimum extraction process of BP-ANN model was as following: Tongbi Gel-Patch was added with 10.8 times 67% ethanol,decocted for 2 times 102 min each time.Conclusion BP ANN modle with orthogonal test can be used to optimize the extraction process of Tongbi Gel-Patch with short time, low cost and low energy consuming. |
| Key words: Tongbi Gel-Patch Orthogonal test BP-ANN |