BP神经网络结合多指标综合评分法优选通脉复方提取工艺
|
摘要
采用BP神经网络结合多指标综合评分法优选通脉复方提取工艺,为该工艺工业化生产提供参考。以黄芪甲苷,毛蕊异黄酮葡萄糖苷、马钱苷、毛蕊花糖苷、丹酚酸B、羟基红花黄色素A、总多糖含量及干膏率为综合评价指标,按照L_9(3~4)正交表安排进行试验,考察提取次数、提取时间、加水量对综合评分的影响,以正交试验数据为基础,采用BP人工神经网络优化通脉复方提取工艺。结果表明最佳提取工艺为11倍水,提取4次,每次40 min;马钱苷,毛蕊异黄酮葡萄糖苷、羟基红花黄色素A、毛蕊花糖苷、丹酚酸B、黄芪甲苷提取量平均值分别为6.9754,0.3315,14.8921,0.2956,55.2634,0.8647 mg·g~(-1);总多糖得率平均值为615 mg·g~(-1),干膏率平均值44.21%。BP人工神经网络结合综合评分可用于优选通脉复方提取工艺,无需多次试验便可为优化条件提供参考。
To optimize the extraction process of Tongmai Formula by back propagation(BP) neural network and multi-index evaluation.Orthogonal test L_9(3~4) was conducted for the experiment of extraction process.Loganin,calycosin-7-O-beta-D-glucoside,hydroxysafflor yellow A,acteoside,salvianolic acid B,astragaloside IV,polysaccharide and extractum were selected as marker components to investigate the effect of extraction time,extraction times and distilled water on multi-index comprehensive weighted score.Based on the data from orthogonal test,take back propagation(BP) neural network to optimize extraction process of Tongmai Formula.The result showed that the optimal extraction process was as follows:Tongmai Formula was added with 10 times distilled water,decocted for 40 min for 4 times.Extraction volume on average was 6.9754 mg·g~(-1)for loganin,0.3315 mg·g~(-1)for calycosin-7-O-beta-D-glucoside,14.8921 mg·g~(-1)for hydroxysafflor yellow A,0.2956 mg·g~(-1)for acteoside,55.2634 mg · g~(-1)for salvianolic acid B,0.8647 mg · g~(-1)for astragaloside IV,615 mg · g~(-1)for polysaccharide,44.21% for extractum,respectively.BP neural network with orthogonal test can be used to optimize the extraction process of Tongmai Formula,providing the reference for optimization without more experiment times.
To optimize the extraction process of Tongmai Formula by back propagation(BP) neural network and multi-index evaluation.Orthogonal test L_9(3~4) was conducted for the experiment of extraction process.Loganin,calycosin-7-O-beta-D-glucoside,hydroxysafflor yellow A,acteoside,salvianolic acid B,astragaloside IV,polysaccharide and extractum were selected as marker components to investigate the effect of extraction time,extraction times and distilled water on multi-index comprehensive weighted score.Based on the data from orthogonal test,take back propagation(BP) neural network to optimize extraction process of Tongmai Formula.The result showed that the optimal extraction process was as follows:Tongmai Formula was added with 10 times distilled water,decocted for 40 min for 4 times.Extraction volume on average was 6.9754 mg·g~(-1)for loganin,0.3315 mg·g~(-1)for calycosin-7-O-beta-D-glucoside,14.8921 mg·g~(-1)for hydroxysafflor yellow A,0.2956 mg·g~(-1)for acteoside,55.2634 mg · g~(-1)for salvianolic acid B,0.8647 mg · g~(-1)for astragaloside IV,615 mg · g~(-1)for polysaccharide,44.21% for extractum,respectively.BP neural network with orthogonal test can be used to optimize the extraction process of Tongmai Formula,providing the reference for optimization without more experiment times.
引文
[1]国家药典委员会.中华人民共和国药典:一部[S].北京:中国医药科技出版社,2015.
[2]单丽芳,杨红梅,曹蕾,等.多指标综合评分法优选参芪复方颗粒提取工艺及其热稳定性考察[J].中国实验方剂学杂志,2016,22(7):24-27.
[3]杨红梅,曹蕾,单丽芳,等.多指标正交试验优选双黄二仙颗粒的提取工艺[J].中国实验方剂学杂志,2016,22(7):28-30.
[4]张杰,李春艳,李劲平,等.蒽酮硫酸法与苯酚硫酸法测定竹节参多糖含量的比较研究[J].中南药学,2012,10(6):421-424.
[5]陈恒文,张翠英,任伟光,等.多指标综合评分法优化宣痹安痛方的水提醇沉工艺[J].中国中药杂志,2016,41(1):70-74.
[6]Zheng H,Jiang L,Lou H,et al.Application of Artificial Neural Network(ANN)and Partial Least-Squares Regression(PLSR)to Predict the Changes of Anthocyanins,Ascorbic Acid,Total Phenols,Flavonoids,and Antioxidant Activity during Storage of Red Bayberry Juice Based on Fractal Analysis and Red,Green,and Blue(RGB)Intensity Values[J].Agric Food Chem,2011,59(2):592-600.
[7]张超,韩丽,杨秀梅,等.BP神经网络结合正交试验优化苦参方中荆芥挥发油的提取工艺[J].中成药,2015,37(1):70-74.
[8]赵宏,谢晓玲,万金志,等.山药的化学成分及药理研究进展[J].今日药学,2009,19(03):49-52.
[9]吴梅,谭睿.黄芪多糖研究进展[J].川北医学院学报,2013,28(1):17-22.
[10]张程荣,曹岗,张云,等.山茱萸的化学、药理与炮制研究进展[J].中华中医药学刊,2011,29(9):2002-2005.
[11]王志江,魏国栋,马思缇.地黄多糖的化学和药理作用研究进展[J].中国实验方剂学杂志,2015,21(16):231-235.
[12]邵方元,付腾飞,薛兴阳,等.正交试验和BP人工神经网络模型优化胰蛋白酶提取鱼腥草多糖的工艺[J].中国实验方剂学杂志,2013,19(11):14-17.
[2]单丽芳,杨红梅,曹蕾,等.多指标综合评分法优选参芪复方颗粒提取工艺及其热稳定性考察[J].中国实验方剂学杂志,2016,22(7):24-27.
[3]杨红梅,曹蕾,单丽芳,等.多指标正交试验优选双黄二仙颗粒的提取工艺[J].中国实验方剂学杂志,2016,22(7):28-30.
[4]张杰,李春艳,李劲平,等.蒽酮硫酸法与苯酚硫酸法测定竹节参多糖含量的比较研究[J].中南药学,2012,10(6):421-424.
[5]陈恒文,张翠英,任伟光,等.多指标综合评分法优化宣痹安痛方的水提醇沉工艺[J].中国中药杂志,2016,41(1):70-74.
[6]Zheng H,Jiang L,Lou H,et al.Application of Artificial Neural Network(ANN)and Partial Least-Squares Regression(PLSR)to Predict the Changes of Anthocyanins,Ascorbic Acid,Total Phenols,Flavonoids,and Antioxidant Activity during Storage of Red Bayberry Juice Based on Fractal Analysis and Red,Green,and Blue(RGB)Intensity Values[J].Agric Food Chem,2011,59(2):592-600.
[7]张超,韩丽,杨秀梅,等.BP神经网络结合正交试验优化苦参方中荆芥挥发油的提取工艺[J].中成药,2015,37(1):70-74.
[8]赵宏,谢晓玲,万金志,等.山药的化学成分及药理研究进展[J].今日药学,2009,19(03):49-52.
[9]吴梅,谭睿.黄芪多糖研究进展[J].川北医学院学报,2013,28(1):17-22.
[10]张程荣,曹岗,张云,等.山茱萸的化学、药理与炮制研究进展[J].中华中医药学刊,2011,29(9):2002-2005.
[11]王志江,魏国栋,马思缇.地黄多糖的化学和药理作用研究进展[J].中国实验方剂学杂志,2015,21(16):231-235.
[12]邵方元,付腾飞,薛兴阳,等.正交试验和BP人工神经网络模型优化胰蛋白酶提取鱼腥草多糖的工艺[J].中国实验方剂学杂志,2013,19(11):14-17.