BP神经网络算法多指标优化酸枣仁汤提取工艺OA北大核心CSTPCD

In order to optimize the extraction process of Suanzaoren decoction,the extraction time,extraction times and solid-liquid ratio were taken as the investigation factors,and the extraction rate of total flavonoids,total saponins,total phenols and polysaccharides were taken as the evaluation index,and the entropy weight method was used for comprehensive evaluation.On the basis of single factor experiment,Box-Behnken response surface design and BP neural network were used to optimize the extraction process of the chemical components of Suanzaoren decoction,and the process verification was carried out.The results show that the comprehensive score of the optimal extraction process predicted by BP neural network algorithm is 149.11,which is better than 137.16 of Box-Behnken response surface.The optimal extraction process verified by BP neural network is determined as follows:extraction time 80 min,extraction times 2 times,solid-liquid ratio 1∶7 g/mL.The extraction rates of total flavonoids,total saponins,total phenols and polysaccharides are(10.73±0.63)mg/g,(73.34±1.77)mg/g,(16.73±0.56)mg/g,(413.08±8.34)mg/g,respectively.The study provides a basis for the optimization of the extraction process of Suanzaoren Decoction.