BP神经网络在沙姜总黄酮提取中的应用  

Application of BP neural network on extraction of flavonoids from kacmpferia galanga

作　　者：吴淦洲[1] 张玲[1] 王伟城[1] 

机构地区：[1]广东石油化工学院理学院,广东茂名525000

出　　处：《计算机与应用化学》2013年第4期435-438,共4页Computers and Applied Chemistry

摘　　要：为获得沙姜总黄酮提取的优化工艺,以乙醇体积分数、溶剂用量、提取温度和提取时间作为影响因素,进行了沙姜总黄酮提取的单因素试验和正交试验。以获得的25个单因素试验结果和9个正交试验结果作为训练样本,设计和训练了4-5-1结构的3层BP人工神经网络。再运用训练好的网络对设计的这些工艺组合进行预测,获得了2组总黄酮得率较高的提取工艺。结果显示,通过人工神经网络获得沙姜中总黄酮物质的最佳提取工艺为:以80%的乙醇溶液为溶剂,料液比为1:90,在75℃下提取10h,在此条件下得到的总黄酮得率为0.950mg/g,比正交试验最佳工艺条件下获得的总黄酮得率提高了2.4%。Single factor tests and orthogonal experiments were taken to acquire optimum extraction craft of flavonoids from Kaempferia galangal, and the alcohol concentration, solvent amount, extraction temperature and extraction time were considered to be influencing factors. The results of 25 single tests and 9 orthogonal experiments were used as training samples to design and train a 4-5-1 three-layers BP neural network. Then these designed extraction conditions was predicted by the terminative neural network, and 2 combinations of extraction conditions which could bring about high yield of flavonoids were acquired. The study results showed that in the condition of 80% ethanol as the solvent, the solid-liquid ratio of 1：90, at temperature 75 ℃and extracting 9 h. Under these conditions, the yield of flavonoids was 0.950 mg/g, obtaining about 2.4% more flavonoids than that of orthogonal experiments.

关 键 词：BP人工神经网络 沙姜 总黄酮 优化 

分 类 号：O29[理学—应用数学]