基于遗传算法的土香薷总黄酮提取工艺优化研究

摘要

Objective] To optimize the extraction process in Origanum vulgare Linn. [Methods] With ethanol concentration, solid-liquid ratio, extraction time and extraction times to examine factors, using the content of total flavonoids as the objective function, the orthogonal experiment was optimized the extraction process. SPSS18.0 software was used to establish its quadratic regression model from orthogonal test data. Using genetic algorithm to search the optimum process conditions and the optimization of extraction technology of the total flavones was carried out by GA. [Results]Its quadratic regression model was Y=51.039+2.164X3X4-0.007X1X2. The optimum extraction conditions were confirmed as fol owing: 61 times volume ethanol, 21% of the ethanol, to extract 3 times and 2h for each time. Under these conditions, the yield of total flavonoids was 54.37mg/g, little difference in predictive value (53.71mg/g). [Conclusion] The advantages of this process are efficiency, stability and reliability, which can be applied in the optimization of extraction process of total flavonoids in Origanum vulgare Linn.%[目的]确定土香薷总黄酮的最优提取工艺。[方法]以乙醇浓度、料液比、提取时间和提取次数为考察因素，以总黄酮含量为目标函数，运用SPSS18.0软件建立其正交试验数据的二次回归模型，并进一步采用遗传算法对其进行优化，确定土香薷总黄酮的最优工艺条件。[结果]建立总黄酮含量（Y）与各因素的二次回归模型为：Y=51.039+2.164X3X4-0.007X1X2；遗传算法优化后得其最优提取工艺条件为：加药材21倍量浓度为61%的乙醇提取3次，每次2h。；此条件下土香薷总黄酮的平均含量为54.37 mg/g，RSD为1.29%，与预测值（53.71mg/g）基本相近。[结论]所选最优提取条件提取率高、稳定性好，可用于土香薷总黄酮的提取。