基于R语言的养阴通脑颗粒总黄酮提取优化

Optimization extraction technique of total flavones in Yangyin Tongnao Granules based on R language

目的:运用R语言结合BP神经网络和遗传算法优化复方养阴通脑颗粒中总黄酮提取工艺。方法:在单因素试验的基础上采用4因素3水平的正交试验设计方法,提取并测得养阴通脑颗粒中总黄酮的含量。通过R语言,建立BP神经网络模型,再利用遗传算法对网络进行目标寻优,从而得到养阴通脑颗粒总黄酮的最佳提取工艺。结果:养阴通脑颗粒中总黄酮的最优工艺条件为乙醇浓度55%、提取时间1.5h、提取温度80℃、液料比15∶1,模型预测值为119.04mg,而实验平均值为120.35mg,相对误差为1.09%,因此具有良好的网络预测性。结论:此数学模型可用来对养阴通脑颗粒中总黄酮的提取工艺进行分析和预测,为实现中药有效部位和成分目标寻优提供了新的参考。

Objective： To optimize the extraction technology of total flavones in Yangyin Tongnao Granules by R language and BP neural network and genetic algorithm. Methods： Based on the single factor experiment, four factors and three levels of orthogonal experiment design method was adopted to extract and test the content of total flavones in Yangyin Tongnao Granules. BP neural network model has established based on R language and optimization design in network was screened by using genetic algorithm, in order to obtain the optimization extraction technology of total flavones in Yangyin Tongnao Granules. Results： The optimization of extraction technology was as follows： ethanol concentration was 55%; extraction time was 1.5h; extraction temperature was 80℃; liquid-solid ratio was 15∶1. Under the condition, model predictive value was 119.04 mg, the experimental average value was 120.35 mg, and the relative error was 1.09%. So it had a good network prediction. Conclusion： This mathematical model could be used to analyze and predict the extraction technology of total flavones in Yangyin Tongnao Granules and provide a new reference for screening optimization of Chinese medicine effective parts and components.