论文部分内容阅读
目的:应用最小二乘支持向量机(LS-SVM)技术优化秦皮的提取工艺。方法:以秦皮甲素和秦皮乙素为指标,采用均匀设计安排提取试验,并用最小二乘支持向量机建立关系模型。结果:最小二乘支持向量机对秦皮甲素和秦皮乙素的拟和相关系数分别为0.999 7与0.999 9;得到的最优工艺条件为提取温度100℃、乙醇浓度50%、液固比11、提取时间70min,机器模型在此条件下的预测值为秦皮甲素提取量为9.291mg·g~(-1),秦皮乙素提取量为2.241mg·g~(-1),和实际测量值的相对误差仅为～2.97%和2.66%,具有较好的预测性。结论:最小二乘支持向量机可用优化秦皮提取工艺。
Objective: To optimize the extraction process of Cortex Fraxinus L. using Least Square Support Vector Machine (LS-SVM) technology. Methods: Using aesculin and apienin as indicators, the experiments were uniformly designed and arranged, and the relational model was established using least squares support vector machine. RESULTS: The quasi-coefficient correlation coefficients of aesculin and aesculin in least squares support vector machines were 0.999 7 and 0.999 9 respectively. The optimum conditions were as follows: extraction temperature 100°C, ethanol concentration 50%, and liquid-solid ratio 11 The extraction time was 70min. The predicted value of the machine model under this condition was 9.291 mg·g -1 for cortex A, 2.241 mg·g -1 for cortex A, and actual measurement. The relative error of the value is only ~2.97% and 2.66%, which has better predictability. Conclusion: Least squares support vector machine can be used to optimize the extraction process of cortex Fraxinus.