基于内源性致香物质和化学计量学的烟草感官评价研究

摘要

The principal component analysis method combined with genetic algorithm and neural network was used to establish the sensory evaluation model based on the endogenous aromatic components of tobacco.GC-MS method was used to qualitatively and quantitatively analyze the endogenous aromatic components of tobacco essential oil obtained by supercritical extraction and molecular distillation.Initially,principal component analysis(PCA) was used to analyse endogenous aromatic components.Scores of the five extracted principal components and sensory evaluation were then used as the input and output variables,respectively.Back-propagation (BP) neural network was uscd to establish the prediction model.Genetic algorithm (GA) was further applied to optimize the neural network weights and thresholds.The experimental results showed that the performance of GA-BP model was better than that of BP.The correlation coefficient between the predicted value by the GA-BP model and the experimental value was 0.96,and the root mean square error of prediction (RMSEP) was 1.81.The GA-BP model showed better fitting ability and prediction ability.The model could effectively predict the sensory quality of the essential oil.%采用主成分分析法结合遗传算法和神经网络,建立了基于烟草内源性致香物质的感官质量评价预测模型.利用气相色谱-质谱(GC-MS)技术对超临界萃取-分子蒸馏所得烟草精油中的内源性致香组分进行定性定量分析,汇总各类致香指标后,对其进行主成分分析;以提取所得5个主成分的得分作为输入变量,感官评吸分数作为输出变量,分别使用标准BP神经网络和遗传算法(GA)优化的BP神经网络建立预测模型.对比实验结果表明,GA优化后的模型预测效果更优,其预测值与实验值间的相关系数为0.96,预测均方根误差为1.81,说明GA-BP模型具有更好的拟合能力和预测能力,该模型能有效地预测烟草精油的感官品质.