黄酒挥发性风味香气成分保留时间的神经网络模型  

A Neural Network Model of the Retention Time of Volatile Flavor and Aroma Constituent in Yellow Rice Wine

作　　者：堵锡华[1] 徐艳[1] 田林[1] 李靖[1] 陈艳[1] DU Xihua;XU Yan;TIAN Lin;LI Jing;CHEN Yan(School of Materials and Chemical Engineering,Xuzhou Institute of Technology,Xuzhou 221018,China)

机构地区：[1]徐州工程学院材料与化学工程学院,江苏徐州221018

出　　处：《徐州工程学院学报（自然科学版）》2023年第3期78-85,共8页Journal of Xuzhou Institute of Technology(Natural Sciences Edition)

基　　金：国家自然科学基金项目(21472071);江苏省自然科学基金项目(BK20171168);江苏省高校自然科学基金重大项目(18KJA430015)。

摘　　要：黄酒含有多种活性成分,营养丰富,具有降低胆固醇、血糖、血压和防癌等较好的保健作用.对延安糜子黄酒和绍兴糯米黄酒挥发性风味香气成分的色谱保留时间,构建其神经网络方法的定量结构-保留关系预测模型,计算了延安糜子黄酒和绍兴糯米黄酒两种酒的挥发性风味香气成分的分子连接性指数(m X),选取其中的0X、2X、5Xc 3种结构参数,作为反向传播神经网络方法的输入节点层变量,延安糜子黄酒挥发性风味香气成分的气相色谱保留时间作为输出节点层变量,采用3-6-1的网络结构,构建的神经网络模型的总相关系数r达到0.9896,根据该预测模型进行计算得到的色谱保留时间(tr)的预测值与文献所列实验值两者的平均误差为1.132;根据相同方法选取分子连接性指数中的1X、3X、5X、4Xpc 4种参数,将这4种指数与绍兴糯米黄酒挥发性风味香气成分的色谱保留时间(tr)进行相关性研究,采用了4-3-1的神经网络网络结构,得到的预测模型总相关系数r达到0.9862,色谱保留时间(tr)的预测值与文献的实验值两者的平均误差为1.413.结果表明,碳原子数量及基团的连接方式是影响黄酒挥发性风味香气成分色谱保留时间的主要因素.Yellow rice wine contains various active ingredients with rich nutrients,and has good health effects such as lowering cholesterol,blood sugar,blood pressure,and cancer prevention.In order to establish neural network model of quantitative structure-retention relationship(QSRR)for the chromatographic retention time of volatile flavor and aroma constituent of Yan'an rice wine and Shaoxing glutinous rice wine,the molecular connectivity indices(mX)of volatile flavor and aroma constituent of the two yellow rice wine were calculated.Three configuration parameters,0X,2X,5X_(c),were selected and used as input layer variables and the chromatographic retention time of Yan'an rice wine was used as the output layer variable.The 3-6-1 network structure was adopted and BP neural network method was used to establish a neural network model.The total correlation coefficient r was 0.9896.The mean error between the predicted values and the experimental values of retention time was 1.132.Using the same method,four configuration parameters of molecular connectivity indices 1X,3X,5X,4Xpc,were selected.The 4-3-1 network structure was adopted and BP neural network method was used to establish a model to study chromatographic retention time of volatile flavor and aroma constituent of Shaoxing glutinous rice wine.The total correlation coefficient r was 0.9862.The mean error between the predicted values and the experimental values of retention time was 1.413.The results showed that the amount and connection mode of carbon atom and groups were the main factors that influenced the chromatographic retention time of volatile flavor and aroma constituent in Yan'an rice wine and Shaoxing glutinous rice wine.

关 键 词：黄酒 挥发性风味香气成分 神经网络 定量结构-保留相关 

分 类 号：O604[理学—化学] TS262[轻工技术与工程—发酵工程]