温度和还原糖对红葡萄酒总酚含量的影响及预测模型构建与验证  

Influence of Temperature and Reducing Sugar on Total Phenolic Content of Red Wine and Construction and Validation of Prediction Model

作　　者：李旭辉 赫润天 张溥天 廖祖宋 张军翔[4] 宋育阳[1,5] LI Xuhui;HE Runtian;ZHANG Putian;LIAO Zusong;ZHANG Junxiang;SONG Yuyang(College of Enology,Northwest A&F University,Yangling 712100,China;College of Mechanical and Electronic Engineering,Northwest A&F University,Yangling 712100,China;Xige Estate,Wuzhong 751100,China;School of Enology and Horticulture,Ningxia University,Yinchuan 750021,China;Ningxia Helan Mountain Eastern Foot Grape Wine Experimental Demonstration Station of Northwest A&F University,Yongning 750104,China)

机构地区：[1]西北农林科技大学葡萄酒学院,陕西杨凌712100 [2]西北农林科技大学机械与电子工程学院,陕西杨凌712100 [3]宁夏西鸽酒庄有限公司,宁夏吴忠751100 [4]宁夏大学葡萄酒与园艺学院,宁夏银川750021 [5]西北农林科技大学宁夏贺兰山东麓葡萄酒试验示范站,宁夏永宁750104

出　　处：《中外葡萄与葡萄酒》2024年第6期73-79,共7页Sino-Overseas Grapevine & Wine

基　　金：宁夏重点研发项目(2023BCF01002);西安市重点产业链项目(23NYGG0011);陕西省重点产业链项目(2023-ZDLNY-31);国家现代农业产业技术体系(CARS-29-jg-03);陕西省种业项目(K3031223092)。

摘　　要：发酵温度和还原糖含量等因素影响酚类物质在发酵过程中的浸出及稳定性。本研究旨在探讨温度及还原糖对总酚浸出及稳定性的影响并建立预测模型。采用10 L小规格(训练集)和5 t大规格(验证集)发酵试验相结合,设置不同发酵温度和还原糖含量,并检测葡萄酒的总酚含量。结果显示,适当升温有助于增加总酚含量,而温度过高会影响酚类物质的稳定性导致其含量下降。还原糖通过影响发酵温度和酒精生成等因素间接影响总酚含量。以小规格试验数据为基础,大规格试验数据为验证,基于决策树回归、随机森林回归、极端梯度提升模型算法构建总酚含量预测模型,研究结果表明基于极端梯度提升算法的模型表现最佳,R^(2)为0.88,RMSE为60.13,其对大规格发酵总酚预测误差为17.8%。本研究为红葡萄酒生产过程中总酚含量的预测和调控提供了一种新的思路。Factors such as fermentation temperature and reducing sugar content affect the leaching and stability of phenolic substances during the fermentation process.This study aims to investigate the effects of temperature and reducing sugar content on the extraction and stability of total phenols and to establish a predictive model.The experiment combined 10-liter small-scale(training set)and 5-ton large-scale(validation set)fermentation trials with different fermentation temperatures and reducing sugar contents,measuring the total phenol content in the wine.Results indicated that moderate temperature increases contribute to higher total phenol content,while excessively high temperatures can decrease phenolic compound stability,leading to reduced content.Reducing sugar content indirectly affected total phenol content by influencing fermentation temperature and alcohol production.Based on small-scale test datas and validated by large-scale test datas,the total phenol content prediction model was constructed based on decision tree regression,random forest regression and extreme gradient improvement model.The results show that the model based on the extreme gradient improvement algorithm performed the best,R^(2) is 0.88,RMSE is 60.13,This prediction model has a prediction error of 17.8%for the total phenolic content of large bulk alcoholic fermentation.This study provides a new approach for predicting and controlling total phenol content in the production process of red wine.

关 键 词：红葡萄酒 温度 还原糖 总酚 预测模型 

分 类 号：TS262.61[轻工技术与工程—发酵工程]