陈 希
（内蒙古伊利实业集团股份有限公司，内蒙古 呼和浩特 010110）
摘要：在碳酸饮料产品的设计阶段，针对含气量在货架期内变化情况进行预测，以确保满足消费者的口感需求。在固定规格、包装材质的情况下，研究了货架期内碳酸饮料的容积倍数即含气量和初始充气量、储存温度、储存时间的关系，基于多元线性回归建立了预测模型并进行了验证。实验结果表明，货架期含气量与初始含气量呈高度正相关，与储存温度及储存时间呈中度负相关，模型调整R2 为0.915，回归方程的F 值为154.827，P<0.001，模型的拟合优度较好，因变量与自变量整体之间存在着显著的线性回归关系，且不存在多重共线性。随机选取自变量进行验证，货架期含气量的模型预测结果与实测结果的相对误差绝对值在10%以内，具有较高的准确性。
关键词：碳酸饮料；货架期；含气量；线性回归
Establishment of Predictive Model for Shelf-life CO2 Content of Carbonated Beverages Packagedwith PET Bottles
CHEN Xi
（Inner Mongolia Yili Industrial Group Co. Ltd.，Hohhot 010110，China）
Abstract: To ensure carbonated beverages can deliver satisfying taste through shelf life，it’s important to have thecapability to predict beverages’CO2 content in product design phase. This research investigated CO2 content duringshelf life vs. the initial CO2 content，storage temperature and storage length. The bottle size and package materialwere constant in the research. Multiple linear regression method was used to establish predictive model. Theexperiment results showed positive correlation between CO2 content during shelf life vs. the initial CO2 content whilenegative correlation between CO2 content during shelf life vs. storage temperature and storage length. The adjusted Rsquare of the model is 0.915，the F value of regression equation is 154.827 with a significant value <0.001，indicating a relatively high proportion of variance in the dependent variable accounted for by the entire regressionmodel and an overall good level of fitness. The validation results show the relative error of prediction is within 10%，which proved the model has good level of accuracy.
Key words: carbonated beverages；shelf life；CO2 content；linear regression
中图分类号：TS275.3 文献标志码：A 文章编号：1007-7871（2023）01-0070-04
2023，Vol.26，No.1