基于鲜烟叶理化特征的烤后烟叶品质预测

Quality prediction for flue-cured tobacco leaves based on their physicochemical properties from freshly harvested state

为明确鲜烟叶理化特征与烘烤后烟叶品质的关系,采用大田不同营养状况耦合采收成熟度试验,分析鲜烟叶的理化特征差异及其与烘烤后烟叶外观品质和主要化学成分的相关性,并建立基于鲜烟叶理化指标的烤后烟叶化学成分回归预测模型。结果表明:①不同成熟度鲜烟叶的理化特征差异明显且中部和上部规律基本一致,欠熟鲜烟叶呈现高氮、高含水率和低淀粉含量(质量分数)的特点,尚熟和成熟鲜烟叶理化特征差异不大。②在研究范围内随鲜烟叶的总氮和烟碱含量的提高,烘烤损失率提高,烤后烟叶的油润感减弱;鲜烟叶的淀粉含量越高,烘烤损失率越低,烤后烟叶的油润感越强。③基于鲜烟叶总氮含量建立的烤后烟叶还原糖和总氮含量回归预测模型的决定系数分别为0.629和0.579,基于鲜烟叶淀粉含量建立的烤后烟叶还原糖含量回归预测模型的决定系数为0.632。④利用鲜烟叶的总氮含量可实现烤后烟叶总氮含量的准确预测,平均相对误差小于7%;基于鲜烟叶的淀粉和总氮含量,可建立烤后烟叶还原糖和糖碱比值的预测模型,平均相对误差分别小于14%和18%。鲜烟叶的理化特征与烤后烟叶品质密切相关,利用鲜烟叶的总氮、淀粉含量可实现烤后烟叶总氮、还原糖和糖碱比值的预测。

To clarify the relationships between physicochemical properties of freshly harvested tobacco leaves and the quality of flue-cured tobacco leaves,the harvest maturity test coupled with different nutrient conditions was conducted in the field.The differences in the physicochemical properties of fresh leaves and their correlations with appearance and major chemical compositions of the cured leaves were analyzed.Regression prediction models of chemical compositions in the flue-cured leaves were established based on physicochemical indexes of the fresh leaves.The results showed that:1)The physicochemical properties of the fresh leaves with different maturity degrees were different,and the rules of middle and upper leaves were basically the same.The nitrogen and moisture contents(mass fractions)in unripe fresh leaves were relatively higher,while their starch contents were lower.There was little difference in the physicochemical properties between mature and ripe fresh leaves.2)Within the scope of this study,the curing loss rate of the fresh leaves increased and the oiliness of the flue-cured leaves decreased with the increase of total nitrogen and nicotine contents in the fresh leaves,while the curing loss rate of the fresh leaves decreased and the oiliness of the flue-cured leaves increased with the increase of starch content in the fresh leaves.3)The coefficients of determination of the established regression prediction model based on total nitrogen content in the fresh leaves for reducing sugar and total nitrogen contents in the flue-cured leaves were 0.629 and 0.579,respectively;and that of the regression prediction model based on starch content in the fresh leaves for reducing sugar content in the flue-cured leaves was 0.632.4)The total nitrogen content in the fresh leaves could be used to accurately predict the total nitrogen content in the flue-cured leaves with the average relative error less than 7%.Based on the starch and total nitrogen contents in the fresh leaves,the prediction model for reducing sugar content and sugar-nicotine ratio of the flue-cured leaves could be established with the average relative errors less than 14%and 18%,respectively.The physicochemical properties of the fresh tobacco leaves were closely related to the quality of the flue-cured tobacco leaves,the total nitrogen and starch contents in the fresh leaves can be used to predict the total nitrogen and reducing sugar contents and sugar-nicotine ratio of the flue-cured leaves.