论文部分内容阅读
Complicated changes are happening during the aging process of flue-cured tobacco (FCT) and understanding of the changing components is of great significance in improving the quality,shortening aging time,and enhancing production and economic efficiency in tobacco industry.The aging samples of FCT from four main producing areas of China,i.e.,Yunnan,Sichuan,Hubei,and Dongbei,were used to study the changing rule of the components such as alkali,acid,and carbohydrate as well as the aroma and their relationship with sensory quality;and based on the correlation among the components and the sensory quality index,multivariable models were established to predict the aging quality.The results showed that the sensory quality of FCT increased steadily during the aging time,and at the same time,the acidic components (total organic acids,volatile acids) increased gradually,while the alkaline substances (nicotine,volatile alkali),carbohydrate (total sugars,reducing sugar) and pH values showed a downward trend.Correlation analysis found that the sensory quality and pH values were negatively correlated (P<0.05),while the sensory quality with total organic acids and aroma were positively correlated.The optimal model for predicting the quality of FCT was y=56.75-0.63x12+50.09x2-13.27x22,(y:sensory quality;x1:pH;x2:total organic acids).The average predicating error was 1.04% with maximum predicating error of 2.89% and the accuracy rate of above 97%.
Complicated changes are happening during the aging process of flue-cured tobacco (FCT) and understanding of the changing components is of great significance in improving the quality, shortening aging time, and enhancing production and economic efficiency in tobacco industry. Aging materials of FCT from four main producing areas of China, ie, Yunnan, Sichuan, Hubei, and Dongbei, were used to study the changing rule of the components such as alkali, acid, and carbohydrate as well as the aroma and their relationship with sensory quality; and based on the correlation among the components and the sensory quality index, multivariable models were established to predict the aging quality. The results showed that the sensory quality of FCT increased steadily during the aging time, and at the same time, the acidic components (total organic acids, volatile acids) increased gradually, while the alkaline substances (nicotine, volatile alkali), carbohydrate (total sugars, reducing sugar) and pH values showed a downw ard trend. Correlation analysis found that the sensory quality and pH values were negatively correlated (P <0.05), while the sensory quality with total organic acids and aroma were positively correlated. The optimal model for predicting the quality of FCT was y = 56.75- 0.63x12 + 50.09x2-13.27x22, (y: sensory quality; x1: pH; x2: total organic acids). The average predicative error was 1.04% with maximum predicative error of 2.89% and the accuracy rate of above 97%.