| Tobacco was one of the floret, annual or perennial herb. The distribution range was wide from latitude 60 degrees to 45 degrees south of the vast majority of countries and regions. China is one of the largest countries in the world producing tobacco, and the annual production is more than 1000000 hectares, the annual output is about 2000000 tons. In our country, the tobacco is planted in most provinces and regions in China, and tobacco is an important economic crop, the tobacco industry’s tax is an important part of our country’s government revenue.Tobacco as an important industrial raw material, the famous tobacco expert Zhu Zunquan academicians believes that the availability to describe the value of tobacco is more appropriate. The availability of tobacco leaf includes 5 aspects, such as appearance factor, physical property, sensory evaluation, chemical composition and safety. The availability of tobacco is determined by the quality of tobacco leaves. The quality of tobacco leaf is a comprehensive concept which reflects the balance of the essential characteristics of tobacco. It is the appearance characteristic and the intrinsic characteristic of the tobacco leaf itself, the quality and the usability and the objective need of the tobacco leaf in the industry. Therefore, the study and evaluation of the tobacco leaf quality should be considered comprehensively.Due to the strong adaptability of tobacco to the ecological environment, it is widely planted in the world, but as the leaves of the plant, the appearance, structure, material transformation and accumulation of the leaves of the leaves of the plant have a direct impact on the quality. From the quality aspect, tobacco is very sensitive to the ecological environment changes. The ecological environment differences and cultivation not only affect the morphological and agronomic characters of tobacco plant, but also lead to the change of chemical composition of tobacco leaves, and then affect the quality of tobacco. In a long period, a lot of researches have been carried out on the relationship among ecological environment, variety, cultivation measures, modulation, etc., the contribution rate of ecological environment; variety and cultivation technology to tobacco leaf quality was 56%,32% and 10% respectively. Ecological environment is the most important factor to determine the quality of tobacco leaf, which is mainly reflected in the ecological environment and the content of aroma components of flue-cured tobacco, and the main characteristics of tobacco aroma and taste. Ecological factors, climatic factors impact tobacco flavor and sensory quality index and contribution rate maximum. A series of studies was carried out on relationship between climatic factors and the tobacco aroma substances, but most of them are limited in a particular area of single climatic factors and flue cured tobacco aroma substances. Common climatic factors include light, humidity, precipitation, temperature, etc.,In this thesis, the quality of tobacco leaves (including appearance quality, physical characteristics, sensory quality and chemical components) was studied. The effects of environmental factors on the quality of tobacco leaves were analyzed by means of the system analysis of the two stages of environmental factors (altitude, air temperature, humidity, rainfall and illumination) and storage environment (temperature, humidity and microbe).(1) Analysis of quality characteristics of tobacco leaves.Chemical:chemical char act eristic index and style from the chemical composition and sensory quality in different environmental conditions of tobacco types and quality characteristics were characterized by sensory evaluation style oriented, the selection of representative index system:quality characteristic index related to the representation of the chemical index of total sugar, total nitrogen, potassium, total polyphenols, Solanone, megastigmatrienone A, geranyl acetone, benzene alcohol, octanoic acid and style characteristic index related to the representative index of total sugar, total nitrogen, potassium, total polyphenols, megastigmatrienone D, took possession of aldehyde, geranyl acetone, benzene acetaldehyde, malonic acid, malic acid.The classification method on the quality of tobacco leaves was studied by means of representative chemical indicators, including:contour analysis, cluster analysis and discriminant analysis. With the contour analysis method, the characteristic contour map of the sample under different environmental conditions was obtained, which indicated the specificity of the chemical components in the sample. By different flavor characteristics of contour map, compared to the faint scent type tobacco and other flavor of tobacco, total sugar, total nitrogen, total polyphenols, Tibetan, citral, phenyl acetaldehyde and malic acid six representative chemical composition content is higher. Application of clustering analysis method, samples of tobacco leaf quality division, Fen, Luzhou flavor liquor and middle flavor tobacco samples is divided into three categories. At the same time, according to the distance between the samples of the tobacco leaf the differences in the style of the tobacco leaf can be compared. Stepwise discriminant analysis was used to establish discriminant function, the quality of tobacco leaf was classified. The discriminant function is used to verify the accuracy of the method, and the accuracy of the decision is good. At the same time the quality of different types of tobacco samples, better classification results in different flavor types of tobacco samples.(2) Study on the effect of planting environment on the quality of tobacco leaves.Correlation analysis, multiple comparison analysis, partial least squares discriminant analysis, regression analysis and neural network analysis, and other methods were used to select the indicators of environmental factors affecting the quality of tobacco leaf.Altitude impact factors index are sweet, nutty aroma, fragrance type, middle flavor, flavor type, style index, citric acid, palmitic acid, linoleic acid, isovaleric acid,2-methyl butyric acid, volatile acids, lutein, beta carotene, rutin, alcohols and acids for a total of 17 indicators. Through regression analysis, the above 17 indexes are affected by low, medium and high altitude, and the correct rate of cross validation is 77.8%,83% and 85% respectively.Annual average air temperature of main indexes are:concentration, a sweet, nutty aroma, sweet aroma, Radix Aucklandiae, honey sweet, fragrance, nicotine and chlorogenic acids and alcohols for a total of 12 indicators. The regression analysis shows that, by cross validated correct rates of the low, medium and high temperature influencing 12 indicators respectively are:80%,81.25% and 80%.The main indexes of the annual average humidity were:reducing sugar, total nitrogen, potassium, citric acid, palm acid,2-methyl butyrate, lutein, beta carotene, chlorogenic acid, total polyphenol and alcohol. The results showed that the 12 indexes were affected by low, medium and high humidity, and the correct rate of cross validation was 91.7%,92.9% and 83.3% respectively.By average annual rainfall impact indicators are:Radix Aucklandiae, nicotine, total sugar, potassium, malic acid, linoleic acid, non-volatile acids, benzoic acid, octanoic acid, lutein, beta carotene, scopoletin, rutin, total polyphenols, aldehydes and heterocyclic a total of 16 indicators, through regression analysis shows that, the correct rate of the influence of the 16 indicators by the low, medium and high rainfall, cross validation were:83.8%,86.6% and 83.3%, respectively.Annual average sunshine impact indicators are:aroma quantity, concentration, mixed gas, clean, moist, quality index, sweet smell, the smell of hay, nutty aroma, honey sweet, middle flavor and style index, valeric acid, caproic acid and benzoic acid, chlorogenic acid and rutin, a total of 17 indicators, through regression analysis shows that the 17 indexes, low, and high impact of sunshine, cross validation accuracy respectively:100%,83.3% and 83.3%.For the further study on joint influences of environmental factors (elevation, mean annual temperature, average annual rainfall, annual average humidity and annual average sunshine) to the quality index of tobacco, the neural network optimization analysis by environmental factors than the index, with BP neural network through iterative analysis, the Fen, Luzhou flavor and sweetness index prediction model is established, the three indicators of training set and the correlation coefficients were:98.16%,96.89%,85.52% and 79.29%, and the test set is better, for the prediction of the above three indexes. Through this chapter, the effect of environmental factors on the quality of tobacco leaves was studied. The effect of environmental factors on the physiological and biochemical, growth and development of tobacco leaves was studied.(3) Studies on the effects of molds and their effects on the quality of tobacco in the storage conditionFirst fungi of microorganism separation, purification and test of pathogenicity, screened in tobacco cause mildew fungi, and the microbial culture and morphological identification of judgment fungal species, in the research of tobacco leaf molds were mainly Aspergillus and Penicillium genus.The chemical composition of tobacco mildew was analyzed. The analysis on nutritional components in tobacco leaves, indicates that the mildew in tobacco samples of C source and N source and a mineral class chemical composition change differently. Sugar, starch, cellulose, wood elements, organic acids and polyphenols C source material content is reduced greatly, chloride ion, sulfate, nitrate and phosphate anion element content did not change basically; through tobacco mildew of volatile and semi volatile components analysis table Ming tobacco mildew new compounds with butanol,2-methyl butanol, isoamyl alcohol, cyclohexanol, linoleic acid acetate, flax acid ethyl ester, stearic acid ethyl ester. Changes of the content of these compounds with a large degree:palm acid ethyl ester, palm acid methyl ester,3-methyl butanol,4-methyl phenol,3-hydroxy-2-butanone,5 HMF, indole, flax acid methyl ester, benzene ethanol, furfural, benzyl alcohol and furfuryl alcohol. The 19 kinds of tobacco mildew characteristic volatile components were filtered out to build a model and normal samples discriminant model for the prediction of accurate rate was 95.0%, moldy samples of the prediction accuracy rate was 94.4%. The prediction accuracy can meet the practical application, and the application of the method can effectively determine the tobacco mildew.In the system analysis of tobacco mildew chemical composition changes and on the basis of mildewing degree decision method, the establishment and validation of both near infrared spectroscopy technique of tobacco mildew and quantitation prediction model of ergosterol was applied. In the range of 780 nm nm-2500, the NIR spectra were collected to obtain the basic spectral data of tobacco leaf samples. GBA algorithm is established, and the characteristics of the basic spectral data are selected, and the PLS-DA discriminant model is established. The discriminant model has a smaller Wilks’0.216 (P 0.001) and an error rate of 2.92%. The model was applied to the identification of tobacco leaf samples, and the accuracy rate was 95.79%. Especially, satisfactory recognition in stage I of tobacco leaf mildew can be achieved, which is applicable to the real-time monitoring of tobacco mildew prediction and storage of tobacco. In the situation of mildew tobacco or early stage of mildew,ventilation, cooling, dehumidification, double stack, threshing and redrying measures the moisture content reduction to prevent mildew further spread and spread, the maximum possible reduce losses. Near infrared spectroscopy (NIRS) of ergosterol content in tobacco leaves was predicted to provide a fast detection speed, simple and convenient operation, forecasting the accurate and rapid method for the determination and prediction for the tobacco leaf mildew has practical significance. |
PDF Full Text Request