Method for predicting feeding tobacco stem structure

Abstract

The application discloses a method for predicting tobacco stem structure, comprising the following steps: S1, measuring stem structure; S2, measuring tobacco stem containing stem ratio; S3, analyzing tobacco stem containing stem ratio and stem structure by using grey correlation method; S4, analyzing tobacco stem containing stem ratio and stem structure by using clustering analysis; S5, analyzing characteristic size of stem, tobacco stem and stem structure by using regression analysis; and S6, predicting tobacco stem structure. The application analyzes the correlation between blended stem structure and tobacco stem containing stem ratio, establishes a regression equation of tobacco stem containing stem ratio and flavored stem structure, predicts tobacco stem structure based on tobacco stem containing stem ratio, determines a stem stem ratio removing scheme suitable for tobacco stem structure, optimizes the structure of tobacco stem in a processing section, improves the applicability of stem size in the blending process, and improves the uniformity and stability of stem structure.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of tobacco stem, in particular to a feeding tobacco stem structure prediction method. BACKGROUND

[0002] As an important component of tobacco leaf, the weight of tobacco stem accounts for about one fourth of the weight of tobacco leaf. Because of the same chemical composition as tobacco leaf, tobacco stem can effectively increase the filling capacity and combustibility of tobacco after processing and blending into tobacco, so as to improve the utilization rate of tobacco raw materials. With the implementation of the measures of reducing and controlling tobacco tar in the tobacco industry and the continuous improvement of the technology level of tobacco processing, the blending ratio of stem is also increasing. The uniformity and stability of stem structure are directly related to the process quality and sensory quality of cigarettes.

[0003] At present, the research on stem structure is mainly focused on the different forms of stem, the structure of stem and the influence of stem on the quality of cigarettes. Some researches are also focused on the influence of stem processing technology on the form and quality of stem. Shen Ya and Qi Lin respectively studied the influence of stem on end drop and the difference in volatile chemical substances of stem of different lengths, which provides a reference for the influence of stem structure on the stability of cigarette quality. There is less research on the stability of tobacco stem and the influence of stem bend on the structure of stem.

[0004] At present, there are 20 brands of cigarettes produced in our factory, among which 14 brands use stem. The maximum blending ratio of some brands reaches 30%. The amount of stem produced each year reaches 2100 batches (2200 kg / batch). The blending ratio of stem is a key parameter in the design of cigarette formula and the production process of tobacco. In the production process, the forms of tobacco stem entering the cigarette production workshop are different. There is a certain proportion of long stem (length > 20 mm), short stem (length ≤ 20 mm), stem bend (stem head diameter generally > 6 mm) and broken stem (length generally < 10 mm, diameter generally < 2.5 mm). The main chemical component of stem bend is lignin, which is easy to absorb water and break in the production process. At the same time, blending into tobacco can affect the taste and combustibility of cigarette products. Therefore, many enterprises now screen out stem bend and broken stem before producing stem, so as to ensure the quality of stem and the consistency of the form of stem and tobacco.

[0005] In the production process of stem in our factory, only broken stem is screened out, and stem bend is still used for production. The form of stem tends to be sheet-shaped, and more than 70% of the length of stem is concentrated in 4.00-5.60 mm and 2.00-2.80 mm. Based on the above reasons, we propose a research on the correlation between the stem bend rate of tobacco stem and the structure of stem. SUMMARY

[0006] To solve the above problems, the present application provides a feeding tobacco stem structure prediction method.

[0007] The present application establishes a regression equation of tobacco stem containing stem bend rate and stem structure after flavoring by analyzing the correlation between blended stem structure and tobacco stem containing stem bend rate of raw material, determines the tobacco stem containing stem bend rate of raw material, improves the applicability of stem size in the blending process, optimizes the incoming material structure of tobacco stem processing section, determines the stem bend removal scheme suitable for the tobacco cut structure, and improves the uniformity and stability of stem structure.

[0008] The technical scheme adopted by the present application to solve the technical problems is:

[0009] The method for predicting the structure of raw material tobacco stem comprises

[0010] S1: measuring the stem structure

[0011] Randomly select 20 batches of general stems for testing, and test them according to the production batches;

[0012] After the stem flavoring process is stable, take the stem sample at the outlet of the stem flavoring machine, take 5 samples per batch, the sampling interval is 1 min, the sampling amount is 400 g, and the stem sample is divided into(100.0 g±10.0 g) screen samples by using the quartering method;

[0013] Use the screen instrument to screen according to the size of the screen mesh aperture, and divide the stem sample into S1: >8.00 mm, S2: 6.70-8.00 mm, S3: 5.60-6.70 mm, S4: 4.75-5.60 mm, S5: 4.00-4.75 mm, S6: 3.35-4.00 mm, S7: 2.80-3.35 mm, S8: 2.00-2.80 mm, S9: 1.40-2.00 mm, S 10 : 0.71-1.40 mm, and S 11 : <0.71 mm according to the size of the screen mesh aperture;

[0014] The running parameters of the screen instrument are: 4 min for each screening, the speed is 230 r / min, the direction is changed every 1 min, and then the weight of the stem on each layer screen is weighed;

[0015] The least square method is used to fit the stem size, calculate the characteristic size S 12 , and the uniformity coefficient S 13 :

[0016] Tobacco cut size distribution equation:

[0017] Solve a and p, and substitute them into the equation:

[0018] In the formula: F is the cumulative mass fraction of stems on the screen, %; a is the characteristic size of the stems, representing the stem size corresponding to a cumulative stem content of 50% on the upper layer of the screen; p is the uniformity coefficient of stem size distribution; q is the screen aperture, mm;

[0019] S2 tobacco stem inflection rate measurement

[0020] 1000g samples were randomly selected from the incoming tobacco stems. The tobacco stems and stems containing tobacco stems were picked out, weighed, and recorded. Each batch of samples was tested in triplicate. The calculation formula is as follows:

[0021]

[0022] Where: G is the stalk inflection rate of tobacco stem, %; g1 is the weight of tobacco stem and tobacco stem containing tobacco stem, g; g is the weight of the sample, g;

[0023] Grey relational analysis of stem inflection rate and stem structure in S3 tobacco stems

[0024] Using the stem inflection rate of tobacco stems as the parent sequence X and the stem fiber structure as the child sequence Y, calculate the grey relational degree between the two and sort them according to the magnitude of the relational degree:

[0025] γ(S 13 )>γ(S5)>γ(S 12 )>γ(S8)>γ(S4)>γ(S9)>γ(S 10 )>γ(S2)>γ(S7)>γ(S6)>γ(S1)>γ(S3)>γ(

[0026] S 11 );

[0027] S4 tobacco stem cluster analysis of stem inflection rate and stem structure

[0028] The gray correlation between stem inflection rate and stem structure was analyzed by clustering according to the strength of strong correlation (0.7<γ<1), medium correlation (0.4<γ<0.7), and weak correlation (γ<0.4). The correlation strength between different stem structure sizes and stem inflection rate was analyzed.

[0029] Wherein: size cluster C1 = {5.60~6.70mm, <0.71mm}, size cluster C2 = {6.70~8.00mm, 2.80~3.35mm, 3.35~4.00mm, >8.00mm}, size cluster C3 = {characteristic size, 4.00~4.75mm, uniformity coefficient, 2.00~2.80mm, 4.75~5.60mm, 1.40~2.00mm, 0.71~1.40mm};

[0030] Combining the clustering analysis result, it can be known that the size of 4.75-5.60 mm, 4.00-4.75 mm, 2.00-2.80 mm, 1.40-2.00 mm and 0.71-1.40 mm stem is strongly related to the stem content of tobacco stem;

[0031] Regression analysis of the characteristic size of stem and the structure of tobacco stem and stem

[0032] Taking the stem content of tobacco stem X1 as the dependent variable, the size of stem (4.75-5.60 mm, X5, 4.00-4.75 mm, X6, 2.00-2.80 mm, X9, 1.40-2.00 mm, X 10 , 0.71-1.40 mm, X 11 ), the uniformity coefficient of stem X 13 and the characteristic size of stem X 14 as the independent variables, stepwise regression analysis is carried out to obtain a multiple regression equation and a significance level;

[0033] The regression analysis equation is y(X1)=1.941X9+2.118X 10 -0.180X 13 +0.424X 14 -1.582; R 2 =0.851, P<0.01;

[0034] S6 prediction of the structure of feeding tobacco stem

[0035] A prediction model y(X1) of the stem content of tobacco stem is established by the regression analysis method to predict the structure of feeding tobacco stem.

[0036] When the stem content of feeding tobacco stem is known, the characteristic size of stem is calculated, and then the stem content removal scheme suitable for the structure of tobacco is determined by combining the production process control parameters of stem, such as stem content removal, stem sign screening and the like to improve the characteristic size of stem.

[0037] The beneficial effects brought by the present application are as follows:

[0038] The present application is based on the structure and size distribution of stem suitable for the process of cigarette rolling, combines the uniformity of stem blending in finished tobacco and the size adaptability of tobacco, establishes the most suitable use interval of stem structure, calculates the grey correlation degree value γ(S 13) = 0.909, so the stem content of the stem contains the inflection rate and the characteristic size of the stem has strong correlation; considering the uniformity of the stem and the tobacco blending in the actual production process, the regression model of the stem content of the stem is established by the stepwise regression analysis method, which is used for the prediction of the feeding stem structure in advance, when the stem content of the stem is known, the characteristic size of the stem can be calculated, and the characteristic size of the stem is improved in the form of stem inflection removal and stem screening, so as to provide certain guarantee for the uniformity and applicability of the subsequent stem and tobacco blending.

[0039] The present application takes the influence of the feeding tobacco stem morphology on the stem structure after processing as the breakthrough point, establishes a stepwise regression model, so that the cigarette enterprise can optimize the incoming structure of the tobacco stem processing section according to the predicted stem structure, determine the stem inflection removal scheme suitable for the tobacco structure, and provide data support for improving the uniformity and stability of the stem structure. BRIEF DESCRIPTION OF DRAWINGS

[0040] The present application will be further described below in combination with the drawings and specific embodiments:

[0041] Figure 1 It is a schematic diagram of the correlation degree of the stem content of the stem and the stem structure.

[0042] Figure 2 It is a cluster analysis tree diagram of the stem content of the stem and the stem structure. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0044] Embodiment 1

[0045] The feeding tobacco stem structure prediction method comprises

[0046] S1 stem structure determination

[0047] Test tobacco stem: general stem used in the production of the tobacco processing workshop;

[0048] Test stem: general stem; sampling point: stem addition outlet falling place;

[0049] Test instrument: Retsch AS400 sieve instrument, screen mesh is 8.00, 6.70, 5.60, 4.75, 4.00, 3.35, 2.80, 2.00, 1.40, 0.71 mm, balance: maximum range 2100 g, sensitivity: 0.01 g; 0.1 g;

[0050] Randomly selected 20 batches of general stem for testing, according to the production batch test number to carry out stem structure test;

[0051] After the stem adding flavor process is stable, the stem sample is taken at the outlet of the stem adding flavor machine, 5 times per batch, the sampling interval is 1 min, the sampling amount is 400g, and the stem sample is reduced to 100.0g±10.0g of the screened sample by using the quartering method;

[0052] The screen instrument is used to screen according to the size of the screen aperture, and the stem sample is divided according to the size of the screen aperture:

[0053] S1: >8.00mm, S2: 6.70~8.00mm, S3: 5.60~6.70mm, S4: 4.75~5.60mm, S5: 4.00~4.75mm, S6: 3.35~4.00mm, S7: 2.80~3.35mm, S8: 2.00~2.80mm, S9: 1.40~2.00mm, S 10 : 0.71~1.40mm, S 11 : <0.71mm, a total of 11 length intervals;

[0054] The operation parameters of the screening instrument are: 4min for each screening, the speed is 230r / min, the direction is changed every 1min, and then the weight of the stem on each layer screen is weighed;

[0055] The percentage of the weight of the stem on each layer screen to the total sample weight is the stem structure of each batch, and the characteristic size and uniformity coefficient are obtained by calculation;

[0056] The least square method is used to fit the stem size, and the characteristic size S 12 and the uniformity coefficient S 13 are calculated:

[0057] Tobacco size distribution equation:

[0058] The a and p are substituted into the equation:

[0059] In the formula: F is the cumulative mass fraction of the stem on the screen, %; a is the characteristic size of the stem, which represents the stem size corresponding to the cumulative amount of the stem on the screen layer being 50%; p is the uniformity coefficient of the stem size distribution; q is the screen aperture, mm;

[0060] Table 1 Stem structure distribution

[0061]

[0062] S2 Tobacco stem content rate measurement

[0063] Randomly take 1000g sample in raw tobacco stem, pick out tobacco stem and tobacco stem containing tobacco stem in tobacco stem, weigh and record, carry out 3 parallel experiments for each batch of sample, and the calculation formula is as follows:

[0064]

[0065] In the formula: G is the tobacco stem containing stem rate, %; g1 is the weight of tobacco stem and tobacco stem containing tobacco stem, g; g is the sample weight, g;

[0066] The tobacco stem sample is analyzed according to the above calculation method to obtain the tobacco stem containing stem rate, and the percentage of stem weight in total sample weight is the tobacco stem containing stem rate, and the results are shown in Table 2;

[0067] Table 2 Tobacco stem containing stem rate statistical results

[0068]

[0069]

[0070] S3 tobacco stem containing stem rate and stem structure gray correlation method analysis

[0071] The tobacco stem containing stem rate is taken as the mother sequence X, and the stem structure is taken as the subsequence Y, the gray correlation degree of the two is calculated, and the order is sorted according to the correlation degree:

[0072] γ(S 13 )>γ(S5)>γ(S 12 )>γ(S8)>γ(S4)>γ(S9)>γ(S 10 )>γ(S2)>γ(S7)>γ(S6)>γ(S1)>γ(S3)>γ(

[0073] S 11 );

[0074] For example Figure 1As shown, the stem content of tobacco stem contained in the stem ratio and stem structure has different degrees of relevance, and the correlation degree is mostly concentrated in 0.50-0.95. Among them, the stem content of tobacco stem contained in the stem ratio and the correlation of the characteristic size of the stem is the strongest, and the correlation degree is 0.909. The stem of 6.70-8.00 mm, 4.75-5.60 mm, 4.00-4.75 mm, 2.80-3.35 mm, 2.00-2.80 mm, 1.40-2.00 mm and 0.71-1.40 mm is strongly correlated with the stem content of tobacco stem contained in the stem ratio, and the stem of >8.00 mm, 5.60-6.70 mm, 3.35-4.00 mm and <0.71 mm is moderately correlated with the stem content of tobacco stem contained in the stem ratio. As a characteristic quantity for measuring the overall size of the stem, the larger the value, the longer the overall size of the stem, and the increase of the stem content of tobacco stem contained in the stem will lead to poor cutting effect, thereby affecting the overall distribution of the size of the stem structure;

[0075] S4 Cluster analysis of stem content of tobacco stem contained in the stem ratio and stem structure

[0076] The gray correlation degree of the stem content of tobacco stem contained in the stem ratio and the stem structure is clustered according to the action intensity of strong correlation (0.7<γ<1), relatively strong correlation (0.4<γ<0.7) and weak correlation (γ<0.4), and the correlation intensity of different sizes of the stem structure and the stem content of tobacco stem contained in the stem ratio is analyzed.

[0077] Through the tree chart Figure 2 Analysis, wherein: size cluster C1={5.60-6.70 mm, <0.71 mm}, size cluster C2={6.70-8.00 mm, 2.80-3.35 mm, 3.35-4.00 mm, >8.00 mm}, size cluster C3={characteristic size, 4.00-4.75 mm, uniformity coefficient, 2.00-2.80 mm, 4.75-5.60 mm, 1.40-2.00 mm, 0.71-1.40 mm};

[0078] Combined with the cluster analysis results, it can be known that the sizes of 4.75-5.60 mm, 4.00-4.75 mm, 2.00-2.80 mm, 1.40-2.00 mm and 0.71-1.40 mm have strong correlation with the stem content of tobacco stem contained in the stem ratio;

[0079] S5 Regression analysis of the characteristic size of the stem and the structure of the stem and the stem

[0080] Further explore the influence of the stem content of tobacco stem contained in the stem ratio on the stem structure, taking the stem content of tobacco stem contained in the stem ratio X1 as the dependent variable, and the sizes of the stem (4.75-5.60 mm, X5, 4.00-4.75 mm, X6, 2.00-2.80 mm, X9, 1.40-2.00 mm, X 10, 0.71~1.40mm, X 11 ), stem uniformity coefficient X 13 , stem characteristic size is X 14 is the independent variable, stepwise regression analysis, SPSS analysis obtained multiple regression equation and significance level;

[0081] Regression equation is y (X1) =1.941X9+2.118X 10 -0.180X 13 +0.424X 14 -1.582; R 2 =0.851, P<0.01;

[0082] Multiple regression gray correlation degree is selected above 0.80 parameter as independent variable, stem rate is dependent variable, the calculation result shows R 2 =0.851, with statistical significance;

[0083] S6 feeding tobacco stem structure prediction

[0084] Through regression analysis method, the prediction model of tobacco stem containing stem rate is established, which is used for the prediction of feeding tobacco stem structure.

[0085] When the tobacco stem containing stem rate of incoming material is known, the characteristic size of stem is calculated, and then combined with the stem production process control parameters, the stem stem removal scheme suitable for the tobacco stem structure is determined, such as stem stem removal, stem stem screening, etc. to improve the characteristic size of stem.

[0086] Example 2

[0087] The feeding tobacco stem structure prediction system comprises

[0088] ①Stem structure measuring unit, which is used for randomly selecting 20 batches of general stems for testing, and testing the stem structure according to the production batch number to obtain 11 optimal length intervals of stem structure, and the characteristic size and uniformity coefficient are obtained by calculation.

[0089] First, after the stem adding flavor process is stable, the stem sample is taken at the outlet of the stem adding flavor machine, 5 times per batch, the sampling interval is 1 min, the sampling amount is 400g, and the stem sample is reduced to 100.0g±10.0g by using the quarter method.

[0090] Then, the screen instrument is used to screen according to the size of the screen mesh aperture, and the stem sample is divided into:

[0091] S1: >8.00 mm, S2: 6.70-8.00 mm, S3: 5.60-6.70 mm, S4: 4.75-5.60 mm, S5: 4.00-4.75 mm, S6: 3.35-4.00 mm, S7: 2.80-3.35 mm, S8: 2.00-2.80 mm, S9: 1.40-2.00 mm, S 10 : 0.71-1.40 mm, S 11 : <0.71 mm, a total of 11 length intervals; screening instrument operating parameters: 4 min per screening, speed 230 r / min, change direction every 1 min;

[0092] Then, the weight of the upper layer of the tobacco stem on each layer of screen is weighed; the percentage of the weight of the tobacco stem on each layer of screen in the total sample weight, i.e. the tobacco stem structure of each batch, is calculated to obtain the characteristic size and uniformity coefficient.

[0093] 2) Tobacco stem containing stem and leaf rate measurement unit, which is used to randomly extract a quantitative sample from the incoming tobacco stem, pick out tobacco stems and tobacco stems containing tobacco stems from the tobacco stem, weigh and record, and calculate the tobacco stem containing stem and leaf rate by performing 3 parallel experiments on each batch of sample: the percentage of the weight of the stem and leaf in the total sample weight is the tobacco stem containing stem and leaf rate.

[0094] 3) Grey correlation analysis unit, which is used for grey correlation degree analysis of tobacco stem containing stem and leaf rate and stem structure;

[0095] Taking the tobacco stem containing stem and leaf rate as the mother sequence X and the stem structure as the child sequence Y, the grey correlation degree of the two is calculated, and the sequences are sorted according to the correlation degree:

[0096] γ(S 13 )>γ(S5)>γ(S 12 )>γ(S8)>γ(S4)>γ(S9)>γ(S 10 )>γ(S2)>γ(S7)>γ(S6)>γ(S1)>γ(S3)>γ(S 11 ).

[0097] 4) Cluster analysis unit, which is used for cluster analysis of tobacco stem containing stem and leaf rate and stem structure;

[0098] The grey correlation degree of the tobacco stem containing stem and leaf rate and the stem structure is clustered according to the action intensity of strong correlation (0.7<γ<1), relatively strong correlation (0.4<γ<0.7), and weak correlation (γ<0.4), and the correlation intensity of different sizes of stem structure and tobacco stem containing stem and leaf rate is analyzed;

[0099] The cluster analysis result: the size of 4.75-5.60 mm, 4.00-4.75 mm, 2.00-2.80 mm, 1.40-2.00 mm and 0.71-1.40 mm stem is strongly related to the stem content of tobacco stem;

[0100] 5. A multiple regression analysis unit for regression analysis of the characteristic size of stem and the structure of tobacco stem and stem, taking the stem content of tobacco stem X1 as the dependent variable, the size of stem (4.75-5.60 mm, X5, 4.00-4.75 mm, X6, 2.00-2.80 mm, X9, 1.40-2.00 mm, X 10 , 0.71-1.40 mm, X 11 ) strongly related to the stem content of tobacco stem in the cluster analysis, the uniformity coefficient of stem X 13 , the characteristic size of stem X 14 as the independent variable, and performing stepwise regression analysis to establish a regression analysis equation y(X1) = 1.941X9 + 2.118X 10 -0.180X 13 +0.424X 14 -1.582;

[0101] 6. A feeding tobacco stem structure prediction unit for establishing a prediction model of the stem content of tobacco stem by the regression analysis method, and applying it to the prediction of the feeding tobacco stem structure: when the stem content of feeding tobacco stem is known, the characteristic size of stem is calculated, and then the stem content removal scheme suitable for the tobacco stem structure is determined by combining the production process control parameters of stem, such as stem content removal, stem sign screening, etc. to improve the characteristic size of stem.

[0102] It should be noted that the above only describes the preferred embodiments of the present application, and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features, and any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting a feeding tobacco stem structure, characterized by: Comprising S1 determination of cut stem structure Randomly select several batches of general stems for testing, and test them according to production batches; At the outlet of the cut stem flavoring machine, cut stem samples are taken, and the cut stem samples are reduced to 100.0g±10.0g of the screened samples; According to the size of the screen mesh aperture, the cut stem samples are divided into: S1: >8.00 mm, S2: 6.70~8.00 mm, S3: 5.60~6.70 mm, S4: 4.75~5.60 mm, S5: 4.00~4.75 mm, S6: 3.35~4.00 mm, S7: 2.80~3.35 mm, S8: 2.00~2.80 mm, S9: 1.40~2.00 mm, S 10 : 0.71~1.40 mm, S 11 : <0.71 mm 11 length intervals; The weight of stem on each layer of screen is weighed, and the percentage of the weight of stem on each layer of screen in the total sample weight, i.e. the stem structure of each batch, is calculated. The stem size is fitted by using the least square method, and the characteristic size S is calculated 12 and the uniformity coefficient S 13 : Tobacco cut size distribution equation: ; Solve for a, p, and substitute into the equation: ; In the formula: F is the cumulative mass fraction of the cut stem screen, %; d is the characteristic size of the cut stem, which represents the cut stem size corresponding to the cumulative amount of 50% on the screen; p is the uniformity coefficient of the cut stem size distribution; q is the screen mesh aperture, mm; S2 measurement of tobacco stem containing stem rate Randomly select samples from incoming tobacco stems, pick out tobacco stems and tobacco stems containing tobacco stems, weigh and record, and perform multiple parallel experiments on each batch of samples, with the calculation formula as follows: ； In the formula: G is the tobacco stem containing stem rate, %; g1 is the weight of tobacco stems and tobacco stems containing tobacco stems, g; g is the sample weight, g; S3 gray correlation analysis of tobacco stem containing stem rate and cut stem structure Take the tobacco stem containing stem rate as the mother sequence X and the cut stem structure as the child sequence Y, calculate the gray correlation degree of the two, and sort them according to the correlation degree; S4 cluster analysis of tobacco stem containing stem rate and cut stem structure The grey correlation degree of the tobacco stem containing rate and the stem structure was clustered according to the action intensity of strong correlation 0.7 <0.4, and the correlation intensity of different sizes of the stem structure and the tobacco stem containing rate was analyzed. <0.4 <0.4, and the correlation intensity of different sizes of the stem structure and the tobacco stem containing rate was analyzed. S5 regression analysis of cut stem characteristic size and tobacco stem and cut stem structure With the ratio of tobacco stem containing stem ratio X1 as the dependent variable, the stem size and the uniformity coefficient X 13 of stem in cluster analysis with strong correlation with the ratio of tobacco stem containing stem ratio as the independent variable, and the characteristic size of stem as X 14 , stepwise regression analysis was carried out to obtain the multiple regression equation y(X1) and the significance level. S6 prediction of incoming tobacco stem structure A prediction model of the tobacco stem containing stem rate is established by regression analysis method for prediction of the incoming tobacco stem structure. When the incoming tobacco stem containing stem rate is known, the characteristic size of the cut stem is calculated, and combined with the cut stem production process control parameters, a stem stem removal scheme suitable for the cut tobacco structure is determined to improve the characteristic size of the cut stem.

2. The incoming tobacco stem structure prediction method according to claim 1, wherein: In S1, after the cut stem flavoring process is stable, cut stem samples are taken at the outlet of the cut stem flavoring machine, 5 times of sampling are taken per batch, the sampling time interval is 1 min, the sampling amount is 400g, and the cut stem samples are reduced to 100.0g±10.0g of the screened samples by using the quartering method.

3. The incoming tobacco stem structure prediction method according to claim 1, wherein: In S1, the cut stem samples are screened by using a screening instrument according to the size of the screen mesh aperture; The screening instrument operating parameters are: 4 min for each screening, speed 230r / min, and direction change every 1 min.

4. The incoming tobacco stem structure prediction method according to claim 1, wherein: In S6, the stem stem removal scheme includes stem stem removal and stem signature screening.

Images