Method for substituting tobacco leaves in cigarette formula on basis of integration of static and dynamic quality characteristics of tobacco leaves

Abstract

A method for substituting tobacco leaves in a cigarette formula on the basis of the integration of static and dynamic quality characteristics of tobacco leaves, the method comprising: (1) fusing near-infrared feature spectral data with thermogravimetric curve feature spectral data, so as to obtain a quality feature matrix X of tobacco leaves in a cigarette formula; (2) by using the same method, obtaining a quality feature matrix Y of different tobacco leaves in stock; (3) calculating a Mahalanobis distance value d between the matrix X and the matrix Y; and (4) according to the Mahalanobis distance value d, selecting tobacco leaves in stock to serve as substitute tobacco leaves. The method is scientific and efficient, and comprehensively ensures the quality and flavor of cigarettes.

Description

A cigarette formula tobacco leaf replacement method integrating static and dynamic quality characteristics of tobacco leaves Technical Field

[0001] The present invention belongs to the technical field of cigarettes, and in particular relates to a method for replacing tobacco leaves in cigarette formulas that integrates the static and dynamic quality characteristics of tobacco leaves. Background technology:

[0002] The quality and distinctive style of cigarettes are primarily determined by product designers through the blending of tobacco leaves from different origins, varieties, and grades. When a particular tobacco leaf runs out, manual selection of similar quality leaves from hundreds of stockpiled grades is necessary, relying on recipe experience and sensory evaluation. This process is highly subjective and labor-intensive, and due to the lack of objective, scientific, and efficient technical means, it is difficult to fully guarantee the stability of cigarette quality and style. Using instrumental testing methods, objective data, and scientific techniques to assist in blend substitution would be crucial for improving cigarette quality stability.

[0003] In recent years, near-infrared (NIR) technology has been widely used in tobacco quality analysis and evaluation, as well as formulation design and maintenance, due to its rapidity, high efficiency, and rich quality information. It can effectively characterize the intrinsic chemical composition of tobacco leaves, effectively characterizing their quality. However, NIR analysis only considers the correlation between tobacco quality and leaf formulation under "static conditions," ignoring the heat transfer behavior during the combustion process of cigarette consumption. This technology cannot truly represent the smoking quality of tobacco leaves, specifically the quality characteristics of the smoke. Therefore, formulation substitution technology based on NIR analysis has certain limitations.

[0004] Thermal analysis technology is one of the most widely used methods in the industry to bridge the gap between tobacco leaves and smoke. Thermogravimetric analysis (TG) measures the relationship between sample weight (mass) and temperature or time under programmed temperature control and in different atmospheres, indirectly reflecting the pyrolysis and combustion behavior of different tobaccos. Applications of thermal analysis have been reported to identify tobacco thermal conversion characteristics, conduct tobacco blend substitution trials, and evaluate the batch quality stability of reconstituted tobacco and cigarette paper.

[0005] To comprehensively characterize the smoking quality of cigarette blends and tobacco leaf raw materials, this present invention utilizes objective data and scientific techniques based on near-infrared spectroscopy data combined with thermogravimetric analysis data, replacing the traditional tobacco leaf replacement method that relies on sensory evaluation and expert experience.

[0006] Summary of the Invention

[0007] The present invention uses a multimodal fusion algorithm based on a convolutional neural network to merge and project the two aspects of quality information, near-infrared spectral data (static quality characteristics) and tobacco leaf thermogravimetric curve data (dynamic quality characteristics), into the same multimodal representation space, thereby forming a feature matrix that can fully characterize the smoking quality of tobacco leaves; on this basis, the quality similarity measurement and calculation of the tobacco leaves in the formula and the tobacco leaves in stock are carried out to realize the judgment and intelligent recommendation of whether the tobacco leaves in stock can replace the tobacco leaves in the cigarette formula.

[0008] The technical solutions of the present invention are as follows:

[0009] A method for replacing tobacco leaves in cigarette formulas that integrates the static and dynamic quality characteristics of tobacco leaves comprises the following steps:

[0010] (1) The near-infrared characteristic spectrum data and thermogravimetric curve characteristic spectrum data of tobacco leaves in the cigarette formula are combined with a multimodal fusion algorithm based on a convolutional neural network to obtain the tobacco leaf quality feature matrix X in the cigarette formula;

[0011] (2) The near-infrared characteristic spectrum data and thermogravimetric curve characteristic spectrum data of different stock tobacco leaves are combined with a multimodal fusion algorithm based on a convolutional neural network to obtain the stock tobacco leaf quality feature matrix Y;

[0012] (3) Calculate the Mahalanobis distance d between matrices X and Y;

[0013] (4) Selecting stock tobacco leaves with a Mahalanobis distance value d not greater than a certain value as replacement tobacco leaves for tobacco leaves in cigarette formulas.

[0014] Preferably, the near infrared characteristic spectrum data of tobacco leaves is: 4350-4450cm -1 , 4500-4700cm -1 , 4800-4900cm -1 , 5050-5100cm -1 , 5220-5300cm -1 , 6050-6250cm -1 , 6400-6600cm -1 , 6950-7220cm -1 Taking the content of the main chemical components of tobacco leaves as characteristic values, including total sugar, total nicotine, reducing sugar, total nitrogen and other indicators, combined with the absorption peak position of each molecular functional group in the near-infrared spectrum, as shown in Figure 1, the characteristic spectrum segments of the near-infrared spectrum of tobacco leaves were screened respectively.

[0015] Preferably, the characteristic spectrum data of the thermogravimetric curve of the tobacco leaf are: 190°C-240°C, 320°C-360°C.

[0016] Preferably, the steps of using the multimodal fusion algorithm based on convolutional neural network are as follows: 1) fully connecting the near infrared characteristic spectrum data of tobacco leaves to the full connection layer f m and tobacco leaf thermogravimetric curve characteristic spectrum data fully connected layer f n Multiply together to form a two-dimensional feature fusion transformation matrix A m×n 2) A m×n Perform maximum pooling to form m feature nodes as the new fusion feature layer B m ;3) Apply softmax layer to B m Logarithmic normalization is performed to compress the variables and return them to the range [0, 1], thus obtaining the final fusion feature layer.

[0017] Preferably, the Mahalanobis distance calculation formula is as follows: Where X is the tobacco quality characteristic matrix X in the cigarette formula, Y is the thermal property quality characteristic matrix Y of the stock tobacco leaves, and Σ is the covariance matrix of X and Y.

[0018] Preferably, stock tobacco leaves with a Mahalanobis distance value d not greater than 2.5000 are selected as substitute tobacco leaves for tobacco leaves in cigarette formulations.

[0019] Beneficial effects of the present invention:

[0020] The present invention's tobacco leaf replacement method for cigarette blends, integrating static and dynamic tobacco leaf quality characteristics, utilizes a multimodal fusion algorithm based on convolutional neural networks for the first time. This algorithm combines and projects characteristic spectral segments from tobacco leaf near-infrared spectra and thermogravimetric curves into a single multimodal representation space, forming a feature matrix that fully characterizes the tobacco leaf's smoking quality. Furthermore, the Mahalanobis distance is used to measure the quality similarity between the tobacco leaf to be replaced in the blend and the tobacco leaf in stock, enabling intelligent judgment and recommendation of tobacco leaf replacements in cigarette blends. This objective, scientific, and efficient method enables judgment of whether tobacco leaf in stock can be substituted for a tobacco leaf in a cigarette blend and intelligently recommends replacements, fully ensuring the stability of cigarette quality and flavor. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] FIG1 is a diagram showing the screening of characteristic spectrum segments of tobacco leaf near-infrared spectrum according to the present invention.

[0022] FIG2 is a schematic diagram of the tobacco leaf static and dynamic quality feature fusion algorithm of the present invention.

[0023] FIG3 is a comparison of the main chemical composition indicators of the tobacco formula before and after the tobacco leaf replacement in the embodiment.

[0024] FIG4 is a comparison of sensory evaluation differences of the tobacco leaf formula before and after replacement in the embodiment. DETAILED DESCRIPTION

[0025] The present invention will be further described in detail below through specific embodiments, but this should not be construed as limiting the scope of the present invention to the following examples. Various substitutions and modifications made according to common technical knowledge and customary means in the art without departing from the spirit of the present invention are intended to be included within the scope of the present invention.

[0026] Example: Determining whether stockpiled tobacco leaves can replace tobacco leaves in cigarette formulations.

[0027] The present invention takes four cigarette formulations as examples, and the tobacco leaf information in these four cigarette formulations is shown in Table 1 below.

[0028] Table 1 Tobacco leaf information in four cigarette formulations

[0029] The specific steps for determining whether the tobacco leaves in stock can replace the tobacco leaves in the four cigarette formulas are as follows:

[0030] (1) The near-infrared spectral characteristic band data and the thermogravimetric curve characteristic band data of the tobacco leaves in the cigarette formula are combined with a multimodal fusion algorithm based on a convolutional neural network, as shown in Figure 2, to obtain the tobacco leaf quality feature matrix X in the cigarette formula; specifically:

[0031] The tobacco leaf samples from the four cigarette formulas were evenly divided into two parts; one part was used for near-infrared spectroscopy analysis, and the other part was used for thermogravimetric analysis.

[0032] Near-infrared spectroscopy analysis of tobacco leaves: Place the tobacco leaf sample in a 40°C oven to dry for 2 hours, grind it through a 40-mesh sieve with a cyclone mill, weigh 3g of tobacco leaf powder and place it in a sample cup. Scan the 4000-9000cm -1 The spectrum of the band was repeated 10 times for each sample and the spectrum average was taken. The content of the main chemical components of tobacco leaves, including total sugar, total nicotine, reducing sugar, total nitrogen and other indicators, was used as the characteristic value. Combined with the absorption peak position of each molecular functional group in the near-infrared spectrum, the characteristic spectrum band of tobacco leaves near-infrared spectrum was screened: 4350-4450cm -1 , 4500-4700cm -1 , 4800-4900cm -1 , 5050-5100cm -1 , 5220-5300cm -1 , 6050-6250cm -1 , 6400-6600cm -1 , 6950-7220cm -1 .

[0033] Tobacco leaf thermogravimetric analysis: Using a TGA / DSC 1 LF thermogravimetric analyzer from Mettler Toledo, Germany, (5.00±0.05) mg of sample was weighed and placed in a thermogravimetric alumina crucible. The temperature was raised from 30°C to 900°C at a rate of 10°C / min under a nitrogen flow rate of 30 mL / min to obtain the tobacco leaf thermogravimetric curve. The characteristic spectrum of the tobacco leaf thermogravimetric curve was screened based on the main heat transfer weight loss stage during tobacco leaf combustion: 190°C-240°C, 320°C-360°C

[0034] The convolutional neural network algorithm is used to perform feature fusion on the near infrared characteristic spectrum data of the tobacco leaf sample in the recipe and the characteristic spectrum data of the thermogravimetric curve. The steps are as follows: 1) the near infrared characteristic spectrum fully connected layer f m and the fully connected layer f of the characteristic spectrum of the thermogravimetric curve n Multiply together to form a two-dimensional feature fusion transformation matrix A m×n 2) A m×n Perform maximum pooling to form m feature nodes as the new fusion feature layer B m ;3) Apply softmax layer to B m Logarithmic normalization is performed to compress the variables and return them to the range [0, 1], thereby obtaining the final fusion feature matrix X, as shown in Table 2.

[0035] Table 2 Tobacco leaf fusion feature matrix X in the recipe

[0036] (2) The same steps as step (1) were followed for the 12 different stock tobacco leaves to obtain the fusion feature matrix of the stock tobacco leaves. The 12 different stock tobacco leaves are shown in Table 3.

[0037] (3) Calculate the Mahalanobis distance d of the thermal property quality matrix of the tobacco leaves in the formula and the tobacco leaves in stock. The results are shown in Table 3.

[0038] Table 3 Calculation results of Mahalanobis distance d of the fusion feature matrix of tobacco leaves in the formula and tobacco leaves in stock

[0039] (4) As can be seen from Table 3, only C-3 has a Mahalanobis distance greater than 2.5000 between the tobacco leaves in the formula and the tobacco leaves in stock, which is 2.9150. This tobacco leaf in stock cannot be used as a substitute for tobacco leaves in cigarette formulas. The other 11 tobacco leaves in stock from the four formulas can be used as substitutes for tobacco leaves in the formula.

[0040] In order to verify the validity of the determination results of tobacco leaf replacement, the differences in the main chemical composition indicators and sensory evaluation of the tobacco leaf replacement were compared before and after.

[0041] A comparison of the main chemical composition indicators of the four cigarette formulas before and after the replacement is shown in Figure 3. As can be seen from Figure 3, the main chemical composition indicators of the cigarettes with the replacement formula of the stock tobacco leaves and the original formula are relatively small, with only slight differences in the total sugar, reducing sugar, and potassium-chloride ratio indicators of "Recommended Formula C 3" (C-3).

[0042] Furthermore, sensory evaluation was used to assess the differences before and after tobacco leaf substitution. Using the sensory quality difference evaluation method described in Table 4, the cigarette samples were sensory evaluated before and after tobacco leaf substitution. The results are shown in Figure 4. As can be seen from Figure 4, only "Recommended Solution 3 of Formulation C" (C-3) showed a significant difference in evaluation (the Mahalanobis distance between the tobacco leaves before and after substitution was 2.9150), which is consistent with the judgment method of the present invention. This demonstrates that the judgment and recommendation methods of the present invention are consistent with the actual sensory evaluation results, and therefore the judgment method of the present invention is objective, scientific, and efficient.

[0043] Table 4 Evaluation of sensory quality differences of formulas

[0044] illustrate:

[0045] No difference: The quality difference is not perceived and the substituted formula can be used in cigarette production;

[0046] Slight difference: The quality difference cannot be accurately and stably perceived, and the substituted formula can be used in cigarette production;

[0047] ● There are differences: There are quality differences, and the replaced formula cannot be used in cigarette production.

[0048] The embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for replacing tobacco leaves in cigarette formulas that combines the static and dynamic quality characteristics of tobacco leaves. It is characterized in that The steps include: (1) The near-infrared characteristic spectrum data and the thermogravimetric curve characteristic spectrum data of tobacco leaves in the cigarette formula are combined with a multimodal fusion algorithm based on a convolutional neural network to obtain a tobacco leaf quality characteristic matrix X in the cigarette formula; (2) The near-infrared characteristic spectrum data and thermogravimetric curve characteristic spectrum data of different stock tobacco leaves are combined with a multimodal fusion algorithm based on a convolutional neural network to obtain the stock tobacco leaf quality characteristic matrix Y; (3) Calculate the Mahalanobis distance d between matrices X and Y; (4) Selecting stock tobacco leaves whose Mahalanobis distance value d is not greater than a certain value as substitute tobacco leaves for tobacco leaves in cigarette formula.

2. The method according to claim 1, It is characterized in that The near infrared characteristic spectrum data of tobacco leaves is: 4350-4450cm -1 , 4500-4700cm -1 , 4800-4900cm -1 , 5050-5100cm -1 , 5220-5300cm -1 , 6050-6250cm -1 , 6400-6600cm -1 , 6950-7220cm -1 .

3. The method according to claim 1, It is characterized in that Characteristic spectral data of the thermogravimetric curve of tobacco leaves: 190℃-240℃, 320℃-360℃.

4. The method according to claim 1, It is characterized in that The steps of using the multimodal fusion algorithm based on convolutional neural network are as follows: 1) fully connect the near-infrared characteristic spectrum data of tobacco leaves to the full connection layer f m The fully connected layer f of the characteristic spectrum data of the thermogravimetric curve of tobacco leaves n Multiply them together to form a two-dimensional feature fusion transformation matrix A m×n ; 2) For A m×n Perform maximum pooling to form m feature nodes as the new fusion feature layer B m ; 3) Apply softmax layer to B m Logarithmic normalization is performed to compress the variables and return them to [0, 1], thus obtaining the final fusion feature layer.

5. The method according to claim 1, It is characterized in that The Mahalanobis distance calculation formula is as follows: Among them, X is the tobacco quality feature matrix X in the cigarette formula, Y is the thermal property quality feature matrix Y of the stock tobacco, and Σ is the covariance matrix of X and Y.

6. The method according to claim 1, It is characterized in that The stock tobacco leaves with a Mahalanobis distance value d not greater than 2.5000 are selected as substitute tobacco leaves for tobacco leaves in cigarette formulas.

Images