Tobacco product production place proportion prediction method and combination formula replacement method

Abstract

The invention relates to a tobacco product producing area proportion prediction method and a combination formula replacement method, and relates to the technical field of tobacco production. The method comprises the following steps: acquiring first thermogravimetric analysis data of a plurality of first pyrolysis temperatures of a pyrolysis process of the tobacco product, a quality change rate of the tobacco product corresponding to each first pyrolysis temperature, and first chemical component data of the content of each chemical component in a plurality of chemical components of the tobacco product; determining first characteristic data related to a characteristic temperature in the plurality of first pyrolysis temperatures and second characteristic data related to a mass change rate corresponding to the characteristic temperature from the first thermogravimetric analysis data; and based on at least one of the first feature data and the second feature data and the first chemical component data, utilizing a trained prediction model corresponding to the target production place to predict the mass ratio of the tobacco product raw materials of the target production place in the tobacco product.

Description

Technical Field

[0001] The present disclosure relates to the technical field of tobacco production, and in particular to a method for predicting the origin ratio of tobacco products and a method for replacing a combination formula. Background Art

[0002] Tobacco products are made from a variety of raw materials combined according to specific formulas. For example, in the case of tobacco leaves, which are used as raw materials and cigarettes, the raw tobacco leaves can have a specific origin. Tobacco raw materials from different origins vary in aroma, taste, and other aspects. Understanding the weight percentage of raw materials in tobacco products can help scientifically blend raw materials from different origins, ultimately improving the quality of the final tobacco product. Summary of the Invention

[0003] According to some embodiments of the present disclosure, a method for predicting the origin ratio of tobacco products is provided, comprising: obtaining first thermogravimetric analysis data and first chemical composition data of the tobacco product, the first thermogravimetric analysis data comprising a plurality of first pyrolysis temperatures of the pyrolysis process of the tobacco product and a mass change rate of the tobacco product corresponding to each of the plurality of first pyrolysis temperatures, and the first chemical composition data comprising the content of each chemical component among the plurality of chemical components of the tobacco product; determining, from the first thermogravimetric analysis data, first characteristic data related to a characteristic temperature among the plurality of first pyrolysis temperatures, and second characteristic data related to the mass change rate corresponding to the characteristic temperature; and predicting, based on at least one of the first characteristic data and the second characteristic data and the first chemical composition data, the mass proportion of tobacco product raw materials from the target origin in the tobacco products using a trained prediction model corresponding to the target origin.

[0004] According to other embodiments of the present disclosure, a method for replacing a combination formula of a tobacco product is provided, comprising: for each candidate combination formula among a plurality of candidate combination formulas, obtaining first thermogravimetric analysis data and first chemical composition data of the candidate tobacco product corresponding to each candidate combination formula, wherein the first thermogravimetric analysis data includes a plurality of first pyrolysis temperatures of a pyrolysis process of the candidate tobacco product and a mass change rate of the candidate tobacco product corresponding to each first pyrolysis temperature among the plurality of first pyrolysis temperatures, and the first chemical composition data includes the content of each chemical component among the plurality of chemical components of the candidate tobacco product; and determining, from the first thermogravimetric analysis data, a characteristic temperature associated with one of the plurality of first pyrolysis temperatures. first characteristic data, and second characteristic data related to the mass change rate corresponding to the characteristic temperature; based on at least one of the first characteristic data and the second characteristic data and the first chemical composition data, using a trained prediction model corresponding to the target origin to predict the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco products; a candidate combination formula in which the difference between the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco products and the mass proportion of tobacco product raw materials from the target origin in a known combination formula to be replaced in a known tobacco product corresponding to the known combination formula is less than a specified threshold is determined as a target combination formula for replacing the known combination formula.

[0005] According to some further embodiments of the present disclosure, a device for predicting the origin ratio of tobacco products is provided, comprising: an acquisition module configured to acquire first thermogravimetric analysis data and first chemical composition data of the tobacco product, the first thermogravimetric analysis data comprising a plurality of first pyrolysis temperatures of the pyrolysis process of the tobacco product and a mass change rate of the tobacco product corresponding to each of the plurality of first pyrolysis temperatures, and the first chemical composition data comprising the content of each chemical component among the plurality of chemical components of the tobacco product; a determination module configured to determine, from the first thermogravimetric analysis data, first characteristic data related to a characteristic temperature among the plurality of first pyrolysis temperatures, and second characteristic data related to the mass change rate corresponding to the characteristic temperature; and a prediction module configured to predict, based on at least one of the first characteristic data and the second characteristic data and the first chemical composition data, the mass proportion of tobacco product raw materials from the target origin in the tobacco products using a trained prediction model corresponding to the target origin.

[0006] According to some further embodiments of the present disclosure, a device for replacing a combination formula of a tobacco product is provided, comprising: an acquisition module configured to, for each candidate combination formula in a plurality of candidate combination formulas, acquire first thermogravimetric analysis data and first chemical composition data of a candidate tobacco product corresponding to the each candidate combination formula, the first thermogravimetric analysis data comprising a plurality of first pyrolysis temperatures of a pyrolysis process of the candidate tobacco product and a mass change rate of the candidate tobacco product corresponding to each of the plurality of first pyrolysis temperatures, and the first chemical composition data comprising the content of each chemical component in the plurality of chemical components of the candidate tobacco product; a first determination module configured to determine, from the first thermogravimetric analysis data, first characteristic data associated with a characteristic temperature in the plurality of first pyrolysis temperatures, and second characteristic data associated with the mass change rate corresponding to the characteristic temperature. data; a prediction module configured to predict the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco products based on at least one of the first feature data and the second feature data and the first chemical composition data using a trained prediction model corresponding to the target origin; a second determination module configured to determine, as a target combination formula for replacing the known combination formula, a candidate combination formula in which the difference between the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco products and the mass proportion of tobacco product raw materials from the target origin in the known tobacco products corresponding to the known combination formula to be replaced is less than a specified threshold.

[0007] According to some other embodiments of the present disclosure, an electronic device is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute the method for predicting the origin ratio of tobacco products or the method for replacing the combination formula of tobacco products in any of the above-mentioned embodiments based on instructions stored in the memory device.

[0008] According to some other embodiments of the present disclosure, a computer-readable storage medium is provided, on which computer instructions are stored. When the instructions are executed by a processor, the method for predicting the origin ratio of tobacco products or the method for replacing the combined formula of tobacco products in any of the above embodiments is implemented.

[0009] According to some further embodiments of the present disclosure, a computer program product is also provided, comprising instructions, which, when executed by a processor, cause the processor to execute the method for predicting the origin ratio of tobacco products or the method for replacing the combination formula of tobacco products according to any of the above embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0011] The present disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0012] Figure 1 A flowchart showing a method for predicting the origin ratio of tobacco products according to some embodiments of the present disclosure;

[0013] Figure 2 A schematic diagram showing a DTG curve according to some embodiments of the present disclosure;

[0014] Figure 3 A flowchart illustrating a method for training a prediction model according to some embodiments of the present disclosure is provided;

[0015] Figure 4 A schematic diagram showing the structure of a convolutional neural network model according to some embodiments of the present disclosure is shown;

[0016] Figure 5 A flowchart illustrating a method for training a prediction model according to other embodiments of the present disclosure is shown;

[0017] Figure 6 A flow chart illustrating an alternative method of combining formulations of tobacco products according to some embodiments of the present disclosure;

[0018] Figure 7 A block diagram illustrating prediction of the origin ratio of tobacco products according to some embodiments of the present disclosure;

[0019] Figure 8 A block diagram illustrating an alternative apparatus for combining formulations of tobacco products according to some embodiments of the present disclosure;

[0020] Figure 9 A block diagram illustrating an electronic device according to some embodiments of the present disclosure is shown;

[0021] Figure 10 A block diagram of an electronic device according to some other embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0022] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0023] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0024] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0025] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0026] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0027] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0028] As mentioned above, combining different tobacco product raw materials according to a specific combination formula can form tobacco products that exhibit the style of a specific origin.

[0029] The combined formula of multiple tobacco raw materials in a tobacco product may include the mass proportions of the multiple tobacco raw materials in the tobacco product. The mass proportions may be presented as absolute or relative values. In the form of absolute values, the combined formula may, for example, include the mass of each tobacco raw material (e.g., in units of grams, milligrams, micrograms, etc.), thereby indirectly reflecting the mass proportions. In the case of relative values, the combined formula may, for example, include the ratio obtained by dividing the mass of each tobacco raw material by the mass of the tobacco product (e.g., in percentages), thereby directly reflecting the mass proportions.

[0030] To help understand the mass proportion of tobacco product raw materials from different origins in tobacco products, the present disclosure proposes a method for predicting the origin ratio of tobacco products. The method can use a trained prediction model corresponding to the target origin to fuse two types of data: thermogravimetric analysis data and chemical composition data of tobacco products, and accurately predict the mass proportion of tobacco product raw materials from the target origin in tobacco products, thereby providing a theoretical basis for the subsequent optimization of the design of combination formulas.

[0031] Figure 1 A flow chart showing a method for predicting the origin ratio of tobacco products according to some embodiments of the present disclosure is shown. Figure 1 As shown, the method for predicting the origin ratio of tobacco products includes steps 110 to 130.

[0032] In step 110, first thermogravimetric analysis data and first chemical composition data of the tobacco product are obtained.

[0033] In this disclosure, tobacco products are products made wholly or partially from tobacco leaves and intended for smoking, chewing, snorting, or other uses. Tobacco products may include, but are not limited to, cigarettes, cigars, pipe tobacco, or hookah. Cigarettes, based on their tobacco leaf formulation and processing techniques, may include, but are not limited to, flue-cured tobacco, burley tobacco, oriental tobacco, or sun-cured tobacco. Tobacco products release chemicals such as nicotine (commonly known as "nicotine") through heating (e.g., heat-not-burn tobacco products) and / or combustion (e.g., cigarettes, cigars, etc.). Tobacco products may or may not be wrapped in paper and may or may not have filters. In this disclosure, for ease of description, cigarettes are sometimes used as an example of tobacco products. However, it should be understood that the various features or limitations described herein with respect to cigarettes also apply to other types of tobacco products.

[0034] Tobacco product raw materials may include tobacco leaves in various forms, such as tobacco leaves or tobacco shreds, tobacco flakes, tobacco powder, or tobacco blocks processed from tobacco leaves. In some embodiments, the tobacco product raw materials are tobacco leaves from a specific origin. Here, the multiple tobacco product raw materials may be tobacco leaves from different origins, or from different growing parts of the same origin (for example, divided into "upper part," "middle part," and "lower part" according to the growing part of the tobacco plant), or from different batches of tobacco leaves from the same growing part of the same origin.

[0035] In some embodiments, the plurality of tobacco feedstocks can include tobacco feedstocks from a plurality of different origins.

[0036] The first thermogravimetric analysis data of the tobacco product includes a plurality of first pyrolysis temperatures of the tobacco product during a pyrolysis process and a mass change rate of the tobacco product corresponding to each of the plurality of first pyrolysis temperatures.

[0037] In some embodiments, the first thermogravimetric analysis data of the tobacco product can be obtained based on the second thermogravimetric analysis data of each of the plurality of tobacco raw materials constituting the tobacco product.

[0038] In some implementations, the first thermogravimetric analysis data for a tobacco product can be determined based on the average of the second thermogravimetric analysis data for multiple tobacco product raw materials. For example, a tobacco product is known to consist of five tobacco leaves A from Yunnan, three tobacco leaves B from Fujian, and two tobacco leaves C from Henan, each weighing 100g. The second thermogravimetric analysis data for the five tobacco leaves A are (S11, S12, S13, S14, S15), the second thermogravimetric analysis data for the three tobacco leaves B are (S21, S22, S23), and the second thermogravimetric analysis data for the two tobacco leaves C are (S31, S32). The first thermogravimetric analysis data S of the tobacco product can be the average of the second thermogravimetric analysis data of 5 tobacco leaves A, the second thermogravimetric analysis data of 3 tobacco leaves B, and the second thermogravimetric analysis data of 2 tobacco leaves C, that is, S = (S11+S12+S13+S14+S15+ S21+S22+S23+ S31+S32) / 10.

[0039] It should be noted that the origin style of tobacco products may undergo uncertain changes over time. To accurately analyze the causes of such changes, given a known tobacco product's formula, the first thermogravimetric analysis data for the tobacco product can be obtained based on the aforementioned calculation method. Then, using a prediction model corresponding to the target origin, the weight percentage of tobacco product raw materials from the target origin in the tobacco product can be predicted. This method can clarify whether changes in the origin style of a tobacco product are due to changes in the weight percentage of tobacco product raw materials from the target origin over time. For example, the original formula of a tobacco product is: Yunnan tobacco leaves account for 50% by weight, Fujian tobacco leaves account for 30% by weight, and Henan tobacco leaves account for 20% by weight. The tobacco product based on this formula exhibits a typical Yunnan origin style. However, over time, the origin style of the tobacco product may undergo uncertain changes. In this case, by predicting whether the weight percentage of Yunnan tobacco leaves in the tobacco product remains at 50%, it can be determined whether the change in the origin style of the tobacco product is due to changes in the weight percentage of Yunnan tobacco leaves over time.

[0040] Once the cause of this change is determined, the blend can be scientifically and rationally adjusted. For example, if the quality proportion of Yunnan tobacco is found to have decreased while the quality proportion of Fujian tobacco has increased, resulting in a change in the origin style, the original origin style can be restored by increasing the quality proportion of Yunnan tobacco.

[0041] In other words, given a clear tobacco product formula, accurately predicting the mass proportions of tobacco raw materials from different origins within that formula can provide a deeper understanding of the origin-specific style retention of these raw materials. This retention refers to the degree to which the stylistic characteristics (such as aroma and taste) of tobacco raw materials from different origins are retained and reflected in the finished tobacco product. This predictive analysis clearly identifies whether changes in tobacco product origin style are related to changes in the mass proportions of raw materials from different origins, thereby facilitating timely adjustments to the formula to maintain product style stability and, in turn, enhance product quality and market competitiveness.

[0042] It should be understood that the thermogravimetric analysis data of the tobacco raw material can be obtained by conducting a pyrolysis experiment on the tobacco raw material and recording multiple pyrolysis temperatures during the pyrolysis process and the mass change rate of the tobacco raw material corresponding to each pyrolysis temperature. The mass change rate can be determined by calculating the derivative of the mass of the tobacco raw material with respect to temperature during the pyrolysis process.

[0043] The following uses tobacco leaves as an example of the raw material for tobacco products to illustrate the process of conducting a pyrolysis experiment on the raw material for tobacco products.

[0044] First, a tobacco leaf is placed in a constant temperature and humidity chamber for a period of time. The leaf is then ground, filtered through a mesh, and thoroughly mixed. A 20 mg sample, for example, is weighed. In a thermal analyzer, the sample is heated at a constant rate (e.g., 10 K / min) to a specific temperature (e.g., 373 K) under a nitrogen atmosphere (e.g., a nitrogen flow rate of 100 mL / min (i.e., a carrier gas flow rate of 100 mL / min)). This temperature is then maintained at a constant temperature for a period of time to eliminate the effects of varying moisture content in the leaf. The sample is then heated at the same rate to a higher temperature (e.g., 873 K). The relationship between sample mass and temperature is recorded during the heating process to generate a weight loss (TG) curve. To make tobacco pyrolysis behavior more visible, the TG curve is differentiated to produce a Derivative Thermo-Gravimetry (DTG) curve. This curve reflects the relationship between the rate of mass change and temperature, referred to herein as "thermogravimetric analysis data." When thermogravimetric analysis data is presented as a graph (i.e., a DTG curve), the horizontal axis represents temperature (referred to herein as "pyrolysis temperature") in Kelvin (K), and the vertical axis represents the rate of mass change during pyrolysis in percentage per Kelvin (% / K). This curve is also known as a thermogram.

[0045] In some embodiments, the first thermogravimetric analysis data for a tobacco product can be obtained by conducting a thermogravimetric analysis experiment (i.e., a pyrolysis experiment) on the tobacco product. For example, in some cases, the tobacco product's composition (i.e., the tobacco product raw materials used and their mass ratios) may not be known, or even if the raw materials are known, the thermogravimetric analysis data for the raw materials are unknown, making it impossible to obtain the first thermogravimetric analysis data for the tobacco product using the aforementioned calculation method. In such cases, it is desirable to determine the mass ratio of tobacco product raw materials from a specific target production area, thereby quantifying the quality characteristics of the tobacco product. Furthermore, this method is more accurate than first thermogravimetric analysis data obtained through calculation, helping to improve the accuracy of subsequent predictions.

[0046] Here, the pyrolysis experiment conducted on tobacco products is similar to the pyrolysis experiment conducted on tobacco product raw materials. For specific instructions, please refer to the relevant description of the pyrolysis experiment conducted on tobacco product raw materials above.

[0047] Figure 2 A schematic diagram showing a DTG curve according to some embodiments of the present disclosure.

[0048] For example, in the above example, 800 data points between 401.138 K and 801.138 K can be taken as the first thermogravimetric analysis data of the tobacco product with a step size of 0.5 K, and displayed in the form of a curve to obtain the following: Figure 2 The DTG curve is shown as a black solid line.

[0049] It can be understood that the pyrolysis temperature in thermogravimetric analysis data (i.e., the horizontal axis of the DTG curve) can reflect the temperature range of the tobacco product during the pyrolysis process. Different pyrolysis temperatures may correspond to different pyrolysis stages in the pyrolysis reaction. The mass change rate in thermogravimetric analysis data (i.e., the vertical axis of the DTG curve) can reflect the intensity of the pyrolysis reaction of the tobacco product, thereby revealing the kinetic characteristics of the pyrolysis reaction of the tobacco product. In other words, the pyrolysis temperature and mass change rate in the thermogravimetric analysis data of the tobacco product can reflect the pyrolysis characteristics of the tobacco product from two different dimensions.

[0050] The first chemical component data of the tobacco product includes the content of each chemical component of the multiple chemical components of the tobacco product.

[0051] In some embodiments, the plurality of chemical components may include at least one of total nitrogen, nicotine, total sugars, reducing sugars, potassium, and chlorine.

[0052] In some embodiments, the first chemical composition data of the tobacco product can be obtained based on the second chemical composition data of each of the multiple tobacco product raw materials that constitute the tobacco product. For example, a continuous flow method can be used to obtain the content of each of the multiple chemical components in each tobacco product raw material.

[0053] As some implementations, the first chemical component data of the tobacco product may be determined based on an average value of the second chemical component data of a plurality of tobacco product raw materials.

[0054] Continuing with the above example, assuming the chemical composition includes total nitrogen, let's assume the total nitrogen content in the second chemical composition data for five tobacco leaves A is (Q11, Q12, Q13, Q14, Q15), the total nitrogen content in the second chemical composition data for three tobacco leaves B is (Q21, Q22, Q23), and the total nitrogen content in the second chemical composition data for two tobacco leaves C is (Q31, Q32). The total nitrogen content Q in the first chemical composition data for the tobacco product is (Q11+Q12+Q13+Q14+Q15+Q21+Q22+Q23+Q31+Q32) / 10.

[0055] In some embodiments, the first chemical component data of the tobacco product can be obtained by performing chemical component analysis experiments corresponding to multiple chemical components of the tobacco product. For example, a continuous flow method can be used to obtain the content of each of the multiple chemical components in the tobacco product.

[0056] It should be understood that if the tobacco product's formula is known, the first chemical composition data can be obtained using a calculation method similar to the above-mentioned calculation of the first thermogravimetric analysis data. However, if the tobacco product's formula is unknown (i.e., the tobacco product raw materials used and their weight ratios), the chemical composition data can be obtained experimentally to deconstruct the tobacco product's formula. Furthermore, this experimental approach is more accurate than first chemical composition data obtained through calculation, helping to improve the accuracy of subsequent predictions.

[0057] In step 120 , first characteristic data related to a characteristic temperature among the plurality of first pyrolysis temperatures and second characteristic data related to a mass change rate corresponding to the characteristic temperature are determined from the first thermogravimetric analysis data.

[0058] In some embodiments, a portion of characteristic points may be selected from a DTG curve corresponding to the first thermogravimetric analysis data. The values ​​of the abscissas of the portion of characteristic points in the DTG curve are determined as characteristic temperatures in the characteristic data, and the values ​​of the ordinates of the portion of characteristic points in the DTG curve are determined as the mass change rates corresponding to the characteristic temperatures in the characteristic data.

[0059] In step 130, based on at least one of the first feature data and the second feature data and the first chemical composition data, a trained prediction model corresponding to the target origin is used to predict the mass proportion of tobacco product raw materials in tobacco products of the target origin.

[0060] In some embodiments, at least one of the first feature data and the second feature data and the first chemical composition data can be used together as input to a trained prediction model corresponding to the target origin to obtain the mass proportion of tobacco product raw materials in tobacco products of the target origin output by the prediction model.

[0061] For example, the first feature data, the second feature data, and the first chemical composition data can be used together as inputs to a trained prediction model to obtain the mass proportion of tobacco product raw materials in tobacco products from the target origin output by the prediction model.

[0062] In the above embodiment, the characteristic data determined from the thermogravimetric analysis data of the tobacco product includes two types of data capable of characterizing the pyrolysis characteristics of the tobacco product: first characteristic data associated with a characteristic temperature and second characteristic data associated with a rate of mass change corresponding to the characteristic temperature. A trained prediction model corresponding to the target production location is used to process at least one of the two types of data capable of characterizing the pyrolysis characteristics of the tobacco product, as well as the chemical composition data of the tobacco product, to predict the mass percentage of tobacco raw materials in the tobacco product from the target production location.

[0063] In this way, since combining thermogravimetric analysis data and chemical composition data of tobacco products can more comprehensively reflect the origin characteristics of tobacco products, processing these two types of data using a trained prediction model corresponding to the target origin can accurately predict the mass proportion of tobacco product raw materials in tobacco products from the target origin, thereby helping to understand the preservation of the origin style of tobacco product raw materials from different origins. By accurately analyzing the mass proportion of tobacco product raw materials from different origins and their impact on the origin style of tobacco products, we can better understand the changing trends of tobacco product styles, thereby providing a scientific basis for the subsequent design of optimized combination formulas.

[0064] The characteristic data in step 120 is exemplarily described below with reference to some embodiments.

[0065] In some embodiments, the characteristic temperature includes at least one of a starting temperature and an ending temperature in the plurality of first pyrolysis temperatures.

[0066] The first characteristic data determined from the first thermogravimetric analysis data includes at least one of a starting temperature and an ending temperature, and the second characteristic data includes at least one of a mass change rate corresponding to the starting temperature and a mass change rate corresponding to the ending temperature.

[0067] The starting temperature in the plurality of first pyrolysis temperatures can be used to indicate the start of the pyrolysis process, ie, the tobacco product begins to thermally decompose. The ending temperature in the plurality of first pyrolysis temperatures can be used to indicate the end of the pyrolysis process, ie, the tobacco product has completed thermal decomposition.

[0068] For example, see Figure 2 As mentioned above, with a step size of 0.5 K, a total of 801 data between 401.138 K and 801.138 K were taken as the thermogravimetric analysis data of the tobacco leaf to obtain the following Figure 2 The DTG curve is shown by the black solid line. Figure 2 In the example shown, the pyrolysis process starts at 401.138 K and ends at 801.138 K.

[0069] Select the starting point A1 and the ending point A2 from the DTG curve as characteristic points. The horizontal coordinate value of the starting point A1 is the starting temperature TA1, and the vertical coordinate value of the starting point A1 is the mass change rate WA1 corresponding to the starting temperature. The horizontal coordinate value of the ending point A2 is the ending temperature TA2, and the vertical coordinate value of the ending point A2 is the mass change rate WA2 corresponding to the ending temperature.

[0070] For example, any set of data among (TA1, TA2), (WA1, WA2), (TA1, TA2, WA1, WA2) and the first chemical component data of the tobacco product can be input into the trained prediction model corresponding to the target origin to predict the mass proportion of tobacco product raw materials in the tobacco products of the target origin.

[0071] In the above embodiment, since the starting temperature of the pyrolysis process indicates that the tobacco product begins to thermally decompose, and the ending temperature of the pyrolysis process indicates that the tobacco product has completed thermal decomposition, the characteristic data includes the starting temperature and the mass change rate corresponding to the starting temperature, the ending temperature and the mass change rate corresponding to the ending temperature. The thermal stability characteristics of the tobacco product can be effectively used to predict the mass proportion of the tobacco product raw materials from the target origin in the multiple tobacco product raw materials that constitute the tobacco product, thereby further improving the prediction accuracy of the mass proportion of the tobacco product raw materials from the target origin in the tobacco product.

[0072] In some embodiments, the plurality of first pyrolysis temperatures form one or more temperature intervals, and the mass change rate corresponding to the characteristic temperature includes at least one extreme value of the mass change rate of the tobacco product within the one or more temperature intervals.

[0073] The first characteristic data determined from the first thermogravimetric analysis data includes at least one extreme value corresponding to a pyrolysis temperature, and the second characteristic data includes at least one extreme value. Here, the at least one extreme value may include a maximum value and / or a minimum value. For example, if the mass change rate is derived with respect to temperature, the extreme value is the mass change rate corresponding to a derivative of 0. Alternatively, if the pyrolysis temperature and mass change rate are plotted as a curve, the extreme value corresponds to a peak or valley on the curve.

[0074] For example, Figure 2 As shown in the figure, the temperature range formed by the multiple first pyrolysis temperatures has a lower limit of 401.138 K and an upper limit of 801.138 K. Multiple points (B1-B5) with a slope of 0 are selected on the DTG curve as characteristic points. B1 and B4 are trough points on the DTG curve, while B2, B3, and B5 are peak points.

[0075] The vertical coordinate values ​​of these characteristic points (WB1~WB5) are the extreme values ​​of the mass change rate of tobacco products in the temperature range of 401.138 K to 801.138 K, and the horizontal coordinate values ​​of these characteristic points (TB1~TB5) are the temperatures corresponding to the extreme values ​​(i.e., characteristic temperatures).

[0076] For example, any set of data among (TB1~TB5), (WB1~WB5), ((TB1, WB1)~(TB5, WB5)) and the first chemical component data of the tobacco product can be input into the trained prediction model corresponding to the target origin to predict the mass proportion of tobacco product raw materials in the tobacco products of the target origin.

[0077] In the above embodiment, since the peaks and troughs in the curve corresponding to the thermogravimetric analysis data can reflect the main thermal decomposition stages of the tobacco product, the characteristic data includes one or more extreme values ​​of the tobacco product's mass change rate within the temperature range of the pyrolysis process and the corresponding characteristic temperatures. This effectively utilizes the characteristics of the tobacco product at each thermal decomposition stage to predict the mass proportion of tobacco raw materials from the target origin among the multiple tobacco raw materials that make up the tobacco product. This further improves the accuracy of the prediction of the mass proportion of tobacco raw materials from the target origin in the tobacco product.

[0078] In some embodiments, the plurality of first pyrolysis temperatures form adjacent first and second temperature intervals.

[0079] The mass change rate corresponding to the characteristic temperature includes a midpoint between a first extreme value of the mass change rate of the tobacco product within the first temperature interval and a second extreme value of the mass change rate of the tobacco product within the second temperature interval. In other words, the mass change rate corresponding to the characteristic temperature may include a midpoint between two extreme values ​​of the mass change rate of the tobacco product within two adjacent temperature intervals.

[0080] In some embodiments, the characteristic temperature includes a first temperature corresponding to an intermediate value in the first thermogravimetric analysis data, a temperature corresponding to a first extreme value is lower than a temperature corresponding to a second extreme value, and the first temperature is higher than the temperature corresponding to the first extreme value and lower than the temperature corresponding to the second extreme value. In other words, the characteristic temperature corresponding to an intermediate value in the first thermogravimetric analysis data may include a first temperature between the temperature corresponding to the first extreme value and the temperature corresponding to the second extreme value.

[0081] The first characteristic data determined from the first thermogravimetric analysis data includes a first temperature, and the second characteristic data includes an intermediate value.

[0082] By further including the characteristic data in the middle value of the two extreme values ​​of the tobacco product's mass change rate within two adjacent temperature intervals, as well as the first temperature corresponding to the middle value, the characteristics of the tobacco product at each stage of thermal decomposition can be further effectively utilized for prediction. This can further improve the accuracy of the prediction of the mass proportion of tobacco raw materials in tobacco products from the target production area.

[0083] In some embodiments, the characteristic temperature includes a second temperature corresponding to the intermediate value in the first thermogravimetric analysis data. The second temperature is less than the temperature corresponding to the first extreme value or greater than the temperature corresponding to the second extreme value. In other words, the characteristic temperature corresponding to the intermediate value may include a second temperature less than the temperature corresponding to the first extreme value or a second temperature greater than the temperature corresponding to the second extreme value.

[0084] The first characteristic data determined from the first thermogravimetric analysis data may further include a second temperature.

[0085] In this way, considering that the intermediate value may correspond to multiple different temperatures in the thermogravimetric analysis data, further including the intermediate value and the corresponding second temperature, which is different from the first temperature, in the characteristic data can further effectively utilize the characteristics of tobacco product raw materials at various stages of thermal decomposition for prediction. This can further improve the accuracy of the prediction of the mass percentage of tobacco product raw materials in tobacco products from the target production area.

[0086] The following combination Figure 2 The case where the characteristic data includes the middle value of two extreme values ​​of the mass change rate of the tobacco product in two adjacent temperature intervals and the temperature corresponding to the middle value is further described.

[0087] like Figure 2 As shown in the figure, A1 and A2 are the starting point and ending point of the DTG curve respectively. B1 and B4 are the trough points of the DTG curve, and B2, B3, and B5 are the peak points of the DTG curve.

[0088] For example, B1 and B2 are extreme points in two adjacent temperature ranges. The ordinate value of feature point B1 can be taken as the first extreme value WB1, and the ordinate value of feature point B2 can be taken as the second extreme value WB2. Points C1 and D1 corresponding to the intermediate values ​​of the ordinate values ​​of feature point B1 and feature point B2 on the DTG curve are selected as feature points. Among them, the value of the horizontal coordinate of feature point C1 is between the values ​​of the horizontal coordinates corresponding to feature points B1 and B2, that is, the value of the horizontal coordinate of feature point C1 is the first temperature TC1 corresponding to the intermediate value WC1; the value of the horizontal coordinate of feature point D1 is less than the value of the horizontal coordinate corresponding to feature point B1, that is, the value of the horizontal coordinate of feature point D1 is the second temperature TD1 corresponding to the intermediate value WD1. And so on.

[0089] It is understood that the point D1' corresponding to the intermediate value between the ordinate value of the characteristic point B1 and the ordinate value of the characteristic point B2 on the DTG curve may also be selected as the characteristic point. The value of the abscissa of the characteristic point D1' is greater than the value of the abscissa corresponding to the characteristic point B2, that is, the value of the abscissa of the characteristic point D1' is the second temperature TD1' corresponding to the intermediate value WD1'.

[0090] Figure 2 The diagram schematically illustrates eight selected characteristic points (also known as "half-peak points"), C1 to C4 and D1 to D4. Any combination of one or both of the temperature and mass change rate data at these eight characteristic points can be used as feature data input into a trained prediction model corresponding to the target production location to predict the mass percentage of tobacco raw materials in tobacco products from that location.

[0091] For example, any set of data from (TC1~TC4), (WC1~WC4), ((TC1, WC1)~(TC4, WC4)) and the chemical composition data of tobacco products can be input into a trained prediction model corresponding to the target origin to predict the mass proportion of tobacco product raw materials in tobacco products from the target origin.

[0092] For example, any set of data from (TD1~TD4), (WD1~WD4), ((TD1, WD1)~(TD4, WD4)) and the chemical composition data of tobacco products can be input into the trained prediction model corresponding to the target origin to predict the mass proportion of tobacco product raw materials in tobacco products from the target origin.

[0093] For example, any set of data among (TC1~TC4 and TD1~TD4), (WC1~WC4 and WD1~WD4), ((TC1, WC1)~(TC4, WC4), (TD1, WD1)~(TD4, WD4)) and the chemical composition data of tobacco products can be input into the trained prediction model corresponding to the target production area to predict the mass proportion of tobacco product raw materials in tobacco products from the target production area.

[0094] In some embodiments, the first characteristic data determined from the first thermogravimetric analysis data further includes a difference between the second temperature and the first temperature.

[0095] like Figure 2 As shown, Figure 2 The diagram schematically illustrates eight selected characteristic points (also called "half-peak points"), C1 to C4 and D1 to D4. The difference between the first temperature TC1 (i.e., the abscissa value of characteristic point C1) and the second temperature TD1 (i.e., the abscissa value of characteristic point D1) is ∆T1 = TC1 - TD1; the difference between the first temperature TC2 (i.e., the abscissa value of characteristic point C2) and the second temperature TD2 (i.e., the abscissa value of characteristic point D2) is ∆T2 = TC2 - TD2; the difference between the first temperature TC3 (i.e., the abscissa value of characteristic point C3) and the second temperature TD3 (i.e., the abscissa value of characteristic point D3) is ∆T3 = TC3 - TD3; and the difference between the first temperature TC4 (i.e., the abscissa value of characteristic point C4) and the second temperature TD4 (i.e., the abscissa value of characteristic point D4) is ∆T4 = TC4 - TD4. The first characteristic data determined from the first thermogravimetric analysis data may also include ∆T1, ∆T2, ∆T3, and ∆T4.

[0096] In some embodiments, the second characteristic data determined from the first thermogravimetric analysis data includes a difference between at least one extreme value and an intermediate value. In other words, the second characteristic data includes a difference between an intermediate value and either of two extreme values ​​of the mass change rate of the tobacco product corresponding to two adjacent temperature intervals.

[0097] Continue to refer Figure 2 , Figure 2 The diagram schematically illustrates the five selected extreme value points B1 through B5. For example, if the ordinate value of feature point B1 is the first extreme value WB1 and the ordinate value of feature point B2 is the second extreme value WB2, the second feature data can include the difference between the first extreme value WB1 and the intermediate value WC1, ∆W1 = WC1 - WB1. Similarly, the second feature data can also include ∆W2 = WC2 - WB3, ∆W3 = WC3 - WB4, and ∆W4 = WC4 - WB5.

[0098] In the above embodiment, taking into account the different thermal weight loss rates in different temperature sections (i.e., the heterogeneity of thermal weight loss rates), adding the difference between the second temperature and the first temperature and / or the difference between at least one extreme value and the intermediate value as characteristic data of the tobacco product can more accurately quantify the pyrolysis characteristics of the tobacco product. Therefore, prediction based on such characteristic data can further improve the accuracy of the prediction of the mass proportion of tobacco product raw materials in tobacco products from the target production area.

[0099] Table 1 below schematically illustrates the feature data (also referred to as “feature parameters”) obtained based on these feature points.

[0100] Table 1

[0101] Please continue to refer to Figure 2 The curve obtained by curve fitting based on the 15 feature points A1, A2, B1~B5, C1~C4 and D1~D4 is as follows Figure 3 The dashed line indicates this. As can be seen, the curve fitted based on these 15 characteristic points follows a similar trend to the actual DTG curve. This indicates that the 38 characteristic data points derived from these 15 characteristic points can fairly accurately reflect the pyrolysis characteristics of tobacco products during the thermal decomposition process. Therefore, using this characteristic data can help accurately predict the weight percentage of tobacco raw materials in tobacco products from a target production area.

[0102] In some embodiments, the first chemical component data of the tobacco product may further include a ratio of the content of different chemical components among the multiple chemical components in the tobacco product. For example, the multiple chemical components may include total nitrogen, nicotine, total sugar, reducing sugar, potassium, and chlorine. The ratio of the content of different chemical components among the multiple chemical components may include at least one of a potassium-chlorine ratio (i.e., the ratio of potassium to chlorine content), a nitrogen-base ratio (i.e., the ratio of total nitrogen to nicotine content), a sugar-base ratio (i.e., the ratio of total sugar to nicotine content), and a two-sugar ratio (i.e., the ratio of reducing sugar to total sugar content).

[0103] In this way, considering that the content and ratio of various chemical components in tobacco products can also reflect the production characteristics of tobacco products to a certain extent, incorporating this type of data into the input of the prediction model will help further improve the prediction accuracy of the mass proportion of tobacco product raw materials in tobacco products from the target production area.

[0104] It is important to realize that in Figure 2 In the example, the pyrolysis temperatures and the number of characteristic points (characteristic temperatures and corresponding mass change rates) are only exemplary.

[0105] The previous article described how to use a trained prediction model to predict the mass percentage of tobacco raw materials in tobacco products from a target origin. The following section further describes how to train a prediction model for a target origin.

[0106] Figure 3 Flowchart showing a method for training a prediction model according to some embodiments of the present disclosure. Figure 3 As shown, for example, the prediction model training method may include steps 310 to 350.

[0107] Among them, the prediction model is used to predict the mass proportion of tobacco product raw materials in tobacco products from the target production area.

[0108] At step 310, a plurality of combination formulations of tobacco product samples are designed.

[0109] In each of the multiple combination formulas, the mass proportion of the tobacco product raw material samples from the target origin in the tobacco product samples is different.

[0110] In some embodiments, the mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples in multiple combination formulas can be in the form of an arithmetic progression. For example, the mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples can be varied within the range of 10% to 90% at intervals of 10%. The mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples in multiple combination formulas can be 10%, 20%, 30%, 40% ... 90%, respectively.

[0111] In step 320, tobacco product raw material samples from the target origin are selected according to the mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples in each combination formula, and tobacco product raw material samples from other origins are selected according to the mass proportion of tobacco product raw material samples from other origins in the tobacco product samples in each combination formula, so as to form tobacco product raw material samples corresponding to each combination formula.

[0112] In some embodiments, for each combination formula, a Monte Carlo sampling method can be used to select tobacco product raw material samples from the target origin from the tobacco product sample raw material library according to the mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples, and select tobacco product raw material samples from other origins according to the mass proportion of tobacco product raw material samples from other origins in the tobacco product samples, so as to form tobacco product raw material samples corresponding to the combination formula.

[0113] It should be understood that the tobacco product raw material samples corresponding to each combination formula include multiple tobacco product raw material samples from different origins (ie, the target origin and other origins).

[0114] In order to improve the generalization ability of model training, when selecting tobacco product raw material samples, it is possible to consider selecting tobacco product raw material samples covering a variety of different flavor styles. In some embodiments, tobacco product raw material samples from the target production area and / or other production areas can cover the three flavor styles of light fragrance, medium fragrance and strong fragrance. For similar purposes, when selecting tobacco product raw material samples, it is also possible to further consider selecting tobacco product raw material samples covering multiple growth parts for each production area. Here, the growth part of the tobacco product raw material is, for example, the growth part of the tobacco leaves on the tobacco plant, such as the upper part, the middle part or the lower part. In some embodiments, tobacco product raw material samples from the target production area and / or other production areas can include upper tobacco leaves, middle tobacco leaves and lower tobacco leaves.

[0115] In this way, the tobacco product raw material samples corresponding to each combination formula can more comprehensively cover the characteristics of various types of tobacco product raw materials. Consequently, subsequent training of the prediction model using tobacco product samples composed of tobacco product raw material samples corresponding to each combination formula can help the prediction model learn a wider range of characteristics of tobacco product raw materials from different production areas, improving the model's generalization ability and thus contributing to improved prediction performance.

[0116] In step 330 , third thermogravimetric analysis data and third chemical composition data of tobacco product samples consisting of tobacco product raw material samples corresponding to each combination recipe are obtained.

[0117] The third thermogravimetric analysis data includes a plurality of third pyrolysis temperatures of the tobacco product sample during pyrolysis and a mass change rate corresponding to each of the plurality of third pyrolysis temperatures. The third chemical composition data includes the content of each of the plurality of chemical components in the tobacco product sample.

[0118] It can be understood here that the third thermogravimetric analysis data of the tobacco product raw material sample can be obtained by performing a thermogravimetric analysis experiment on the tobacco product raw material sample or can be calculated based on the thermogravimetric analysis data of the tobacco product raw material sample corresponding to each combination formula. The specific implementation is similar to the method of obtaining the first thermogravimetric analysis data of the tobacco product mentioned above. For relevant instructions, please refer to the description in the relevant embodiments mentioned above, and will not be repeated here.

[0119] The third chemical composition data of the tobacco product raw material sample can be obtained by conducting chemical composition analysis experiments corresponding to multiple chemical components of the tobacco product raw material sample, or can be obtained by calculating the chemical composition data of the tobacco product raw material sample corresponding to each combination formula. The specific implementation is similar to the method of obtaining the first chemical composition data of the tobacco product mentioned above. For relevant instructions, please refer to the description in the relevant embodiments mentioned above, and will not be repeated here.

[0120] In step 340 , third characteristic data related to the sample characteristic temperature among the plurality of third pyrolysis temperatures and fourth characteristic data related to the mass change rate corresponding to the sample characteristic temperature are determined from the third thermogravimetric analysis data.

[0121] The method of selecting characteristic data of tobacco product samples in the process of training the prediction model is similar to the method of selecting characteristic data of tobacco products in the process of applying the prediction model for prediction, and the specific description can be found in the above-mentioned related embodiments.

[0122] In step 350, the prediction model is trained with at least one of the third characteristic data and the fourth characteristic data and the fourth chemical composition data as input and the mass proportion of the tobacco product raw material sample of the target origin in the tobacco product sample as output until the training end condition is met.

[0123] The training termination condition may be, for example, that the training has been completed a specified number of times and / or that the difference between the predicted value of the mass percentage of tobacco product raw material samples from the target origin in the tobacco product samples and the actual value of the mass percentage of tobacco product raw material samples from the target origin in the tobacco product samples is less than a threshold. For example, the difference between the predicted value and the actual value of the mass percentage can be represented by the loss function value of the prediction model.

[0124] It can be understood here that the type of data input into the prediction model during the process of training the prediction model is consistent with the type of data input into the prediction model during the process of applying the prediction model to make predictions.

[0125] In some embodiments, the prediction model may include a convolutional neural network model.

[0126] Figure 4 A schematic diagram of the structure of a convolutional neural network model according to some embodiments of the present disclosure is shown.

[0127] like Figure 4 As shown in Figure 1, the convolutional neural network model is mainly composed of convolutional layers, pooling layers, and fully connected layers. The input layer receives feature data representing tobacco product samples.

[0128] For example, the input feature dimension can be batch size × 1 × 48. The third feature data, fourth feature data, and fourth chemical composition data are taken as input. After the input third feature data, fourth feature data, and fourth chemical composition data are calculated through the first convolution layer (for example, a one-dimensional convolution kernel of size 3), the feature dimension becomes batch size × 16 × 48. After the second convolution layer (for example, a one-dimensional convolution kernel of size 3), the feature dimension remains unchanged. The multi-dimensional feature vector is then flattened into a one-dimensional feature vector and input into the fully connected layer. The ratio data output by the fully connected layer is used as the predicted value of the mass proportion of the tobacco product raw material sample in the tobacco product sample from the target origin.

[0129] Here, the batch size can be set based on the hardware resources of the convolutional neural network model. For example, if the memory is large, a larger batch size can be set to increase the training speed.

[0130] In some embodiments, according to the mass proportion of tobacco product raw material samples from other origins in the tobacco product samples in each combination formula, at least one tobacco product raw material is randomly selected from the tobacco product raw materials from multiple origins other than the target origin in the tobacco product raw material library as the tobacco product raw material sample from other origins.

[0131] That is, tobacco product raw material samples from other production areas can be selected from multiple production areas other than the target production area.

[0132] For example, taking the example that the mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples is 90%, and the mass proportion of tobacco product raw material samples from other origins in the tobacco product samples is 10% (hereinafter referred to as "target origin: other origins = 90%: 10%"), one or more tobacco product raw material samples from the target origin (such as Yunnan) can be selected from the tobacco product sample raw material library, and one or more tobacco product raw material samples from multiple origins other than the target origin (such as Fujian, Hunan, and Henan) can be selected as tobacco product raw material samples from other origins, thereby forming tobacco product raw material samples corresponding to the combined formula of "target origin: other origins = 90%: 10%".

[0133] In the above embodiment, multiple combination formulas are obtained by designing different proportions between the target origin and other origins, and for each combination formula, tobacco product raw material samples from the target origin and other origins are accurately selected according to the designed proportions, thereby forming tobacco product samples with specific origin proportion characteristics.

[0134] The tobacco product samples obtained in this way can effectively simulate the situation after tobacco product raw materials from different origins are mixed during the tobacco production process, providing a reliable experimental basis for studying the interaction between tobacco product raw materials from different origins in the combined formula and its impact on the style of the final product. The trained prediction model can accurately predict the mass proportion of tobacco product raw materials from the predicted target origin in the tobacco products, providing a scientific basis for understanding the changes in the style of tobacco products caused by the origin, and helping to optimize the design of the combined formula in the future.

[0135] Figure 5 Flowchart showing a method for training a prediction model according to other embodiments of the present disclosure. Figure 5 As shown, for example, the training method of the prediction model may include steps 510 to 580. The method may be used as Figure 3 A specific implementation of the method is executed.

[0136] In step 510, tobacco product raw material samples are selected. For example, tobacco product raw material samples are selected from eight ecological zones according to light aroma, medium aroma, and strong aroma. Each production zone contains at least 30 different tobacco product raw material samples, and the selected tobacco product raw material samples include upper tobacco leaves, middle tobacco leaves, and lower tobacco leaves.

[0137] In step 520, chemical composition data for each tobacco product raw material sample is obtained.

[0138] For example, the chemical composition data of each tobacco product raw material sample obtained by the continuous flow method may include total nitrogen, nicotine, total sugar, reducing sugar, potassium, chlorine, potassium-chlorine ratio, nitrogen-alkali ratio, sugar-alkali ratio, and two-sugar ratio.

[0139] In step 530, a pyrolysis experiment is performed on each tobacco product raw material sample to obtain thermogravimetric analysis data of each tobacco product raw material sample.

[0140] In step 540, multiple combination formulas of tobacco product samples are designed according to the proportional gradient principle. Here, the proportional gradient principle means that the combination formulas are designed according to the production area combination of "target production area: other production areas", and the weight proportions of the tobacco product raw material samples from the target production area in the tobacco product samples in the multiple combination formulas are in the form of an arithmetic progression.

[0141] In step 550, tobacco product raw material samples from the target origin are selected according to the mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples in each combination formula, and tobacco product raw material samples from other origins are selected according to the mass proportion of tobacco product raw material samples from other origins in the tobacco product samples in each combination formula, so as to form tobacco product raw material samples corresponding to each combination formula.

[0142] In step 560, third thermogravimetric analysis data and third chemical composition data of the tobacco product samples consisting of the tobacco product raw material samples corresponding to each combination recipe are obtained.

[0143] For example, the third thermogravimetric analysis data can be calculated based on the thermogravimetric analysis data of each tobacco product raw material sample obtained in step 530, and the third chemical composition data can be calculated based on the chemical composition data of each tobacco product raw material sample obtained in step 520.

[0144] In step 570, third characteristic data related to the sample characteristic temperature among the plurality of third pyrolysis temperatures and fourth characteristic data related to the mass change rate corresponding to the sample characteristic temperature are determined from the third thermogravimetric analysis data to obtain a training data set.

[0145] In step 580, the convolutional neural network model is trained based on the training data set to obtain a prediction model corresponding to the target origin (also known as an origin ratio prediction model).

[0146] It is understood that if it is necessary to predict the mass proportion of tobacco product raw materials in tobacco products from each of multiple production areas, the Figures 3 to 5 The illustrated training method trains a prediction model corresponding to each of a plurality of origins.

[0147] For example, multiple prediction models can be trained, such as a prediction model for predicting the mass proportion of tobacco product raw materials in tobacco products in Yunnan, a prediction model for predicting the mass proportion of tobacco product raw materials in tobacco products in Fujian, and a prediction model for predicting the mass proportion of tobacco product raw materials in tobacco products in Hunan, so as to obtain the mass proportion of tobacco product raw materials in tobacco products from different origins.

[0148] For more embodiments of the training method of the prediction model provided in the embodiments of the present disclosure, reference may be made to the above description of the method for predicting the origin ratio of tobacco products and its embodiments.

[0149] Next, some embodiments are used to introduce related applications of the above-mentioned method for predicting the ratio of tobacco products by origin. For example, the above-mentioned method for predicting the ratio of tobacco products by origin can provide a basis for replacing the combination formula of tobacco products.

[0150] Figure 6 A flow chart illustrating an alternative method for combining formulations of tobacco products according to some embodiments of the present disclosure. Figure 6 As shown, the alternative method for the combined formulation of tobacco products includes steps 610 to 640.

[0151] In step 610 , for each candidate combination formula among a plurality of candidate combination formulas, first thermogravimetric analysis data and first chemical composition data of a candidate tobacco product corresponding to each candidate combination formula are obtained.

[0152] Here, the first thermogravimetric analysis data includes a plurality of first pyrolysis temperatures of the candidate tobacco product during pyrolysis and a mass change rate of the candidate tobacco product corresponding to each of the plurality of first pyrolysis temperatures. The first chemical composition data includes the content of each of the plurality of chemical components of the candidate tobacco product.

[0153] At step 620 , first characteristic data related to a characteristic temperature among the plurality of first pyrolysis temperatures and second characteristic data related to a mass change rate corresponding to the characteristic temperature are determined from the first thermogravimetric analysis data.

[0154] In step 630, based on at least one of the first feature data and the second feature data and the first chemical composition data, a trained prediction model corresponding to the target origin is used to predict the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco products.

[0155] It can be understood that steps 610 to 630 are the same as those in the previous Figure 1 The steps 110 to 130 shown in FIG are similar, and can be referred to in the previous text. Figure 1 The method of the related embodiment is similar to that of the embodiment of the present invention, and the specific description can be found in the related embodiment above. No further details will be given here.

[0156] In step 640, a candidate combination formula in which the difference between the mass proportion of the tobacco product raw materials from the target origin in the candidate tobacco product and the mass proportion of the tobacco product raw materials from the target origin in the known tobacco product corresponding to the known combination formula to be replaced is less than a specified threshold is determined as the target combination formula for replacing the known combination formula.

[0157] Taking Yunnan as an example, assuming that the mass proportion of tobacco product raw materials from the target origin in the known tobacco product to be replaced is 50% in the known combination formula to be replaced, the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco product can be predicted by implementing steps 610 to 630. Candidate combination formulas for which the difference between the predicted mass proportion of tobacco product raw materials from the target origin in the candidate tobacco product and 50% is less than a specified threshold (e.g., the difference from 50% is less than 5%) are determined as target combination formulas.

[0158] In the above embodiment, it is considered that during the tobacco production process, due to factors such as insufficient inventory of some tobacco product raw materials, it may be necessary to find alternative tobacco product raw materials. In this case, in order to ensure that the tobacco products made based on the replaced combination formula can be as consistent as possible with the tobacco products to be replaced in terms of production style, the prediction model proposed in this disclosure can be used to accurately predict the mass proportion of tobacco product raw materials from the target production area (i.e., the production area of ​​the insufficient inventory of tobacco product raw materials) in the final tobacco product for each candidate combination formula. This prediction can not only help the manufacturer quickly evaluate the feasibility of the replacement plan, but also provide a scientific basis for optimizing the formula, thereby effectively improving the scientific nature and accuracy of combination formula replacement in the case of raw material shortages.

[0159] The following, in combination with some embodiments, exemplifies the effect of using the prediction model corresponding to the target origin trained by the training method of the present invention and using the prediction model corresponding to the target origin to predict the mass proportion of tobacco product raw materials in tobacco products from the target origin.

[0160] 434 tobacco product raw material samples from 8 production areas, namely, Production A, Production B, Production C, Production D, Production E, Production F, Production G, and Production H, were selected.

[0161] A continuous flow method was used to measure 10 chemical components of each of the 434 tobacco product raw material samples. For example, these 10 chemical components include total nitrogen, nicotine, total sugars, reducing sugars, potassium, chlorine, potassium-chlorine ratio, nitrogen-alkali ratio, sugar-alkali ratio, and di-sugar ratio.

[0162] For each of the 434 tobacco product raw material samples, corresponding tobacco powder was selected for pyrolysis experiments to obtain thermogravimetric analysis data for each tobacco product raw material sample.

[0163] Multiple combination formulas of tobacco product samples are designed according to the proportional gradient principle. Furthermore, for each combination formula, a Monte Carlo sampling method is used to select tobacco product raw material samples from the tobacco product sample raw material library based on the mass proportion of tobacco product raw material samples from the target production area in the tobacco product samples, and tobacco product raw material samples from other production areas are selected based on the mass proportion of tobacco product raw material samples from other production areas in the tobacco product samples, thereby forming the tobacco product raw material samples corresponding to the combination formula.

[0164] Thermogravimetric analysis data and chemical composition data for tobacco product samples, consisting of tobacco product raw material samples corresponding to each combination formula, are obtained. Feature data related to characteristic temperatures among multiple pyrolysis temperatures, as well as feature data related to the mass change rate corresponding to the characteristic temperatures, are extracted from these data. These two feature data, along with the chemical composition data, are input into a convolutional neural network for training to generate a prediction model corresponding to the target production area.

[0165] For example, for the origin combination of "target origin: other origins", 10 proportion relationships are designed, including 90%:10%, 80%:20%, 70%:30%, 60%:40%, 50%:50%, 40%:60%, 30%:70%, 20%:80%, 10%:90%, and 0%:100%.

[0166] Taking Origin A as the target origin, Origin B, Origin C, Origin D, Origin E, Origin F, Origin G, and Origin H are all considered "other origins." Taking the mass proportion of tobacco product raw material samples from Origin A in tobacco product samples as the prediction target, a prediction model corresponding to Origin A can be trained. Similarly, prediction models corresponding to Origin B, Origin C, Origin D, Origin E, Origin F, Origin G, and Origin H can be trained separately to obtain eight prediction models.

[0167] The eight trained prediction models were used to predict tobacco product samples composed of tobacco product raw material samples corresponding to a certain combination formula. The results are shown in Table 2.

[0168] Table 2

[0169] As shown in Table 2, the prediction errors for the mass fraction of tobacco product raw materials across the eight production areas using the eight prediction models were all within ±10%. The prediction accuracy for Region C was 75.83%, and for Region F was 75.48%, both of which were relatively low. In contrast, the prediction accuracy for Region B was 90.74%, and for Region H was 89.68%, both of which were relatively high. Furthermore, the average accuracy for the predictions of the mass fraction of tobacco product raw materials across the eight production areas using the eight prediction models reached 82.17%.

[0170] It can be seen that tobacco raw materials from different origins retain different levels of their own regional characteristics when combined into tobacco products based on a combination formula. Using the training method proposed in this disclosure, a prediction model corresponding to a target origin can accurately predict the mass percentage of tobacco raw materials from the target origin in tobacco products, thereby providing a theoretical basis for the subsequent optimization of the combination formula design.

[0171] Figure 7 A block diagram illustrating prediction of the origin ratio of tobacco products according to some embodiments of the present disclosure.

[0172] like Figure 7 As shown, the tobacco product origin ratio prediction device 700 includes an acquisition module 701 , a determination module 702 and a prediction module 703 .

[0173] The acquisition module 701 can be configured to acquire first thermogravimetric analysis data and first chemical composition data of the tobacco product. Here, the first thermogravimetric analysis data includes a plurality of first pyrolysis temperatures during the pyrolysis process of the tobacco product and a mass change rate of the tobacco product corresponding to each of the plurality of first pyrolysis temperatures. The first chemical composition data includes the content of each of the plurality of chemical components of the tobacco product.

[0174] The determination module 702 may be configured to determine, from the first thermogravimetric analysis data, first characteristic data associated with a characteristic temperature among the plurality of first pyrolysis temperatures and second characteristic data associated with a mass change rate corresponding to the characteristic temperature.

[0175] The prediction module 703 can be configured to predict the mass proportion of tobacco product raw materials in tobacco products from the target origin based on at least one of the first feature data and the second feature data and the first chemical composition data using a trained prediction model corresponding to the target origin.

[0176] In some embodiments, the tobacco product origin ratio prediction device 700 may further include executing the above Figures 1 to 5 Other modules for other operations in the illustrated embodiments.

[0177] Figure 8 A block diagram illustrating an alternative apparatus for combining formulations of tobacco products according to some embodiments of the present disclosure.

[0178] like Figure 8 As shown, the device 800 for replacing the combination formula of tobacco products includes an acquisition module 801 , a first determination module 802 , a prediction module 803 and a second determination module 804 .

[0179] Acquisition module 801 can be configured to acquire, for each candidate combination formula in a plurality of candidate combination formulas, first thermogravimetric analysis data and first chemical composition data of a candidate tobacco product corresponding to each candidate combination formula. Here, the first thermogravimetric analysis data includes a plurality of first pyrolysis temperatures during the pyrolysis process of the candidate tobacco product and a mass change rate of the candidate tobacco product corresponding to each of the plurality of first pyrolysis temperatures. The first chemical composition data includes the content of each of the plurality of chemical components in the candidate tobacco product.

[0180] The first determination module 802 may be configured to determine, from the first thermogravimetric analysis data, first characteristic data associated with a characteristic temperature among the plurality of first pyrolysis temperatures and second characteristic data associated with a mass change rate corresponding to the characteristic temperature.

[0181] The prediction module 803 can be configured to predict the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco products based on at least one of the first feature data and the second feature data and the first chemical composition data using a trained prediction model corresponding to the target origin.

[0182] The second determination module 804 can be configured to determine a candidate combination formula, whose difference between the mass proportion of tobacco product raw materials from the target origin in the candidate tobacco product and the mass proportion of tobacco product raw materials from the target origin in the known tobacco product corresponding to the known combination formula to be replaced is less than a specified threshold, as a target combination formula for replacing the known combination formula.

[0183] In some embodiments, the device 800 for replacing the combination of tobacco products may further include the following steps: Figure 6 Other modules for other operations in the illustrated embodiments.

[0184] Figure 9 A block diagram of an electronic device according to some embodiments of the present disclosure is shown.

[0185] like Figure 9 As shown, the electronic device 900 of this embodiment includes: a memory 901 and a processor 902 coupled to the memory 901, and the processor 902 is configured to execute the method for predicting the origin ratio of tobacco products or the method for replacing the combination formula of tobacco products in any embodiment of the present disclosure based on instructions stored in the memory 901.

[0186] The memory 901 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, a database, and other programs.

[0187] Figure 10 A block diagram of an electronic device according to some other embodiments of the present disclosure is shown.

[0188] like Figure 10 As shown, the electronic device 1000 of this embodiment includes: a memory 1001 and a processor 1002 coupled to the memory 1001, and the processor 1002 is configured to execute the method for predicting the origin ratio of tobacco products or the method for replacing the combination formula of tobacco products in any of the aforementioned embodiments based on instructions stored in the memory 1001.

[0189] The memory 1001 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, application programs, a boot loader, and other programs.

[0190] Electronic device 1000 may also include input / output interfaces 1003, a network interface 1004, a storage interface 1005, and the like. These interfaces 1003, 1004, and 1005, as well as the memory 1001 and the processor 1002, may be connected, for example, via a bus 1006. The input / output interface 1003 provides a connection interface for input / output devices such as a display, mouse, keyboard, touch screen, microphone, and speakers. The network interface 1004 provides a connection interface for various networked devices. The storage interface 1005 provides a connection interface for external storage devices such as SD cards and USB flash drives.

[0191] The embodiments of the present disclosure also provide a computer-readable storage medium comprising computer program instructions, which, when executed by a processor, implement the method for predicting the origin ratio of tobacco products or the method for replacing the combined formula of tobacco products in any of the above embodiments.

[0192] The embodiments of the present disclosure also provide a computer program product, including a computer program, which, when executed by a processor, implements the method for predicting the origin ratio of tobacco products or the method for replacing the combined formula of tobacco products in any of the above embodiments.

[0193] Those skilled in the art will appreciate that embodiments of the present disclosure may be provided as methods, systems, or computer program products. Thus, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable, non-transitory storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0194] Thus far, the method for predicting the origin ratio of tobacco products and the replacement method for the combined formulation according to the present disclosure have been described in detail. To avoid obscuring the concept of the present disclosure, some details well known in the art have not been described. Based on the above description, those skilled in the art can fully understand how to implement the technical solutions disclosed herein.

[0195] The methods and systems of the present disclosure may be implemented in many ways. For example, the methods and systems of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present disclosure may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers recording media that store programs for executing the methods according to the present disclosure.

[0196] Although some specific embodiments of the present disclosure have been described in detail by way of examples, those skilled in the art will appreciate that the above examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Those skilled in the art will appreciate that modifications may be made to the above embodiments without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A method for predicting the origin ratio of tobacco products, comprising: Obtaining first thermogravimetric analysis data and first chemical composition data of a tobacco product, wherein the first thermogravimetric analysis data includes a plurality of first pyrolysis temperatures during a pyrolysis process of the tobacco product and a mass change rate of the tobacco product corresponding to each of the plurality of first pyrolysis temperatures, and the first chemical composition data includes a content of each of a plurality of chemical components of the tobacco product; determining, from the first thermogravimetric analysis data, first characteristic data associated with a characteristic temperature among the plurality of first pyrolysis temperatures, and second characteristic data associated with a mass change rate corresponding to the characteristic temperature; Based on at least one of the first feature data and the second feature data and the first chemical composition data, a trained prediction model corresponding to the target origin is used to predict the mass proportion of tobacco product raw materials from the target origin in the tobacco products.

2. The method for predicting the production area ratio according to claim 1, wherein: The plurality of first pyrolysis temperatures form one or more temperature intervals, and the mass change rate corresponding to the characteristic temperature includes at least one extreme value of the mass change rate of the tobacco product within the one or more temperature intervals. The first characteristic data includes the pyrolysis temperature corresponding to the at least one extreme value, and the second characteristic data includes the at least one extreme value.

3. The method for predicting the production area ratio according to claim 1, wherein: The plurality of first pyrolysis temperatures form adjacent first and second temperature intervals, The mass change rate corresponding to the characteristic temperature includes an intermediate value between a first extreme value of the mass change rate of the tobacco product in the first temperature range and a second extreme value of the mass change rate of the tobacco product in the second temperature range. The characteristic temperature includes a first temperature corresponding to the intermediate value in the first thermogravimetric analysis data, a temperature corresponding to the first extreme value is lower than a temperature corresponding to the second extreme value, and the first temperature is higher than a temperature corresponding to the first extreme value and lower than a temperature corresponding to the second extreme value. The first characteristic data includes the first temperature, and the second characteristic data includes the intermediate value.

4. The method for predicting the production area ratio according to claim 3, wherein: The characteristic temperature includes a second temperature corresponding to the intermediate value in the first thermogravimetric analysis data, the second temperature being lower than the temperature corresponding to the first extreme value or higher than the temperature corresponding to the second extreme value, The first characteristic data further includes the second temperature.

5. The method for predicting the production area ratio according to claim 4, wherein: The first characteristic data further includes a difference between the second temperature and the first temperature.

6. The method for predicting the production area ratio according to claim 3, wherein: The mass change rate corresponding to the characteristic temperature includes at least one extreme value of the mass change rate of the tobacco product in the first temperature range and the second temperature range. The second characteristic data includes a difference between the at least one extreme value and the intermediate value.

7. The method for predicting the origin ratio according to any one of claims 1 to 6, wherein: The characteristic temperature includes at least one of a starting temperature and an ending temperature among the plurality of first pyrolysis temperatures, The first characteristic data includes at least one of the starting temperature and the ending temperature, The second characteristic data includes at least one of a mass change rate corresponding to the starting temperature and a mass change rate corresponding to the ending temperature.

8. The method for predicting the origin ratio according to any one of claims 1 to 6, wherein: The first chemical component data also includes ratios of the contents of different chemical components among the multiple chemical components in the tobacco product.

9. The method for predicting the origin ratio according to any one of claims 1 to 6, wherein: The tobacco product comprises a combination of multiple tobacco product raw materials, and obtaining first thermogravimetric analysis data and first chemical composition data of the tobacco product comprises: Obtaining second thermogravimetric analysis data and second chemical composition data for each of the plurality of tobacco product raw materials, the second thermogravimetric analysis data including a plurality of second pyrolysis temperatures during the pyrolysis process of each tobacco product raw material and a mass change rate of each tobacco product raw material corresponding to each of the plurality of second pyrolysis temperatures, and the second chemical composition data including the content of each chemical component among the plurality of chemical components of each tobacco product raw material; determining the first thermogravimetric analysis data according to an average value of the second thermogravimetric analysis data; The first chemical composition data is determined according to an average value of the second chemical composition data.

10. The method for predicting the origin ratio according to any one of claims 1 to 6, wherein: The tobacco product comprises a combination of multiple tobacco product raw materials, and obtaining first thermogravimetric analysis data and first chemical composition data of the tobacco product comprises: performing a thermogravimetric analysis experiment on the tobacco product to obtain the first thermogravimetric analysis data; and A chemical component analysis experiment corresponding to the multiple chemical components is performed on the tobacco product to obtain the first chemical component data.

11. The method for predicting the origin ratio according to any one of claims 1 to 6, wherein: The trained prediction model corresponding to the target origin is trained as follows: designing multiple combination formulas of the tobacco product sample, wherein the weight proportion of the tobacco product raw material sample from the target production area in the tobacco product sample is different in each combination formula of the multiple combination formulas; Selecting tobacco product raw material samples from the target origin according to the mass proportion of tobacco product raw material samples from the target origin in the tobacco product samples in each combination formula, and selecting tobacco product raw material samples from other origins according to the mass proportion of tobacco product raw material samples from other origins in the tobacco product samples in each combination formula, so as to form tobacco product raw material samples corresponding to each combination formula; Obtaining third thermogravimetric analysis data and third chemical composition data of the tobacco product sample composed of the tobacco product raw material sample corresponding to each combination formula, the third thermogravimetric analysis data including a plurality of third pyrolysis temperatures during the pyrolysis process of the tobacco product sample and a mass change rate corresponding to each of the plurality of third pyrolysis temperatures, and the third chemical composition data including the content of each of the plurality of chemical components in the tobacco product sample; determining, from the third thermogravimetric analysis data, third characteristic data associated with a sample characteristic temperature among the plurality of third pyrolysis temperatures, and fourth characteristic data associated with a mass change rate corresponding to the sample characteristic temperature; The prediction model is trained with at least one of the third characteristic data and the fourth characteristic data and the fourth chemical composition data as input and the mass proportion of the tobacco product raw material sample from the target origin in the tobacco product sample as output until the training end condition is met.

12. The method for predicting the production area ratio according to claim 11, wherein: The selecting of the tobacco product raw material samples from other origins according to the mass proportion of the tobacco product raw material samples from other origins in each combination formula in the tobacco product samples includes: According to the mass proportion of the tobacco product raw material samples from other origins in the tobacco product samples in each combination formula, at least one tobacco product raw material is randomly selected from the tobacco product raw materials from multiple origins other than the target origin in the tobacco product raw material library as the tobacco product raw material sample from other origins.

13. The method for predicting the production area ratio according to claim 11, wherein: The prediction model includes a convolutional neural network model.

14. The method for predicting the origin ratio according to any one of claims 1 to 6, wherein: The plurality of chemical components include at least one of total nitrogen, nicotine, total sugar, reducing sugar, potassium and chlorine.

15. A method for replacing a combination formulation of a tobacco product, comprising: For each candidate combination formula among a plurality of candidate combination formulas, obtaining first thermogravimetric analysis data and first chemical composition data of a candidate tobacco product corresponding to each candidate combination formula, the first thermogravimetric analysis data including a plurality of first pyrolysis temperatures during a pyrolysis process of the candidate tobacco product and a mass change rate of the candidate tobacco product corresponding to each of the plurality of first pyrolysis temperatures, and the first chemical composition data including the content of each of a plurality of chemical components of the candidate tobacco product; determining, from the first thermogravimetric analysis data, first characteristic data associated with a characteristic temperature among the plurality of first pyrolysis temperatures, and second characteristic data associated with a mass change rate corresponding to the characteristic temperature; Based on at least one of the first feature data and the second feature data and the first chemical composition data, using a trained prediction model corresponding to a target production area, predicting a mass proportion of tobacco product raw materials from the target production area in the candidate tobacco products; A candidate combination formula whose difference between the mass proportion of the tobacco product raw materials from the target origin in the candidate tobacco product and the mass proportion of the tobacco product raw materials from the target origin in the known tobacco product corresponding to the known combination formula to be replaced is less than a specified threshold is determined as the target combination formula for replacing the known combination formula.

16. A device for predicting the origin ratio of tobacco products, comprising: an acquisition module configured to acquire first thermogravimetric analysis data and first chemical composition data of the tobacco product, wherein the first thermogravimetric analysis data includes a plurality of first pyrolysis temperatures during a pyrolysis process of the tobacco product and a mass change rate of the tobacco product corresponding to each of the plurality of first pyrolysis temperatures, and the first chemical composition data includes a content of each of a plurality of chemical components of the tobacco product; a determination module configured to determine, from the first thermogravimetric analysis data, first characteristic data associated with a characteristic temperature among the plurality of first pyrolysis temperatures, and second characteristic data associated with a mass change rate corresponding to the characteristic temperature; The prediction module is configured to predict the mass proportion of tobacco product raw materials from the target origin in the tobacco products based on at least one of the first feature data and the second feature data and the first chemical composition data using a trained prediction model corresponding to the target origin.

17. A device for replacing a combination of tobacco products, comprising: an acquisition module configured to acquire, for each candidate combination formula in a plurality of candidate combination formulas, first thermogravimetric analysis data and first chemical composition data of a candidate tobacco product corresponding to the candidate combination formula, the first thermogravimetric analysis data including a plurality of first pyrolysis temperatures during a pyrolysis process of the candidate tobacco product and a mass change rate of the candidate tobacco product corresponding to each of the plurality of first pyrolysis temperatures, and the first chemical composition data including a content of each of a plurality of chemical components of the candidate tobacco product; a first determining module configured to determine, from the first thermogravimetric analysis data, first characteristic data related to a characteristic temperature among the plurality of first pyrolysis temperatures, and second characteristic data related to a mass change rate corresponding to the characteristic temperature; a prediction module configured to predict, based on at least one of the first feature data and the second feature data and the first chemical composition data, a mass proportion of tobacco product raw materials from the target production area in the candidate tobacco product using a trained prediction model corresponding to the target production area; The second determination module is configured to determine, as a target combination formula for replacing the known combination formula, a candidate combination formula in which the difference between the mass proportion of the tobacco product raw materials from the target origin in the candidate tobacco product and the mass proportion of the tobacco product raw materials from the target origin in the known tobacco product corresponding to the known combination formula to be replaced is less than a specified threshold.

18. An electronic device comprising: Memory; and A processor coupled to the memory, the processor being configured to execute the method for predicting the origin ratio of tobacco products according to any one of claims 1 to 14 or the method for replacing the combination formulation of tobacco products according to claim 15 based on instructions stored in the memory.

19. A computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the method for predicting the origin ratio of tobacco products according to any one of claims 1 to 14 or the method for replacing the combined formulation of tobacco products according to claim 15.

20. A computer program product comprising instructions, which, when executed by a processor, cause the processor to execute the method for predicting the origin ratio of tobacco products according to any one of claims 1 to 14 or the method for replacing the combined formulation of tobacco products according to claim 15.

Images