Processing scheme recommendation method and device and computer readable storage medium

Abstract

The invention relates to a processing scheme recommendation method and device and a computer readable storage medium, and relates to the technical field of tobacco production. The method comprises the following steps: constructing a combined formula-processing scheme database according to each candidate combined formula in a plurality of candidate combined formulas; obtaining and screening out a plurality of first combined formulas according to multi-dimensional characteristics of candidate tobacco products corresponding to each candidate combined formula; determining a processing scheme corresponding to each first combined formula in the plurality of first combined formulas; storing each first combined formula and the processing scheme corresponding to each first combined formula into a combined formula-processing scheme database; and searching a first target combination formula matched with the to-be-processed second combination formula from the combination formula-processing scheme database, and recommending a corresponding target processing scheme for the second combination formula according to the processing scheme corresponding to the first target combination formula. Therefore, an appropriate processing scheme can be efficiently and reliably recommended for the combined formula at low cost.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of tobacco production, and in particular, to a processing scheme recommendation method and device, and a computer readable storage medium. BACKGROUND

[0002] Group processing of tobacco products is a process technology that performs targeted processing according to different combination formulas of tobacco products.

[0003] In the related art, for a specific combination formula, a corresponding processing scheme is determined by relying on the experience of process personnel. SUMMARY

[0004] The present inventors have found that the above related art has the following problem: it is difficult to quickly and reliably determine a corresponding processing scheme for a specific combination formula.

[0005] To solve the above problem, the embodiments of the present disclosure provide the following solutions.

[0006] According to some embodiments of the present disclosure, a processing scheme recommendation method is provided, including: according to each candidate combination formula in a plurality of candidate combination formulas, performing the following steps to construct a combination formula-processing scheme database: obtaining a multi-dimensional feature of a candidate tobacco product corresponding to each candidate combination formula, the multi-dimensional feature including a plurality of first volatile fraction scores, first fiber class scores, first chemical component data, and origin styles, the first chemical component data including the content of each chemical component in a plurality of chemical components of the candidate tobacco product; according to the multi-dimensional feature of the candidate tobacco product, screening a plurality of first combination formulas from the plurality of candidate combination formulas; determining a processing scheme corresponding to each first combination formula in the plurality of first combination formulas, the processing scheme including a processing path and a processing intensity; storing each first combination formula and the processing scheme corresponding to each first combination formula to the combination formula-processing scheme database; retrieving a first target combination formula matching a second combination formula to be processed from the combination formula-processing scheme database, and recommending a corresponding target processing scheme for the second combination formula according to the processing scheme corresponding to the first target combination formula.

[0007] In some embodiments, the multi-dimensional features include at least one of the first volatile component score, the first fiber class score, and the first chemical composition data and the origin style, and the filtering the plurality of candidate combination recipes according to the multi-dimensional features of the candidate tobacco products comprises: filtering a plurality of groups of first tobacco products presenting a plurality of target origin styles from a plurality of candidate tobacco products corresponding to the plurality of candidate combination recipes, different groups of tobacco products present different target origin styles, each group of first tobacco products in the plurality of groups of first tobacco products comprises a plurality of first tobacco products; filtering a plurality of second tobacco products from each group of first tobacco products according to at least one of the first volatile component score, the first fiber class score, and the first chemical composition data of each first tobacco product in the plurality of first tobacco products in the each group of first tobacco products; and determining the plurality of first combination recipes according to a plurality of combination recipes corresponding to the plurality of second tobacco products.

[0008] In some embodiments, the filtering a plurality of second tobacco products from each group of first tobacco products according to at least one of the first volatile component score, the first fiber class score, and the first chemical composition data of each first tobacco product in the plurality of first tobacco products in the each group of first tobacco products to obtain a plurality of groups of second tobacco products comprises: filtering the plurality of second tobacco products from the each group of first tobacco products according to at least one of a first difference between the first volatile component scores, a second difference between the first fiber class scores, and a third difference between the first chemical composition data of different first tobacco products in the each group of first tobacco products.

[0009] In some embodiments, each group of first tobacco products comprises a plurality of subgroups, and first tobacco products in different subgroups of the each group of first tobacco products present different probabilities of presenting the target origin style corresponding to the each group of first tobacco products.

[0010] In some embodiments, the determining the processing scheme corresponding to each first combination recipe in the plurality of first combination recipes comprises: performing a first processing experiment on a tobacco product corresponding to the each first combination recipe according to different processing paths, and obtaining sensory evaluation results of the tobacco product after the first processing experiment; and determining the processing path corresponding to the each first combination recipe according to the sensory evaluation results of the tobacco product after the first processing experiment.

[0011] In some embodiments, the determining the processing scheme corresponding to each first combined formula of the plurality of first combined formulas comprises: performing a second processing experiment on the tobacco product corresponding to each first combined formula according to different processing intensities of a same processing path, and obtaining sensory evaluation results of the tobacco product after the second processing experiment; and determining the processing intensity corresponding to each first combined formula according to the sensory evaluation results of the tobacco product after the second processing experiment.

[0012] In some embodiments, the candidate tobacco product comprises a plurality of tobacco raw materials, and the obtaining the multi-dimensional features of the candidate tobacco product corresponding to each candidate combined formula comprises: obtaining thermal gravimetric analysis data and second chemical component data of each tobacco raw material of the plurality of tobacco raw materials, the thermal gravimetric analysis data indicating a relationship between a mass change rate and a temperature of each tobacco raw material in a pyrolysis process, and the second chemical component data comprising contents of each chemical component of each tobacco raw material; determining a second volatile matter score and a second fiber class score of each tobacco raw material according to the thermal gravimetric analysis data; determining the first volatile matter score and the first fiber class score according to a plurality of second volatile matter scores and a plurality of second fiber class scores of the plurality of tobacco raw materials; determining the first chemical component data according to the second chemical component data and each candidate combined formula; and determining the origin style of the candidate tobacco product according to the thermal gravimetric analysis data, the first chemical component data, and each candidate combined formula.

[0013] In some embodiments, the determining the first volatile matter score and the first fiber class score according to the plurality of second volatile matter scores and the plurality of second fiber class scores of the plurality of tobacco raw materials comprises: performing fusion processing on the plurality of second volatile matter scores to determine the first volatile matter score; and performing fusion processing on the plurality of second fiber class scores to determine the first fiber class score.

[0014] In some embodiments, the retrieving, from the combined formula-processing scheme database, a first target combined formula matching a second combined formula to be processed comprises: obtaining the multi-dimensional features of a tobacco product to be tested corresponding to the second combined formula; calculating similarities between the multi-dimensional features of the tobacco product to be tested and the multi-dimensional features of the tobacco product corresponding to each first combined formula; and determining the first target combined formula according to the similarities.

[0015] In some embodiments, the first target combination formulation includes multiple formulations, and the step of recommending a target processing scheme corresponding to the second combination formulation based on the processing scheme corresponding to the first target combination formulation includes: determining the sensory evaluation result of each of the multiple formulations; determining the second target combination formulation from the multiple formulations based on the sensory evaluation result of each formulation; and determining the processing scheme corresponding to the second target combination formulation as the target processing scheme.

[0016] In some embodiments, the plurality of chemical components includes total sugars and nicotine.

[0017] According to some embodiments of this disclosure, a processing scheme recommendation apparatus is provided, comprising: a database construction module configured to perform the following steps to construct a combination formula-processing scheme database based on each of a plurality of candidate combination formulas: obtaining multidimensional features of candidate tobacco products corresponding to each candidate combination formula, the multidimensional features including multiple of a first volatile component score, a first fiber score, first chemical component data, and origin style, the first chemical component data including the content of each chemical component among multiple chemical components of the candidate tobacco product; screening a plurality of first combination formulas from the plurality of candidate combination formulas based on the multidimensional features of the candidate tobacco product; determining a processing scheme corresponding to each of the plurality of first combination formulas, the processing scheme including a processing path and a processing intensity; storing each first combination formula and the processing scheme corresponding to each first combination formula in the combination formula-processing scheme database; and a recommendation module configured to retrieve a first target combination formula matching a second combination formula to be processed from the combination formula-processing scheme database, and recommend a corresponding target processing scheme for the second combination formula based on the processing scheme corresponding to the first target combination formula.

[0018] According to further embodiments of this disclosure, a processing scheme recommendation apparatus is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute the recommendation method of any of the above embodiments based on instructions stored in the memory device.

[0019] According to further embodiments of the present disclosure, a computer-readable storage medium is provided having computer instructions stored thereon that, when executed by a processor, implement the recommended methods in any of the above embodiments.

[0020] According to further embodiments of this disclosure, a computer program product is also provided, including instructions that, when executed by a processor, cause the processor to perform the recommended method according to any of the foregoing embodiments.

[0021] In the above embodiments, by constructing a database of representative combination formulations and corresponding processing schemes, when a processing scheme needs to be determined for a new combination formulation, it is only necessary to search for a matching combination formulation in the database to quickly and reliably recommend a verified and effective processing scheme as the processing scheme corresponding to the new combination formulation. This eliminates the need for extensive small-scale / pilot-scale experiments and reduces the risk of human error, thus recommending suitable processing schemes for combination formulations in a low-cost, efficient, and reliable manner. Attached Figure Description

[0022] The accompanying drawings, which form part of this specification, illustrate embodiments of this disclosure and, together with the specification, serve to explain the principles of this disclosure.

[0023] This disclosure will become clearer with reference to the accompanying drawings and the following detailed description, wherein:

[0024] Figure 1 A flowchart illustrating a recommended method for processing schemes according to some embodiments of the present disclosure;

[0025] Figure 2 A flowchart illustrating a method for obtaining multidimensional features according to some embodiments of the present disclosure is shown;

[0026] Figure 3 A flowchart illustrating a recommended method for processing schemes according to other embodiments of the present disclosure;

[0027] Figure 4 A block diagram illustrating a recommended apparatus for a processing scheme according to some embodiments of the present disclosure;

[0028] Figure 5 A block diagram illustrating a recommended apparatus for a processing scheme according to other embodiments of the present disclosure;

[0029] Figure 6 A block diagram of a recommended apparatus for a processing scheme according to some embodiments of the present disclosure is shown. Detailed Implementation

[0030] 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, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0031] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0032] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.

[0033] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0034] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0035] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0036] As mentioned above, determining a suitable processing scheme for a new combination formulation often requires relying on experience or conducting numerous small-scale / pilot-scale experiments, which is time-consuming and labor-intensive. In view of this, embodiments of this disclosure provide a method for recommending processing schemes. By constructing a database storing representative combination formulations and corresponding processing schemes, when a processing scheme needs to be determined for a new combination formulation, the method can quickly and reliably recommend a proven and effective processing scheme as the processing scheme corresponding to the new combination formulation simply by retrieving a matching combination formulation from the database.

[0037] In this way, there is no need to conduct a large number of small-scale / pilot-scale experiments, and the risk of human experience misjudgment is reduced, thus recommending suitable processing schemes for combined formulations in a low-cost, efficient and reliable manner.

[0038] Figure 1 A flowchart illustrating a recommended method for processing schemes according to some embodiments of the present disclosure is provided.

[0039] like Figure 1 As shown, in step 110, steps 111 to 114 are performed to construct a combination formula-processing scheme database based on each candidate combination formula among multiple candidate combination formulas.

[0040] In step 111, the multidimensional features of the candidate tobacco products corresponding to each candidate combination formulation are obtained.

[0041] In this disclosure, tobacco products are products made wholly or partially from tobacco leaves and intended for smoking, chewing, snorting, or otherwise use. Tobacco products may include, but are not limited to, cigarettes, cigars, pipe tobacco, or hookah. Cigarettes, distinguished by their tobacco leaf formulation and processing method, may include, but are not limited to, flue-cured tobacco, burley tobacco, aromatic tobacco, or sun-cured tobacco. Tobacco products release chemical substances such as nicotine (a common term for "nicotine") through heating (e.g., heated tobacco products) and / or combustion (e.g., cigarettes, cigars, etc.). Tobacco products may or may not have a cigarette wrapper and may or may not have a filter.

[0042] Tobacco product raw materials may include tobacco leaves in various forms, such as tobacco leaves or tobacco shreds, tobacco sheets, tobacco powder, or tobacco blocks processed from tobacco leaves. In some embodiments, tobacco product raw materials are tobacco leaves from a specific origin. For example, multiple tobacco product raw materials constituting a candidate tobacco product may be tobacco leaves from different origins, or tobacco leaves from different growing parts of the same origin (e.g., classified as "upper," "middle," and "lower" according to their growing part on the tobacco plant), or tobacco leaves from different batches from the same growing part of the same origin.

[0043] The multidimensional features of candidate tobacco products include multiple features from the first volatile component score, the first fiber score, the first chemical composition data, and the regional style. For example, the multidimensional features of candidate tobacco products may include any two, three, or more features from the first volatile component score, the first fiber score, the first chemical composition data, and the regional style.

[0044] From a material composition analysis perspective, tobacco product raw materials include volatile components, cellulose, hemicellulose, and lignin, among which cellulose, hemicellulose, and lignin are collectively referred to as fibers. In pyrolysis experiments, volatile components undergo pyrolysis first and are released, followed by the release of fibers through pyrolysis at higher temperatures. The char produced from the pyrolysis of fibers then undergoes a combustion reaction, releasing the aroma of the tobacco leaves and forming smoke.

[0045] The first volatile component score of a candidate tobacco product is related to the content of volatile components in the candidate tobacco product and is used to represent the quality grade of the candidate tobacco product. For example, the higher the first volatile component score of a candidate tobacco product, the higher the quality grade of the candidate tobacco product.

[0046] The first fiber score of a candidate tobacco product is related to the fiber content in the product and is used to represent its combustion performance. For example, a higher first fiber score indicates weaker combustion performance, or a lower combustion rating.

[0047] The first chemical composition data includes the content of each chemical component in a variety of chemical components of the candidate tobacco product.

[0048] For example, multiple chemical components may include total sugar and nicotine. For example, the first chemical composition data of a candidate tobacco product may include the total sugar content and nicotine content of the candidate tobacco product.

[0049] In step 112, based on the multidimensional characteristics of the candidate tobacco products, multiple first combination formulations are selected from multiple candidate combination formulations.

[0050] In step 113, a processing scheme corresponding to each of the plurality of first combination formulations is determined. For example, the processing scheme includes a processing path and a processing intensity.

[0051] In step 114, each first combination formula and the corresponding processing scheme are stored in the combination formula-processing scheme database.

[0052] In step 120, a target processing scheme is recommended for the second combined formulation to be processed.

[0053] For example, a first target combination formula that matches the second combination formula to be processed is retrieved from the combination formula-processing scheme database, and the target processing scheme corresponding to the second combination formula is determined according to the processing scheme corresponding to the first target combination formula.

[0054] In the above embodiments, multiple representative first combination formulations are selected based on the characteristics of any multiple dimensions among the first volatile component score, first fiber score, first chemical composition data, and regional style of the candidate tobacco products corresponding to each candidate combination formulation, in order to construct a combination formulation-processing scheme database. When it is necessary to determine the corresponding processing scheme for a specific second combination formulation, it is only necessary to search the constructed database for the processing scheme of the matching combination formulation to determine the target processing scheme.

[0055] In this way, since the first volatile component score can reflect the quality grade of tobacco products, the first fiber score can reflect the combustion performance of tobacco products, the first chemical composition data can reflect the chemical characteristics of tobacco products, and the origin style can reflect the sensory style of tobacco products, the first combination formula selected based on multiple different dimensions of characteristics can demonstrate high representativeness in multiple dimensions of quality grade, combustion performance, chemical characteristics, and sensory style, thereby improving the scientificity and diversity of combination formulas in the constructed database.

[0056] Based on this, the first target combination formula matched in the database and the second combination formula have a high degree of consistency in multiple dimensions such as quality grade, combustion performance, chemical properties and sensory style. This makes it possible to process the tobacco products corresponding to the second combination formula according to the target processing scheme recommended by the processing scheme corresponding to the first target combination formula to achieve the expected quality, thereby obtaining a reliable target processing scheme in a low-cost and efficient manner.

[0057] The following examples illustrate the method of constructing the combined formula-processing scheme database in step 110.

[0058] First, we will use some examples to illustrate the multidimensional features of the candidate tobacco products corresponding to each candidate combination formula in step 111.

[0059] Figure 2 A flowchart illustrating a method for obtaining multidimensional features according to some embodiments of the present disclosure is shown.

[0060] like Figure 2 As shown, in step 210, thermogravimetric analysis data and second chemical composition data of each tobacco product raw material in multiple tobacco product raw materials are obtained.

[0061] Here, "multiple tobacco product raw materials" refers to the multiple tobacco product raw materials that make up the candidate tobacco product corresponding to each candidate combination formulation. Thermogravimetric analysis data indicates the relationship between the rate of mass change of each tobacco product raw material and temperature during pyrolysis, and the second chemical composition data includes the content of each chemical component in each tobacco product raw material.

[0062] For example, a continuous flow method can be used to obtain the content of each chemical component among multiple chemical components in each tobacco product raw material.

[0063] For example, various chemical components may include total sugar and nicotine. Secondary chemical composition data for tobacco product raw materials may include the total sugar content and nicotine content in the tobacco product raw materials.

[0064] For example, thermogravimetric analysis data for tobacco product raw materials can be obtained by conducting pyrolysis experiments on the raw materials and recording multiple temperatures (also known as pyrolysis temperatures) during the pyrolysis process and the corresponding rate of mass change of the raw materials at each temperature. The rate of mass change can be determined by calculating the derivative of the mass of the raw materials with respect to temperature during pyrolysis.

[0065] The following example uses a certain type of tobacco leaf as the raw material for tobacco products to illustrate the process of conducting a pyrolysis experiment on the raw material for tobacco products.

[0066] First, a suitable amount of tobacco powder from the tobacco leaves was weighed as a sample. The sample was heated to a certain temperature at a certain rate under a nitrogen atmosphere and kept at a constant temperature for a period of time to eliminate the influence of different moisture contents of the tobacco leaves. Then, the sample was heated to an even higher temperature at the same rate. The relationship between sample mass and temperature during the heating process was recorded, thus obtaining the thermogravimetry (TG) curve. To make the pyrolysis behavior of tobacco more obvious, the derivative of the TG curve was obtained to obtain the derivative thermogravimetry (DTG) curve, which reflects the relationship between the rate of mass change and temperature, i.e., the "thermogravimetry data" mentioned in this paper. If the thermogravimetry data is displayed as a curve (i.e., the DTG curve), the horizontal axis represents the pyrolysis temperature in Kelvin (K), and the vertical axis represents the rate of mass change of the tobacco leaf during the pyrolysis process in percentage / Kelvin (% / K). In this paper, this curve is also referred to as the thermal analysis spectrum.

[0067] In step 220, the second volatile component score and the second fiber score of each tobacco product raw material are determined based on thermogravimetric analysis data.

[0068] In the thermal analysis spectrum of tobacco product raw materials, two pyrolysis peaks corresponding to two temperature ranges will appear. The first pyrolysis peak, corresponding to the lower first temperature range, is related to the content of volatile components. It represents the quality grade of the tobacco product raw material and directly affects its aroma and taste. The second pyrolysis peak, corresponding to the higher second temperature range, is related to the content of fibers. It represents the combustion performance of the tobacco product raw material and indirectly affects its aroma and taste. The content of volatile components and fibers varies in different tobacco product raw materials, so the peak values ​​of the first and second pyrolysis peaks in the corresponding thermal analysis spectrum may also differ. It should be understood that the term "peak" here is described with the absolute value of the rate of mass change. Since the rate of mass change is actually negative, the "peak" here can be understood as a "valley" from the perspective of the thermal analysis spectrum curve. In some embodiments, the first temperature range can be 400K to 500K, and the second temperature range can be 500K (excluding 500K) to 800K.

[0069] To determine the volatile matter score and fiber score of each tobacco product raw material, the corresponding volatile matter score can be calculated based on the peak value of the first pyrolysis peak of each tobacco product raw material, and the corresponding fiber score can be calculated based on the peak value of the second pyrolysis peak of each tobacco product raw material.

[0070] The following describes a non-limiting exemplary calculation process using tobacco leaves as the raw material for tobacco products.

[0071] First, the maximum and minimum values ​​of the first pyrolysis peak in the first temperature range and the maximum and minimum values ​​of the second pyrolysis peak in the second temperature range can be determined from the thermogravimetric analysis data of multiple tobacco leaves.

[0072] The maximum and minimum values ​​of the first pyrolysis peak correspond to the endpoints of the score range for volatile components. For example, the minimum value of the first pyrolysis peak can be set as the maximum value of the corresponding volatile component score, i.e., the upper limit of the score range, and the maximum value of the first pyrolysis peak can be set as the minimum value of the corresponding volatile component score, i.e., the lower limit of the score range. The score range for volatile components can be, for example, [0, 100], meaning that the minimum value of the first pyrolysis peak corresponds to a volatile component score of 100, and the maximum value of the first pyrolysis peak corresponds to a volatile component score of 0. This means that the smaller the peak value of the first pyrolysis peak, the higher the corresponding volatile component score.

[0073] The maximum and minimum values ​​of the second pyrolysis peak correspond to the endpoints of the score range for fiber components. For example, the minimum value of the second pyrolysis peak can be set as the minimum score for the corresponding fiber component, i.e., the lower limit of the score range, and the maximum value of the second pyrolysis peak can be set as the maximum score for the corresponding volatile component, i.e., the upper limit of the score range. The score range for fiber components can also be [0, 100], i.e., the minimum value of the second pyrolysis peak corresponds to a fiber component score of 0, and the maximum value of the second pyrolysis peak corresponds to a fiber component score of 100. This means that the smaller the peak value of the second pyrolysis peak, the lower the corresponding fiber component score.

[0074] Next, the volatile matter score is calculated within the volatile matter score range based on the ratio of the peak value of the first pyrolysis peak of each tobacco leaf to the aforementioned maximum and minimum values ​​of the first pyrolysis peak. For example, if the peak value of the first pyrolysis peak of a tobacco leaf is exactly in the middle between the aforementioned maximum and minimum values ​​of the first pyrolysis peak, then proportionally, the volatile matter score is 50 in the score range [0, 100]. Similarly, the fiber score is calculated within the fiber score range based on the ratio of the peak value of the second pyrolysis peak of each tobacco leaf to the aforementioned maximum and minimum values ​​of the second pyrolysis peak. For example, if the peak value of the second pyrolysis peak of a tobacco leaf is exactly in the middle between the aforementioned maximum and minimum values ​​of the second pyrolysis peak, then proportionally, the fiber score is 50 in the score range [0, 100].

[0075] For example, the volatile matter score and fiber score of a certain tobacco leaf i can be calculated proportionally using the following formula:

[0076]

[0077]

[0078] Among them, the peak value of the first pyrolysis peak of tobacco leaf i is denoted as P1i, the peak value of the second pyrolysis peak is denoted as P2i, and the score of volatile components is denoted as... The score for fiber category is recorded as .

[0079] In step 230, the first volatile component score and the first fiber score are determined based on the multiple second volatile component scores and multiple second fiber scores of the multiple tobacco product raw materials.

[0080] In some embodiments, multiple second volatile component scores are fused to determine the first volatile component score of the candidate tobacco product.

[0081] For example, multiple second volatile component scores can be merged by calculating the sum of multiple second volatile component scores, and this sum can be determined as the first volatile component score of the candidate tobacco product. Alternatively, multiple second volatile component scores can be merged by calculating a weighted average of multiple second volatile component scores, and this weighted average can be determined as the first volatile component score of the candidate tobacco product.

[0082] In the above embodiments, the first volatile component score of the candidate tobacco product is determined by integrating the second volatile component scores of multiple tobacco product raw materials that make up the candidate tobacco product, so that the first volatile component score can reliably reflect the overall volatility of the candidate tobacco product.

[0083] In this way, the score of the first volatile component can more reliably reflect the quality level of candidate tobacco products, which helps to improve the reliability of the first combination formulation used to build the database, thereby helping to improve the reliability of the target processing scheme recommended by the database.

[0084] In some embodiments, multiple second fiber category scores are fused to determine the first fiber category score of the candidate tobacco product.

[0085] For example, multiple second fiber category scores can be merged by summing the scores of multiple second fiber categories, and this sum can be determined as the first fiber category score of the candidate tobacco product. Alternatively, multiple second fiber category scores can be merged by calculating a weighted average of the scores, and this weighted average can be determined as the first fiber category score of the candidate tobacco product.

[0086] In the above embodiments, the first fiber score of the candidate tobacco product is determined by integrating the second fiber scores of multiple tobacco product raw materials that make up the candidate tobacco product, so that the first fiber score can reliably reflect the overall combustion performance of the candidate tobacco product.

[0087] In this way, the first fiber category score can more reliably reflect the combustion rating of candidate tobacco products, which helps to improve the reliability of the first combination formulation used to build the database, thereby helping to improve the reliability of the target processing scheme recommended by the database.

[0088] In step 240, the first chemical composition data is determined based on the second chemical composition data and each candidate combination formulation.

[0089] In some embodiments, the candidate formulation corresponding to the candidate tobacco product may include the mass percentage of multiple tobacco product ingredients in the candidate tobacco product. The mass percentage may be presented as an absolute value or a relative value. In the absolute value form, the formulation may include, for example, the mass of each tobacco product ingredient (e.g., in grams, milligrams, micrograms, etc.), thereby indirectly reflecting the mass percentage. In the relative value form, the formulation may include, for example, the ratio of the mass of each tobacco product ingredient to the mass of the candidate tobacco product (e.g., in percentage form), thereby directly reflecting the mass percentage.

[0090] In some embodiments, the first chemical composition data of a candidate tobacco product can be obtained based on the second chemical composition data of each of the plurality of tobacco product raw materials that make up the candidate tobacco product.

[0091] As one implementation method, the mass percentage of each tobacco product raw material in the candidate tobacco product in the combined formulation is used as the weight of the second chemical component data, the first weighted sum of the second chemical component data is calculated, and the first weighted sum is determined as the first chemical component data.

[0092] Taking nicotine as an example, a candidate tobacco product corresponding to a certain combination formula is known to be composed of tobacco leaves A from Yunnan, tobacco leaves B from Fujian, and tobacco leaves C from Henan. Assume the nicotine content in the second chemical component data of tobacco leaf A is W1, the nicotine content in the second chemical component data of tobacco leaf B is W2, and the nicotine content in the second chemical component data of tobacco leaf C is W3. Using 'a' as the weight for the nicotine content W1 in tobacco leaf A, 'b' as the weight for the nicotine content W2 in tobacco leaf B, and 'c' as the weight for the nicotine content W3 in tobacco leaf C, the nicotine content W in the first chemical component data of the candidate tobacco product can be calculated as: W = a × W1 + b × W2 + c × W3.

[0093] In step 250, the origin style of the candidate tobacco products is determined based on the thermogravimetric analysis data of each tobacco product raw material, the first chemical composition data, and each candidate combination formulation.

[0094] In some embodiments, thermogravimetric analysis (TGA) data for candidate tobacco products are determined based on TGA data for each tobacco product raw material and each candidate combination formulation. Based on the TGA data and first chemical composition data of the candidate tobacco products, a trained prediction model is used to predict the regional style of the candidate tobacco products.

[0095] In some embodiments, the thermogravimetric analysis data of the candidate tobacco product can be determined based on the thermogravimetric analysis data of each tobacco product raw material and the mass percentage of each tobacco product raw material in the candidate tobacco product.

[0096] As one implementation method, the mass percentage of each tobacco product raw material in the candidate tobacco product is used as the weight of the thermogravimetric analysis data of each tobacco product raw material. A second weighted sum of the thermogravimetric analysis data of multiple tobacco product raw materials is calculated, and the second weighted sum is determined as the thermogravimetric analysis data of the candidate tobacco product.

[0097] Continuing with the example above, assume that the mass percentage (e.g., percentage) of tobacco leaf A in the candidate tobacco product is *a*, the mass percentage of tobacco leaf B in the candidate tobacco product is *b*, and the mass percentage of tobacco leaf C in the candidate tobacco product is *c*. Using *a* as the weight of the first thermogravimetric analysis data S1 for tobacco leaf A, using *b* as the weight of the first thermogravimetric analysis data S2 for tobacco leaf B, and using *c* as the weight of the first thermogravimetric analysis data S3 for tobacco leaf C, we can calculate the thermogravimetric analysis data S of the candidate tobacco product as: S = a × S1 + b × S2 + c × S3.

[0098] It is important to clarify that the regional style of tobacco products refers to the overall sensory characteristics and style that emerge after blending tobacco leaves from different regions. In other words, the regional style of tobacco products reflects the overall style characteristics of the tobacco leaves from various regions after blending. For example, if the regional style of a tobacco product is Fujian, it means that the tobacco product as a whole exhibits the style characteristics (such as aroma) of tobacco leaves from Fujian.

[0099] In some embodiments, thermogravimetric analysis data and first chemical composition data of candidate tobacco products can be input into a trained prediction model to obtain the regional style of the candidate tobacco products output by the prediction model. As an exemplary implementation, different regional styles can correspond to different prediction labels. For example, the number 1 can represent the regional style as Yunnan, and the number 2 can represent the regional style as Fujian. In response to the prediction model outputting a prediction label of 1, the regional style of the candidate tobacco product can be determined to be Yunnan.

[0100] In the above embodiments, the thermogravimetric analysis data and chemical composition data of multiple tobacco raw materials constituting the candidate tobacco products are used to obtain multi-dimensional characteristics of the candidate tobacco products, including volatile matter scores, fiber scores, chemical composition data, and regional style. This makes the objective basis for screening the first combination formulation more scientific and accurate. This improves the reliability of the first combination formulation used to construct the database, thereby helping to improve the reliability of the target processing scheme recommended through the database.

[0101] Next, the method of screening multiple first combination formulations in step 112 will be illustrated by way of example with reference to some embodiments.

[0102] In some embodiments, the multidimensional features of a candidate tobacco product include at least one of a first volatile component score, a first fiber score, and first chemical composition data, as well as a country of origin style. For example, the multidimensional features of a candidate tobacco product may include a first volatile component score, a first fiber score, first chemical composition data, and a country of origin style.

[0103] In these embodiments, multiple first combination formulations can be screened in the following manner.

[0104] First, multiple groups of first tobacco products exhibiting styles from multiple candidate tobacco products corresponding to multiple candidate combination formulations are selected. Here, different groups of tobacco products exhibit styles from different target origins, and each group of first tobacco products includes multiple first tobacco products.

[0105] Then, based on at least one of the first volatile component score, first fiber score, and first chemical composition data of each first tobacco product in each group of first tobacco products, multiple second tobacco products are selected from each group of first tobacco products. For example, by selecting multiple second tobacco products for each group of first tobacco products, multiple groups of second tobacco products corresponding to multiple groups of first tobacco products can be obtained, that is, multiple groups of second tobacco products presenting multiple target origin styles can be obtained.

[0106] Subsequently, based on the multiple combination formulas corresponding to the multiple second tobacco products, multiple first combination formulas are determined.

[0107] For example, from multiple candidate tobacco products, a group of first tobacco products X1 exhibiting Yunnan style, a group of first tobacco products X2 exhibiting Fujian style, and a group of first tobacco products X3 exhibiting Henan style were selected.

[0108] Then, based on the first volatile component score, first fiber score, and first chemical component data of each first tobacco product in X1, multiple second tobacco products are selected from X1 to obtain a set of second tobacco products Y1; based on the first volatile component score, first fiber score, and first chemical component data of each first tobacco product in X2, multiple second tobacco products are selected from X2 to obtain a set of second tobacco products Y2; based on the first volatile component score, first fiber score, and first chemical component data of each first tobacco product in X1, multiple second tobacco products are selected from X3 to obtain a set of second tobacco products Y3.

[0109] Subsequently, all the combined formulations corresponding to the second tobacco products in the three groups of second tobacco products Y1, Y2, and Y3 were determined as multiple first combined formulations.

[0110] In the above embodiments, multiple candidate tobacco products are first screened based on their origin style to identify and retain multiple groups of first tobacco products that can exhibit characteristics of multiple target origin styles. Subsequently, within each group of first tobacco products, a second round of refined screening is conducted by combining features from at least one dimension of their first volatile component score, first fiber score, and first chemical composition data. Through this two-stage screening strategy, multiple groups of second combination formulations with different target origin styles and excellent performance in one or more relevant indicators such as quality grade, combustion performance, and chemical properties are finally obtained, serving as the data foundation for constructing a combination formulation-processing scheme database.

[0111] In this way, the first combination of formulas selected not only has the representative style of the place of origin, but also has been cross-validated by multi-dimensional physicochemical indicators, effectively taking into account the consistency of style characteristics and intrinsic quality, and significantly improving the practicality and generalization ability of the constructed database in subsequent application scenarios such as processing scheme migration and product optimization.

[0112] In some embodiments, each group of first tobacco products comprises multiple subgroups, and the first tobacco products in different subgroups have different probabilities of exhibiting the target origin style corresponding to each group of first tobacco products. The first tobacco products in the same subgroup may have the same or different probabilities of exhibiting the target origin style corresponding to each group of first tobacco products.

[0113] The different probabilities of tobacco products exhibiting the style of their target origin indicate varying degrees of emphasis on that style. For example, the higher the probability of a tobacco product exhibiting the style of its target origin, the greater the degree to which it emphasizes that style.

[0114] Continuing with the example above, a group of first tobacco products X1 exhibiting a Yunnan style can contain three subgroups (X11, X12, X13), with each subgroup containing multiple first tobacco products. The probability of the first tobacco products in subgroup X11 exhibiting a Yunnan style can be between 60% and 70%, the probability in subgroup X12 can be between 70% and 80%, and the probability in subgroup X13 can be between 80% and 90%. Here, among the three subgroups, the first tobacco products in subgroup X13 exhibit the highest degree of Yunnan style, followed by the first tobacco products in subgroup X12, while the first tobacco products in subgroup X11 exhibit the lowest degree of Yunnan style.

[0115] In this way, by selecting first tobacco products from multiple candidate tobacco products that exhibit different intensities of style prominence for each target origin, the subsequent construction of a combination formulation-processing scheme database has a richer hierarchical structure and broader coverage. This improves the diversity and representativeness of the combination formulations in the constructed database, laying a solid data foundation for the recommendation of subsequent processing schemes.

[0116] In some embodiments, a plurality of second tobacco products are selected from each group of first tobacco products based on at least one of a first difference between the first volatile scores of different first tobacco products in each group, a second difference between the first fiber scores, and a third difference in the first chemical composition data.

[0117] As one implementation method, multiple second tobacco products are selected from each group of first tobacco products based on the first difference between the first volatile scores, the second difference between the first fiber scores, and the third difference between the first chemical composition data of different first tobacco products in each group of first tobacco products.

[0118] For example, a first difference can be determined based on the difference between the first volatile matter scores of different first tobacco products. A second difference can be determined based on the difference between the first fiber scores of different first tobacco products. A third difference can be determined based on the difference between the content of the same chemical component in the first chemical composition data of different first tobacco products. For example, a third difference may include the difference between the total sugar content and / or the nicotine content in different first tobacco products.

[0119] For example, multiple first tobacco products that satisfy at least one of the following conditions—a first difference greater than a first threshold, a second difference greater than a second threshold, and a third difference greater than a third threshold—can be identified as multiple second tobacco products. That is, multiple first tobacco products that satisfy at least one of the following conditions—a significant difference in a first volatile matter score, a significant difference in a first fiber score, and a significant difference in a first chemical composition data—can be identified as multiple second tobacco products.

[0120] In the above embodiments, multiple second tobacco products are selected from each group of first tobacco products based on at least one of the following differences: a first difference between first volatile scores, a second difference between first fiber scores, and a third difference between first chemical composition data.

[0121] In this way, the selected secondary tobacco products exhibit high differentiation across one or more key dimensions, such as quality grade, combustion performance, and chemical properties. This not only enriches the diversity of blended formulations but also enhances their representativeness in terms of coverage. Consequently, the database stores more comprehensive blended formulations, thereby contributing to improving the scientific rigor and reliability of subsequent recommended processing schemes.

[0122] Next, the method of determining the processing scheme corresponding to each first combination formula in step 113 will be illustrated by way of example with reference to some embodiments.

[0123] In some embodiments, a first processing experiment is performed on the tobacco products corresponding to each first combination formulation according to different processing paths, and the sensory evaluation results of the tobacco products after the first processing experiment are obtained. Based on the sensory evaluation results of the tobacco products after the first processing experiment, the processing path corresponding to each first combination formulation is determined.

[0124] For example, the processing path may include a drum processing process and an airflow processing process. The first processing experiment includes two processing experiments performed on the tobacco products corresponding to each first combination formulation, according to both the drum processing process and the airflow processing process. The processing path that yields the optimal sensory evaluation results for the tobacco products after the processing experiments is determined as the processing path corresponding to that first combination formulation.

[0125] For example, processing experiments corresponding to both the drum processing process and the airflow processing process are performed on a tobacco product corresponding to a certain first combination formula M. If the sensory evaluation results of the tobacco product processed by the drum processing process are better than those of the tobacco product processed by the airflow processing process, then the processing path in the processing scheme corresponding to the first combination formula M can be determined to be the drum processing process.

[0126] In this way, the sensory evaluation results of tobacco products processed according to different processing paths can effectively reveal the differences in sensory performance of the same formula under different processing paths. Based on this, a corresponding processing scheme can be determined for each first combination formula in the constructed database, which significantly improves the effectiveness of the processing schemes stored in the database in terms of sensory performance, thereby improving the scientificity and reliability of recommending target processing schemes in subsequent practical applications.

[0127] In some embodiments, a second processing experiment is performed on the tobacco products corresponding to each first combination formulation according to different processing intensities along the same processing path, and the sensory evaluation results of the tobacco products after the second processing experiment are obtained. Based on the sensory evaluation results of the tobacco products after the second processing experiment, the processing intensity corresponding to each first combination formulation is determined.

[0128] For example, the second processing experiment includes multiple processing intensities performed on tobacco products corresponding to each first combination formulation according to the same processing path (such as roller processing or airflow processing). The processing intensity that yields the optimal sensory evaluation results for the tobacco products after the processing experiment is determined as the processing intensity corresponding to that first combination formulation. For example, the processing intensity can be changed by adjusting process parameters (such as equipment parameters) under the same processing path to perform multiple processing experiments with different intensities.

[0129] Following the example above, if the processing path in the processing scheme corresponding to the first combination formula M is determined to be a roller processing process, processing experiments can be performed on the tobacco products corresponding to the first combination formula M according to different processing intensities (e.g., low, medium, and high processing intensities). If the sensory evaluation results of the tobacco product processed with high processing intensity are better than those processed with other processing intensities, then the processing path in the processing scheme corresponding to the first combination formula M can be determined to be a high processing intensity process.

[0130] In this way, by conducting multiple processing experiments of different intensities on tobacco products corresponding to the same formula under the same processing path and obtaining the corresponding sensory evaluation results, the appropriate processing intensity can be accurately identified for each formula, thereby establishing a multi-dimensional mapping relationship between "combined formula - processing path - processing intensity - sensory performance". This significantly improves the effectiveness of the processing schemes stored in the database in terms of sensory performance, thus enhancing the scientific rigor and reliability of recommending target processing schemes in subsequent practical applications.

[0131] The method of recommending the target processing scheme in step 120 is illustrated by way of example with reference to some embodiments.

[0132] In some embodiments, the multidimensional features of the tobacco product to be tested corresponding to the second combination formula are obtained, the similarity between the multidimensional features of the tobacco product to be tested and the multidimensional features of the tobacco product corresponding to each first combination formula is calculated, and the first target combination formula is determined based on the similarity.

[0133] For example, multidimensional features may include the first volatile component score, the first fiber type score, the first chemical composition data, and the place of origin style.

[0134] The following similarities can be calculated: first similarity between the first volatile component score of the tested tobacco product and the first volatile component score of the tobacco product corresponding to each first combination formulation; second similarity between the first fiber score of the tested tobacco product and the first fiber score of the tobacco product corresponding to each first combination formulation; third similarity between the first chemical composition data of the tested tobacco product and the first chemical composition data of the tobacco product corresponding to each first combination formulation; and fourth similarity between the regional style of the tested tobacco product and the regional style of the tobacco product corresponding to each first combination formulation.

[0135] Then, the first target combination formulation can be determined based on the first similarity, second similarity, third similarity, and fourth similarity. For example, the first combination formulation in which the first similarity, second similarity, third similarity, and fourth similarity all meet the corresponding criteria is determined as the first target combination formulation.

[0136] In this way, by calculating the similarity of features across multiple dimensions, the determined first target combination formula can more accurately reflect the overall attribute characteristics of the second combination formula for the processing scheme to be determined. As a result, the target processing scheme determined based on the first target combination formula has a higher degree of adaptability, thus improving the accuracy and feasibility of the processing scheme recommendation.

[0137] In some embodiments, the first target combination formulation includes only one formulation. That is, the number of first target combination formulations matched in the combination formulation-processing scheme database is one.

[0138] In these embodiments, the processing scheme corresponding to the first target combination formulation can be determined as the target processing scheme corresponding to the second combination formulation. For example, the similarity between each first combination formulation and the second combination formulation to be processed in the combination formulation-processing scheme database can be calculated, and then the processing scheme corresponding to the first target combination formulation with the highest similarity can be determined as the target processing scheme corresponding to the second combination formulation.

[0139] In some embodiments, the first target combination formulation includes multiple formulations. That is, the number of first target combination formulations matched in the combination formulation-processing scheme database is multiple.

[0140] In these embodiments, sensory evaluation results for each of multiple formulations can be determined. Based on the sensory evaluation results of each formulation, a second target combination formulation is determined from the multiple formulations, and the processing scheme corresponding to the second target combination formulation is determined as the target processing scheme. For example, the similarity between each first combination formulation and the second combination formulation to be processed in the combination formulation-processing scheme database can be calculated, and multiple formulations with similarity satisfying a preset threshold can be determined as the first target combination formulation. Then, based on the sensory evaluation results of each formulation in the first target combination formulation, the formulation with the best sensory evaluation results can be selected from the multiple formulations of the first target combination formulation as the second target combination formulation, and the processing scheme corresponding to the second target combination formulation is determined as the target processing scheme corresponding to the second combination formulation.

[0141] In this way, when there are multiple matching initial target combination formulations, the sensory evaluation results of different formulations can be combined to determine the final recommended target processing scheme. Therefore, the recommended target processing scheme will better meet the requirements of sensory quality, thereby improving the scientific rigor and practicality of determining the target processing scheme.

[0142] The implementation of the technical solutions proposed in this disclosure will be illustrated below with reference to some embodiments.

[0143] Figure 3 A flowchart illustrating a recommended method for processing schemes according to other embodiments of this disclosure is provided. Figure 3 As shown, this method can be used as Figure 1 This is a specific implementation of the method.

[0144] In step 310, select several representative first combination formulations.

[0145] In some embodiments, multidimensional characteristics (such as first volatile component score, first fiber score, first chemical composition data, and origin style) of the candidate tobacco products corresponding to each candidate combination formulation are obtained. Based on the multidimensional characteristics of the candidate tobacco products, multiple representative first combination formulations are selected from multiple candidate combination formulations. For example, the first chemical composition data includes the total sugar content and nicotine content in the candidate tobacco products.

[0146] In this way, the combined formulation-processing scheme database, built based on multiple representative first combination formulations, comprehensively covers formulation types of key dimensions such as different origin styles, quality grades, combustion performance and chemical characteristics, forming a rich, diverse and widely covered combined formulation system.

[0147] In step 320, the processing scheme corresponding to each first combination formula is determined to construct a combination formula-processing scheme database.

[0148] In some embodiments, the system performs multi-gradient processing experiments for each selected first combination formulation. The multi-gradient processing experiments include a first processing experiment performed on the tobacco product corresponding to each first combination formulation along different processing paths, and a second processing experiment performed on the tobacco product corresponding to each first combination formulation along the same processing path but with different processing intensities.

[0149] For example, suppose that nine representative first combination formulations are selected through step 310, and multi-gradient simulation processing experiments of the system are executed on the simulation processing platform according to the gradient settings shown in Table 1 below.

[0150] Table 1

[0151]

[0152] As shown in Table 1, the processing paths in the machining experiments included roller machining and airflow machining. Three different machining intensities were set within the same processing path. For example, the machining intensities for the roller machining process included T1, T2, and T3, with T1 having the highest intensity, followed by T2, and T3 having the lowest. The machining intensities for the airflow machining process included Q1, Q2, and Q3, with Q1 having the highest intensity, followed by Q2, and Q3 having the lowest.

[0153] In step 330, a combination formula matching is performed. For example, a first target combination formula that matches the second combination formula to be processed is retrieved from the combination formula-processing scheme database.

[0154] For example, taking the second combination formula as formula A, we can calculate the similarity between the features of each dimension of formula A (first volatile component score, first fiber score, first chemical composition data, and origin style) and the features of each first combination formula in the combination formula-processing scheme database, so as to determine the first target combination formula that matches the formula.

[0155] In step 340, a target processing scheme is recommended. For example, based on the processing scheme corresponding to the first target combination formulation, a corresponding target processing scheme is recommended for the second combination formulation.

[0156] Following the example above, assuming there are multiple first target combination formulations that match formulation A, but the sensory evaluation result of combination formulation III is the best (e.g., the highest sensory quality score), then the processing scheme corresponding to combination formulation III (e.g., roller processing with a processing intensity of T2) can be determined as the target processing scheme for formulation A.

[0157] To verify the effectiveness of transferring the processing scheme corresponding to combination formulation III to formulation A, a pilot-scale test of the product can be carried out, and the sensory evaluation results of formulation A under different processing intensities under roller processing technology (also known as roller line) and airflow processing technology (also known as airflow line) can be statistically analyzed (as shown in the form of sensory quality scores), as shown in Table 2.

[0158] Table 2

[0159]

[0160] As shown in Table 2, formulation III exhibited the highest sensory quality score (61.8 points) under the T2 processing intensity of the roller line, while formulation A also exhibited the highest sensory quality score (62.4 points) under the T2 processing intensity of the roller line.

[0161] This result demonstrates that the processing scheme recommendation method provided in this disclosure can effectively and accurately transfer and adapt processing schemes that have been successfully validated on specific formulations to new combination formulations. This technical solution not only significantly improves the success rate and reusability of processing scheme design but also ensures the consistency and stability of product sensory quality, successfully achieving a complete closed loop from intelligent recommendation of digital processing schemes to verification of actual production effects.

[0162] The implementation of the technical solutions proposed in this disclosure will be further illustrated below with reference to some embodiments.

[0163] For the new combination formulation B, based on the characteristics of multiple dimensions such as the first volatile component score, the first fiber score, the first chemical composition data, and the origin style, the first target combination formulation that matches combination formulation B is retrieved from the combination formulation-processing scheme database.

[0164] Assuming the first target formulation to be matched is Formulation II, which is highly consistent with Formulation B in terms of style characteristics (such as origin style) and intrinsic properties (such as quality grade, combustion performance, and chemical properties), based on this high similarity of multi-dimensional characteristics, it is reasonable to recommend the processing scheme of Formulation II (i.e., the roller line T2 processing strength) as the processing scheme of Formulation B.

[0165] The processing scheme corresponding to Compound Formula II was transferred to Compound Formula B, and pilot-scale production was conducted in the same manner. Sensory evaluation results of the pilot-scale samples showed that the tobacco product samples corresponding to Compound Formula B, processed under the T2 processing intensity of the roller conveyor, performed excellently in key sensory dimensions such as aroma characteristics, smoke characteristics, and taste characteristics, indicating that the product design style of Compound Formula B was accurately realized.

[0166] This result demonstrates that the processing scheme recommendation method provided in this disclosure can effectively improve the success rate and reusability of processing scheme design, and ensure the consistency and stability of product sensory quality, providing reliable technical support for the research and development of intelligent and efficient tobacco products.

[0167] Figure 4 A block diagram illustrating a recommended apparatus for a processing scheme according to some embodiments of the present disclosure is shown.

[0168] like Figure 4 As shown, the recommendation device 400 includes a database construction module 401 and a recommendation module 402.

[0169] The database building module 401 is configured to perform the following steps to build a combination formula-processing scheme database based on each of the multiple candidate combination formulas.

[0170] First, obtain the multidimensional features of the candidate tobacco products corresponding to each candidate combination formula. The multidimensional features include multiple features such as the first volatile component score, the first fiber score, the first chemical component data, and the origin style. The first chemical component data includes the content of each chemical component among the various chemical components of the candidate tobacco products.

[0171] Then, based on the multidimensional characteristics of the candidate tobacco products, multiple first combination formulations are selected from multiple candidate combination formulations, and a processing scheme corresponding to each of the multiple first combination formulations is determined. The processing scheme includes processing path and processing intensity.

[0172] Then, each first combination formula and the corresponding processing scheme are stored in the combination formula-processing scheme database.

[0173] The recommendation module 402 is configured to retrieve a first target combination formula that matches the second combination formula to be processed from the combination formula-processing scheme database, and recommend a corresponding target processing scheme for the second combination formula based on the processing scheme corresponding to the first target combination formula.

[0174] In some embodiments, the multidimensional features include at least one of a first volatile component score, a first fiber type score, and first chemical composition data, as well as place of origin style.

[0175] The database construction module 401 is configured to select multiple groups of first tobacco products that present multiple target origin styles from multiple candidate tobacco products corresponding to multiple candidate combination formulas, with different groups of tobacco products presenting different target origin styles, and each group of first tobacco products including multiple first tobacco products; select multiple second tobacco products from each group of first tobacco products based on at least one of the first volatile component score, first fiber score, and first chemical component data of each first tobacco product in each group of first tobacco products; and determine multiple first combination formulas based on multiple combination formulas corresponding to multiple second tobacco products.

[0176] In some embodiments, the database construction module 401 is configured to select a plurality of second tobacco products from each group of first tobacco products based on at least one of a first difference between the first volatile scores of different first tobacco products in each group, a second difference between the first fiber scores, and a third difference between the first chemical composition data.

[0177] In some embodiments, each group of first tobacco products includes multiple subgroups, and the first tobacco products in different subgroups of each group of first tobacco products have different probabilities of exhibiting the target origin style corresponding to each group of first tobacco products.

[0178] In some embodiments, the database construction module 401 is configured to perform a first processing experiment on the tobacco product corresponding to each first combination formula according to different processing paths, and obtain the sensory evaluation results of the tobacco product after the first processing experiment; and determine the processing path corresponding to each first combination formula based on the sensory evaluation results of the tobacco product after the first processing experiment.

[0179] In some embodiments, the database construction module 401 is configured to perform a second processing experiment on the tobacco products corresponding to each first combination formula according to different processing intensities of the same processing path, and obtain the sensory evaluation results of the tobacco products after the second processing experiment; and determine the processing intensity corresponding to each first combination formula based on the sensory evaluation results of the tobacco products after the second processing experiment.

[0180] In some embodiments, the candidate tobacco product includes multiple tobacco product raw materials. The database construction module 401 is configured to acquire thermogravimetric analysis (TGA) data and second chemical composition data for each of the multiple tobacco product raw materials. The TGA data indicates the relationship between the rate of mass change of each tobacco product raw material and temperature during pyrolysis. The second chemical composition data includes the content of each chemical component in each tobacco product raw material. The module further configures itself to determine a second volatile component score and a second fiber score for each tobacco product raw material based on the TGA data. It then determines a first volatile component score and a first fiber score based on the multiple second volatile component scores and multiple second fiber scores of the multiple tobacco product raw materials. Finally, it determines first chemical composition data based on the second chemical composition data and each candidate combination formulation. Finally, it determines the regional style of the candidate tobacco product based on the TGA data, the first chemical composition data, and each candidate combination formulation.

[0181] In some embodiments, the database construction module 401 is configured to perform fusion processing on multiple second volatile component scores to determine a first volatile component score; and to perform fusion processing on multiple second fiber type scores to determine a first fiber type score.

[0182] In some embodiments, the recommendation module 402 is configured to obtain the multidimensional features of the tobacco product to be tested corresponding to the second combination formula; calculate the similarity between the multidimensional features of the tobacco product to be tested and the multidimensional features of the tobacco product corresponding to each first combination formula; and determine the first target combination formula based on the similarity.

[0183] In some embodiments, the first target combination formulation includes multiple formulations. The recommendation module 402 is configured to determine the sensory evaluation results of each of the multiple formulations; determine the second target combination formulation from the multiple formulations based on the sensory evaluation results of each formulation; and determine the processing scheme corresponding to the second target combination formulation as the target processing scheme.

[0184] In some embodiments, the chemical components include total sugar and nicotine.

[0185] Figure 5 A block diagram illustrating a recommended apparatus for a processing scheme according to other embodiments of the present disclosure is shown.

[0186] like Figure 5 As shown, the recommended apparatus 500 of this embodiment includes a memory 501 and a processor 502 coupled to the memory 501. The processor 502 is configured to execute the recommended method in any embodiment of this disclosure based on instructions stored in the memory 501.

[0187] The memory 501 may include, for example, system memory, fixed non-volatile storage media, etc. The system memory may store, for example, an operating system, application programs, a boot loader, a database, and other programs.

[0188] Figure 6 A block diagram of a recommended apparatus for a processing scheme according to some embodiments of the present disclosure is shown.

[0189] like Figure 6 As shown, the recommended apparatus 600 of this embodiment includes a memory 601 and a processor 602 coupled to the memory 601. The processor 602 is configured to execute the recommended method in any of the foregoing embodiments based on instructions stored in the memory 601.

[0190] The memory 601 may include, for example, system memory, fixed non-volatile storage media, etc. The system memory may store, for example, an operating system, application programs, a boot loader, and other programs.

[0191] The recommended device 600 may also include an input / output interface 603, a network interface 604, and a storage interface 605. These interfaces 603, 604, and 605, as well as the memory 601 and processor 602, can be connected, for example, via a bus 606. Specifically, the input / output interface 603 provides a connection interface for input / output devices such as a monitor, mouse, keyboard, touchscreen, microphone, and speakers. The network interface 604 provides a connection interface for various networked devices. The storage interface 605 provides a connection interface for external storage devices such as SD cards and USB flash drives.

[0192] This disclosure also provides a computer-readable storage medium including computer program instructions that, when executed by a processor, implement the recommended method of any of the above embodiments.

[0193] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the recommended method of any of the above embodiments.

[0194] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0195] The recommended technical solutions for processing according to this disclosure have now been described in detail. To avoid obscuring the concept of this disclosure, some details known in the art have not been described. Those skilled in the art can fully understand how to implement the technical solutions disclosed herein based on the above description.

[0196] The methods and systems of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the specific order described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.

[0197] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A method for recommending a processing scheme, comprising: For each candidate combination formulation from a pool of candidate formulations, perform the following steps to build a formulation-processing scheme database: Obtain multidimensional features of the candidate tobacco products corresponding to each candidate combination formulation. The multidimensional features include multiple of the following: first volatile component score, first fiber score, first chemical component data, and origin style. The first chemical component data includes the content of each chemical component among the multiple chemical components of the candidate tobacco products. Based on the multidimensional characteristics of the candidate tobacco products, a plurality of first combination formulations are selected from the plurality of candidate combination formulations; Determine a processing scheme corresponding to each of the plurality of first combination formulations, the processing scheme including processing path and processing intensity; Each first combination formula and the corresponding processing scheme are stored in the combination formula-processing scheme database; The system retrieves a first target combination formula that matches the second combination formula to be processed from the combination formula-processing scheme database, and recommends a corresponding target processing scheme for the second combination formula based on the processing scheme corresponding to the first target combination formula.

2. The recommendation method according to claim 1, wherein, The multidimensional features include at least one of the first volatile component score, the first fiber score, and the first chemical composition data, and the place of origin style. The step of selecting multiple first combination formulations from the multiple candidate combination formulations based on the multidimensional characteristics of the candidate tobacco products includes: From the multiple candidate tobacco products corresponding to the multiple candidate combination formulations, multiple groups of first tobacco products that present multiple target origin styles are selected. Different groups of tobacco products present different target origin styles. Each group of first tobacco products includes multiple first tobacco products. Based on at least one of the first volatile component score, first fiber score, and first chemical component data of each of the plurality of first tobacco products in each group of first tobacco products, a plurality of second tobacco products are selected from each group of first tobacco products; The plurality of first combination formulations are determined based on the plurality of combination formulations corresponding to the plurality of second tobacco products.

3. The recommendation method according to claim 2, wherein, The step involves selecting multiple second tobacco products from each group of first tobacco products based on at least one of the first volatile component score, first fiber score, and first chemical composition data of each first tobacco product in each group of first tobacco products, to obtain multiple groups of second tobacco products including: The plurality of second tobacco products are selected from each group of first tobacco products based on at least one of the following differences: a first difference between the first volatile matter scores, a second difference between the first fiber scores, and a third difference between the first chemical composition data.

4. The recommendation method according to claim 2, wherein, Each group of first tobacco products comprises multiple subgroups, and the first tobacco products in different subgroups of each group of first tobacco products exhibit different probabilities of exhibiting the target origin style corresponding to each group of first tobacco products.

5. The recommended method according to any one of claims 1-4, wherein, The step of determining the processing scheme corresponding to each of the plurality of first combination formulations includes: A first processing experiment was performed on the tobacco products corresponding to each first combination formula according to different processing paths, and the sensory evaluation results of the tobacco products after the first processing experiment were obtained. Based on the sensory evaluation results of the tobacco products after the first processing experiment, the processing path corresponding to each first combination formula is determined.

6. The recommendation method according to any one of claims 1-4, wherein, The step of determining the processing scheme corresponding to each of the plurality of first combination formulations includes: A second processing experiment was performed on the tobacco products corresponding to each of the first combination formulations according to different processing intensities along the same processing path, and the sensory evaluation results of the tobacco products after the second processing experiment were obtained. Based on the sensory evaluation results of the tobacco products after the second processing experiment, the processing intensity corresponding to each first combination formula is determined.

7. The recommended method according to any one of claims 1-4, wherein, The candidate tobacco products include multiple tobacco product raw materials, and obtaining the multidimensional features of the candidate tobacco products corresponding to each candidate combination formulation includes: Thermogravimetric analysis data and second chemical composition data of each of the plurality of tobacco product raw materials are obtained. The thermogravimetric analysis data indicates the relationship between the mass change rate of each tobacco product raw material and temperature during the pyrolysis process. The second chemical composition data includes the content of each chemical component in each tobacco product raw material. The second volatile component score and the second fiber score of each tobacco product raw material are determined based on the thermogravimetric analysis data. The first volatile component score and the first fiber score are determined based on the multiple second volatile component scores and multiple second fiber scores of the multiple tobacco product raw materials; The first chemical component data is determined based on the second chemical component data and each candidate combination formulation; Based on the thermogravimetric analysis data, the first chemical composition data, and each candidate combination formulation, the origin style of the candidate tobacco products is determined.

8. The recommended method according to claim 7, wherein, The step of determining the first volatile component score and the first fiber score based on the multiple second volatile component scores and multiple second fiber scores of the multiple tobacco product raw materials includes: The scores of the multiple second volatile components are fused together to determine the score of the first volatile component; The scores of the multiple second fiber categories are fused to determine the score of the first fiber category.

9. The recommended method according to any one of claims 1-4, wherein, The step of retrieving a first target combined formulation that matches the second combined formulation to be processed from the combined formulation-processing scheme database includes: Obtain the multidimensional features of the tobacco product to be tested corresponding to the second combination formula; Calculate the similarity between the multidimensional features of the tobacco product to be tested and the multidimensional features of the tobacco product corresponding to each first combination formula; The first target combination formulation is determined based on the similarity.

10. The recommendation method according to any one of claims 1-4, wherein, The first target combination formulation includes multiple formulations. The step of recommending a target processing scheme for the second combination formula based on the processing scheme corresponding to the first target combination formula includes: Determine the sensory evaluation results for each of the plurality of formulations; Based on the sensory evaluation results of each formulation, a second target combination formulation is determined from the plurality of formulations; The processing scheme corresponding to the second target combination formula is determined as the target processing scheme.

11. The recommendation method according to any one of claims 1-4, wherein, The various chemical components include total sugars and nicotine.

12. A recommended apparatus for a processing scheme, comprising: The database building module is configured to perform the following steps to build a combination recipe-processing scheme database for each of a plurality of candidate combination recipes: Obtain multidimensional features of the candidate tobacco products corresponding to each candidate combination formulation. The multidimensional features include multiple of the following: first volatile component score, first fiber score, first chemical component data, and origin style. The first chemical component data includes the content of each chemical component among the multiple chemical components of the candidate tobacco products. Based on the multidimensional characteristics of the candidate tobacco products, a plurality of first combination formulations are selected from the plurality of candidate combination formulations; Determine a processing scheme corresponding to each of the plurality of first combination formulations, the processing scheme including processing path and processing intensity; Each first combination formula and the corresponding processing scheme are stored in the combination formula-processing scheme database; The recommendation module is configured to retrieve a first target combination formula that matches the second combination formula to be processed from the combination formula-processing scheme database, and recommend a corresponding target processing scheme for the second combination formula based on the processing scheme corresponding to the first target combination formula.

13. A recommended apparatus for a processing scheme, comprising: Memory; and A processor coupled to the memory, the processor being configured to perform the recommended method according to any one of claims 1-10 based on instructions stored in the memory.

14. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the recommended method according to any one of claims 1-10.

15. A computer program product comprising instructions that, when executed by a processor, cause the processor to perform the recommended method according to any one of claims 1-10.

Images