Cigarette formula design method and device based on multiple constraint conditions and storage medium

Abstract

The application discloses a cigarette formula design method and device based on multiple constraint conditions and a storage medium, and the method comprises the following steps: acquiring aroma type, price cost, inventory, production year, origin and grade data of single-grade tobacco to be used; setting constraint conditions for all single-grade tobacco to be used through preset formula cost, preset feeding amount and tobacco production year data to perform preliminary screening, and forming a candidate tobacco set; dividing the candidate tobacco set into four subsets according to the aroma type; setting an upper limit of the total grade number of tobacco used in the formula and an upper limit of the grade number of tobacco used in each aroma type, and randomly selecting the grade number of tobacco in each aroma type in the four subsets to combine into a formula; and based on the Monte Carlo method, the proportioning of each tobacco in the formula is optimized to meet the formula cost and feeding amount constraint conditions, and finally, a tobacco group formula is obtained. The application realizes the design of the tobacco group formula based on non-artificial experience, and improves the design efficiency of the tobacco group formula.

Description

Technical Field

[0001] This invention relates to the field of cigarette and cigarette product quality testing technology, and in particular to a method, apparatus and storage medium for designing cigarette leaf blend formulations based on multiple constraints. Background Technology

[0002] Cigarette blends are the foundation and core of cigarette production in tobacco companies. A cigarette blend is composed of various single-grade tobacco leaves mixed in specific proportions; typically, a finished cigarette blend consists of 15 to 30 different single-origin tobaccos blended in a certain ratio. During cigarette production, the cigarette blend is crucial for maintaining the stability of cigarette brand quality. Traditional cigarette blend maintenance is a complex task, requiring sensory evaluation and comprehensive assessment using measuring instruments to determine whether the sensory and smoke indicators of the replaced cigarette blend meet requirements. This usually requires experienced tasters with extensive training and practical experience. The health and mental state of the tasters also significantly impact the evaluation results. This method, relying on expert evaluation and analysis, is inefficient, time-consuming, and costly. Therefore, developing intelligent cigarette blend maintenance technologies and methods is necessary and urgent.

[0003] Over the years, cigarette companies have accumulated a wealth of basic and R&D data. While this data is highly valuable, it suffers from characteristics such as scattered storage, low integration, insufficient correlation, and poor consistency. In recent years, with the widespread application and development of machine learning, data mining, and artificial intelligence across various fields, utilizing these methods to mine and process data can uncover its immense value. This can enhance our understanding of the inherent laws governing tobacco production, provide support for cigarette blend design, and realize the digitalization and scientification of cigarette product design. This is of significant practical importance for cigarette companies to design and produce cigarette products scientifically and efficiently, avoid repetitive work, improve work efficiency, enhance the stability of cigarette production, thereby increasing market competitiveness and promoting sustainable development.

[0004] This invention aims to constrain factors in tobacco leaf design, such as tobacco cost, batch feed quantity, proportion of light-aroma tobacco, proportion of medium-aroma tobacco, proportion of strong-aroma tobacco, proportion of imported tobacco, number of formulation grades, and tobacco year, and recommend several leaf blend formulation sheets that meet the constraints. This will enable the design of cigarette leaf blend formulations based on non-human experience and improve the efficiency of cigarette leaf blend formulation design. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a cigarette formulation design method, apparatus, and storage medium based on multiple constraints, aiming to achieve cigarette leaf blend formulation design based on non-human experience and improve the efficiency of cigarette leaf blend formulation design.

[0006] Firstly, a method for cigarette formulation design based on multiple constraints is provided, including:

[0007] S1: Obtain data on the aroma type, price cost, inventory, production year, origin, and grade of the single-grade tobacco leaves to be used;

[0008] S2: By setting constraints on all single-grade tobacco leaves to be used through preset formula costs, preset feed amounts, and tobacco production year data, a preliminary screening is conducted to form a candidate tobacco leaf set.

[0009] S3: Based on the aroma of the tobacco leaves, the candidate tobacco leaf set is divided into four subsets: C 清 C 浓 C 中 C 进 Among them, C 清 Indicates the collection of candidate tobacco leaves for the light aroma type, C 浓 Indicates the set of candidate tobacco leaves for strong aroma, C 中 Indicates the set of candidate tobacco leaves for intermediate aroma type, C 进 This indicates a collection of imported candidate tobacco leaves;

[0010] S4: Set the upper limit on the total quantity of tobacco leaves of each grade used in the formula and the upper limit on the quantity of tobacco leaves of each aroma type used, in C 清 C 浓 C 中 C 进 From four subsets, randomly select N tobacco leaves of each aroma type. 清 1, N 浓 1, N 中 1, N 进 There are N, which are combined to form formula P; where N 清 N indicates the quantity of light-aroma tobacco leaves used in the formula. 浓 N indicates the quantity of aromatic tobacco leaves used in the formula. 中 N indicates the quantity of intermediate-aroma tobacco leaves used in the formula. 进 Indicates the grade and quantity of imported tobacco leaves used in the formula;

[0011] S5: Based on the Monte Carlo method, the proportion of each tobacco leaf in formula P is optimized to meet the preset formula cost and preset feed amount constraints, so as to obtain the final tobacco leaf group formula.

[0012] Furthermore, the aroma profile of the single-grade tobacco leaves to be used in step S1 is obtained through the following method:

[0013] Sensory evaluation of the single-grade tobacco leaves to be used was conducted. The evaluation mainly included scores for four aroma characteristics: light sweetness, honey sweetness, mellow sweetness, and caramel sweetness. Based on the regional information of the tobacco production area, the single-grade tobacco leaves to be used were classified according to the evaluation results as follows:

[0014] In the evaluation results, tobacco leaves with the highest score for sweet aroma are defined as light-aroma tobacco leaves; tobacco leaves with the highest score for caramel sweet aroma are defined as strong-aroma tobacco leaves; tobacco leaves with the highest score for honey sweet aroma or mellow sweet aroma are defined as medium-aroma tobacco leaves; and tobacco leaves from foreign countries are defined as imported tobacco leaves.

[0015] Furthermore, in step S2, constraints are set for all single-grade tobacco leaves to be used based on preset formula costs, tobacco leaf inventory, and tobacco leaf production year data. These constraints specifically include:

[0016] The production year of the required single-grade tobacco leaf x is manually set; the unit price of the required single-grade tobacco leaf x × a < the preset formula cost; the inventory of the required single-grade tobacco leaf x > the preset feed quantity × b; where a and b are empirical parameters.

[0017] Furthermore, the value range of a is [10, 50], and the value range of b is [0.01, 0.2].

[0018] Furthermore, in step S4, constraints are set on the total upper limit of the number of tobacco grades used in the formula and the upper limit of the number of tobacco grades used for each aroma type. These constraints specifically include:

[0019] N 清 +N 浓 +N 中 +N 进 ≤N; N 清 ≤ The maximum quantity of light-aroma tobacco leaves can be manually set, N 浓 ≤ The maximum quantity of strongly aromatic tobacco leaves can be manually set, N 中 ≤ The maximum number of intermediate-aroma tobacco leaf grades can be manually set, N 进 ≤ The maximum number of imported tobacco leaf grades can be manually set; where N is the preset maximum number of grades.

[0020] Further, step S5 specifically includes:

[0021] Randomly generate n sets of proportion data and assign them to formula P; where n is a preset value;

[0022] If each set of proportioning data fails to make formula P meet the preset formula cost and preset feed amount constraints, then discard the n sets of proportioning data, regenerate n sets of proportioning data randomly, and determine whether each set of proportioning data makes formula P meet the preset formula cost and preset feed amount constraints; otherwise, for the proportioning data that meet the preset formula cost and preset feed amount constraints, assign one set to formula P by averaging or randomly selecting one set, to obtain the final tobacco leaf formula.

[0023] Secondly, a cigarette formulation design device based on multiple constraints is provided, comprising:

[0024] The data acquisition module is used to acquire data on the aroma, price, inventory, production year, origin, and grade of the single-grade tobacco leaves to be used.

[0025] The initial screening module is used to initially screen all single-grade tobacco leaves to be used by setting constraints based on preset formula costs, preset feed amounts, and tobacco production year data, forming a candidate tobacco leaf set.

[0026] The candidate tobacco leaf set partitioning module is used to divide the candidate tobacco leaf set into four subsets based on the aroma type of the tobacco leaves: C 清 C 浓 C 中 C 进 Among them, C 清 Indicates the collection of candidate tobacco leaves for the light aroma type, C 浓 Indicates the set of candidate tobacco leaves for strong aroma, C 中 Indicates the set of candidate tobacco leaves for intermediate aroma type, C 进 Indicates a collection of imported candidate tobacco leaves;

[0027] The recipe generation module is used to set the upper limit on the total quantity of tobacco leaves of each grade used in the recipe, as well as the upper limit on the quantity of tobacco leaves of each aroma type used. (In C...) 清 C 浓 C 中 C 进 From four subsets, randomly select N tobacco leaves of each aroma type. 清 1, N 浓 1, N 中 1, N 进 There are N, which are combined to form formula P; where N 清 N indicates the quantity of light-aroma tobacco leaves used in the formula. 浓 N indicates the quantity of aromatic tobacco leaves used in the formula. 中 N indicates the quantity of intermediate-aroma tobacco leaves used in the formula. 进 Indicates the grade and quantity of imported tobacco leaves used in the formula;

[0028] The formulation optimization module is used to optimize the proportion of each tobacco leaf in formulation P based on the Monte Carlo method, so that it meets the preset formulation cost and preset feed amount constraints, and obtains the final tobacco leaf group formulation.

[0029] Thirdly, a cigarette formulation design device based on multiple constraints is provided, comprising:

[0030] A memory that stores computer programs;

[0031] A processor is used to execute the computer program to implement the cigarette formulation design method based on multiple constraints as described above.

[0032] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the cigarette formulation design method based on multiple constraints as described above.

[0033] Beneficial effects

[0034] This invention proposes a cigarette formulation design method, device, and storage medium based on multiple constraints. By constraining factors in tobacco leaf design, such as tobacco cost, batch feed quantity, proportion of light-aroma tobacco, proportion of medium-aroma tobacco, proportion of strong-aroma tobacco, proportion of imported tobacco, number of formulation grades, and tobacco year, a cigarette leaf blend formulation sheet that meets the constraints is generated. This achieves cigarette leaf blend formulation design based on non-human experience, thereby improving the efficiency of cigarette leaf blend formulation design. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart of a cigarette leaf blend formulation design method based on multiple constraints provided in an embodiment of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0038] like Figure 1 As shown, one embodiment of the present invention provides a cigarette formulation design method based on multiple constraints, including:

[0039] S1: Obtain the aroma type, price cost, inventory, production year, origin, and grade data of the single-grade tobacco leaves to be used.

[0040] The aroma profile of the single-grade tobacco leaves to be used is obtained through the following method:

[0041] Sensory evaluation of the single-grade tobacco leaves to be used was conducted. The evaluation mainly included scores for four aroma characteristics: light sweetness, honey sweetness, mellow sweetness, and caramel sweetness. Based on the regional information of the tobacco production area, the single-grade tobacco leaves to be used were classified according to the evaluation results as follows:

[0042] ① The tobacco leaves with the highest score in sweet aroma in the evaluation results are defined as light-aroma tobacco leaves; ② The tobacco leaves with the highest score in caramel sweet aroma in the evaluation results are defined as strong-aroma tobacco leaves; ③ The tobacco leaves with the highest score in honey sweet aroma or mellow sweet aroma in the evaluation results are defined as medium-aroma tobacco leaves; ④ Tobacco leaves produced abroad are defined as imported tobacco leaves.

[0043] S2: Preliminary screening of all single-grade tobacco leaves to be used is conducted by setting constraints based on preset formula costs, preset feed amounts, and tobacco production year data, forming a candidate tobacco leaf set. The specific process includes:

[0044] Preset conditions are set as follows: ① The production year of the required single-grade tobacco leaf x is manually set; ② The unit price of the required single-grade tobacco leaf x × a < the preset formula cost; ③ The inventory of the required single-grade tobacco leaf x × b > the preset feed quantity; where a and b are empirical parameters, a represents the empirical value of the number of single-grade tobacco leaves in the formula, with a value range of [10, 50], and b represents the empirical value of the proportion of the minimum single-grade tobacco leaf used in the formula, with a value range of [0.01, 0.2]. In this embodiment, a = 33, b = 2%. Single-grade tobacco leaf x that meets the above constraints enters the candidate tobacco leaf set, denoted as C.

[0045] S3: Based on the aroma of the tobacco leaves, the candidate tobacco leaf set is divided into four subsets: C 清 C 浓 C 中 C 进 Among them, C 清 Indicates the collection of candidate tobacco leaves for the light aroma type, C 浓 Indicates the set of candidate tobacco leaves for strong aroma, C 中 Indicates the set of candidate tobacco leaves for intermediate aroma type, C 进 This indicates a collection of imported candidate tobacco leaves.

[0046] S4: Set the upper limit on the total quantity of tobacco leaves of each grade used in the formula and the upper limit on the quantity of tobacco leaves of each aroma type used, in C 清 C 浓 C 中 C 进 From four subsets, randomly select N tobacco leaves of each aroma type. 清 1, N 浓 1, N 中 1, N 进 There are N, which are combined to form formula P; where N 清 N indicates the quantity of light-aroma tobacco leaves used in the formula.浓 N indicates the quantity of aromatic tobacco leaves used in the formula. 中 N indicates the quantity of intermediate-aroma tobacco leaves used in the formula. 进 This indicates the quantity of imported tobacco leaf grades used in the formula. It includes constraints setting upper limits on the total quantity of tobacco leaf grades used in the formula and the quantity of each aroma type of tobacco leaf used. These constraints specifically include:

[0047] ①N 清 +N 浓 +N 中 +N 进 ≤N, where N is the maximum number of levels; ②N 清 ≤ The maximum quantity of light-aroma tobacco leaves can be manually set, N 浓 ≤ The maximum quantity of strongly aromatic tobacco leaves can be manually set, N 中 ≤ The maximum number of intermediate-aroma tobacco leaf grades can be manually set, N 进 ≤ The maximum quantity of imported tobacco leaves of different grades can be manually set.

[0048] S5: Based on the Monte Carlo method, the proportions of each tobacco leaf in formula P are optimized to meet the preset formula cost and preset feed quantity constraints, resulting in the final cigarette leaf blend formula. Specifically, this includes:

[0049] Randomly generate n sets of proportion data and assign them to formula P; where n is a preset value, preferably n>1000;

[0050] If each set of proportioning data fails to make formula P meet the preset formula cost and preset feed amount constraints, then discard the n sets of proportioning data, regenerate n sets of proportioning data randomly, and determine whether each set of proportioning data makes formula P meet the preset formula cost and preset feed amount constraints; otherwise, for the proportioning data that meet the preset formula cost and preset feed amount constraints, assign one set to formula P by averaging or randomly selecting one set, to obtain the final tobacco leaf formula.

[0051] The preset formula cost and preset feed quantity constraints here correspond to the preset formula cost and preset feed quantity in step S2, that is, the sum of the product of the unit price and the ratio of each grade of tobacco leaf in the formula does not exceed the preset formula cost, and the inventory of each grade of tobacco leaf is not less than the product of the preset feed quantity and its ratio.

[0052] Another embodiment of the present invention provides a cigarette formulation design device based on multiple constraints, comprising:

[0053] The data acquisition module is used to acquire data on the aroma, price, inventory, production year, origin, and grade of the single-grade tobacco leaves to be used.

[0054] The initial screening module is used to initially screen all single-grade tobacco leaves to be used by setting constraints based on preset formula costs, preset feed amounts, and tobacco production year data, forming a candidate tobacco leaf set.

[0055] The candidate tobacco leaf set partitioning module is used to divide the candidate tobacco leaf set into four subsets based on the aroma type of the tobacco leaves: C 清 C 浓 C 中 C 进 Among them, C 清 Indicates the collection of candidate tobacco leaves for the light aroma type, C 浓 Indicates the set of candidate tobacco leaves for strong aroma, C 中 Indicates the set of candidate tobacco leaves for intermediate aroma type, C 进 Indicates a collection of imported candidate tobacco leaves;

[0056] The recipe generation module is used to set the upper limit on the total quantity of tobacco leaves of each grade used in the recipe, as well as the upper limit on the quantity of tobacco leaves of each aroma type used. (In C...) 清 C 浓 C 中 C 进 From four subsets, randomly select N tobacco leaves of each aroma type. 清 1, N 浓 1, N 中 1, N 进 There are N, which are combined to form formula P; where N 清 N indicates the quantity of light-aroma tobacco leaves used in the formula. 浓 N indicates the quantity of aromatic tobacco leaves used in the formula. 中 N indicates the quantity of intermediate-aroma tobacco leaves used in the formula. 进 Indicates the grade and quantity of imported tobacco leaves used in the formula;

[0057] The formulation optimization module is used to optimize the proportion of each tobacco leaf in formulation P based on the Monte Carlo method, so that it meets the preset formulation cost and preset feed amount constraints, and obtains the final tobacco leaf group formulation.

[0058] Other specific implementations in this embodiment can be found in the foregoing embodiments, and will not be repeated here.

[0059] Another embodiment of the present invention provides a cigarette formulation design device based on multiple constraints, comprising:

[0060] A memory that stores computer programs;

[0061] A processor is used to execute the computer program to implement the cigarette formulation design method based on multiple constraints as described in the above embodiments.

[0062] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the cigarette formulation design method based on multiple constraints as described in the above embodiment.

[0063] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0064] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0065] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0066] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0067] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0068] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0069] The technical solution of the present invention will be further explained below with reference to a specific example:

[0070] Example

[0071] 1. Obtain data on 500 single-grade tobacco leaves from major domestic and international tobacco-producing regions in Hunan Tobacco's inventory, including price, cost, inventory, production year, origin, and grade.

[0072] 2. Sensory evaluation was conducted on the above 500 single-grade tobacco leaves to obtain the aroma scores of four types of aroma: light sweetness, honey sweetness, mellow sweetness, and caramel sweetness. Based on the aroma scores and origin information of each type of tobacco leaf, according to the aroma classification standard of this invention, each grade of tobacco leaf was classified into light aroma type, strong aroma type, intermediate aroma type, and imported tobacco leaf.

[0073] 3. The following constraints are set for the leaf blend formulation: the tobacco leaves used in the formulation are produced in 2016, 2017, and 2018; the preset formulation cost is 4000 yuan / large carton; the preset feed amount is 3000 kg; the maximum number of tobacco grades N used in the formulation is 30; and the maximum number of light-aroma tobacco grades N used in the formulation is [not specified]. 清 The formula contains 15 grades of aromatic tobacco leaves, N. 浓 The value is 3, and the number of grades N of intermediate-aroma tobacco leaves used in the formula is 3. 中 The formula contains 9 grades of imported tobacco leaves, N. 进 It is 3;

[0074] 4. Using the leaf group formulation design scheme provided in this invention, the following leaf group formulation sheet is finally obtained:

[0075]

[0076]

[0077] According to the formula of the present invention, the production year and inventory of tobacco leaves, the grade and quantity of each type of tobacco leaf, the actual formula cost, the overall formula structure, etc., all meet the requirements and basically meet the usage standards of cigarette leaf group formula.

[0078] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A cigarette formulation design method based on multiple constraints, characterized in that, include: S1: Obtain data on the aroma type, price cost, inventory, production year, origin, and grade of the single-grade tobacco leaves to be used; S2: By setting constraints on all single-grade tobacco leaves to be used through preset formula costs, preset feed amounts, and tobacco production year data, a preliminary screening is conducted to form a candidate tobacco leaf set. S3: Based on the aroma of the tobacco leaves, the candidate tobacco leaf set is divided into four subsets: C 清 C 浓 C 中 C 进 ; Among them, C 清 Indicates the collection of candidate tobacco leaves for the light aroma type, C 浓 Indicates the set of candidate tobacco leaves for strong aroma, C 中 Indicates the set of candidate tobacco leaves for intermediate aroma type, C 进 Indicates a collection of imported candidate tobacco leaves; S4: Set the upper limit on the total quantity of tobacco leaves of each grade used in the formula and the upper limit on the quantity of tobacco leaves of each aroma type used, in C 清 C 浓 C 中 C 进 From four subsets, randomly select N tobacco leaves of each aroma type. 清 1, N 浓 1, N 中 1, N 进 There are N, which are combined to form formula P; where N 清 N indicates the quantity of light-aroma tobacco leaves used in the formula. 浓 N indicates the quantity of aromatic tobacco leaves used in the formula. 中 N indicates the quantity of intermediate-aroma tobacco leaves used in the formula. 进 Indicates the grade and quantity of imported tobacco leaves used in the formula; S5: Based on the Monte Carlo method, the proportion of each tobacco leaf in formula P is optimized to meet the preset formula cost and preset feed amount constraints, so as to obtain the final tobacco leaf group formula. Step S5 specifically includes: Randomly generate n sets of proportion data and assign them to formula P; where n is a preset value; If each set of proportioning data fails to make formula P meet the preset formula cost and preset feed amount constraints, then discard the n sets of proportioning data, regenerate n sets of proportioning data randomly, and determine whether each set of proportioning data makes formula P meet the preset formula cost and preset feed amount constraints; otherwise, for the proportioning data that meet the preset formula cost and preset feed amount constraints, assign one set to formula P by averaging or randomly selecting one set, to obtain the final tobacco leaf formula.

2. The cigarette formulation design method based on multiple constraints according to claim 1, characterized in that, The aroma profile of the single-grade tobacco leaves to be used in step S1 is obtained through the following method: Sensory evaluation was conducted on the single-grade tobacco leaves to be used. The evaluation included scores for four aroma characteristics: light sweetness, honey sweetness, mellow sweetness, and caramel sweetness. Based on the regional information of the tobacco production area, the single-grade tobacco leaves to be used were classified according to the evaluation results as follows: In the evaluation results, tobacco leaves with the highest score for sweet aroma are defined as light-aroma tobacco leaves; tobacco leaves with the highest score for caramel sweet aroma are defined as strong-aroma tobacco leaves; tobacco leaves with the highest score for honey sweet aroma or mellow sweet aroma are defined as medium-aroma tobacco leaves; and tobacco leaves from foreign countries are defined as imported tobacco leaves.

3. The cigarette formulation design method based on multiple constraints according to claim 1, characterized in that, In step S2, constraints are set for all single-grade tobacco leaves to be used based on preset formula costs, tobacco leaf inventory, and tobacco leaf production year data. These constraints specifically include: Manually set the production year of the required single-grade tobacco leaves (x); the unit price of the required single-grade tobacco leaves (x). a <Preset formula cost; required inventory of single-grade tobacco leaves>Preset feed quantity b ;in, a and b All of these are empirical parameters.

4. The cigarette formulation design method based on multiple constraints according to claim 3, characterized in that, a The value range is [10, 50]. b The value range is [0.01, 0.2].

5. The cigarette formulation design method based on multiple constraints according to claim 1, characterized in that, In step S4, constraints are set regarding the upper limit of the total quantity of tobacco leaves of each grade used in the formula and the upper limit of the quantity of tobacco leaves of each aroma type. These constraints specifically include: N 清 +N 浓 +N 中 +N 进 ≤N; N 清 ≤ The maximum quantity of light-aroma tobacco leaves can be manually set, N 浓 ≤ The maximum quantity of strongly aromatic tobacco leaves can be manually set, N 中 ≤ The maximum number of intermediate-aroma tobacco leaf grades can be manually set, N 进 ≤ The maximum number of imported tobacco leaf grades can be manually set; where N is the preset maximum number of grades.

6. A cigarette formulation design device based on multiple constraints, characterized in that, include: The data acquisition module is used to acquire data on the aroma, price, inventory, production year, origin, and grade of the single-grade tobacco leaves to be used. The initial screening module is used to initially screen all single-grade tobacco leaves to be used by setting constraints based on preset formula costs, preset feed amounts, and tobacco production year data, forming a candidate tobacco leaf set. The candidate tobacco leaf set partitioning module is used to divide the candidate tobacco leaf set into four subsets based on the aroma type of the tobacco leaves: C 清 C 浓 C 中 C 进 ; Among them, C 清 Indicates the collection of candidate tobacco leaves for the light aroma type, C 浓 Indicates the set of candidate tobacco leaves for strong aroma, C 中 Indicates the set of candidate tobacco leaves for intermediate aroma type, C 进 Indicates a collection of imported candidate tobacco leaves; The recipe generation module is used to set the upper limit on the total quantity of tobacco leaves of each grade used in the recipe, as well as the upper limit on the quantity of tobacco leaves of each aroma type used. (In C...) 清 C 浓 C 中 C 进 From four subsets, randomly select N tobacco leaves of each aroma type. 清 1, N 浓 1, N 中 1, N 进 There are N, which are combined to form formula P; where N 清 N indicates the quantity of light-aroma tobacco leaves used in the formula. 浓 N indicates the quantity of aromatic tobacco leaves used in the formula. 中 N indicates the quantity of intermediate-aroma tobacco leaves used in the formula. 进 Indicates the grade and quantity of imported tobacco leaves used in the formula; The formulation optimization module, based on the Monte Carlo method, optimizes the proportions of each tobacco leaf in formulation P to meet preset formulation cost and preset feed quantity constraints, thus obtaining the final cigarette leaf blend formulation; specifically including: Randomly generate n sets of proportion data and assign them to formula P; where n is a preset value; If each set of proportioning data fails to make formula P meet the preset formula cost and preset feed amount constraints, then discard the n sets of proportioning data, regenerate n sets of proportioning data randomly, and determine whether each set of proportioning data makes formula P meet the preset formula cost and preset feed amount constraints; otherwise, for the proportioning data that meet the preset formula cost and preset feed amount constraints, assign one set to formula P by averaging or randomly selecting one set, to obtain the final tobacco leaf formula.

7. A cigarette formulation design device based on multiple constraints, characterized in that, include: A memory that stores computer programs; A processor is configured to execute the computer program to implement the cigarette formulation design method based on multiple constraints as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the cigarette formulation design method based on multiple constraints as described in any one of claims 1 to 5.

Images