Cigar flavor dynamic regulation and control system and method

Abstract

The invention discloses a cigar flavor dynamic regulation and control system and method, and relates to the technical field of cigar production. Comprises: a flavor analysis module for analyzing volatile components in cigar tobacco leaves, performing sensory evaluation on the overall flavor of cigars, and establishing a cigar flavor fingerprint spectrum; the demand acquisition module is used for acquiring demands of current consumers and collecting personal information and cigar consumption habit data of the consumers; and the consumer habit database is used for storing personal information of consumers and cigar consumption habit data, and establishing a correlation model between consumer habits and cigar flavor characteristics through a data mining and analysis technology. According to the system, a personalized cigar formula can be generated according to personal information, consumption habits and specific preferences of consumers, the personalized requirements of the consumers for flavor are met, and the consumption experience is improved; through a mathematical model of entrance depth and flavor perception, in combination with personal habits of consumers, formula adjustment strategies are refined, and the sense of participation and satisfaction of the consumers are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of cigar production, and in particular to a system and method for dynamically controlling cigar flavor. Background Art

[0002] The traditional cigar manufacturing industry has long relied on inherited experience and craftsmanship. While this ensures the unique appeal of its products, it has struggled to meet the increasingly diverse demands of consumers. With technological advancements and consumers' pursuit of personalized, high-quality products, the cigar industry urgently needs new technologies that can precisely control flavor and rapidly respond to market changes.

[0003] A search revealed Chinese patent application number CN202111632686.2, which discloses a machine vision-based device and method for inspecting and sorting the appearance quality of cigars. The device comprises a cigar sampling system, a roller blanking system, a cigar sampling platform, a cigar storage system, an image acquisition and detection system, and a control system. The method utilizes two rotating rollers on the sampling platform to rotate the cigar sample in the same direction. A CCD industrial camera captures the cigar surface image, which is then processed and stitched together using a computer image processing system to obtain a panoramic image of the cigar surface. The cigar image is then inspected for noise and color, and the length and width of the cigars are calculated to identify defective and qualified cigars. Finally, a control system controls the direction and speed of the two rotating rollers to sort the defective and qualified cigars into different storage boxes. The sorting device described in the aforementioned patent has the following drawbacks: while it enables intelligent production, it lacks flavor customization and cannot meet the individual needs of different consumers. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a cigar flavor dynamic control system and method.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A cigar flavor dynamic control system, comprising: The flavor analysis module analyzes the volatile components in cigar tobacco leaves, performs sensory evaluation on the overall flavor of the cigar, and establishes a cigar flavor fingerprint; Demand collection module, used to collect current consumer demand, personal information and cigar consumption habit data; A consumer habits database stores consumers' personal information and cigar consumption habits data, and uses data mining and analysis techniques to establish a correlation model between consumer habits and cigar flavor characteristics; The recipe generation module generates personalized cigar recipes based on the association model in the consumer habit database and the specific information and preferences of the current consumer; Production management module, which manages production based on the generated personalized cigar recipe; The feedback adjustment module conducts flavor testing and consumer trial evaluation on the product after the cigar production is completed, and provides feedback and adjustments.

[0006] Preferably, the flavor analysis module analyzes volatile components in the following manner: The volatile components in cigar tobacco leaves were analyzed by GC-MS. Assume that the content of a certain flavor substance in the cigar tobacco leaf sample is , where i = 1, 2, ..., n, n is the number of flavor substances, and the peak area of ​​the substance is obtained by GC-MS detection , the content was calculated using the standard curve method; the standard curve equation is ,in and It is a coefficient obtained by measuring a standard of known concentration.

[0007] Preferably, the flavor analysis module performs sensory evaluation in the following manner: The overall flavor of cigars is evaluated using an electronic nose and an electronic tongue. The sensor array based on the electronic nose has different response values ​​to different odor molecules. Let sensor j, where j = 1, 2, ..., m, where m is the number of sensors, and the response value to the cigar sample is , the response value is processed by principal component analysis method to obtain the principal component score representing the cigar odor characteristics ; The principal component score of the electronic tongue representing the taste characteristics was obtained through principal component analysis , where k=1,2,…,p, and p is the number of electronic tongue sensors; the data from GC-MS, electronic nose, and electronic tongue are integrated to establish a cigar flavor fingerprint.

[0008] Preferably, the demand collection module collects consumers' personal information and cigar consumption habit data as follows: Consumer information includes age, gender, years of smoking, and region. Let age be Y, gender coefficient be G, years of smoking be X, and region coefficient be D. Cigar consumption habit data includes favorite cigar brands, common sizes, daily smoking volume, and preference ratings for descriptors of different flavors.

[0009] Preferably, when establishing a correlation model between consumer habits and cigar flavor characteristics in the consumer habit database, the details are as follows: Assume that the preference score for richness is , the light preference score is , the sweetness preference score is , the spiciness preference score is ; Through data mining and analysis technology, establish the correlation model between consumer habits and cigar flavor characteristics; use the multivariate linear regression model to establish the consumer's preference for a certain flavor Relationship with consumer information and consumption habit data: ; in: 、 、 、 、 、 、 、 、 is the regression coefficient, is the random error term.

[0010] Preferably, the recipe generation module generates a personalized cigar recipe in the following manner: Assume that the type ratio vector of tobacco raw materials in the generated formula is , the additive type ratio vector is ; Recipe generation is based on the following optimization objective function: ; Among them, q is the number of tobacco raw material types, r is the number of additive types, is the actual content of flavor substance i in the cigar after production according to the recipe, The expected content is determined based on consumer preferences; and are the principal component scores of the actual and expected electronic nose sensor j, respectively; and are the actual and expected principal component scores of the electronic tongue sensor k, respectively; λ and μ are weight coefficients used to balance the importance of chemical component content and sensory evaluation indicators; by solving this optimization problem, the optimal formula that meets the personalized needs of consumers is obtained.

[0011] Preferably, the feedback adjustment module is: After cigar production is completed, the product is subjected to flavor testing and consumer trial evaluation; the test results are compared with the expected values ​​in the recipe generation module to calculate the flavor deviation : ; At the same time, feedback from consumers who tried the product, including satisfaction ratings on the flavor, was collected. , improvement suggestions; adjust and optimize the parameters of the recipe generation module based on flavor deviation and consumer feedback.

[0012] Preferably, when performing feedback adjustment, the feedback adjustment module further includes establishing a mathematical model of flavor score and inlet depth through trial puff experiments and using data analysis methods, and adjusting the strategy, as follows: Through consumer test smoking experiments, data were collected at different entrance depths. Under this circumstance, the consumer's rating data for various flavors of cigars; let the consumer's rating for the kth flavor at different entrance depths be , where k = 1, 2, …, s, s is the number of flavor types; Using data analysis methods, a mathematical model of flavor score and depth of entry was established, and multiple regression analysis was performed to obtain the following: ; in, 、 、 、 、 is the regression coefficient, is a random error term; through the model, we can understand the influence of different entrance depths on the perception of various flavors.

[0013] Preferably, the feedback adjustment module adjusts the recipe according to the consumer's entry depth when performing feedback adjustment; and formulates a personalized recipe adjustment strategy based on the personal information and preferences in the consumer habit database and the association model between entry depth and flavor.

[0014] Preferably, the control method of the system comprises the following steps: S1: Data acquisition and analysis; S11: Collect flavor data of cigars of different brands and origins, analyze them using GC-MS, electronic nose, and electronic tongue devices in the flavor analysis module, and establish a cigar flavor database; S12: Collect consumers’ personal information and cigar consumption habit data, clean, organize and analyze the data, and build a consumer habit database; S2: Formulation design and production; S21: The demand collection module collects consumers' personalized customization needs and inputs consumers' relevant information and preference data into the system; S22: The recipe generation module generates a personalized cigar recipe based on the association model in the consumer habit database and the optimization objective function; S23: The production management module produces corresponding products based on the generated personalized cigar recipe; S3: Quality inspection and feedback; S31: After production, the customized cigars undergo flavor testing, including GC-MS analysis, electronic nose and electronic tongue evaluation, and consumer trial smoking. S32: The feedback adjustment module calculates the flavor deviation and satisfaction score based on the test results and consumer feedback; if the product does not meet the requirements, the relevant information is fed back to the recipe generation module, and the parameters are adjusted and re-produced until the flavor effect that satisfies the consumer is achieved.

[0015] The beneficial effects of the present invention are: 1. The system of the present invention can generate personalized cigar recipes based on consumers' personal information, consumption habits and specific preferences, meet consumers' personalized needs for flavor, and enhance the consumption experience.

[0016] 2. The present invention combines analytical techniques such as GC-MS, electronic nose, and electronic tongue to comprehensively analyze the volatile components and sensory characteristics of cigars, establish an accurate flavor fingerprint, and provide a scientific basis for formula design.

[0017] 3. This invention uses big data analysis technology to explore the correlation model between consumer habits and cigar flavor characteristics, realize intelligent and precise formula generation, and improve production efficiency and product quality.

[0018] 4. Through post-production quality testing and consumer test-sniffing feedback, the system can calculate flavor deviation in real time and dynamically adjust the formula and production process based on the feedback results to ensure that the product continues to meet consumer expectations. By using a mathematical model of ingestion depth and flavor perception, combined with consumer personal habits, the formula adjustment strategy is further refined to enhance consumer participation and satisfaction, thereby improving brand image.

[0019] 5. During the formula generation process, the present invention comprehensively considers the chemical component content, sensory evaluation indicators and consumer preferences, and achieves comprehensive optimization of product flavor and quality through a multi-objective optimization algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a framework diagram of a cigar flavor dynamic control system proposed by the present invention; Figure 2 This is a flow chart of a method for dynamically controlling cigar flavor proposed by the present invention. DETAILED DESCRIPTION

[0021] The technical solution of the present invention will be further described in detail below in conjunction with specific implementation methods. Example

[0022] A cigar flavor dynamic control system, comprising: The flavor analysis module analyzes the volatile components in cigar tobacco leaves, performs sensory evaluation of the overall flavor of the cigar, and establishes a cigar flavor fingerprint for subsequent flavor analysis and regulation; Demand collection module, used to collect current consumer demand, personal information and cigar consumption habit data; A consumer habits database stores consumers' personal information and cigar consumption habits data, and uses data mining and analysis techniques to establish a correlation model between consumer habits and cigar flavor characteristics; The recipe generation module generates personalized cigar recipes based on the association model in the consumer habit database and the specific information and preferences of the current consumer; Production management module, which manages production based on the generated personalized cigar recipe; The feedback adjustment module conducts flavor testing and consumer trial evaluation on the product after the cigar production is completed, and provides feedback and adjustments.

[0023] The flavor analysis module analyzes volatile components in the following manner: The volatile components in cigar tobacco leaves were analyzed using GC-MS (gas chromatography-mass spectrometry); the content of a certain flavor substance in the cigar tobacco leaf sample was (i=1,2,…,n, n is the number of flavor substances), the peak area of ​​the substance is obtained by GC-MS detection , the content was calculated using the standard curve method; the standard curve equation is ,in and It is a coefficient obtained by measuring a standard of known concentration.

[0024] The flavor analysis module performs sensory evaluation in the following manner: The overall flavor of cigars is evaluated using an electronic nose and an electronic tongue. The sensor array based on the electronic nose has different response values ​​to different odor molecules. Suppose the response value of sensor j (j = 1, 2, ..., m, where m is the number of sensors) to the cigar sample is , the response values ​​are processed by methods such as principal component analysis (PCA) to obtain the principal component scores that characterize the cigar odor characteristics The electronic tongue is similar to the one used to obtain the principal component score representing the taste characteristics. (k=1,2,…,p, where p is the number of electronic tongue sensors). By integrating data from GC-MS, electronic nose, and electronic tongue, a cigar flavor fingerprint was established for subsequent flavor analysis and control.

[0025] The demand collection module collects consumers' personal information and cigar consumption habit data as follows: Consumer information includes basic information such as age, gender, years of smoking, and region. Let age be Y, gender coefficient (male is 1, female is 0) be G, years of smoking be X, and region coefficient (based on different regions, such as North China is 1, South China is 2, etc.) be D; Cigar consumption habit data includes favorite cigar brands, common sizes, daily smoking volume, and preference scores for different flavor descriptors (such as strong, light, sweet, spicy, etc.) (scoring range is 1-5 points).

[0026] The consumer habit database, when establishing a correlation model between consumer habits and cigar flavor characteristics, is specifically as follows: Assume that the preference score for richness is , the light preference score is , the sweetness preference score is , the spiciness preference score is etc.; establish the correlation model between consumer habits and cigar flavor characteristics through data mining and analysis technology; use the multivariate linear regression model to establish the consumer preference for a certain flavor Relationship with consumer information and consumption habit data: ; in: is the regression coefficient, is the random error term.

[0027] The recipe generation module generates a personalized cigar recipe in the following manner: Assume that the type ratio vector of tobacco raw materials in the generated formula is (q is the number of tobacco leaf raw material types), the additive type ratio vector is (r is the number of additive types); recipe generation is based on the following optimization objective function: ; in, is the actual content of flavor substance i in the cigar after production according to the recipe, The expected content is determined based on consumer preferences; and are the principal component scores of the actual and expected electronic nose sensor j, respectively; and are the actual and expected principal component scores of the electronic tongue sensor k, respectively; λ and μ are weight coefficients used to balance the importance of chemical component content and sensory evaluation indicators; by solving this optimization problem, the optimal formula that meets the personalized needs of consumers is obtained.

[0028] The feedback adjustment module is specifically: After cigar production is completed, the product is subjected to flavor testing and consumer trial evaluation; the test results are compared with the expected values ​​in the recipe generation module to calculate the flavor deviation : ; At the same time, feedback from consumers who tried the product, including satisfaction ratings on the flavor, was collected. (scoring range is 1-10 points), improvement suggestions, etc.; according to the flavor deviation and consumer feedback, the parameters of the recipe generation module are adjusted and optimized.

[0029] For example, if the flavor deviation exceeds the set threshold δ and the consumer's satisfaction with a certain flavor is low, the proportion of relevant raw materials or processing parameters in the formula will be adjusted, and production and testing will be carried out again until the consumer's personalized needs are met.

[0030] Example 2: A cigar flavor dynamic control system is provided. This embodiment is based on the first embodiment. The feedback adjustment module in this embodiment further includes establishing a mathematical model of flavor score and inhalation depth through trial puff experiments and utilizing data analysis methods, and adjusting the strategy as follows: Through consumer test smoking experiments, data were collected at different entrance depths. Under this condition, the consumer's rating data on various flavors of cigars (such as richness, fragrance, sweetness, etc.); let the consumer's rating of the kth flavor at different entrance depths be (k=1,2,…,s, s is the number of flavor types); Using data analysis methods, a mathematical model of flavor score and depth of entry was established, and multiple regression analysis was performed to obtain the following: ; in, are regression coefficients, is a random error term; through these models, we can gain a deeper understanding of the influence of different entrance depths on the perception of various flavors.

[0031] Adjust the recipe according to the depth of consumer intake; formulate a personalized recipe adjustment strategy based on the personal information and preferences in the consumer habit database, as well as the association model between intake depth and flavor; if the consumer has a deep intake habit and prefers a strong flavor, appropriately increase the proportion of tobacco raw materials or the amount of additives with strong flavor release characteristics in the recipe generation module; suppose the proportion of a certain strong flavor tobacco raw material in the original recipe is , the ratio adjusted according to the inlet depth is ,in, is the adjustment amount calculated according to the model, which is related to the inlet depth It is related to the consumer's preference for the flavor, e.g. ( Score consumers' preference for strong flavors. 、 is the adjustment factor).

[0032] Example 3: A method for dynamically controlling cigar flavor comprises the following steps: S1: Data acquisition and analysis; S11: Collect flavor data of cigars of different brands and origins, analyze them using GC-MS, electronic nose, and electronic tongue devices in the flavor analysis module, and establish a cigar flavor database; S12: Collect consumers’ personal information and cigar consumption habit data, clean, organize and analyze the data, and build a consumer habit database; S2: Formulation design and production; S21: The demand collection module collects consumers' personalized customization needs and inputs consumers' relevant information and preference data into the system; S22: The recipe generation module generates a personalized cigar recipe based on the association model in the consumer habit database and the optimization objective function; S23: The production management module produces corresponding products based on the generated personalized cigar recipe; S3: Quality inspection and feedback; S31: After production is complete, the customized cigars undergo flavor testing, including GC-MS analysis, electronic nose and electronic tongue evaluation, and consumer trial smoking. S32: The feedback adjustment module calculates the flavor deviation and satisfaction score based on the test results and consumer feedback; if the product does not meet the requirements, the relevant information is fed back to the recipe generation module, and the parameters are adjusted and re-produced until the flavor effect that satisfies the consumer is achieved.

[0033] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A cigar flavor dynamic control system, characterized in that: include: The flavor analysis module analyzes the volatile components in cigar tobacco leaves, performs sensory evaluation on the overall flavor of the cigar, and establishes a cigar flavor fingerprint; Demand collection module, used to collect current consumer demand, personal information and cigar consumption habit data; A consumer habits database stores consumers' personal information and cigar consumption habits data, and uses data mining and analysis techniques to establish a correlation model between consumer habits and cigar flavor characteristics; The recipe generation module generates personalized cigar recipes based on the association model in the consumer habit database and the specific information and preferences of the current consumer; Production management module, which manages production based on the generated personalized cigar recipe; The feedback adjustment module conducts flavor testing and consumer trial evaluation on the product after the cigar production is completed, and provides feedback and adjustments.

2. The cigar flavor dynamic control system according to claim 1, characterized in that: The flavor analysis module analyzes volatile components in the following manner: The volatile components in cigar tobacco leaves were analyzed by GC-MS. Assume that the content of a certain flavor substance in the cigar tobacco leaf sample is , where i = 1, 2, ..., n, n is the number of flavor substances, and the peak area of ​​the substance is obtained by GC-MS detection , the content was calculated using the standard curve method; the standard curve equation is ,in and It is a coefficient obtained by measuring a standard of known concentration.

3. The cigar flavor dynamic control system according to claim 2, characterized in that: The flavor analysis module performs sensory evaluation in the following manner: The overall flavor of cigars is evaluated using an electronic nose and an electronic tongue. The sensor array based on the electronic nose has different response values ​​to different odor molecules. Let sensor j, where j = 1, 2, ..., m, where m is the number of sensors, and the response value to the cigar sample is , the response value is processed by principal component analysis method to obtain the principal component score representing the cigar odor characteristics ; The principal component score of the electronic tongue representing the taste characteristics was obtained through principal component analysis. , where k=1,2,…,p, p is the number of electronic tongue sensors; The data from GC-MS, electronic nose and electronic tongue were integrated to establish a cigar flavor fingerprint.

4. The cigar flavor dynamic control system according to claim 3, characterized in that: The demand collection module collects consumers' personal information and cigar consumption habit data as follows: Consumer information includes age, gender, years of smoking, and region. Let age be Y, gender coefficient be G, years of smoking be X, and region coefficient be D. Cigar consumption habit data includes favorite cigar brands, common sizes, daily smoking volume, and preference ratings for descriptors of different flavors.

5. The cigar flavor dynamic control system according to claim 4, characterized in that: The consumer habit database establishes a correlation model between consumer habits and cigar flavor characteristics as follows: Assume that the preference score for richness is , the light preference score is , the sweetness preference score is , the spiciness preference score is ; Through data mining and analysis technology, establish the correlation model between consumer habits and cigar flavor characteristics; use the multivariate linear regression model to establish the consumer's preference for a certain flavor Relationship with consumer information and consumption habit data: ； in: is the regression coefficient, is the random error term.

6. The cigar flavor dynamic control system according to claim 5, characterized in that: The recipe generation module generates a personalized cigar recipe in the following manner: Assume that the type ratio vector of tobacco raw materials in the generated formula is , the additive type ratio vector is ; Recipe generation is based on the following optimization objective function: ； Among them, q is the number of tobacco raw material types, r is the number of additive types, is the actual content of flavor substance i in the cigar after production according to the recipe, The expected content is determined based on consumer preferences; and are the principal component scores of the actual and expected electronic nose sensor j, respectively; and are the actual and expected principal component scores of the electronic tongue sensor k, respectively; λ and μ are weight coefficients used to balance the importance of chemical component content and sensory evaluation indicators; by solving this optimization problem, the optimal formula that meets the personalized needs of consumers is obtained.

7. The cigar flavor dynamic control system according to claim 6, characterized in that: The feedback adjustment module is specifically: After cigar production is completed, the product is subjected to flavor testing and consumer trial evaluation; the test results are compared with the expected values ​​in the recipe generation module to calculate the flavor deviation : ； At the same time, feedback from consumers who tried the product, including satisfaction ratings on the flavor, was collected. , improvement suggestions; adjust and optimize the parameters of the recipe generation module based on flavor deviation and consumer feedback.

8. The cigar flavor dynamic control system according to claim 1, characterized in that: The feedback adjustment module also includes establishing a mathematical model of flavor score and inlet depth through trial puff experiments and using data analysis methods when making feedback adjustments, and adjusting the strategy as follows: Through consumer test smoking experiments, data were collected at different entrance depths. Under this circumstance, the consumer's rating data for various flavors of cigars; let the consumer's rating for the kth flavor at different entrance depths be , where k = 1, 2, …, s, s is the number of flavor types; Using data analysis methods, a mathematical model of flavor score and depth of entry was established, and multiple regression analysis was performed to obtain the following: ； in, 、 、 、 、 is the regression coefficient, is a random error term; through the model, we can understand the influence of different entrance depths on the perception of various flavors.

9. The cigar flavor dynamic control system according to claim 8, characterized in that: The feedback adjustment module adjusts the recipe according to the consumer's depth of entry when performing feedback adjustment; and formulates a personalized recipe adjustment strategy based on the personal information and preferences in the consumer habit database and the association model between the depth of entry and flavor.

10. The cigar flavor dynamic control system according to claim 9, characterized in that: The control method of the system comprises the following steps: S1: Data acquisition and analysis; S11: Collect flavor data of cigars of different brands and origins, analyze them using GC-MS, electronic nose, and electronic tongue devices in the flavor analysis module, and establish a cigar flavor database; S12: Collect consumers’ personal information and cigar consumption habit data, clean, organize and analyze the data, and build a consumer habit database; S2: Formulation design and production; S21: The demand collection module collects consumers' personalized customization needs and inputs consumers' relevant information and preference data into the system; S22: The recipe generation module generates a personalized cigar recipe based on the association model in the consumer habit database and the optimization objective function; S23: The production management module produces corresponding products based on the generated personalized cigar recipe; S3: Quality inspection and feedback; S31: After production, the customized cigars undergo flavor testing, including GC-MS analysis, electronic nose and electronic tongue evaluation, and consumer trial smoking. S32: The feedback adjustment module calculates the flavor deviation and satisfaction score based on the test results and consumer feedback; if the product does not meet the requirements, the relevant information is fed back to the recipe generation module, and the parameters are adjusted and re-produced until the flavor effect that satisfies the consumer is achieved.

Images