Baijiu sensory quality concrete expression system construction and evaluation method

Abstract

The invention relates to the technical field of baijiu, and discloses a baijiu sensory quality concrete expression system construction and evaluation method which comprises the following steps: acquiring basic data; according to the basic data, obtaining description word data of characteristics of each flavor dimension of the Baijiu, performing deletion according to the occurrence frequency, determining description words of a concrete flavor wheel of the Baijiu, and drawing an overall concrete flavor wheel of the Baijiu based on the description words of the concrete flavor wheel and corresponding concrete reference substances; based on the overall concrete flavor wheel of the white spirit, preparing a corresponding concrete reference object, and under the guidance of the overall concrete flavor wheel of the white spirit and the concrete reference object, obtaining a flavor description word obtained by evaluating the white spirit by an evaluator and the strength of the flavor description word; and drawing a flavor profile representing the flavor characteristics of the Baijiu. According to the method, drawing of the concrete flavor wheel of the Baijiu is realized for the first time, so that a user can intuitively and scientifically feel the flavor characteristics of the Baijiu product.

Description

Technical Field

[0001] The present invention relates to the technical field of liquor, and in particular to a method for constructing a concrete expression system for the sensory quality of liquor and an evaluation method. Background Art

[0002] As a unique distilled liquor in China, the study of the flavor characteristics of baijiu is of great significance to improving product quality and meeting consumer needs.

[0003] The complexity of baijiu flavors makes accurate description and communication of its sensory characteristics challenging. Traditional baijiu sensory evaluation often relies on subjective evaluation by sensory panelists, lacking unified standards and concrete representations. Specifically, sensory panelists have varying perceptions of the same aroma and its intensity, making it difficult to effectively evaluate the unique flavor qualities of certain baijiu. Therefore, systematic, standardized, and concrete evaluation of baijiu's sensory qualities remains an urgent challenge in this field. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for constructing a concrete expression system for the sensory quality of liquor and for evaluating it, so as to solve the problems existing in the current sensory evaluation of liquor flavor, such as subjective reliance, lack of unified standards and concrete expression methods.

[0005] In order to achieve the above purpose, the technical solutions adopted are as follows:

[0006] In a first aspect, the present invention provides a method for constructing a concrete expression system for the sensory quality of liquor, the method comprising:

[0007] Acquire basic data; wherein the basic data includes liquor flavor descriptor data and concrete reference data, the liquor flavor descriptor data includes a plurality of liquor flavor descriptors, the concrete reference data includes a plurality of concrete references, one liquor flavor descriptor corresponds to at least one concrete reference, and the types of liquor flavor descriptors include aroma descriptors and taste descriptors;

[0008] Based on the basic data, descriptive word data of the characteristics of each flavor dimension of the liquor is obtained, and descriptive words are deleted according to the frequency of occurrence to determine the descriptive words of the figurative flavor wheel of the liquor; based on the descriptive words of the figurative flavor wheel and the corresponding figurative reference objects, an overall figurative flavor wheel of the liquor is drawn;

[0009] Based on the overall figurative flavor wheel of the liquor, corresponding figurative reference objects are prepared, and guided by the overall figurative flavor wheel of the liquor and the figurative reference objects, flavor descriptors and their intensities obtained by evaluators when evaluating the liquor are obtained;

[0010] The flavor descriptors are sorted out to draw a flavor profile representing the flavor characteristics of the liquor.

[0011] Furthermore, based on the overall figurative flavor wheel of the liquor, corresponding figurative reference objects are prepared. Under the guidance of the overall figurative flavor wheel of the liquor and the figurative reference objects, flavor descriptors and their intensities obtained by sensory panelists when evaluating the liquor are obtained, including:

[0012] At room temperature, the liquor sample is poured into a plurality of first transparent containers with random coding marks, and the amount of liquor poured into each transparent container is between 15 and 20 mL. The plurality of first transparent containers are presented to evaluators in a random order, and the evaluators independently smell the liquor sample to be tested without external interference, and select the aroma descriptor most relevant to the liquor sample to be tested from the table as the flavor descriptor according to the liquor reference flavor sensory description vocabulary; wherein, the liquor reference flavor sensory description vocabulary is determined according to the overall concrete flavor wheel of the liquor.

[0013] Furthermore, when the evaluators independently smell the liquor samples to be tested without external interference and select the aroma descriptors most relevant to the liquor samples to be tested from the liquor reference flavor sensory description vocabulary as flavor descriptors, each round of tasting time is within 30 minutes, and coffee beans are provided to relieve olfactory fatigue for the evaluators.

[0014] Furthermore, based on the overall figurative flavor wheel of the liquor, corresponding figurative reference objects are prepared. Under the guidance of the overall figurative flavor wheel of the liquor and the figurative reference objects, flavor descriptors and their intensities obtained by sensory panelists when evaluating the liquor are obtained, including:

[0015] At room temperature, the liquor sample is poured into a plurality of second transparent containers with random coding marks, with the amount of liquor poured into each transparent container being equal. The plurality of second transparent containers are presented to evaluators in a random order, and the evaluators independently evaluate the liquor sample to be tested without external interference, and select the taste descriptors most relevant to the liquor sample to be tested as flavor descriptors based on the list of alcoholic beverage taste terms; wherein the list of alcoholic beverage taste terms is determined based on the overall concrete flavor wheel of the liquor.

[0016] Furthermore, when the evaluators independently evaluate the liquor samples to be tested without external interference and select the taste descriptors most relevant to the liquor samples to be tested as flavor descriptors based on the list of alcoholic beverage taste terms, a rest time is set after each sample is evaluated, and oral hygiene is performed.

[0017] Furthermore, the method of organizing the flavor descriptors includes:

[0018] Based on the flavor descriptors, hedonic terms, synonyms, antonyms, pleasure terms, quantitative terms, and terms unrelated to the aroma and taste of liquor were deleted, and descriptors with an occurrence frequency of less than 3 were deleted to obtain the sorted flavor descriptors.

[0019] In a second aspect, the present invention provides a method for evaluating the sensory quality of liquor. Based on the method for constructing a concrete expression system for the sensory quality of liquor as described above, the method for evaluating the sensory quality of liquor comprises:

[0020] Based on the figurative flavor wheel of liquor, the flavor descriptors of different liquor samples were determined;

[0021] Determine the quantitative description of flavor descriptors of different liquor samples based on the set liquor aroma reference corresponding intensity table and the liquor taste reference corresponding intensity table;

[0022] Based on the flavor descriptors of different liquor samples, contribution value screening, clustering and principal component analysis were performed to obtain multiple flavor profile descriptors;

[0023] Determine the quantitative descriptions of flavor profile descriptors for different liquor samples and draw flavor profile graphs.

[0024] Furthermore, the contribution values of the flavor descriptors of different liquor samples were screened in the following way:

[0025] A geometric mean is calculated for each flavor descriptor, and the flavor descriptors are arranged from large to small according to the size of the geometric mean. The top N flavor descriptors are selected for subsequent clustering and principal component analysis, where N is a preset positive integer value.

[0026] Furthermore, the geometric mean is calculated as:

[0027]

[0028] Where M represents the geometric mean, F represents the percentage of the number of times a descriptor is used to the total number of times the descriptor may be used, and I represents the percentage of the actual intensity of the descriptor given to the maximum possible intensity of the descriptor.

[0029] Furthermore, the drawn flavor profiles include aroma profile profiles of different liquor samples and taste profile profile profiles of different liquor samples.

[0030] The beneficial effects of the present invention are:

[0031] 1. This invention, by extensively collecting baijiu flavor descriptors and their corresponding physical and chemical references, breaks the limitations and abstract nature of traditional flavor evaluation, which relies solely on verbal communication of flavor characteristics. This ensures the standardization and visualization of sensory data. By comprehensively collecting baijiu flavor descriptors, gathering their corresponding physical references, and formulating corresponding chemical references, this approach comprehensively, systematically, and concretely enhances sensory personnel's understanding and perception of baijiu flavor.

[0032] 2. This invention is the first to create a figurative flavor wheel for liquor, enabling users to intuitively and scientifically understand the flavor characteristics of liquor products. By providing a figurative flavor wheel with images of physical and chemical references, it systematically and specifically improves the accuracy of evaluators' perception of liquor flavor.

[0033] 3. The present invention accurately, quickly and scientifically characterizes the flavor differences between different liquors by establishing a concrete expression system for the sensory quality of liquor and applying a liquor sensory quality evaluation method. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a method for constructing a concrete expression system for the sensory quality of liquor according to an embodiment of the present invention.

[0035] Figure 2 A reference figure showing a physical object of a liquor aroma descriptor according to an embodiment of the present invention is shown.

[0036] Figure 3 A schematic diagram of a flavor wheel representing the flavor characteristics of liquor drawn according to an embodiment of the present invention is shown.

[0037] Figure 4 A flow chart of a method for evaluating the sensory quality of liquor according to an embodiment of the present invention is shown.

[0038] Figure 5 Shown are aroma profile profiles of different liquor samples according to an embodiment of the present invention.

[0039] Figure 6 Shown are profile diagrams of the taste of different liquor samples according to embodiments of the present invention. DETAILED DESCRIPTION

[0040] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0041] The specific implementation of the present invention is further described in detail below with reference to the accompanying drawings and examples.

[0042] The embodiment of the present invention provides a method for constructing a concrete expression system for the sensory quality of liquor. Figure 1 As shown, the method includes the following steps:

[0043] S101: Obtain basic data; wherein, the basic data includes liquor flavor descriptor data and concrete reference data, the liquor flavor descriptor data includes multiple liquor flavor descriptors, the concrete reference data includes multiple concrete references, one liquor flavor descriptor corresponds to at least one concrete reference, and the types of liquor flavor descriptors include aroma descriptors and taste descriptors.

[0044] The purpose of step S101 is to comprehensively collect flavor descriptors and definitions related to baijiu, as well as concrete references associated with these flavor descriptors. These references, combined with the flavor descriptors, can help guide evaluators in evaluating baijiu. This guidance includes training evaluators and helping them better formulate relevant descriptors. Specifically, a comprehensive list of baijiu aroma descriptors and a comprehensive list of baijiu taste descriptors are shown in Tables 1 and 2.

[0045] Table 1 Summary of descriptive words for liquor aroma

[0046]

[0047]

[0048] Table 2 Summary of descriptors for liquor taste

[0049]

[0050]

[0051] like Figure 2 The following table shows the reference table of the aroma descriptors of liquor, and the corresponding chemical substances of the aroma descriptors of liquor are shown in Table 3. The chemical and physical reference table of the taste descriptors of liquor are shown in Table 4.

[0052] Table 3 Reference table of chemical substances for descriptive words of liquor aroma

[0053]

[0054]

[0055] Table 4 Chemical and physical reference table of liquor taste description words

[0056] smell Reference Materials Mass concentration (g / L) acid Citric acid 2.5 sweet sucrose 10.00 bitter Quinine 0.1 salty anhydrous sodium chloride 4.00 fresh monosodium glutamate 2.00 burning sensation 30% ethanol in water - Astringent feeling aluminum sulfate 3.00 astringency Wine samples - sticky feeling Carboxymethyl cellulose 2.00 mellow milk -

[0057] S102: Based on the basic data, obtain descriptive word data of the characteristics of each flavor dimension of the liquor, delete them according to the frequency of occurrence, determine the descriptive words of the concrete flavor wheel of the liquor, and draw the overall concrete flavor wheel of the liquor based on the descriptive words of the concrete flavor wheel and the corresponding concrete reference objects.

[0058] In this embodiment, based on the data collected in step S101, a descriptive analysis is performed on the characteristics of each flavor dimension of a certain category of liquor. The obtained descriptive word data is integrated and filtered, and deleted based on frequency of occurrence. Descriptive words for the specific brand of liquor's figurative flavor wheel are determined. Combined with physical and chemical references, a comprehensive figurative flavor wheel for the specific category of liquor is constructed. This comprehensive figurative flavor wheel is used to guide evaluators in generating corresponding flavor descriptors, and its specific function will be detailed in subsequent steps.

[0059] S103: Based on the overall concrete flavor wheel of the liquor, corresponding concrete reference objects are prepared. Under the guidance of the overall concrete flavor wheel of the liquor and the concrete reference objects, flavor descriptors and their intensities obtained by evaluators when evaluating the liquor are obtained.

[0060] In this example, physical references and chemical references for the above-mentioned flavor descriptors were prepared to help sensory panelists identify and understand the aroma and taste characteristics. Subsequently, sensory evaluation of a series of liquor brands was conducted. The aroma and taste evaluations were conducted in stages to avoid olfactory and taste fatigue of the sensory panelists.

[0061] ① Aroma Evaluation Method: Samples were evaluated in secret. At room temperature (25°C), the team leader poured 15-20 mL of Qinghua Fenjiu into a 50 mL, tulip-shaped, transparent tasting cup marked with a three-digit random code. Each cup was immediately presented to the evaluators in a random order. The evaluators then independently smelled the sample without interference. Using a reference list of flavor sensory descriptors for baijiu, they selected the most relevant descriptors for the sample and recorded them (with minor additions). Each evaluation round lasted no more than 30 minutes, and coffee beans were provided to alleviate olfactory fatigue. Sensory panelists were asked not to communicate with each other during the tasting process.

[0062] ② Taste evaluation method: The wine samples were evaluated in secret. At room temperature of 25°C, the liquor samples were placed in disposable transparent tasting cups. Each sample was injected with the same volume of 3mL using a pipette, and then a random three-digit code was provided to the evaluator.

[0063] The evaluators take the wine sample and spread it throughout their mouth (both sides of the tongue, the tongue surface, and extending to the throat) to experience the different flavors and mouthfeel characteristics. The wine stays in the mouth for 4 seconds, carefully feeling the wine quality and noting the flavors and mouthfeel characteristics at each stage. To eliminate the interference of aroma, evaluators are required to wear a nose clip to focus on the oral sensation. Finally, the wine can be swallowed or spit out, and the evaluators slowly open their mouths and inhale, allowing the wine vapor to be exhaled through the nasal cavity with the breath, to judge the wine's aftertaste (aftertaste, aftertaste). At the same time, to avoid a lack of evaluation vocabulary, imagination, and memory, a list of alcoholic beverage taste terms is provided as an auxiliary reference and can be supplemented. Samples are presented randomly. After each sample is tasted, they rest for 30 seconds to 1 minute and are provided with mineral water, skim milk, soda crackers, etc. to help them clean their mouths.

[0064] S104: Arrange the flavor descriptors and their intensities to draw a flavor profile representing the flavor characteristics of the liquor.

[0065] In this embodiment, according to the above step S103, different wine samples are evaluated for aroma and taste respectively and relevant descriptors are given. The descriptors obtained by the evaluators are counted, the descriptions are filtered, and hedonic terms, synonyms, antonyms, pleasure terms, quantitative terms and terms unrelated to the aroma and taste of liquor are deleted. Descriptors with an appearance frequency of less than 3 are deleted; the sorted descriptive terms are classified and combined with physical references and chemical substance references; according to the classification, a concrete flavor wheel representing the flavor characteristics of a certain brand or category of liquor is drawn. The drawn concrete flavor wheel is as follows: Figure 3 shown.

[0066] Example 2:

[0067] The embodiment of the present invention provides a method for evaluating the sensory quality of liquor, based on the method for constructing a concrete expression system for the sensory quality of liquor as described in Example 1. Figure 4 As shown, the sensory quality evaluation method of liquor includes:

[0068] S401: Determine flavor descriptors of different liquor samples based on the liquor flavor wheel.

[0069] In this embodiment, the figurative flavor wheel of liquor is constructed using the method provided in Example 1, and the flavor descriptors of different liquor samples can be directly extracted from the figurative flavor wheel of liquor.

[0070] S402: Determine the quantitative description of flavor descriptors of different liquor samples based on the set liquor aroma reference corresponding intensity table and the liquor taste reference corresponding intensity table.

[0071] In this embodiment, the table of corresponding intensity of liquor aroma references and the table of corresponding intensity of liquor taste references each provide a numerical value for each flavor descriptor (e.g., aroma descriptors and taste descriptors) for quantification. These numerical values are described herein as reference sample aroma intensity and reference sample taste intensity, depending on the aroma descriptor and taste descriptor. Therefore, each flavor descriptor can be assigned a specific numerical value, enabling quantitative description of flavor descriptors for different varieties of liquor. Specific tables of corresponding intensity of liquor aroma references and corresponding intensity of liquor taste references are shown in Tables 5 and 6.

[0072] Table 5 Corresponding intensity of liquor aroma references

[0073]

[0074]

[0075]

[0076] Table 6 Corresponding intensity of liquor taste references

[0077] Taste descriptors Reference Materials Mass concentration (g / L) Reference sample taste intensity acid Citric acid 2.5 4 sweet sucrose 10.00 4 bitter Quinine 0.1 4 salty anhydrous sodium chloride 4.00 4 fresh monosodium glutamate 2.00 4 burning sensation 30% ethanol in water - 4 Astringent feeling aluminum sulfate 3.00 4 astringency Wine samples - 4 sticky feeling Carboxymethyl cellulose 2.00 4 mellow milk - 4

[0078] Corresponding to Tables 5 and 6 above, this embodiment also provides a liquor aroma intensity scale and a liquor taste intensity scale as shown in Tables 7 and 8, respectively.

[0079] Table 7 Liquor aroma intensity scale

[0080] scale 0 1 2 3 4 5 6 7 8 9 degree weak medium powerful

[0081] Table 8 Liquor Taste Intensity Scale

[0082]

[0083]

[0084] S403: Perform contribution value screening, clustering, and principal component analysis based on the flavor descriptors of different liquor samples to obtain multiple flavor profile descriptors.

[0085] In this embodiment, a contribution value screening method is provided: referring to the international standard ISO11035 sensory descriptor selection method, the flavor descriptors obtained by sensory personnel are refined, and the geometric mean (M) of each descriptor in the flavor wheel is calculated:

[0086]

[0087] F: The percentage of times a descriptive word is used in the total number of times the descriptive word may be used;

[0088] I: The actual intensity of the descriptor given as a percentage of the maximum possible intensity of the descriptor.

[0089] Clustering and principal component analysis were performed using origin, and flavor profile descriptors were finally screened. In this embodiment, 12 aroma descriptors and 12 taste descriptors were finally determined, and profile diagrams of a series of liquor samples will be drawn subsequently.

[0090] The purpose of cluster analysis is to group descriptive words with similar semantic or statistical characteristics, reduce redundancy, and extract representative words. The implementation steps can be, for example:

[0091] S4031: Feature matrix construction.

[0092] For each descriptor (flavor descriptors filtered by contribution value), a feature vector is constructed, including statistical and semantic features. Statistical features can include mean, variance, skewness, kurtosis, etc. Semantic features can be extracted using word embedding (Word2Vec / BERT). Statistical and semantic features are combined to construct a feature matrix.

[0093] S4032: Select a clustering algorithm.

[0094] Clustering algorithms include Hierarchical Clustering and K-means Clustering. Hierarchical Clustering calculates the similarity matrix (Euclidean distance or cosine similarity) between descriptors and uses Ward's method or Complete Linkage to generate a dendrogram.

[0095] K-means clustering: Determine the optimal number of clusters based on the Silhouette Score (e.g., aroma words are divided into 5-6 groups, taste words are divided into 5-6 groups).

[0096] S4033: Selection of representative words within a cluster.

[0097] Calculate the contribution of each word in the cluster (such as variance contribution, random forest importance). Select the 1-2 words with the highest contribution in each cluster as representatives.

[0098] S4034: Output result: Aroma and taste descriptors are clustered into 5-6 groups, and 1-2 representative words are retained in each group.

[0099] The purpose of principal component analysis is to verify whether the descriptors selected by cluster analysis cover the main flavor dimensions and to further optimize the dimensionality reduction. For example, the steps to implement principal component analysis can be:

[0100] S4035: Data standardization.

[0101] The filtered descriptor rating matrix is centered (mean returned to zero) and standardized (unit variance).

[0102] S4036: PCA dimensionality reduction.

[0103] Calculate the covariance matrix and extract principal components (PCs). Determine the number of principal components (e.g., the first 5-8) based on their cumulative variance contribution (usually >80%).

[0104] S4037: Load matrix analysis.

[0105] The loading of each principal component relative to the original descriptor represents its contribution weight. Descriptors with an absolute loading value greater than 0.5 were selected to ensure the interpretability of the principal components.

[0106] S4038: Orthogonality verification.

[0107] Check whether the loadings of the final retained descriptors on the principal components are orthogonal (low correlation).

[0108] S4039: Output 12 aroma descriptors and 12 taste descriptors, which must cover the main principal components (PC1-PC5) and have low correlation between descriptors of different principal components (Pearson correlation coefficient <0.3).

[0109] S404: Determine the quantitative description of the flavor profile descriptors of different liquor samples and draw a flavor profile graph.

[0110] In this embodiment, the flavor profile descriptors of different liquor samples are all screened from the flavor descriptors. Therefore, the quantitative descriptions of the flavor profile descriptors of different liquor samples can be obtained based on the quantitative descriptions of the flavor descriptors of different liquor samples determined in step S402. The drawn flavor profiles include aroma profile profiles of different liquor samples and taste profile profiles of different liquor samples, as shown in FIG. Figure 5 and Figure 6 shown.

[0111] The above embodiments are only used to illustrate the present invention, and are not intended to limit the present invention. Ordinary technicians in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, all equivalent technical solutions also fall within the scope of the present invention. The scope of patent protection of the present invention should be defined by the claims.

Claims

1. A method for constructing a concrete expression system for the sensory quality of liquor, characterized in that: The method comprises: Acquire basic data; wherein the basic data includes liquor flavor descriptor data and concrete reference data, the liquor flavor descriptor data includes a plurality of liquor flavor descriptors, the concrete reference data includes a plurality of concrete references, one liquor flavor descriptor corresponds to at least one concrete reference, and the types of liquor flavor descriptors include aroma descriptors and taste descriptors; Based on the basic data, descriptive word data of the characteristics of each flavor dimension of the liquor is obtained, and descriptive words are deleted according to the frequency of occurrence to determine the descriptive words of the figurative flavor wheel of the liquor; based on the descriptive words of the figurative flavor wheel and the corresponding figurative reference objects, an overall figurative flavor wheel of the liquor is drawn; Based on the overall figurative flavor wheel of the liquor, corresponding figurative reference objects are prepared, and guided by the overall figurative flavor wheel of the liquor and the figurative reference objects, flavor descriptors and their intensities obtained by evaluators when evaluating the liquor are obtained; The flavor descriptors and their intensities are collated to create a flavor profile representing the flavor characteristics of the liquor.

2. The method for constructing a concrete expression system for the sensory quality of liquor according to claim 1, characterized in that: Based on the overall figurative flavor wheel of the liquor, corresponding figurative reference objects are prepared. Under the guidance of the overall figurative flavor wheel of the liquor and the figurative reference objects, the flavor descriptors and their intensities obtained by sensory panelists when evaluating the liquor are obtained, including: At room temperature, the liquor sample is poured into a plurality of first transparent containers with random coding marks, and the amount of liquor poured into each transparent container is between 15 and 20 mL. The plurality of first transparent containers are presented to evaluators in a random order, and the evaluators independently smell the liquor sample to be tested without external interference, and select the aroma descriptor most relevant to the liquor sample to be tested from the table as the flavor descriptor according to the liquor reference flavor sensory description vocabulary; wherein, the liquor reference flavor sensory description vocabulary is determined according to the overall concrete flavor wheel of the liquor.

3. The method for constructing a concrete expression system for the sensory quality of liquor according to claim 2, characterized in that: When the evaluators independently smell the liquor samples to be tested without external interference and select the aroma descriptors most relevant to the liquor samples to be tested from the liquor reference flavor sensory description vocabulary table as flavor descriptors, each round of tasting time is within 30 minutes, and coffee beans are provided to relieve olfactory fatigue for the evaluators.

4. The method for constructing a concrete expression system for the sensory quality of liquor according to claim 1, characterized in that: Based on the overall figurative flavor wheel of the liquor, corresponding figurative reference objects are prepared. Under the guidance of the overall figurative flavor wheel of the liquor and the figurative reference objects, the flavor descriptors and their intensities obtained by sensory panelists when evaluating the liquor are obtained, including: At room temperature, the liquor sample is poured into a plurality of second transparent containers with random coding marks, with the amount of liquor poured into each transparent container being equal. The plurality of second transparent containers are presented to evaluators in a random order, and the evaluators independently evaluate the liquor sample to be tested without external interference, and select the taste descriptors most relevant to the liquor sample to be tested as flavor descriptors based on the list of alcoholic beverage taste terms; wherein the list of alcoholic beverage taste terms is determined based on the overall concrete flavor wheel of the liquor.

5. The method for constructing a concrete expression system for the sensory quality of liquor according to claim 4, characterized in that: When the evaluators independently evaluate the liquor samples to be tested without external interference and select the taste descriptors most relevant to the liquor samples to be tested as flavor descriptors based on the list of alcoholic beverage taste terms, they will take a set break after each sample is tasted and perform oral cleaning.

6. The method for constructing a concrete expression system for the sensory quality of liquor according to claim 1, characterized in that: Methods for organizing the flavor descriptors include: Based on the flavor descriptors, hedonic terms, synonyms, antonyms, pleasure terms, quantitative terms, and terms unrelated to the aroma and taste of liquor were deleted, and descriptors with an occurrence frequency of less than 3 were deleted to obtain the sorted flavor descriptors.

7. A method for evaluating the sensory quality of liquor, characterized in that: Based on the method for constructing a concrete expression system for the sensory quality of liquor according to any one of claims 1 to 4, the method for evaluating the sensory quality of liquor comprises: Based on the figurative flavor wheel of liquor, the flavor descriptors of different liquor samples were determined; Determine the quantitative description of flavor descriptors of different liquor samples based on the set liquor aroma reference corresponding intensity table and the liquor taste reference corresponding intensity table; Based on the flavor descriptors of different liquor samples, contribution value screening, clustering and principal component analysis were performed to obtain multiple flavor profile descriptors; Determine the quantitative descriptions of the flavor profile descriptors of the liquor samples and draw flavor profile graphs of different liquor samples.

8. The method for evaluating the sensory quality of liquor according to claim 7, wherein: The contribution values of flavor descriptors of different liquor samples were screened in the following way: A geometric mean is calculated for each flavor descriptor, and the flavor descriptors are arranged from large to small according to the size of the geometric mean. The top N flavor descriptors are selected for subsequent clustering and principal component analysis, where N is a preset positive integer value.

9. The method for evaluating the sensory quality of liquor according to claim 8, wherein: The geometric mean is calculated as: Where M represents the geometric mean, F represents the percentage of the number of times a descriptor is used to the total number of times the descriptor may be used, and I represents the percentage of the actual intensity of the descriptor given to the maximum possible intensity of the descriptor.

10. The method for evaluating the sensory quality of liquor according to claim 7, wherein: The drawn flavor profiles include aroma profile profiles of different liquor samples and taste profile profiles of different liquor samples.

Images