Method and system for intelligently blending Liupao tea

Abstract

The invention relates to the technical field of geological exploration, in particular to a method and system for intelligently blending Liupao tea, and the method comprises the following steps: detecting the concentration of tea base components, setting a heating interval, collecting dissolution data, analyzing a change trend, extracting a key release area, calculating a difference identification coordination area, and constructing a unit set through vectoring numbering. According to the method, dissolution data of ingredients such as polyphenols, caffeine and amino acid in the tea base at different temperatures are collected, the dynamic relation between the temperature zone and the ingredient concentration is constructed, accurate recognition of the key release zone and the flavor coordination zone is achieved, and the accuracy of the tea base flavor matching is improved. A standardized vector is constructed based on component release concentration, a number set is generated in combination with a source and a processing sequence, quantifiable and comparable flavor feature units are provided, and data driving and accurate matching of a blending decision are realized through distance comparison with target parameters and screening of an optimal combination. Quantitative analysis and standard expression of flavor composition are realized in the whole process.

Description

Technical Field

[0001] This invention relates to the field of geological exploration technology, and in particular to a method and system for intelligent blending of Liubao tea. Background Technology

[0002] Product data management technology involves the structured organization, centralized maintenance, and collaborative management of data flow at each stage of a company's product lifecycle. This technology primarily includes modeling product formula data, standardizing raw material parameters, defining configuration logic, version tracking, material substitution, process correlation, and data interface management with business processes such as production, procurement, and sales. This field typically achieves unified management of product information and multi-system linkage through methods such as establishing data dictionaries, defining data coding rules, configuring attribute templates, and defining data flow paths. Traditional Liubao tea blending refers to the qualitative combination of tea leaves from different origins, years, aroma types, and fermentation levels during Liubao tea processing, based on experience, to achieve stable flavor, consistent quality, and a clear consumer positioning. This technical aspect typically involves determining the blending ratio through manual sampling, based on the blending technician's judgment of sensory attributes and historical formula records, including sensory evaluation methods such as the dry aroma characteristics of the tea sample, liquor color, taste harmony, and leaf condition. The blending operation is then completed through weighing, distributing, and manual mixing.

[0003] In existing technologies, traditional Liubao tea blending relies on technicians' experience and sensory evaluation to combine samples, lacking quantitative analysis of the internal chemical characteristics of the tea samples. Blending judgments are largely based on subjective perceptions of dry aroma, liquor color, taste, and leaf appearance, making it difficult to avoid fluctuations caused by individual differences in perception. In determining blending ratios, there are batch-to-batch differences between historical records and current tea samples, and sensory characteristics may not be entirely equivalent, leading to flavor deviations and making it difficult to maintain stable quality control over the long term. Furthermore, when facing target consumer needs, traditional blending methods lack a clear data matching mechanism, making it difficult to proactively align with the target flavor model. In daily operations, if the aroma of a certain type of tea sample changes or parameters are unclear during blending, it is difficult to trace the source or make adjustments, affecting blending efficiency and flavor consistency. For example, the same blending ratio may yield completely different sensory responses from raw materials of different years or fermentation levels, and the lack of component data to support adjustments leads to repeated trials in the optimization process, increasing labor costs and material waste. Overall, under the existing technological model, blending decisions are constrained by experience accumulation and sensory variability, lacking system support and data reusability, making it difficult to meet the high requirements of modern consumption scenarios for flavor standardization and batch consistency. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for intelligent blending of Liubao tea, comprising the following steps: S1: Detect the room temperature concentrations of polyphenols, caffeine, and amino acids in Liubao tea leaves, set the heating range, collect dissolution data at temperature nodes, establish the relationship between temperature and component concentration, and generate a high-temperature dissolution behavior information set for tea samples. S2: Based on the concentration change trend of the high-temperature dissolution behavior information set of the tea sample, calculate the release rate of the differential temperature zone, identify the rate change segment, mark it as the key release zone, and generate a key release temperature zone structure table of the tea sample. S3: Based on the segment distribution in the key release temperature zone structure table of the tea sample, calculate the component release difference value of the temperature zone, identify the segments with concentrated differences and similar composition, mark them as flavor feature coordination areas, screen the temperature zone segments with high flavor composition ratio, and generate tea sample flavor structure feature segment intervals. S4: Extract the component release concentration data from the flavor structure feature segment interval of the tea sample, standardize and combine them into a structural vector, construct a set of flavor feature units according to the source and processing order of the tea sample, and generate a set of flavor component unit numbers. S5: Based on the comparison between the vector features of the flavor component number set and the target parameters, filter the combination with the smallest distance value, summarize the flavor parameters, and generate the blended flavor parameter input structure.

[0005] As a further aspect of the present invention, the high-temperature dissolution behavior information set of the tea sample includes temperature nodes, component concentrations, and dissolution trends; the key release temperature zone structure table of the tea sample includes key temperature zone locations, component release rates, and temperature zone types; the flavor structure feature segment intervals of the tea sample include coordinated region distribution, flavor composition ratios, and component difference values; the flavor composition unit number set includes flavor structure vectors, tea sample numbers, and processing order information; and the blended flavor parameter input structure includes minimum distance combination numbers, target flavor parameters, and flavor data summaries.

[0006] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Obtain the room temperature dissolution amount of polyphenols, caffeine and amino acids in Liubao tea blank samples, extract the component concentration values ​​in the dissolution liquid, calculate the dissolution ratio based on the change relationship between the initial component content and the dissolution concentration, integrate the characteristic information under the current temperature conditions, and generate a room temperature dissolution reference coefficient matrix. S102: Based on the room temperature dissolution reference coefficient matrix, set the heating range, collect the component concentration values ​​in the dissolution solution at temperature nodes in stages, calculate the corresponding ratio based on the difference between the node dissolution concentration and the reference coefficient, sort out the change trend of the components in stages, and generate a set of dissolution distribution values ​​at temperature nodes. S103: Extract the concentration change sequence of components at different temperature nodes based on the dissolution distribution value set at the temperature nodes, calculate the concentration change rate and associate it with the corresponding temperature range, construct a data structure that associates the concentration change trend with temperature, and generate a high-temperature dissolution behavior information set of tea samples.

[0007] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Extract the concentration change trend of components in the temperature range based on the high-temperature dissolution behavior information set of the tea sample, organize the concentration change data according to the temperature range, calculate the release rate of components in the different temperature ranges, and generate a component release rate sequence value set. S202: Based on the component release rate sequence value set, identify the variation range of release rate in the temperature range, screen the temperature range where the rate change exceeds the release rate difference threshold, mark it as a key release segment, and generate a key temperature range identification segment value. S203: Based on the key temperature zone identification segment value, organize the key release segments in the tea sample according to the classification method, extract the temperature zone segments corresponding to the difference type, construct the correspondence structure between temperature zone and tea sample type, and generate the key release temperature zone structure table of tea sample.

[0008] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Based on the distribution of the segments in the key release temperature zone structure table of the tea sample, call the component concentration value and release rate value in each temperature zone, calculate the release difference value between components in the same segment, and generate a set of temperature zone release difference values ​​by combining the release difference distribution under the segment. S302: Based on the temperature range release difference value set, the release difference value and component ratio composition value of the segment are jointly judged, and the segments that meet the release difference distribution concentration threshold and component ratio fitting degree threshold are selected to obtain the flavor feature coordination segment value. S303: Extract the temperature range segments with concentrated flavor composition proportions based on the flavor feature coordination segment values, establish a mapping relationship structure between flavor composition concentration and segment intervals, and organize them to form the flavor feature segment intervals of the tea sample under the temperature range, generating the flavor structure feature segment intervals of the tea sample.

[0009] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Call the component release concentration data corresponding to the segment in the flavor structure feature segment interval of the tea sample, perform standard deviation scaling operation on the release concentration values ​​in the different segments according to the minimum and maximum values ​​in the segment, unify the scale of components in the same interval, combine them into a multi-dimensional vector with consistent structural order, and generate a flavor release structure vector set. S402: Based on the flavor release structure vector set, call the tea sample source number and processing sequence code, add the source number and processing sequence value to each structure vector respectively, and perform position matching and number classification on the multi-dimensional vectors after the additional identifier to obtain the flavor structure mapping identifier value group. S403: Based on the flavor structure mapping identifier value group, index and divide it according to the source number, extract the structural vectors with the same source and summarize them into a unified numbering unit, organize them to form a flavor vector composition structure of the combination and screening, and generate a flavor composition unit number set.

[0010] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Extract vector features based on the flavor component number set, compare the differences with the preset flavor target parameters, calculate the offset value between the combination number and the target parameter, and filter according to the offset value order to obtain the combination number with the lowest degree of difference, and generate the number group with the smallest flavor distance. S502: Extract the corresponding flavor vector based on the flavor distance minimum number group, call the dimensional flavor component parameter values, calculate the average parameter under the same dimension, integrate them into a complete parameter set, and obtain the component parameter mean structure. S503: Organize all parameter contents according to the mean structure of the component parameters, arrange them uniformly into a standard structure form, and label the corresponding dimension information to generate the blended flavor parameter input structure.

[0011] As a further aspect of the present invention, the Liubao tea base is a raw tea leaf used for preliminary drying before processing Liubao tea or for mid-stage fermentation, and has clear origin, degree of fermentation, moisture content and basic data of flavor substances; The polyphenols, caffeine, and amino acids mentioned refer to the three main flavor substances commonly found in tea. Among them, polyphenols contribute to the astringency of tea soup, caffeine affects bitterness, and theanine contributes to freshness. All of them can be quantitatively determined by high performance liquid chromatography or ultraviolet spectrophotometry. The heating range refers to the temperature change range set during the experiment to simulate brewing or processing scenarios. The range generally covers the temperature from room temperature to the temperature before boiling and is divided into multiple fixed temperature nodes for data acquisition. The temperature nodes are specific temperature points used for recording and sampling during the heating process, and each node corresponds to a detection operation of the tea sample dissolution amount. The dissolution data are recorded values ​​of the concentration of the target component in the solution, collected through experimental means. The temperature range where the release rate changes significantly refers to the temperature range in which the increase in the release rate of the target component between consecutive temperature nodes is greater than the average rate of change of the component. The critical release zone refers to the temperature range in which the release rate of components in the tea sample fluctuates significantly or where the total amount of components released accounts for a high percentage.

[0012] As a further aspect of the present invention, the difference concentration refers to the standard deviation of the release difference of different components within the target segment being less than 0.005 mg / mL; The term "similar composition" means that the difference in the concentration ratio of each component among the tea samples does not exceed ±10%. The segment with similar composition refers to the segment within the same interval where the concentration ratio of multiple target components has a low variance or deviation rate among tea samples. The flavor characteristic coordination region refers to the temperature range in which the rate of change of component release ratio between adjacent temperature nodes is less than 10%, and the rate of difference in component ratio between tea samples in the same temperature range does not exceed ±5%. The temperature range segment in the flavor composition ratio refers to the temperature range in which, within a certain temperature range, the proportion of polyphenols, caffeine, and amino acids in the total flavor components of the tea sample is significantly higher than that of the temperature range, and the variance of the variation among the proportions of the three is small. The structure vector is a fixed-dimensional vector formed by combining the normalized values ​​of polyphenols, caffeine, and amino acids in sequence. The set of flavor feature units refers to a structured information set consisting of multiple structural vectors and their source tea sample numbers; The flavor composition unit number set is a set of numbers formed by generating a unique number for each flavor feature unit according to the tea sample source, temperature zone segment number and composition vector combination information. The distance value refers to the numerical difference between the flavor component vector and the target flavor parameter vector. The blended flavor parameter input structure refers to the structured blending information composed of selected flavor component unit numbers and their component parameter sets; The distance value refers to the Euclidean distance between the structure vector of the flavor constituent unit and the target flavor parameter vector.

[0013] A system for intelligent blending of Liubao tea includes: The tea sample data acquisition module acquires Liubao tea blank samples, collects the concentration values ​​of three components—polyphenols, caffeine, and amino acids—at room temperature, sets multiple temperature control nodes and records the dissolution concentration of components at each node in sequence, and organizes the component behavior information of the sample by the correspondence between component concentration and temperature nodes, generating a tea sample high-temperature dissolution behavior information set. The temperature zone release identification module calculates the release rate value of the temperature zone based on the concentration change trend of the concentrated components of the high-temperature dissolution behavior information of the tea sample under the differential temperature control node, identifies the continuous segments with prominent release rate changes, marks and classifies the samples, and organizes and generates a key release temperature zone structure table of the tea sample. The flavor segment extraction module extracts the difference in release concentration of components within the segment based on the temperature segment in the key release temperature zone structure table of the tea sample, identifies segments with concentrated difference distribution and similar component proportions, and filters and numbers the segments that meet the conditions to generate flavor structure feature segment intervals of the tea sample. The flavor vector construction module calls the release concentration data of components in the flavor structure feature segment interval of the tea sample, performs standardization processing, assembles them into a structure vector, assigns numbers according to the tea sample source information and processing order, and generates a set of flavor component unit numbers. The flavor parameter screening module calls preset flavor parameters based on the structural vector features of the flavor component number set, compares the vector distance values, filters the combination numbers with the smallest differences, integrates the corresponding component concentration features, and generates a blended flavor parameter input structure.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by collecting dissolution data of polyphenols, caffeine, amino acids and other components in tea leaves at different temperatures, a dynamic relationship between temperature range and component concentration is constructed to achieve accurate identification of key release segments and flavor coordination regions. Based on the component release concentration, a standardized vector is constructed, and a set of numbers is generated by combining the source and processing order to provide quantifiable and comparable flavor feature units. By comparing the distance with the target parameters, the optimal combination is screened to achieve data-driven and precise matching of blending decisions. The whole process achieves quantitative analysis and standardized expression of flavor composition. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] Please see Figure 1 This invention provides a method for intelligent blending of Liubao tea, comprising the following steps: S1: Obtain room temperature detection values ​​of polyphenols, caffeine, and amino acid components in Liubao tea leaves, set the heating range and collect dissolution data at temperature nodes in sequence, and form a high-temperature dissolution behavior information set of tea samples based on the correspondence between temperature and component concentration. Liubao tea base is the raw tea used for preliminary drying or mid-stage fermentation before processing Liubao tea. It has clear origin, degree of fermentation, moisture content and basic data of flavor substances, and can be used as the basic unit raw material before blending. Polyphenols, caffeine, and amino acid components are three common main flavor substances in tea. Polyphenols contribute to the astringency of tea soup, caffeine affects bitterness, and theanine contributes to freshness. All of them can be quantitatively determined by high performance liquid chromatography or ultraviolet spectrophotometry. The heating range refers to the temperature change range set during the experiment to simulate brewing or processing scenarios. This range generally covers the temperature from room temperature to the temperature before boiling and is divided into multiple fixed temperature nodes for data acquisition. Temperature nodes are specific temperature points used for recording and sampling during the heating process. Each node corresponds to a detection operation of the dissolution amount of the tea sample, which is used to analyze the release behavior of components under different thermal conditions. Dissolution data are recorded values ​​of the concentration of the target component in the solution collected through experimental means. They are used to describe the release capacity of tea sample components at different temperature points and are usually expressed in mg / L. S2: Based on the component concentration change trend of the high-temperature dissolution behavior information of tea samples, calculate the release rate in the temperature difference zone, identify the temperature zone with prominent release rate changes and mark it as the key release zone, classify and organize the tea samples to form a mapping structure, and generate a key release temperature zone structure table of tea samples. Temperature ranges with significant changes in release rate refer to temperature segments where the increase in the release rate of the target component between consecutive temperature nodes is greater than the average rate of change of the component. These ranges are used to identify locations where the release behavior of the component changes abruptly. The critical release range refers to the temperature range in which the release rate of components in a tea sample fluctuates significantly or the total amount of components released is high. It is used for the extraction and analysis of flavor structure characteristics. The component release difference value refers to the interval data of the maximum and minimum difference in the release concentration of the same component among different tea samples within the target temperature range, which is used to reflect the stability and consistency of flavor. S3: Based on the segment distribution in the key release temperature zone structure table of tea sample, calculate the component release difference value in each temperature zone, determine the segment with concentrated difference distribution and similar composition ratio, mark it as the flavor feature coordination area, screen the temperature zone segment with concentrated flavor composition ratio, and generate the flavor structure feature segment interval of tea sample. Differential concentration refers to the standard deviation of the release difference of different components within the target segment being less than 0.005 mg / mL; Similar composition means that the difference in the concentration ratio of each component among the tea samples does not exceed ±10%; Segments with similar composition ratios refer to segments within the same range where the concentration ratios of multiple target components exhibit low variance or deviation among tea samples, and are used to screen segments with balanced flavor structures. The flavor characteristic coordination zone refers to the temperature range in which the rate of change of component release ratio between adjacent temperature nodes is less than 10%, and the rate of difference in component ratio between tea samples in the same temperature range does not exceed ±5%. The temperature range segment where the proportion of flavor composition is concentrated refers to the temperature range in which polyphenols, caffeine and amino acids account for a significantly higher proportion of the total flavor components in the tea sample than in the temperature range, and the variance of the variation among the proportions of the three is small. S4: Call the release concentration data of components in the flavor structure feature segment interval of the tea sample, perform standardization processing and combine them into a structure vector, assign numbers according to the source of the tea sample and the processing order, construct a set of flavor feature units for blending and screening, and generate a set of flavor component unit numbers; The structural vector is a fixed-dimensional vector formed by combining the normalized values ​​of polyphenols, caffeine and amino acids in sequence, and is used to characterize the flavor composition features of tea samples. The set of flavor feature units for blending screening refers to a structured information set consisting of multiple structural vectors and their source tea sample numbers, which is used to construct a blending comparison sample pool and screen flavor-matching objects. The flavor composition unit number set is a set of numbers generated by generating a unique number for each flavor feature unit according to the tea sample source, temperature zone segment number and composition vector combination information. It is used for parameter comparison and blending assembly. S5: Based on the vector features of the flavor component unit number set, compare the differences with the preset flavor target parameters, select the combination number with the smallest distance value from the target flavor parameter, summarize the corresponding flavor parameter data, and uniformly form the blended flavor parameter input structure. The target flavor parameters are a set of flavor indicators constructed based on the design or sensory standards of Liubao tea finished products, including reference values ​​for bitterness, freshness, and astringency within a specified range. Distance value is a numerical difference measure between the flavor component vector and the target flavor parameter vector. It is used to assess the proximity between flavor units and the target and is commonly calculated using Euclidean distance. The blending flavor parameter input structure refers to the structured blending information consisting of the selected flavor component unit numbers and their component parameter sets, which is used to guide the setting of blending ratios and the generation of schemes; The distance value refers to the Euclidean distance between the structural vector of the flavor constituent unit (i.e., the multidimensional normalized flavor parameter vector) and the target flavor parameter vector.

[0023] The tea sample high-temperature dissolution behavior information set includes temperature nodes, component concentrations, and dissolution trends. The key release temperature zone structure table of tea samples includes the location of key temperature zones, component release rates, and temperature zone types. The tea sample flavor structure feature segment intervals include the distribution of coordinated regions, flavor composition ratios, and component difference values. The flavor composition unit number set includes flavor structure vectors, tea sample numbers, and processing order information. The blended flavor parameter input structure includes the minimum distance combination number, target flavor parameters, and flavor data summary.

[0024] Please see Figure 2 The specific steps of S1 are as follows: S101: Obtain the room temperature dissolution amount of polyphenols, caffeine and amino acids in Liubao tea blank samples, extract the component concentration values ​​in the dissolution liquid, calculate the dissolution ratio based on the change relationship between the initial component content and the dissolution concentration, integrate the characteristic information under the current temperature conditions, and generate a room temperature dissolution reference coefficient matrix. To obtain the dissolution rates of polyphenols, caffeine, and amino acids in Liubao tea samples at room temperature, select tea samples that meet batch management requirements from the sample warehouse, weigh 5g, place them in a 250mL beaker, add 150mL of deionized water, let stand at room temperature for 10 minutes, then stir at 300rpm for 5 minutes. Obtain the dissolution solution through a vacuum filter and use it for component content determination. Use a high-performance liquid chromatography (HPLC) system with a reversed-phase C18 column, a detection wavelength of 280nm, a mobile phase of methanol and water in a 30:70 ratio, a flow rate of 1mL / min, and repeat the test three times for each batch of samples, taking the average value as the final concentration data. If the measured polyphenol concentration is 0... The concentrations of polyphenols, caffeine, and amino acids were 0.35 mg / mL, 0.12 mg / mL, and 0.22 mg / mL, respectively. The dissolution ratios need to be calculated based on the initial content of the tea leaves. If the initial contents are 8.5 mg / g, 2 mg / g, and 3.8 mg / g, the total dissolution amount of each component is obtained by multiplying the concentration by the volume. Dividing this by the sample mass and the initial content yields the corresponding percentage dissolution ratio. For example, if a sample has a polyphenol concentration of 0.35 mg / mL, caffeine of 0.12 mg / mL, and amino acid of 0.22 mg / mL, the dissolution ratios calculated based on the initial contents and the volume and mass used are approximately 95% for polyphenols, 85% for caffeine, and 78% for amino acids. These data are all within the theoretical upper limit of 100%, indicating that the experimental process was well controlled. If the dissolution rate of individual components is close to or exceeds 100%, it may be due to the deviation in the initial content estimation of the sample, the error in the recording of solution volume, or the detection error. The sample quality and detection process should be re-verified, and the experiment should be repeated if necessary to obtain true and valid dissolution data to ensure that the constructed room temperature reference structure is accurate and comparable.

[0025] S102: Based on the room temperature dissolution reference coefficient matrix, the heating range is set, the component concentration values ​​in the dissolution solution are collected in stages at temperature nodes, the corresponding ratio is calculated according to the difference between the node dissolution concentration and the reference coefficient, the change trend of the components in each stage is sorted out, and the dissolution distribution value set at the temperature node is generated. Based on the dissolution ratio data of the three components obtained under normal temperature conditions, the experimental temperature range for the heating stage was set from 30℃ to 100℃. Sampling was carried out in stages of increasing temperature by 10℃. The specific operation included using the same mass of tea sample and an equal volume of deionized water for dissolution treatment at each temperature node. Water bath heating was used to stabilize the sample liquid temperature at the target node temperature, controlling the deviation to not exceed 0.5℃. Each stage of the experiment was completed under the same soaking time and stirring rate. The dissolution concentrations of polyphenols, caffeine, and amino acids were measured separately. The dissolution amount at each temperature was calculated as a ratio and compared with the dissolution ratio at normal temperature to obtain the changes in dissolution performance at different temperatures. For example, the dissolution ratio of polyphenols at a certain node was 82%, which was about 19% higher than 63% at normal temperature. This indicates that the temperature range has a significant promoting effect on the release of polyphenols. After sorting out all the node data, a set of ratios that changed with temperature was formed. Multiple sets of numerical sequences were arranged according to the components to form a set of dissolution distribution values ​​of the three components at each temperature, which were used for trend calculation in the next stage.

[0026] S103: Extract the concentration change sequence of components at different temperature nodes based on the dissolution distribution value set at temperature nodes, calculate the concentration change rate and associate it with the corresponding temperature range, construct a data structure that associates the concentration change trend with temperature, and generate a high-temperature dissolution behavior information set of tea samples. From the compiled set of dissolution distributions, the concentration sequences of polyphenols, caffeine, and amino acids at different temperature nodes were extracted. The concentration change between adjacent temperature nodes was compared, and the change rate was obtained by dividing the change value by the temperature difference. For example, if polyphenols are 0.29 mg / mL at 30℃ and 0.46 mg / mL at 40℃, then the concentration increases by 0.017 mg / mL for every 1℃ increase. All node pairs were processed in this way to obtain a complete concentration change trend sequence. Temperature was then used as a sequence index and combined with the corresponding change value to form a trend structure. The trend was visualized as a line graph using data processing tools. By observing the curve shape, it was determined whether the change was continuously increasing, tending to flatten, or showing a turning point. Fitting calculations were performed on the trend of each component to generate a correlation model expressing the relationship between the concentration change rate and temperature. This was recorded as behavioral description information of each component and integrated into a unified high-temperature dissolution behavior set to reflect the characteristic trends exhibited by different components during the heating process.

[0027] Please see Figure 3 The specific steps of S2 are as follows: S201: Extract the concentration change trend of components in the temperature range based on the high-temperature dissolution behavior information set of tea samples, organize the concentration change data according to the temperature range, calculate the release rate of components in the different temperature ranges, and generate a set of component release rate sequence values. When extracting the concentration trends of various components in different temperature ranges based on the high-temperature dissolution behavior information set of tea samples, it is necessary to first organize the concentration data of polyphenols, caffeine, and amino acids at different temperature nodes, selecting a temperature range of 30℃ to 100℃, with each interval being 10℃. The concentration difference between two adjacent temperature nodes is compared with the corresponding temperature difference. The release rate per unit temperature is obtained by dividing the concentration difference by the temperature difference. For example, if polyphenols are 0.46 mg / mL at 40℃ and 0.55 mg / mL at 50℃, the release rate is 0.009 mg / mL / ℃. This method is repeated for each temperature range, and the concentrations of each component are recorded. The release rate values ​​in each temperature zone are categorized by component to form multiple sets of rate value sequences indexed by temperature zone. In practice, it is necessary to maintain consistency between sample quality, dissolution time, volume, and detection method to ensure the comparability of concentration data. For example, in a set of samples, the release rate of caffeine is 0.004 mg / mL / ℃ in the 30℃ to 40℃ range, while it reaches 0.016 mg / mL / ℃ in the 70℃ to 80℃ range, indicating that the release trend changes significantly with increasing temperature. The rate values ​​of all components need to be formed into a standard format data structure, arranged in order of temperature zone, for subsequent analysis of the relationship between rate change magnitude and temperature response.

[0028] S202: Identify the variation range of release rate within a temperature range based on the component release rate sequence value set, screen temperature ranges where the rate change exceeds the release rate difference threshold, mark them as key release segments, and generate key temperature range identification segment values; Based on the obtained set of component release rate sequences, the variation in release rate between each temperature zone needs to be calculated sequentially. This is done by subtracting the rate value of the previous zone from the release rate of the next zone, obtaining the variation, and then comparing this variation with a preset release rate difference threshold. If the variation exceeds this threshold, it is identified as a critical release segment. This threshold is set by multiplying the average standard deviation of the rate variation between reference samples by a correction factor. For example, if the standard deviation is 0.003 and the correction factor is 2, then the threshold is 0.006 mg / mL / ℃. For instance, this applies to the release of polyphenols at 60℃ to 70℃. The release rate was 0.014 mg / mL / ℃, while it was 0.007 mg / mL / ℃ in the 50℃ to 60℃ range. The difference was 0.007, which exceeded the threshold. Therefore, 60℃ to 70℃ was determined to be the critical release range for polyphenols. All components were processed in this way to extract temperature zones with significant variations in different temperature ranges and to establish a label list. Each record should include the component name, corresponding temperature zone, release rate, variation value, and the result of whether the threshold was exceeded. Finally, the critical release range of multiple components was obtained to identify the temperature zones with active release in different samples.

[0029] S203: Based on the key temperature zone identification segment values, organize the key release segments in the tea sample according to the classification method, extract the temperature zone segments corresponding to the difference type, construct the correspondence structure between temperature zone and tea sample type, and generate a key release temperature zone structure table of tea sample. Based on the critical temperature range values, the corresponding critical release ranges in each tea sample are categorized and organized. The classification criteria can be based on the tea sample's processing technology, fermentation degree, or raw material source. For example, tea samples can be divided into fermented and non-fermented types. The critical temperature ranges exhibited in each type of sample are statistically analyzed. If fermented samples show a significant increase in polyphenol release rate concentrated in the 70℃ to 80℃ range, while non-fermented samples show a significant difference in amino acid release rate mainly in the 40℃ to 50℃ range, then the critical temperature range is extracted based on the sample characteristics. A correspondence structure is established between sample type and critical temperature range, including sample identification, component name, temperature range start and end points, and classification. At the same time, overlapping or similar temperature ranges should be merged to ensure the integration and comparability of the structural data. Finally, a critical release temperature range structure table for tea samples is generated. Each item should reflect the components actively released by different tea samples in different temperature ranges and their classification, providing classification data support for subsequent temperature-controlled tea sample evaluation.

[0030] Please see Figure 4 The specific steps of S3 are as follows: S301: Based on the distribution of segments in the key release temperature zone structure table of the tea sample, call the component concentration value and release rate value in each temperature zone, calculate the release difference value between components in the same segment, and generate a set of temperature zone release difference values ​​by combining the release difference distribution under the segment. Based on the distribution information of each temperature range in the key release temperature zone structure table of tea samples, the concentration and release rate values ​​of various components within each temperature range were retrieved one by one. Different components within the same range were compared pairwise to obtain the release difference value. The release difference between each component pair was the absolute difference in the corresponding release rate. For example, in the 60℃ to 70℃ range, the release rate of polyphenols was 0.016 mg / mL / ℃, and that of amino acids was 0.010 mg / mL / ℃, so the release difference value was 0.006 mg / mL / ℃. All component combinations within the range were processed sequentially, and all release difference values ​​were recorded. Then, a concentration analysis was performed on all difference values ​​within the range, including the maximum, minimum, average, and standard values. The standard deviation is used to establish the distribution data of release differences within the range. Each record should include the range number, the combination of components involved, the concentration and rate values ​​of each component, the release difference value and its ranking within the range. The entire process should maintain a uniform sample volume, dissolution time and analytical method to ensure data comparability. For example, in a range of 70℃ to 80℃, if the release rate of caffeine is 0.013 mg / mL / ℃ and that of polyphenols is 0.007 mg / mL / ℃, then the difference is 0.006 mg / mL / ℃. The mean and standard deviation of all difference values ​​in this range are then calculated for subsequent analysis of the difference distribution structure. All data are organized using the temperature range as an index to establish a structured set of temperature range release difference values.

[0031] S302: Based on the temperature zone release difference value set, the release difference value and component ratio composition value of the segment are jointly judged, and the segments that meet the release difference distribution concentration threshold and component ratio fitting degree threshold are selected to obtain the flavor characteristic coordination segment value. For each segment of the above-mentioned temperature-zone release difference value concentration, the distribution of its release difference value and the component ratio are jointly judged. Segments with relatively concentrated release differences and similar component concentrations are extracted. During the process, the standard deviation of the release difference values ​​of all components in each segment is first calculated. If the standard deviation is lower than the preset difference concentration threshold, such as 0.004 mg / mL / ℃, it is considered as a concentration of difference values. Then, the concentration ratio of each component in this segment is calculated. If the difference between the ratios of the main components is small and the overall distribution is similar, the ratio fit is used for judgment. The ratio fit threshold is set at 90%. That is, if the matching degree between the component composition reaches 90% or more, it is considered as a similar ratio composition. For example, in the 70℃ to 80℃ range, the proportion of polyphenols is 42%, amino acids are 40%, and caffeine is 18%. The one-dimensional difference between this ratio structure and the preset flavor reference model (the three ratios are 40%, 40%, and 20%) is within ±2%, and the overall relative error is small. The "proportion fit" here describes how closely the sample component proportions match the target model. It can be calculated by averaging or standard deviationing the differences between the proportions of the three component classes. When the deviation of all component proportions is less than the set tolerance (e.g., 5%), the fit is considered to be above 90%. Considering that the release difference in this range is between 0.005 and 0.007 mg / mL / ℃, with a standard deviation of 0.001, which is below the set difference concentration threshold, this temperature range can be marked as a flavor characteristic coordination zone.

[0032] S303: Extract temperature zone segments with concentrated flavor composition proportions based on flavor feature coordination zone values, establish a mapping relationship structure between flavor composition concentration and zone intervals, and organize them into flavor feature segment intervals of tea samples under temperature zones to generate flavor structure feature segment intervals of tea samples. Within the selected flavor characteristic coordination ranges, temperature zones with higher concentrations of dominant flavor components are further extracted as temperature zones with concentrated flavor composition. During processing, the concentration ratio of the highest-proportion component in each zone is statistically analyzed, and a standard for judging flavor composition concentration is established. For example, in a certain temperature zone segment, polyphenols account for 80%, amino acids for 15%, and caffeine for 5%. According to the concentration standard, this segment should be classified as a polyphenol-dominant segment with "concentrated composition." A correspondence is established between this flavor composition concentration and the start and end ranges of the temperature zones, and comparisons are performed among samples. If multiple tea samples exhibit a concentrated distribution characteristic of polyphenols exceeding 75% in the 60℃ to 70℃ range, they can be uniformly classified as "polyphenol-dominant concentrated composition" temperature zone segments. Based on this classification, a structured relationship table is constructed using dominant flavor components, concentration levels, and temperature zone segment numbers as indexes. Furthermore, flavor structure characteristic segment intervals for each tea sample in different temperature zones are formed for subsequent flavor combination optimization and blending modeling. In this invention, the concentration of flavor composition is divided into two categories: when a certain component accounts for more than 75% of the temperature range segment, it is determined to be 'concentrated'; when its proportion exceeds 85%, it is determined to be 'highly concentrated'; and cases where the proportion does not exceed 75% are not treated as concentrated segments.

[0033] Please see Figure 5 The specific steps of S4 are as follows: S401: Call the component release concentration data corresponding to the segment in the flavor structure feature segment interval of the tea sample, perform standard deviation scaling operation on the release concentration values ​​in the difference segment according to the minimum and maximum values ​​in the segment, unify the scale of components in the same interval, combine them into a multi-dimensional vector with consistent structural order, and generate a flavor release structure vector set. The algorithm retrieves the component release concentration data corresponding to each segment within the flavor structure feature segment of the tea sample. For each component in a differential segment, it extracts all release concentration values ​​within that segment and determines the minimum and maximum values ​​of that component in that segment as scaling reference boundaries. For example, in a certain temperature range, the lowest polyphenol concentration is 0.40 mg / mL and the highest is 0.85 mg / mL. The scaling operation adjusts the values ​​within this range. The difference between each original release concentration value and the minimum value of that segment is calculated, and then divided by the difference between the maximum and minimum values ​​to obtain a standardized value for that value in the range of 0 to 1. For example, when the original release concentration is 0.65 mg / mL, its scaled value... The value is approximately 0.56. The above process is repeated to process all components with different performance in this segment. After standard deviation scaling, the component values ​​in each segment are arranged and combined into a multidimensional vector according to a predetermined structural order. The structural order must be consistent. For example, the fixed order is polyphenols, amino acids, caffeine, etc. If the standardized values ​​of the three components in this segment are 0.56, 0.71, and 0.38 respectively, they are combined into a set of three-dimensional vectors. All flavor feature segment intervals are processed in this way. Finally, a flavor release structure vector set consisting of multiple vectors with consistent structure and the same dimension is generated. Each set of vectors represents the standardized release structure of a tea sample in a specific temperature range.

[0034] S402: Based on the flavor release structure vector set, call the source number and processing sequence code of the tea sample, add the source number and processing sequence value to each structure vector, and perform position matching and number classification on the multi-dimensional vector after the additional label to obtain the flavor structure mapping label value group. Based on the flavor release structure vector set, each vector is appended with the source number and processing sequence code of the tea sample to which it belongs. First, the source number of each structural vector is identified and its position in the processing flow is determined. For example, if a vector belongs to sample number T101 and is the 3rd segment processed, the source number T101 and the processing sequence value P3 are appended to the front of the vector to form an information vector with a number identifier. Then, all the labeled vectors are sorted according to the source number and processing sequence. All vectors belonging to the same tea sample number are grouped together, and the position of the vectors in the group is arranged according to the processing order. For example, number T102 contains 5 structural vectors, corresponding to processing sequences P1 to P5. After being arranged in order, the numbering is assigned to a unified unit. Further, all vector groups corresponding to different tea sample numbers are classified and labeled to complete the vector numbering and grouping process, generating flavor structure mapping identifier value groups. Each record contains the tea sample source number, the processing sequence of the corresponding segment, and the standardized multidimensional vector information of that segment, which is used for structural positioning and source tracing in subsequent processing.

[0035] S403: Based on the flavor structure mapping identifier value group, index and divide it according to the source number, extract the structure vector with the same source and summarize it into a unified number unit, organize it to form a flavor vector composition structure for matching and screening, and generate a flavor composition unit number set. Based on the flavor structure mapping identifier value group, all multidimensional structure vectors are categorized and organized according to their source numbers, establishing structural units indexed by the tea sample source number. Each source number corresponds to a number unit, and all structural vectors under that source are extracted and combined into this unit to form a flavor vector composition structure. In this structure, each vector represents the component release structure of the tea sample at a certain temperature range. The arrangement order of the vectors strictly follows the processing sequence to maintain consistency in time or processing order. For example, if tea sample number T201 contains 5 flavor fragments, corresponding to 5 vectors, these 5 vectors are combined in sequence to form a structure set to present the flavor performance characteristics of the tea sample at different stages. This structure set must ensure consistency in all dimensions and must not have missing items. If there are null values ​​or data anomalies, they must be filled or removed before combination processing. After the classification of all tea sample structure sets is completed, the output is a flavor composition unit number set. Each number set contains a flavor structure vector group of all valid fragments under a tea sample source and its number index information.

[0036] Please see Figure 6 The specific steps of S5 are as follows: S501: Extract vector features based on the flavor component unit number set, compare the differences with the preset flavor target parameters, calculate the offset value between the combination number and the target parameter, and filter according to the offset value order to obtain the combination number with the lowest degree of difference, generating the number group with the smallest flavor distance. After extracting the corresponding vector features based on the flavor component number set, the multidimensional flavor parameters under each number group need to be compared one by one with the preset flavor target parameters. First, extract the flavor structure vector corresponding to each number group. The vector contains the standardized release values ​​of multiple components. For example, in the vector corresponding to a certain number, the first dimension represents the polyphenol release intensity, the second dimension represents amino acids, and the third dimension represents caffeine, with values ​​of 0.62, 0.53, and 0.47 respectively. The corresponding target flavor parameters are 0.60, 0.55, and 0.50. Calculate the corresponding differences from each dimension and merge them to obtain the multidimensional flavor structure vector. Differences across all dimensions are unified into a single comprehensive offset value. After performing the same operation on all flavor components, the components are sorted by comprehensive offset value from smallest to largest. The combination of numbers with the smallest offset value is selected as the group of numbers whose flavor performance is closest to the target. To ensure that the degree of offset is based on actual evidence, a reference range can be set. For example, an offset less than 0.05 is considered to be minimal, 0.05 to 0.15 is considered to be moderate, and an offset greater than 0.15 is considered to be significant. When multiple combinations fall within the range of minimal difference, the group with the lowest value is selected as the optimal numbering group, ultimately forming the numbering group with the smallest flavor distance.

[0037] S502: Extract the corresponding flavor vector based on the flavor distance minimum number group, call the dimensional flavor component parameter values, calculate the average parameter under the same dimension, integrate them into a complete parameter set, and obtain the component parameter mean structure; Based on the flavor distance minimum numbering group, all relevant flavor vectors are extracted. Parameters are collected and averaged for each dimension. First, all combination vectors in the numbering group are identified, and the parameter values ​​for each dimension are extracted one by one. For example, a combination contains three vectors with values ​​of 0.61, 0.59, and 0.60 for polyphenols, 0.56, 0.57, and 0.58 for amino acids, and 0.52, 0.51, and 0.50 for caffeine. All values ​​for the same dimension are summarized, and the mean parameter value under that dimension is calculated to represent the overall characteristic value of the combination in that dimension. This process is performed sequentially on all dimensions, ultimately obtaining a complete set of structural parameters composed of the mean values ​​of each component dimension. In operation, it is recommended to classify by dimension before performing averaging to avoid dimension confusion. When the number of samples exceeds ten groups, a spreadsheet tool can be used for column-level summarization, and the integrated structure can be output by dimension. The final output structure should ensure that each dimension has valid values ​​without missing or abnormal biases.

[0038] S503: Organize all parameter content according to the mean structure of component parameters, arrange them uniformly into a standard structural form, and label the corresponding dimension information to generate the blended flavor parameter input structure; Based on the mean structure of the component parameters obtained in the previous stage, all parameters need to be organized into a unified input structure in a fixed order, and each parameter should be labeled with its corresponding dimension information. During the organization process, the dimensions should be arranged in a preset order, such as polyphenols first, then amino acids, and then caffeine, forming a standard arrangement format. The dimension name should be paired with the corresponding value to express the information. For example, polyphenols are 0.60, amino acids are 0.57, and caffeine is 0.51. If new components such as theaflavins or aromatic components are added later, they should also be processed in the same way to ensure that each parameter has a clear dimension label and avoid confusion caused by unclear dimensions during subsequent blending. To maintain structural consistency, key-value pairs or itemized labels can be used to represent the content of each parameter. At the same time, attention should be paid to the accuracy and completeness of each dimension name. Abbreviations or vague abbreviations should not be used. Each set of blended flavor parameter input structures should include all dimension information to ensure that the structure is complete and without omissions.

[0039] Please see Figure 7 A system for intelligent blending of Liubao tea, comprising: The tea sample data acquisition module acquires Liubao tea blank samples, collects the concentration values ​​of three components—polyphenols, caffeine, and amino acids—at room temperature, sets multiple temperature control nodes and records the dissolution concentration of components at each node in sequence, and organizes the component behavior information of the sample by the correspondence between component concentration and temperature nodes, generating a tea sample high-temperature dissolution behavior information set. The temperature zone release identification module calculates the release rate value of the temperature zone based on the concentration change trend of the concentrated components at different temperature control nodes according to the high temperature dissolution behavior information of tea samples, identifies continuous segments with prominent release rate changes, marks and classifies the samples, and generates a key release temperature zone structure table of tea samples. The flavor segment extraction module extracts the difference in release concentration of components within the segment based on the temperature segment in the key release temperature zone structure table of the tea sample. It identifies segments with concentrated differences and similar component proportions, and filters and numbers the segments that meet the criteria to generate flavor structure feature segment intervals of the tea sample. The flavor vector construction module calls the release concentration data of components in the flavor structure feature segment interval of the tea sample, performs standardization processing, assembles them into a structure vector, assigns numbers according to the tea sample source information and processing order, and generates a set of flavor component unit numbers; The flavor parameter screening module calls preset flavor parameters based on the structural vector features of the flavor component number set, compares the vector distance values, filters the combination numbers with the smallest differences, integrates the corresponding component concentration features, and generates the blended flavor parameter input structure.

[0040] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for intelligent blending of Liubao tea, characterized in that, Includes the following steps: S1: Detect the room temperature concentrations of polyphenols, caffeine, and amino acids in Liubao tea leaves, set the heating range, collect dissolution data at temperature nodes, establish the relationship between temperature and component concentration, and generate a high-temperature dissolution behavior information set for tea samples. S2: Based on the concentration change trend of the high-temperature dissolution behavior information set of the tea sample, calculate the release rate of the differential temperature zone, identify the rate change segment, mark it as the key release zone, and generate a key release temperature zone structure table of the tea sample. S3: Based on the segment distribution in the key release temperature zone structure table of the tea sample, calculate the component release difference value of the temperature zone, identify the segments with concentrated differences and similar composition, mark them as flavor feature coordination areas, screen the temperature zone segments with high flavor composition ratio, and generate tea sample flavor structure feature segment intervals. S4: Extract the component release concentration data from the flavor structure feature segment interval of the tea sample, standardize and combine them into a structural vector, construct a set of flavor feature units according to the source and processing order of the tea sample, and generate a set of flavor component unit numbers. S5: Based on the comparison between the vector features of the flavor component number set and the target parameters, filter the combination with the smallest distance value, summarize the flavor parameters, and generate the blended flavor parameter input structure.

2. The method for intelligent blending Liubao tea according to claim 1, characterized in that, The tea sample high-temperature dissolution behavior information set includes temperature nodes, component concentrations, and dissolution trends. The tea sample key release temperature zone structure table includes key temperature zone locations, component release rates, and temperature zone types. The tea sample flavor structure feature segment intervals include coordinated region distribution, flavor composition ratios, and component difference values. The flavor composition unit number set includes flavor structure vectors, tea sample numbers, and processing order information. The blended flavor parameter input structure includes minimum distance combination numbers, target flavor parameters, and flavor data summaries.

3. The method for intelligent blending Liubao tea according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Obtain the room temperature dissolution amount of polyphenols, caffeine and amino acids in Liubao tea blank samples, extract the component concentration values ​​in the dissolution liquid, calculate the dissolution ratio based on the change relationship between the initial component content and the dissolution concentration, integrate the characteristic information under the current temperature conditions, and generate a room temperature dissolution reference coefficient matrix. S102: Based on the room temperature dissolution reference coefficient matrix, set the heating range, collect the component concentration values ​​in the dissolution solution at temperature nodes in stages, calculate the corresponding ratio based on the difference between the node dissolution concentration and the reference coefficient, sort out the change trend of the components in stages, and generate a set of dissolution distribution values ​​at temperature nodes. S103: Extract the concentration change sequence of components at different temperature nodes based on the dissolution distribution value set at the temperature nodes, calculate the concentration change rate and associate it with the corresponding temperature range, construct a data structure that associates the concentration change trend with temperature, and generate a high-temperature dissolution behavior information set of tea samples.

4. The method for intelligent blending Liubao tea according to claim 3, characterized in that, The specific steps of S2 are as follows: S201: Extract the concentration change trend of components in the temperature range based on the high-temperature dissolution behavior information set of the tea sample, organize the concentration change data according to the temperature range, calculate the release rate of components in the different temperature ranges, and generate a component release rate sequence value set. S202: Based on the component release rate sequence value set, identify the variation range of release rate in the temperature range, screen the temperature range where the rate change exceeds the release rate difference threshold, mark it as a key release segment, and generate a key temperature range identification segment value. S203: Based on the key temperature zone identification segment value, organize the key release segments in the tea sample according to the classification method, extract the temperature zone segments corresponding to the difference type, construct the correspondence structure between temperature zone and tea sample type, and generate the key release temperature zone structure table of tea sample.

5. The method for intelligent blending Liubao tea according to claim 4, characterized in that, The specific steps for S3 are as follows: S301: Based on the distribution of the segments in the key release temperature zone structure table of the tea sample, call the component concentration value and release rate value in each temperature zone, calculate the release difference value between components in the same segment, and generate a set of temperature zone release difference values ​​by combining the release difference distribution under the segment. S302: Based on the temperature range release difference value set, the release difference value and component ratio composition value of the segment are jointly judged, and the segments that meet the release difference distribution concentration threshold and component ratio fitting degree threshold are selected to obtain the flavor feature coordination segment value. S303: Extract the temperature range segments with concentrated flavor composition proportions based on the flavor feature coordination segment values, establish a mapping relationship structure between flavor composition concentration and segment intervals, and organize them to form the flavor feature segment intervals of the tea sample under the temperature range, generating the flavor structure feature segment intervals of the tea sample.

6. The method for intelligent blending Liubao tea according to claim 5, characterized in that, The specific steps of S4 are as follows: S401: Call the component release concentration data corresponding to the segment in the flavor structure feature segment interval of the tea sample, perform standard deviation scaling operation on the release concentration values ​​in the different segments according to the minimum and maximum values ​​in the segment, unify the scale of components in the same interval, combine them into a multi-dimensional vector with consistent structural order, and generate a flavor release structure vector set. S402: Based on the flavor release structure vector set, call the tea sample source number and processing sequence code, add the source number and processing sequence value to each structure vector respectively, and perform position matching and number classification on the multi-dimensional vectors after the additional identifier to obtain the flavor structure mapping identifier value group. S403: Based on the flavor structure mapping identifier value group, index and divide it according to the source number, extract the structural vectors with the same source and summarize them into a unified numbering unit, organize them to form a flavor vector composition structure of the combination and screening, and generate a flavor composition unit number set.

7. The method for intelligent blending Liubao tea according to claim 6, characterized in that, The specific steps of S5 are as follows: S501: Extract vector features based on the flavor component number set, compare the differences with the preset flavor target parameters, calculate the offset value between the combination number and the target parameter, and filter according to the offset value order to obtain the combination number with the lowest degree of difference, and generate the number group with the smallest flavor distance. S502: Extract the corresponding flavor vector based on the flavor distance minimum number group, call the dimensional flavor component parameter values, calculate the average parameter under the same dimension, integrate them into a complete parameter set, and obtain the component parameter mean structure. S503: Organize all parameter contents according to the mean structure of the component parameters, arrange them uniformly into a standard structure form, and label the corresponding dimension information to generate the blended flavor parameter input structure.

8. The method for intelligent blending of Liubao tea according to claim 1, wherein the Liubao tea base is raw tea leaves used for preliminary drying or mid-stage fermentation before processing Liubao tea, and has clear origin, degree of fermentation, moisture content and basic data of flavor substances; The polyphenols, caffeine, and amino acids mentioned refer to the three main flavor substances commonly found in tea. Among them, polyphenols contribute to the astringency of tea soup, caffeine affects bitterness, and theanine contributes to freshness. All of them can be quantitatively determined by high performance liquid chromatography or ultraviolet spectrophotometry. The heating range refers to the temperature change range set during the experiment to simulate brewing or processing scenarios. The range generally covers the temperature from room temperature to the temperature before boiling and is divided into multiple fixed temperature nodes for data acquisition. The temperature nodes are specific temperature points used for recording and sampling during the heating process, and each node corresponds to a detection operation of the tea sample dissolution amount. The dissolution data are recorded values ​​of the concentration of the target component in the solution, collected through experimental means. The temperature range where the release rate changes significantly refers to the temperature range in which the increase in the release rate of the target component between consecutive temperature nodes is greater than the average rate of change of the component. The critical release zone refers to the temperature range in which the release rate of components in the tea sample fluctuates significantly or where the total amount of components released accounts for a high percentage.

9. The method for intelligent blending Liubao tea according to claim 1, characterized in that, The difference concentration refers to the standard deviation of the release difference of different components within the target segment being less than 0.005 mg / mL; The term "similar composition" means that the difference in the concentration ratio of each component among the tea samples does not exceed ±10%. The segment with similar composition refers to the segment within the same interval where the concentration ratio of multiple target components has a low variance or deviation rate among tea samples. The flavor characteristic coordination region refers to the temperature range in which the rate of change of component release ratio between adjacent temperature nodes is less than 10%, and the rate of difference in component ratio between tea samples in the same temperature range does not exceed ±5%. The temperature range segment in the flavor composition ratio refers to the temperature range in which, within a certain temperature range, the proportion of polyphenols, caffeine, and amino acids in the total flavor components of the tea sample is significantly higher than that of the temperature range, and the variance of the variation among the proportions of the three is small. The structure vector is a fixed-dimensional vector formed by combining the normalized values ​​of polyphenols, caffeine, and amino acids in sequence. The set of flavor feature units refers to a structured information set consisting of multiple structural vectors and their source tea sample numbers; The flavor composition unit number set is a set of numbers formed by generating a unique number for each flavor feature unit according to the tea sample source, temperature zone segment number and composition vector combination information. The distance value refers to the numerical difference between the flavor component vector and the target flavor parameter vector. The blended flavor parameter input structure refers to the structured blending information composed of selected flavor component unit numbers and their component parameter sets; The distance value refers to the Euclidean distance between the structure vector of the flavor constituent unit and the target flavor parameter vector.

10. A system for intelligent blending of Liubao tea, characterized in that, The system is used to implement the intelligent blending method of Liubao tea according to any one of claims 1-9, the system comprising: The tea sample data acquisition module acquires Liubao tea blank samples, collects the concentration values ​​of three components—polyphenols, caffeine, and amino acids—at room temperature, sets multiple temperature control nodes and records the dissolution concentration of components at each node in sequence, and organizes the component behavior information of the sample by the correspondence between component concentration and temperature nodes, generating a tea sample high-temperature dissolution behavior information set. The temperature zone release identification module calculates the release rate value of the temperature zone based on the concentration change trend of the concentrated components of the high-temperature dissolution behavior information of the tea sample under the differential temperature control node, identifies the continuous segments with prominent release rate changes, marks and classifies the samples, and organizes and generates a key release temperature zone structure table of the tea sample. The flavor segment extraction module extracts the difference in release concentration of components within the segment based on the temperature segment in the key release temperature zone structure table of the tea sample, identifies segments with concentrated difference distribution and similar component proportions, and filters and numbers the segments that meet the conditions to generate flavor structure feature segment intervals of the tea sample. The flavor vector construction module calls the release concentration data of components in the flavor structure feature segment interval of the tea sample, performs standardization processing, assembles them into a structure vector, assigns numbers according to the tea sample source information and processing order, and generates a set of flavor component unit numbers. The flavor parameter screening module calls preset flavor parameters based on the structural vector features of the flavor component number set, compares the vector distance values, filters the combination numbers with the smallest differences, integrates the corresponding component concentration features, and generates a blended flavor parameter input structure.

Images