New tea drink syrup raw material ratio detection method and system based on fingerprint spectrum and storage medium

Abstract

The invention relates to the technical field of food component detection, and particularly discloses a new tea drink syrup raw material ratio detection method and system based on a fingerprint spectrum and a storage medium, and the method comprises the following steps: (1) constructing a standard raw material multi-dimensional feature database; (2) performing HPLC detection on the syrup sample to be detected; (3) extracting multi-dimensional detection parameters of the syrup sample to be detected; (4) inputting the multi-dimensional detection parameters into a pre-established proportion calculation model; and (5) outputting the raw material ratio of the syrup sample to be detected. According to the method, a multi-dimensional HPLC fingerprint database fusing main characteristic peaks, trace characteristic peaks and characteristic ratios is constructed, and an intelligent matching calculation model with self-checking and iterative optimization capabilities is established, so that automatic and high-precision dual reduction of names and proportions of specific raw materials in the new tea drink syrup is realized; the method effectively breaks through the limitation of the traditional method in the recognition and quantification of the complex carbohydrate source, and provides an integrated technical solution from raw material quality control, formula research and development to competitive product analysis for the industry.

Description

Technical Field

[0001] This invention relates to the field of food component detection technology, and specifically discloses a new method, system and storage medium for detecting the proportion of raw materials for tea beverage syrups based on fingerprint spectroscopy. Background Technology

[0002] The new-style tea beverage industry has flourished in recent years, with its product flavors highly dependent on the complex formula of the base syrup. Syrups are typically composed of white sugar, glucose, fructose syrup, water, and various functional sweeteners or flavorings. Within the industry chain, whether for raw material quality control, product consistency monitoring, formula reverse engineering, or competitor analysis, achieving precise and rapid traceability and proportioning of all ingredients in the finished syrup is of paramount commercial and technological value.

[0003] Currently, the industry typically uses the following methods for component analysis of syrup products: (1) Traditional traceability method relying on supply chain records: This method relies entirely on production records, ingredient lists, and certification documents provided by suppliers. Its drawbacks are obvious: First, its authenticity and accuracy are entirely based on the integrity of human records, and it cannot verify the physicochemical properties of the syrup itself, making it susceptible to interference from false information or opaque supply chains; Second, this method cannot perform any effective component analysis for syrups with unknown formulas (such as competing products).

[0004] (2) Inference methods based on single index detection: For example, using an Abbe refractometer to measure total solids content, or using a polarimeter to measure total optical rotation. These methods can only provide a single macroscopic index of the syrup as a whole, and cannot distinguish which specific raw materials contributed to it, let alone calculate the precise proportion of multiple sugar sources. For the analysis of complex formulas, such methods are inadequate.

[0005] (3) Conventional chromatographic analysis: High-performance liquid chromatography (HPLC) and other techniques have been used for carbohydrate analysis, capable of separating and detecting single components such as fructose, glucose, and sucrose in samples. However, existing technologies are mostly limited to qualitative identification of components or determination of the content of single components. For example, HPLC can detect the presence of sucrose in a sample, but it cannot effectively distinguish whether the sucrose comes from pure white sugar or is an inherent component of other natural syrups (such as maple syrup). More importantly, when faced with a complex raw material such as high-fructose corn syrup, which is itself a mixture of fructose and glucose, traditional methods struggle to answer a crucial question: do the detected fructose and glucose come from the added high-fructose corn syrup, or from separately added pure fructose and pure glucose? This constitutes a blind spot in component identification for existing chromatographic technologies in traceability applications.

[0006] Furthermore, existing technical solutions generally lack a systematic approach to accurately and automatically correlate and map the multidimensional information obtained from chromatographic separation (such as main peak, trace characteristic peaks, and characteristic peak ratios) with the identity and proportion of raw materials. Most methods rely on empirical judgment or simple linear calculations, failing to construct an intelligent computational model with self-verification capabilities. This results in a significant decrease in the accuracy and reliability of analytical results when faced with complex scenarios such as fluctuations in raw material purity, the coexistence of multiple sugar sources, or the presence of undeclared components.

[0007] Therefore, it is necessary to propose a new tea beverage syrup ingredient ratio detection method based on fingerprint spectrum that can identify the types of ingredients, more accurately trace the specific syrup ingredient names, accurately quantify the mass ratio of each ingredient, and handle complex formulas and identify unknown ingredients. Summary of the Invention

[0008] To address the aforementioned issues, this invention provides a method, system, and storage medium for detecting the proportion of raw materials in new tea beverage syrups based on fingerprint spectroscopy. By constructing a multidimensional HPLC fingerprint spectroscopy library that integrates main characteristic peaks, trace characteristic peaks, and characteristic ratios, and establishing an intelligent proportion calculation model with self-verification and iterative optimization capabilities, it achieves automated and high-precision dual reconstruction of the specific raw material names and proportions in new tea beverage syrups. This effectively overcomes the limitations of traditional methods in identifying and quantifying complex sugar sources, providing the industry with an integrated technical solution from raw material quality control and formula development to competitor analysis.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows: A method for detecting the proportion of raw materials for new tea beverage syrups based on fingerprint spectroscopy, comprising the following steps: (1) Construct a multidimensional characteristic database of standard raw materials, which stores HPLC characteristic information of white sugar, glucose and fructose syrup; (2) Perform HPLC analysis on the syrup sample to be tested to obtain its HPLC fingerprint spectrum; (3) Based on the HPLC fingerprint, extract multi-dimensional detection parameters of the syrup sample to be tested. The multi-dimensional detection parameters include the area ratio of the main characteristic peak, information on at least one trace characteristic peak, and at least one characteristic peak ratio. (4) Input the multi-dimensional detection parameters into the pre-established ratio calculation model; (5) Using the ratio calculation model, output the raw material ratio of at least white sugar, glucose and fructose syrup in the syrup sample to be tested.

[0010] The technical principles of this solution consist of two parts: (1) Establishing a "personal identification card" for raw materials. Each raw material, due to its unique chemical composition and processing, exhibits a unique spectral fingerprint in HPLC analysis. We record not only its most prominent features (main characteristic peaks, such as the DP2 peak of granulated sugar), but also its trace, unique markers (auxiliary characteristic peaks, such as specific oligosaccharide peaks in glucose syrup), and the stable relationships of its internal components (characteristic ratios, such as the fixed fructose / glucose ratio in glucose syrup). This transforms the database from a simple list of ingredients into a three-dimensional, cross-referenced "identity recognition system." For example, detecting only fructose and glucose doesn't determine the source; however, detecting both in proportions matching those of high-fructose corn syrup... This confirms that the ingredient comes from high-fructose corn syrup rather than a mixture of other monosaccharides.

[0011] (2) Multi-dimensional parameter extraction and model calculation The HPLC chromatogram of the syrup in the analyte is a composite signal resulting from the superposition of fingerprints from all raw materials. The core task of the proportioning calculation model is to decompose this composite signal and reconstruct the contribution of each raw material.

[0012] The model first performs preliminary proportion calculations based on the main characteristic peaks of each raw material. For example, based on the intensity of the DP2 peak and the purity of the white sugar, the proportion of white sugar (A) is initially calculated.

[0013] Cross-validation and decoupling are key to the method. For example, the detected total glucose ( The glucose may originate from two sources: pure glucose raw material (B) and the glucose fraction in fructose syrup (C × 40%). The model utilizes constraints such as eigenvalue ratios to establish a system of equations: Total glucose detected = B × 98% + C × 40% Total fructose detected = C × 55% Total solids S = A + B + C The model uses an iterative algorithm to continuously adjust the estimated values ​​of A, B, and C, minimizing the difference between the theoretical spectrum (including characteristic ratios, optical rotation, etc.) calculated based on these estimates and the actual detection data of the syrup being tested. This process effectively decouples the contributions of different raw materials to the same signal, thereby deriving the most probable combination of proportions.

[0014] Furthermore, the construction of the standard raw material multidimensional feature database includes: Record the main characteristic peak and its theoretical purity for each standard raw material, wherein the main characteristic peak is either DP1 peak or DP2 peak; Record at least one auxiliary characteristic peak for each standard raw material, wherein the auxiliary characteristic peak is a trace component peak or a specific degree of polymerization peak that distinguishes it from other raw materials; Record the characteristic ratio of at least one standard raw material, wherein the characteristic ratio is the peak height ratio or peak area ratio of two different characteristic peaks in its HPLC chromatogram.

[0015] By constructing a multi-dimensional feature database integrating main characteristic peaks, auxiliary characteristic peaks, and characteristic ratios, its core function is to expand the HPLC information of each standard raw material from a single indicator to a three-dimensional identification system: using the main characteristic peak for preliminary identification and baseline quantification of the raw material, using auxiliary characteristic peaks (such as trace components or specific degree of polymerization peaks) to solve the problem of distinguishing structurally similar raw materials, and using the characteristic ratio as an inherently stable parameter to achieve accurate identification and cross-validation of composite raw materials (such as fructose syrup). This multi-dimensional design achieves a leap from rough judgment relying on a single fingerprint to accurate traceability by integrating multiple information, effectively improving the distinguishability of raw material identification, anti-interference ability, and the accuracy and reliability of ratio calculation.

[0016] Furthermore, the standard raw material includes fructose syrup, whose multidimensional characteristic information includes: the main characteristic peak is the DP1 peak, and the theoretical purity is ≥95%; the characteristic ratio is the ratio of the fructose peak to the glucose peak in its DP1 peak. The ratio ranges from 1.2 to 1.4.

[0017] For high-fructose corn syrup, a key complex raw material, this study simultaneously defines its main characteristic peak (DP1 peak) and the ratio of its intrinsic characteristics (DP1 peak). A dual verification mechanism was constructed. Its core lies in the fact that while the DP1 peak alone can determine the presence of fructose and glucose, it cannot trace the origin back to the specific raw material, high-fructose corn syrup; therefore, the inherent and stable fructose / glucose peak ratio is introduced. (1.2-1.4) Serves as an identification marker, fundamentally distinguishing it from separately added monosaccharide mixtures. This approach enables precise qualitative identification of fructose syrup and allows for quantitative calculation of it as a whole unit through this fixed ratio, effectively solving the problem of tracing and quantifying complex raw materials in syrups and significantly improving the accuracy and reliability of ratio analysis.

[0018] Furthermore, the standard ingredients also include honey and / or maple syrup, wherein: The multidimensional characteristic information of honey includes: the main characteristic peak is the DP1 peak, and the auxiliary characteristic peaks of organic acids or amino acids that distinguish it from fructose syrup; The multidimensional characteristic information of maple syrup includes: the main characteristic peak is the DP2 peak, and the auxiliary characteristic peaks of characteristic phenolic substances that distinguish it from white sugar.

[0019] By incorporating honey and maple syrup into a standard raw material database and precisely defining their multidimensional characteristic information, accurate traceability of natural complex sugar sources has been achieved. Although honey and fructose syrup share the DP1 main characteristic peak, and maple syrup and white sugar share the DP2 main characteristic peak, the introduction of auxiliary characteristic peaks of organic acids / amino acids unique to honey and characteristic phenolic substances unique to maple syrup as "chemical fingerprints" has successfully overcome the technical bottleneck of traditional methods that cannot distinguish different raw material sources based solely on the main peak. This significantly improves the distinguishing power and specificity of the detection method, enabling the scheme to accurately identify whether these high-value natural raw materials have been used in syrups, effectively preventing misjudgment of ingredients and label fraud, and greatly expanding the application scope of the technology in the quality control and traceability of high-end syrup products.

[0020] Furthermore, the method for establishing the proportioning calculation model includes: The initial proportioning was calculated based on the area ratio of the main characteristic peaks of each raw material. Constraints based on the aforementioned characteristic ratios are introduced to verify and correct the initial proportion calculation results. By using an iterative optimization algorithm, the difference between the feature ratios predicted by the model and the feature ratios measured from the HPLC chromatogram of the syrup to be tested is minimized, thereby outputting the optimal raw material ratio.

[0021] The core of the proportioning calculation model constructed in this scheme lies in its three-level progressive calculation logic—initial estimation, constraint verification, and iterative optimization—upgrading traditional unidirectional linear calculation into an intelligent system with self-correcting capabilities. The model first performs a preliminary proportioning estimate based on the main characteristic peak. Then, it introduces the inherent characteristic ratios of the raw materials as key constraints to forcibly verify the initial results. Finally, it dynamically adjusts the proportions of each raw material through an iterative optimization algorithm to minimize the difference between the model's predicted values ​​and the measured values ​​of the samples. This design completely changes the traditional mode that relies on a single parameter and manual calculation. Its effect is a significant improvement in the accuracy and reliability of proportioning analysis for complex syrup systems. It can effectively compensate for the impact of raw material purity fluctuations and detection errors, ultimately outputting a proportioning result that is most reasonable and optimized from a chemometric perspective.

[0022] Furthermore, the method also includes introducing verification parameters from an independent detection instrument, including the sample optical rotation measured by a polarimeter; The ratio calculation model is further configured to: calculate the predicted optical rotation based on the output raw material ratio and the theoretical specific optical rotation of each raw material, and compare the predicted optical rotation with the measured optical rotation; When the deviation between the predicted value and the measured value exceeds a preset threshold, the model is triggered to recalculate or output an early warning message.

[0023] By introducing the sample optical rotation measured by a polarimeter as a validation parameter independent of HPLC, a cross-validation mechanism across detection technologies was constructed. The principle is to utilize the unique specific rotation of different sugar sources—a physicochemical property—to forcibly compare the theoretical predictions output by the proportioning calculation model with the instrument's measured values. When the deviation exceeds a threshold, the inconsistency triggers a recalculation of the model or issues an early warning. This design forms a closed-loop feedback system, significantly improving the robustness and reliability of the entire detection method. It effectively identifies and corrects errors caused by batch differences in raw materials, interference from coexisting substances, or model calculation biases, ultimately ensuring the scientific validity and reliability of the proportioning traceability results.

[0024] Furthermore, the method also includes a residual analysis step: After calculating the proportions of all known raw materials, the residual peak signals in the HPLC fingerprint that were not explained by the characteristic peaks of the known raw materials were analyzed. If the residual peak signal exceeds a preset threshold, it is determined that there is an undeclared sugar source or additive, and the residual peak signal is compared with the standard raw material multidimensional feature database to assist in identification.

[0025] The residual analysis step introduced in this scheme leverages the completeness of HPLC fingerprint information to proactively detect and warn of undeclared components by monitoring and quantifying residual signals that cannot be explained by known raw material characteristic peaks. This step extends the analytical logic from verifying the known to discovering the unknown. When the residual signal exceeds a threshold, the system can not only determine the presence of undeclared sugar sources or additives but also provide crucial clues for identifying unknown components by comparing them with a multidimensional feature database. This design significantly enhances the method's depth of insight and risk warning value, effectively endowing the detection scheme with the powerful ability to detect formulation fraud, identify non-standard additives, and assist in the identification of new components in complex practical applications. This provides decisive support for comprehensive quality control and accurate formulation analysis.

[0026] Furthermore, a system for implementing a new method for detecting the proportion of raw materials in tea beverage syrups based on fingerprint spectroscopy includes: High-performance liquid chromatography (HPLC) is used to obtain HPLC fingerprints of standard raw materials and syrup samples to be tested. The data processing unit is configured as follows: a. Construct and store a multidimensional feature database of the standard raw materials; b. Extract the multi-dimensional detection parameters from the HPLC fingerprint of the syrup to be tested; c. Run the ratio calculation model and output the raw material ratio results.

[0027] By systematically integrating a high-performance liquid chromatograph (HPLC) with a dedicated data processing unit, an automated detection system was constructed, encompassing the entire process from data acquisition and feature extraction to intelligent computation and result output. Its core principle lies in the deep integration of the chemical analysis capabilities of multidimensional fingerprint spectroscopy with the intelligent decision-making functions of computer software. The chromatograph accurately acquires the underlying chemical signals of raw materials and samples, while the data processing unit executes a series of core algorithms, including database construction, multidimensional parameter extraction, and the running of a self-verifying proportioning model. This integrated hardware and software design successfully transforms the complex, experience-dependent process of syrup proportioning analysis into an efficient, objective, and reproducible standardized automated operation, significantly improving the method's practicality, accuracy, and its value for industrial application.

[0028] Furthermore, the system also includes a polarimeter, which is communicatively connected to the data processing unit and is used to provide the verification parameters to the proportioning calculation model.

[0029] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for detecting the proportion of raw materials for new tea beverage syrups based on fingerprint mapping.

[0030] The beneficial effects of this invention are specifically reflected in: (1) A qualitative leap from component identification to the accuracy of raw material traceability Traditional methods can only answer "what it contains" (e.g., sucrose, fructose, glucose). This method can accurately answer where it comes from. It can clearly distinguish whether sucrose comes from white sugar or maple syrup, and can clearly determine whether fructose and glucose come from high-fructose corn syrup or are separate single sugar sources, completely eliminating blind spots in component identification.

[0031] (2) A significant improvement in analytical ability from qualitative judgment to precise quantitative analysis By leveraging the mathematical decoupling capabilities of the intelligent model, this method can not only identify products containing high-fructose corn syrup, but also accurately calculate its proportion in the final product, achieving a quantitative restoration of complex mixed syrup ratios—something that traditional empirical judgment or simple linear calculations cannot accomplish.

[0032] (3) It has robustness against interference and self-verification. The model incorporates multi-dimensional parameters such as eigenvalue ratio and optical rotation for cross-validation, enabling the system to self-detect and correct biases. When the initial calculation results do not match a certain validation parameter, the model automatically optimizes, significantly reducing errors caused by fluctuations in raw material batch purity or minor detection errors, resulting in more reliable results.

[0033] (4) Powerful early warning and expansion capabilities for unknown objects Through residual analysis, this method can detect anomalous signals in the chromatogram that cannot be explained by known raw materials, thus providing early warning of the possible presence of undeclared raw materials or additives in the formulation, offering crucial clues for quality control and competitor analysis. Furthermore, the multidimensional feature database is an extensible system that can be expanded at any time with the addition of new raw materials (such as honey and mogrosides), continuously broadening the method's applicability.

[0034] In summary, this technical solution combines multidimensional HPLC fingerprinting with intelligent computing models to construct a precise, reliable, and automated syrup ratio analysis tool, providing unprecedented core technical support for the new tea beverage industry in terms of quality control, formula innovation, and supply chain management. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the method for detecting the proportion of raw materials for new tea syrup based on fingerprint spectrum in Embodiment 1 of the present invention.

[0036] Figure 2 This is a schematic diagram of the fingerprint spectrum-based new tea syrup ingredient ratio detection system in Embodiment 1 of the present invention.

[0037] Figure 3 This is a schematic diagram of the process of running the fingerprint spectrum-based method for detecting the proportion of raw materials for new tea syrups in Experiment Example 1 of the present invention. Detailed Implementation

[0038] The specific implementation method is described below with reference to the accompanying drawings.

[0039] Example 1 The implementation process of a new method, system, and storage medium for detecting the proportion of raw materials in tea beverage syrups based on fingerprint spectroscopy is as follows: First, a multidimensional characteristic database of standard raw materials is constructed, and high-performance liquid chromatography (HPLC) analysis is performed on three basic raw materials: granulated sugar, glucose, and fructose syrup. The characteristic information of granulated sugar includes: the main characteristic peak is DP2, and the theoretical purity is 99%; the characteristic information of glucose includes: the main characteristic peak is DP1, and the theoretical purity is 98.5%; the characteristic information of fructose syrup includes: the main characteristic peak is DP1, and the theoretical purity is 96%, with a characteristic ratio of... It is 1.32.

[0040] Take 1.0 g of the syrup sample to be tested, dilute to 10 mL with ultrapure water, filter through a 0.45 μm microporous membrane, and then perform high-performance liquid chromatography (HPLC) analysis. The chromatographic conditions were: amino column (4.6 mm × 250 mm, 5 μm), column temperature 35℃, mobile phase acetonitrile-ultrapure water (75:25, v / v), flow rate stable at 1.0 mL / min, and a refractive index detector (RID) with the detection temperature consistent with the column temperature. Simultaneously, the total solids content S of the sample was measured using an Abbe refractometer (S = 69.7 g / 100 g, i.e., the total percentage of carbohydrates; water content W = 100%). S). After obtaining the HPLC fingerprint of the sample, the following multi-dimensional detection parameters were extracted: the area ratio of DP1 peak was X1=65.2%, the area ratio of DP2 peak was X2=31.5%, and the measured ratio of fructose peak to glucose peak in DP1 peak. The value is 1.30. Inputting the above parameters into the proportion calculation model, the model first performs an initial proportion calculation based on the proportion of the main characteristic peak. The core formula is as follows: (1) Percentage of white sugar (based on total syrup): , in =99% is the theoretical purity of the main characteristic peak of white sugar. Substituting the data, we get... .

[0041] (2) Proportion of high fructose syrup: First, based on the proportion of high fructose syrup... =1.32, therefore the fructose content in its DP1 peak is... Then through the formula calculate( The percentage of DP1 fructose peak in solids, determined by... have to (For the theoretical purity of fructose syrup), substituting into the equation yields... .

[0042] (3) Glucose percentage: due to the total percentage of carbohydrates Therefore .

[0043] (4) Moisture content: .

[0044] After the initial calculation, the model further uses "prediction" Compared with actual measurement With the objective of minimizing the difference, the objective function is as follows, obtained through an iterative optimization algorithm:

[0045] in, =98.5% is the theoretical purity of glucose.

[0046] After optimization and correction, the final output raw material ratio is: 30.1% white sugar, 4.2% glucose, 35.4% fructose syrup, and 30.3% moisture. The entire testing process was completed within 40 minutes. After verification by three parallel experiments, the relative standard deviation (RSD) of the results was ≤1.5%, demonstrating good accuracy and repeatability.

[0047] Example 2 This embodiment provides another implementation process for a new method of detecting the proportion of raw materials for tea syrups based on fingerprint spectroscopy. Building upon Embodiment 1, honey and maple syrup are incorporated into a multidimensional characteristic database of standard raw materials. The characteristic information of honey includes: the main characteristic peak is DP1, theoretical purity is 95%, and it contains a unique auxiliary characteristic peak of malic acid-citric acid mixed organic acids (retention time 8.2-8.5 min). The characteristic information of maple syrup includes: the main characteristic peak is DP2, theoretical purity is 97%, and it contains a unique auxiliary characteristic peak of Quebecol phenolic substances (retention time 12.1-12.4 min). The theoretical specific rotation of the two raw materials is also recorded: honey... Maple syrup (white sugar High-fructose corn syrup ).

[0048] The experiment selected a sample of compound tea beverage syrup from a certain brand and analyzed it according to the pretreatment method and HPLC detection conditions in Example 1. Simultaneously, the optical rotation of the sample was measured using a polarimeter. When extracting parameters from HPLC fingerprints, in addition to the usual DP1 peak area ratio... DP,2 peak area ratio In DP1 The concentration was 1.28, and a significant organic acid auxiliary characteristic peak (peak area percentage 0.8%) was detected at 8.3 min. No Quebecol phenolic peak was detected. Based on this, it was determined that the sample contained honey but not maple syrup.

[0049] When the proportioning calculation model is running, honey is first included as an independent raw material unit in the calculation. The core formula is adjusted based on Example 1 as follows: (1) Percentage of white sugar (based on total syrup): (S was measured to be 74.8% by a refractometer) =99%), substituting into the equation, we get .

[0050] (2) Percentage of fructose syrup: ( Substituting into the equation yields... .

[0051] (3) Honey percentage: because honey (Database preset), fructose content in DP1 Furthermore, the organic acid peaks originate solely from honey, therefore , ( The percentage of fructose peaks contributing to honey was determined through iterative optimization, and the final calculation was... .

[0052] (4) Moisture content: .

[0053] Then, the formula for verifying optical rotation is introduced: Substituting the data, we get:

[0054] The deviation from the measured value was only 0.2°, within the allowable error range of ±2%, so no recalculation was required. The final output ingredient ratio was: 25.6% white sugar, 15.2% honey, 29.8% fructose syrup, and 29.4% water, successfully achieving precise analysis of complex formulas containing natural syrups.

[0055] Experiment Example 1: Verification Experiment of the Accuracy of Basic Sugar Source Ratio To verify the accuracy of this method in detecting the basic sugar source ratios, three standard samples with known accurate ratios were prepared for testing. All samples were analyzed under the chromatographic conditions described in Example 1, with each sample measured in triplicate, and the average value was taken as the final result. The sample design covered the typical ratio ranges of granulated sugar, glucose, and high-fructose corn syrup to simulate common syrup formulations in actual production. The experimental results are shown in the table below. The data show that the relative error of the detection of all raw materials was controlled within 1.5%, indicating that the detection of the ratios of these three basic sugar sources—granulated sugar, glucose, and high-fructose corn syrup—has excellent accuracy and repeatability, and can meet the quality control requirements of conventional syrup products. The experimental results are shown in Table 1 below: Table 1. Results of verification of the accuracy of the basic sugar source ratio

[0056] Experiment Example 2: Verification Experiment of the Distinguishing Ability of Composite Materials To verify the ability of this method to distinguish between raw materials with similar characteristics, a complex formulation sample containing honey and maple syrup was specifically designed. These natural syrups have similar main characteristic peaks to the basic sugar source, but contain unique auxiliary characteristic peaks. The method of Example 2 was used for detection, focusing on the system's ability to identify the characteristic peaks of organic acids in honey and phenolic substances in maple syrup. The experimental results are shown in Table 2 below. The results demonstrate that this method can accurately identify and quantify the amount of honey and maple syrup added, and maintains high detection accuracy even when coexisting with fructose syrup and white sugar, overcoming the limitation of traditional methods that cannot distinguish between different sugar sources.

[0057] Table 2. Results of the verification of the distinguishing ability of composite raw materials

[0058] Experiment Example 3: Method Anti-interference Test and Actual Sample Detection Experiment To comprehensively verify the reliability of this method in practical applications, a systematic anti-interference test was conducted, and various commercially available syrup samples were collected for testing. Six spiked samples were designed, with common tea additives (such as citric acid, vitamin C, and tea polyphenols) artificially added to standard samples with known proportions to simulate the complex matrix of real samples. Simultaneously, optical rotation verification parameters were used to correct the results and examine the method's anti-interference ability in complex matrices. As shown in Table 3 below, even in the presence of interfering substances, this method maintains high detection accuracy, with the detection error of each component controlled within 2.0%, demonstrating the strong adaptability and reliability of this method in the detection of real complex samples.

[0059] Table 3. Anti-interference performance and test results of actual samples

Claims

1. A method for detecting the proportioning of new tea drink syrup raw materials based on fingerprint, characterized in that, The method comprises the following steps: (1) constructing a standard raw material multi-dimensional feature database, the database storing HPLC feature information of white granulated sugar, glucose and high fructose corn syrup; (2) detecting a to-be-tested syrup sample by HPLC to obtain an HPLC fingerprint spectrum thereof; (3) extracting multi-dimensional detection parameters of the to-be-tested syrup sample based on the HPLC fingerprint spectrum, the multi-dimensional detection parameters comprising a main feature peak area proportion, at least one trace feature peak information and at least one feature peak ratio; (4) inputting the multi-dimensional detection parameters into a pre-established proportioning calculation model; (5) outputting, by the proportioning calculation model, at least a raw material proportioning of white granulated sugar, glucose and high fructose corn syrup in the to-be-tested syrup sample.

2. The method of claim 1, wherein, The construction of the standard raw material multi-dimensional feature database comprises: recording a main feature peak and a theoretical purity of each standard raw material, wherein the main feature peak is a DP1 peak or a DP2 peak; recording at least one auxiliary feature peak of each standard raw material, the auxiliary feature peak being a trace component peak or a specific polymerization degree peak that is distinguished from other raw materials; recording at least one feature ratio of a standard raw material, the feature ratio being a peak height ratio or a peak area ratio of two different feature peaks in an HPLC spectrum thereof.

3. The method of claim 2, wherein, The standard raw material includes fructose-glucose syrup, and the multi-dimensional characteristic information of the fructose-glucose syrup includes: a main characteristic peak is a DP1 peak, a theoretical purity is greater than or equal to 95%, and a characteristic ratio is a ratio of a fructose peak to a glucose peak in the DP1 peak The ratio ranges from 1.2 to 1.

4.

4. The method of claim 2, wherein, The standard raw materials further comprise honey and / or maple syrup, wherein: the multi-dimensional feature information of the honey comprises a main feature peak of a DP1 peak and an auxiliary feature peak of an organic acid or an amino acid that is distinguished from high fructose corn syrup; the multi-dimensional feature information of the maple syrup comprises a main feature peak of a DP2 peak and an auxiliary feature peak of a characteristic phenolic substance that is distinguished from white granulated sugar.

5. The method of claim 1, wherein, The establishment method of the proportioning calculation model comprises: performing initial proportioning calculation based on a main feature peak area proportion of each raw material; introducing a constraint condition based on the feature ratio to verify and correct the initial proportioning calculation result; minimizing a difference between a feature ratio predicted by the model and a feature ratio actually measured from an HPLC spectrum of the to-be-tested syrup by an iterative optimization algorithm, thereby outputting an optimized raw material proportioning.

6. The method of claim 5, wherein, The method further comprises introducing a verification parameter from an independent detection instrument, the verification parameter comprising a sample optical rotation degree measured by an optical rotation instrument; the proportioning calculation model is further configured to calculate a predicted optical rotation degree according to the output raw material proportioning and a theoretical specific optical rotation degree of each raw material, and compare the predicted optical rotation degree with an actually measured optical rotation degree; when a deviation between the predicted value and the actually measured value exceeds a preset threshold, triggering a re-calculation of the model or outputting a warning information.

7. The method of claim 1, wherein, The method further comprises a residual analysis step: after calculating the proportioning of all known raw materials, analyzing residual peak signals in the HPLC fingerprint spectrum that are not explained by the feature peaks of the known raw materials; if the residual peak signals exceed a preset threshold, determining that there is an undeclared sugar source or additive, and comparing the residual peak signals with the standard raw material multi-dimensional feature database to assist in identification.

8. A system for implementing the method for detecting the proportion of new tea drink syrup raw materials based on fingerprint spectrum according to any one of claims 1-7, characterized in that, The method comprises: an HPLC instrument for obtaining HPLC fingerprint spectra of standard raw materials and to-be-tested syrup samples; a data processing unit configured to: a. construct and store the standard raw material multi-dimensional feature database; b. extracting the multi-dimensional detection parameters from the HPLC fingerprint of the to-be-tested syrup; c. running the proportioning calculation model to output the raw material proportioning result.

9. The system of claim 8, wherein, The system further comprises a polarimeter in communication connection with the data processing unit, configured to provide the verification parameters to the proportioning calculation model.

10. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any one of claims 1-7.

Images