Honey nutritional ingredient detection method and system

Abstract

The invention relates to the technical field of component detection and analysis, in particular to a honey nutrient component detection method and system. The method comprises the following steps: performing honey matrix characteristic analysis of disturbance state response based on honey matrix state data to generate honey matrix characteristic data; based on the honey matrix characteristic data, performing microenvironment excitation induced nutrition medium response characteristic spectrum analysis to generate honey nutrition medium induced response characteristic spectrum data; performing honey nutrition medium time sequence level response characteristic analysis on the honey nutrition medium induction response characteristic spectrum data to generate honey nutrition medium time sequence level response characteristic data; performing honey nutritional ingredient characterization characteristic analysis based on the honey nutritional medium time sequence hierarchy response characteristic data to generate honey nutritional ingredient characterization characteristic data; and performing honey nutrient content intelligent detection on the honey nutrient characterization characteristic data to generate honey nutrient content detection data. According to the invention, efficient detection of honey nutrients is realized.

Description

Technical Field

[0001] This invention relates to the field of component detection and analysis technology, and in particular to a method and system for detecting the nutritional components of honey. Background Technology

[0002] Honey, as a natural nutritional food, contains various sugars, amino acids, vitamins, minerals, and bioactive substances. Accurate detection of honey's nutritional components is a crucial technical foundation for honey quality evaluation, authenticity identification, nutritional value assessment, and quality supervision. However, existing methods for detecting honey's nutritional components mostly employ physicochemical analysis or instrumental detection, typically targeting single or a few nutrients for qualitative or quantitative analysis. While these methods can achieve a certain level of accuracy under ideal conditions, the results are easily affected by the complexity of the honey matrix when dealing with honey samples from different sources and with varying matrix characteristics. Furthermore, the matrix state parameters of honey, such as its internal structure, rheological properties, density distribution, and optical properties, vary significantly between different samples. These matrix differences can interfere with the signal response of nutrients, making the correspondence between the detected signal and the actual nutrient content unstable. Simultaneously, the lack of systematic analysis of the response behavior of nutrient media under changing external conditions makes it difficult to reveal the dynamic response characteristics of different nutrients during microenvironmental changes. This makes it difficult to effectively distinguish and accurately characterize nutrient media with complex compositions and similar response characteristics, hindering stable and reliable nutrient content detection under complex honey sample conditions. Summary of the Invention

[0003] Based on this, the present invention provides a method and system for detecting the nutritional components of honey, in order to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for detecting the nutritional components of honey includes the following steps: Step S1: Obtain the honey sample group to be tested; perform honey matrix state data analysis on the honey sample group to be tested to generate honey matrix state data; perform honey matrix characteristic analysis on perturbation state response based on honey matrix state data to generate honey matrix characteristic data; Step S2: Based on the honey matrix characteristic data, perform microenvironment-induced nutrient medium response characteristic spectrum analysis to generate honey nutrient medium-induced response characteristic spectrum data; Step S3: Perform time-series hierarchical response characteristic analysis on the honey nutrient medium induced response characteristic spectrum data to generate honey nutrient medium time-series hierarchical response characteristic data; Step S4: Based on the time-series hierarchical response characteristic data of honey nutrient medium, perform honey nutrient component characterization characteristic analysis to generate honey nutrient component characterization characteristic data; Step S5: Perform intelligent detection processing on the honey nutrient content of the honey nutrient characterization data to generate honey nutrient content detection data.

[0005] Furthermore, step S1 includes the following steps: Step S11: Obtain the honey sample group to be tested; Step S12: Collect original honey sample group characterization data to generate original honey sample group characterization data. Step S13: Extract the original honey structural state characteristic data based on the original honey sample group characterization data, and analyze the honey matrix state data through the original honey structural state characteristic data; Step S14: Perform temperature-controlled dilution on the honey sample group to be tested to generate temperature-controlled diluted honey samples, and perform temperature-controlled diluted honey state response analysis on the honey matrix state data based on the temperature-controlled diluted honey samples to generate temperature-controlled diluted honey state response data. Step S15: Perform mechanical heterogeneity perturbation on the temperature-controlled diluted honey sample to generate mechanical honey sample, analyze the honey rheological fluctuation characteristics of the mechanical honey sample, and perform mechanical perturbation honey state response analysis on the honey matrix state data based on the honey rheological fluctuation characteristics data to generate mechanical perturbation honey state response data. Step S16: Based on the state response data of temperature-controlled diluted honey and the state response data of mechanically disturbed honey, perform a state response characteristic analysis of honey matrix under external disturbance, and generate state response characteristic data of disturbed honey matrix. Step S17: Perform multi-scale response feature decoupling processing on the disturbed honey matrix state response feature data to generate honey matrix state response feature decoupled data; Step S18: Perform honey matrix feature analysis by decoupling the honey matrix state response feature data to generate honey matrix feature data.

[0006] Furthermore, the original honey structural state characteristic data mentioned in step S13 includes honey initial viscosity data, honey density distribution data, honey optical transmission characteristics data, and honey flow behavior data.

[0007] Furthermore, step S2 includes the following steps: Step S21: Analyze the response window type of honey nutrient media based on the honey matrix characteristic data to generate honey nutrient media response window type data; Step S22: Perform microenvironment-induced stimulation nutrient medium response window type analysis on the honey nutrient medium response window type data to generate microenvironment-induced stimulation nutrient medium response window data; Step S23: Analyze the nutrient medium-induced response signals based on the microenvironment-induced nutrient medium response window data to generate nutrient medium-induced response signal data; Step S24: Perform nutrient medium-induced response sensitivity analysis based on the nutrient medium-induced response signal data to generate nutrient medium-induced response sensitivity data; Step S25: Analyze the nutrient medium-type-specific response behavior of the nutrient medium-induced response sensitivity data to generate nutrient medium-induced specific response behavior data; Step S26: Analyze the nutrient-mediated induced response characteristic spectrum of honey using the nutrient-mediated induced specific response behavior data to generate nutrient-mediated induced response characteristic spectrum data of honey.

[0008] Furthermore, step S23 includes the following steps: Based on the microenvironment-induced response window data of honey nutrient media, a nutrient media spectrum response analysis was performed to generate microenvironment-induced nutrient media spectrum response data. Then, the response signals of each nutrient media induced by the microenvironment were extracted from the microenvironment-induced nutrient media spectrum response data to generate nutrient media-induced response signal data.

[0009] Furthermore, step S3 includes the following steps: Step S31: Perform time-series response gradient analysis on the nutrient medium-induced response characteristic spectrum data of honey to generate time-series response gradient data of nutrient medium type; Step S32: Perform nonlinear time-domain decomposition processing on the nutrient medium type induced response gradient data to generate nutrient medium type induced response time-domain decomposition data; Step S33: Perform response stage segmentation processing on the time-domain decomposition data of nutrient medium type induced response to generate nutrient medium type induced response stage data; Step S34: Perform time-series hierarchical response characteristic analysis of honey nutrient media based on the nutrient media type-induced response stage data, and generate time-series hierarchical response characteristic data of honey nutrient media.

[0010] Furthermore, step S4 includes the following steps: Step S41: Extract stability response signals based on the time-series hierarchical response characteristic data of honey nutrient media, generate time-series stability response signal data of nutrient media, and mark the time-series stability response signal data of nutrient media as intrinsic characteristic signal data of nutrient media. Step S42: Perform cross-level correlation analysis on the intrinsic characteristic signal data of the nutrient medium to generate hierarchical correlation data of the intrinsic characteristic signal of the nutrient medium; Step S43: Perform dynamic feature analysis on the intrinsic characteristic signal data of the nutrient medium to generate dynamic feature data of the intrinsic characteristic signal of the nutrient medium; Step S44: Analyze the characterization features of honey nutrients by using the hierarchical correlation data of the intrinsic characteristic signals of the nutrient medium and the dynamic characteristic data of the intrinsic characteristic signals of the nutrient medium, and generate characterization feature data of honey nutrients.

[0011] Furthermore, step S5 includes the following steps: Step S51: Extract the response signal relationship data of honey nutritional components based on the honey nutritional component characterization characteristic data; Step S52: Based on the relationship data of honey nutrient component characterization response signals, establish the multivariate nonlinear response function of honey nutrient component characterization signals and content to obtain a preliminary honey nutrient component content mapping model; Step S53: Analyze the nutritional component matrix correction factors based on the honey matrix characteristic data to generate nutritional component matrix correction factors. Step S54: Obtain a priori characterization and content assessment data of honey nutritional components; Step S55: Use the nutrient matrix correction factor to correct the nutrient content offset parameter of the preliminary honey nutrient content mapping model due to differences in honey matrix, generate a corrected honey nutrient content mapping model, and use the honey nutrient prior characterization-content assessment data to train the model parameters of the corrected honey nutrient content mapping model, generate a honey nutrient content mapping model. Step S56: Transmit the honey nutrient characterization feature data to the honey nutrient content mapping model for intelligent detection and processing of honey nutrient content, and generate honey nutrient content detection data.

[0012] Furthermore, the honey nutrient component characterization response signal relationship data in step S51 includes nutrient component characterization response signal amplitude relationship data, nutrient component characterization response signal delay relationship data, and nutrient component characterization response signal stability time relationship data.

[0013] This specification provides a honey nutrient composition detection system for performing the honey nutrient composition detection method as described above. The honey nutrient composition detection system includes: The honey matrix feature analysis module is used to acquire the honey sample group to be tested; to analyze the honey matrix state data of the honey sample group to be tested and generate honey matrix state data; and to perform honey matrix feature analysis on the perturbation state response based on the honey matrix state data and generate honey matrix feature data. The honey nutrient medium induced response analysis module is used to perform microenvironment-induced nutrient medium response characteristic spectrum analysis based on honey matrix characteristic data, and generate honey nutrient medium induced response characteristic spectrum data. The time-series hierarchical response feature analysis module is used to perform time-series hierarchical response feature analysis on the induced response feature spectrum data of honey nutrient media, and generate time-series hierarchical response feature data of honey nutrient media; The honey nutrient component characterization feature analysis module is used to perform honey nutrient component characterization feature analysis based on honey nutrient medium time-series hierarchical response feature data, and generate honey nutrient component characterization feature data. The honey nutrient content detection module is used to intelligently detect and process the nutrient content of honey based on the characteristic data of honey nutrient composition, and generate honey nutrient content detection data.

[0014] The beneficial effects of this application are as follows: This invention analyzes the matrix state data of honey samples and introduces external perturbation conditions such as temperature-controlled dilution and mechanical heterogeneity disturbance to systematically analyze the response behavior of the honey matrix state under different perturbations, thereby constructing honey matrix characteristic data. By collecting original honey structural state characteristic data such as initial viscosity, density distribution, optical transmission characteristics, and flow behavior, and combining multi-scale response characteristic decoupling processing, it can comprehensively characterize the matrix state differences of honey as a high-viscosity, multiphase natural system, effectively separating interference responses unrelated to nutrients in the matrix state, making the honey matrix characteristics more stable and comparable, thus providing a unified and reliable matrix reference basis for subsequent nutrient response analysis, and significantly reducing the impact of matrix differences between different honey samples on the detection results. By constructing a honey nutrient medium response characteristic spectrum induced by microenvironment stimulation, the response window types of different nutrient media are analyzed, and differentiated response signals are generated by stimulating nutrient media under specific microenvironment induction conditions, thereby systematically analyzing the induced response signals, response sensitivity, and induced specific response behavior of each nutrient medium. This method effectively amplifies the response differences of different nutrient media during microenvironmental changes, overcoming the limitations of existing technologies that rely solely on single or static detection signals and struggle to distinguish nutrients with similar response characteristics. The resulting honey nutrient media-induced response characteristic spectrum simultaneously includes response amplitude, response sensitivity, and specific behavioral information, thereby improving the identifiability and characterization accuracy of honey nutrients in complex systems. Time-series hierarchical response characteristic analysis is performed on the honey nutrient media-induced response characteristic spectrum data. Through time-series response gradient analysis of nutrient media types, nonlinear time-domain decomposition, and response stage division, a systematic analysis of the dynamic behavior of each nutrient media at different time scales and response stages is achieved. This reveals the nonlinear variation laws and multi-stage hierarchical characteristics of nutrient media responses, clearly distinguishing the response signals of different nutrient media under microenvironmental induction. This enhances the ability to identify the time-series characteristics of complex nutrient systems and provides a reliable time-series data foundation for subsequent stability analysis and characterization feature extraction, significantly improving the accuracy and resolvability of dynamic characterization of honey nutrient media. Stability response signals were extracted from time-series hierarchical response characteristic data of honey's nutrient media. Cross-hierarchical correlation analysis and dynamic feature analysis of intrinsic characteristic signals were then performed to fully extract the stability signals and dynamic feature information inherently related to nutrient components in the nutrient media response. This effectively eliminated the influence of environmental interference and instantaneous fluctuations, achieving a comprehensive characterization of the intrinsic properties of various nutrients in honey. Through systematic analysis of cross-hierarchical correlations and dynamic response characteristics, the reliability, robustness, and information completeness of honey nutrient characterization were significantly improved, providing a scientific and modelable feature foundation for subsequent intelligent content detection.This research employs intelligent processing to analyze the nutritional component characterization data of honey, extracting the amplitude, delay, and stability time-relationship data of the nutrient component characterization response signals, and establishing a preliminary content mapping model based on a multivariate nonlinear response function. Subsequently, a nutrient component matrix correction factor is generated by combining honey matrix characteristic analysis to correct for matrix differences in the model. The model is then trained using prior characterization-content data to form a corrected content mapping model adaptable to different honey samples. This achieves accurate mapping of honey nutrient component characterization features to quantitative content, effectively reducing the interference of sample matrix differences on detection results, and improving the accuracy, stability, and generalization ability of honey nutrient component detection. This enables intelligent, high-precision, and repeatable nutrient component content detection, meeting the application needs of complex honey samples.

[0015] Therefore, the honey nutrient component detection method of this invention analyzes the honey matrix state of the honey sample group to be tested, and extracts the honey matrix state response characteristics under external perturbation conditions such as temperature-controlled dilution and mechanical disturbance. This achieves effective characterization of the complexity and sample variability of the honey matrix, and can quantitatively describe the matrix state of different honey samples before detection, thereby reducing the interference of honey matrix differences on subsequent nutrient component detection results and improving the consistency and stability of detection results in honey samples from different sources. A microenvironment-induced nutrient medium response analysis mechanism is introduced. By constructing a honey nutrient medium-induced response characteristic spectrum, the response behavior of different nutrient media under specific microenvironment conditions is systematically analyzed, effectively amplifying the response differences between different nutrients. This overcomes the problem of existing technologies relying solely on static detection signals and having difficulty distinguishing nutrient media with similar response characteristics, improving the identification ability and characterization accuracy of complex honey nutrient component systems. This study analyzes the temporal hierarchical response characteristics of honey induced by nutrient media, decomposing and hierarchically classifying the response processes of different nutrient media. This reveals the nonlinear variation characteristics of the response behavior of nutrient media from a temporal perspective, enabling the characterization of nutrients to move beyond single response points or instantaneous signals and instead rely on comprehensive analysis of stable response characteristics across multiple stages and levels. This significantly improves the reliability and anti-interference ability of nutrient characterization. Based on this, a honey nutrient content mapping model is constructed, and a nutrient matrix correction factor based on honey matrix characteristics is introduced to correct for matrix differences between different honey samples. This allows the established content mapping model to adapt to honey samples with different matrix conditions, enhancing its generalization ability and applicability in practical detection applications. This approach enables intelligent and highly accurate detection of honey nutrients under complex honey sample conditions, balancing detection precision, stability, and adaptability, and is suitable for the nutrient content detection needs of honey samples from multiple sources and of multiple types. Attached Figure Description

[0016] Figure 1This is a schematic diagram of the steps in the method for detecting the nutritional components of honey according to the present invention; Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S5. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0018] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. Functional entities may be implemented in software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods. The term "and / or" as used herein includes any and all combinations of one or more of the associated items listed.

[0019] To achieve the above objectives, please refer to Figures 1 to 2 This invention provides a method and system for detecting the nutritional components of honey. In the embodiments of this invention, please refer to... Figure 1 The diagram shown is a flowchart illustrating the steps of a method for detecting the nutritional components of honey according to the present invention. The method includes the following steps: Step S1: Obtain the honey sample group to be tested; perform honey matrix state data analysis on the honey sample group to be tested to generate honey matrix state data; perform honey matrix characteristic analysis on perturbation state response based on honey matrix state data to generate honey matrix characteristic data; In this embodiment of the invention, a systematic analysis of the honey matrix state of the honey sample group is performed to obtain complete honey matrix characteristic data. First, each sample group is homogenized. A temperature-controlled dilution technique is used to dilute the honey sample to a specific viscosity range. For example, honey with an initial viscosity between 2000 mPa·s and 4500 mPa·s is diluted to approximately 1000 mPa·s by heating to 40°C and slowly mixing, ensuring the repeatability of the measurement without damaging the internal structure. Subsequently, multi-parameter matrix state data are acquired, including viscosity, density distribution, optical transmittance, rheological properties, shear response behavior, and microparticle distribution. Viscosity is measured using a rotational viscometer to obtain shear stress-shear rate curves at different shear rates. Density distribution is determined using the density gradient column separation method. Optical transmittance is measured using a spectrophotometer to determine the transmitted light intensity as a function of wavelength. Rheological properties are obtained by frequency scanning to obtain storage modulus and loss modulus. Microparticle distribution is analyzed using microscopic imaging and a particle size analyzer to statistically determine the honey particle size and distribution characteristics. The acquired honey matrix state data is then subjected to perturbation response analysis. Honey samples were subjected to both mechanical and temperature-controlled perturbations. Mechanical perturbation applied periodic micro-shear forces using a precision vibration table to simulate the rheological fluctuations of honey under natural transport or stirring conditions, monitoring viscosity changes, shear rate response, and nonlinear hysteresis effects. Temperature-controlled perturbation involved heating samples to different temperature gradients (e.g., 25℃, 35℃, 45℃) and maintaining these temperatures for a certain time, recording viscosity versus temperature curves, structural recovery time, and fluidity changes. Multi-scale decoupling analysis of the mechanical and temperature-controlled perturbation response data was performed to distinguish honey matrix characteristics at the rapid response, intermediate response, and long-term response levels, forming a honey matrix characteristic data matrix. Rows in the matrix represent sample numbers, and columns represent amplitude, delay, and stability time characteristics under different perturbation parameters.

[0020] Step S2: Based on the honey matrix characteristic data, perform microenvironment-induced nutrient medium response characteristic spectrum analysis to generate honey nutrient medium-induced response characteristic spectrum data; In this embodiment of the invention, the induced response characteristic spectrum of honey nutrient media under microenvironment excitation is analyzed based on honey matrix characteristic data. Each honey sample is exposed to controlled microenvironment conditions, including temperature gradient, humidity change, light intensity, and ion concentration change. For example, the sample is placed under constant temperature of 37°C, relative humidity of 50%, and light intensity of 5000 lux, with a specific ion concentration gradient added to induce responses from glucose, fructose, sucrose, and amino acid nutrient media in the honey. The amplitude changes, response delays, and duration of stability of the nutrient media under microenvironment excitation are recorded using a high-sensitivity spectrometer and rheological sensor. Subsequently, characteristic spectrum analysis is performed on the response data. Characteristic spectrum analysis includes nutrient media response window division, time-series peak analysis, nonlinear coupling analysis, and microenvironment-specific response extraction. The response window is divided according to the amplitude change threshold and peak duration; for example, the interval where the amplitude reaches 80% of the total peak value is defined as the main response window, and a duration exceeding 5 seconds is considered a stable response segment. Nonlinear coupling analysis determines the interactive response characteristics of each nutrient media by comparing the amplitude and delay differences between glucose and fructose within the same window. By classifying response signals according to nutrient medium type and labeling them according to microenvironmental stimulation conditions, a honey nutrient medium-induced response characteristic spectrum data matrix is ​​generated. The rows of the matrix represent sample numbers, and the columns represent different nutrient media and their characteristic parameters, including amplitude, delay, and stability time.

[0021] Step S3: Perform time-series hierarchical response characteristic analysis on the honey nutrient medium induced response characteristic spectrum data to generate honey nutrient medium time-series hierarchical response characteristic data; In this embodiment of the invention, time-series hierarchical response characteristic analysis is performed on the nutrient-mediated induced response characteristic spectrum data of honey. The characteristic spectrum data is divided into rapid response level, intermediate response level, and long-term response level, and gradient analysis is performed on the nutrient media at each level. Gradient analysis obtains the dynamic response characteristics of each nutrient media at different time scales by calculating the rate of change of amplitude over time, peak delay gradient, and stability time gradient. For example, the rate of change of glucose amplitude in the rapid level can reach 0.8 mV / s, the delay gradient is 0.15 s / s, and the stability time change rate is 0.02 s / s; the rate of change of fructose amplitude amplitude in the intermediate level is 0.3 mV / s, the delay gradient is 0.08 s / s, and the stability time change rate of sucrose in the long-term level is 0.01 s / s. Subsequently, the nonlinear response is decomposed in the time domain, and the response signal of each nutrient media is decomposed into different frequency components and multi-level contribution terms. The rapid response, slow adaptation, and long-term stable stages are identified through the decomposition results. Based on the principle of response stage division, the signal is divided into three stages: initial excitation, peak hold, and decay recovery. The average amplitude, peak duration, and delay of each stage are calculated. By combining the staged features with the hierarchical response gradient, a time-series hierarchical response feature data matrix of honey nutrient media is formed, where rows represent sample numbers and columns represent different nutrient media and response features at each level.

[0022] Step S4: Based on the time-series hierarchical response characteristic data of honey nutrient medium, perform honey nutrient component characterization characteristic analysis to generate honey nutrient component characterization characteristic data; In this embodiment of the invention, nutrient component characterization feature analysis is performed based on the time-series hierarchical response characteristic data of honey nutrient media. The stability response signal of each nutrient medium is extracted as an intrinsic characteristic signal, and the stable signals of rapid response, intermediate response, and long-term response are represented hierarchically to generate multi-level feature vectors. For example, the stability signal of glucose at the rapid level is 35 seconds, at the intermediate level is 8 minutes, and at the long-term level is 28 minutes. Subsequently, cross-level correlation analysis is performed on the intrinsic characteristic signals to calculate the amplitude ratio, delay difference, and stability time ratio between different levels, forming a hierarchical correlation matrix. For example, the amplitude ratio between the rapid level and the intermediate level is 1.12, the delay difference is 2 seconds, and the stability time ratio is 0.90. Furthermore, through dynamic feature analysis, the nonlinearity index, amplitude volatility, and delay change rate of the nutrient medium during the response process are extracted to characterize the dynamic behavior of the nutrient components. For example, the nonlinearity index of glucose is 0.38, the amplitude volatility is 5%, and the delay change rate is 0.02 s / s. By combining the hierarchical correlation matrix and dynamic feature data, a honey nutritional component characterization feature data matrix is ​​generated. The rows represent sample numbers, and the columns represent each nutrient component and its characterization feature parameters, including stability signals, hierarchical correlation coefficients, and dynamic behavior indicators.

[0023] Step S5: Perform intelligent detection processing on the honey nutrient content of the honey nutrient characterization data to generate honey nutrient content detection data.

[0024] In this embodiment of the invention, intelligent detection and processing are performed on the characteristic data of honey nutritional components. A multivariate nonlinear mapping relationship is established based on the characteristic data, using features such as amplitude, delay, and stability time as inputs, with the response characteristics of each nutrient as the independent variable and the actual nutrient content as the dependent variable. A preliminary mapping model is generated through nonlinear function fitting. For example, the relationship between glucose content and amplitude, delay, and stability time is modeled using a quadratic polynomial and exponential coupling terms; similar nonlinear coupling models are used for fructose and sucrose. Subsequently, the preliminary mapping model is corrected using a matrix correction factor to adjust the amplitude, delay, and stability time parameters, ensuring that the model output reflects the actual content under different honey matrix conditions. After correction, prior characterization-content assessment data is used for parameter training to optimize the fitting coefficients and interaction terms, minimizing the sum of squared residuals, achieving a fitting accuracy R² ≥ 0.95. The final output is a matrix of honey nutrient content detection data. Rows represent sample numbers, columns represent nutrient types, and each cell records the content value, fitting residual, and stability coefficient. For example, glucose content is 42.3g / 100g, fructose is 38.5g / 100g, sucrose is 5.7g / 100g, and the stability coefficient is ≥0.90. This detection method achieves high-precision, repeatable, and intelligent detection of nutrient content from multidimensional features, ensuring reliable results under different honey sample and matrix conditions.

[0025] Furthermore, step S1 includes the following steps: Step S11: Obtain the honey sample group to be tested; In this embodiment of the invention, a honey sample group is obtained to ensure its representativeness and comparability. The honey sample group includes honey samples from different nectar-producing plants, different geographical regions, and different collection times. Each sample is collected in a highly airtight glass bottle, which is dried and cleaned at high temperature before collection to eliminate external contamination. The honey sample group is immediately sealed and stored after collection under low-temperature, light-protected conditions to inhibit enzyme activity and microbial growth, preventing nutrient degradation or alteration of chemical structure. Each sample is collected in a quantity of at least 50 grams to ensure sufficient sampling volume for subsequent analysis and to meet the requirements for repeated experiments. A unified coding system is used to identify the samples. Each code includes nectar source information, collection area, collection date, and batch number to facilitate the correlation analysis between the sample matrix characteristics and subsequent nutrient component detection results. After obtaining the samples, environmental parameters, including temperature, humidity, and light conditions, should be recorded immediately to exclude interference from environmental factors on the honey matrix state and nutrient response in subsequent analysis. The entire collection process followed standard operating procedures to ensure the integrity, stability, and traceability of the samples, enabling the obtained honey sample population to fully reflect the diversity and complexity of honey under natural conditions and providing a reliable basis for subsequent matrix state analysis.

[0026] Step S12: Collect original honey sample group characterization data to generate original honey sample group characterization data. In this embodiment of the invention, a raw honey sample group characterization process is performed on the honey sample group to obtain multi-dimensional characteristic information of the honey samples in their unprocessed state. The characterization process includes comprehensive measurements of the appearance, color, transparency, initial viscosity, density distribution, flow behavior, and optical transmission properties of each honey sample. Appearance characterization involves observing the color and particle uniformity of the honey sample under standard light source conditions, and measuring the transmittance and absorbance curves of the honey using a spectrophotometer to quantify the optical properties of the honey. Viscosity measurement uses a high-precision rotational viscometer to measure the viscosity curves of the honey fluid at different shear rates under isothermal conditions, recording the shear rate-viscosity correspondence to reflect the rheological properties of the honey. Density distribution is measured at multiple points on the sample using a vibrating densitometer to generate a honey density distribution curve as a function of location, reflecting the uniformity of the honey's internal composition. Flow behavior characterization employs rheological analysis methods, performing isothermal shear, step shear, and instantaneous stress tests in a standard isothermal device, recording the flow rate and stress response of the sample under different external forces, and obtaining the honey shear stress-shear rate curve and resilience data. These measurements systematically organized the collected multi-dimensional raw characterization data into a raw characterization dataset that can be used for subsequent matrix state analysis. To ensure data stability and repeatability, the temperature was maintained at 25℃±0.5℃ and the light intensity was controlled at 500 lux during the measurement process. All measuring instruments were calibrated, and three repeated measurements were performed and the average value was taken to eliminate random errors. In addition, each honey sample was taken at least 30 grams to ensure sufficient sample volume for multiple physical property tests. The sample batch number, collection date, and honey source information were recorded to enable data traceability in subsequent analysis.

[0027] Step S13: Extract the original honey structural state characteristic data based on the original honey sample group characterization data, and analyze the honey matrix state data through the original honey structural state characteristic data; In this embodiment of the invention, structural state characteristic data of the original honey are extracted based on the characterization data of the original honey sample group, and the honey matrix state is analyzed through this characteristic data. The structural state characteristic data of the original honey includes initial viscosity data, density distribution data, optical transmission properties data, and flow behavior data. Initial viscosity data was obtained by analyzing the shear rate-viscosity curves measured by a rotational viscometer and fitting them using a power-law model to determine the viscosity coefficient k and the non-Newtonian exponent n, thus reflecting the influence of the internal molecular arrangement and sugar content of honey on its rheological properties. Density distribution data was obtained by high-precision density measurements of different volume segments (5 mL each) of the sample, calculating the average and standard deviation of each segment's density to generate a density uniformity coefficient, used to assess the uniformity of solute distribution within the sample and potential particle sedimentation. Optical transmission characteristics data were obtained by measuring the transmittance and absorbance curves of honey in the 400-700 nm wavelength range using a spectrophotometer, further calculating the optical uniformity index to reveal the optical uniformity of the sample at different wavelengths and the possible impurity content. Flow behavior data were obtained by rheological analysis of the flow rate and stress recovery ability of honey under constant shear, step shear, and stress release conditions, and the nonlinear rheological response characteristics of honey under different mechanical disturbances were quantified using amplitude fluctuation rate and stress recovery index. The above four types of structural state characteristic data were comprehensively analyzed to form a honey matrix state data matrix. The matrix includes viscoelastic parameters (calculated from storage modulus G' and loss modulus G''), shear sensitivity parameters (represented by the non-Newtonian exponent n and viscosity coefficient k), density uniformity coefficient, optical uniformity index, and mechanical perturbation response parameters (amplitude fluctuation rate and stress recovery index). This matrix can quantify matrix differences between different samples, providing a standard reference for subsequent temperature-controlled perturbation and mechanical perturbation analysis. It can also identify internal structural factors that may affect the nutrient medium response, such as sugar crystallization, particle deposition, and uneven moisture distribution.

[0028] Step S14: Perform temperature-controlled dilution on the honey sample group to be tested to generate temperature-controlled diluted honey samples, and perform temperature-controlled diluted honey state response analysis on the honey matrix state data based on the temperature-controlled diluted honey samples to generate temperature-controlled diluted honey state response data. In this embodiment of the invention, a temperature-controlled dilution process is performed on a group of honey samples to obtain temperature-diluted honey samples. The temperature-controlled dilution state response data of the honey matrix is ​​then analyzed based on the diluted samples. The temperature-controlled dilution process involves heating the honey to a specified temperature range (e.g., 40°C) in a constant-temperature water bath and adding purified water in a fixed ratio for uniform mixing. This appropriately reduces the viscosity of the honey, ensuring the repeatability of rheological measurements while preventing thermal degradation or changes in the chemical structure of the honey components. During the dilution process, continuous stirring is used to ensure thorough mixing of the honey and water, eliminating localized concentration unevenness. After processing, the temperature-diluted honey samples undergo rheological, density, and optical measurements to obtain viscosity curves, density distribution, and transmission characteristics under temperature-controlled dilution conditions. These data are then compared with the original matrix state data. By calculating the viscosity change rate, density uniformity change coefficient, and optical transmission change amplitude, temperature-controlled dilution honey state response data is generated. The response data reflects the matrix adjustability and stability of honey under external temperature control and dilution conditions, providing a reference for further mechanical perturbation experiments. It also reveals the sensitivity of each sample to matrix state under microenvironmental changes, providing a basis for subsequent nutrient medium-induced response analysis.

[0029] Step S15: Perform mechanical heterogeneity perturbation on the temperature-controlled diluted honey sample to generate mechanical honey sample, analyze the honey rheological fluctuation characteristics of the mechanical honey sample, and perform mechanical perturbation honey state response analysis on the honey matrix state data based on the honey rheological fluctuation characteristics data to generate mechanical perturbation honey state response data. In this embodiment of the invention, temperature-controlled diluted honey samples are subjected to mechanical heterogeneous perturbation, and their rheological fluctuation characteristics are analyzed. The perturbation involves applying various mechanical loading methods, such as stepped shear, pulsed stress, and oscillating shear, in a rheometer that controls shear rate and shear time. This subjects the internal structure of the honey to a combined effect of shear force, vibration, and micro-displacement, thereby causing a microscopic rearrangement of intermolecular forces, particle distribution, and water aggregation. During the perturbation process, shear stress, strain, and viscosity change curves are collected in real time to generate rheological fluctuation characteristic data of the mechanically perturbed honey sample, including the nonlinear response of shear rate to viscosity, stress relaxation time, and flow recovery ability. These data are compared with the state response data of temperature-controlled diluted honey to analyze the magnitude and response law of changes in the honey matrix state under external mechanical perturbation, generating mechanically perturbed honey state response data. This data can quantify the dynamic stability of the honey matrix structure under different mechanical environments.

[0030] Step S16: Based on the state response data of temperature-controlled diluted honey and the state response data of mechanically disturbed honey, perform a state response characteristic analysis of honey matrix under external disturbance, and generate state response characteristic data of disturbed honey matrix. In this embodiment of the invention, the state response characteristics of honey matrix under external disturbances are analyzed based on the state response data of temperature-controlled diluted honey and the state response data of honey under mechanical disturbances. The analysis first involves uniformly organizing various physical parameters collected in the temperature-controlled dilution and mechanical disturbance experiments, including viscosity change rate, shear stress response curve, stress relaxation time, density uniformity change coefficient, and optical transmission characteristic change amplitude. Then, by establishing a multidimensional response matrix, various characteristic data of the same honey sample under different disturbance conditions are compared and analyzed to calculate the response amplitude, response delay, and nonlinear variation law of the honey matrix to temperature control and mechanical disturbances. Correlation analysis is performed between the response amplitude and the original matrix characteristics, and matrix operations are used to generate honey matrix state disturbance sensitivity indices, including the sensitivity to viscosity and density changes under different temperature control conditions, and the stability coefficient of the shear stress-strain curve under different mechanical disturbance conditions. Simultaneously, by fitting the trend of optical transmission characteristics with disturbance, an optical uniformity disturbance response curve is generated to quantify the internal structure adjustment of the sample under external disturbances. These indicators collectively form characteristic data of the state response of perturbed honey matrix, which can describe the dynamic stability and nonlinear response characteristics of honey matrix under various types of perturbations, providing a quantifiable and comparable matrix response basis for subsequent multi-scale decoupling analysis. Furthermore, by statistically analyzing the perturbation response data of multiple sample groups, a matrix response distribution model can be established, revealing the response differences of different honey sources and different batches of samples under the same perturbation conditions.

[0031] Step S17: Perform multi-scale response feature decoupling processing on the disturbed honey matrix state response feature data to generate honey matrix state response feature decoupled data; In this embodiment of the invention, multi-scale response feature decoupling processing is performed on the perturbed honey matrix state response characteristic data. The decoupling process first classifies the perturbed response data according to time scale, stress level, and physical property type, constructing a multi-scale feature matrix containing three levels: short-term transient response, medium-term staged change, and long-term stable response. By independently analyzing the response features within each level, the response signals directly caused by changes in the matrix's internal structure and the additional responses caused by external perturbations are separated, generating a feature subset after removing interference. Signal decomposition methods are used on the viscosity, density, and optical transmission response curves to identify the main frequency components, fluctuation amplitudes, and nonlinear trends. The response information at different scales is then normalized to form decoupled features that can be compared between different samples. Furthermore, by constructing a multi-physical quantity response correlation matrix, the coupling relationship between the viscosity response, density response, and optical transmission response is quantified, identifying the independent contributions and interactions of each response feature, thereby obtaining multi-scale independent features of the honey matrix state response. The decoupled feature data can not only quantify the internal structural stability of honey under complex perturbations, but also reveal the basis for the possible differential responses of different nutrient media under the same matrix conditions.

[0032] Step S18: Perform honey matrix feature analysis by decoupling the honey matrix state response feature data to generate honey matrix feature data.

[0033] In this embodiment of the invention, honey matrix characteristic analysis is performed using decoupled data of honey matrix state response characteristics. The matrix characteristic analysis first quantitatively models the response characteristics of multiple physical quantities such as viscosity, density, and optical transmittance after decoupling, calculating the viscoelastic index, shear sensitivity coefficient, density uniformity index, and optical uniformity index of the honey matrix to form a multidimensional matrix feature vector. Subsequently, through feature clustering and statistical analysis, typical patterns and abnormal samples of honey matrix state in the sample group are identified, and a honey matrix feature space is established to quantify the differences between samples from different honey sources, batches, or processing conditions. Further, by combining perturbation response characteristics with the original matrix characteristic data, the response law of the honey matrix to changes in the microenvironment and mechanical perturbations is analyzed, generating a honey matrix response stability evaluation index to describe the matrix's ability to regulate the response to the nutrient medium. Finally, the multi-scale characteristics of various physical quantities, the independent contributions after decoupling, and the stability index are integrated to form a complete honey matrix characteristic data set, including the matrix state parameters and response law parameters of each sample under different perturbation conditions. This honey matrix characteristic data can comprehensively reflect the physical structural state and dynamic stability of honey.

[0034] Furthermore, the original honey structural state characteristic data mentioned in step S13 includes honey initial viscosity data, honey density distribution data, honey optical transmission characteristics data, and honey flow behavior data.

[0035] Furthermore, step S2 includes the following steps: Step S21: Analyze the response window type of honey nutrient media based on the honey matrix characteristic data to generate honey nutrient media response window type data; In this embodiment of the invention, based on honey matrix characteristic data, including multidimensional physical quantities such as viscosity, density distribution, flow behavior, and optical transmission characteristics, a multi-physical quantity coupling matrix is ​​established for each honey sample. By performing multi-level quantification of viscosity gradient, density uniformity index, flow recovery time, and optical transmission rate of change, the response sensitivity intervals of honey samples under different physical environments are defined. These sensitivity intervals are used to delineate the response windows that nutrient media may produce under different matrix conditions. For example, in the high viscosity, low density variation range, the chemical activity or optical signal response of specific sugar or protein nutrient media shows peak changes, while in the low viscosity, high density uniformity range, it shows a low amplitude response. By performing cluster analysis on the multidimensional response data of all samples, response window types with unified characteristics are formed. Each window type corresponds to a certain range of physical characteristic combinations and its preliminary prediction of the response of nutrient media. The analysis process identifies high-variance and low-variance regions by calculating statistical parameters such as the standard deviation, skewness, and kurtosis of viscosity, density, and optical indicators. It also quantifies and records the range of matrix physical parameters, possible response intensity, and time delay for each window type, thereby generating honey nutrient medium response window type data. This data can comprehensively characterize the response potential of nutrient media in different samples under specific matrix conditions, providing a scientific basis for microenvironment-induced excitation and ensuring comparability and stability of subsequent steps in dynamic and nonlinear response analysis.

[0036] Step S22: Perform microenvironment-induced stimulation nutrient medium response window type analysis on the honey nutrient medium response window type data to generate microenvironment-induced stimulation nutrient medium response window data; In this embodiment of the invention, based on the inherent correlation between honey matrix characteristics and nutrient medium response window types, the time intervals and state intervals for effective responses of different nutrient media are clarified through quantitative control and temporal effects of external microenvironment induction conditions. Based on the honey nutrient medium response window type data, the baseline response intervals of various nutrient media in the uninduced state are confirmed. These baseline response intervals include the initial response start point, the stable response amplitude interval, and the response decay start point, used to characterize the inherent response range of the nutrient media under static conditions of the honey matrix. Subsequently, microenvironmental induction and excitation conditions matching the honey matrix characteristic analysis results in step S1 are introduced to apply controlled microenvironmental changes to the honey sample. These microenvironmental induction conditions include at least temperature gradient changes, local concentration gradient changes, and microscale mechanical perturbation intensity changes. The temperature change range is limited to 20℃ to 45℃, and the rate of change is controlled to no more than 1℃ per minute to avoid irreversible changes to the overall honey structure. The mechanical perturbation intensity is limited to the low-shear to medium-shear range to maintain the continuity of the honey matrix. During the microenvironment-induced process, the response behavior of nutrient media under various response window types was continuously monitored. By simultaneously analyzing the rate of change of response amplitude, the offset of response start time, and the extent of response duration extension, the activation effect of microenvironment induction on the original response window was identified. When, under specific induction conditions, the response amplitude of the nutrient media first exceeds the upper limit of the baseline response range, and the duration of this change exceeds a preset stability threshold, the corresponding time interval is defined as the initial segment of the microenvironment-induced activation response window. When the response amplitude enters a relatively stable range and the fluctuation amplitude is lower than the baseline fluctuation range, the corresponding interval is defined as the stable segment of the microenvironment-induced activation response window. When the response amplitude shows a continuous decline and approaches the baseline response range, the corresponding interval is defined as the decay segment of the microenvironment-induced activation response window. Through this method, the response windows formed by each type of nutrient media under different microenvironment-induced conditions are fully characterized. The generated microenvironment-induced response window data for nutrient media is indexed by nutrient media type, constrained by microenvironment induction parameters, and focuses on the start time, stable time interval, decay time interval, and corresponding response amplitude range of the excitation response window. The system reflects the effective response time period and state range of various nutrient media in honey under controlled microenvironment induction. This enhances the distinguishing ability and detection stability of honey nutrient component detection methods under complex honey matrix conditions.

[0037] Step S23: Analyze the nutrient medium-induced response signals based on the microenvironment-induced nutrient medium response window data to generate nutrient medium-induced response signal data; In this embodiment of the invention, response signals induced by various nutrient media are analyzed based on microenvironment-induced nutrient media response window data. The signals of various nutrient media under microenvironment-induced conditions are quantified in the time domain and amplitude, including response onset time, peak amplitude, stable amplitude, response delay, and signal recovery speed. By comparing the response signals of the same nutrient media under different microenvironmental conditions, the response sensitivity, amplitude variation range, and time-series characteristics of each nutrient media are calculated. Simultaneously, the response signals are coupled with matrix characteristic data to identify the degree to which the signals are affected by matrix characteristics. Using multivariate nonlinear analysis, the response signals of each nutrient media are decomposed to distinguish between specific response signals induced by the microenvironment and background noise or random fluctuation signals, with specific responses being the primary analysis object. The amplitude variation, delay characteristics, and stability time of the signals are statistically summarized to form a multidimensional nutrient media-induced response signal matrix, including peak amplitude data, delay time data, and stability time data. These data comprehensively reflect the dynamic response characteristics of each nutrient media under different microenvironmental conditions, providing an accurate signal basis for subsequent response sensitivity analysis and induced specific behavior analysis, enabling the quantitative characterization and comparison of the behavior of honey nutrients in complex microenvironments.

[0038] Step S24: Perform nutrient medium-induced response sensitivity analysis based on the nutrient medium-induced response signal data to generate nutrient medium-induced response sensitivity data; In this embodiment of the invention, nutrient medium-induced response sensitivity analysis is performed based on nutrient medium-induced response signal data. The amplitude, response delay, and stability time of each nutrient medium under different microenvironmental conditions are used as analytical parameters. By calculating the amplitude change rate, time response gradient, and response stability coefficient, the sensitivity of each nutrient medium to microenvironmental disturbances is quantified. Further analysis of response differences under different microenvironmental conditions such as temperature, pH, ionic strength, and humidity is conducted to form nutrient medium response sensitivity curves and matrices. By comparing the slope, peak amplitude, and duration of the response sensitivity curves, the nutrient medium most sensitive to microenvironmental changes and its characteristic response range are identified. Combined with matrix characteristic data, response sensitivity correction coefficients for each nutrient medium under different matrix states are calculated, standardizing the sensitivity indicators and making the sensitivity of nutrient media from different samples and nectar sources comparable. The generated nutrient medium-induced response sensitivity data matrix contains multidimensional features of the response amplitude, response delay, and stability time of each nutrient medium under microenvironmental induction, providing a complete quantitative basis for subsequent analysis of induced specific response behavior and ensuring that the response characteristics of different nutrient media can be systematically and standardizedly described and compared.

[0039] Step S25: Analyze the nutrient medium-type-specific response behavior of the nutrient medium-induced response sensitivity data to generate nutrient medium-induced specific response behavior data; In this embodiment of the invention, the sensitivity data of nutrient medium-induced responses are categorized and organized according to the type of nutrient medium. The nutrient medium types include at least monosaccharide nutrient media, disaccharide nutrient media, polysaccharide nutrient media, amino acid nutrient media, and nutrient media containing trace active components. For each type of nutrient medium, the rate of change of response amplitude, the amount of change of response delay, the proportion of change of stability time, and the nonlinear response index are statistically analyzed under different microenvironmental induction conditions, thereby constructing a multidimensional data structure of nutrient medium type—induction conditions—sensitivity parameters. Based on this, the sensitivity parameters of the same nutrient medium type under different induction conditions are compared and analyzed. By calculating the range, direction, and concentration of change of each parameter under different induction conditions, the response characteristics of the nutrient medium to specific induction conditions are identified. For example, glucose nutrient media exhibits a response amplitude change concentrated within ±8% under temperature perturbation conditions, while showing a significantly shortened response delay under mechanical perturbation conditions. This response pattern is then marked as one of the specific induction response behaviors of glucose nutrient media. For different nutrient media types, response patterns with stable and repeatable characteristics under microenvironment-induced conditions were extracted. The consistency of these response patterns across different samples was statistically verified, and random fluctuations were eliminated, retaining only response behaviors highly correlated with the intrinsic properties of the nutrient media. The resulting nutrient media-induced specific response behavior data, indexed by nutrient media type and focusing on response amplitude variation range, response delay variation characteristics, stability time distribution characteristics, and nonlinear exponential distribution characteristics, is used to characterize the discriminative and stable response behaviors exhibited by different nutrient media under microenvironment-induced conditions.

[0040] Step S26: Analyze the nutrient-mediated induced response characteristic spectrum of honey using the nutrient-mediated induced specific response behavior data to generate nutrient-mediated induced response characteristic spectrum data of honey.

[0041] In this embodiment of the invention, the induced specific response behaviors corresponding to different nutrient media types are uniformly arranged according to the dimensions of induction conditions and time, constructing a three-dimensional feature framework of nutrient media—induction conditions—response characteristics. In this framework, each nutrient media corresponds to a set of stable response characteristic parameters under specific induction conditions, including the amplitude range of the induced response, the delay characteristics of the induced response, the time distribution of the stability of the induced response, and the nonlinear response morphology parameters. Subsequently, the specific response behaviors of each nutrient media are spectrally processed, that is, the discrete response behavior parameters are continuously mapped according to preset amplitude ranges, time ranges, and nonlinear exponential ranges, so that the response characteristics of different nutrient media under the same induction conditions can be expressed in a unified coordinate system. In this way, different nutrient media form comparable response characteristic distributions under microenvironmental induction conditions, thus constituting the honey nutrient media induced response characteristic spectrum. During the characteristic spectrum construction process, the response characteristics of the same nutrient media under different induction conditions are superimposed and analyzed to form multiple induced response spectra for that nutrient media; simultaneously, the response characteristics of different nutrient media under the same induction conditions are compared and analyzed to form nutrient media distinguishing spectra corresponding to the induction conditions. The final generated honey nutrient media-induced response characteristic spectrum data, with nutrient media type as the main axis and induction conditions and time scale as extended dimensions, systematically reflects the overall response characteristic distribution of various nutrient media in honey under the induction of the microenvironment. This characteristic spectrum data not only completely preserves the detailed information of the specific response behavior induced by nutrient media, but also enhances the distinguishability between different nutrient media through spectral expression, providing stable, systematic and discriminative feature inputs for subsequent steps such as nutrient media time-series hierarchical response analysis and the establishment of nutrient content mapping models.

[0042] Furthermore, step S23 includes the following steps: Based on the microenvironment-induced response window data of honey nutrient media, a nutrient media spectrum response analysis was performed to generate microenvironment-induced nutrient media spectrum response data. Then, the response signals of each nutrient media induced by the microenvironment were extracted from the microenvironment-induced nutrient media spectrum response data to generate nutrient media-induced response signal data.

[0043] In this embodiment of the invention, based on the microenvironment-induced excitation response window data of the nutrient medium, a systematic analysis is performed on the response signals of each nutrient medium under different microenvironmental conditions to quantify the influence of the microenvironment on the spectral response of the nutrient medium. Microenvironmental conditions include temperature, pH, ionic strength, humidity, and trace additives, wherein the temperature is controlled within the range of 25℃ to 45℃, the pH is adjusted between 3.5 and 6.5, the ionic strength is adjusted to the range of 0.01M to 0.1M using sodium chloride or other inorganic salts, and the humidity is controlled between 40% and 70%. The experiment obtains complete time-series spectral response data by continuously observing the chemical activity, optical signal response, and kinetic change curves of the nutrient medium under the above-mentioned different microenvironmental conditions. Specific operations include maintaining stable microenvironmental conditions for the honey sample in a constant temperature and humidity device, measuring the change in optical transmittance over time using a spectrophotometer, determining the viscosity and stress response of the sample under microenvironmental disturbances using a rheometer, and simultaneously using a multi-point detection method to obtain optical and rheological data at different locations to reflect the internal homogeneity of the sample. By comparing the response curves of the same nutrient medium under different microenvironmental conditions, indicators such as response amplitude, response onset time, peak time, steady-state amplitude, and recovery time are calculated to form a microenvironment-induced nutrient medium spectral response matrix. The rows of the matrix represent different nutrient medium types, the columns represent microenvironmental parameters, and the cells contain multidimensional response characteristic data under that condition, including optical signal peak value, viscosity change rate, and response stability coefficient. Through statistical analysis and multivariate correlation analysis, the enhancing or inhibiting effects of the microenvironment on the response of each nutrient medium are identified, providing a quantitative basis for the specific analysis of nutrient medium responses. Simultaneously, the response spectrum patterns under different microenvironmental conditions are established, providing a clear spectral foundation for subsequent microenvironment-induced signal extraction. This yields spectral response data on the influence of the microenvironment on nutrient media, fully recording the response amplitude, delay, steady-state period, and nonlinear dynamic changes of each nutrient medium under different microenvironments, laying the foundation for high-precision and comparable nutrient medium signal extraction. Based on the spectral response data on the influence of the microenvironment on nutrient media, the microenvironment-induced response signals of each nutrient medium are accurately extracted. The extraction process first involves baseline correction and noise filtering of the time-series signals of each nutrient medium under different environmental conditions within the microenvironment spectral response matrix to eliminate the influence of optical instrument drift, rheological measurement fluctuations, and environmental disturbances. Subsequently, based on the response amplitude, delay, and settling-point characteristics, the main response signal segments induced by the microenvironment are identified, and the peak response, half-maximum width at half-maximum, onset response time, maximum rate point, and recovery time are calculated. These parameters are then combined into a nutrient medium-induced response signal vector. Furthermore, by performing weighted averaging and variance analysis on the response signals of the same nutrient medium under multiple microenvironment conditions, representative induced response signals are extracted to ensure that the signals reflect the essential dynamic behavior of the nutrient medium in response to microenvironmental changes.During the extraction process, multi-dimensional correlation analysis of the response signal is also required, coupling optical signals, viscosity changes, and stress responses to quantify the amplitude and temporal consistency of the nutrient medium response. Simultaneously, the response sensitivity coefficients under different microenvironmental conditions are calculated, providing a foundation for subsequent specific response analysis. The generated nutrient medium-induced response signal data not only includes the peak amplitude, response delay, and stationary period information for each nutrient medium under different microenvironmental conditions, but also the nonlinear characteristics and dynamic trends of the response curves. This provides a complete and quantifiable time-series data foundation for nutrient medium sensitivity analysis, specific response analysis, and characteristic spectrum construction. This allows for clear separation and standardized description of the microenvironment-induced nutrient medium response signals, providing reliable input data for high-precision and systematic characterization of honey nutrients.

[0044] Furthermore, step S3 includes the following steps: Step S31: Perform time-series response gradient analysis on the nutrient medium-induced response characteristic spectrum data of honey to generate time-series response gradient data of nutrient medium type; In this embodiment of the invention, based on the nutrient-mediated response characteristic spectrum data of honey, including the response amplitude, response delay, steady-state period, and nonlinear dynamic characteristics of each nutrient medium under different microenvironments, gradient calculation is performed on the time series of each nutrient medium. Specifically, the response spectrum data is segmented by time, and the instantaneous rate of change of the response amplitude, the signal rise gradient, the fall gradient, and the steady-state period change rate are calculated for each time segment to obtain the gradient feature vector for each time period. The gradient calculation formula is ΔR / Δt, where ΔR represents the change in response amplitude and Δt represents the time interval. The calculation results generate a continuous time-series gradient curve, reflecting the dynamic change rate and direction of the nutrient medium under microenvironment excitation. To quantify the sensitivity of different nutrient media in the response process, gradient analysis also combines matrix characteristic data, coupling the response change rate with changes in honey viscosity, density uniformity, and optical transmission to form a multidimensional gradient matrix. By analyzing the peak value, zero-crossing point, rise rate, and fall rate of the gradient matrix, it is possible to distinguish between nutrient media with rapid and significant responses and those with slow and stable responses, providing an input basis for subsequent nonlinear time-domain decomposition. Furthermore, by standardizing the gradient matrix, the response gradients of different nutrient media are unified to the same dimension, enabling the gradient analysis results to be compared between samples, batches, and nectar sources. This generates systematic time-series response gradient data for different nutrient media types, providing quantifiable and comparable time dynamic characteristics for subsequent time-domain decomposition and response stage division.

[0045] Step S32: Perform nonlinear time-domain decomposition processing on the nutrient medium type induced response gradient data to generate nutrient medium type induced response time-domain decomposition data; In this embodiment of the invention, multi-scale analysis is performed on the time-series response gradient data of different nutrient media types. The response signal is decomposed into rapid change segments, moderate change segments, and slow change segments. The signal within each time period is described by characteristic parameters such as amplitude change rate, duration of the steady-state period, and response delay. During the decomposition process, numerical integration and differentiation methods are used to calculate the cumulative effect of the signal and the instantaneous response deviation. Simultaneously, the nonlinear characteristics are quantified using the ratio of peak amplitude to delay time, obtaining nonlinear indices for each time period, including parameters such as the convexity of the response curve, slope change, and peak width. The nonlinear characteristics are coupled with microenvironmental conditions and honey matrix characteristics for analysis. A multidimensional matrix is ​​used to represent the nonlinear time-domain decomposition results of each nutrient medium at different time scales, including indicators such as peak amplitude, response delay, steady-state period, and nonlinear exponent for each segment. This processing not only decomposes complex continuous time-series signals into analyzable nonlinear segments but also quantifies the proportion of each segment's contribution to the overall response, thereby revealing the dynamic behavior of nutrient media under microenvironmental stimulation. The nutrient media type-induced response time-domain decomposition data generated through this step can be used to identify nutrient media types with rapid response, delayed response, and slow adaptation, providing accurate time-domain input data for subsequent response stage division and hierarchical feature analysis, and ensuring that the dynamic response characteristics of honey nutrient media are fully quantified.

[0046] Step S33: Perform response stage segmentation processing on the time-domain decomposition data of nutrient medium type induced response to generate nutrient medium type induced response stage data; In this embodiment of the invention, time series analysis of nonlinear time-domain decomposition data is performed to divide the response process of each nutrient medium into four typical stages: initial response stage, peak response stage, stable response stage, and recovery stage. The initial response stage is defined as the time period during which the response signal amplitude exceeds the baseline by 3% to 10%, reflecting the rapid initiation response of the nutrient medium after changes in the microenvironment. The peak response stage is defined as the time period during which the amplitude reaches its maximum value and fluctuates within a 10% range, quantifying the maximum response amplitude and peak duration. The stable response stage is defined as the time period during which the response amplitude fluctuation is less than ±5%, assessing the sustained stable behavior of the nutrient medium under specific microenvironments. The recovery stage is defined as the time period during which the signal falls back to ±5% of the baseline level, describing the dynamic recovery characteristics of the nutrient medium after the response ends. For each stage, quantitative characteristics of each nutrient medium at each stage are generated by calculating the amplitude change rate, time delay, stability coefficient, and nonlinearity exponent. Furthermore, the stage characteristics are coupled with microenvironmental conditions and honey matrix state to generate a multidimensional response stage matrix. The rows of the matrix represent different nutrient medium types, the columns represent different response stages, and the cells contain the amplitude, delay, and stability parameters for each stage.

[0047] Step S34: Perform time-series hierarchical response characteristic analysis of honey nutrient media based on the nutrient media type-induced response stage data, and generate time-series hierarchical response characteristic data of honey nutrient media.

[0048] In this embodiment of the invention, the quantitative characteristics of the initial response, peak response, stable response, and recovery phase are integrated into a multi-level response matrix. Time series, amplitude changes, nonlinear exponents, and stability coefficients are mapped to different levels, including a rapid level (seconds), an intermediate level (minutes), and a long-term level (hours). Through multi-level statistical analysis, the average response amplitude, response duration, nonlinear exponent, and stability coefficient of each nutrient medium at different levels are calculated, generating a hierarchical response feature vector. Further analysis of the dynamic behavior patterns of different nutrient media at each level identifies typical behavior types such as rapid but short-lived responses, slow but sustained responses, or multi-peak responses, and quantifies them into comparable parameters. This analysis reveals the hierarchical structure and nonlinear distribution of the dynamic response of nutrient media under microenvironmental stimulation, providing complete time-series and hierarchical feature data for subsequent honey nutrient characterization and intelligent detection. The final honey nutrient medium time-series hierarchical response characteristic data includes a multidimensional quantitative feature matrix. The matrix rows represent the nutrient medium type, the columns represent the hierarchy and response stage, and the cells record the amplitude, delay, stability and nonlinearity index, so that the dynamic behavior of honey nutrient medium under microenvironment and matrix conditions can be comprehensively and quantitatively described and compared.

[0049] Furthermore, step S4 includes the following steps: Step S41: Extract stability response signals based on the time-series hierarchical response characteristic data of honey nutrient media, generate time-series stability response signal data of nutrient media, and mark the time-series stability response signal data of nutrient media as intrinsic characteristic signal data of nutrient media. In this embodiment of the invention, amplitude and temporal stability analysis is performed on the time-series response curves of each nutrient medium in the time-series hierarchical response characteristic data at different levels (rapid response level, minute-level intermediate level, and long-term response level). The stability characteristics of each response segment are identified by calculating the standard deviation, peak-to-peak rate of change, and maximum amplitude change ratio of the signal during the stable phase. A stable response segment is defined as an interval where the response amplitude fluctuation is less than 5% of the overall peak amplitude over a continuous time period, and the duration exceeds 20% of the total response time. For each stable segment, the mean response amplitude, response delay time, duration of stable period, and nonlinear exponent are calculated, and these parameters are combined to form an intrinsic characteristic signal vector of the nutrient medium to quantify the intrinsic stable behavior of each nutrient medium. Furthermore, by comparing the stability characteristics of the rapid and long-term levels, the response consistency and dynamic maintenance ability of the nutrient medium at different time scales are evaluated. For example, in glucose-based nutrient media, the peak value may be reached in the rapid response level and remain stable for approximately 30 to 60 seconds, while the stable period in the intermediate level lasts for approximately 5 to 10 minutes. In the long-term level, the stable period may extend to over 30 minutes, with an amplitude variation rate of less than 0.05. The resulting time-series stability response signal data for nutrient media not only reflects the dynamic stability of each nutrient medium but also serves as a core indicator for measuring its intrinsic response characteristics under different microenvironments and matrix conditions, providing a reliable foundation for subsequent cross-level correlation analysis and dynamic feature extraction.

[0050] Step S42: Perform cross-level correlation analysis on the intrinsic characteristic signal data of the nutrient medium to generate hierarchical correlation data of the intrinsic characteristic signal of the nutrient medium; In this embodiment of the invention, cross-level correlation analysis is performed on the intrinsic characteristic signal data of nutrient media to generate hierarchical correlation data of nutrient media intrinsic characteristic signals. This analysis compares the stability signal vectors of the rapid response level, intermediate level, and long-term response level at multiple levels. By calculating amplitude correlation coefficients, delay time correlation coefficients, and nonlinear exponential correlation coefficients, the consistency of response and the degree of hierarchical coupling between different levels are quantified. During implementation, the mean amplitude, peak duration, and nonlinear exponent of each nutrient media at different levels are normalized to construct a three-dimensional correlation matrix. The rows of the matrix represent nutrient media types, the columns represent hierarchical pairs, and the cells represent correlation indicators between levels. Furthermore, through multivariate statistical analysis, the hierarchical correlation of each nutrient media is quantitatively classified to identify nutrient media with high consistency or high differences at different levels. For example, the amplitude correlation coefficient of sucrose nutrient media between the rapid and intermediate levels may reach 0.92, while the correlation coefficient between the rapid and long-term levels is approximately 0.75, reflecting a certain dynamic change between the rapid response and long-term maintenance phases, but overall exhibiting high hierarchical correlation. This hierarchical correlation analysis not only reveals the dynamic synergistic characteristics of nutrient media at different time scales, but also provides quantitative indicators for characterizing the multi-level dynamic behavior of nutrients in complex honey samples, thus forming a complete hierarchical correlation data matrix of intrinsic nutrient media signals, providing basic data for the next step of dynamic feature analysis and nutrient characterization.

[0051] Step S43: Perform dynamic feature analysis on the intrinsic characteristic signal data of the nutrient medium to generate dynamic feature data of the intrinsic characteristic signal of the nutrient medium; In this embodiment of the invention, dynamic feature analysis is performed on the intrinsic characteristic signal data of the nutrient medium to generate dynamic feature data of the intrinsic characteristic signal of the nutrient medium. This analysis is performed by subdividing the stability response signal of each nutrient medium in a time series and calculating dynamic features such as the instantaneous response amplitude change rate, nonlinear change curve, peak duration, and signal decay rate. In the implementation process, each stability signal segment is first subdivided into 1-second time intervals, and the amplitude change rate ΔR / Δt and nonlinear exponent are calculated, where ΔR represents the amplitude change and Δt represents the time interval. The dynamic change range of the signal and the instantaneous nonlinear peak value are obtained by the cumulative distribution method. Furthermore, the signals at different levels are combined to analyze the dynamic coupling law between each level, including peak synchronization, delay time difference, and amplitude fluctuation pattern. For example, in a glucose and fructose mixed sample, the glucose response peak lasts for about 45 seconds in the rapid level and the nonlinear exponent is about 0.38, while the fructose response peak lasts for about 55 seconds and the nonlinear exponent is about 0.42. Through hierarchical combination analysis, the dynamic coupling relationship between the two sugars in the rapid level and the delay response difference in the intermediate level can be identified. The final generated dynamic characteristic data of the intrinsic properties of the nutrient medium includes the amplitude change rate, nonlinear exponent, response peak and duration of each nutrient medium at each level, forming a multidimensional dynamic characteristic matrix. This provides accurate input for the quantitative description of the time dynamic behavior of the nutrient medium under the stimulation of the microenvironment, enabling subsequent characterization of honey nutrients to be performed with high precision and multidimensional analysis based on the intrinsic dynamic characteristics.

[0052] Step S44: Analyze the characterization features of honey nutrients by using the hierarchical correlation data of the intrinsic characteristic signals of the nutrient medium and the dynamic characteristic data of the intrinsic characteristic signals of the nutrient medium, and generate characterization feature data of honey nutrients.

[0053] In this embodiment of the invention, the nutritional component characterization features of honey are analyzed by combining hierarchical correlation data and dynamic feature data of the intrinsic characteristic signals of the nutritional media. During the analysis, the amplitude correlation, delay correlation, and nonlinear exponential correlation in the hierarchical correlation matrix are comprehensively calculated with the instantaneous change rate, peak duration, and signal decay rate in the dynamic feature matrix to form a multidimensional feature fusion matrix. The rows of the feature fusion matrix represent the nutritional media type, and the columns represent feature parameters, including the hierarchical consistency index, dynamic change amplitude, peak duration, nonlinear exponent, and stability coefficient. Through statistical analysis of the matrix, a comprehensive characterization feature value for each nutritional media is calculated. For example, the comprehensive feature value of glucose is formed by combining the rapid hierarchical amplitude correlation (0.91), the intermediate hierarchical amplitude correlation (0.87), the peak duration (45 seconds), the nonlinear exponent (0.38), and the stability coefficient (0.95), providing quantitative indicators for nutritional component characterization. The nutritional component characterization data of honey not only reflects the static characteristics of the nutrient medium, but also quantifies the dynamic behavior and hierarchical response relationship induced by the microenvironment. This makes the characterization of different nutrients in honey highly accurate and comparable, providing complete input for subsequent nutrient content mapping and intelligent detection, and realizing comprehensive, quantitative and dynamic characterization of the nutrient medium in honey samples.

[0054] Furthermore, as an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart illustrating the implementation steps of step S5 is provided in this embodiment. Step S5 includes: Step S51: Extract the response signal relationship data of honey nutritional components based on the honey nutritional component characterization characteristic data; In this embodiment of the invention, based on the characteristic data of honey nutritional components, the characteristic response signals of each nutrient are extracted. The amplitude, peak duration, nonlinear exponent, and stability characteristics of each nutrient at the rapid response level, intermediate level, and long-term level are uniformly organized into a standardized vector, forming a multidimensional feature matrix. Subsequently, the amplitude relationship of each nutrient at different levels and different samples is calculated by amplitude comparison. The amplitude relationship is characterized by ratios or percentage changes. For example, the amplitude ratio of glucose to fructose at the rapid response level is 1.15, the peak duration ratio is 0.85, and the stability time difference is 10 seconds. The amplitude relationship data is used to quantify the difference in response intensity of different nutrients under the same microenvironmental stimulation, thereby distinguishing nutrients with high signal amplitude and rapid changes from those with low amplitude and slow changes. In the extraction of response delay relationships, the delay value of the peak response of each nutrient relative to the microenvironment excitation start time is calculated, and the delay relationship matrix is ​​obtained by comparing the delay time differences of different nutrients. The delay relationship matrix consists of rows representing nutrient types, columns representing sample numbers, and cells recording peak delay differences. This delay relationship data reflects the sequence and dynamic characteristics of each nutrient's response under microenvironmental stimulation, providing time-dimensional input parameters for establishing nonlinear response functions. In the stability time relationship extraction, time periods where the amplitude fluctuation of the response signal for each nutrient is less than 5% of the total peak amplitude are statistically analyzed to form a stability time relationship matrix. The matrix rows represent nutrient types, columns represent sample numbers, and cells record the length of the stable time. Stability time relationship data reflects the retention ability and dynamic stability of nutrients after microenvironmental stimulation, and can be used to quantify the persistence characteristics of nutrients. By comprehensively constructing response signal relationship data for honey nutrients using amplitude relationships, delay relationships, and stability time relationships, a complete quantitative description of the response characteristics of nutrients under multi-level, multi-timescale, and different sample conditions is achieved.

[0055] Step S52: Based on the relationship data of honey nutrient component characterization response signals, establish the multivariate nonlinear response function of honey nutrient component characterization signals and content to obtain a preliminary honey nutrient component content mapping model; In this embodiment of the invention, based on the response signal relationship data of honey nutrient components, a multivariate nonlinear response function for the characterization signals and content of honey nutrients is established. Amplitude, delay, and stability time relationships are used as independent variables, and the actual content of the nutrients is used as the dependent variable. A mapping relationship is established through a nonlinear function. The nonlinear function includes polynomial terms, exponential terms, and interactive coupling terms to describe the coupling effect and nonlinear response between signals of different nutrients. For example, for glucose, the response amplitude and content are fitted using a quadratic nonlinear function, and the relationship between peak delay and content is fitted using an exponential function. Simultaneously, the interactive influence of glucose and fructose at the rapid level is considered, forming a coupling term. During the establishment process, the amplitude, delay, and stability time of each nutrient at different levels are normalized to eliminate the influence of differences in dimensions and amplitudes, ensuring the overall comparability of the response signal relationship matrix. Subsequently, the coefficients of the nonlinear response function are optimized using a fitting algorithm. The fitting accuracy is evaluated using the coefficient of determination R², requiring R² ≥ 0.95. Simultaneously, the sum of squared residuals is minimized to ensure the fitting error is less than 5% of the total peak amplitude. When establishing the multivariate nonlinear response function, the interaction interference between different nutrients is considered. For example, the amplitude signal of glucose may be affected by fructose, sucrose, and the viscosity of the honey matrix. This is corrected by adding a nonlinear coupling term, enabling the mapping model to reflect the multivariate nonlinear characteristics of nutrients in complex honey samples. In the output matrix of the preliminary honey nutrient content mapping model, rows represent sample numbers, columns represent nutrient types, and each cell records the mapped nutrient content value and the fitting accuracy parameters of the nonlinear function, such as glucose content 42.3g / 100g, fitting residual standard deviation 1.7%, and nonlinear exponent 0.38. This mapping model can accurately map the characteristic response signal to the actual content, forming a complete multivariate nonlinear correspondence.

[0056] Step S53: Analyze the nutritional component matrix correction factors based on the honey matrix characteristic data to generate nutritional component matrix correction factors. In this embodiment of the invention, based on honey matrix characteristic data, a matrix correction factor analysis is performed on the characterization data of honey nutrients. The matrix characteristic parameters of each honey sample, including initial viscosity, density distribution, rheological properties, optical transmittance, and temperature-controlled dilution response characteristics, are systematically matched with the characterization data of each nutrient in the corresponding sample. The analysis logic of the matrix correction factor is based on the amplitude shift and response delay changes of the nutrient characterization signal with varying matrix conditions. For example, when the honey viscosity increases from 2000 mPa·s to 4000 mPa·s, the amplitude of the glucose rapid response level decreases by approximately 12%, the peak delay increases by approximately 2 seconds, and the stability time decreases by approximately 3 seconds. By calculating the amplitude shift ratio, delay shift, and stability time correction ratio, a matrix correction factor for glucose under this viscosity condition is formed. For different nutrients, such as fructose, sucrose, and amino acids, correction factor matrices are established based on their respective response amplitudes, delays, and stability times. Multi-level, multi-timescale analysis was performed on all samples. Matrix shifts of nutrient components were calculated based on the rapid response level, intermediate level, and long-term level, and the relationship between matrix characteristic parameters and characterization signal shifts was described using a nonlinear function. Matrix correction factors for each nutrient component included amplitude correction coefficients, delay correction coefficients, and stability time correction coefficients, forming a multi-dimensional matrix. Rows represented nutrient component types, columns represented sample matrix characteristic parameters, and cells recorded the corresponding quantified correction values. These matrix correction factors enabled the quantification and correction of response shifts caused by differences in viscosity, density, and rheological properties in different honey samples. This provided a foundation for accurate predictions in subsequent mapping models, ensuring high accuracy and repeatability in the detection of honey nutrient content under complex matrix conditions.

[0057] Step S54: Obtain a priori characterization and content assessment data of honey nutritional components; In this embodiment of the invention, high-precision analytical methods are used to quantitatively detect target nutrients in typical honey samples, including glucose, fructose, sucrose, amino acids, vitamins, and trace elements. High-performance liquid chromatography (HPLC), gas chromatography (GC), mass spectrometry (MS), and nuclear magnetic resonance (NMR) techniques are employed to determine the standard content value of each nutrient, with measurements repeated at least three times. The average value and standard deviation are calculated to ensure that the measurement error is less than 5% of the total content. Simultaneously, the matrix characteristic parameters of each sample are recorded, including viscosity, density distribution, optical transmittance, temperature-controlled dilution response characteristics, and rheological fluctuation characteristics. The rows of the prior characterization-content assessment data matrix represent nutrient types, the columns represent different sample numbers, and the cells contain the standard content value of the nutrient and the corresponding matrix parameters. For example, the standard content of glucose in a certain honey source sample is 42.5 g / 100 g, the viscosity is 3500 mPa·s, the density is 1.42 g / cm³, the optical transmittance is 85%, and the rheological fluctuation characteristic index is 0.38. This matrix structure not only provides a practical content benchmark for model calibration but also links matrix parameters to nutrient content, allowing for quantitative correction of matrix differences during model training. Prior data provides a realistic, accurate, and multi-dimensional training reference for the honey nutrient content mapping model, ensuring high accuracy and stability under different honey sample conditions.

[0058] Step S55: Use the nutrient matrix correction factor to correct the nutrient content offset parameter of the preliminary honey nutrient content mapping model due to differences in honey matrix, generate a corrected honey nutrient content mapping model, and use the honey nutrient prior characterization-content assessment data to train the model parameters of the corrected honey nutrient content mapping model, generate a honey nutrient content mapping model. In this embodiment of the invention, a nutrient matrix correction factor is used to correct the nutrient content offset parameters of the preliminary honey nutrient content mapping model due to matrix differences, and the model parameters are trained in conjunction with prior characterization-content assessment data. The nutrient content output by the preliminary mapping model is corrected for amplitude, delay, and stability time using the matrix correction factor. For example, the glucose amplitude correction coefficient is 1.12, and the output value of the preliminary mapping model is multiplied by this coefficient to correct the problem of low amplitude in high-viscosity honey samples; the delay correction coefficient is 0.95, which corresponds to a linear adjustment of the peak delay; the stability time correction coefficient is 1.05, which adjusts the stability time output by the model to be consistent with the actual honey matrix conditions. The corrected mapping model is further trained using prior characterization-content assessment data. The training process optimizes the model parameters based on the least squares method to minimize the sum of squared residuals between the mapping model output value and the standard content value. During the training process, multiple iterations are performed for each nutrient, and fitting is performed at the fast response level, intermediate level, and long-term level to ensure that the model maintains high accuracy under different levels and sample conditions. Through training, the model not only adjusted the matrix offset but also optimized the fitting coefficients and interactive coupling terms of the nonlinear function. This enabled the corrected honey nutrient content mapping model to accurately predict the nutrient content under different matrix conditions. For example, the error between the predicted glucose value and the prior standard value was no more than ±1.5%, and the error between the predicted fructose value and the standard value was no more than ±2%. The generated mapping model achieved comprehensive correction of amplitude, delay, and stability time in complex honey samples, ensuring high-precision mapping of multivariable, nonlinear, and coupling effects.

[0059] Step S56: Transmit the honey nutrient characterization feature data to the honey nutrient content mapping model for intelligent detection and processing of honey nutrient content, and generate honey nutrient content detection data.

[0060] In this embodiment of the invention, based on the characteristic data of honey nutritional components, a corrected honey nutritional component content mapping model is subjected to intelligent detection processing of honey nutritional component content. The characteristic vectors of each honey sample at the fast response level, intermediate level, and long-term level are used as input to provide the mapping model with complete multidimensional information, including the amplitude of the nutritional component characterization signal, response delay, stability time, nonlinear exponent, and cross-level correlation characteristics. Based on the corrected nonlinear multivariate function, the mapping model calculates the multidimensional mapping relationship between each input vector and the nutritional component response and content established in the model one by one. The nonlinear function outputs the content value of each nutritional component in the sample, and simultaneously outputs the corresponding fitting residual, confidence interval, and stability index.

[0061] During implementation, the content calculation of each nutrient component employs a three-dimensional coupled mapping relationship of amplitude, delay, and stability. For example, glucose has an amplitude of 120mV, a peak delay of 5 seconds, and a stability time of 35 seconds in the rapid level. Calculated using the nonlinear function of the mapping model, the glucose content is output as 42.3g / 100g, with a fitting residual of 1.7% and a stability coefficient of 0.95. Fructose has an amplitude of 90mV, a delay of 7 seconds, and a stability time of 8 minutes in the intermediate level. The content calculated by the mapping model is 38.5g / 100g, with a fitting residual of 2.0% and a stability coefficient of 0.92. Sucrose, amino acids, vitamins, and trace elements are also calculated using multi-dimensional input vectors to determine their actual content and response indicators, ensuring that the dynamic response characteristics of each honey sample's nutrients at different levels completely correspond to their actual content. The mapping model dynamically corrects for differences in the matrix characteristics of honey samples during processing. For example, for samples with high viscosity, low transmittance, or heterogeneous distribution, the model corrects the amplitude, delay, and stability time parameters through a matrix correction factor to ensure that the output content reflects the actual concentration of nutrients, rather than a response offset influenced by the state of the honey matrix. This processing logic guarantees high repeatability and accuracy of the detection results under different sources, honey sources, and matrix conditions. The generated honey nutrient content detection data matrix shows the sample number in rows, the nutrient type in columns, and the cell records the content value of each nutrient, along with the corresponding fitting residual and stability coefficient. For example, in the matrix, glucose content is 42.3g / 100g, fructose is 38.5g / 100g, sucrose is 5.7g / 100g, and total amino acids are 1.2g / 100g. Each value includes the fitting residual and stability index. This test data can comprehensively reflect the actual content of various nutrients in honey samples under different levels and microenvironmental responses, achieving high-precision, quantifiable, and comparable nutrient detection, and providing reliable data support for honey quality assessment, standardized production, nutrient monitoring, and scientific research analysis.

[0062] Furthermore, the honey nutrient component characterization response signal relationship data in step S51 includes nutrient component characterization response signal amplitude relationship data, nutrient component characterization response signal delay relationship data, and nutrient component characterization response signal stability time relationship data.

[0063] This specification provides a honey nutrient composition detection system for performing the honey nutrient composition detection method as described above. The honey nutrient composition detection system includes: The honey matrix feature analysis module is used to acquire the honey sample group to be tested; to analyze the honey matrix state data of the honey sample group to be tested and generate honey matrix state data; and to perform honey matrix feature analysis on the perturbation state response based on the honey matrix state data and generate honey matrix feature data. The honey nutrient medium induced response analysis module is used to perform microenvironment-induced nutrient medium response characteristic spectrum analysis based on honey matrix characteristic data, and generate honey nutrient medium induced response characteristic spectrum data. The time-series hierarchical response feature analysis module is used to perform time-series hierarchical response feature analysis on the induced response feature spectrum data of honey nutrient media, and generate time-series hierarchical response feature data of honey nutrient media; The honey nutrient component characterization feature analysis module is used to perform honey nutrient component characterization feature analysis based on honey nutrient medium time-series hierarchical response feature data, and generate honey nutrient component characterization feature data. The honey nutrient content detection module is used to intelligently detect and process the nutrient content of honey based on the characteristic data of honey nutrient composition, and generate honey nutrient content detection data.

[0064] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0065] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for detecting the nutritional components of honey, characterized in that, Includes the following steps: Step S1: Obtain the honey sample group to be tested; Analyze the honey matrix state data of the honey sample group to be tested to generate honey matrix state data; Based on the honey matrix state data, a honey matrix characteristic analysis of the perturbation state response is performed to generate honey matrix characteristic data; Step S2: Based on the honey matrix characteristic data, perform microenvironment-induced nutrient medium response characteristic spectrum analysis to generate honey nutrient medium-induced response characteristic spectrum data; Step S3: Perform time-series hierarchical response characteristic analysis on the honey nutrient medium induced response characteristic spectrum data to generate honey nutrient medium time-series hierarchical response characteristic data; Step S4: Based on the time-series hierarchical response characteristic data of honey nutrient medium, perform honey nutrient component characterization characteristic analysis to generate honey nutrient component characterization characteristic data; Step S5: Perform intelligent detection processing on the honey nutrient content of the honey nutrient characterization data to generate honey nutrient content detection data.

2. The method for detecting the nutritional components of honey according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain the honey sample group to be tested; Step S12: Collect original honey sample group characterization data to generate original honey sample group characterization data. Step S13: Extract the original honey structural state characteristic data based on the original honey sample group characterization data, and analyze the honey matrix state data through the original honey structural state characteristic data; Step S14: Perform temperature-controlled dilution on the honey sample group to be tested to generate temperature-controlled diluted honey samples, and perform temperature-controlled diluted honey state response analysis on the honey matrix state data based on the temperature-controlled diluted honey samples to generate temperature-controlled diluted honey state response data. Step S15: Perform mechanical heterogeneity perturbation on the temperature-controlled diluted honey sample to generate mechanical honey sample, analyze the honey rheological fluctuation characteristics of the mechanical honey sample, and perform mechanical perturbation honey state response analysis on the honey matrix state data based on the honey rheological fluctuation characteristics data to generate mechanical perturbation honey state response data. Step S16: Based on the state response data of temperature-controlled diluted honey and the state response data of mechanically disturbed honey, perform a state response characteristic analysis of honey matrix under external disturbance, and generate state response characteristic data of disturbed honey matrix. Step S17: Perform multi-scale response feature decoupling processing on the disturbed honey matrix state response feature data to generate honey matrix state response feature decoupled data; Step S18: Perform honey matrix feature analysis by decoupling the honey matrix state response feature data to generate honey matrix feature data.

3. The method for detecting the nutritional components of honey according to claim 2, characterized in that, The original honey structural state characteristic data mentioned in step S13 includes honey initial viscosity data, honey density distribution data, honey optical transmission characteristics data, and honey flow behavior data.

4. The method for detecting the nutritional components of honey according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Analyze the response window type of honey nutrient media based on the honey matrix characteristic data to generate honey nutrient media response window type data; Step S22: Perform microenvironment-induced stimulation nutrient medium response window type analysis on the honey nutrient medium response window type data to generate microenvironment-induced stimulation nutrient medium response window data; Step S23: Analyze the nutrient medium-induced response signals based on the microenvironment-induced nutrient medium response window data to generate nutrient medium-induced response signal data; Step S24: Perform nutrient medium-induced response sensitivity analysis based on the nutrient medium-induced response signal data to generate nutrient medium-induced response sensitivity data; Step S25: Analyze the nutrient medium-type-specific response behavior of the nutrient medium-induced response sensitivity data to generate nutrient medium-induced specific response behavior data; Step S26: Analyze the nutrient-mediated induced response characteristic spectrum of honey using the nutrient-mediated induced specific response behavior data to generate nutrient-mediated induced response characteristic spectrum data of honey.

5. The method for detecting the nutritional components of honey according to claim 4, characterized in that, Step S23 includes the following steps: Based on the microenvironment-induced response window data of honey nutrient media, a nutrient media spectrum response analysis was performed to generate microenvironment-induced nutrient media spectrum response data. Then, the response signals of each nutrient media induced by the microenvironment were extracted from the microenvironment-induced nutrient media spectrum response data to generate nutrient media-induced response signal data.

6. The method for detecting the nutritional components of honey according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Perform time-series response gradient analysis on the nutrient medium-induced response characteristic spectrum data of honey to generate time-series response gradient data of nutrient medium type; Step S32: Perform nonlinear time-domain decomposition processing on the nutrient medium type induced response gradient data to generate nutrient medium type induced response time-domain decomposition data; Step S33: Perform response stage segmentation processing on the time-domain decomposition data of nutrient medium type induced response to generate nutrient medium type induced response stage data; Step S34: Perform time-series hierarchical response characteristic analysis of honey nutrient media based on the nutrient media type-induced response stage data, and generate time-series hierarchical response characteristic data of honey nutrient media.

7. The method for detecting the nutritional components of honey according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Extract stability response signals based on the time-series hierarchical response characteristic data of honey nutrient media, generate time-series stability response signal data of nutrient media, and mark the time-series stability response signal data of nutrient media as intrinsic characteristic signal data of nutrient media. Step S42: Perform cross-level correlation analysis on the intrinsic characteristic signal data of the nutrient medium to generate hierarchical correlation data of the intrinsic characteristic signal of the nutrient medium; Step S43: Perform dynamic feature analysis on the intrinsic characteristic signal data of the nutrient medium to generate dynamic feature data of the intrinsic characteristic signal of the nutrient medium; Step S44: Analyze the characterization features of honey nutrients by using the hierarchical correlation data of the intrinsic characteristic signals of the nutrient medium and the dynamic characteristic data of the intrinsic characteristic signals of the nutrient medium, and generate characterization feature data of honey nutrients.

8. The method for detecting the nutritional components of honey according to claim 1, characterized in that, Step S5 includes the following steps: Step S51: Extract the response signal relationship data of honey nutritional components based on the honey nutritional component characterization characteristic data; Step S52: Based on the relationship data of honey nutrient component characterization response signals, establish the multivariate nonlinear response function of honey nutrient component characterization signals and content to obtain a preliminary honey nutrient component content mapping model; Step S53: Analyze the nutritional component matrix correction factors based on the honey matrix characteristic data to generate nutritional component matrix correction factors. Step S54: Obtain a priori characterization and content assessment data of honey nutritional components; Step S55: Use the nutrient matrix correction factor to correct the nutrient content offset parameter of the preliminary honey nutrient content mapping model due to differences in honey matrix, generate a corrected honey nutrient content mapping model, and use the honey nutrient prior characterization-content assessment data to train the model parameters of the corrected honey nutrient content mapping model, generate a honey nutrient content mapping model. Step S56: Transmit the honey nutrient characterization feature data to the honey nutrient content mapping model for intelligent detection and processing of honey nutrient content, and generate honey nutrient content detection data.

9. The method for detecting the nutritional components of honey according to claim 8, characterized in that, The honey nutrient component characterization response signal relationship data in step S51 includes nutrient component characterization response signal amplitude relationship data, nutrient component characterization response signal delay relationship data, and nutrient component characterization response signal stability time relationship data.

10. A honey nutritional component detection system, characterized in that, For performing the honey nutrient detection method as described in claim 1, the honey nutrient detection system comprises: The honey matrix feature analysis module is used to acquire the honey sample group to be tested; to analyze the honey matrix state data of the honey sample group to be tested and generate honey matrix state data; and to perform honey matrix feature analysis on the perturbation state response based on the honey matrix state data and generate honey matrix feature data. The honey nutrient medium induced response analysis module is used to perform microenvironment-induced nutrient medium response characteristic spectrum analysis based on honey matrix characteristic data, and generate honey nutrient medium induced response characteristic spectrum data. The time-series hierarchical response feature analysis module is used to perform time-series hierarchical response feature analysis on the induced response feature spectrum data of honey nutrient media, and generate time-series hierarchical response feature data of honey nutrient media; The honey nutrient component characterization feature analysis module is used to perform honey nutrient component characterization feature analysis based on honey nutrient medium time-series hierarchical response feature data, and generate honey nutrient component characterization feature data. The honey nutrient content detection module is used to intelligently detect and process the nutrient content of honey based on the characteristic data of honey nutrient composition, and generate honey nutrient content detection data.

Images