Intelligent analysis system and method based on atomized liquid components

Abstract

The invention belongs to the technical field of atomized liquid component analysis, and discloses an intelligent analysis system and method based on atomized liquid components, and the method comprises the following steps: carrying out multi-parameter sampling on atomized liquid to obtain pyrolysis data; performing temperature gradient analysis to generate a pyrolysis spectrogram; identifying trace markers to generate a feature library; analyzing the component evolution track to generate a dynamic evolution spectrum; carrying out partition demarcation processing to generate an atomization characteristic spectrum; performing thermodynamic analysis to generate a reaction map; analyzing component diffusion to generate quantized data; constructing a sensory response prediction engine; performing intelligent analysis to generate response feature data; evaluating and generating a quality report; and performing optimization processing to generate formula suggestions and executing accurate blending. According to the method, accurate analysis and evaluation of atomized liquid components are achieved through multi-dimensional analysis, a scientific basis is provided for product optimization, and safety and taste experience are effectively balanced.

Description

Technical Field

[0001] The present invention relates to the technical field of atomized liquid composition analysis. More specifically, the present invention relates to an intelligent analysis system and method for atomized liquid composition. Background Art

[0002] With the emergence of various atomized liquid products, the complexity and diversity of their components have also increased. At present, the analysis of atomized liquid components mainly relies on traditional chemical analysis methods such as gas chromatography-mass spectrometry and high-performance liquid chromatography. Although these methods are accurate and reliable, they have obvious defects; traditional methods cannot detect the thermal degradation products generated during the high-temperature atomization process and their dynamic change laws in real time. When the product works under different powers and different usage modes, the chemical components in the atomized liquid will undergo complex thermal chemical reactions, generating a variety of intermediate products and final products. The composition and content of these products are significantly different from those of the initial atomized liquid. Traditional analysis methods can often only provide the composition data of static samples and cannot capture the dynamic changes during actual use. It is difficult for existing technologies to identify and quantify ultra-trace specific thermal degradation markers. Some specific fragrance components (such as certain aldehyde and ketone compounds) will generate extremely trace but specific degradation products with significant biological activities during the atomization process. Although the content of these products is extremely low, they may have unique physiological effects on users. For example, vanillin may generate 4-methylguaiacol at a specific temperature, and this substance can cause respiratory irritation reactions in specific populations at extremely low concentrations. However, due to its extremely low content and co-elution with other components, it is very difficult for traditional analysis methods to accurately separate and quantify it. Traditional analysis methods cannot establish an accurate correlation between the atomized liquid composition fingerprint and the subjective feelings of users. Different users may have completely different subjective evaluations of the same atomized liquid. This difference is partly due to the different individual sensory sensitivities and partly due to the complex interactions of trace components in the atomized liquid. Existing technologies lack a method that can correlate objective chemical analysis with the subjective experience of users, which leads to a large amount of blindness and repetition in the development process of atomized liquid formulations.

[0003] In view of this, the present invention proposes an intelligent analysis system and method for atomized liquid composition to solve the above problems. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art and achieve the above object, the present application provides an intelligent analysis method for atomized liquid composition, including: Step S1: Sampling multi-parameter atomization conditions of the atomized liquid to obtain original pyrolysis data; performing temperature gradient deconstruction analysis on the original pyrolysis data to generate a multi-dimensional pyrolysis spectrum; Step S2: Perform precise identification processing of trace markers based on the multi-dimensional pyrolysis spectrogram to generate a trace marker feature library; perform multi-dimensional component evolution trajectory analysis based on the trace marker feature library to generate a dynamic component evolution spectrogram; Step S3: Perform partitioning and demarcation processing of the atomization process based on the dynamic component evolution spectrogram to generate a partitioned atomization feature spectrogram; perform component-temperature interaction thermodynamics analysis based on the partitioned atomization feature spectrogram to generate a component thermodynamics reaction map; Step S4: Perform component diffusion kinetics analysis based on the partitioned atomization feature spectrogram to generate component diffusion quantification data; construct a sensory response prediction engine based on the component thermodynamics reaction map to obtain a sensory response prediction engine; transmit the component diffusion quantification data to the sensory response prediction engine in real time for intelligent analysis of personalized sensory experience to generate sensory response feature data; perform intelligent evaluation of the global atomized liquid characteristics based on the sensory response feature data to generate an atomized liquid quality evaluation report; perform safety and taste balance optimization processing on the original pyrolysis data according to the atomized liquid quality evaluation report to generate an optimized formulation recommendation, and perform precise preparation of the atomized liquid through the optimized formulation recommendation.

[0005] This application provides an intelligent analysis system based on atomized liquid components for executing the intelligent analysis method based on atomized liquid components as described above, including: Multi-dimensional thermal spectrum construction module: Sample multi-parameter atomization conditions of the atomized liquid to obtain original pyrolysis data; perform temperature gradient deconstruction analysis based on the original pyrolysis data to generate a multi-dimensional pyrolysis spectrogram; Dynamic trajectory analysis module: Perform precise identification processing of trace markers based on the multi-dimensional pyrolysis spectrogram to generate a trace marker feature library; perform multi-dimensional component evolution trajectory analysis based on the trace marker feature library to generate a dynamic component evolution spectrogram; Thermodynamic interaction modeling module: Perform partitioning and demarcation processing of the atomization process based on the dynamic component evolution spectrogram to generate a partitioned atomization feature spectrogram; perform component-temperature interaction thermodynamics analysis based on the partitioned atomization feature spectrogram to generate a component thermodynamics reaction map; Intelligent evaluation and optimization module: Perform component diffusion kinetics analysis based on the partitioned atomization feature spectrogram to generate component diffusion quantification data; construct a sensory response prediction engine based on the component thermodynamics reaction map to obtain a sensory response prediction engine; transmit the component diffusion quantification data to the sensory response prediction engine in real time for intelligent analysis of personalized sensory experience to generate sensory response feature data; perform intelligent evaluation of the global atomized liquid characteristics based on the sensory response feature data to generate an atomized liquid quality evaluation report; perform safety and taste balance optimization processing on the original pyrolysis data according to the atomized liquid quality evaluation report to generate an optimized formulation recommendation, and perform precise preparation of the atomized liquid through the optimized formulation recommendation.

[0006] Technical effects and advantages of an intelligent analysis system and method based on the composition of atomized liquid in the present invention: Through multi-parameter atomization condition sampling and temperature gradient deconstruction analysis, the present invention realizes a comprehensive characterization of the pyrolysis components generated during the atomization process, and can convert the complex pyrolysis process into a visual multi-dimensional pyrolysis spectrogram, thereby enhancing the monitoring accuracy of the changes in the composition of the atomized liquid. By constructing a trace marker feature library and generating a dynamic component evolution spectrum, it can effectively identify and track ultra-trace active components, achieve precise quantitative analysis of bioactive substances, ensure comprehensive monitoring of trace components that may have a potential impact on human health, greatly improve the accuracy of the safety assessment of the atomized liquid, and avoid the problem that trace components are easily overlooked in traditional analysis methods. Through the zoning and demarcation processing of the atomization process and the component-temperature interaction thermodynamics analysis, the system can comprehensively master the thermodynamic behavior characteristics of the atomized liquid components in different temperature ranges, cope with the component change rules under different atomization conditions, and improve the system's analytical ability and accuracy for the complex thermodynamic changes in the atomization process through the analysis of the thermal degradation critical point and the analysis of the molecular recombination activity. The introduction of reaction path network diagrams and non-linear thermokinetic behavior analysis realizes the multi-level information fusion of complex reaction systems, providing a theoretical basis for component safety assessment. By analyzing the component diffusion kinetics and constructing a sensory response prediction engine, it can analyze the diffusion characteristics of atomized components in real time, predict the possible sensory experiences, dynamically evaluate the impact of different component combinations on the user's senses, ensure a high-quality sensory experience even when the atomization conditions change, and automatically generate a personalized quality assessment report and optimize the formula according to the real-time atomization characteristics based on the automated assessment analysis of individual perception difference data, improving the system's ability to adapt to different user needs, significantly enhancing the user experience, and ensuring product safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 Schematic diagram of an intelligent analysis method based on the composition of atomized liquid according to the present invention; Figure 2 Schematic diagram of an intelligent analysis system based on the composition of atomized liquid according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0008] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0009] Embodiment 1

[0010] Please refer to Figure 1As shown in the figure, an intelligent analysis method for the composition of atomized liquid in this embodiment includes: Step S1: Sampling the atomized liquid under multi-parameter atomization conditions to obtain original pyrolysis data; performing temperature gradient deconstruction analysis on the original pyrolysis data to generate a multi-dimensional pyrolysis spectrum; Step S2: Performing precise identification processing of trace markers based on the multi-dimensional pyrolysis spectrum to generate a trace marker feature library; analyzing the multi-dimensional composition evolution trajectory based on the trace marker feature library to generate a dynamic composition evolution spectrum; Step S3: Performing partition demarcation processing on the atomization process based on the dynamic composition evolution spectrum to generate a partition atomization feature spectrum; performing component-temperature interaction thermodynamics analysis based on the partition atomization feature spectrum to generate a component thermodynamics reaction map; Step S4: Performing component diffusion kinetics analysis based on the partition atomization feature spectrum to generate component diffusion quantification data; constructing a sensory response prediction engine based on the component thermodynamics reaction map to obtain a sensory response prediction engine; transmitting the component diffusion quantification data to the sensory response prediction engine in real time for intelligent analysis of personalized sensory experience to generate sensory response feature data; performing intelligent evaluation of the global atomized liquid characteristics based on the sensory response feature data to generate an atomized liquid quality evaluation report; performing safety and taste balance optimization processing on the original pyrolysis data according to the atomized liquid quality evaluation report to generate an optimized formula suggestion, and performing precise preparation of the atomized liquid through the optimized formula suggestion.

[0011] Preferably, Step S1 includes: Performing dynamic sampling of the atomized liquid under multi-parameter atomization conditions to generate original pyrolysis data; Performing classification processing on the types of pyrolysis products based on the original pyrolysis data to generate pyrolysis product type feature data; Performing multi-stage pyrolysis kinetics characterization based on the pyrolysis product type feature data to generate pyrolysis kinetics characterization data; Performing temperature gradient deconstruction analysis based on the pyrolysis kinetics characterization data to generate a multi-dimensional pyrolysis spectrum.

[0012] In the embodiments of the present invention, an adjustable atomization device is adopted, and multiple groups of atomization parameters are set, including atomization temperature (200°C - 350°C, at intervals of 5°C), heating power (10W to 50W, at intervals of 5W), gas flow rate (10 to 30 mL / s, at intervals of 2 mL / s), etc., to perform dynamic sampling on the atomization liquid to be measured. During the sampling process, pyrolysis products generated by the atomization liquid at different temperatures are captured through pyrolysis mass spectrometry technology and gas chromatography - mass spectrometry technology. Sampling is continuously carried out for 30 seconds at each temperature point, and mass spectrometry data is recorded every 5 seconds to form time - series pyrolysis data. At the same time, auxiliary parameters such as the resistance change of the atomization device and power output fluctuation are recorded to ensure precise control of the sampling conditions. All the collected data is numbered and saved in the order of temperature and time to generate an original pyrolysis data set containing the complete pyrolysis process. The collected original pyrolysis data is classified. According to molecular mass, structural characteristics, and chemical properties, the pyrolysis products are divided into multiple categories such as alcohols, aldehydes, ketones, acids, esters, aromatic compounds, heterocyclic compounds, etc. Preliminary classification is carried out using characteristic ion fragments in the mass spectrometry data (such as mass - to - charge ratio: 31 represents alcohols, mass - to - charge ratio: 43 represents ketones, mass - to - charge ratio: 91 represents aromatic). For each type of compound, characteristic data of pyrolysis product types are generated based on their appearance frequency and intensity at different temperatures. For example, at 250°C, the characteristic peak intensity of the detected alcohol compounds is 12000 counts, that of ketones is 8500 counts, and that of aromatic is 3200 counts; while at 300°C, these values become 9500, 12800, and 7600 counts respectively, indicating that the increase in temperature leads to a decrease in alcohols and an increase in ketones and aromatic compounds. For each type of pyrolysis product, the formation kinetic parameters at different temperature stages are analyzed, including formation rate, peak temperature, half - life, etc. The apparent activation energy and pre - exponential factor of various compounds are calculated through the Arrhenius equation to evaluate the temperature sensitivity of the pyrolysis reaction. For example, the apparent activation energy of alcohol compounds is 145 kJ / mol, and the pre - exponential factor is / s; while the apparent activation energy of ketones is 120 kJ / mol, and the pre - exponential factor is / s, indicating that ketones are more likely to form at lower temperatures. Based on the pyrolysis kinetic data, a relationship curve between the reaction rate constant and temperature is constructed to compare the relative formation amounts of various compounds at different temperatures, generating pyrolysis kinetic characterization data. The temperature gradient deconstruction technique is used to transform the pyrolysis kinetic characterization data into a three-dimensional spectrogram, where the x-axis represents temperature (200 °C to 350 °C), the y-axis represents retention time or mass-to-charge ratio, and the z-axis represents signal intensity. Through mathematical processing methods such as Fourier transform and wavelet transform, noise reduction and resolution enhancement are performed on the original spectrogram to extract the fine structure of the temperature-response relationship. In the spectrogram, the relative abundance changes of each compound at different temperatures are represented by color gradients or contour lines, with dark regions indicating high abundance and light regions indicating low abundance. The spectrogram data of each temperature point are integrated to generate a multi-dimensional pyrolysis spectrogram covering the entire temperature range, which intuitively shows the complete profile of the changes in the atomized liquid composition with temperature, providing a comprehensive data basis for subsequent identification of trace markers.

[0013] Preferably, step S2 includes: Extract pyrolysis intensity characteristics from the multi-dimensional pyrolysis spectrogram to generate pyrolysis intensity spectrum data; Extract molecular fragment spectrum characteristics from the multi-dimensional pyrolysis spectrogram to generate molecular fragment spectrum data; Use a preset multi-dimensional spectrogram deconvolution algorithm to perform initial trace marker identification and analysis on the pyrolysis intensity spectrum data and molecular fragment spectrum data, generating initial trace marker data; Perform low signal-to-noise ratio signal enhancement processing based on the initial trace marker data to generate a trace marker feature library; Analyze the multi-dimensional component evolution trajectory according to the trace marker feature library to generate a dynamic component evolution spectrogram.

[0014] In the embodiments of the present invention, the total ion current intensity (TIC) data, i.e., the sum of mass-to-charge ratios, of each temperature point is extracted from the multi-dimensional pyrolysis spectrum to quantify the overall intensity of the pyrolysis reaction at different temperatures. The TIC values of each temperature point (such as 200°C, 205°C, 210°C up to 350°C) are normalized to eliminate the influence caused by the difference in sample amount. The TIC curve is processed by a smoothing algorithm to reduce the influence of random noise. The first derivative and the second derivative of the TIC at each temperature point are calculated to identify the inflection points of the change in pyrolysis intensity and the characteristics of the acceleration change. These inflection points usually correspond to the start or completion temperatures of key chemical reactions, such as the initial pyrolysis temperature and the complete pyrolysis temperature of the main components. For example, an obvious peak of the first derivative of the TIC is detected at 270°C, indicating that the pyrolysis reaction rate reaches the maximum at this temperature point; the second derivative is detected to change from positive to negative at 320°C, indicating that the pyrolysis rate begins to slow down. The pyrolysis intensity data of all temperature points are integrated to generate pyrolysis intensity spectrum data, which includes the quantitative relationship between the atomization temperature and the pyrolysis intensity. The mass spectrometry data in the multi-dimensional pyrolysis spectrum is analyzed in detail, and the change in signal intensity of each mass-to-charge ratio at different temperatures is extracted to generate molecular fragment spectrum data. Characteristic molecular fragments are concerned, such as mass-to-charge ratio: 31 (characteristic fragment of alcohols), mass-to-charge ratio: 43 (characteristic fragment of ketones), mass-to-charge ratio: 91 (characteristic fragment of aromatics), etc., and the intensity change profiles of these characteristic fragments in the temperature range of 200°C to 350°C are recorded. Through multivariate statistical methods such as principal component analysis (PCA) or non-negative matrix factorization (NMF), the data dimension is reduced to extract the main variation patterns. The peak shape analysis is performed on the temperature-intensity curve of each characteristic fragment to extract parameters such as peak height, peak width, peak area, and peak shape asymmetry. These parameters reflect the pyrolysis behavior characteristics of the molecular fragments. For example, the mass-to-charge ratio: 31 shows a broad peak in the low-temperature region (200 to 250°C) with a peak area of 8500 units; while in the high-temperature region (300 to 350°C), it shows a sharp peak and the peak area drops to 3200 units, indicating that alcohols decompose rapidly at high temperatures. The data of all characteristic fragments are integrated to generate molecular fragment spectrum data, which describes the fragmentation rules of various molecular structures during pyrolysis. Using a pre-designed multi-dimensional spectrum deconvolution algorithm, combined with the pyrolysis intensity spectrum data and the molecular fragment spectrum data, the identification of trace markers is carried out. The deconvolution algorithm is based on the non-linear least squares method, which can separate the complex mixed spectrum into the contributions of individual components, and is particularly suitable for processing overlapping spectrum signals. The deconvolution parameters are set as follows: the upper limit of the number of iterations is 1000 times, the convergence threshold is 0.001, and a fitting residual less than 5% is regarded as effective separation. The algorithm processes the spectrum data of each temperature point to extract the possible trace marker signals, and even if their content is extremely low (as low as 0.1% of the total signal), they can be detected. For each potential marker obtained by deconvolution, it is matched with the standard spectrum library according to its mass spectrometry characteristics, and the matching degree score (0 to 100 points) is calculated. A matching degree exceeding 85 points is regarded as reliable identification.Through deconvolution processing, dozens of potential trace markers are identified from the original spectrogram, including common spice components such as nicotine, vanillin, menthol, etc., and various potential pyrolysis by-products such as toluene, naphthalene, benzo[a]pyrene, etc. The relative content changes of each identified trace marker at different temperatures are recorded to generate initial trace marker data. The low signal-to-noise ratio signals in the initial trace marker data are enhanced to improve the detection accuracy of trace components. Adaptive wavelet transform is used to estimate and suppress the noise background, and corresponding wavelet basis functions are used according to the noise characteristics in different frequency regions. The signal-to-noise ratio threshold is set at 3:1. Signals below this threshold are enhanced by cumulative averaging (100 scans), and mathematical enhancement techniques such as correlation integration and peak detection algorithms are applied to particularly weak signals (signal-to-noise ratio below 2:1) to improve signal distinguishability. Isotope ratio verification technology is used to verify the authenticity of trace marker signals by checking the isotope peak ratios of elements such as carbon, hydrogen, and oxygen, and to exclude instrument noise or background interference. For example, the 13C / 12C ratio of carbon should be close to 1.1%. If the measured ratio deviates significantly, it is regarded as a suspicious signal. After signal enhancement processing, a trace marker feature library is constructed, which contains the retention time, mass spectrometry characteristics, characteristic ion fragments, appearance rules at different temperatures, and co-occurrence relationships with other components of each marker. Each marker entry in the feature library contains information such as a unique ID (e.g., MK001), chemical name, molecular formula, mass-to-charge ratio, maximum response temperature, apparent activation energy, etc. Based on the trace marker feature library, the concentration change trajectories of each marker at different temperatures are traced, and the rules of its generation, transformation, and disappearance are analyzed. For each key marker, a temperature-concentration curve is plotted, marking its starting temperature, maximum concentration temperature, and complete disappearance temperature. The transformation relationships between markers are analyzed to identify precursor-product pairs. For example, when a precursor substance A starts to decrease at 250°C and at the same time product B starts to increase, it indicates that there may be a transformation relationship of A→B. Multivariate statistical methods (such as cluster analysis and principal component analysis) are used to explore the correlations between markers and identify co-varying groups, which usually reflect common chemical reaction pathways. For example, cluster analysis shows that markers MK003, MK007, and MK012 form a co-varying group, and they increase synchronously in the temperature range of 270 to 290°C, indicating that they may be derived from the decomposition of the same precursor substance. The evolution trajectories of all markers are integrated to construct a complete dynamic component evolution spectrum, which shows the whole process evolution rules of each component in the atomizing liquid from the initial state to complete pyrolysis, providing a data basis for subsequent atomization process zoning.

[0015] Preferably, based on the initial trace marker data, low signal-to-noise ratio signal enhancement processing is performed to generate a trace marker feature library, including: Perform bioactivity threshold analysis based on the initial trace marker data to generate bioactivity threshold data, where the bioactivity threshold data includes physiological response trigger concentration data and marker-specific indices; Perform ultra-trace active ingredient enrichment processing based on the physiological response trigger concentration data to generate ultra-trace active ingredient data; Perform bioactivity marker correlation processing based on the ultra-trace active ingredient data and the marker-specific indices to generate bioactivity marker data; Perform marker screening processing on the initial trace marker data based on the bioactivity marker data to generate a trace marker feature library.

[0016] In the embodiments of the present invention, the biological activity thresholds of the components in the initial trace biomarker data are analyzed to determine the lowest concentration level that can cause a physiological response in the human body. By referring to literature data and experimental verification, a biological activity threshold database for each biomarker is established. For flavor biomarkers (such as vanillin, menthol), their taste thresholds (the lowest concentration that can be perceived, usually at the ppb level) and odor thresholds (the lowest concentration that can be detected by olfaction) are recorded. For example, the taste threshold of vanillin is 2 ppb, and the odor threshold is 0.5 ppb; the taste threshold of menthol is 5 ppb, and the odor threshold is 1 ppb. For potentially harmful biomarkers (such as formaldehyde, benzo[a]pyrene), their health impact thresholds are recorded, including the short-term exposure limit (STEL) and the long-term exposure limit (TWA). For example, the STEL of formaldehyde is 2 ppm, and the TWA is 0.75 ppm; the TWA of benzo[a]pyrene is 0.2 μg / m³. According to the chemical structure and receptor affinity of the biomarker, the biomarker specificity index is calculated to quantify its binding ability to specific biological receptors. The specificity index uses a scoring system from 0 to 100, and a higher score indicates stronger specificity. For example, the specificity index of nicotine with the nicotinic acetylcholine receptor is 95, the specificity index of menthol with the TRPM8 cold receptor is 87, and the specificity index of vanillin with the vanilloid receptor is 76. Combining the above data, biological activity threshold data are generated, including the physiological response trigger concentration and specificity index of each biomarker, providing a threshold reference for subsequent analysis. Based on the physiological response trigger concentration data, ultra-trace components with concentrations below the detection limit but potentially having biological activity are enriched. The solid-phase microextraction (SPME) technique is used, and extraction heads with different polarities (such as PDMS, Carboxen, DVB, etc.) are used to selectively adsorb the atomized aerosol. The extraction conditions are: temperature 300 °C, adsorption time 10 minutes, and desorption time 2 minutes. The cold trap technique (-20 °C) is used to condense and enrich the pyrolysis products, and then the enriched substances are gradually released by programmed temperature increase (from -20 °C to 300 °C at 5 °C / min). High-sensitivity analysis is performed on the enriched samples, and the selected ion monitoring (SIM) mode is used to increase the instrument sensitivity to 100 times that of the conventional full-scan mode. For example, for trace nicotine metabolites, characteristic ions with mass-to-charge ratios of 84, 133, 161, etc. are monitored, and the detection limit can reach 0.1 ppb. Through the above enrichment and detection techniques, a variety of ultra-trace active components are extracted from the initial biomarker data, including nicotine alkaloids, polycyclic aromatic hydrocarbons, carbonyl compounds, etc. Although their concentrations are extremely low (usually at the ppb or ppt level), they have significant biological activity due to high specificity. Ultra-trace active component data are generated, recording the exact concentrations and the temperature ranges in which these components appear. Combining the ultra-trace active component data and the biomarker specificity index, the biological activity correlation between different biomarkers is analyzed. Through multivariate correlation analysis and receptor simulation software, the binding mode and synergistic effect of the biomarker with the biological receptor are predicted.Marker combination models are established respectively for taste sensations (such as sweet, sour, bitter, salty, umami), olfactory sensations (such as floral, woody, tobacco, fruit scents, etc.) and other sensations (such as cooling, warming, pungency). For example, for the "cooling" experience, menthol (specificity index 87) and menthene (specificity index 65) show a strong synergistic effect, and the cooling sensation is the strongest when the ratio of the two is 3:1. There is a complementary effect between nicotine (specificity index 95) and certain pyrazine compounds (specificity index 73 to 82), which can enhance the integrity of the tobacco sensory characteristics. Based on the results of the correlation analysis, bioactive marker data are generated, including the synergistic relationship, antagonistic relationship and threshold change effect between key markers, providing a basis for subsequent sensory experience prediction. According to the bioactive marker data, the initial trace marker data are screened to construct a complete trace marker feature library. The screening rules are set as follows: markers with a specificity index greater than 50 and a concentration greater than the physiological response trigger concentration are retained; markers with a specificity index greater than 80 are retained even if the concentration is slightly lower than the trigger concentration; for marker combinations with a synergistic effect, even if the individual components do not meet the conditions, as long as the combined effect is significant, they will also be retained. Through these rules, markers that have an important impact on the sensory characteristics and physiological effects of the e-liquid are screened out, and interfering components with insignificant bioactivity are removed. The retained markers are classified and cataloged, and a hierarchical structure is established according to the chemical structure (such as alcohols, ketones, aldehydes, etc.) and the type of bioactivity (such as aroma contribution, taste contribution, sensory stimulation, etc.). A unique feature code is assigned to each marker, and detailed information such as its chemical properties, spectral characteristics, and bioactive parameters is recorded. Finally, a trace marker feature library is generated, which contains dozens to hundreds of trace components that have an important impact on the characteristics of the e-liquid, providing a complete marker reference system for subsequent analysis.

[0017] Preferably, step S3 includes: Performing partition delimitation processing on the atomization process according to the dynamic component evolution spectrum to generate a partition atomization characteristic spectrum; Detecting the thermodynamic stability limit according to the partition atomization characteristic spectrum to generate thermodynamic stability data; Performing component-temperature interactive thermodynamic analysis according to the thermodynamic stability data to generate a component thermodynamic reaction map.

[0018] In the embodiments of the present invention, the dynamic component evolution spectrum is analyzed to find the key transition points during the atomization process, and the entire atomization process is divided into several characteristic regions. The inflection point detection algorithm (such as the second derivative method) is used to identify the mutation points in the component spectrum, which usually correspond to the changes in the reaction mechanism. Through cluster analysis, the natural partition of the atomization process is determined. For example: Preheating zone (200, 230] °C: The main components remain stable, and only a small amount of volatile components are released; Mild pyrolysis zone (230, 270] °C: The low-boiling components volatilize rapidly, and some compounds start to pyrolyze; Moderate pyrolysis zone (270, 310] °C: A large amount of the main components pyrolyze to generate a variety of secondary pyrolysis products; Deep pyrolysis zone (310, 350] °C: High molecular weight polycyclic compounds and low molecular fragments are formed, and more harmful substances are generated at the same time. The characteristic parameters of each region are calculated, such as the change rate of the main components, the number of types of newly generated substances, the pyrolysis intensity, etc. For example, in the mild pyrolysis zone, 27 newly generated substances are detected, and the degradation rate of the main components is 15%; while in the deep pyrolysis zone, the number of newly generated substances increases to 56, and the degradation rate of the main components reaches 85%. The component spectrum of each partition is reconstructed to highlight the characteristic components and transformation relationships in this region, and a partitioned atomization characteristic spectrum is generated, which includes the typical component spectrum and transformation characteristics of each temperature interval. Thermodynamic stability analysis is performed on each temperature interval in the partitioned atomization characteristic spectrum to determine the stability boundaries of each component at different temperatures. The thermal degradation onset temperature (T-onset) and complete degradation temperature (T-complete) of each key component are measured to construct a temperature-stability curve. For example, the T-onset of glycerol is 240 °C and the T-complete is 275 °C; the T-onset of propylene glycol is 225 °C and the T-complete is 260 °C. The relationship between the thermal degradation rate constant k and the temperature T is calculated, and the activation energy Ea is determined through the Arrhenius equation. The higher the activation energy, the better the thermal stability of the substance. For example, the Ea of glycerol is 125 kJ / mol, and the Ea of propylene glycol is 115 kJ / mol. According to the results of thermodynamic analysis, a safe atomization temperature range is determined for each component, that is, within this temperature range, the component remains stable or undergoes controllable degradation without generating harmful by-products. Thermodynamic stability data are generated, including the stability parameters, safe temperature range, and degradation kinetic constants of each component. Based on the thermodynamic stability data, the interactions and temperature dependencies between different components during pyrolysis are analyzed. A component-temperature interaction matrix is constructed to quantify the interaction intensity between each pair of components at different temperatures. For example, the interaction intensity between glycerol and nicotine at 250 °C is 0.75 (a positive value indicates mutual promotion of degradation), while the interaction intensity between glycerol and hydroxymethylfurfural at the same temperature is -0.4 (a negative value indicates mutual inhibition of degradation). Through pyrolysis mass spectrometry technology, the decomposition paths of each component in the mixture are tracked to identify pyrolysis intermediates and final products.For example, at 270 °C, the concentration of propionaldehyde produced by the mixture of glycerol and propylene glycol is 30% higher than that during pyrolysis alone, indicating a synergistic decomposition effect between the two. A reaction energy diagram is plotted, marking the activation energy and reaction heat of key conversion steps to help understand the reaction rate and product distribution. For example, the energy barrier for the N-oxidation reaction of nicotine is 110 kJ / mol, while the energy barrier for the demethylation reaction is 130 kJ / mol, indicating that N-oxidation is the dominant reaction path at lower temperatures. Integrating the above thermodynamic analysis results, a component thermodynamic reaction map is generated, which details the decomposition paths, interactions, and product distributions of each component in the e-liquid at different temperatures, providing a thermodynamic basis for subsequent sensory response prediction.

[0019] Preferably, thermodynamic stability limit detection is performed based on the partition atomization characteristic spectrum to generate thermodynamic stability data, including: Performing thermal degradation critical point analysis based on the partition atomization characteristic spectrum to generate thermal degradation critical point data; Performing molecular recombination activity analysis based on the partition atomization characteristic spectrum to generate molecular recombination activity data; Constructing a component-temperature phase change matrix based on the thermal degradation critical point data and the molecular recombination activity data to generate a component-temperature phase change matrix; Performing thermodynamic stability limit detection based on the component-temperature phase change matrix to generate thermodynamic stability data.

[0020] In the embodiments of the present invention, the thermal degradation behaviors of the main components in the partition atomization characteristic spectrum are analyzed to determine the critical transition points of their thermal stabilities. For each key component, three key temperature points of its thermal degradation are measured through a temperature scan experiment: the thermal degradation onset temperature (T-onset), which is the temperature at which the component starts to decompose significantly; the thermal degradation acceleration temperature (T-acceleration), which is the temperature at which the decomposition rate increases significantly; the thermal degradation complete temperature (T-complete), which is the temperature at which the component is completely decomposed. These temperature points are accurately measured using differential thermogravimetric analysis (DTG) and thermogravimetry-mass spectrometry (TG-MS) techniques, with a resolution of ±1 °C. For example, for glycerol, T-onset = 245 °C, T-acceleration = 262 °C, and T-complete = 278 °C are measured; for propylene glycol, T-onset = 232 °C, T-acceleration = 248 °C, and T-complete = 265 °C are measured. Calculate the thermal degradation kinetic parameters of each component near the critical point, including the zero-order reaction rate constant and the first-order reaction rate constant and the activation energy Ea. For example, the = 0.028 / s, Ea = 125 kJ / mol. Plot the thermal degradation rate - temperature curve, mark the positions of the critical points, and calculate the temperature gradients and reaction gradients between the critical points. Generate thermal degradation critical point data, including the critical temperature values, corresponding degradation rates, and kinetic parameters of each component. Study the recombination activity of molecular fragments during the atomization process and analyze the ability of the pyrolysis products to form new compounds. Use high-resolution mass spectrometry technology to monitor the newly generated molecular ions during the atomization process, especially focusing on compounds with molecular weights greater than the original components, which usually result from the recombination reactions of molecular fragments. For example, at 280 °C, a newly generated substance with a mass-to-charge ratio of 284.1362 was detected and determined to be the product formed by the aldol condensation of two aldehyde fragments through accurate mass analysis. For each temperature point, calculate the molecular recombination index (MRI), defined as the ratio of the number of newly generated compounds to the number of original degradation products. For example, at 250 °C, MRI = 0.15, indicating that approximately 15 out of every 100 original degradation products participated in the recombination reaction; while at 300 °C, MRI increased to 0.42, indicating that high temperature promoted the molecular recombination activity. Through isotope labeling experiments (such as using 13C-labeled components), track the migration and recombination paths of specific molecular fragments to determine the key recombination mechanisms. For example, the 13C-labeling experiment showed that at 290 °C, the rate of the addition reaction between the propylene glycol carbon chain fragment and the aromatic compound was 2.3 times that of direct oxidation. Generate molecular recombination activity data, including the molecular recombination index, main recombination reaction types, and relative abundances of recombination products at different temperatures. Combine the thermal degradation critical point data and the molecular recombination activity data to construct a component-temperature phase change matrix to describe the phase change behavior of each component in the atomized liquid under temperature changes. The rows of the matrix represent different components, and the columns represent temperature points (from 200 °C to 350 °C, with a step of 5 °C), and the matrix elements represent the phase change state codes under the corresponding conditions. Define the phase change state codes: 0 represents the stable state (no obvious change); 1 represents the mild decomposition state (<10% degradation); 2 represents the moderate decomposition state (10% to 50% degradation); 3 represents the severe decomposition state (50% to 90% degradation); 4 represents the complete decomposition state (>90% degradation). For each temperature point, calculate the overall phase change index (TPI) as the weighted average of the phase change states of all components. For example, at 270 °C, the phase change state of glycerol is 3 (severe decomposition), the phase change state of propylene glycol is 4 (complete decomposition), and the phase change state of nicotine is 2 (moderate decomposition), and the overall phase change index TPI = 3.1, indicating that the overall is in a severe decomposition state at this temperature. Through cluster analysis, identify groups of components with similar behaviors in the phase change matrix to provide a basis for formulation optimization. Generate the component-temperature phase change matrix to comprehensively describe the phase change behavior of each component in the atomized liquid with temperature changes. Based on the component-temperature phase change matrix, comprehensively evaluate the thermodynamic stability of the atomized liquid and determine the stability limit.Define key stability indicators: the Safe Vaporization Temperature Range (SVTR), which is the temperature range within which the main components remain stable or undergo controlled degradation; the Critical Instability Temperature (CIT), which is the lowest temperature at which the system begins to produce significant amounts of harmful substances; and the Optimal Vaporization Temperature Point (OVT), which is the temperature within the safe range that releases the best sensory characteristics. Calculate these indicators based on phase change matrix data. For example, for a certain formulation, the SVTR is 235 to 265 °C, the CIT is 282 °C, and the OVT is 255 °C. Calculate the Thermal Stability Coefficient (TSC) for each component, defined as the percentage of temperature points within the SVTR at which the component remains in a stable state (phase change state ≤ 2). For example, the TSC of glycerol = 0.75, indicating that the component remains relatively stable at 75% of the temperature points within the SVTR. Analyze the thermal stability synergistic effects between different components. Some components may exhibit higher or lower stability in the mixed state than when present alone. For example, experimental data show that adding 5% glycerol ester can increase the T-onset of propylene glycol by 8 °C, showing a positive synergistic effect. Generate thermodynamic stability data, including the safe temperature range of the atomized liquid, the stability coefficients of each component, the unstable critical points, and the synergistic effect parameters between components, laying the foundation for subsequent thermodynamic analysis and sensory prediction.

[0021] Preferably, perform a component-temperature interactive thermodynamic analysis based on the thermodynamic stability data to generate a component thermodynamic reaction map, including: Extract multi-dimensional thermodynamic parameters from the thermodynamic stability data to generate a set of thermodynamic parameters including the rate of change of Gibbs free energy, the activation energy distribution, and the entropy change gradient; Construct a potential energy field of component interactions based on the set of thermodynamic parameters to generate potential energy data of component interactions; Perform a temperature sensitivity grading process on the potential energy data of component interactions to generate temperature sensitivity grading data; Analyze the non-linear thermokinetic behavior based on the temperature sensitivity grading data to generate non-linear thermokinetic behavior data including the critical phase change point, the reaction path weight, and the energy barrier height; Analyze the temperature-component synergistic effect on the non-linear thermokinetic behavior data to generate a temperature-component synergistic effect matrix; Perform a microscopic thermodynamic equilibrium state analysis based on the temperature-component synergistic effect matrix to generate microscopic equilibrium state data; Trace the thermodynamic reaction path based on the microscopic equilibrium state data to generate a reaction path network map; Perform a multi-level information fusion process on the reaction path network map and the non-linear thermokinetic behavior data to generate a component thermodynamic reaction map.

[0022] In the embodiments of the present invention, key thermodynamic parameters are extracted from thermodynamic stability data to construct a thermodynamic description framework for the pyrolysis process of the atomized liquid. The change rate of Gibbs free energy of each main component at each temperature point is calculated to characterize the temperature sensitivity and the spontaneous change trend of the reaction. For example, in the temperature range of 250 to 260 °C, the change rate of Gibbs free energy of glycerol is -2.3 kJ / (mol·K), indicating that the increase in temperature significantly promotes the spontaneity of its decomposition reaction. By measuring the reaction rates at different temperatures, the activation energy distribution of each main reaction is calculated in combination with the Arrhenius equation, and an activation energy spectrum is plotted. According to the van't Hoff equation, the change of the reaction equilibrium constant at different temperatures is calculated to evaluate the influence of temperature on the equilibrium. A set of thermodynamic parameters is generated, which contains comprehensive thermodynamic description parameters, providing a theoretical basis for subsequent analysis. Based on the extracted set of thermodynamic parameters, a potential energy field model describing the interactions between the components in the atomized liquid is constructed. Molecular simulation techniques (such as molecular dynamics simulation or Monte Carlo simulation) are used to calculate the interaction energies between the components in the mixed system, including hydrogen bond interactions, van der Waals forces, electrostatic interactions, etc. For example, through simulation calculations, it is known that the average interaction energy between glycerol and aromatic compounds is -18.5 kJ / mol (a negative value indicates an attractive interaction). A component-component interaction matrix is constructed to quantify the interaction strength between each pair of components and identify strong interaction pairs (interaction energy < -25 kJ / mol) and weak interaction pairs (interaction energy > -10 kJ / mol). The influence of temperature on the interaction potential energy is analyzed, and the change rate of the interaction energy with temperature is calculated. For example, the interaction energy between nicotine and propylene glycol decreases at a rate of 0.15 kJ / (mol·°C) in the temperature range of 230 to 290 °C, indicating that the increase in temperature weakens the interaction between them. Based on quantum chemical calculations, the bond dissociation energy and formation energy of key chemical bonds are analyzed to predict the dominant reaction path during the pyrolysis process. Component interaction potential energy data is generated to describe the energy relationship network and temperature dependence between the components in the atomized liquid. The temperature sensitivity of the component interaction potential energy data is evaluated, and each component is classified according to its sensitivity to temperature during the pyrolysis behavior. The temperature sensitivity classification criteria are set as follows: Level 1 (low sensitivity): In the temperature change range of 10 °C, the pyrolysis rate change < 10%; Level 2 (medium sensitivity): In the temperature change range of 10 °C, the pyrolysis rate change range is (10%, 30%]; Level 3 (high sensitivity): In the temperature change range of 10 °C, the pyrolysis rate change range is (30%, 50%]; Level 4 (extremely high sensitivity): In the temperature change range of 10 °C, the pyrolysis rate change > 50%. Each main component is evaluated. For example, the temperature sensitivity of propylene glycol in the temperature range of 240 to 250 °C is Level 3 (high sensitivity), while it drops to Level 1 (low sensitivity) in the temperature range of 270 to 280 °C, indicating that its decomposition is almost complete at high temperatures.Calculate the overall temperature sensitivity index (TSI) for each temperature point, taking the sensitivity of each component as the weighted average. For example, at 245 °C, the overall TSI = 2.8, indicating that the atomized liquid at this temperature point is relatively sensitive to temperature changes. Analyze the transition points of temperature sensitivity, which usually correspond to the starting temperatures of key thermochemical reactions. Generate temperature sensitivity classification data to provide a basis for precisely controlling the atomization temperature. Based on the temperature sensitivity classification data, deeply analyze the non-linear thermokinetic behavior during the pyrolysis process of the atomized liquid. Identify the critical phase transition points on the temperature-reaction rate curve, which represent the mutation of the reaction mechanism or the conversion of the dominant reaction. For example, in a certain formulation, 260 °C is detected as the key critical point. Below this temperature, unimolecular decomposition reactions mainly occur, and above this temperature, free radical chain reactions are dominant. Through kinetic modeling, calculate the weight coefficients of different reaction paths to quantify the contribution of each path to the overall pyrolysis process. For example, at 270 °C, the decomposition paths of propylene glycol include: dehydration reaction (weight 0.45), C-C bond cleavage (weight 0.30), and oxidation reaction (weight 0.25). Determine the height of the energy barrier of the key reaction to predict the influence of temperature changes on reaction selectivity. For example, the energy barrier of the propylene glycol dehydration reaction is 112 kJ / mol, while the energy barrier of C-C bond cleavage is 125 kJ / mol, indicating that the dehydration reaction is more dominant at low temperatures. Analyze the non-linear characteristics of reaction kinetics, such as autocatalysis, reaction inhibition effect, and induction period phenomenon. Generate non-linear thermokinetic behavior data, including critical phase transition points, reaction path weights, and energy barrier heights, to reveal the complex kinetic characteristics of the pyrolysis of the atomized liquid. Analyze the synergistic effect between temperature changes and component interactions, and study the synergistic behavior of different component combinations under specific temperature conditions. Construct a temperature-component synergistic effect matrix, where the rows represent different temperature points, the columns represent component pairs, and the matrix elements represent the synergistic coefficients. The synergistic coefficient is defined as the ratio of the actual reaction rate in the mixed state to the sum of the reaction rates when present alone. A value greater than 1 indicates positive synergy (promotion), and a value less than 1 indicates negative synergy (inhibition). For example, the synergistic coefficient of glycerol and glyceride at 255 °C is 1.35, indicating that the decomposition rate of glycerol increases by 35% when the two are mixed. Through experiments and modeling, determine the temperature range and component combination with the most significant synergistic effect. For example, it is found that nicotine and certain aldehydes exhibit the strongest negative synergistic effect (synergistic coefficient 0.72) in the range of 265 to 275 °C, which can reduce the generation of aldehydes. Analyze the molecular mechanism of the synergistic effect, such as free radical capture, reaction intermediate stabilization, etc. Generate the temperature-component synergistic effect matrix to provide a basis for formulation optimization. Based on the temperature-component synergistic effect matrix, analyze the thermodynamic equilibrium state of each component in the atomized liquid at the microscale. For each temperature point, calculate the total free energy and chemical potential distribution of the system to determine the thermodynamic equilibrium conditions.For example, at 258 °C, the total free energy of the system reaches a local minimum of -1245 kJ / mol, indicating that the system tends to a stable state at this temperature. Analyze the activities and fugacity coefficients of each component at different temperatures to evaluate the degree of deviation from thermodynamic equilibrium. For example, in the high-temperature region (>300 °C), the activity coefficient of nicotine increases from 1.15 to 1.85, indicating that its thermodynamic behavior deviates more and more from the ideal solution. Use phase equilibrium theory to analyze the distribution behavior of each component in the liquid and gas phases, and calculate the variation of the distribution coefficient with temperature. For example, the gas-liquid distribution coefficient of vanillin is 2.8 at 240 °C and increases to 4.5 at 280 °C, indicating that high temperature promotes its transfer to the gas phase. Generate micro-equilibrium state data to describe the thermodynamic equilibrium characteristics of each component in the atomized liquid at different temperatures. Based on the micro-equilibrium state data, trace the thermodynamic reaction paths of each component in the atomized liquid and construct a complete reaction network. Use reaction flow analysis techniques to identify the main reaction paths and minor paths, and calculate the fluxes of each path. For example, the analysis shows that propylene glycol mainly degrades through the path of first dehydrating to form propionaldehyde and then oxidizing to propionic acid, and the flux of this path accounts for 65% of the total degradation flux. Draw a reaction path diagram, connecting all the main components and intermediate products into a network, with the arrow width indicating the magnitude of the reaction flux. For example, the path diagram of glycerol shows that the path flux of decomposing into formaldehyde is 3.2 times that of decomposing into propionaldehyde. Mark the thermodynamically controlled paths (determined by the change in Gibbs free energy) and kinetically controlled paths (determined by the activation energy), and analyze the influence of temperature on path selection. For example, in the low-temperature region (<250 °C), the decomposition of glycerol is mainly kinetically controlled, preferentially generating products with low activation energy; while in the high-temperature region (>280 °C), it is mainly thermodynamically controlled, preferentially generating products with a smaller change in Gibbs free energy. Generate a reaction path network map to visually display the reaction transformation relationship during the pyrolysis process of the atomized liquid. Combine the reaction path network map with the non-linear thermokinetic behavior data, and construct a comprehensive component thermodynamic reaction map through multi-level information fusion technology. Adopt a hierarchical data structure to correlate the macroscopic reaction behaviors (such as pyrolysis rate, product distribution) with the microscopic reaction mechanisms (such as bond-breaking mechanism, radical reaction). Integrate four dimensions of information in the map: temperature dimension, component dimension, reaction type dimension, and time dimension to form a multi-dimensional visual display. Use color coding to represent the reaction types (such as blue for hydrolysis reactions and red for oxidation reactions), the line thickness to represent the reaction rate, and the node size to represent the substance content. Integrate the thermodynamic stability data and mark the safe temperature range and critical instability points for each component. For example, mark the upper limit of the safe temperature of glycerol on the map as 255 °C, and harmful aldehyde substances will be rapidly generated above this temperature. Through an interactive design, allow on-demand viewing of detailed information on specific temperatures, specific components, or specific reaction types. Generate a component thermodynamic reaction map, which comprehensively displays the thermodynamic behavior and reaction network of the atomized liquid over the entire temperature range, providing a theoretical basis for subsequent sensory response prediction and formulation optimization.

[0023] Preferably, step S4 includes: Performing component diffusion kinetics analysis based on the partition atomization characteristic spectrum to generate component diffusion quantification data; Performing multi-dimensional sensory influence factor evaluation based on the component thermodynamics reaction map to generate sensory influence factor data, and designing a perception response decision node through the sensory influence factor data; Constructing a neural perception mapping for associating sensory experience and physiological response for the perception response decision node and the component thermodynamics reaction map based on a preset sensory feature extraction algorithm to obtain a sensory response prediction engine; Transmitting the component diffusion quantification data to the sensory response prediction engine in real time for intelligent analysis of personalized sensory experience to generate sensory response characteristic data; Performing intelligent evaluation of the global atomization liquid characteristics based on the sensory response characteristic data to generate an atomization liquid quality evaluation report; Performing safety and taste balance optimization processing on the original pyrolysis data according to the atomization liquid quality evaluation report to generate an optimized formula recommendation, and performing precise preparation of the atomization liquid through the optimized formula recommendation.

[0024] In the embodiments of the present invention, based on the partition atomization characteristic spectrum, the diffusion kinetic characteristics of the atomized liquid components in the gas phase are analyzed. The pulsed laser-induced fluorescence (PLIF) technique is used to measure the spatial distribution and diffusion rate of flavor molecules in the gas phase. The experimental conditions are set as follows: the atomization temperature ranges from 200 to 350 °C, with an interval of 10 °C; the measurement distance ranges from 0 to 10 cm, with an interval of 1 cm; the sampling time ranges from 0 to 5 s, with an interval of 0.5 s. According to the measurement data, the diffusion coefficient D of each key component is calculated, and this coefficient characterizes the rate of molecular diffusion in the gas phase. For example, at 280 °C, the diffusion coefficient of nicotine is 0.082 cm² / s, and the diffusion coefficient of vanillin is 0.065 cm² / s. The Fick diffusion equation is used to simulate the concentration distribution curves of each component at different temperatures and times. For example, the simulation shows that nicotine can diffuse to 4.5 cm within 2 seconds, while vanillin can only diffuse to 3.8 cm. The half-life distance (the distance required for the concentration to drop to half of the initial value) and the peak time (the time required to reach the maximum concentration) of each component are calculated. For example, the half-life distance of nicotine is 3.2 cm, and the peak time is 1.8 s; while the half-life distance of menthol is 2.5 cm, and the peak time is 1.2 s, indicating that the minty feeling is perceived faster but lasts for a shorter time. Combining the atomization characteristic spectrum data, the influence of temperature on the diffusion behavior is analyzed. For example, the experimental data shows that for every 10 °C increase in temperature, the diffusion coefficient of nicotine increases by about 8%, and the diffusion coefficient of vanillin increases by about 12%. Diffusion quantification data of the components are generated, including the diffusion parameters and spatio-temporal distribution characteristics of each component at different temperatures, providing a kinetic basis for the sensory experience analysis. According to the component thermodynamic reaction map, the influence of each component in the atomized liquid on the sensory experience is comprehensively evaluated. Multi-dimensional evaluations are carried out for the five basic tastes (sweet, sour, bitter, salty, umami), three-dimensional aromas (top note, middle note, base note), and tactile characteristics (such as irritation, cold feeling, warm feeling, lubrication feeling). Through human sensory tests and literature data, a database of sensory influence parameters for each component is established. For example, the sweetness influence coefficient of vanillin is 0.85 (full score 1.0), the top note aroma influence coefficient is 0.65, and the irritation influence coefficient is 0.15; the cold feeling influence coefficient of menthol is 0.92, and the middle note aroma influence coefficient is 0.78. Multiple regression analysis is used to quantify the non-linear relationship between the concentration of each component and the sensory intensity, and a mathematical model is established. For example, the sweetness perception model of a certain flavor component is: sweetness = 0.85×log(concentration) + 0.23, indicating that the perception intensity is proportional to the logarithm of the concentration. The sensory synergistic and antagonistic effects between components are analyzed. For example, the experimental data shows that when menthol is combined with certain sweeteners, the cold feeling is enhanced by 25%, while when combined with certain sour agents, the cold feeling is reduced by 15%. According to the evaluation results, sensory influence factor data are generated, and corresponding perception response decision nodes are designed. Each decision node represents a specific sensory attribute (such as "cold feeling", "sweetness", "throat feeling", etc.), and includes the influence weights of related components and threshold judgment rules.For example, the rule of the "cool feeling" decision node is: if the menthol concentration > 5 ppm and the ambient temperature < 28 °C, then the cool feeling score is 0.8 times the menthol concentration coefficient plus 0.2 times the temperature coefficient. Based on the preset sensory feature extraction algorithm and the designed perceptual response decision nodes, a neural perception mapping model connecting the e-liquid components and the human sensory experience is constructed. Using deep learning architectures such as multi-layer perceptron (MLP) or convolutional neural network (CNN), with the component data in the component thermodynamic reaction map as the input and the sensory experience score as the output, the neural network model is trained. Structure design of the neural network model: The input layer contains n neurons, corresponding to the concentrations of n key components and their interaction parameters; the hidden layer adopts a three-layer structure, with each layer containing 64, 32, and 16 neurons respectively, and the ReLU activation function is used; the output layer contains m neurons, corresponding to the intensity scores of m sensory attributes. Model training uses historical sensory evaluation data, including the component data of 300 different formulations of e-liquids at different temperatures and the corresponding human sensory scores. 80% of the data is used for training and 20% for validation. The mean squared error (MSE) is used as the loss function, and the Adam optimizer is used for parameter optimization. During the training process, emphasis is placed on the recognition ability of components with low concentrations but high sensory impact factors to ensure that the model can accurately predict the sensory contributions of trace components. After training and validation, the prediction accuracy of the model on the test set reaches 87%, and the average sensory intensity prediction error is less than 0.15 (with a full score of 1.0). After completing the construction of the neural perception mapping, it is integrated into a sensory response prediction engine, which can predict the corresponding sensory experience characteristics in real time based on the input component data. The generated component diffusion quantification data is input into the sensory response prediction engine in real time for personalized sensory experience analysis. Data transmission uses cache technology to ensure a response speed of milliseconds to meet the requirements of real-time analysis. The prediction engine calculates the changes in sensory experience over time series based on the input component diffusion dynamic data, combined with environmental parameters such as temperature and air flow. For example, the engine can predict that at an atomization temperature of 280 °C, within 0.5 seconds from the start of inhalation, the cool feeling of mint and the top note aroma are mainly felt, from 1 to 2 seconds, the sweetness and middle note aroma are felt, and from 2 to 4 seconds, the tobacco aroma and aftertaste are felt. For different user groups (such as factors like age, gender, smoking history, etc.), the engine can adjust the prediction parameters to provide personalized sensory experience predictions. For example, for long-term smokers, the engine will appropriately increase the threshold parameters of certain flavors to reflect the changes in their sensory sensitivity. Generate sensory response feature data, including the sensory intensity curve in the time dimension, the sensory attribute distribution map, and the satisfaction prediction index, providing a basis for the quality evaluation of e-liquids. Based on the sensory response feature data, a comprehensive quality evaluation of e-liquids is carried out, taking into account aspects such as safety, sensory experience, and use stability.Design an atomization liquid quality assessment system, which includes the following core indicators: Safety Index (SI): Evaluate the generation of harmful substances during atomization, with a full score of 100 points. <60 points indicates unsafe, 60 - 80 points indicates basically safe, and >80 points indicates highly safe; Sensory Balance (SB): Evaluate the coordination of various sensory attributes, with a full score of 100 points. <60 points indicates uncoordinated, 60 - 80 points indicates basically coordinated, and >80 points indicates highly coordinated; Durability and Stability (DS): Evaluate the duration and stability of the sensory experience, with a full score of 100 points; Thermal Sensitivity (TS): Evaluate the sensitivity of the atomization liquid to temperature changes, and the lower the score, the better; Overall Satisfaction (OS): The weighted average of the above indicators is used as the overall evaluation. Score the current atomization liquid sample according to the predefined scoring rules. For example, the score of a certain sample is: SI = 75 (basically safe), SB = 82 (highly coordinated), DS = 68 (moderate durability), TS = 42 (moderately sensitive), OS = 72 (good). Analyze the scoring results to identify the key aspects that need improvement. For example, the low TS score of the above sample indicates that it is more sensitive to temperature changes, and a stabilizer needs to be added to the formula to improve heat resistance. Generate an atomization liquid quality assessment report, which includes detailed scoring data, improvement suggestions, and recommended applicable temperature ranges. According to the atomization liquid quality assessment report, propose optimization strategies for the safety risks and sensory balance problems in the original pyrolysis data. For safety issues, analyze the generation temperature and precursor substances of harmful substances (such as aldehydes and polycyclic aromatic hydrocarbons) in the pyrolysis data, and propose methods to reduce risks. For example, if it is found that a large amount of glycerol is converted into acrolein above 280°C, it is recommended to reduce the glycerol content and add antioxidants, or control the recommended atomization temperature below 270°C. For sensory balance problems, analyze the proportional relationship between various sensory attributes and propose adjustment plans. For example, if the cold feeling is too strong and the sweetness is insufficient, it is recommended to reduce the proportion of menthol and increase the proportion of vanillin or maltitol. According to the thermodynamic reaction diagram, predict the effects of various optimization plans and select the optimal plan. For example, through simulation prediction, adding 2% antioxidant to the basic formula and adjusting the glycerol / propylene glycol ratio from 30:70 to 20:80 can increase the SI by 12 points while only slightly affecting the sensory balance (SB decreases by 3 points). Generate an optimized formula recommendation, which includes specific ingredient adjustment ratios, recommended atomization temperature ranges, and expected effects. According to the optimization recommendations, adjust the atomization liquid formula to achieve precise preparation. Implement the formula adjustment, such as adding antioxidants (such as vitamin E, concentration 0.5% - 2%) to reduce aldehyde generation; adding stabilizers (such as glycerol esters, concentration 1% - 5%) to improve thermal stability; adjusting the ratio of the basic solvent (such as adjusting the glycerol / propylene glycol ratio from 30:70 to 20:80) to optimize safety and taste. Adjust the spice ratio, such as reducing the menthol content (from 1.2% to 0.8%) to reduce the overly strong cold feeling; increasing the vanillin content (from 0.5% to 0.7%) to enhance the sweet taste.Control the nicotine concentration within a safe range and select the most stable form of nicotine salt according to thermodynamic data. Conduct small-batch production tests on the adjusted formula to verify whether its actual effect meets expectations. If the effect is satisfactory, complete the precise preparation of the e-liquid and form the final formula; if there are still deficiencies, further fine-tune the formula until the target quality standard is achieved.

[0025] Preferably, based on a preset sensory feature extraction algorithm, construct a neural perception mapping for associating sensory experience and physiological response for the perception response decision node and the component thermodynamic reaction map, to obtain a sensory response prediction engine, including: Design a neural perception mapping architecture for associating sensory experience and physiological response for the perception response decision node based on a preset sensory feature extraction algorithm, and generate a sensory response prediction framework; Conduct an individual perception difference analysis based on the component thermodynamic reaction map to generate individual perception difference data; Perform a relevance weight learning process for associating sensory experience and physiological response on the sensory response prediction framework according to the individual perception difference data to generate a sensory response prediction engine.

[0026] In the embodiment of the present invention, based on a pre-designed sensory feature extraction algorithm, construct a neural perception mapping architecture connecting the e-liquid components and the human sensory experience. The sensory feature extraction algorithm includes the following core components: a multi-dimensional sensory feature vectorization module that encodes various sensory attributes (such as sweetness, acidity, coolness, irritation, etc.) into standardized numerical vectors; a temporal sensory experience model that describes the evolution law of sensory experience over time; a component-sensory association matrix that records the contribution weights of each component to different sensory attributes. Design a hierarchical neural network architecture, including: an input layer: receiving parameters such as component concentration, temperature, time, etc., and the number of neurons is determined according to the input parameters (usually 50 to 100); a feature extraction layer: adopting a multi-layer convolutional network (CNN) structure to extract the implicit features of the component combination, including 3 convolutional layers, with 64, 128, and 64 convolutional kernels in each layer respectively; a sensory mapping layer: adopting a fully connected network structure to map the extracted features to the sensory space, including 3 fully connected layers, with 128, 64, and 32 neurons in each layer respectively; a temporal prediction layer: adopting a recurrent neural network (RNN) or LSTM structure to predict the temporal evolution of sensory experience, including 2 layers of LSTM units, with 16 neurons in each layer; an output layer: generating multi-dimensional sensory experience prediction results, including the same number of neurons as the preset sensory attributes. For each perception response decision node, design a dedicated neural network branch to form a multi-task learning architecture. For example, the "coolness" decision node is connected to a dedicated hidden layer containing 32 neurons, while the "throat feeling" decision node is connected to another dedicated hidden layer containing 24 neurons. The loss function of the network is designed as a multi-task weighted loss: , where is the prediction loss of the th sensory attribute, The sensory weights of the sensory attributes are initially set to be evenly distributed. Generate a sensory response prediction framework, including a complete neural network structure, parameter initialization scheme, and training strategy, to provide the infrastructure for subsequent personalized training. Analyze the data in the compositional thermodynamic reaction map to study the differences in sensory perception among individuals. According to the data on sensory receptor gene polymorphisms, the population is divided into different types of perception abilities. For example, according to the TAS2R38 genotype, the population is divided into "taste-sensitive type" (able to strongly perceive the bitterness of PTC) and "taste-insensitive type" (weak perception of the bitterness of PTC); according to the polymorphism of the TRPM8 gene, the population is divided into "high cold-sensation type" and "low cold-sensation type". Design and execute a sensory difference test experiment, inviting 100 volunteers with different backgrounds (age, gender, smoking history, etc.) to conduct blind tests and score standard e-liquid samples. The test contents include: basic taste recognition test (recognition thresholds for the five basic tastes of sweet, sour, bitter, salty, and umami); special sensation test (perception intensity of cold sensation, heat sensation, and irritation); aroma recognition test (recognition ability of top notes, middle notes, and base notes of aroma); overall preference score (preference score for different vaping temperatures and composition combinations). Analyze the test data and calculate the difference coefficients of each population in different sensory dimensions. For example, the perception intensity of menthol in the "high cold-sensation type" population is 45% higher than that in the "low cold-sensation type" population; the taste threshold of nicotine for long-term smokers is 2.8 times higher than that for non-smokers. According to the results of the difference analysis, divide the users into 5 to 10 typical perception pattern groups and establish a standardized set of perception parameters for each group. Generate individual perception difference data, including the sensory thresholds, perception intensity coefficients, and preference patterns of different types of users, to provide a basis for the personalized training of the neural network. According to the individual perception difference data, conduct targeted training on the pre-designed sensory response prediction framework to learn the sensory-physiological response association patterns of different user groups. Design a weight learning strategy to adjust the association weights of the neural network for different user groups. For example, for "high cold-sensation type" users, increase the weight of the menthol-cold sensation pathway; for long-term smokers, reduce the weight of the nicotine-irritation pathway. Use the supervised learning method to train the network, using the labeled data collected from the sensory tests as the training set. Composition of the training set: The sensory score data of each user group for 50 different formulations at 5 different temperature points, forming a total of 1250 groups of training samples. Each group of samples contains input features (ingredient concentration, temperature, etc.) and output labels (sensory scores). Training parameter settings: batch size = 32, initial learning rate = 0.001, adopt a learning rate decay strategy (reduce by 10% every 50 epochs), maximum number of training epochs = 500, and the early stopping condition is that the validation set loss has not improved for 10 consecutive epochs. For each user group, train a dedicated set of weights separately to form a multi-model integration system. Integrate the prediction capabilities of the group models to achieve personalized prediction for specific users.For example, the system can automatically select or fuse the most suitable prediction model based on the user's basic information (age, gender, smoking history, etc.). After training, the model is evaluated, and the average prediction accuracy rate on the test set reaches 91%, and the average prediction error of the sensory intensity is reduced to 0.08 (with a full score of 1.0). A complete sensory response prediction engine is generated, which can receive the component thermodynamics reaction data and diffusion kinetics data as inputs and output personalized sensory experience prediction results, including the change of sensory intensity in the time series, the coordination score of each sensory attribute, and the overall satisfaction prediction.

[0027] In this embodiment, through multi-parameter atomization condition sampling and temperature gradient deconstruction analysis, a comprehensive characterization of the pyrolysis components generated during the atomization process is achieved. The complex pyrolysis process can be transformed into a visual multi-dimensional pyrolysis spectrogram, thereby enhancing the monitoring accuracy of the changes in the atomized liquid components. Through the construction of a trace marker feature library and the generation of a dynamic component evolution spectrum, ultra-trace active components can be effectively identified and tracked, realizing the accurate quantitative analysis of bioactive substances, ensuring the comprehensive monitoring of trace components that may have a potential impact on human health, greatly improving the accuracy of the safety assessment of the atomized liquid, and avoiding the problem that trace components are easily overlooked in traditional analysis methods. Through the partition boundary treatment of the atomization process and the component-temperature interaction thermodynamics analysis, the system can comprehensively master the thermodynamic behavior characteristics of the atomized liquid components in different temperature ranges, cope with the component change rules under different atomization conditions, and improve the system's analytical ability and accuracy for the complex thermodynamic changes in the atomization process through the analysis of the thermal degradation critical point and the analysis of the molecular recombination activity. The introduction of the reaction path network map and the non-linear thermodynamics behavior analysis realizes the multi-level information fusion of the complex reaction system, providing a theoretical basis for the component safety assessment. Through the analysis of the component diffusion kinetics and the construction of the sensory response prediction engine, the diffusion characteristics of the atomized components are analyzed in real time, the possible sensory experience is predicted, the influence of different component combinations on the user's senses is dynamically evaluated, ensuring a high-quality sensory experience can still be achieved when the atomization conditions change. According to the automated evaluation analysis of individual perception difference data, a personalized quality assessment report can be automatically generated and the formula can be optimized according to the real-time atomization characteristics, improving the system's ability to adapt to different user needs, significantly enhancing the user experience, and ensuring product safety.

[0028] Embodiment 2

[0029] Please refer to Figure 2 as shown. For the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. A smart analysis system based on atomized liquid components is provided, including: Multi-dimensional thermal spectrum construction module: Perform multi-parameter atomization condition sampling on the atomized liquid to obtain the original pyrolysis data; perform temperature gradient deconstruction analysis according to the original pyrolysis data to generate a multi-dimensional pyrolysis spectrogram; Dynamic Trajectory Analysis Module: Based on the multi-dimensional pyrolysis spectrum, it accurately identifies trace markers and generates a trace marker feature library; based on the trace marker feature library, it analyzes the multi-dimensional component evolution trajectory and generates a dynamic component evolution spectrum; Thermodynamic Interaction Modeling Module: Based on the dynamic component evolution spectrum, it divides and demarcates the atomization process to generate a partitioned atomization feature spectrum; based on the partitioned atomization feature spectrum, it conducts component-temperature interaction thermodynamic analysis and generates a component thermodynamic reaction map; Intelligent Evaluation and Optimization Module: Based on the partitioned atomization feature spectrum, it analyzes the component diffusion kinetics to generate component diffusion quantification data; based on the component thermodynamic reaction map, it constructs a sensory response prediction engine to obtain the sensory response prediction engine; it transmits the component diffusion quantification data to the sensory response prediction engine in real time for intelligent analysis of personalized sensory experience and generates sensory response feature data; based on the sensory response feature data, it conducts intelligent evaluation of the global atomized liquid characteristics and generates an atomized liquid quality evaluation report; based on the atomized liquid quality evaluation report, it optimizes the balance between safety and taste of the original pyrolysis data, generates an optimized formula suggestion, and performs precise preparation of the atomized liquid through the optimized formula suggestion.

[0030] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent substitution on some of the technical features. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0031] It should be noted that in this article, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0032] In the description of the present invention, it should be understood that terms such as "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0033] In the description of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0034] In the description of the present invention, the meaning of "several" is one or more, and the meaning of "a large number" is two or more.

[0035] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0036] For the formulas in this specification, the dimensional quantities are removed and only the numerical values are calculated. The formula is obtained by collecting a large amount of data and performing software simulation to get a formula closest to the actual situation. The preset parameters and threshold selection in the formula are set by those skilled in the art according to the actual situation.

[0037] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for intelligent analysis of atomized liquid components, characterized in that: include: Step S1: sampling the atomized liquid under multi-parameter atomization conditions to obtain original pyrolysis data; Perform temperature gradient deconstruction analysis based on the original pyrolysis data to generate a multi-dimensional pyrolysis spectrum; Step S2: accurately identifying trace markers according to the multi-dimensional pyrolysis spectrum to generate a trace marker feature library; performing multi-dimensional component evolution trajectory analysis based on the trace marker feature library to generate a dynamic component evolution spectrum; Step S3: performing partition demarcation processing on the atomization process according to the dynamic component evolution spectrum to generate a partition atomization characteristic spectrum; Conduct component-temperature interaction thermodynamic analysis based on the partitioned atomization characteristic spectrum to generate a component thermodynamic reaction map; Step S4: Analyze the component diffusion dynamics according to the partition atomization characteristic spectrum to generate component diffusion quantitative data; A sensory response prediction engine is constructed based on the thermodynamic reaction map of the ingredients to obtain a sensory response prediction engine; the ingredient diffusion quantification data is transmitted to the sensory response prediction engine in real time to perform intelligent analysis of personalized sensory experience and generate sensory response characteristic data; Based on the sensory response characteristic data, an intelligent evaluation of the global atomized liquid characteristics is performed to generate an atomized liquid quality evaluation report. Based on the atomized liquid quality evaluation report, the original pyrolysis data is optimized for safety and taste balance, and optimized formula recommendations are generated. The atomized liquid is then precisely prepared through the optimized formula recommendations.

2. The intelligent analysis method based on atomized liquid composition according to claim 1 is characterized in that: Step S1 includes: Perform dynamic sampling of atomized liquid under multi-parameter atomization conditions to generate raw pyrolysis data; Perform pyrolysis product type classification processing according to the original pyrolysis data to generate pyrolysis product type characteristic data; Perform multi-stage pyrolysis kinetic characterization based on the pyrolysis product type characteristic data to generate pyrolysis kinetic characterization data; Temperature gradient deconstruction analysis is performed based on the pyrolysis kinetic characterization data to generate a multidimensional pyrolysis spectrum.

3. The method for intelligent analysis of atomized liquid components according to claim 1, characterized in that: Step S2 includes: Extract pyrolysis intensity features based on multi-dimensional pyrolysis spectra to generate pyrolysis intensity spectrum data; Extract molecular fragment spectrum features based on multi-dimensional pyrolysis spectra to generate molecular fragment spectrum data; Using a preset multi-dimensional spectrum deconvolution algorithm to perform initial trace marker identification analysis on the pyrolysis intensity spectrum data and the molecular fragmentation spectrum data, to generate initial trace marker data; Perform low signal-to-noise ratio signal enhancement processing based on initial trace marker data to generate a trace marker feature library; The multi-dimensional component evolution trajectory is analyzed based on the trace marker feature library to generate a dynamic component evolution spectrum.

4. The method for intelligent analysis of atomized liquid components according to claim 3, characterized in that: The method of performing low signal-to-noise ratio signal enhancement processing based on the initial trace marker data to generate a trace marker feature library includes: Performing a bioactivity threshold analysis based on the initial trace marker data to generate bioactivity threshold data, wherein the bioactivity threshold data includes physiological response triggering concentration data and a marker specificity index; Perform ultra-trace active ingredient enrichment processing based on physiological response triggering concentration data to generate ultra-trace active ingredient data; Perform bioactive marker correlation processing based on ultra-trace active ingredient data and marker specificity index to generate bioactive marker data; The initial trace marker data are subjected to marker screening processing according to the bioactive marker data to generate a trace marker feature library.

5. The method for intelligent analysis of atomized liquid components according to claim 1, characterized in that: Step S3 includes: According to the dynamic component evolution spectrum, the atomization process is partitioned and demarcated to generate a partitioned atomization characteristic spectrum; Detect the thermodynamic stability limit based on the partitioned atomization characteristic spectrum and generate thermodynamic stability data; The component-temperature interaction thermodynamic analysis is performed based on the thermodynamic stability data to generate the component thermodynamic reaction map.

6. The method for intelligent analysis of atomized liquid components according to claim 5, characterized in that: The thermodynamic stability limit detection is performed according to the partition atomization characteristic spectrum to generate thermodynamic stability data, including: Perform thermal degradation critical point analysis based on the partitioned atomization characteristic spectrum to generate thermal degradation critical point data; Perform molecular recombination activity analysis based on the partitioned atomization characteristic spectrum to generate molecular recombination activity data; The composition-temperature phase change matrix is ​​constructed based on the thermal degradation critical point data and the molecular reorganization activity data to generate the composition-temperature phase change matrix; Thermodynamic stability limits are detected based on the composition-temperature phase transition matrix to generate thermodynamic stability data.

7. The method for intelligent analysis of atomized liquid components according to claim 5, characterized in that: The component-temperature interaction thermodynamic analysis is performed according to the thermodynamic stability data to generate a component thermodynamic reaction map, including: Extract multi-dimensional thermodynamic parameters from thermodynamic stability data to generate a set of thermodynamic parameters including Gibbs free energy change rate, activation energy distribution, and entropy change gradient; Constructing the component interaction potential energy field based on the thermodynamic parameter set to generate component interaction potential energy data; Performing temperature sensitivity classification processing on component interaction potential energy data to generate temperature sensitivity classification data; Nonlinear thermodynamic behavior analysis is performed based on temperature sensitivity classification data to generate nonlinear thermodynamic behavior data including critical phase transition points, reaction path weights, and energy barrier heights; Conduct temperature-component synergistic effect analysis on nonlinear thermodynamic behavior data and generate a temperature-component synergistic effect matrix; Conduct micro-thermodynamic equilibrium analysis based on the temperature-component synergistic effect matrix to generate micro-equilibrium data; Tracing thermodynamic reaction paths based on microscopic equilibrium data to generate reaction path network maps; The reaction path network map and nonlinear thermodynamic behavior data are subjected to multi-level information fusion processing to generate a component thermodynamic reaction map.

8. The method for intelligent analysis of atomized liquid components according to claim 1, characterized in that: The step S4 comprises: Analyze the component diffusion dynamics based on the partitioned atomization characteristic spectrum to generate component diffusion quantitative data; Conduct multi-dimensional sensory impact factor evaluation based on the thermodynamic reaction map of the ingredients, generate sensory impact factor data, and design perceptual response decision nodes based on the sensory impact factor data; Based on the preset sensory feature extraction algorithm, a neural perception mapping is constructed to associate sensory experience with physiological response for the sensory response decision nodes and the component thermodynamic response map, so as to obtain a sensory response prediction engine; The component diffusion quantification data is transmitted in real time to the sensory response prediction engine for intelligent analysis of personalized sensory experience to generate sensory response characteristic data; Perform intelligent evaluation of global atomized liquid characteristics based on sensory response characteristic data and generate an atomized liquid quality evaluation report; According to the atomized liquid quality evaluation report, the original pyrolysis data is optimized for safety and taste balance, an optimized formula recommendation is generated, and the atomized liquid is precisely prepared through the optimized formula recommendation.

9. The method for intelligent analysis of atomized liquid components according to claim 8, characterized in that: The sensory response prediction engine is obtained by constructing a neural perception mapping of the sensory response decision nodes and the component thermodynamic reaction map based on a preset sensory feature extraction algorithm to associate sensory experience with physiological response, including: Based on the preset sensory feature extraction algorithm, a neural perception mapping architecture is designed to associate sensory experience with physiological response at the perception response decision node, generating a sensory response prediction framework; Conduct individual perception difference analysis based on the component thermodynamic reaction spectrum to generate individual perception difference data; According to the individual perception difference data, the sensory response prediction framework is processed by learning the correlation weights between sensory experience and physiological response to generate a sensory response prediction engine.

10. A system for intelligent analysis of atomized liquid components, which is used to implement the method for intelligent analysis of atomized liquid components according to any one of claims 1 to 9, characterized in that: include: Multi-dimensional thermal spectrum building module: multi-parameter atomization condition sampling of the atomized liquid to obtain raw pyrolysis data; Perform temperature gradient deconstruction analysis based on the original pyrolysis data to generate a multi-dimensional pyrolysis spectrum; Dynamic trajectory analysis module: accurately identify trace markers based on multi-dimensional pyrolysis spectra and generate a trace marker feature library; perform multi-dimensional component evolution trajectory analysis based on the trace marker feature library and generate a dynamic component evolution spectrum; Thermodynamic interactive modeling module: partition and demarcate the atomization process according to the dynamic component evolution spectrum to generate partition atomization characteristic spectrum; Conduct component-temperature interaction thermodynamic analysis based on the partitioned atomization characteristic spectrum to generate a component thermodynamic reaction map; Intelligent evaluation and optimization module: analyzes the component diffusion dynamics according to the partitioned atomization characteristic spectrum and generates component diffusion quantitative data; A sensory response prediction engine is constructed based on the thermodynamic reaction map of the ingredients to obtain a sensory response prediction engine; the ingredient diffusion quantification data is transmitted to the sensory response prediction engine in real time to perform intelligent analysis of personalized sensory experience and generate sensory response characteristic data; Based on the sensory response characteristic data, an intelligent evaluation of the global atomized liquid characteristics is performed to generate an atomized liquid quality evaluation report. Based on the atomized liquid quality evaluation report, the original pyrolysis data is optimized for safety and taste balance, and optimized formula recommendations are generated. The atomized liquid is then precisely prepared through the optimized formula recommendations.

Images