Analysis method for rapidly evaluating barrier property of lining paper for cigarettes

Abstract

The invention discloses an analysis method for rapidly evaluating the barrier property of lining paper for cigarettes, and relates to the field of performance analysis of the lining paper for cigarettes, and the analysis method comprises the following steps: after coating the surface of the lining paper for cigarettes to be detected with a liquid crystal solution, generating a change curve of a contact angle and a liquid film thickness distribution diagram through water vapor to obtain a nanoscale wetting dynamic data set; determining a characteristic signal range for performing multispectral monitoring on the lining paper for the cigarette according to the nanoscale wetting dynamics data set, performing characteristic extraction to obtain a multispectral characteristic parameter data set, performing dimension reduction on the multispectral characteristic parameter data set, and inputting the multispectral characteristic parameter data set into the constructed comprehensive barrier performance model, according to the method, through principal component analysis dimensionality reduction and support vector machine modeling, complex multi-source data is converted into the reliable barrier performance index, and the problem that a traditional method depends on a single index and data isolation is solved.

Description

Technical Field

[0001] The invention relates to the field of performance analysis of cigarette liner paper, in particular to an analysis method for rapidly evaluating the barrier properties of cigarette liner paper. Background Art

[0002] Cigarette liner paper, a crucial component of cigarette packaging, has a significant impact on the shelf life, aroma retention, and sensory quality of cigarette products through its barrier properties. In recent years, with the growing demand for high-quality packaging materials in the cigarette industry, significant progress has been made in evaluating the barrier properties of cigarette liner paper. Traditional evaluation methods rely primarily on standardized physical tests for water vapor transmission rate (WVTR) and gas chromatography-mass spectrometry-based oxygen transmission rate (OTR) and volatile organic compound (VOC) retention. These methods simulate packaging environments to quantify a material's barrier properties to moisture, oxygen, and aroma components, and are widely used in industrial production and quality control. Furthermore, surface analysis techniques such as scanning electron microscopy (SEM) and atomic force microscopy (AFM) have been introduced to characterize the microscopic morphology and pore structure of cigarette liner paper, thereby revealing the microscopic mechanisms underlying its barrier properties.

[0003] However, there is still room for improvement in the existing evaluation technology for the barrier properties of cigarette liner paper. For example, traditional methods such as ASTM-E96 and GC-MS have long testing cycles and cannot meet the modern cigarette industry's demand for rapid quality testing. They can only evaluate single substances and cannot fully characterize the synergistic barrier properties of moisture, oxygen and aroma components. In addition, although multispectral analysis can detect multiple substances, existing methods mostly stay in a single spectral range, and lack detection sensitivity and data comprehensiveness. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an analytical method for rapidly evaluating the barrier properties of cigarette liner paper to solve the problem that only a single substance can be evaluated and the detection sensitivity and data comprehensiveness are insufficient.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an analytical method for rapidly evaluating the barrier properties of cigarette liner paper, comprising:

[0008] After the liquid crystal solution is applied to the surface of the cigarette liner paper to be tested, water vapor is passed through to generate a contact angle change curve and a liquid film thickness distribution map, thereby obtaining a nanoscale wetting dynamics data set.

[0009] Determine the characteristic signal range of multispectral monitoring of cigarette liner paper based on the nanoscale wetting dynamics data set and perform feature extraction to obtain a multispectral characteristic parameter data set;

[0010] After dimensionality reduction of the multispectral feature parameter dataset, the constructed comprehensive barrier performance model was input to obtain the comprehensive barrier performance index of cigarette liner paper.

[0011] The observation parameters of the polarization optical microscope are optimized by using the comprehensive barrier performance index to generate polarization image sequences and liquid crystal film characteristic datasets. A visualization evaluation model is then constructed based on the comprehensive barrier performance index and the liquid crystal film characteristic dataset to obtain the final barrier performance index.

[0012] Draw multi-dimensional visualization charts and generate analysis reports for cigarette liner paper.

[0013] As a preferred embodiment of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to the present invention, the method comprises the following steps: after coating the surface of the cigarette liner paper to be tested with a liquid crystal solution, water vapor is passed through the surface to generate a contact angle change curve and a liquid film thickness distribution diagram, thereby obtaining a nanoscale wetting dynamics data set.

[0014] A liquid crystal solution is coated on the surface of the cigarette liner paper to be tested to form a liquid crystal film. The liquid crystal film-coated cigarette liner paper sample is fixed on the sample holder of a multifunctional testing device, and the initial surface topology is recorded using an atomic force microscope.

[0015] The permeation chamber was activated, water vapor was introduced, and an atomic force microscope was used to observe in situ the nanoscale liquid film formed by the adsorption of water molecules. The contact angle and thickness of the liquid film were calculated using NanoScope-Analysis, generating a contact angle curve and a liquid film thickness distribution map.

[0016] The initial surface topology, contact angle variation curve, and liquid film thickness distribution map are combined into a nanoscale wetting dynamics dataset.

[0017] As a preferred embodiment of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to the present invention, the method comprises the following steps: determining the characteristic signal range of multispectral monitoring of cigarette liner paper based on the nanoscale wetting dynamics data set and performing feature extraction to obtain a multispectral characteristic parameter data set.

[0018] Read the nanoscale wetting dynamics data set for statistical analysis, generate a table of barrier performance-related parameters, and determine the characteristic signal range of multispectral monitoring based on the table of barrier performance-related parameters;

[0019] According to the determined characteristic signal range of multispectral monitoring, the multispectral equipment is used to extract the terahertz wave characteristic peak displacement, infrared absorption peak intensity change and ultraviolet absorption peak area, and combine them to obtain a multispectral characteristic parameter data set.

[0020] As a preferred embodiment of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to the present invention, the multispectral characteristic parameter data set is subjected to dimensionality reduction and then input into the constructed comprehensive barrier performance model to obtain the comprehensive barrier performance index of the cigarette liner paper, specifically comprising the following steps:

[0021] The comprehensive dataset was obtained by merging the multispectral characteristic parameter dataset and the nanoscale wetting dynamics dataset;

[0022] Use PCA to reduce the dimensionality of the comprehensive data set to obtain a comprehensive data set after dimensionality reduction;

[0023] Based on the dimensionality-reduced comprehensive dataset and the sample dataset, a comprehensive barrier performance model is obtained using a support vector machine and optimized through cross-validation. The sample dataset refers to the water vapor transmission rate and oxygen transmission rate.

[0024] The comprehensive data set after dimensionality reduction was input into the comprehensive barrier performance model to obtain the comprehensive barrier performance index of the cigarette liner paper sample fixed and coated with liquid crystal film.

[0025] As a preferred solution of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to the present invention, the following steps are specifically included: optimizing the observation parameters of a polarizing optical microscope by comprehensively analyzing the barrier performance index to generate a polarization image sequence and a liquid crystal film characteristic data set:

[0026] Correlation analysis was performed between the comprehensive barrier performance index and the multispectral characteristic parameter data set to determine the optimal observation parameters of the polarization optical microscope;

[0027] A polarization optical microscope with optimized observation parameters was used to record the color and texture changes of liquid crystal films of cigarette liner paper samples under the action of water vapor, oxygen and terpenes, generate polarization image sequences, and obtain liquid crystal film feature datasets through feature extraction.

[0028] As a preferred solution of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to the present invention, the step of obtaining a liquid crystal film feature dataset by feature extraction refers to extracting visual features of a polarized image sequence by image processing to generate a liquid crystal film feature dataset that characterizes changes in the orientation of liquid crystal molecules.

[0029] As a preferred solution of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to the present invention, the visualization evaluation model is constructed based on the comprehensive barrier performance index and the liquid crystal film characteristic data set to obtain the final barrier performance index by merging the comprehensive barrier performance index with the visual features in the liquid crystal film characteristic data set into a comprehensive evaluation data set, and using a linear regression method to construct a visualization evaluation model to generate the final barrier performance index.

[0030] As a preferred solution of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to the present invention, the drawing of multidimensional visualization charts and the generation of a cigarette liner paper analysis report refer to integrating a nanoscale wetting dynamics data set, a multispectral characteristic parameter data set, a comprehensive barrier performance index, a liquid crystal film characteristic data set, and the final barrier performance, using the Matplotlib library to draw a multidimensional visualization chart, and generating a comprehensive cigarette liner paper analysis report.

[0031] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper as described in the first aspect of the present invention is implemented.

[0032] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the analytical method for rapidly evaluating the barrier properties of cigarette liner paper as described in the first aspect of the present invention is implemented.

[0033] The beneficial effects of the present invention are as follows: by coating a thermochromic liquid crystal film and combining it with in-situ observation using an atomic force microscope, the nanoscale liquid film formation process that cannot be identified by traditional methods is captured, significantly improving the detection sensitivity of trace penetration behavior. At the same time, the synergistic effects of multiple penetrating substances are indirectly reflected through the wetting dynamics characteristics. Moreover, based on the simultaneous monitoring of terahertz waves, infrared and ultraviolet multi-spectra, the limitations of traditional single substance detection are broken through, and the multi-dimensional barrier performance characterization of water vapor, oxygen and volatile organic compounds is achieved. In addition, through principal component analysis dimensionality reduction and support vector machine modeling, complex multi-source data is converted into a reliable barrier performance index, solving the problems of relying on a single indicator and data isolation. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 Flowchart of the analytical method for rapid evaluation of the barrier properties of cigarette liner paper.

[0036] Figure 2 Schematic diagram of nanoscale wetting dynamics data acquisition.

[0037] Figure 3 Flowchart for multispectral monitoring and model building.

[0038] Figure 4 Flowchart of the visual evaluation model. DETAILED DESCRIPTION

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0040] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0041] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0042] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides an analytical method for rapidly evaluating the barrier properties of cigarette liner paper, comprising the following steps:

[0043] S1. After coating the surface of the liner paper for the cigarette to be tested with a liquid crystal solution, an initial nanoscale surface topology dataset is collected. By introducing water vapor, a contact angle variation curve and a liquid film thickness distribution map are generated to obtain a nanoscale wetting dynamics dataset.

[0044] The specific steps include:

[0045] S1.1. Select the cigarette lining paper to be tested and cut it into square samples using a cutting tool. Inspect the surface to ensure there are no wrinkles, scratches, or contaminants. Gently wipe the surface of the cigarette lining paper sample with a dust-free cloth dipped in analytical-grade ethanol. Rinse with deionized water and dry in a vacuum drying oven to obtain a clean cigarette lining paper sample. Prepare a thermochromic liquid crystal solution and evenly apply it to the surface of the cigarette lining paper sample using a spin coater to form a liquid crystal film. Place the LC film-coated cigarette lining paper sample in an oven to cure, ensuring the LC film is uniform and free of solvent residue. Inspect the surface of the LC film using an optical microscope to confirm the presence of bubbles or cracks. Secure the qualified LC film-coated cigarette lining paper sample to the sample holder of a customized multifunctional testing device, which integrates an atomic force microscope with a probe, a multispectral instrument, and a polarization optical microscope. Adjust the sample position to ensure alignment with the test components. The multifunctional testing device is placed in a constant temperature and humidity chamber. One side of the sample holder is connected to a permeation chamber, which allows water vapor, oxygen and volatile organic compounds to enter, simulating the internal environment of cigarette packaging.

[0046] S1.2. Scan the surface of the fixed and liquid crystal film-coated cigarette liner paper sample using an atomic force microscope to record the initial surface topology, including surface roughness, porosity, and microscopic pore size distribution. Use NanoScope-Analysis software to generate an initial nanoscale surface topology dataset and three-dimensional topology image, which are saved in CSV format and include timestamps and environmental conditions. NanoScope-Analysis software is data analysis software used in conjunction with an atomic force microscope.

[0047] The permeation chamber was activated, water vapor was introduced, and the humidity was gradually increased using environmental control equipment, with each 10% RH increment maintained for 5 minutes until it reached 80% RH. Under each humidity condition, an atomic force microscope was used in tapping mode to observe in situ the nanoscale liquid film formed by water molecules adsorbed on the surface of the liquid crystal film. The height distribution data of the nanoscale liquid film was collected every 30 seconds for 30 minutes. NanoScope-Analysis software was used to calculate the liquid film contact angle and thickness based on the height distribution, and the liquid film profile information was extracted to generate a profile dataset containing the contact angle and thickness.

[0048] Based on the profile data set, NanoScope-Analysis software was used to calculate the change in the liquid film contact angle over time, and the contact angle change rate was obtained, which was used to generate the contact angle change curve. NanoScope-Analysis software was also used to analyze the change in the liquid film area over time and calculate the liquid film expansion rate, which was used to generate the liquid film thickness distribution map.

[0049] NanoScope-Analysis software was used to integrate the contact angle change rate and liquid film expansion rate data to generate a contact angle change curve and a liquid film thickness distribution diagram. The contact angle change curve shows the trend of the contact angle change rate with time and humidity, and the liquid film thickness distribution diagram shows the spatial distribution of the liquid film thickness. The data were saved in CSV format, including timestamp, humidity, contact angle, and liquid film thickness, for subsequent multispectral monitoring.

[0050] The three-dimensional topological image, contact angle change rate, liquid film thickness, surface roughness, as well as the contact angle change curve and liquid film thickness distribution map are combined into a nanoscale wetting dynamics dataset.

[0051] S2. determining a characteristic signal range for multispectral monitoring of cigarette liner paper based on the nanoscale wetting dynamics dataset and performing feature extraction to obtain a multispectral characteristic parameter dataset;

[0052] The specific steps include:

[0053] S2.1. Analyze the contact angle change rate and liquid film thickness data based on the nanoscale wetting dynamics dataset. The analysis process is as follows: Use Python software to read the contact angle change rate and liquid film thickness data in the nanoscale wetting dynamics dataset and perform statistical analysis. Calculate the mean, standard deviation, maximum, minimum, and quartiles for the contact angle change rate and liquid film thickness, respectively. Calculate the sliding window mean for the contact angle change rate and liquid film thickness data. Smooth the data to identify long-term trends. Determine the barrier performance trends related to water vapor, oxygen, and terpene penetration. Generate a barrier performance-related parameter table for selecting the characteristic signal range. Based on the barrier performance-related parameter table, use Python software to analyze the correlation between the contact angle change rate and liquid film thickness and the spectral signal. Determine the characteristic signal range for multispectral monitoring of the cigarette liner paper sample that has been fixed and coated with a liquid crystal film. The characteristic signal range includes the terahertz wave characteristic peak, infrared absorption peak, and ultraviolet absorption peak.

[0054] The multi-spectral instrumentation integrated into the multifunctional test device includes a terahertz time-domain spectrometer, a Fourier transform infrared spectrometer, and a UV-visible spectrophotometer. The optical paths of the terahertz time-domain spectrometer, Fourier transform infrared spectrometer, and UV-visible spectrophotometer were calibrated to ensure synchronous acquisition of the three spectral signals. Baseline correction was performed using polyethylene film to verify the signal stability of the multi-spectral instrument.

[0055] It is further explained that correlating nanoscale wetting dynamics data with spectral signals builds a bridge from microscopic interface behavior to macroscopic barrier performance, significantly improving detection sensitivity.

[0056] S2.2. Using the determined characteristic signal range of multispectral monitoring and the calibrated multispectral equipment, real-time spectral signal acquisition is performed on the cigarette lining paper sample fixed and coated with liquid crystal film. The specific process is as follows: keep the permeation chamber running, introduce water vapor, oxygen and terpenes, use a terahertz time-domain spectrometer to monitor the characteristic peak displacement and intensity change of water vapor after passing through the cigarette lining paper sample fixed and coated with liquid crystal film, use a Fourier transform infrared spectrometer to monitor the intensity change of the characteristic absorption peak of terpenes, and use a UV-visible spectrophotometer to monitor the area of ​​the oxygen absorption peak, and obtain three groups of Spectral datasets. Python software was used to perform denoising and normalization on the three spectral datasets. Denoising involved applying a sliding average filter to remove random noise and high-frequency interference to enhance the stability and clarity of the spectral signals of the cigarette liner paper sample, which had been fixed and coated with a liquid crystal film. Normalization involved linearly scaling the intensity values ​​of the three denoised spectral datasets to the [0, 1] range to eliminate dimensional differences between different spectral instruments. The terahertz peak shift, infrared absorption peak intensity change, and ultraviolet absorption peak area were extracted from the three normalized spectral datasets. These values, along with the test conditions, were saved together to form a multispectral characteristic parameter dataset.

[0057] It is further explained that by determining the characteristic signal range of multi-spectral monitoring and using calibrated multi-spectral equipment, real-time spectral signal acquisition is carried out on cigarette liner paper samples that are fixed and coated with liquid crystal film, thereby achieving a comprehensive characterization of multi-dimensional barrier properties such as moisture, oxygen and volatile organic compounds, overcoming the limitation of traditional single detection methods that can only monitor a single substance.

[0058] S3. Performing dimensionality reduction on the multispectral feature parameter data set and inputting the data into the constructed comprehensive barrier performance model to obtain a comprehensive barrier performance index of the cigarette liner paper;

[0059] The specific steps include:

[0060] S3.1. Use the Pandas library to read the multispectral characteristic parameter dataset and the nanoscale wetting dynamics dataset, extract the contact angle change rate in the nanoscale wetting dynamics dataset, and align it with the timestamp of the multispectral characteristic parameter dataset. Then, use the merge function of the Pandas library to merge the multispectral characteristic parameter dataset and the contact angle change rate by timestamp to generate a comprehensive dataset containing the terahertz wave characteristic peak displacement, infrared absorption peak intensity change, ultraviolet absorption peak area and contact angle change rate, and perform normalization.

[0061] S3.2. Perform dimensionality reduction on the standardized comprehensive data set, set the number of principal components to 2, call the PCA fit method, perform eigendecomposition on the standardized comprehensive data set, calculate the two main eigenvectors and the eigenvalues ​​corresponding to the two main eigenvectors, generate a principal component analysis model, and save the eigenvector data of the principal component analysis model as a temporary CSV file, which contains the weight coefficients of the two principal components. Generate preliminary dimensionality reduction data. Each row includes two principal component values, which are directly used for variance explanation rate calculation. Calculate the variance explanation rate of the two principal components and sum them to obtain the cumulative variance explanation rate. Compare the cumulative variance explanation rate with the 95% threshold (customized according to personal needs). If it reaches or exceeds 95%, confirm that the principal component analysis model is valid, and obtain a principal component analysis model that has been verified by the variance explanation rate, which contains two principal component values ​​and variance explanation rate information.

[0062] The transformed method is used to perform dimensionality reduction on the validated, standardized, and standardized synthetic dataset, based on a principal component analysis model validated for variance explanation. All data in the synthetic dataset is projected onto two principal component spaces, generating reduced-dimensional data with each row containing two principal component values. The Pandas library is used to integrate the reduced-dimensional data with the synthetic dataset to form a reduced-dimensional synthetic dataset. This dataset contains the timestamp, the test condition, and the two principal component values.

[0063] It is further explained that simplifying the original four-dimensional features into two principal component values ​​significantly reduces the computational complexity, reduces the impact of multicollinearity, improves the efficiency and stability of subsequent support vector machine training, and provides high-quality input data for the construction of a comprehensive barrier performance model.

[0064] S3.3. The process of building a comprehensive barrier performance model is as follows: Based on the comprehensive dataset after dimensionality reduction and combined with the sample datasets of known water vapor transmission rate and oxygen transmission rate, an initial comprehensive barrier performance model is built using the scikit-learn library, and initial weights are set:

[0065] P=w1·A+w2·B+w3·C+w4·D;

[0066] Among them, P represents the comprehensive barrier performance index, A represents the displacement of the characteristic peak of terahertz wave, B represents the change in the intensity of the infrared absorption peak, C represents the area of ​​the ultraviolet absorption peak, D represents the rate of change of the contact angle, w1 represents the weight of the displacement of the characteristic peak of terahertz wave, w2 represents the weight of the change in the intensity of the infrared absorption peak, w3 represents the weight of the area of ​​the ultraviolet absorption peak, and w4 represents the weight of the rate of change of the contact angle.

[0067] The Pandas library was used to load the reduced comprehensive dataset, extract the two principal component values ​​as input features, and load the sample dataset as the target output labels. The support vector machine (SVM) was used with the two principal component values ​​as input and the water vapor transmission rate and oxygen transmission rate as output to construct the initial comprehensive barrier performance model. Next, the initial comprehensive barrier performance model was optimized: weight coefficients w1 to w4 were optimized, support vector machine hyperparameters were set, model performance was evaluated using 5-fold cross-validation, and R was calculated. 2 value, ensure R 2 The value is greater than 0.95, if R 2 If it does not reach 95%, adjust the support vector machine hyperparameters and retrain until it meets R 2 Greater than 0.95, the final comprehensive barrier performance model is completed, R 2 It is an indicator used in statistics and machine learning to evaluate the predictive performance of regression models, usually ranging from 0 to 1. 2 The closer the value is to 1, the more consistent the model prediction is with the actual value, and the better the model performance is.

[0068] The final comprehensive barrier performance model is loaded, and the two principal component values ​​in the comprehensive data set after dimensionality reduction are used as input features and input into the final comprehensive barrier performance model. The final comprehensive barrier performance model predicts the two principal component values ​​through the optimized weight coefficients and calculates the comprehensive barrier performance index of the cigarette liner paper sample that has been fixed and coated with liquid crystal film.

[0069] It is further explained that the comprehensive barrier performance model uses support vector machines to efficiently learn the nonlinear relationship between the two principal component values ​​and water vapor transmission rate and oxygen transmission rate to generate a comprehensive barrier performance index, which provides a reliable preliminary barrier performance indicator for the subsequent visual evaluation of liquid crystal molecular orientation changes.

[0070] S4. Optimize the observation parameters of the polarization optical microscope through the comprehensive barrier performance index, generate polarization image sequences, videos, and liquid crystal film characteristic datasets, and construct a visualization evaluation model based on the comprehensive barrier performance index and liquid crystal film characteristic dataset to obtain the final barrier performance index.

[0071] The specific steps include:

[0072] S4.1. Based on the multispectral characteristic parameter data set and the comprehensive barrier performance index, the Pandas library is used to analyze the correlation between the comprehensive barrier performance index and the multispectral characteristic parameter data set, and the observation parameter range of the polarization optical microscope is determined. The process of determining the observation parameter range is as follows: perform Pearson correlation analysis on the comprehensive analysis data set, calculate the correlation coefficient between the comprehensive barrier performance index and the multispectral characteristic parameter data set, generate a correlation coefficient table containing correlation coefficient values, ranging from -1 to 1, and based on the correlation coefficient table, screen parameters with an absolute value greater than 0.7 relative to the comprehensive barrier performance index. Determine the observation parameter range of the polarization optical microscope based on the physical properties of the screened parameters (for example, the terahertz wave characteristic peak displacement corresponding to the water vapor permeability characteristics). The parameters with an absolute value greater than 0.7 mentioned here refer to the terahertz wave characteristic peak displacement, infrared absorption peak intensity change, and ultraviolet absorption peak area in the multispectral characteristic parameter data set. The observation parameter range of the polarization optical microscope mentioned includes the angle between the polarizer and the analyzer, the wavelength light source, and the magnification.

[0073] It is further explained that through correlation analysis, the multi-spectral features that are highly correlated with the barrier performance are accurately identified, the observation parameters of the polarization optical microscope are scientifically determined, and the angles of the polarizer and analyzer, the wavelength light source and the magnification are optimized to match the liquid crystal molecular orientation change characteristics of the cigarette liner paper sample that is fixed and coated with liquid crystal film, thereby improving the acquisition quality of polarization image sequences and videos, and providing high-resolution and reliable visual data for subsequent feature extraction.

[0074] After the polarizing optical microscope is adjusted according to the determined observation parameter range, the permeation chamber of the multifunctional testing device is kept running, and water vapor, oxygen and terpenes are introduced to simulate the internal environment of the cigarette packaging. The polarizing optical microscope is used to record the color and texture changes of the liquid crystal film, generate polarization image sequences and corresponding videos, and save them in PNG and AVI formats, including timestamps and test conditions.

[0075] ImageJ was used to process polarization image sequences and extract the visual features of the cigarette liner paper sample, which had been fixed and coated with liquid crystal film, frame by frame: RGB values ​​and texture features. The processing process was as follows: the texture contrast of each frame was calculated using the Gray Level Co-occurrence Matrix plugin in ImageJ, generating a liquid crystal film feature dataset containing timestamps, test conditions, RGB values, and texture contrast values. Using the NumPy library, the color change rate was calculated by dividing the difference between the RGB values ​​of adjacent frames by the time interval. The texture contrast change value was calculated by dividing the difference between the texture contrast values ​​of adjacent frames by the time interval. The expressions for this are:

[0076]

[0077]

[0078] Among them, Ri Indicates the color change rate of the i-th frame, L i Indicates the texture contrast change value of the i-th frame, U i Represents the red channel pixel value of the i-th frame, V i Indicates the green channel pixel value of the i-th frame, S i represents the blue channel pixel value of the i-th frame, T i Represents the texture contrast value of the i-th frame, T i+1 represents the texture contrast value of the i+1th frame, and Δt represents the time interval.

[0079] The color change rate and texture contrast change values ​​are added to the liquid crystal film feature dataset.

[0080] S4.2. Based on the liquid crystal film characteristic dataset and the comprehensive barrier performance index, a visual evaluation model was constructed using Python software. The specific process was as follows: the color change rate and texture contrast change value in the liquid crystal film characteristic dataset were combined with the comprehensive barrier performance index to form a comprehensive evaluation dataset; linear regression was used with the color change rate, texture contrast change value, and comprehensive barrier performance index as input features, and a barrier dataset containing known water vapor transmission rate, oxygen transmission rate, and aroma retention rate as the target output label. The visual evaluation model was trained, the weight coefficients of the color change rate, texture contrast change value, and comprehensive barrier performance index were optimized, and 5-fold cross-validation was used to evaluate the model performance and calculate R. 2 value, ensure R 2 If the value is greater than 0.90, and if it does not reach 90%, adjust the comprehensive evaluation data set and retrain until R 2 If the value is greater than 0.90, the construction of the visual evaluation model is completed, including the optimized color change rate, texture contrast change value, and weight coefficient and linear regression parameters of the comprehensive barrier performance index:

[0081] Q = k1·E+k2·Y+k3·P;

[0082] Wherein, Q represents the final barrier performance index, E represents the color change rate, Y represents the texture contrast change value, P represents the comprehensive barrier performance index, k1 represents the weight of the color change rate, k2 represents the weight of the texture contrast change value, and k3 represents the comprehensive barrier performance index;

[0083] The comprehensive evaluation dataset was input into the constructed visual evaluation model to obtain the final barrier performance index. This data set, including color change rate, texture contrast change, comprehensive barrier performance index, and final barrier performance index, was generated. Ten cigarette liner paper samples, fixed and coated with liquid crystal film, were selected. Water vapor transmission rate (WVTR) was measured using the traditional gravimetric method, while oxygen transmission rate (OTR) and aroma retention rate (ARR) were measured using gas chromatography-mass spectrometry. The experiment was repeated three times, and repeatability was calculated and compared with the final barrier performance index. Deviation was calculated to ensure that the deviation was less than 5%. A deviation analysis table and repeatability evaluation table were generated, including sample number, test conditions, measured values ​​for WVTR, OTR, and ARR, the final barrier performance index, and deviation percentage.

[0084] It is further explained that the visual evaluation model integrates the visual characteristics of liquid crystal film and the previous comprehensive barrier performance index, generates a high-precision model through linear regression and 5-fold cross-validation, and refines the barrier performance evaluation of cigarette liner paper samples that are fixed and coated with liquid crystal film. Compared with the comprehensive barrier performance index, it more comprehensively reflects the water vapor, oxygen and aroma retention performance.

[0085] S5. Draw a multi-dimensional visualization chart and generate an analysis report on cigarette liner paper.

[0086] The specific steps include:

[0087] Pandas was used to align the polarization image sequence, color change rate, texture contrast change value, final barrier performance index, comprehensive barrier performance index, multispectral feature parameters, contact angle change rate, and liquid film thickness by timestamp and merge them into a chart dataset.

[0088] Matplotlib was used to create multidimensional visualizations based on the chart dataset. The process involved plotting time series of the terahertz wave characteristic peak displacement, infrared absorption peak intensity change, and ultraviolet absorption peak area, using the timestamp as the horizontal axis, to reflect the dynamics of the multispectral characteristics. The color change rate and texture contrast change values ​​were plotted as color change curves and texture contrast change curves of the liquid crystal film, demonstrating the changes in the visual characteristics of the liquid crystal film and the dynamics of its texture characteristics.

[0089] The chart data sets, multi-dimensional visualization charts, 3D topological images, deviation analysis tables, and repeatability evaluation tables are organized into a cigarette liner paper analysis report, which includes sample number, test conditions, contact angle change curve, liquid film thickness distribution diagram, polarization image, final barrier performance index and its corresponding measured value, comparison results with traditional methods, error analysis, and repeatability evaluation.

[0090] This embodiment also provides a computer device suitable for the analysis method of quickly evaluating the barrier properties of cigarette liner paper, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the analysis method of quickly evaluating the barrier properties of cigarette liner paper as proposed in the above embodiment.

[0091] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0092] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the analysis method for rapidly evaluating the barrier properties of cigarette liner paper as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0093] In summary, the present invention: by coating a thermochromic liquid crystal film and combining it with in situ atomic force microscopy observation, it captures the nanoscale liquid film formation process that cannot be identified by traditional methods, significantly improving the detection sensitivity of trace penetration behavior, while indirectly reflecting the synergistic influence of multiple penetrating substances through wetting dynamic characteristics. Moreover, based on the simultaneous monitoring of terahertz waves, infrared and ultraviolet multispectroscopy, it breaks through the limitations of traditional single substance detection and realizes the multi-dimensional barrier performance characterization of water vapor, oxygen and volatile organic compounds. In addition, through principal component analysis dimensionality reduction and support vector machine modeling, complex multi-source data is converted into a reliable barrier performance index, which solves the problem of traditional methods relying on single indicators and isolated data.

[0094] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for rapidly evaluating the barrier properties of cigarette liner paper, characterized by: include, After the liquid crystal solution is applied to the surface of the cigarette liner paper to be tested, water vapor is passed through to generate a contact angle change curve and a liquid film thickness distribution map, thereby obtaining a nanoscale wetting dynamics data set. Determine the characteristic signal range of multispectral monitoring of cigarette liner paper based on the nanoscale wetting dynamics data set and perform feature extraction to obtain a multispectral characteristic parameter data set; After dimensionality reduction of the multispectral feature parameter dataset, the constructed comprehensive barrier performance model was input to obtain the comprehensive barrier performance index of cigarette liner paper. The observation parameters of the polarization optical microscope are optimized by using the comprehensive barrier performance index to generate polarization image sequences and liquid crystal film characteristic datasets. A visualization evaluation model is then constructed based on the comprehensive barrier performance index and the liquid crystal film characteristic dataset to obtain the final barrier performance index. Draw multi-dimensional visualization charts and generate analysis reports for cigarette liner paper.

2. The analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to claim 1, characterized in that: After the liquid crystal solution is applied to the surface of the inner lining paper of the cigarette to be tested, water vapor is passed through to generate a contact angle change curve and a liquid film thickness distribution map to obtain a nanoscale wetting dynamics data set, which specifically includes the following steps: A liquid crystal solution is coated on the surface of the cigarette liner paper to be tested to form a liquid crystal film. The liquid crystal film-coated cigarette liner paper sample is fixed on the sample holder of a multifunctional testing device, and the initial surface topology is recorded using an atomic force microscope. The permeation chamber was activated, water vapor was introduced, and an atomic force microscope was used to observe in situ the nanoscale liquid film formed by the adsorption of water molecules. The contact angle and thickness of the liquid film were calculated using NanoScope-Analysis, generating a contact angle curve and a liquid film thickness distribution map. The initial surface topology, contact angle variation curve, and liquid film thickness distribution map are combined into a nanoscale wetting dynamics dataset.

3. The analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to claim 2, characterized in that: The characteristic signal range of multispectral monitoring of cigarette liner paper is determined based on the nanoscale wetting dynamics data set and feature extraction is performed to obtain a multispectral characteristic parameter data set, specifically including the following steps: Read the nanoscale wetting dynamics data set for statistical analysis, generate a table of barrier performance-related parameters, and determine the characteristic signal range of multispectral monitoring based on the table of barrier performance-related parameters; According to the determined characteristic signal range of multispectral monitoring, the multispectral equipment is used to extract the terahertz wave characteristic peak displacement, infrared absorption peak intensity change and ultraviolet absorption peak area, and combine them to obtain a multispectral characteristic parameter data set.

4. The analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to claim 3, characterized in that: After dimensionality reduction of the multispectral feature parameter data set, the constructed comprehensive barrier performance model is input to obtain the comprehensive barrier performance index of the cigarette liner paper. The specific steps include the following: The comprehensive dataset was obtained by merging the multispectral characteristic parameter dataset and the nanoscale wetting dynamics dataset; Use PCA to reduce the dimensionality of the comprehensive data set to obtain a comprehensive data set after dimensionality reduction; Based on the dimensionality-reduced comprehensive dataset and the sample dataset, a comprehensive barrier performance model is obtained using a support vector machine and optimized through cross-validation. The sample dataset refers to the water vapor transmission rate and oxygen transmission rate. The comprehensive data set after dimensionality reduction was input into the comprehensive barrier performance model to obtain the comprehensive barrier performance index of the cigarette liner paper sample fixed and coated with liquid crystal film.

5. The analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to claim 4, characterized in that: By optimizing the observation parameters of the polarization optical microscope through the comprehensive barrier performance index, a polarization image sequence and a liquid crystal film characteristic data set are generated. The specific steps include the following: Correlation analysis was performed between the comprehensive barrier performance index and the multispectral characteristic parameter data set to determine the optimal observation parameters of the polarization optical microscope; A polarization optical microscope with optimized observation parameters was used to record the color and texture changes of liquid crystal films of cigarette liner paper samples under the action of water vapor, oxygen and terpenes, generate polarization image sequences, and obtain liquid crystal film feature datasets through feature extraction.

6. The analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to claim 5, characterized in that: The obtaining of the liquid crystal film feature data set by feature extraction refers to extracting visual features of a polarization image sequence by image processing to generate a liquid crystal film feature data set representing changes in the orientation of liquid crystal molecules.

7. The analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to claim 6, characterized in that: The visualization evaluation model is constructed based on the comprehensive barrier performance index and the liquid crystal film characteristic data set to obtain the final barrier performance index. This is achieved by merging the comprehensive barrier performance index and the visual features in the liquid crystal film characteristic data set into a comprehensive evaluation data set, and using a linear regression method to construct the visualization evaluation model to generate the final barrier performance index.

8. The analytical method for rapidly evaluating the barrier properties of cigarette liner paper according to claim 7, characterized in that: The drawing of multi-dimensional visualization charts and generation of an analysis report for cigarette liner paper refers to integrating a nano-scale wetting dynamics data set, a multi-spectral characteristic parameter data set, a comprehensive barrier performance index, a liquid crystal film characteristic data set and a final barrier performance, using the Matplotlib library to draw a multi-dimensional visualization chart, and generating a comprehensive analysis report for cigarette liner paper.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the analysis method for rapidly evaluating the barrier properties of cigarette liner paper according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the analysis method for rapidly evaluating the barrier properties of cigarette liner paper according to any one of claims 1 to 8 are implemented.

Images