Component detection method of waterproof material for building external wall

Abstract

The invention discloses a component detection method of a waterproof material for a building external wall. In the invention, a machine learning algorithm is used to automatically identify the chromatographic peak and remove noise and background interference, and the method has significant beneficial effects on the component detection of the waterproof material for the building exterior wall. Through accurate component analysis, the detection precision and reliability can be improved, and it is ensured that various components, including polymers, additives and the like, in the waterproof material are accurately recognized and quantified. The detection process is accelerated through automatic processing, the working efficiency of a laboratory is improved, and manpower and cost are saved. The reliability and the accuracy of the detection method are improved through the standardized process and repeatability, product quality control and formula optimization are facilitated, and in the component detection method of the waterproof material for the building outer wall, the detection efficiency and the detection precision are improved through the machine learning algorithm, the product quality control is enhanced, and the detection accuracy is improved. And research, development and innovation of products are supported, so that the competitiveness of the whole industry is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of waterproof material detection, and in particular relates to a method for detecting the composition of waterproof materials for building exterior walls. Background Art

[0002] Waterproof materials for building exterior walls are functional materials specially used to protect the exterior walls of buildings and prevent water intrusion and leakage. Its main function is to improve the durability of buildings, ensure the dryness and comfort of the indoor environment, and enhance the overall beauty and structural safety of the building. The following are the main features of waterproof materials for building exterior walls: Good waterproof performance: The material has excellent hydrophobicity and can form a solid waterproof barrier on the surface of the building exterior wall, effectively blocking rainwater and groundwater from invading the wall. High-efficiency durability: The waterproof material has strong UV resistance and aging resistance, and can maintain the waterproof effect for a long time even under harsh climatic conditions. Environmental protection and safety: The waterproof material for building exterior walls is non-toxic, harmless, does not contain harmful substances, and is friendly to the human body and the environment. In the research and development and production process of waterproof materials for building exterior walls, component detection is a crucial link. This link involves a series of complex pretreatment steps for the sample, including grinding, extraction, purification, etc., to ensure that subsequent chromatographic analysis, spectral analysis and other means can accurately reveal the composition of the sample.

[0003] However, traditional component detection methods usually involve sample pretreatment, such as grinding, extraction, purification, etc., and then component analysis by chromatography, spectral analysis, etc. However, these methods have some limitations, such as complex pretreatment steps, cumbersome operations, and long analysis cycles. Summary of the invention

[0004] The purpose of the present invention is to provide a method for detecting the composition of waterproof materials for building exterior walls in order to solve the above-mentioned problems.

[0005] The technical solution adopted by the present invention is as follows: a method for detecting the composition of waterproof materials for building exterior walls, the method comprising the following steps:

[0006] S1: Collect samples of waterproof materials and store them in clean and dry containers to avoid contamination and moisture absorption;

[0007] S2: Use a ball mill to crush the dried sample into a fine powder; ensure the homogenization of the sample by high-speed stirring or ultrasonic treatment to pre-treat the sample material;

[0008] S3: diluting the concentrated sample obtained by pretreatment with an appropriate solvent to a concentration suitable for HPLC analysis, and then performing high performance liquid chromatography analysis;

[0009] S4: The extract or the directly injected sample is subjected to derivatization treatment and then analyzed by gas chromatography-mass spectrometry;

[0010] S5: Mix the dried powder sample with an appropriate amount of potassium bromide, press it into a transparent infrared spectrum sheet, and then perform infrared spectrum analysis;

[0011] S6: Use NMR technology to analyze the organic molecular structure in the sample and infer the molecular structure through chemical shift and coupling constant information;

[0012] S7: Observe the microscopic morphology and structure of the sample using a transmission electron microscope (TEM) or a scanning electron microscope (SEM), and perform elemental analysis using an energy dispersive spectrometer.

[0013] S8: Data integration, integrating the data obtained from all analytical instruments and comparing the results obtained from different techniques;

[0014] S9: Use statistical software to process the data and perform principal component analysis to identify the main components and impurities;

[0015] S10: Verify the accuracy of the analysis results by repeated tests and standard material comparison methods;

[0016] S11: The analysis results are recorded and saved, thereby completing the entire composition detection process of the waterproof material for the building exterior wall.

[0017] In a preferred embodiment, in step S1, the sample is cut into small pieces to ensure the uniformity of the sample; if the sample is uneven, it is fully mixed; the sample is placed in an electric constant temperature drying oven and dried to constant weight at 105°C; for heat-sensitive samples, freeze drying is used.

[0018] In a preferred embodiment, in step S2, the pre-processing method includes:

[0019] S2-1. Select appropriate acetone or toluene solvent for liquid-liquid extraction to extract organic components in the sample; for polymers in waterproof materials, use a Soxhlet extractor for long-term extraction;

[0020] S2-2. Filter the extract using filter paper or a filter membrane to remove insoluble impurities, and further purify the extract by an activated carbon column or a solid phase extraction column;

[0021] S2-3. The extract was concentrated by a rotary evaporator to remove the solvent to obtain a higher concentration of the target compound;

[0022] S2-4. According to the properties of the target compound, a derivatization reaction is performed to improve its response on GC-MS;

[0023] S2-5. Dilute the concentrated sample to a concentration suitable for instrument detection; aliquot the sample and prepare for subsequent instrument analysis.

[0024] In a preferred embodiment, the step S3 specifically includes:

[0025] S2-1: Automated sample preparation, using an automated dilution system to dilute the concentrated sample to a concentration suitable for HPLC analysis according to a preset program. The automated filtration device is equipped with a 0.22 μm filter membrane to filter the sample to ensure that there is no interference from particulate matter and protein. For samples that need to be pre-concentrated, an online S2-PE-HPLC system is used for automated pre-concentration;

[0026] S2-2: Column and mobile phase selection: select a suitable UHPLC column with ultrafine particles C18 column according to the sample characteristics, use expert system software to automatically optimize the mobile phase composition, including the ratio of aqueous phase and organic phase, and set the gradient elution program to ensure the effective separation of different components;

[0027] S2-3: Intelligent setting of chromatographic conditions, automatic adjustment of flow rate to obtain optimal separation efficiency and peak shape, setting column temperature, usually between 30-40°C to reduce thermal degradation of samples, and setting appropriate detection wavelength according to detector type;

[0028] S2-4: Instrument operation and data acquisition: start the autosampler, inject samples according to the preset injection sequence and volume, start the data acquisition system, ensure the high resolution and sensitivity of the chromatogram, monitor the system performance in real time, including pressure and column temperature, and automatically adjust the parameters through the feedback control system;

[0029] S2-5: Chromatographic peak detection and optimization, using machine learning algorithms to automatically identify chromatographic peaks, remove noise and background interference, perform peak shape fitting on chromatographic peaks, and ensure the accuracy of integration;

[0030] S2-6: Compound identification: Combining retention time and spectral data, the AI-assisted library matching system is used to identify components; for unknown compounds, spectral analysis software is used to infer structures;

[0031] S2-7: Quantitative analysis, using advanced integration methods for accurate integration, using internal or external standard methods for quantification, and calculating the concentration of each component;

[0032] S2-8: Data analysis and report generation, using statistical software for data analysis, including precision, accuracy and repeatability assessment, and automatically generating detailed analysis reports including sample information, analysis conditions, results and conclusions.

[0033] In a preferred embodiment, in the step S2-5, data preprocessing is first performed, including data import, cleaning and normalization; then, key chromatographic peak information is extracted through peak detection and feature calculation; on this basis, a noise model is established and background correction is performed to identify and remove interference signals; then, using the training set data, a suitable machine learning algorithm is selected for model training, and the model is optimized through feature selection; the trained model is used to automatically identify chromatographic peaks in new data, and the accuracy of the results is improved through peak confirmation and noise suppression steps; finally, the analysis quality is ensured through result verification and model optimization, and the entire process is automated to establish a continuous monitoring process to achieve efficient and reliable chromatographic data analysis.

[0034] In a preferred embodiment, in step S4, first, the extract or sample is placed in a derivatization reactor, and a suitable derivatization reagent is selected according to the chemical properties of the sample, including a silanization reagent, an acylating reagent or an alkylating reagent; the derivatization reaction is fully carried out under mild heating or room temperature conditions, and it is usually necessary to add a catalyst or use specific reaction conditions to promote the reaction; after the derivatization is completed, the excess derivatization reagent and solvent are removed by nitrogen purging or rotary evaporation, leaving the derivatization product; then, the derivatization product is dissolved in an appropriate solvent for GC-MS injection; in GC-MS analysis, the sample is injected into a gas chromatograph, and after separation by a chromatographic column, different compounds are eluted at different time points according to their boiling point and polarity, and enter a mass spectrometer for detection; the mass spectrometer generates a mass spectrum according to the mass-to-charge ratio of the compound, and identifies and quantifies the components in the sample by comparing the standard spectral library or the calculated retention time.

[0035] In a preferred embodiment, in step S5, the dried powder sample is first mixed with an appropriate amount of potassium bromide; potassium bromide, as an infrared transparent salt, is used to prepare infrared spectroscopic films because it does not absorb light within the infrared spectrum; during the mixing process, it is necessary to ensure that the ratio of the sample and potassium bromide is appropriate so as to obtain sufficient signal intensity without oversaturation; the mixed powder is carefully transferred to a special mold and pressed into a transparent infrared spectroscopic film with the help of a press; during the pressing process, the pressure needs to be evenly distributed to avoid bubbles and cracks, which will affect the transparency of the spectroscopic film and subsequent spectral analysis results; the pressed infrared spectroscopic film is then placed in the sample chamber of an infrared spectrometer; during the analysis process, the infrared light beam passes through the spectroscopic film, and the chemical bonds and functional groups in the sample absorb the infrared light according to their specific vibration frequencies, thereby generating characteristic absorption peaks; by comparing with a standard spectrum or using the known absorption frequency of chemical bonds, the chemical structure and composition in the sample are identified, thereby performing qualitative and quantitative analysis on the sample.

[0036] In a preferred embodiment, in step S6, first, the sample to be analyzed is dissolved in an appropriate solvent to provide a uniform magnetic field environment; the dissolved sample is placed in an NMR tube and placed in the magnetic field of an NMR spectrometer; in the 1H NMR experiment, the magnetic resonance signal of the hydrogen nucleus is excited and recorded, generating a series of chemical shift peaks, the position, shape and integrated area of ​​these peaks provide information about the hydrogen atom environment in the molecule; in the 13C NMR experiment, the signal of the carbon nucleus is also recorded to provide information about the carbon skeleton; by analyzing the chemical shift, coupling constant, and peak splitting mode parameters, the structure of the carbon-hydrogen skeleton in the molecule, the type of functional group, and the connection mode within the molecule are inferred.

[0037] In a preferred embodiment, in step S7, first, the sample is cut or ground into nanometer or micrometer-level flakes or particles and fixed on a dedicated sample stage; in SEM operation, the sample usually needs to be coated, including gold or carbon plating, to improve conductivity and image quality; in TEM operation, the sample needs to be thin enough to allow the electron beam to penetrate; while obtaining the microscopic image, elemental analysis is performed in conjunction with an energy spectrometer; EDS determines the type and relative content of elements in the sample by detecting the characteristic X-rays emitted by the sample after being excited by the electron beam; this analysis can provide chemical composition information of the sample and help researchers understand the relationship between the composition of the sample and its microstructure; through the combined use of SEM and TEM, detailed information on element distribution and chemical state can be obtained while observing the sample morphology.

[0038] In a preferred embodiment, in step S9, the preprocessed data is organized into a matrix, where each row represents a sample and each column represents a variable, and then the covariance matrix is ​​calculated using the following formula:

[0039] Where Xi is the i-th sample and Xˉ is the average of all samples;

[0040] Then perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues ​​and corresponding eigenvectors.

[0041] Calculate the eigenvalues: By solving the characteristic equation:

[0042] det(Cov(X)-λI)=0 to find the eigenvalue λ;

[0043] For each eigenvalue, by solving the equation:

[0044] (Cov(X)-λI)v=0 to find the eigenvector v;

[0045] Then sort the eigenvalues ​​from large to small; select the principal components: select the eigenvectors corresponding to the largest eigenvalues, which represent the main components in the data;

[0046] Finally, the original data matrix is ​​projected onto the selected principal component to obtain the principal component score matrix T, and the score is calculated: T = XW, where W is the matrix of the selected eigenvectors. The eigenvectors corresponding to each principal component are analyzed to determine which original variables contribute the most to the principal component.

[0047] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0048] 1. In the present invention, a machine learning algorithm is used to automatically identify chromatographic peaks and remove noise and background interference, which has a significant beneficial effect on the component detection method of waterproof materials for building exterior walls. First, through precise component analysis, the accuracy and reliability of the detection can be improved, ensuring that various components in waterproof materials, including polymers, additives, etc., are accurately identified and quantified. Secondly, automated processing accelerates the detection process, improves the work efficiency of the laboratory, and saves manpower and costs. In addition, standardized processes and repeatability improve the reliability and accuracy of the detection method, which helps to optimize product quality control and formula optimization. Finally, through precise component detection, manufacturers can better control the quality of waterproof materials, enhance market competitiveness, and improve brand reputation. Therefore, the machine learning algorithm in the component detection method of waterproof materials for building exterior walls not only improves the detection efficiency and accuracy, but also enhances the quality control of the product, supports the research and development and innovation of the product, thereby enhancing the competitiveness of the entire industry.

[0049] 2. In the present invention, the data obtained by different analysis techniques are comprehensively processed through statistical methods such as data integration and principal component analysis (PCA), and the main components and impurities are effectively identified. This data processing method not only improves the analysis efficiency, but also simplifies the complex data set through dimensionality reduction technology, making the results more intuitive and easy to understand. Through statistical analysis, the correlation between different components can be revealed, providing a scientific basis for product formulation optimization and quality control. In addition, the accuracy of the analysis results is verified by repeated experiments and standard substance comparison, ensuring the reliability and repeatability of the detection method. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0051] Example:

[0052] A method for detecting components of waterproof materials for building exterior walls, the method comprising the following steps:

[0053] S1: Collect samples of waterproof materials and store them in clean and dry containers to avoid contamination and moisture absorption;

[0054] S2: Use a ball mill to crush the dried sample into a fine powder. Pre-treat the sample material by high-speed stirring or ultrasonic treatment to ensure sample homogenization;

[0055] S3: diluting the concentrated sample obtained by pretreatment with an appropriate solvent to a concentration suitable for HPLC analysis, and then performing high performance liquid chromatography analysis;

[0056] S4: The extract or the directly injected sample is subjected to derivatization treatment and then analyzed by gas chromatography-mass spectrometry;

[0057] S5: Mix the dried powder sample with an appropriate amount of potassium bromide, press it into a transparent infrared spectrum sheet, and then perform infrared spectrum analysis;

[0058] S6: Use NMR technology to analyze the organic molecular structure in the sample and infer the molecular structure through information such as chemical shift and coupling constant.

[0059] S7: Observe the microscopic morphology and structure of the sample using a transmission electron microscope (TEM) or a scanning electron microscope (SEM), and perform elemental analysis using an energy dispersive spectrometer (EDS);

[0060] S8: Data integration, integrating the data obtained from all analytical instruments and comparing the results obtained from different techniques;

[0061] S9: Use statistical software to process the data, such as principal component analysis (PCA), to identify the main components and impurities;

[0062] S10: Verify the accuracy of the analysis results by repeated tests, comparison with standard substances, etc.;

[0063] S11: The analysis results are recorded and saved, thereby completing the entire composition detection process of the waterproof material for the building exterior wall.

[0064] In step S1, cut the sample into small pieces to ensure the uniformity of the sample. If the sample is not uniform, mix it thoroughly. Place the sample in an electric constant temperature drying oven and dry it to constant weight at 105°C. For heat-sensitive samples, freeze drying can be used.

[0065] In step S2, the preprocessing method includes:

[0066] S2-1. Select appropriate acetone or toluene solvent for liquid-liquid extraction to extract the organic components in the sample. For polymers in waterproof materials, use a Soxhlet extractor for long-term extraction;

[0067] S2-2. Filter the extract using filter paper or filter membrane to remove insoluble impurities, and further purify the extract through an activated carbon column or solid phase extraction (SPE) column;

[0068] S2-3. The extract was concentrated by a rotary evaporator to remove the solvent to obtain a higher concentration of the target compound;

[0069] S2-4. According to the properties of the target compound, a derivatization reaction is performed to improve its response on GC-MS;

[0070] S2-5. Dilute the concentrated sample to a concentration suitable for instrument detection. Aliquot the sample and prepare for subsequent instrument analysis.

[0071] Step S3 specifically includes:

[0072] S2-1: Automated sample preparation, using an automated dilution system to dilute concentrated samples to a concentration suitable for HPLC analysis according to a preset program. An automated filtration device with a 0.22 μm filter membrane is used to filter samples to ensure that there is no interference from particulate matter and protein. For samples that require pre-concentration, an online S2-PE-HPLC system is used for automated pre-concentration.

[0073] S2-2: Selection of chromatographic column and mobile phase: select a suitable UHPLC column according to the sample characteristics, such as an ultrafine particle C18 column, use expert system software to automatically optimize the mobile phase composition, including the ratio of aqueous phase to organic phase, and set the gradient elution program to ensure effective separation of different components.

[0074] S2-3: Intelligent setting of chromatographic conditions, automatic adjustment of flow rate to obtain optimal separation efficiency and peak shape, setting column temperature, usually between 30-40°C to reduce thermal degradation of samples, and setting appropriate detection wavelength according to detector type.

[0075] S2-4: Instrument operation and data acquisition: Start the autosampler, inject samples according to the preset injection sequence and volume, turn on the data acquisition system, ensure the high resolution and sensitivity of the chromatogram, monitor the system performance in real time, such as pressure and column temperature, and automatically adjust the parameters through the feedback control system.

[0076] S2-5: Chromatographic peak detection and optimization, using machine learning algorithms to automatically identify chromatographic peaks, remove noise and background interference, perform peak shape fitting on chromatographic peaks, and ensure the accuracy of integration.

[0077] S2-6: Compound identification: Combining retention time and spectral data, an AI-assisted library matching system is used to identify components; for unknown compounds, spectral analysis software is used for structural inference.

[0078] S2-7: Quantitative analysis: Apply advanced integration methods (such as peak area or peak height) for accurate integration, use internal or external standard methods for quantification, and calculate the concentration of each component.

[0079] S2-8: Data analysis and report generation, using statistical software for data analysis, including precision, accuracy and repeatability assessment, and automatically generating detailed analysis reports including sample information, analysis conditions, results and conclusions.

[0080] In step S2-5, data preprocessing is first performed, including data import, cleaning and normalization. Then, key chromatographic peak information is extracted by peak detection and feature calculation. On this basis, a noise model is established and background correction is performed to identify and remove interference signals. Subsequently, using the training set data, a suitable machine learning algorithm (such as SVM or random forest) is selected for model training, and the model is optimized by feature selection. The trained model is applied to automatically identify the chromatographic peaks in the new data, and the accuracy of the results is improved by peak confirmation and noise suppression steps. Finally, the analysis quality is ensured by result verification and model optimization, and the entire process is automated to establish a continuous monitoring process to achieve efficient and reliable chromatographic data analysis.

[0081] In step S4, first, the extract or sample is placed in a derivatization reactor, and a suitable derivatization agent, such as a silanization agent, an acylating agent or an alkylating agent, is selected according to the chemical properties of the sample. Under mild heating or room temperature conditions, the derivatization reaction is fully carried out, and it is usually necessary to add a catalyst or use specific reaction conditions to promote the reaction. After the derivatization is completed, the excess derivatization agent and solvent are removed by nitrogen purging or rotary evaporation, leaving the derivatization product. Subsequently, the derivatization product is dissolved in an appropriate solvent for GC-MS injection. In GC-MS analysis, the sample is injected into a gas chromatograph, and after separation by a chromatographic column, different compounds are eluted at different time points according to their boiling point and polarity, and enter a mass spectrometer for detection. The mass spectrometer generates a mass spectrum according to the mass-to-charge ratio (m / z) of the compound, and the components in the sample can be identified and quantified by comparing the standard spectral library or the calculated retention time.

[0082] In step S5, the dried powder sample is first mixed with an appropriate amount of potassium bromide (KBr). Potassium bromide, as an infrared transparent salt, is used to prepare infrared spectroscopic films because it does not absorb light in the infrared spectrum. During the mixing process, it is necessary to ensure that the ratio of sample and potassium bromide is appropriate so that sufficient signal intensity is obtained without oversaturation. The mixed powder is carefully transferred to a dedicated mold and pressed into a transparent infrared spectroscopic film with the help of a press. During the pressing process, the pressure needs to be evenly distributed to avoid bubbles and cracks, which will affect the transparency of the spectroscopic film and the subsequent spectral analysis results. The pressed infrared spectroscopic film is then placed in the sample chamber of the infrared spectrometer. During the analysis process, the infrared beam passes through the spectroscopic film, and the chemical bonds and functional groups in the sample absorb infrared light according to their specific vibration frequencies, thereby producing characteristic absorption peaks. By comparing the standard spectrum or using the known absorption frequency of the chemical bond, the chemical structure and composition in the sample can be identified, thereby performing qualitative and quantitative analysis of the sample.

[0083] In step S6, first, the sample to be analyzed is dissolved in an appropriate solvent, usually a deuterated solvent, such as deuterated chloroform (CDCl3) or deuterated water (D2O), to provide a uniform magnetic field environment. The dissolved sample is placed in an NMR tube and placed in the magnetic field of an NMR spectrometer. In the 1H NMR experiment, the magnetic resonance signal of the hydrogen nucleus is excited and recorded, producing a series of chemical shift peaks, the position, shape and integrated area of ​​which provide information about the hydrogen atom environment in the molecule. In the 13C NMR experiment, the signal of the carbon nucleus is also recorded to provide information about the carbon skeleton. By analyzing parameters such as chemical shift, coupling constant, and splitting mode of the peak, the structure of the carbon-hydrogen skeleton in the molecule, the type of functional group, and the connection mode inside the molecule can be inferred. In addition, two-dimensional NMR techniques, such as HSQC and COSY, can further provide correlation information between different nuclei in the molecule, thereby more comprehensively analyzing the molecular structure.

[0084] In step S7, first, the sample is cut or ground into nanometer or micrometer-level flakes or particles and fixed on a dedicated sample stage. In SEM operation, the sample usually needs to be coated, such as gold or carbon, to improve conductivity and image quality. In TEM operation, the sample needs to be thin enough so that the electron beam can penetrate. While obtaining a microscopic image, elemental analysis can be performed in conjunction with an energy dispersive spectrometer (EDS). EDS determines the type and relative content of elements in the sample by detecting the characteristic X-rays emitted by the sample after being excited by the electron beam. This analysis can provide information on the chemical composition of the sample and help researchers understand the relationship between the composition of the sample and its microstructure. By combining SEM and TEM, detailed information about element distribution and chemical state can be obtained while observing the sample morphology.

[0085] In step S9, the preprocessed data is organized into a matrix, where each row represents a sample and each column represents a variable (such as a wavelength point in a spectrum or a retention time point in a chromatogram). The covariance matrix is ​​then calculated using the formula:

[0086] Where Xi is the i-th sample and Xˉ is the average of all samples;

[0087] Then perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues ​​and corresponding eigenvectors.

[0088] Calculate eigenvalues: By solving the characteristic equation

[0089] det(Cov(X)-λI)=0 to find the eigenvalue λ.

[0090] For each eigenvalue, by solving the equation:

[0091] (Cov(X)-λI)v=0 to find the eigenvector v.

[0092] Then sort the eigenvalues ​​from largest to smallest. Select principal components: Select the eigenvectors corresponding to the largest eigenvalues. These eigenvectors represent the main components in the data.

[0093] Finally, the original data matrix is ​​projected onto the selected principal component to obtain the principal component score matrix T, and the score is calculated: T = XW, where W is the matrix of the selected eigenvectors. The eigenvectors corresponding to each principal component are analyzed to determine which original variables contribute the most to the principal component.

[0094] In the present invention, a machine learning algorithm is used to automatically identify chromatographic peaks and remove noise and background interference, which has a significant beneficial effect on the component detection method of waterproof materials for building exterior walls. First, through accurate component analysis, the accuracy and reliability of the detection can be improved, ensuring that various components in waterproof materials, including polymers, additives, etc., are accurately identified and quantified. Secondly, automated processing accelerates the detection process, improves the work efficiency of the laboratory, and saves manpower and costs. In addition, standardized processes and repeatability improve the reliability and accuracy of the detection method, which helps to optimize product quality control and formula optimization. Finally, through accurate component detection, manufacturers can better control the quality of waterproof materials, enhance market competitiveness, and improve brand reputation. Therefore, the machine learning algorithm in the component detection method of waterproof materials for building exterior walls not only improves the detection efficiency and accuracy, but also enhances the quality control of the product, supports the research and development and innovation of the product, thereby enhancing the competitiveness of the entire industry.

[0095] In the present invention, the data obtained by different analytical techniques are comprehensively processed by statistical methods such as data integration and principal component analysis (PCA), and the main components and impurities are effectively identified. This data processing method not only improves the analysis efficiency, but also simplifies the complex data set through dimensionality reduction technology, making the results more intuitive and easy to understand. Through statistical analysis, the correlation between different components can be revealed, providing a scientific basis for product formulation optimization and quality control. In addition, the accuracy of the analysis results is verified by repeated experiments and standard material comparison, ensuring the reliability and repeatability of the detection method.

[0096] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprise" 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 includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0097] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the composition of waterproof materials for building exterior walls, characterized in that: The method comprises the following steps: S1: Collect samples of waterproof materials and store them in clean and dry containers to avoid contamination and moisture absorption; S2: Use a ball mill to crush the dried sample into a fine powder; ensure the homogenization of the sample by high-speed stirring or ultrasonic treatment to pre-treat the sample material; S3: diluting the concentrated sample obtained by pretreatment with an appropriate solvent to a concentration suitable for HPLC analysis, and then performing high performance liquid chromatography analysis; S4: The extract or the directly injected sample is subjected to derivatization treatment and then analyzed by gas chromatography-mass spectrometry; S5: Mix the dried powder sample with an appropriate amount of potassium bromide, press it into a transparent infrared spectrum sheet, and then perform infrared spectrum analysis; S6: Use NMR technology to analyze the organic molecular structure in the sample and infer the molecular structure through chemical shift and coupling constant information; S7: Observe the microscopic morphology and structure of the sample using a transmission electron microscope (TEM) or a scanning electron microscope (SEM), and perform elemental analysis using an energy dispersive spectrometer. S8: Data integration, integrating the data obtained from all analytical instruments and comparing the results obtained from different techniques; S9: Use statistical software to process the data and perform principal component analysis to identify the main components and impurities; S10: Verify the accuracy of the analysis results by repeated tests and standard material comparison methods; S11: The analysis results are recorded and saved, thereby completing the entire composition detection process of the waterproof material for the building exterior wall.

2. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 1, characterized in that: In the step S1, the sample is cut into small pieces to ensure the uniformity of the sample; If the sample is not uniform, mix it thoroughly; place the sample in an electric constant temperature drying oven and dry it to constant weight at 105°C; for heat-sensitive samples, use freeze drying.

3. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 1, characterized in that: In step S2, the preprocessing method includes: S2-1. Select appropriate acetone or toluene solvent for liquid-liquid extraction to extract organic components in the sample; for polymers in waterproof materials, use a Soxhlet extractor for long-term extraction; S2-2. Filter the extract using filter paper or a filter membrane to remove insoluble impurities, and further purify the extract by an activated carbon column or a solid phase extraction column; S2-3. The extract was concentrated by a rotary evaporator to remove the solvent to obtain a higher concentration of the target compound; S2-4. According to the properties of the target compound, a derivatization reaction is performed to improve its response on GC-MS; S2-5. Dilute the concentrated sample to a concentration suitable for instrument detection; aliquot the sample and prepare for subsequent instrument analysis.

4. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 1, characterized in that: The step S3 specifically includes: S2-1: Automated sample preparation, using an automated dilution system to dilute the concentrated sample to a concentration suitable for HPLC analysis according to a preset program. The automated filtration device is equipped with a 0.22 μm filter membrane to filter the sample to ensure that there is no interference from particulate matter and protein. For samples that need to be pre-concentrated, an online S2-PE-HPLC system is used for automated pre-concentration; S2-2: Column and mobile phase selection: select a suitable UHPLC column with ultrafine particles C18 column according to the sample characteristics, use expert system software to automatically optimize the mobile phase composition, including the ratio of aqueous phase and organic phase, and set the gradient elution program to ensure the effective separation of different components; S2-3: Intelligent setting of chromatographic conditions, automatic adjustment of flow rate to obtain optimal separation efficiency and peak shape, setting column temperature, usually between 30-40°C to reduce thermal degradation of samples, and setting appropriate detection wavelength according to detector type; S2-4: Instrument operation and data acquisition: start the autosampler, inject samples according to the preset injection sequence and volume, start the data acquisition system, ensure the high resolution and sensitivity of the chromatogram, monitor the system performance in real time, including pressure and column temperature, and automatically adjust the parameters through the feedback control system; S2-5: Chromatographic peak detection and optimization, using machine learning algorithms to automatically identify chromatographic peaks, remove noise and background interference, perform peak shape fitting on chromatographic peaks, and ensure the accuracy of integration; S2-6: Compound identification: Combining retention time and spectral data, the AI-assisted library matching system is used to identify components; for unknown compounds, spectral analysis software is used to infer structures; S2-7: Quantitative analysis, using advanced integration methods for accurate integration, using internal or external standard methods for quantification, and calculating the concentration of each component; S2-8: Data analysis and report generation, using statistical software for data analysis, including precision, accuracy and repeatability assessment, and automatically generating detailed analysis reports including sample information, analysis conditions, results and conclusions.

5. The method for detecting the composition of the waterproof material for building exterior walls according to claim 1, characterized in that: In the step S2-5, data preprocessing is first performed, including data import, cleaning and normalization; then, key chromatographic peak information is extracted through peak detection and feature calculation; on this basis, a noise model is established and background correction is performed to identify and remove interference signals; then, using the training set data, a suitable machine learning algorithm is selected for model training, and the model is optimized through feature selection; the trained model is used to automatically identify chromatographic peaks in new data, and the accuracy of the results is improved through peak confirmation and noise suppression steps; finally, the analysis quality is ensured through result verification and model optimization, and the entire process is automated to establish a continuous monitoring process to achieve efficient and reliable chromatographic data analysis.

6. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 5, characterized in that: In step S4, first, the extract or sample is placed in a derivatization reactor, and a suitable derivatization reagent is selected according to the chemical properties of the sample, including a silanization reagent, an acylating reagent or an alkylating reagent; the derivatization reaction is fully carried out under mild heating or room temperature conditions, and a catalyst is usually added or specific reaction conditions are used to promote the reaction; After the derivatization is completed, the excess derivatization reagent and solvent are removed by nitrogen purging or rotary evaporation, leaving the derivatized product; then, the derivatized product is dissolved in an appropriate solvent for GC-MS injection; In GC-MS analysis, the sample is injected into the gas chromatograph. After separation by the chromatographic column, different compounds are eluted at different time points according to their boiling points and polarities, and enter the mass spectrometer for detection; the mass spectrometer generates a mass spectrum based on the mass-to-charge ratio of the compound, and identifies and quantifies the components in the sample by comparing it with a standard spectral library or calculated retention time.

7. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 1, characterized in that: In step S5, the dried powder sample is first mixed with an appropriate amount of potassium bromide; potassium bromide is an infrared transparent salt used to prepare infrared spectroscopic films because it does not absorb light in the infrared spectrum; during the mixing process, it is necessary to ensure that the ratio of the sample and potassium bromide is appropriate so as to obtain sufficient signal strength and avoid oversaturation; The mixed powder is carefully transferred to a special mold and pressed into a transparent infrared spectrum sheet with the help of a press. During the pressing process, the pressure needs to be evenly distributed to avoid bubbles and cracks, which will affect the transparency of the spectrum sheet and the subsequent spectral analysis results. The pressed infrared spectrum sheet is then placed in the sample chamber of the infrared spectrometer. During the analysis, the infrared beam passes through the spectrum sheet, and the chemical bonds and functional groups in the sample absorb the infrared light according to their specific vibration frequencies, thereby producing characteristic absorption peaks. By comparing the spectra with standards or using the known absorption frequencies of chemical bonds, the chemical structure and composition of the sample can be identified, allowing for qualitative and quantitative analysis of the sample.

8. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 1, characterized in that: In the step S6, first, the sample to be analyzed is dissolved in an appropriate solvent to provide a uniform magnetic field environment; the dissolved sample is placed in an NMR tube and placed in the magnetic field of the NMR spectrometer; in the 1H NMR experiment, the magnetic resonance signal of the hydrogen nucleus is excited and recorded, generating a series of chemical shift peaks, the position, shape and integrated area of ​​these peaks provide information about the hydrogen atom environment in the molecule; in the 13C NMR experiment, the signal of the carbon nucleus is also recorded to provide information about the carbon skeleton; by analyzing the chemical shift, coupling constant, and peak splitting mode parameters, the structure of the carbon-hydrogen skeleton in the molecule, the type of functional group, and the connection mode within the molecule are inferred.

9. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 1, characterized in that: In step S7, first, the sample is cut or ground into nanometer or micrometer-level flakes or particles and fixed on a dedicated sample stage; in SEM operation, the sample usually needs to be coated, including gold or carbon coating, to improve conductivity and image quality; in TEM operation, the sample needs to be thin enough so that the electron beam can penetrate; While obtaining microscopic images, elemental analysis is performed in conjunction with an energy spectrometer; EDS determines the types of elements and relative contents in the sample by detecting the characteristic X-rays emitted by the sample after being excited by an electron beam; this analysis can provide information on the chemical composition of the sample and help researchers understand the relationship between the composition of the sample and its microstructure; through the combined use of SEM and TEM, detailed information on element distribution and chemical state can be obtained while observing the sample morphology.

10. A method for detecting components of a waterproof material for building exterior walls as claimed in claim 1, characterized in that: In step S9, the preprocessed data is organized into a matrix, where each row represents a sample and each column represents a variable, and then the covariance matrix is ​​calculated using the following formula: Where Xi is the i-th sample and Xˉ is the average of all samples; Then perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues ​​and corresponding eigenvectors. Calculate the eigenvalues: By solving the characteristic equation: de t(Cov(X)-λI)=0 to find the eigenvalue λ; For each eigenvalue, by solving the equation: (Cov(X)-λI)v=0 to find the eigenvector v; Then sort the eigenvalues ​​from large to small; select the principal components: select the eigenvectors corresponding to the largest eigenvalues, which represent the main components in the data; Finally, the original data matrix is ​​projected onto the selected principal component to obtain the principal component score matrix T, and the score is calculated: T = XW, where W is the matrix of the selected eigenvectors. The eigenvectors corresponding to each principal component are analyzed to determine which original variables contribute the most to the principal component.

Images