A method and system for calculating volatile organic compounds in exhaust gas from stationary pollution sources

Abstract

This invention discloses a method and system for measuring volatile organic compounds (VOCs) in exhaust gas from stationary pollution sources. The method includes determining the sampling locations and points for exhaust gas emissions from stationary pollution sources; collecting data on the main product outputs and main raw material or fuel consumption of a pre-defined enterprise and establishing a database; collecting exhaust gas using an exhaust gas sampling device; obtaining exhaust gas chromatography using a mass spectrometer; obtaining low molecular weight components using a flame ionization detector (FID) to obtain VOC components; dividing the low molecular weight components into different concentration series and performing gas chromatography analysis to plot working curves; calculating the correlation coefficients between the low molecular weight components and the response area for different concentration series; analyzing other compounds using a mass spectrometer; calculating the correlation coefficients between other compounds and the response area; calculating the concentration of VOCs; establishing pollution source data based on the database; and performing regression analysis using the database and the VOC concentrations to output the results.

Description

Technical Field

[0001] This invention relates to the field of measurement, and in particular to a method and system for measuring volatile organic compounds in exhaust gas emitted from stationary pollution sources. Background Technology

[0002] Volatile organic compounds (VOCs) have various definitions, one of which refers to the general term for organic chemical substances with a vapor pressure above 13.3 Pa and a boiling point below 260℃ under normal conditions (20℃, 101.3 kPa). Structurally, VOCs include short-chain hydrocarbons such as alkanes, alkenes, aromatic hydrocarbons, and alkynes, as well as oxygen-containing organic compounds such as aldehydes and ketones. These compounds not only participate in photochemical reactions and contribute significantly to ozone formation, but some also have obvious carcinogenic effects. Therefore, the state has formulated a series of policies and standards to strictly control VOC pollution. VOCs in ambient air have a wide range of sources, but one important source is the exhaust gas from stationary sources in key VOC-related industries, i.e., exhaust gases emitted into the ambient air through exhaust stacks. These industries include chemical raw material and chemical product manufacturing, pharmaceutical manufacturing, printing, petroleum, and rubber industries.

[0003] In recent years, comprehensive treatment of key VOCs-related industries has been carried out through special campaigns for air pollution control, resulting in continuous improvement in air quality. VOCs testing techniques are a crucial technical support for comprehensive environmental governance. VOCs testing in organized exhaust gas from stationary pollution sources mainly involves sample collection and analysis. Sample collection requires on-site investigation of the sampling point and carrying sampling instruments to collect samples. Generally, the sampling point is not on the ground but at a certain height, requiring personnel to carry the sampler to the point. During sampling, personnel must remain at the sampling instrument. After sampling, the sampling instrument needs to be moved back to its original location, and the collected samples need to be transported to a designated laboratory for analysis. Analysts, upon receiving the samples, need to debug the analytical instruments and analyze the samples according to established methods. Data processing software within the instrument is used to process the data to determine the required pollutant concentrations. All analyses are completed in the laboratory. The entire process requires significant manpower and resources, resulting in low efficiency and high costs.

[0004] Therefore, there is an urgent need for a new rapid method for calculating volatile organic compounds (VOCs) in organized emissions from stationary sources, covering a wider range of pollutants, including more than 100 types of oxygenated organic compounds such as alkanes, alkenes, aromatic hydrocarbons, alkynes, and aldehydes and ketones, adaptable to the exhaust emissions from stationary pollution sources in more industries and enterprises; utilizing artificial intelligence technology to remotely control on-site sampling and laboratory analysis in real time, saving a significant amount of manpower and resources; and fully utilizing data to extract its value. Summary of the Invention

[0005] The purpose of this invention is to provide a method for calculating volatile organic compounds in exhaust gases emitted from stationary pollution sources.

[0006] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0007] This invention includes the following steps:

[0008] Determine the sampling locations and sampling points for exhaust gas emissions from stationary pollution sources, collect data on the output of major products and the consumption of major raw materials or fuels of the pre-selected enterprises, and establish a database.

[0009] Waste gas is collected by a waste gas sampling device, and the waste gas is analyzed by a mass spectrometer to obtain waste gas chromatography data. Coarse mass spectrometry data is obtained, and the low molecular weight components of the waste gas are analyzed by a flame ionization detector (FID) to obtain fine molecular weight data. Volatile organic compounds are obtained based on the coarse mass spectrometry data and the fine molecular weight data.

[0010] The low molecular weight components were divided into different concentration series and analyzed by gas chromatography to plot working curves. The correlation coefficients between the low molecular weight components and the response area were calculated for different concentration series.

[0011] Other compounds are analyzed using a mass spectrometer detector, the correlation coefficient between other compounds and the response area is calculated, the concentration of the volatile organic compounds is calculated, and pollution source data is established based on the database.

[0012] Regression analysis was performed using the database and the concentration of volatile organic compounds, and the calculation results were output.

[0013] Further, the method for obtaining volatile organic compound components based on the coarse mass spectrometry data and the fine fraction data includes:

[0014] Component stratification analysis was performed on coarse mass spectrometry data and fine fraction data. The first layer: C2-C5 hydrocarbons were identified, and they were quickly matched using a retention time index library. FID quantification was adopted, and a retention time prediction model was introduced to correct column aging shift.

[0015] The second layer: For components not covered by FID, machine learning-based mass spectrometry deconvolution is used to separate co-eluent peaks, and fragment particle matching is combined to verify the compound structure;

[0016] The third layer: For unmatched mass spectrometry peaks, cluster the fragmentation patterns of similar mass spectrometry peaks, and infer unidentified compounds based on the company's raw material database;

[0017] A dynamic time warping algorithm is used to automatically correct the retention time deviation between FID and mass spectrometry. If both FID and mass spectrometry detect signals, but the confidence level of mass spectrometry is greater than 0.9, the mass spectrometry result shall prevail. For FID-sensitive components, mass spectrometry is used for verification and outputs the volatile organic compound components.

[0018] Furthermore, a method for dividing the low molecular weight components into different concentration series and performing gas chromatography analysis includes:

[0019] Based on the actual emission concentration range, multiple concentration points are configured to cover the detection linear range. Pure nitrogen is used as the zero concentration point. Two samples are prepared for each concentration point, with a relative deviation of less than or equal to 5%.

[0020] Set the instrument parameters and select the injection method. Repeat the sampling three times at each concentration point and take the average of the peak values. Lock the compound by retention time. According to the separation requirements of low molecular weight components, set the temperature program. The initial temperature is 40°C and held for 3 minutes. The first temperature is increased to 120°C and held at a rate of 10°C per second for 8 minutes to quickly separate low boiling point compounds. The second temperature is increased to 200°C and held at a rate of 5°C per minute for 16 minutes to improve the separation of medium and high boiling point compounds. Finally, the temperature is held at 200°C for 5 minutes until all target components are in the flow.

[0021] Independently control helium and nitrogen gas, given flow rates, obtain the relative response factors of common compounds using the external standard method, select internal standards and obtain their peak areas, and calculate compound concentrations:

[0022]

[0023] The concentration of the i-th compound is C. i The chromatographic peak area of ​​the i-th compound is A. i The relative response factor of the i-th compound is RRF. i The chromatographic peak area of ​​the internal standard is A. os ;

[0024] After analyzing light hydrocarbons, heavy components are immediately backflushed to shorten the cycle. The co-eluting peaks are separated using center cutting technology. A convolutional neural network model is introduced and trained to identify overlapping peaks, and the gas chromatographic analysis results are output.

[0025] Furthermore, the method for plotting the working curve includes:

[0026] Each concentration point was sampled three times, and the average peak area was taken. The peak area was used to identify the target compound by the retention time.

[0027] Input the target compound into the linear regression model, the expression is:

[0028] A = h × C + z

[0029] The chromatographic peak area of ​​the target compound is A, the concentration of the target compound is C, the intercept of the target compound is z, and the slope of the target compound is h.

[0030] When the correlation coefficient of the target compound is greater than 0.995 and the intercept relative response value is greater than 0.05, the corresponding working curve is plotted.

[0031] Furthermore, the method for calculating the correlation coefficient between the low molecular weight component and the response area for different concentration series includes:

[0032] Obtain the peak area of ​​the concentration level and the corresponding low molecular weight component, where the product of the peak area and the concentration is the response area. Calculate the correlation coefficient between the low molecular weight component and the response area.

[0033]

[0034] The correlation coefficient is R, the number of concentration levels is n, and the concentration of the i-th compound is C. i The peak area of ​​the i-th compound is A. i ;

[0035] If the correlation coefficient is greater than or equal to 0.995, the linear relationship between the low molecular weight component and the response area is valid; otherwise, the linear relationship between the low molecular weight component and the response area is not valid. Check for errors in standard sample preparation, abnormal instrument response, or abnormal chromatographic peak integration. If any of these exist, correct the errors and recalculate the correlation coefficient; otherwise, output the results.

[0036] Further, the method for calculating the concentration of the volatile organic compounds includes:

[0037] Calculate the concentration of a certain class of volatile organic compounds based on the chromatographic peak area of ​​the target compound obtained using an FID instrument:

[0038]

[0039] The concentration of the c-th organic compound is The chromatographic peak area of ​​the cth organic compound is A. c The intercept of the working curve is z, and the slope of the working curve is z.

[0040] Calculate the concentration of a certain type of volatile organic compound for a detector with a known relative response factor value:

[0041]

[0042] The relative response factor of the c-th organic compound is RRF. c The concentration of the internal standard is The chromatographic peak area of ​​the internal standard is Aos ;

[0043] The concentration of volatile organic compounds is obtained by summing the concentrations of a certain type of volatile organic compounds.

[0044] Furthermore, the method for performing regression analysis using the database and the volatile organic compound concentration includes:

[0045] Using data from the enterprise database, regression analysis was conducted on the output of major products, the consumption of major raw materials or fuels, and the concentration of volatile organic compounds (VOCs) in waste gas. A regression model was established, expressing the relationship between VOC concentration in waste gas and the output of major products and the consumption of major raw materials or fuels:

[0046] y = β0 + β1x1 + β2x2 + β3x3 + S

[0047] The emission concentration of volatile organic compounds in the exhaust gas is y, the output of the main product is x1, the consumption of the main raw materials is x2, the fuel consumption is x3, and β0, β1, β3, and S are all constants.

[0048] When the explanatory power of the regression analysis model is greater than 0.7 and the significance of the variables is less than 0.05, the predicted output is the regression analysis result.

[0049] Secondly, a system for measuring volatile organic compounds (VOCs) in exhaust gas from stationary pollution sources includes:

[0050] Exhaust gas sampling module: used to determine the sampling location and sampling point of exhaust gas emissions from stationary pollution sources, collect the output of main products, main raw materials or fuel consumption of preset enterprises and establish a database;

[0051] Component analysis module: used to collect waste gas through a waste gas sampling device, perform component analysis on the waste gas using a mass spectrometer to obtain waste gas chromatography, obtain coarse mass spectrometry data, analyze the low molecular weight components of the waste gas using a hydrogen flame ionization detector to obtain fine molecular weight data, and obtain volatile organic compound components based on the coarse mass spectrometry data and the fine molecular weight data;

[0052] Chromatography-related module: used to divide the low molecular weight components into different concentration series, perform gas chromatography analysis, plot working curves, and calculate the correlation coefficient between the low molecular weight components and the response area for different concentration series;

[0053] Concentration calculation module: used to analyze other compounds using a mass spectrometer detector, calculate the correlation coefficient between other compounds and the response area, calculate the concentration of the volatile organic compounds, and establish pollution source data based on the database;

[0054] Modeling and calculation module: used to perform regression analysis using the database and the concentration of volatile organic compounds, and output the calculation results.

[0055] The beneficial effects of this invention are:

[0056] This invention provides a method and system for calculating volatile organic compounds (VOCs) in exhaust gases from stationary pollution sources. Compared with existing technologies, this invention offers the following technical advantages:

[0057] This invention improves the accuracy of volatile organic compound (VOC) measurement in stationary source emissions by employing steps such as component analysis, molecular low-level component analysis, VOC composition determination, chromatographic analysis, correlation coefficient calculation, and regression analysis. This optimization of VOC measurement significantly saves resources and improves work efficiency. It enables intelligent measurement of VOCs in stationary source emissions, allowing for real-time determination of VOC composition and regression analysis. This method is of great significance for VOC measurement in stationary source emissions, adaptable to different standards and varying VOC measurement needs, and possesses a certain degree of universality. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating the steps of a method for calculating volatile organic compounds in exhaust gas from stationary pollution sources according to the present invention.

[0059] Figure 2 This is a standard curve showing the relationship between different concentrations of low molecular weight molecules and response area in a specific embodiment of the present invention. Detailed Implementation

[0060] The present invention will be further described below through specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0061] The present invention discloses a method and system for calculating volatile organic compounds in exhaust gas from stationary pollution sources, comprising the following steps:

[0062] like Figure 1 As shown, this embodiment includes the following steps:

[0063] Determine the sampling locations and sampling points for exhaust gas emissions from stationary pollution sources, collect data on the output of major products and the consumption of major raw materials or fuels of the pre-selected enterprises, and establish a database.

[0064] In actual assessment, on-site inspections are conducted in accordance with national regulations on waste gas emission outlets; sampling locations and points for waste gas emissions from stationary pollution sources are determined; data on the output of major products and the consumption of major raw materials or fuels of enterprises are collected to establish a database; and waste gas is sampled after verifying that the production equipment and pollutant treatment facilities are operating normally.

[0065] According to the technical requirements, connect the exhaust gas sampling device, turn on the power, turn on the heating button, set the required heating temperature, and start sampling for 10 seconds. Then connect the vacuum bottle. The exhaust gas collection device is a 1L brown glass vacuum bottle with deactivated inner walls and a volume of 1L. Use a sampling gun with heating function to draw gas. The sample gas is passed through a three-way valve according to certain flow restriction requirements. Prepare an inert flow restrictor with a flow rate of 1L / h and collect it into the vacuum sampling bottle. The sampling time is 1 hour. After sampling is completed, push the quick-connect cathode head, remove the glass sampling bottle, and transport the collected sample bottle back to the laboratory at low temperature.

[0066] Waste gas is collected by a waste gas sampling device, and the waste gas is analyzed by a mass spectrometer to obtain waste gas chromatography data. Coarse mass spectrometry data is obtained, and the low molecular weight components of the waste gas are analyzed by a flame ionization detector (FID) to obtain fine molecular weight data. Volatile organic compounds are obtained based on the coarse mass spectrometry data and the fine molecular weight data.

[0067] In the actual evaluation, the low molecular weight components were propane, ethylene, acetylene, ethane and propylene; chromatographic column 1 was a capillary column (60m*0.25mm*1μm); chromatographic column 2 was a capillary column (30m*0.32mm*20μm); chromatographic column 3 was a damping column (0.67m*0.1mm).

[0068] Propane, ethylene, acetylene, ethane, and propylene with low molecular weights were analyzed using a flame ionization detector (FID). Concentration series of propane, ethylene, acetylene, ethane, and propylene at concentrations of 0.2, 0.5, 1, 2, 4, 6, 8, and 10 ppb were prepared for gas chromatography / FID analysis. Based on the experimental data, a standard curve was plotted, with the concentration series in ppb as the x-axis and the corresponding peak area as the y-axis, to establish a working curve.

[0069] The low molecular weight components were divided into different concentration series and analyzed by gas chromatography to plot working curves. The correlation coefficients between the low molecular weight components and the response area were calculated for different concentration series.

[0070] In actual evaluation, the correlation coefficients R between the concentrations of propane, ethylene, acetylene, ethane and propylene and the response area were calculated, and the results were all above 0.9998, indicating a strong linear relationship.

[0071] Other compounds are analyzed using a mass spectrometer detector, the correlation coefficient between other compounds and the response area is calculated, the concentration of the volatile organic compounds is calculated, and pollution source data is established based on the database.

[0072] Regression analysis was performed using the database and the concentration of volatile organic compounds, and the calculation results were output.

[0073] In this embodiment, the method for obtaining volatile organic compound components based on the coarse mass spectrometry data and the fine fraction data includes:

[0074] Component stratification analysis was performed on coarse mass spectrometry data and fine fraction data. The first layer: C2-C5 hydrocarbons were identified, and they were quickly matched using a retention time index library. FID quantification was adopted, and a retention time prediction model was introduced to correct column aging shift.

[0075] The second layer: For components not covered by FID, machine learning-based mass spectrometry deconvolution is used to separate co-eluent peaks, and fragment particle matching is combined to verify the compound structure;

[0076] The third layer: For unmatched mass spectrometry peaks, cluster the fragmentation patterns of similar mass spectrometry peaks, and infer unidentified compounds based on the company's raw material database;

[0077] A dynamic time warping algorithm is used to automatically correct the retention time deviation between FID and mass spectrometry. If both FID and mass spectrometry detect signals, but the confidence level of mass spectrometry is greater than 0.9, the mass spectrometry result shall prevail. For FID-sensitive components, mass spectrometry is used for verification and outputs the volatile organic compound components.

[0078] In this embodiment, the method for dividing the low molecular weight components into different concentration series and performing gas chromatography analysis includes:

[0079] Based on the actual emission concentration range, multiple concentration points are configured to cover the detection linear range. Pure nitrogen is used as the zero concentration point. Two samples are prepared for each concentration point, with a relative deviation of less than or equal to 5%.

[0080] Set the instrument parameters and select the injection method. Repeat the sampling three times at each concentration point and take the average of the peak values. Lock the compound by retention time. According to the separation requirements of low molecular weight components, set the temperature program. The initial temperature is 40°C and held for 3 minutes. The first temperature is increased to 120°C and held at a rate of 10°C per second for 8 minutes to quickly separate low boiling point compounds. The second temperature is increased to 200°C and held at a rate of 5°C per minute for 16 minutes to improve the separation of medium and high boiling point compounds. Finally, the temperature is held at 200°C for 5 minutes until all target components are in the flow.

[0081] Independently control helium and nitrogen gas, given flow rates, obtain the relative response factors of common compounds using the external standard method, select internal standards and obtain their peak areas, and calculate compound concentrations:

[0082]

[0083] The concentration of the i-th compound is C. i The chromatographic peak area of ​​the i-th compound is A. i The relative response factor of the i-th compound is RRF. i The chromatographic peak area of ​​the internal standard is A. os ;

[0084] After analyzing light hydrocarbons, heavy components are immediately backflushed to shorten the cycle. The co-eluting peaks are separated using center cutting technology. A convolutional neural network model is introduced and trained to identify overlapping peaks, and the gas chromatographic analysis results are output.

[0085] In this embodiment, the method for drawing the working curve includes:

[0086] Each concentration point was sampled three times, and the average peak area was taken. The peak area was used to identify the target compound by the retention time.

[0087] Input the target compound into the linear regression model, the expression is:

[0088] A = h × C + z

[0089] The chromatographic peak area of ​​the target compound is A, the concentration of the target compound is C, the intercept of the target compound is z, and the slope of the target compound is h.

[0090] When the correlation coefficient of the target compound is greater than 0.995 and the intercept relative response value is greater than 0.05, the corresponding working curve is plotted.

[0091] In this embodiment, the method for calculating the correlation coefficient between the low molecular weight component and the response area for different concentration series includes:

[0092] Obtain the peak area of ​​the concentration level and the corresponding low molecular weight component, where the product of the peak area and the concentration is the response area. Calculate the correlation coefficient between the low molecular weight component and the response area.

[0093]

[0094] The correlation coefficient is R, the number of concentration levels is n, and the concentration of the i-th compound is C. i The peak area of ​​the i-th compound is A. i ;

[0095] If the correlation coefficient is greater than or equal to 0.995, the linear relationship between the low molecular weight component and the response area is valid; otherwise, the linear relationship between the low molecular weight component and the response area is not valid. Check for errors in standard sample preparation, abnormal instrument response, or abnormal chromatographic peak integration. If any of these exist, correct the errors and recalculate the correlation coefficient; otherwise, output the results.

[0096] In this embodiment, the method for calculating the concentration of the volatile organic compounds includes:

[0097] Calculate the concentration of a certain class of volatile organic compounds based on the chromatographic peak area of ​​the target compound obtained using an FID instrument:

[0098]

[0099] The concentration of the c-th organic compound is The chromatographic peak area of ​​the cth organic compound is A. c The intercept of the working curve is z, and the slope of the working curve is z.

[0100] Calculate the concentration of a certain type of volatile organic compound for a detector with a known relative response factor value:

[0101]

[0102] The relative response factor of the c-th organic compound is RRF. c The concentration of the internal standard is The chromatographic peak area of ​​the internal standard is A os ;

[0103] The concentration of volatile organic compounds is obtained by summing the concentrations of a certain type of volatile organic compounds.

[0104] In this embodiment, the method for performing regression analysis using the database and the concentration of volatile organic compounds includes:

[0105] Using data from the enterprise database, regression analysis was conducted on the output of major products, the consumption of major raw materials or fuels, and the concentration of volatile organic compounds (VOCs) in waste gas. A regression model was established, expressing the relationship between VOC concentration in waste gas and the output of major products and the consumption of major raw materials or fuels:

[0106] y = β0 + β1x1 + β2x2 + β3x3 + S

[0107] The emission concentration of volatile organic compounds in the exhaust gas is y, the output of the main product is x1, the consumption of the main raw materials is x2, the fuel consumption is x3, and β0, β1, β3, and S are all constants.

[0108] When the explanatory power of the regression analysis model is greater than 0.7 and the significance of the variables is less than 0.05, the predicted output is the regression analysis result.

[0109] Secondly, a system for measuring volatile organic compounds (VOCs) in exhaust gas from stationary pollution sources includes:

[0110] Exhaust gas sampling module: used to determine the sampling location and sampling point of exhaust gas emissions from stationary pollution sources, collect the output of main products, main raw materials or fuel consumption of preset enterprises and establish a database;

[0111] Component analysis module: used to collect waste gas through a waste gas sampling device, perform component analysis on the waste gas using a mass spectrometer to obtain waste gas chromatography, obtain coarse mass spectrometry data, analyze the low molecular weight components of the waste gas using a hydrogen flame ionization detector to obtain fine molecular weight data, and obtain volatile organic compound components based on the coarse mass spectrometry data and the fine molecular weight data;

[0112] Chromatography-related module: used to divide the low molecular weight components into different concentration series, perform gas chromatography analysis, plot working curves, and calculate the correlation coefficient between the low molecular weight components and the response area for different concentration series;

[0113] Concentration calculation module: used to analyze other compounds using a mass spectrometer detector, calculate the correlation coefficient between other compounds and the response area, calculate the concentration of the volatile organic compounds, and establish pollution source data based on the database;

[0114] Modeling and calculation module: used to perform regression analysis using the database and the concentration of volatile organic compounds, and output the calculation results.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for measuring volatile organic compounds in exhaust gas emitted from a stationary pollution source, characterized by, The method comprises the following steps: Determine the sampling position and sampling point of the exhaust emission of stationary sources, collect the main product output, main raw material or fuel consumption of the preset enterprise and establish a database; Collect the exhaust gas by an exhaust gas sampling device, analyze the composition of the exhaust gas by a mass spectrometer to obtain an exhaust gas chromatogram, obtain coarse mass spectrum data, analyze the low molecular weight components of the exhaust gas by a hydrogen flame ionization detector (FID) to obtain fine molecular data, and obtain volatile organic components according to the coarse mass spectrum data and the fine molecular data; comprising: Layered analysis of coarse mass spectrum data and fine molecular data, first layer: lock C2-C5 hydrocarbons, quickly match by retention time index library, use FID quantification, introduce retention time prediction model, correct chromatographic column aging offset; Second layer: for components not covered by FID, use machine learning-based mass spectrometry to separate co-elution peaks, combine fragment particle matching to verify compound structure; Third layer: for unmatched mass spectrum peaks, cluster similar mass spectrum peak fragment patterns, and infer uncalibrated compounds according to the enterprise raw material database; Use dynamic time warping algorithm to automatically correct the time deviation generated by different chromatographic analysis processes in the FID and mass spectrometry analysis system, construct a time sequence matching model of the FID detection signal and the mass spectrometry particle flow signal, and dynamically correct the retention time offset of the FID detection signal and the mass spectrometry particle flow signal in the mass spectrum peak process. If both FID and mass spectrometry detect signals, but the mass spectrometry confidence is greater than 0.9, the mass spectrometry result is used as the reference; for FID sensitive components, use mass spectrometry for verification, and output volatile organic components; Distribute the low molecular weight components into different concentration series, perform gas chromatography analysis, and draw a working curve to calculate the correlation coefficient of the low molecular weight components in different concentration series and the response area; Analyze other compounds by a mass spectrometer, calculate the correlation coefficient of other compounds and the response area, calculate the concentration of the volatile organic compounds, and establish a pollution source data according to the database; Use the database and the volatile organic compound concentration for regression analysis, and output the calculation result.

2. The method according to claim 1, wherein the method is used for measuring the concentration of volatile organic compounds in exhaust gas from a stationary pollution source. The method for distributing the low molecular weight components into different concentration series and performing gas chromatography analysis, comprising: According to the actual emission concentration range, configure multiple concentration points to cover the detection linear interval, use pure nitrogen as the zero concentration point, and prepare double samples for each concentration point with a relative deviation of less than or equal to 5%; Set the instrument parameters and select the sampling mode, repeat sampling three times for each concentration point, take the peak value average, lock the compound by retention time, set the temperature rising program according to the separation requirements of the low molecular weight components, and set the initial temperature to 40 degrees for 3 minutes; the first temperature rising is to 120 degrees at a rate of 10 degrees per second for 8 minutes to quickly separate low-boiling-point compounds; the second temperature rising is to 200 degrees at a rate of 5 degrees per minute for 16 minutes to improve the separation degree of medium and high-boiling-point compounds; finally, keep the temperature at 200 degrees for 5 minutes until all target components flow out; Independently control helium and nitrogen, give a flow rate, get the relative response factor of common compounds in external standard method, select an internal standard and get the peak area of the internal standard, and calculate the concentration of the compound: wherein the concentration of the ith compound is , the chromatographic peak area of the ith compound is , the relative response factor of the ith compound is , and the chromatographic peak area of the internal standard is ; Immediately after analyzing light hydrocarbons, blow back heavy components to shorten the cycle, use center cutting technology to separate co-eluting peaks, introduce a convolutional neural network model, train the convolutional neural network model to identify overlapping peaks, and output the gas chromatography analysis results.

3. The method according to claim 1, wherein the method is used for measuring the concentration of volatile organic compounds in stationary source exhaust gas. The method for drawing the working curve comprises: Each concentration point is repeated 3 times, and the peak area is averaged, wherein the peak area is locked by the retention time of the target compound; The target compound is input into a linear regression model, and the expression is: where the peak area of the target compound is , the concentration of the target compound is , the intercept of the target compound is z, and the slope of the target compound is h; When the correlation coefficient of the target compound is greater than 0.995, and the intercept relative response value is greater than 0.05, the corresponding working curve is drawn.

4. The method according to claim 1, wherein the method is used for measuring the concentration of volatile organic compounds in stationary source exhaust gas. The method for calculating the correlation coefficient of the molecular weight low component and the response area at different concentration series comprises: Get the peak area of the concentration level and the corresponding molecular weight low component, wherein the product of the peak area and the concentration is the response area, and calculate the correlation coefficient of the molecular weight low component and the response area: where the correlation coefficient is , the number of concentration levels is , the concentration of the ith compound is , the peak area of the ith compound is ; If the correlation coefficient is greater than or equal to 0.995, the linear relationship between the molecular weight low component and the response area is established; otherwise, the linear relationship between the molecular weight low component and the response area is not established, and whether there is a sample preparation error, an instrument response abnormality, or a chromatographic peak integration abnormality is checked, if there is, the error is corrected and the correlation coefficient is recalculated, otherwise the result is output.

5. The method of claim 1, wherein the method is used for measuring the concentration of volatile organic compounds in exhaust gas from a stationary pollution source. The method for calculating the concentration of the volatile organic compound comprises: For the chromatographic peak area of the target compound obtained by the FID device, the concentration of a certain type of volatile organic compound is calculated: wherein the concentration of the cth organic is , the chromatographic peak area of the cth organic is , the intercept of the working curve is z, and the slope of the working curve is z; For the detector with a known relative response factor value, the concentration of a certain type of volatile organic compound is calculated: wherein the relative response factor of the cth organic is , the concentration of the internal standard is , the chromatographic peak area of the internal standard is ; The concentration of a certain type of volatile organic compound is summed to obtain the concentration of the volatile organic compound.

6. The method of claim 1, wherein the method is used for measuring the concentration of volatile organic compounds in stationary source exhaust gas. The method for performing regression analysis on the database and the volatile organic compound concentration comprises: Using the data in the enterprise database, the main product yield, the main raw material or fuel consumption data, and the volatile organic compound concentration in the waste gas are subjected to regression analysis, and a regression analysis model of the volatile organic compound concentration in the waste gas and the main product yield, the main raw material or fuel consumption is established, and the expression is: wherein the emission concentration of volatile organic compounds in the exhaust gas is , the main product yield is , the main raw material consumption is , the fuel consumption is , , , , , S are constants; When the model explanation of the regression analysis model is greater than 0.7 and the variable significance is less than 0.05, the regression analysis result is predicted and output.

7. A system for measuring volatile organic compounds in exhaust gas from a stationary pollution source for performing the method according to any one of claims 1 to 6, characterized in that It comprises: A waste gas sampling module for determining the sampling position and sampling point of the waste gas emission of a fixed pollution source, collecting the main product yield, the main raw material or fuel consumption of a predetermined enterprise, and establishing a database; A component analysis module for collecting waste gas by a waste gas sampling device, analyzing the components of the waste gas by a mass spectrometer detector to obtain waste gas chromatography, obtaining coarse mass spectrum data, analyzing the molecular weight low components of the waste gas by a hydrogen flame ionization detector to obtain fine molecular data, and obtaining volatile organic compound components according to the coarse mass spectrum data and the fine molecular data; A chromatographic correlation module for distributing the molecular weight low components into different concentration series, performing gas chromatography analysis, drawing a working curve, and calculating the correlation coefficient of the molecular weight low components and the response area at different concentration series. A concentration calculation module is configured to analyze other compounds by using a mass spectrometer, calculate a correlation coefficient of the other compounds and a response area, calculate a concentration of the volatile organic compounds, and establish pollution source data according to the database; A modeling calculation module is configured to perform regression analysis on the database and the concentration of the volatile organic compounds, and output a calculation result.

Images