A VOCs source analysis method, device, medium and product based on flux data

Abstract

The application discloses a VOCs source analysis method and device based on flux data, a medium and a product, relates to the technical field of VOCs source analysis, and comprises the following steps: acquiring VOCs species data in a to-be-tested gas according to mass spectrum peak data; calculating VOCs flux data of the species according to wind speed data, VOCs species name data and corresponding species concentration data; calculating a VOCs flux error matrix according to the VOCs flux data; and inputting the VOCs flux matrix, the VOCs flux error matrix, the VOCs species name data, mass-to-charge ratio data and time axis data into a PMF model for source analysis. The chemical information of flux measurement is affected very little by photochemical reactions, and the primary source of VOCs can be directly identified, so that the application can directly represent the source characteristics of VOCs emission, thereby effectively improving the accuracy of VOCs source analysis.

Description

Technical Field

[0001] This application relates to the field of VOCs source analysis technology, and in particular to a method, device, medium and product for VOCs source analysis based on flux data. Background Technology

[0002] Source apportionment technology is a technique that uses chemical, physical, and mathematical methods to qualitatively or quantitatively identify the sources of air pollutants in the environment. Volatile organic compounds (VOCs) refer to organic chemical substances with a vapor pressure above 13.3 Pa and a boiling point below 260℃ under normal conditions (20℃, 101.3 kPa). Currently, most methods for measuring VOC sources are source apportionment methods based on VOC concentration. The problem with this method is that it is affected by meteorological factors such as boundary layer height, air mass transport, and atmospheric photochemical reactions, and cannot directly measure VOC emissions and deposition. This results in significant uncertainty and analytical difficulty for VOC concentration-based source apportionment methods. Furthermore, this method struggles to accurately quantify the relative contributions of primary emissions and secondary formations of VOCs. In practical applications, the concentration of many VOC species has secondary formation sources, which introduces considerable uncertainty into the VOC source results. Summary of the Invention

[0003] The purpose of this application is to provide a method, device, medium, and product for VOCs source analysis based on flux data, which can effectively improve the accuracy of VOCs source analysis.

[0004] To achieve the above objectives, this application provides the following solution:

[0005] Firstly, this application provides a method for VOCs source analysis based on flux data, including:

[0006] Acquire wind speed data, mass spectrometry peak data of the gas to be measured, and time axis data during mass spectrometry peak measurement;

[0007] VOC species data in the gas to be tested are obtained based on the mass spectrometry peak data; the VOC species data includes: VOC species name data, corresponding species concentration data and corresponding mass-to-charge ratio data;

[0008] The VOC flux data of each species is calculated based on the wind speed data, the VOC species name data, and the corresponding species concentration data.

[0009] The VOCs flux error matrix is ​​calculated based on the VOCs flux data; the VOCs flux error matrix includes: a measurement error matrix, a systematic error matrix, and a random error matrix;

[0010] The VOCs flux data is converted into a VOCs flux matrix;

[0011] The VOCs flux matrix, the VOCs flux error matrix, the VOCs species name data, the mass-to-charge ratio data, and the time axis data are input into an orthogonal matrix factorization model to obtain the number of factors, the source component spectrum of the factors, and the flux time series of the factors.

[0012] The pollution source represented by the factor is determined based on the source component spectrum of the factor;

[0013] The specific emission sources of the factor are determined based on the flux time series of the factor.

[0014] Secondly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the VOCs source analysis method based on flux data as described above.

[0015] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the VOCs source analysis method based on flux data described above.

[0016] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the VOCs source analysis method based on flux data described above.

[0017] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0018] This application provides a method, device, medium, and product for VOCs source apportionment based on flux data. First, wind speed data, mass spectrometry peak data of the gas to be tested, and time axis data during mass spectrometry peak measurement are acquired. Then, VOCs species data in the gas to be tested are obtained based on the mass spectrometry peak data. The VOCs species data includes: VOCs species name data, corresponding species concentration data, and corresponding mass-to-charge ratio data. The VOCs flux data of each species is calculated based on the wind speed data, the VOCs species name data, and the corresponding species concentration data. Finally, the VOCs flux is calculated based on the VOCs flux data. The VOCs flux error matrix includes a measurement error matrix, a systematic error matrix, and a random error matrix. The VOCs flux data are converted into a VOCs flux matrix. The VOCs flux matrix, the VOCs flux error matrix, the VOCs species name data, the mass-to-charge ratio data, and the time axis data are input into a PMF model to obtain the number of factors, the source composition spectrum of the factors, and the flux time series of the factors. The pollution sources represented by the factors are determined based on the source composition spectrum of the factors. The specific emission sources of the factors are determined based on the flux time series of the factors. In flux observation, the characteristic transport time of flux turbulence is affected by the surface friction velocity and the measurement height of the measuring instrument, typically only a few seconds to a few minutes. This means that the chemical information of atmospheric flux measurement is minimally affected by photochemical reactions, thus directly identifying the primary sources of VOCs and eliminating the secondary generation portion of VOCs. Therefore, the measurement method based on flux data in this application can directly characterize the source characteristics of VOCs emissions, thereby effectively improving the accuracy of VOCs source apportionment. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a VOCs source analysis method based on flux data, provided in an embodiment of this application;

[0021] Figure 2 A flowchart illustrating a VOCs source analysis method based on flux data, provided for another embodiment of this application;

[0022] Figure 3 A schematic diagram of a high-throughput species screening mechanism provided in another embodiment of this application;

[0023] Figure 4A schematic diagram showing the proportion of factor time series and the average contribution of five factors to total VOCs flux, provided for another embodiment of this application; Figure 4 (a) is a schematic diagram of the proportion of factor time series. Figure 4 (b) is a schematic diagram showing the average contribution of the five factors to the total VOC flux.

[0024] Figure 5 A schematic diagram illustrating the average contribution of five factors to a specific species in VOCs flux source apportionment, as provided in another embodiment of this application;

[0025] Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] Numerous studies have analyzed the sources of VOCs in different regions (e.g., urban areas, forest areas), but these methods are all based on VOCs concentration source apportionment. However, this approach has limitations. Atmospheric reactive gas concentrations are influenced by meteorological factors such as boundary layer height, air mass transport, and atmospheric photochemical reactions. Therefore, atmospheric concentration cannot directly reflect the emission and deposition of substances, making source apportionment based on concentration data highly uncertain and difficult to analyze. Flux measurement, on the other hand, captures atmospheric turbulence. The classic turbulence timescale ranges from a few seconds to 10 minutes, which means that VOCs are significantly less affected by photochemical processes than VOCs concentration measurements, thus providing a more direct characterization of VOCs emission sources. Although there are some studies on VOCs flux measurement both domestically and internationally, there is a lack of source apportionment studies based on VOCs flux measurement data, resulting in an inability to directly characterize the sources of VOCs emissions. Therefore, in order to overcome the shortcomings of current VOCs concentration source apportionment methods and broaden the scope of VOCs source apportionment methods, this invention uses the orthogonal matrix factorization (PMF) model to perform source apportionment of VOCs flux, identifying the main VOCs flux emission sources and the relative contributions of quantitative emission sources.

[0028] Source apportionment technology is a research method that uses chemical, physical, and mathematical methods to qualitatively or quantitatively identify the sources of air pollutants in the environment. Currently, VOCs source analysis based on measurement is mostly based on VOCs concentration. The biggest problem with this method is that it is affected by meteorological factors such as boundary layer height, air mass transport, and atmospheric photochemical reactions, making it impossible to directly measure VOCs emissions and deposition. This results in significant uncertainty and analytical difficulty in source apportionment based on concentration data. Furthermore, this method struggles to accurately quantify the relative contributions of primary and secondary VOCs emissions. In practical applications, the concentrations of many VOC species have secondary sources, introducing considerable uncertainty into the VOCs source analysis results.

[0029] Therefore, addressing the limitations of existing technologies, this application proposes a VOCs source apportionment method based on VOCs flux data. In flux observation, the characteristic transport time of flux turbulence (i.e., the turbulent timescale between the emission source and sensor detection) is affected by surface friction velocity and the measurement height of the instrument, typically ranging from a few seconds to a few minutes. This means that the chemical information from atmospheric flux measurements is minimally affected by photochemical reactions (compared to concentration measurement methods), directly eliminating the secondary generation of VOCs and thus directly characterizing the source features of VOCs emissions. Addressing the key scientific issue of high uncertainty in understanding VOCs emissions, a VOCs flux source apportionment method based on the PMF model is constructed to identify the main VOCs flux emission sources in different regions and quantify the specific contributions of different emission source types to various VOCs species. A comparison of VOCs source apportionment calculation results based on flux measurement and concentration measurement shows that this method effectively reduces the uncertainty of VOCs source apportionment methods, thereby more accurately identifying and quantifying VOCs sources.

[0030] This application proposes a source apportionment method based on VOCs flux data, which can identify the main VOCs emission sources in different regions, filling the gap in source apportionment technology based on VOCs flux data. Based on VOCs flux data, this application applies an orthogonal matrix factorization (PMF) model for source apportionment to identify and quantify the contributions of major VOCs emission sources in different regions, thus filling the gap in source apportionment technology based on VOCs flux data. Simultaneously, it addresses the problem of VOCs concentration-based source apportionment being affected by meteorological factors and atmospheric photochemical reactions, representing a supplement and advancement to VOCs concentration-based source apportionment methods, and reducing the uncertainty of VOCs source apportionment methods.

[0031] This includes explanations of some technical terms related to the technical solutions:

[0032] Orthogonal matrix factorization (PMF) is a multivariate statistical method widely used in environmental science, particularly in the apportionment of atmospheric pollutant sources. The PMF model decomposes the complex pollution source contributions into different source components and their respective contribution rates by breaking down the sample data matrix collected from receptor sites.

[0033] Oxygenated volatile organic compounds (OVOCs) are a class of organic compounds that play an important role in atmospheric chemical processes. They are mainly composed of aldehydes and ketones, alcohols, ethers, low-molecular-weight organic acids, organic esters, and compounds such as alkenes and ketones.

[0034] Flux: Flux is a physics concept that refers to the amount of physical quantities such as momentum, heat, and matter transported per unit area of ​​an interface per unit time.

[0035] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] In one exemplary embodiment, such as Figure 1 As shown, a VOCs source analysis method based on flux data is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is described using a server as an example, and includes the following steps S1 to S8. Wherein:

[0037] S1. Acquire wind speed data, mass spectrum peak data of the gas to be measured, and time axis data during mass spectrum peak measurement.

[0038] S2. Obtain VOCs species data in the gas to be tested based on the mass spectrometry peak data; the VOCs species data includes: VOCs species name data, corresponding species concentration data and corresponding mass-to-charge ratio data.

[0039] In this embodiment, the raw data measured by PTR-ToF-MS (proton reaction time-of-flight mass spectrometry) were used to perform high-resolution fitting on the measured mass spectrum peaks using mass spectrometry instrument data analysis software to obtain VOC species name data, corresponding mass-to-charge ratio data, and species signal data measured during the mass spectrometry measurement.

[0040] Based on species signal data and combined with species response factors obtained through calibration experiments using chemical standards, the species signal data is converted into species concentration data, and the calculation formula is shown below:

[0041]

[0042] In the formula, [X] represents species concentration, Signal represents species signal, and Sensitivity represents species response factor.

[0043] The limit of detection (LOD) is then calculated using the formula shown below:

[0044]

[0045] In the formula, DL is the detection limit for the species concentration; C f B and B are the average signal and background signal of the species during mass spectrometry measurement; t is the temporal resolution of VOCs flux data during mass spectrometry measurement (i.e., the measurement interval required to obtain one flux data, typically 30 min); SNR is the signal-to-noise ratio of VOCs measurement; and the detection limit is set to 3.

[0046] VOC species with concentrations above the detection limit were screened out by calculating the detection limit, while VOC species with average concentrations below the detection limit were not included in subsequent analyses.

[0047] S3. Calculate the VOC flux data of the species based on the wind speed data, the VOC species name data and the corresponding species concentration data.

[0048] In this embodiment, VOCs flux data are calculated by combining wind speed data measured by a three-dimensional ultrasonic anemometer:

[0049]

[0050] In the formula, w' and c' represent the change in vertical wind speed and the change in concentration of a certain VOCs, respectively.

[0051] Then, based on the species concentration data and the wind speed data, the VOCs species data are labeled with data quality using steady-state testing and complete turbulence characteristic testing methods to obtain data quality labeling results; VOCs species data with quality labeling results higher than a preset threshold are deleted. Labeled data for each VOCs species is obtained, and data points that do not meet the flux measurement conditions (data labeling value greater than 6) are removed.

[0052] In this embodiment, to ensure the quality of VOCs flux data, a data quality labeling method is used to filter the data. The quality labeling method for flux data includes steady-state testing and complete turbulence characteristic testing. The specific operation steps are as follows:

[0053] The first step involves calculating steady-state condition parameters and turbulent development characteristic parameters by combining the measured VOCs concentration data and wind speed data. The parameter results are then obtained based on the data label classification range (Table 1), and the data label values ​​of the parameters are obtained. The calculation methods for steady-state condition parameters and turbulent development condition parameters are shown in the following formulas.

[0054] Table 1 Data Label Classification Range Table

[0055]

[0056] The second step is to select the highest number of data labels from the steady-state test and the complete turbulence characteristic test as the total label (e.g., the steady-state test is 3 and the turbulence development characteristic test is 5, then the total label is 5).

[0057] The third step is to use the data labeled 1-6 for data analysis, while the data labeled 7-10 is removed.

[0058]

[0059] In the formula, x s Represents the horizontal wind component or scalar, RN x Steady-state condition parameters for representative species over an average interval (30 min). The formulas for calculating turbulent development condition parameters are as follows:

[0060]

[0061] In the formula, ITC σ These are characteristic parameters for turbulence development. and These represent the wind steady-state parameters obtained from the theoretical model and the measured wind steady-state parameters under neutral conditions, respectively.

[0062] Based on VOCs concentration and wind speed data, the detection limit for species flux is calculated. Species with average VOCs flux below their detection limit are removed, and the remaining VOCs species meeting their detection limits proceed to the next step of analysis. The formula for calculating the detection limit is:

[0063] LOD flux =3*RE noise ;

[0064] Among them, random error Similar to the formula used to calculate flux, but in terms of data, the wind speed and concentration data are randomly shifted forward or backward (creating data misalignment to introduce noise into the flux measurement) by 300-900 data points. In the formula, LOD... flux and RE noise These represent the flux detection limit and the random error in VOCs flux measurement, respectively.

[0065] S4. Calculate the VOCs flux error matrix based on the VOCs flux data; the VOCs flux error matrix includes: measurement error matrix, systematic error matrix and random error matrix.

[0066] In this embodiment, the error mainly comes from three aspects. By considering the three aspects of measurement error, systematic error and random error, the flux error is obtained. Among them, the matrix is ​​a calculation form.

[0067] S5. Convert the VOCs flux data into a VOCs flux matrix.

[0068] The VOCs flux data are synthesized into a VOCs flux matrix. If there are i VOCs species, and each species has j data points, then a matrix with i columns and j rows is generated. In the matrix, for flux data points that are less than 0 or are null, the median flux of that species is used instead.

[0069] The second step is to calculate the error U of the PTR-ToF-MS measurement of VOCs concentration for all VOCs species after screening. PTR,ij The systematic error in flux measurement, and the random error RE in flux measurement. noise The method for calculating the VOCs concentration error is shown in the following formula:

[0070]

[0071] In the formula, U PTR,ij This refers to the error in measuring the concentration of species j using PTR-ToF-MS within time period i. SNR ij is the signal-to-noise ratio of species j within time period i, used to assess the counting statistical error during instrument measurement, and is calculated using the following formula. ij p is the average concentration of j during time period i (30 min). ij This refers to the error rate during PTR-ToF-MS measurement, set at 20%. Of this, 10% originates from instrument operational fluctuations, such as ion source temperature fluctuations and precursor ion signal fluctuations. The other 10% comes from errors caused by calibration using standard gases during long-term observations. DL is the instrument detection limit for species j, calculated as follows:

[0072]

[0073] In the formula, SNR ij C is the signal-to-noise ratio of species j in the i-th time period; f，j B and B are the average and background signals for species j during mass spectrometry measurements; t is the time interval (30 min) used in the VOCs flux calculation in PMF. Since the systematic error of flux measurement can be estimated from the flux averaging interval, it is considered negligible.

[0074] Calculate the VOCs flux error. The VOCs flux error is obtained by adding the error in VOCs concentration and the random error in flux measurement. However, not all of the VOCs concentration error matrix is ​​propagated to the VOCs flux measurement. It is estimated that approximately half of the VOCs concentration error is propagated to the VOCs flux measurement error. Therefore, the VOCs flux error is calculated as follows:

[0075]

[0076] In the formula, U total,ij The error in measuring the flux of species j in PTR-ToF-MS during time period i, RE noise,ij The random error in measuring the flux of species j within time period i, U PTR,ij It is the error in measuring the concentration of species j by PTR-ToF-MS during time period i.

[0077] A VOCs flux error matrix is ​​synthesized. Based on the calculated VOCs flux error data, for flux data points with concentrations less than 0 or null values, the corresponding error data is set to three times the median flux of the corresponding species. The flux error data of all VOCs species calculated above are synthesized into a VOCs flux error matrix. If there are i VOCs species, and each species has j data points, then a matrix with i columns and j rows is generated. The flux matrix and the flux error matrix should have the same number of rows and columns, and the data should correspond one-to-one.

[0078] S6. Input the VOCs flux matrix, the VOCs flux error matrix, the VOCs species name data, the mass-to-charge ratio data and the time axis data into the PMF model to obtain the number of factors, the source component spectrum of the factors and the flux time series of the factors.

[0079] S7. Determine the pollution source represented by the factor based on the source component spectrum of the factor.

[0080] S8. Determine the specific emission sources of the factor based on the flux time series of the factor.

[0081] The VOCs flux matrix, VOCs flux error matrix, VOCs species name data, mass-to-charge ratio data, and time axis data are input into the PMF Evaluation Tool 3.05 package based on Igor Pro software for source analysis. The specific steps are as follows:

[0082] The first step is to import the VOCs flux matrix, VOCs flux error matrix, VOCs species name data, mass-to-charge ratio data, and time axis data of mass spectrometry measurements into the PMF software. First, the flux matrix needs to be pruned to remove outliers with more than two consecutive points. Second, the error matrix needs to be pruned to remove outliers with more than two consecutive points. Finally, null values ​​and rows containing 0 are removed from the matrix to ensure that the input data meets the model requirements.

[0083] The second step is to calculate the flux signal-to-noise ratio (S / N) for each species to assess data quality, as shown in the following formula:

[0084]

[0085] In the formula, The flux signal-to-noise ratio, C, represents the species. f [X] represents the species response factor, [t] represents the measurement period, and B represents the background signal. The S / N results are divided into three categories: less than 0.2, between 0.2 and 2, and greater than 2. The throughput data in the two categories of less than 0.2 and between 0.2 and 2 are subjected to deduplication to obtain throughput data with further quality assurance.

[0086] The third step is to set the parameters of the PMF model. Generally, the number of factors is set between 1 and 10 to find the optimal solution. The Fpeak value (the parameter that adjusts the degree of freedom of factor rotation) in the PMF model is set to a range of -3 to 3, with an interval of 0.2. After selecting the storage path and performing source resolution, the obtained PMF model results include the following:

[0087] 1. The number of factors (which can be understood as the number of pollution sources analyzed) ranges from 1 to 10, corresponding to the values ​​of Q (residual sum of squares) / Qexp (expected residual sum of squares).

[0088] 2. Under specified factor conditions (e.g., 1 to 10), the flux time series of all factors (referring to a series of VOCs flux data points arranged in chronological order, reflecting the changes in VOCs flux at different time points).

[0089] 3. Given a specified number of factors (e.g., 1 to 10), the source component spectra for all factors, where the component spectra show the compositional distribution of different VOCs components in each factor.

[0090] The number and names of VOC flux sources are determined based on the above PMF model results. This mainly includes determining the number of factors based on Q / Qexp, comparing factor component spectra, comparing tracer species with diurnal variations of factors, and comparing factor component spectra with measured source spectra. The specific operational steps are as follows:

[0091] The first step is to initially determine the number of factors using Q / Qexp. The Q (sum of squared residuals) value is a key parameter used in this embodiment to verify and determine the optimal number of factors and the validation error matrix. It is an important basis for judging whether the PMF error matrix setting is reasonable and for factor selection. The Q value is closely related to the residuals and errors; the larger the data error, the lower its weight in the calculation. The calculation formula is as follows:

[0092]

[0093] In the formula, e ij u is the residual of the j-th species and the i-th sample; ij Let Q be the error of the j-th species and the i-th sample. Q / Qexp is the ratio of the Q of all data points to its expected error Qexp, used to verify the rationality of the error matrix setting. Simultaneously, the appropriate number of factors is selected based on the decrease in Q / Qexp across different factors. If Q / Qexp ~ 1, it indicates that all data points in the matrix are within their expected error range. If Q / Qexp >> 1, it means the error of the input data is severely overestimated. Conversely, if Q / Qexp << 1, it means the error of the input data is severely underestimated. Generally, as the number of factors increases, Q / Qexp continuously decreases. If the ratio is close to 1 when the number of factors is high, it indicates that the flux error matrix setting is reasonable. When the number of factors increases to a certain value, the decrease in Q / Qexp significantly decreases, indicating that when the number of factors is greater than or equal to this value, the VOCs flux data will be over-analyzed; this serves as a basis for judging whether the number of factors is reasonable. When the number of factors increases by 1 (at which point Q / Qexp is close to 1), but the decrease in Q / Qexp is less than 20%, this number of factors is considered the optimal number of factors.

[0094] The second step involves comparing and analyzing the compositional profiles of all factors based on the optimal number of factors (pollution sources) determined in the first step, identifying the component characteristics of different factors, and making a preliminary determination of the pollution sources that different factors may represent.

[0095] The third step involves determining the specific emission source represented by a factor by comparing the diurnal variations of tracer species (compounds that characterize a particular emission source, such as nitrogen oxides indicating vehicle emissions in cities) with those of the factor, given the component characteristics of different factors. The diurnal variations of tracers or factors are calculated based on tracer data or factor data and corresponding time data, calculating the number of hours per day for each time period, and averaging the same number of hours to obtain the average flux for different hours.

[0096] Fourth, if the emission source represented by a certain factor cannot be identified in the second and third steps, the factor component spectrum is compared and analyzed with the VOCs source component spectrum obtained from the pollution source emission experiment. The similarity between the two is calculated (as shown in the following formula). If the angle between the two is less than 30°, it can be determined that the factor component spectrum is consistent with the VOCs source component spectrum obtained from the pollution source emission experiment, and the pollution source represented by the factor can be finally identified.

[0097]

[0098] In the formula, MSA and MSB are the component proportions of the two component spectra used for comparison, and θ is the similarity angle between the two.

[0099] Finally, data analysis was conducted based on the VOCs flux source apportionment results obtained above under the optimal number of factors. These results include the VOCs flux time series (a series of VOCs flux data points arranged in chronological order, reflecting the changes in VOCs flux at different time points) for all factors under the optimal number of factors, as well as the source component spectra for all factors. The analysis of PMF results included analytical steps such as the proportion of factor temporal variation, the proportion of factor contribution, the proportion of factor diurnal variation, and the proportion of factor to individual species.

[0100] The specific operational steps are as follows: First, based on the flux time series of all factors, divide the VOCs flux time series of each factor by the total VOCs flux time series of all factors to obtain the VOCs flux share time series for each factor. This step allows us to understand the changes in the proportion of VOCs sources in different time periods and infer the possible reasons for these changes.

[0101] The second step involves summing all data points from the VOCs flux time series for each factor, based on the flux time series of all factors, to obtain the total flux for each factor. Then, the data points from the VOCs flux time series of all factors are summed to obtain the total flux for all factors. Dividing the total flux for each factor by the total flux for all factors yields the contribution percentage of different factors, thus revealing the relative importance of different VOCs flux sources.

[0102] The third step involves calculating the number of hours per day for each time period based on the factor time variation percentages and corresponding time data obtained in the first step. Averaging of the same number of hours yields the average flux for different time periods. By analyzing the diurnal variation characteristics of flux, the differences between day and night can be compared, and its correlation with environmental factors such as temperature and humidity can be analyzed.

[0103] The fourth step involves combining the factor contribution percentages obtained in the second step with the source component spectra corresponding to all factors to calculate the relative contribution of each factor to a single species in the component spectra. By analyzing the contribution percentages of different factors to a single species, more information can be provided for source tracing studies of typical VOCs species emissions.

[0104] This embodiment uses VOCs flux data as a basis and applies the orthogonal matrix factorization (PMF) model for source apportionment. This includes, but is not limited to, methods for screening VOCs flux data, calculating VOCs flux uncertainty, and calculating the VOCs flux matrix and VOCs flux error matrix. It determines the optimal PMF model output and conducts in-depth analysis of the results. Subsequent methods and parameter calculations are added as needed based on relevant research or experiments. This embodiment proposes a VOCs source apportionment method based on VOCs flux data, using the PMF model to identify the main VOCs emission sources in a region. Because this method can directly apportion sources from the emission of substances, it solves the problem of secondary generation and meteorological factors affecting the analysis in VOCs concentration-based source apportionment methods. Therefore, it can directly identify the primary sources of VOCs and quantify the specific contributions of different emission sources to various VOCs species. Compared with traditional VOCs source apportionment methods based on concentration measurement, it effectively reduces the uncertainty of VOCs source apportionment methods, better analyzes and quantifies the source characteristics of VOCs, and fills the technical gap in existing flux data-based source apportionment. In the future, it may be promoted and applied in the source apportionment of VOCs in different regions, bringing new insights to the prevention and control of air pollutants.

[0105] Based on the same inventive concept, this application also provides an apparatus for implementing the above-described method for VOCs source analysis based on flux data. The solution provided by this apparatus is similar to the solution described in the above-described method.

[0106] In one exemplary embodiment, a VOCs source apportionment device based on flux data is provided, including: a VOCs flux quality control and inspection module, a VOCs flux species determination module, a PMF flux matrix and error matrix generation module, a PMF source apportionment module, a PMF factor number determination and factor identification module, and a PMF result analysis module.

[0107] The VOCs flux quality control and inspection module is designed to ensure the quality of VOCs flux data. It uses a data quality labeling method to screen the data and is applied to the flux data quality control and inspection step of the VOCs flux species identification module.

[0108] The VOCs flux species identification module uses the data screening method designed in this application to obtain flux species data that can ultimately be applied to PMF analysis. The screening mechanism includes: high-resolution fitting of mass spectrometry peaks, concentration quantification, screening based on concentration detection limits, flux calculation, flux data quality control detection, screening based on flux detection limits, and positive flux evaluation.

[0109] The PMF flux matrix and error matrix generation module includes the generation of VOCs flux matrix and VOCs flux error matrix.

[0110] The PMF source resolution module takes the flux matrix, flux error matrix, species name, mass-to-charge ratio, and time axis as inputs to the PMF EvaluationTool 3.05 package based on IgorPro software for source resolution.

[0111] The PMF factor number determination and factor identification module refers to determining the number and name of VOC flux sources based on the above PMF model results. It mainly includes determining the number of factors based on Q / Qexp, comparing factor component spectra, comparing tracer species with diurnal variations of factors, and comparing factor component spectra with measured source spectra.

[0112] The PMF results analysis module refers to the data analysis performed based on the VOCs flux source apportionment results obtained from the PMF source apportionment module and the PMF factor number determination and factor identification module with the optimal number of factors. These results include the VOCs flux time series (a series of VOCs flux data points arranged in chronological order, reflecting the changes in VOCs flux at different time points) for all factors under the optimal number of factors, as well as the source component spectra corresponding to all factors. The analysis of the PMF results includes steps such as the percentage of factor temporal variation, the percentage of factor contribution, the percentage of factor diurnal variation, and the percentage of factors per species.

[0113] In another exemplary embodiment, such as Figure 2 As shown, a VOCs source apportionment method based on flux data is provided, which can identify the main VOCs emission sources in different regions, filling the gap in VOCs flux data-based source apportionment technology. The source apportionment method includes: flux data quality control verification, flux species determination, PMF flux error matrix determination, PMF source apportionment, PMF factor number determination and factor identification, and PMF result analysis.

[0114] The VOCs flux quality control module is designed to ensure the quality of VOCs flux data. It employs data quality labeling methods to filter the data and is used in the flux data quality control step of the VOCs flux species identification module. The quality labeling methods for flux data include steady-state testing and complete turbulence characteristic testing. The specific operation steps are as follows:

[0115] The first step involves calculating steady-state condition parameters and turbulence development characteristic parameters by combining the measured VOCs concentration data and environmental data (wind speed data measured by a three-dimensional ultrasonic anemometer). The parameter results are obtained and the data label values ​​of the parameters are obtained based on the data label classification range (Table 1). The calculation methods for steady-state condition parameters and turbulence development condition parameters are shown in Equations 1-2.

[0116] The second step is to select the highest number of data labels from the steady-state test and the complete turbulence characteristic test as the total label (e.g., the steady-state test is 3 and the turbulence development characteristic test is 5, then the total label is 5).

[0117] The third step is to use the data labeled 1-6 for data analysis, while the data labeled 7-10 is removed.

[0118] The third step is to label the data as 1-6 for data analysis, while discarding the data labeled as 7-10.

[0119]

[0120] In the formula, x s Represents the horizontal wind component or scalar, RN x Steady-state condition parameters for representative species over an average interval (30 min). The formulas for calculating turbulent development condition parameters are as follows:

[0121]

[0122] In the formula, ITC σ These are characteristic parameters for turbulence development. and These represent the wind steady-state parameters obtained from the theoretical model and the measured wind steady-state parameters under neutral conditions, respectively.

[0123] The VOCs flux species identification module uses the data screening method designed in this invention to obtain flux species data that can ultimately be applied to PMF analysis. The screening mechanism includes: high-resolution fitting of mass spectrometry peaks, concentration quantification, screening based on concentration detection limits, flux calculation, flux data quality control detection, screening based on flux detection limits, and positive flux evaluation. The specific operation steps are as follows:

[0124] The first step involved using the raw data from PTR-ToF-MS (proton reaction time-of-flight mass spectrometry) measurements to perform high-resolution fitting on the measured mass spectral peaks using mass spectrometry instrument data analysis software. This yielded the species names of 1778 VOCs measured during the mass spectrometry measurements, along with their corresponding mass-to-charge ratios and species signal data.

[0125] The second step involves converting the species signal into concentration based on the VOCs species signal data obtained in the previous step, combined with the species response factors obtained through calibration experiments using chemical standards. The calculation formula is shown below.

[0126]

[0127] In the formula, [X] represents species concentration, Signal represents species signal, and Sensitivity represents species response factor. The detection limit for species concentration is then calculated using the formula shown below.

[0128]

[0129] In the formula, DL is the detection limit of the VOC species concentration; Cf and B are the average signal and background signal of the species during mass spectrometry measurement; t is the temporal resolution of the VOC flux data during mass spectrometry measurement (i.e., the measurement interval required to obtain one flux data, generally 30 min); SNR is the signal-to-noise ratio of VOC measurement, and its value is set to 3 when calculating the detection limit.

[0130] The third step involves calculating the detection limit of concentration to screen out VOC species that are above the detection limit. VOC species with average concentrations below the detection limit will not be included in subsequent analyses. A total of 766 species with concentrations above the detection limit were obtained.

[0131] The fourth step involves calculating the VOCs flux (Formula 5) based on the VOCs concentration data filtered in the third step, combined with the wind speed data measured by a three-dimensional ultrasonic anemometer.

[0132] The fifth step involves using the VOCs flux quality control and inspection module described above to obtain data markers for each VOCs species, and then removing flux data points that do not meet the flux measurement conditions (data marker values ​​greater than 6).

[0133] Step 6: Calculate the species flux detection limit (Formula 6) based on VOCs concentration data and three-dimensional wind speed data. Remove species whose average VOCs flux is lower than their flux detection limit. The remaining 199 VOCs species that meet the flux detection limit will proceed to the next step of analysis.

[0134] Step 7: Based on the VOCs flux data obtained in the above screening steps, calculate the positive proportion of effective flux data (non-null data points after data quality control) for each VOCs species. Select 193 species with a positive proportion greater than or equal to 70% as the final species that can be used for input into the PMF model. The specific process is as follows: Figure 3 As shown.

[0135]

[0136] In the formula, w' and c' represent the change in vertical wind speed and the change in concentration of a certain VOCs, respectively.

[0137] LOD flux =3*RE noise( 6)

[0138] In the formula, LOD flux and RE noise These represent the flux detection limit and the random error in VOCs flux measurement, respectively.

[0139] The PMF flux matrix and error matrix generation module includes the generation of the VOCs flux matrix and the VOCs flux error matrix. The specific operation steps are as follows:

[0140] The first step involves synthesizing a VOCs flux matrix from the flux data of all VOCs species selected by the VOCs flux species identification module that are suitable for PMF analysis. If there are i VOCs species, and each species has j data points, then a matrix with i columns and j rows is generated. In the matrix, for flux data points that are less than 0 or are null values, the median flux of that species is used instead.

[0141] The second step involves calculating the error UPTR,ij of PTR-ToF-MS measurement of VOC concentrations, the systematic error of flux measurement, and the random error REnoise of flux measurement for all VOC species involved in the first step.

[0142] The method for calculating the error in VOCs concentration is shown in the following formula.

[0143]

[0144] In the formula, UPTR,ij is the error in measuring the concentration of species j by PTR-ToF-MS during time period i. SNRij is the signal-to-noise ratio of species j during time period i, used to evaluate the counting statistical error in the instrument measurement process, and is calculated by the following formula. xij is the average concentration of j during time period i (30 min). pij is the error rate for PTR-ToF-MS measurement, set to 20%. Of this, 10% of the error comes from fluctuations in instrument operation, such as fluctuations in ion source temperature and parent ion signal. The other 10% comes from the error caused by calibration using standard gases during long-term observation. DL is the instrument detection limit for species j, calculated as follows.

[0145]

[0146] In the formula, SNRij is the signal-to-noise ratio of species j in the i-th time period; Cf,j and B are the average and background signals of species j during the mass spectrometry measurement; t is the time interval (30 min) used in the VOC flux calculation in PMF. Since the systematic error of flux measurement can be estimated from the flux averaging interval, it is considered to be negligible.

[0147] The third step is to calculate the VOCs flux error. The VOCs flux error is obtained by adding the error in VOCs concentration and the random error in flux measurement. However, not all of the VOCs concentration error matrix is ​​propagated to the VOCs flux measurement. It is estimated that approximately half of the VOCs concentration error is propagated to the VOCs flux measurement error. Therefore, the VOCs flux error is calculated as follows.

[0148]

[0149] In the formula, U total,ij The error in measuring the flux of species j in PTR-ToF-MS during time period i, RE noise,ij The random error in measuring the flux of species j within time period i, U PTR,ij It is the error in measuring the concentration of species j by PTR-ToF-MS during time period i.

[0150] The fourth step is to synthesize the VOCs flux error matrix. Based on the VOCs flux error data calculated in the third step, for flux data points with concentrations less than 0 or null values, the corresponding error data is set to three times the median flux of the corresponding species. The flux error data of all VOCs species calculated above are then synthesized into a VOCs flux error matrix. If there are i VOCs species, and each species has j data points, then a matrix with i columns and j rows is generated. The flux matrix and the flux error matrix should have the same number of rows and columns.

[0151] The PMF source resolution module performs source resolution by inputting the flux matrix, flux error matrix, species name, mass-to-charge ratio, and time axis into the PMF Evaluation Tool 3.05 package based on Igor Pro software. The specific operation steps are as follows:

[0152] The first step involves importing the flux matrix and flux error matrix generated by the PMF flux matrix and error matrix generation module, along with the VOCs species names, mass-to-charge ratios, and mass spectrometry time axes obtained from the high-resolution fitting step of the VOCs flux species determination module, into the PMF software. First, the flux matrix needs to be processed to remove outliers exceeding two consecutive points. Second, the error matrix, after removing outliers, needs to be processed to remove outliers exceeding two consecutive points. Finally, null values ​​and rows containing zeros are removed from the matrix to ensure that the input data meets the model requirements.

[0153] The second step is to calculate the flux signal-to-noise ratio (S / N) for each species to assess data quality, as shown in the following formula:

[0154]

[0155] In the formula, The flux signal-to-noise ratio, C, represents the species. f [X] represents the species response factor, [t] represents the measurement period, and B represents the background signal. The S / N results are divided into three categories: less than 0.2, between 0.2 and 2, and greater than 2. The throughput data in the two categories of less than 0.2 and between 0.2 and 2 are subjected to deduplication to obtain throughput data with further quality assurance.

[0156] The third step is to set the parameters of the PMF model. Generally, the number of factors is set between 1 and 10 to find the optimal solution. The Fpeak value (the parameter that adjusts the degree of freedom of factor rotation) in the PMF model is set to a range of -3 to 3, with an interval of 0.2. After selecting the storage path and performing source resolution, the obtained PMF model results include the following:

[0157] 1. The number of factors (which can be understood as the number of pollution sources analyzed) ranges from 1 to 10, corresponding to the values ​​of Q (residual sum of squares) / Qexp (expected residual sum of squares).

[0158] 2. Under specified factor conditions (e.g., 1 to 10), the flux time series of all factors (referring to a series of VOCs flux data points arranged in chronological order, reflecting the changes in VOCs flux at different time points).

[0159] 3. Given a specified number of factors (e.g., 1 to 10), the source component spectra for all factors, where the component spectra show the compositional distribution of different VOCs components in each factor.

[0160] Based on the analysis of the PMF model results, the number of factors was determined according to Q / Qexp, comparison of factor component spectra, comparison of tracer species with diurnal variations of factors, and comparison of factor component spectra with measured source spectra. Ultimately, the number of factors was determined to be five: Source1, Source2, Source3, Source4, and Source5. The Q-value calculation formula is as follows:

[0161]

[0162] Taking the example dataset as an example, the PMF results analysis is not limited to the content described above. Based on time series plots with different factor proportions (such as...) Figure 4 As shown in the figure, Source 1 plays a dominant role and its contribution is stable, especially during the period from May 29th to June 7th, indicating that VOCs emissions were significantly influenced by Source 1 during this period. Source 2 and Source 4 showed significant fluctuations throughout the mass spectrometry measurement period, suggesting considerable uncertainty regarding their related emissions. The factor percentage plot shows that Source 1 (33%) is the main contributor, while the emissions from Source 2 (22%) are also not negligible. Figure 5 The figure shows the average contribution of factors to a typical species, where, Figure 5 (a) is isoprene. Figure 5 (b) is toluene. Figure 5 (c) is a C8 aromatic hydrocarbon. Figure 5 (d) is acetone. Figure 5 (e) represents ethanol. Typical aromatic hydrocarbon sources differ; toluene mainly originates from Source 5 (30%), while C8 aromatics mainly originate from Source 4 (38%). The contributions of the five factors to acetone are relatively balanced, while ethanol mainly originates from emissions from Source 1 (38%) and Source 4 (37%). Analysis based on flux-driven PMF results can provide new insights into local VOCs sources and is of great significance for subsequent VOCs control efforts.

[0163] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a VOCs source analysis method based on flux data.

[0164] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0165] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0166] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0167] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0168] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0169] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0170] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0171] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0172] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for VOCs source analysis based on flux data, characterized in that, The VOCs source analysis method based on flux data includes: Acquire wind speed data, mass spectrometry peak data of the gas to be measured, and time axis data during mass spectrometry peak measurement; The VOC species data in the gas to be tested are determined based on the mass spectrometry peak data; the VOC species data includes: VOC species name data, corresponding species concentration data, and corresponding mass-to-charge ratio data; The VOC flux data for each species is calculated based on the wind speed data, the VOC species name data, and the corresponding species concentration data. For each species: The VOCs flux error matrix is ​​calculated based on the VOCs flux data; the VOCs flux error matrix includes: a measurement error matrix, a systematic error matrix, and a random error matrix; The VOCs flux data is converted into a VOCs flux matrix; The VOCs flux matrix, the VOCs flux error matrix, the VOCs species name data, the mass-to-charge ratio data, and the time axis data are input into an orthogonal matrix factorization model to obtain the number of factors, the source component spectrum of the factors, and the flux time series of the factors. The pollution source represented by the factor is determined based on the source component spectrum of the factor; The specific emission sources of the factor are determined based on the flux time series of the factor.

2. The VOCs source analysis method based on flux data according to claim 1, characterized in that, The VOC species data in the gas to be tested are determined based on the mass spectrometry peak data. The VOCs species data includes: VOCs species name data, corresponding species concentration data, and corresponding mass-to-charge ratio data, specifically including: The mass spectrometry peak data were fitted at high resolution using mass spectrometry instrument data analysis software to obtain VOCs species name data, corresponding mass-to-charge ratio data and corresponding species signal data. Species concentration data are calculated based on the species signal data; the formula for calculating the species concentration data is as follows: In the formula, [X] represents species concentration, Signal represents species signal, and Sensitivity represents species response factor.

3. The VOCs source analysis method based on flux data according to claim 1, characterized in that, The VOC flux data for each species is calculated based on the wind speed data, the VOC species name data, and the corresponding species concentration data, specifically including: Delete species data whose concentrations are below the detection limit; the formula for calculating the detection limit is: In the formula, DL is the detection limit for the species concentration; C f B and B are the average signal and background signal of the species during mass spectrometry measurement; t is the temporal resolution of the VOCs flux data during mass spectrometry measurement; and SNR is the signal-to-noise ratio of VOCs measurement. The VOC flux data for each species is calculated based on the wind speed data, the VOC species name data, and the corresponding species concentration data.

4. The VOCs source analysis method based on flux data according to claim 1, characterized in that, The VOCs flux error matrix is ​​calculated based on the VOCs flux data, specifically including: VOCs flux data below the flux detection limit are deleted; wherein, the flux detection limit is calculated using the species concentration data and the wind speed data; Calculate the VOCs flux error matrix based on the VOCs flux data.

5. The VOCs source analysis method based on flux data according to claim 1, characterized in that, The specific emission sources of the factor are determined based on the flux time series of the factor, including: Calculate the daily variation average flux of the factor based on the flux time series of the factor; The emission sources are obtained by matching the average daily variation flux with the average daily variation flux of tracer species in the city.

6. The VOCs source analysis method based on flux data according to claim 1, characterized in that, The VOC flux data for each species is calculated based on the wind speed data, the VOC species name data, and the corresponding species concentration data, specifically including: Based on the species concentration data and the wind speed data, the VOCs species data are labeled with data quality using steady-state testing and complete turbulence characteristic testing methods to obtain data quality labeling results. Delete the VOC species data corresponding to the quality labeling results that are higher than the preset threshold; The VOC flux data for each species is calculated based on the wind speed data, the VOC species name data, and the corresponding species concentration data.

7. The VOCs source analysis method based on flux data according to claim 1, characterized in that, The formula for calculating the VOCs flux data is as follows: Here, w' and c' represent changes in wind speed and VOCs concentration, respectively.

8. A computer device, comprising: The memory and processor contain a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the VOCs source analysis method based on flux data as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the VOCs source analysis method based on flux data as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the VOCs source analysis method based on flux data as described in any one of claims 1-7.

Images