VOCs pollution monitoring system and method for industrial park

Abstract

The invention discloses an industrial park VOCs pollution monitoring system and method, and relates to the technical field of industrial park VOCs pollution monitoring, the system comprises a multi-source data acquisition and fusion module, a non-point source-point source conversion module, a micro-scale CALPUFF modeling module, a pollution traceability inversion module and a traceability result verification module; by collecting cleaning data, a non-point source is discretized into a virtual point source, a CALPUFF model is constructed to simulate diffusion, and a pollution source is inverted and verified. According to the method, high-temporal-spatial-resolution pollution monitoring with 500m * 500m grid precision is realized, the traceability precision is improved, the problem of distortion of traditional non-point source processing is solved, the method adapts to complex terrains and dynamic meteorological conditions, and a precise decision basis is provided for park pollution management and control.

Description

Technical Field

[0001] The present invention relates to the technical field of volatile organic compound pollution monitoring in industrial parks, and in particular to a VOCs pollution monitoring system and method for industrial parks. Background Art

[0002] Currently, VOCs pollution monitoring in industrial parks faces the following technical bottlenecks: Insufficient accuracy in tracing non-point source pollution: Traditional methods make it difficult to effectively fit non-point source pollution in industrial parks into point sources. There is a lack of virtual point source layout algorithms that take into account factors such as terrain and wind direction, resulting in the inability to accurately invert pollution sources within a small-scale space of 500m×500m.

[0003] Weak small-scale diffusion simulation capabilities: The existing CALPUFF model lacks double-nested grid technology and terrain correction mechanism when integrating the WRF meteorological field, making it difficult to simulate the dry / wet deposition and chemical transformation processes of VOCs under complex terrain. The time resolution is generally less than 1 hour, which cannot meet real-time monitoring needs.

[0004] Inefficient multi-source data fusion: There are spatiotemporal deviations between online monitoring, underway observations, and meteorological data. There is a lack of standardized processes for dynamic cleaning, spatiotemporal calibration, and feature extraction, resulting in insufficient reliability of model input data.

[0005] The traceability results lack closed-loop verification: Traditional technologies have not established a characteristic pollutant spectrum matching analysis and parameter iterative correction mechanism, and are unable to verify the traceability accuracy at a small scale, which is prone to misjudgment.

[0006] In addition, the existing technology has the following defects: Regarding non-point source pollution treatment: the combined effects of the sub-region's geometric center, dominant wind direction, and terrain undulation were not considered, and the virtual point source location setting lacked a scientific basis; In terms of model simulation: high-precision coupling between the WRF meteorological field and the CALPUFF model was not achieved, and the calibration of diffusion parameters relied on manual experience, making it difficult to adapt to the dynamically changing atmospheric environment; In terms of data application: the lack of spatiotemporal alignment technology for multi-source heterogeneous data leads to inaccurate extraction of pollution plume features, affecting the accuracy of source tracing and inversion.

[0007] To solve the above problems, the present invention proposes a VOCs pollution monitoring system and method for an industrial park. Summary of the Invention

[0008] The purpose of the present invention is to provide a VOCs pollution monitoring system and method for an industrial park to solve the problems raised in the prior art.

[0009] To achieve the above object, the present invention provides the following technical solutions: A VOCs pollution monitoring system for industrial parks, comprising a multi-source data acquisition and fusion module, a surface source-point source conversion module, a micro-scale CALPUFF modeling module, a pollution source tracing inversion module, and a tracing result verification module; the multi-source data acquisition and fusion module is responsible for real-time collection of VOCs online monitoring, meteorological, pollution source emissions, and cruise observation data in the industrial park, and simultaneously completes dynamic cleaning, spatiotemporal calibration, and feature extraction to form a standardized data set for subsequent modeling and analysis; the surface source-point source conversion module uses a virtual point source processing algorithm to discretize the surface source pollution area of ​​the industrial park into regularly distributed virtual point sources, establishes an equivalent relationship between the pollutant emissions of surface sources and point sources, and provides input parameters for CALPUFF modeling; the micro-scale The CALPUFF modeling module constructs a CALPUFF model based on a 500m×500m grid accuracy, integrates virtual point source parameters and the three-dimensional meteorological field generated by WRF, simulates the diffusion and transmission process of VOCs in a small-scale space, and outputs hourly concentration distribution data; the pollution source tracing and inversion module combines the measured data of the monitoring points with the model concentration field, matches the pollution transmission path through inverse operations, calculates the pollution contribution rate of each virtual point source to the receptor point, and locates the spatial distribution of the pollution source; the tracing result verification module verifies the accuracy of the virtual point source tracing through the matching degree analysis of the characteristic pollutant spectra of the cruise monitoring data and the model inversion results, and iteratively corrects the model parameters to improve the tracing accuracy at the 500m×500m scale.

[0010] The multi-source data acquisition and fusion module includes a real-time data acquisition unit, a WRF data integration unit, a dynamic data cleaning unit, a time-space calibration unit and a feature extraction unit; The real-time data acquisition unit collects VOCs concentration, pollution source emission parameters, three-dimensional meteorological field data and cruise monitoring trajectory data in real time through online monitoring equipment, cruise vehicles and weather stations deployed in the park, and transmits the acquired multi-dimensional raw data to the dynamic data cleaning unit; The WRF data integration unit obtains the three-dimensional meteorological field data output by the mesoscale meteorological model WRF through an API interface, including wind field, temperature and humidity field, and pressure field. The temporal resolution of the data is not less than 1 hour, and the spatial resolution is resampled by a bilinear interpolation algorithm until it matches the spatial resolution of the park's 500m×500m grid; The dynamic data cleaning unit uses an adaptive filtering algorithm to remove outliers and noise from the collected multi-source heterogeneous data, and at the same time fills in missing data through a data integrity verification mechanism to form a preliminary standardized data set; The time-space calibration unit unifies the timestamps and maps the spatial coordinates of data from different sources based on the park's 500m×500m grid coordinate system, eliminating clock bias and spatial positioning errors of monitoring equipment. The feature extraction unit uses wavelet transform to remove spectral noise, uses principal component analysis algorithm to extract meteorological factor coupling characteristics, extracts VOCs characteristic component spectra, pollution plume diffusion characteristics and meteorological factor coupling relationships from the cleaned and calibrated data, and generates a high-temporal and spatial resolution data set suitable for virtual point source modeling. The high-temporal and spatial resolution data set includes a VOCs characteristic spectrum matrix and a meteorological factor load matrix, providing data input for the subsequent surface source-point source conversion module.

[0011] The surface source-point source conversion module includes a surface source area division unit, a virtual point source generation unit, an emission parameter mapping unit and a model adaptation verification unit; The non-point source area division unit divides the non-point source pollution area of ​​the industrial park into a preset number of sub-areas with regular shapes and similar areas using the Delaunay triangulation algorithm based on the park geographic information, land use type and pollution source distribution data provided by the multi-source data acquisition and fusion module; The virtual point source generation unit determines the location of the virtual point source in each sub-area using a virtual point source placement algorithm based on pollutant diffusion characteristics. The algorithm comprehensively considers the geometric center of the sub-area, the dominant wind direction, and the terrain undulation to determine the coordinates of the virtual point source. The calculation formula is as follows: ; Among them, P i is the coordinate of the virtual point source in the i-th sub-area, C i is the geometric center coordinate of the sub-region, W i is the dominant wind direction vector in the area, T i is the terrain relief correction coefficient; the geometric center coordinates of the sub-area C i After dividing the non-point source pollution area of ​​the industrial park by the non-point source area division unit using the Delaunay triangulation algorithm, the geometric center is calculated based on the coordinates of the sub-area polygon vertices; the dominant wind direction vector W i The park meteorological data collected by the multi-source data acquisition and fusion module is statistically analyzed by calculating the wind direction and speed data within the preset time period, and the wind direction frequency and wind speed weighted average are calculated to determine the direction and intensity of the dominant wind direction in the area and represent them in vector form; the terrain undulation correction coefficient T i Based on the digital elevation model data of the park, the standard deviation and range of terrain elevation in the sub-area are calculated. The range of terrain elevation is the difference between the highest elevation and the lowest elevation in the sub-area. After normalizing the standard deviation and the range of terrain elevation, the coefficient used to correct the position of the virtual point source is obtained. The emission parameter mapping unit maps the emission parameters of each pollutant component of the non-point source to the corresponding virtual point source based on the emission data of each pollution source in the non-point source sub-area through the law of conservation of mass and the spatial weight distribution method, establishes an equivalent relationship between the emission parameters of the non-point source and the virtual point source, and enables the virtual point source to reflect the intensity of non-point source pollution; wherein the parameters of the non-point source include but are not limited to the pollutant emission rate and concentration; specifically: According to the law of conservation of mass, the total pollution emissions of the surface source sub-area are calculated by the addition method, so that the total emissions of the virtual point source are consistent with the surface source, and then the Gaussian distance attenuation function w is used. j = exp(-d j 2 / 2ε 2 ) calculate the weight, where d j is the Euclidean distance from the pollution source to the virtual point source, ε is the characteristic scale of the sub-region, which is 1 / 3 of the length of the sub-region diagonal. The closer the pollution source is to the virtual point source, the higher the weight. Then, the emission parameters of each pollutant component of the non-point source are weighted and summed according to the spatial weight to obtain the spatial weighted mapping of the concentration parameters, and generate the standardized emission parameter set of the virtual point source, which includes: Quantitative parameters: emission rate and emission concentration of each pollutant component; Spatial parameters: emission height and emission direction, which are determined by the average emission height and dominant wind direction vector of the surface source sub-area; Time parameters: emission period distribution, directly inheriting the measured emission period data of non-point source pollution sources; The parameter set is stored in HDF5 format, including the coordinate mapping table, component concentration matrix and time series table, and is output to the model adaptation verification unit; The model adaptation verification unit inputs the virtual point source parameters generated by the emission parameter mapping unit into the small microscale CALPUFF model and verifies the rationality of the parameters through the following process: The CALPUFF model is used to simulate the pollution diffusion process of the virtual point source and output the hourly concentration field. The concentration deviation E of the receptor point is calculated by comparing it with the cruise monitoring and fixed-point monitoring data synchronized with the simulation period in the multi-source data acquisition module. The formula is as follows: ; Among them, C obs is the measured value of the multi-source data acquisition module, C model is the pollutant concentration at the receptor site simulated by the CALPUFF model; If E exceeds the preset threshold, the virtual point source position and emission parameters are adaptively adjusted through the error back propagation algorithm; for position adjustment: based on the terrain elevation range and standard deviation, the virtual point source coordinates P are corrected. i = f(C i , Wi , T i ); For emission parameter adjustment: remap the emission rate and concentration of the surface source sub-area, and update the virtual point source parameter set; Iterative adjustments were made until the spatial resolution of 500m×500m was achieved, and the deviation between the model simulation results and the actual monitoring data was controlled within the preset threshold range, providing calibrated point source input data for the subsequent small-scale CALPUFF modeling module.

[0012] The microscale CALPUFF modeling module includes a grid division unit, a meteorological field processing unit, a virtual point source integration unit, a diffusion simulation unit and an output calibration unit; The grid division unit is based on the 500m×500m spatial resolution requirement and uses a double nested grid technology to divide the industrial park and the peripheral buffer area. The peripheral buffer area refers to the outer grid coverage area surrounding the core simulation area, providing boundary meteorological conditions for the inner high-resolution grid; the outer grid provides boundary conditions for the inner grid and controls the accuracy of small-scale simulation; the inner grid covers the core pollution area, and the grid spacing is limited to 500m×500m; The meteorological field processing unit receives the WRF model output results provided by the multi-source data acquisition and fusion module, pre-processes the three-dimensional meteorological field data through the CALMET program module, and generates a meteorological input field suitable for the CALPUFF model by combining the park terrain elevation data and land use type; The virtual point source integration unit performs spatiotemporal matching on the virtual point source parameters of the area source-point source conversion module calibrated by the model adaptation and verification unit with the WRF meteorological field data of the multi-source data acquisition and fusion module. The double nested grid technology is used to first convert the virtual point source coordinates from the park geographic coordinate system to the CALPUFF model grid coordinate system through coordinate transformation, and the emission direction parameters are converted to the wind direction offset angle of the CALPUFF source attribute. Then, linear interpolation is performed on the intermittent emission parameters and the WRF meteorological field to generate an hourly emission rate matrix. Subsequently, the virtual point source, the park fixed point source and the volume source are integrated to generate an HDF5 format mixed source emission inventory containing coordinates, emission parameters and source types. The inventory data structure is SourceInventory = (x, y, z, t, Q i , C i , Type, H s , θ), where H s is the emission height, θ is the emission direction, and Type identifies the source type. At the same time, the diffusion coefficient is pre-calculated based on the corresponding grid meteorological parameters extracted from the virtual point source coordinates. The diffusion coefficient ơ is pre-calculated according to the Pasquill-Gifford classification method based on the emission height Hs and the meteorological stability level from WRF. x ,ơ y,ơ z , ultimately providing spatiotemporally consistent source term input for the diffusion simulation of the microscale CALPUFF modeling module; the HDF5 format contains three levels of data sets, namely coordinate mapping, emission parameters, and meteorological correlation, which are compatible with the CALPUFF model input format; The diffusion simulation unit simulates the pollutant transport process within each 500m×500m grid hourly based on the Gaussian smoke flow model algorithm of the CALPUFF model and the Lagrangian particle tracking method. The Lagrangian particle tracking method is used to calculate the pollutant concentration distribution. The calculation formula is as follows: ; Where C(x,y,z,t) is the pollutant concentration at coordinate (x,y,z) at time t; Q i is the emission rate of the ith virtual point source; (x i ,y i ,z i ) is the virtual point source coordinate; x ,ơ y ,ơ z are the diffusion coefficients in the x, y, and z directions, respectively; The output calibration unit compares the simulated concentration field with the measured data of the multi-source data acquisition module, and uses the same gradient descent algorithm as the traceability result verification module to calibrate the diffusion coefficient ơ x ,ơ y ,ơ z Perform adaptive iterative adjustments; the calibrated concentration field needs to be re-input into the pollution source tracing inversion module, and trigger the source tracing result verification module to calculate the characteristic pollutant spectrum matching degree, until the deviation between the simulated concentration and the measured data at the 500m×500m scale is controlled within the preset threshold, and finally output hourly high temporal and spatial resolution data containing the concentration of each VOCs component to provide model input for the pollution source tracing inversion module, where the iterative calibration trigger condition is that the spectrum matching degree is less than the preset threshold.

[0013] The pollution source tracing inversion module includes a data fusion unit, an inverse operation inversion unit, a contribution rate calculation unit and a spatial positioning unit; The data fusion unit is responsible for performing spatiotemporal alignment processing on the hourly pollutant concentration field data output by the micro-scale CALPUFF modeling module and the measured data of the multi-source data acquisition and fusion module; specifically, the Kalman filter algorithm is used to eliminate the deviation of the two types of data and generate a fused data set C model (x,y,z,t) and C obs (x, y, z, t); where the model concentration field data C model(x, y, z, t) comes from the simulation results of the CALPUFF modeling module based on the virtual point source parameters and the WRF meteorological field, and the measured concentration data C obs (x, y, z, t) Fixed-point monitoring and cruise monitoring equipment from the multi-source data acquisition module, including VOCs concentration, cruise trajectory, and characteristic component spectrum information; The inverse operation inversion unit uses an improved regularized inversion algorithm to solve the inverse trajectory of pollution transmission based on the data set generated by the data fusion unit. Specifically, by minimizing the mean square error between the measured concentration and the model-predicted concentration and introducing a regularized constraint term, the optimal solution of the virtual point source emission rate vector q is obtained. The calculation formula is as follows: ; Among them, C obs is the measured value of the multi-source data acquisition module, C model (q) is the simulated value based on q by the CALPUFF modeling module; q is the virtual point source emission rate vector. The initial value of q is generated by the area source-point source conversion module by mapping the measured area source emission data using the law of conservation of mass, and is subsequently iteratively adjusted as an optimization variable. The regularization parameter λ is determined through cross-validation based on historical monitoring data, and the constraint matrix L presets a priori constraints on the spatial distribution of pollution sources. The contribution rate calculation unit calculates the pollution contribution rate of each virtual point source to the fixed monitoring point receptor point in the park using the source allocation receptor model based on the virtual point source emission parameters solved by the inverse operation inversion unit; the calculation formula is as follows: ; Among them, Q i is the emission rate of the i-th virtual point source obtained by inversion, C contrib,i is the contribution of the point source to the receptor concentration simulated by the CALPUFF model, f i is the contribution rate of the ith virtual point source, Q j is the pollutant emission rate of the jth virtual point source, which is obtained by the inverse operation inversion unit of the pollution source tracing inversion module through the regularized inversion algorithm. Its initial value comes from the emission parameter mapping unit of the surface source to point source conversion module, which is generated by the mass conservation law mapping based on the measured emission data of the surface source; C contrib,j The jth virtual point source is simulated by the CALPUFF model to simulate the pollutant concentration contribution to the receptor point, which represents the concentration contribution value of the pollutant emitted by the point source at the receptor point. It is generated by the diffusion simulation unit of the microscale CALPUFF modeling module based on the virtual point source coordinates, emission rate Q j and WRF meteorological field data, obtained by simulating the pollutant diffusion process through the Gaussian smoke flow model and Lagrangian particle tracking method; The spatial positioning unit spatially matches the coordinates of virtual point sources whose contribution rates exceed a threshold with a virtual point source standardized parameter set and the surface source area division result generated by the surface source-point source conversion module, and generates a pollution source heat map in combination with a GIS geographic information system. The virtual point source coordinates are determined by the virtual point source generation unit of the surface source-point source conversion module, and the virtual point source standardized parameter set includes the sub-region geometric center coordinates output by the surface source area division unit, the dominant wind direction vector, and the emission rate and concentration parameters generated by the emission parameter mapping unit. Through spatial matching and visualization processing, the pollution source is located at a 500m×500m grid scale.

[0014] The traceability result verification module includes a spectrum data acquisition unit, a characteristic spectrum extraction unit, a matching degree calculation unit, a parameter iterative correction unit and a verification report generation unit; The spectrum data acquisition unit establishes a spatiotemporally matched verification data set by synchronously acquiring the virtual point source characteristic spectrum data output by the contribution rate calculation unit in the pollution source tracing and inversion module and the cruise monitoring spectrum data of the real-time data acquisition unit of the multi-source data acquisition and fusion module; wherein the virtual point source characteristic spectrum is the VOCs component concentration distribution vector of each virtual point source, which is generated by diffusion simulation of the micro-scale CALPUFF modeling module; the cruise monitoring spectrum is acquired by a flight mass spectrometer with a time resolution of 1 hour, and the spatial positioning is matched with the 500m×500m grid of the park; The characteristic spectrum extraction unit performs wavelet transform denoising on the original spectrum data to eliminate instrument noise and environmental interference, completes baseline correction through cubic polynomial fitting, then matches characteristic peaks based on the NIST mass spectrum database, extracts pollutant concentration values, and generates normalized characteristic pollutant spectrum vectors to provide standardized data input for subsequent matching degree calculations; The matching degree calculation unit uses the weighted cosine similarity algorithm to calculate the matching degree index WeightefdSimilarity between the virtual point source spectrum and the cruise monitoring spectrum to quantify the accuracy of the traceability result. The calculation formula is as follows: ; Among them, s model,i is the concentration of the i-th characteristic pollutant in the virtual point source spectrum, which is simulated by the CALPUFF model; s obs,i is the concentration of the i-th characteristic pollutant in the cruise monitoring spectrum; w i is the weight coefficient of the i-th pollutant, which is set according to the degree of harm of the pollutant in the industry standard; n is the number of characteristic pollutant components; When the parameter iteration correction unit detects that the matching degree is lower than a preset threshold, the unit triggers the inverse operation inversion unit of the pollution source tracing inversion module to perform parameter iteration optimization; The verification report generation unit integrates the matching degree calculation results and parameter correction records to generate a visual verification report including a matching degree time series, parameter iteration trajectory, GIS comparison diagram of pollution hotspot areas before and after correction, and optimization suggestions; the matching degree time series reflects the traceability accuracy of each time period, the parameter trajectory records the iteration process of q and λ, and the GIS comparison diagram intuitively displays the differences in the location of pollution sources.

[0015] The parameter iteration correction unit further includes the following contents: When the matching index between the virtual point source spectrum and the cruise monitoring spectrum is lower than the preset threshold, the error function Loss = 1 - Weighted Similarity is used as the optimization target to convert the spectrum matching into an optimizable error index. The gradient descent algorithm is used to iteratively update the virtual point source emission rate vector q and the regularization parameter λ until the matching index meets the standard or the number of iterations reaches the upper limit. The calculation formula is as follows: ; Among them, q k is the virtual point source emission rate vector of the kth iteration, and its initial value is generated by the area source-point source conversion module through the mass conservation law mapping; k+1 is the emission rate vector of the k+1th iteration, which gradually approaches the optimal solution through iteration; α is the learning rate, which controls the iteration step size; Loss is the gradient of the error function with respect to the emission rate vector q, which represents the direction and magnitude of the impact of the change in q on the error; λ k is the regularization parameter of the kth iteration, and its initial value is determined by cross-validation of historical monitoring data from the multi-source data acquisition module; k+1 The regularization parameter for the k+1th iteration; β is the learning rate, and the parameter update step size is less than α; The gradient of the Loss error function with respect to the regularization parameter λ represents the effect of the change of λ on the error; The iterated q and λ are fed back to the inverse operation inversion unit of the pollution source tracing inversion module to recalculate the pollution contribution rate of the virtual point source to the receptor point until the matching degree between the tracing result and the measured data meets the 500m×500m scale requirement.

[0016] A method for monitoring VOCs pollution in an industrial park, comprising the following steps: S1. The system first collects the park's VOCs online monitoring data, meteorological data, pollution source emission parameters and navigation observation trajectory data in real time through the multi-source data acquisition module, and simultaneously performs dynamic cleaning, spatiotemporal calibration and feature extraction on the data to form a standardized data set; S2. Use the area source to point source conversion module to discretize the area source pollution area of ​​the industrial park into virtual point sources. The location of the virtual point source is determined by combining the geometric center coordinates of the sub-area, the dominant wind direction vector, and the terrain undulation correction coefficient. The emission equivalence relationship between area source and point source is established through the law of conservation of mass. S3. Build a microscale CALPUFF model based on a 500m×500m grid precision, integrating parameters such as the coordinates of the virtual point source, emission rate, and pollutant composition with the wind field, temperature and humidity field, and pressure field data generated by WRF to simulate the diffusion and transmission process of VOCs in a small-scale space and output hourly concentration distribution data. S4, the pollution source inversion module integrates the concentration field output by the CALPUFF model with the measured data of VOCs concentration monitored at fixed points and characteristic component spectra of cruise monitoring. It solves the reverse trajectory of pollution transmission through a regularized inversion algorithm, calculates the pollution contribution rate of each virtual point source to the receptor point, and locates the spatial distribution of pollution sources. S5. The traceability result verification module obtains the VOCs component concentration distribution vector of the virtual point source and the dynamic spectrum data of the cruise monitoring. After preprocessing, the weighted cosine similarity algorithm is used to calculate the matching degree. When the matching degree is lower than the threshold, the virtual point source emission rate and regularization parameter are iteratively corrected. S6. Finally, based on the calibrated traceability results and combined with the pollution source list, a GIS heat map of the pollution hotspot area, a matching time series curve, and a visual verification report of the parameter iteration trajectory are generated to provide a decision-making basis for pollution control in the park.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. Improved small-scale source tracing accuracy: The virtual point source processing algorithm is used to discretize area sources into regularly distributed virtual point sources. Combined with the 500m×500m grid precision simulation of the micro-scale CALPUFF model, high-precision pollution source tracing with a spatial resolution of 500m×500m and a temporal resolution of 1 hour is achieved, improving positioning accuracy by more than 30% compared to traditional methods.

[0018] 2. Innovation in area source-point source equivalent conversion: The area source-point source conversion module uses the Delaunay triangulation algorithm to divide the area source area, combines the dominant wind direction vector and the terrain undulation correction coefficient to determine the position of the virtual point source, maps the emission parameters through the law of conservation of mass, establishes an equivalent relationship between area source and point source, and solves the distortion problem of traditional area source processing.

[0019] 3. Enhanced adaptability to complex scenarios: The Lagrangian particle tracking method integrating the WRF three-dimensional meteorological field and the CALPUFF model takes into account the influence of dry / wet deposition, chemical conversion and terrain, and realizes the simulation of VOCs diffusion under complex terrain and dynamic meteorological conditions. Compared with traditional models, the simulation accuracy in complex scenarios is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a workflow diagram of an industrial park VOCs pollution monitoring system according to the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] Example: Figure 1 As shown, the present invention provides a technical solution. A VOCs pollution monitoring system for industrial parks, comprising a multi-source data acquisition and fusion module, a surface source-point source conversion module, a micro-scale CALPUFF modeling module, a pollution source tracing inversion module, and a tracing result verification module; the multi-source data acquisition and fusion module is responsible for real-time collection of VOCs online monitoring, meteorological, pollution source emissions, and cruise observation data in the industrial park, and simultaneously completes dynamic cleaning, spatiotemporal calibration, and feature extraction to form a standardized data set for subsequent modeling and analysis; the surface source-point source conversion module uses a virtual point source processing algorithm to discretize the surface source pollution area of ​​the industrial park into regularly distributed virtual point sources, establishes an equivalent relationship between the pollutant emissions of surface sources and point sources, and provides input parameters for CALPUFF modeling; the micro-scale The CALPUFF modeling module constructs a CALPUFF model based on a 500m×500m grid accuracy, integrates virtual point source parameters and the three-dimensional meteorological field generated by WRF, simulates the diffusion and transmission process of VOCs in a small-scale space, and outputs hourly concentration distribution data; the pollution source tracing and inversion module combines the measured data of the monitoring points with the model concentration field, matches the pollution transmission path through inverse operations, calculates the pollution contribution rate of each virtual point source to the receptor point, and locates the spatial distribution of the pollution source; the tracing result verification module verifies the accuracy of the virtual point source tracing through the matching degree analysis of the characteristic pollutant spectra of the cruise monitoring data and the model inversion results, and iteratively corrects the model parameters to improve the tracing accuracy at the 500m×500m scale.

[0023] The multi-source data acquisition and fusion module includes a real-time data acquisition unit, a WRF data integration unit, a dynamic data cleaning unit, a time-space calibration unit and a feature extraction unit; The real-time data acquisition unit collects VOCs concentration, pollution source emission parameters, three-dimensional meteorological field data and cruise monitoring trajectory data in real time through online monitoring equipment, cruise vehicles and weather stations deployed in the park, and transmits the acquired multi-dimensional raw data to the dynamic data cleaning unit; The WRF data integration unit obtains the three-dimensional meteorological field data output by the mesoscale meteorological model WRF through an API interface, including wind field, temperature and humidity field, and pressure field. The temporal resolution of the data is not less than 1 hour, and the spatial resolution is resampled by a bilinear interpolation algorithm until it matches the spatial resolution of the park's 500m×500m grid; The dynamic data cleaning unit uses an adaptive filtering algorithm to remove outliers and noise from the collected multi-source heterogeneous data, and at the same time fills in missing data through a data integrity verification mechanism to form a preliminary standardized data set; The time-space calibration unit unifies the timestamps and maps the spatial coordinates of data from different sources based on the park's 500m×500m grid coordinate system, eliminating clock bias and spatial positioning errors of monitoring equipment. The feature extraction unit uses wavelet transform to remove spectral noise, uses principal component analysis algorithm to extract meteorological factor coupling characteristics, extracts VOCs characteristic component spectra, pollution plume diffusion characteristics and meteorological factor coupling relationships from the cleaned and calibrated data, and generates a high-temporal and spatial resolution data set suitable for virtual point source modeling. The high-temporal and spatial resolution data set includes a VOCs characteristic spectrum matrix and a meteorological factor load matrix, providing data input for the subsequent surface source-point source conversion module.

[0024] The surface source-point source conversion module includes a surface source area division unit, a virtual point source generation unit, an emission parameter mapping unit and a model adaptation verification unit; The non-point source area division unit divides the non-point source pollution area of ​​the industrial park into a preset number of sub-areas with regular shapes and similar areas using the Delaunay triangulation algorithm based on the park geographic information, land use type and pollution source distribution data provided by the multi-source data acquisition and fusion module; The virtual point source generation unit determines the location of the virtual point source in each sub-area using a virtual point source placement algorithm based on pollutant diffusion characteristics. The algorithm comprehensively considers the geometric center of the sub-area, the dominant wind direction, and the terrain undulation to determine the coordinates of the virtual point source. The calculation formula is as follows: ; Among them, P i is the coordinate of the virtual point source in the i-th sub-area, C i is the geometric center coordinate of the sub-region, W i is the dominant wind direction vector in the area, T iis the terrain relief correction coefficient; the geometric center coordinates of the sub-area C i After dividing the non-point source pollution area of ​​the industrial park by the non-point source area division unit using the Delaunay triangulation algorithm, the geometric center is calculated based on the coordinates of the sub-area polygon vertices; the dominant wind direction vector W i The park meteorological data collected by the multi-source data acquisition and fusion module is statistically analyzed by calculating the wind direction and speed data within the preset time period, and the wind direction frequency and wind speed weighted average are calculated to determine the direction and intensity of the dominant wind direction in the area and represent them in vector form; the terrain undulation correction coefficient T i Based on the digital elevation model data of the park, the standard deviation and range of terrain elevation in the sub-area are calculated. The range of terrain elevation is the difference between the highest elevation and the lowest elevation in the sub-area. After normalizing the standard deviation and the range of terrain elevation, the coefficient used to correct the position of the virtual point source is obtained. The emission parameter mapping unit maps the emission parameters of each pollutant component of the non-point source to the corresponding virtual point source based on the emission data of each pollution source in the non-point source sub-area through the law of conservation of mass and the spatial weight distribution method, establishes an equivalent relationship between the emission parameters of the non-point source and the virtual point source, and enables the virtual point source to reflect the intensity of non-point source pollution; wherein the parameters of the non-point source include but are not limited to the pollutant emission rate and concentration; specifically: According to the law of conservation of mass, the total pollution emissions of the surface source sub-area are calculated by the addition method, so that the total emissions of the virtual point source are consistent with the surface source, and then the Gaussian distance attenuation function w is used. j = exp(-d j 2 / 2ε 2 ) calculate the weight, where d j is the Euclidean distance from the pollution source to the virtual point source, ε is the characteristic scale of the sub-region, which is 1 / 3 of the length of the sub-region diagonal. The closer the pollution source is to the virtual point source, the higher the weight. Then, the emission parameters of each pollutant component of the non-point source are weighted and summed according to the spatial weight to obtain the spatial weighted mapping of the concentration parameters, and generate the standardized emission parameter set of the virtual point source, which includes: Quantitative parameters: emission rate and emission concentration of each pollutant component; Spatial parameters: emission height and emission direction, which are determined by the average emission height and dominant wind direction vector of the surface source sub-area; Time parameters: emission period distribution, directly inheriting the measured emission period data of non-point source pollution sources; The parameter set is stored in HDF5 format, including the coordinate mapping table, component concentration matrix and time series table, and is output to the model adaptation verification unit; The model adaptation verification unit inputs the virtual point source parameters generated by the emission parameter mapping unit into the small microscale CALPUFF model and verifies the rationality of the parameters through the following process: The CALPUFF model is used to simulate the pollution diffusion process of the virtual point source and output the hourly concentration field. The concentration deviation E of the receptor point is calculated by comparing it with the cruise monitoring and fixed-point monitoring data synchronized with the simulation period in the multi-source data acquisition module. The formula is as follows: ; Among them, C obs is the measured value of the multi-source data acquisition module, C model is the pollutant concentration at the receptor site simulated by the CALPUFF model; If E exceeds the preset threshold, the virtual point source position and emission parameters are adaptively adjusted through the error back propagation algorithm; for position adjustment: based on the terrain elevation range and standard deviation, the virtual point source coordinates P are corrected. i = f(C i , W i , T i ); For emission parameter adjustment: remap the emission rate and concentration of the surface source sub-area, and update the virtual point source parameter set; Iterative adjustments were made until the spatial resolution of 500m×500m was achieved, and the deviation between the model simulation results and the actual monitoring data was controlled within the preset threshold range, providing calibrated point source input data for the subsequent small-scale CALPUFF modeling module.

[0025] The microscale CALPUFF modeling module includes a grid division unit, a meteorological field processing unit, a virtual point source integration unit, a diffusion simulation unit and an output calibration unit; The grid division unit is based on the 500m×500m spatial resolution requirement and uses a double nested grid technology to divide the industrial park and the peripheral buffer area. The peripheral buffer area refers to the outer grid coverage area surrounding the core simulation area, providing boundary meteorological conditions for the inner high-resolution grid; the outer grid provides boundary conditions for the inner grid and controls the accuracy of small-scale simulation; the inner grid covers the core pollution area, and the grid spacing is limited to 500m×500m; The meteorological field processing unit receives the WRF model output results provided by the multi-source data acquisition and fusion module, pre-processes the three-dimensional meteorological field data through the CALMET program module, and generates a meteorological input field suitable for the CALPUFF model by combining the park terrain elevation data and land use type; The virtual point source integration unit performs spatiotemporal matching on the virtual point source parameters of the area source-point source conversion module calibrated by the model adaptation and verification unit with the WRF meteorological field data of the multi-source data acquisition and fusion module. The double nested grid technology is used to first convert the virtual point source coordinates from the park geographic coordinate system to the CALPUFF model grid coordinate system through coordinate transformation, and the emission direction parameters are converted to the wind direction offset angle of the CALPUFF source attribute. Then, linear interpolation is performed on the intermittent emission parameters and the WRF meteorological field to generate an hourly emission rate matrix. Subsequently, the virtual point source, the park fixed point source and the volume source are integrated to generate an HDF5 format mixed source emission inventory containing coordinates, emission parameters and source types. The inventory data structure is SourceInventory = (x, y, z, t, Q i , C i , Type, H s , θ), where H s is the emission height, θ is the emission direction, and Type identifies the source type. At the same time, the diffusion coefficient is pre-calculated based on the corresponding grid meteorological parameters extracted from the virtual point source coordinates. The diffusion coefficient ơ is pre-calculated according to the Pasquill-Gifford classification method based on the emission height Hs and the meteorological stability level from WRF. x ,ơ y ,ơ z , ultimately providing spatiotemporally consistent source term input for the diffusion simulation of the microscale CALPUFF modeling module; the HDF5 format contains three levels of data sets, namely coordinate mapping, emission parameters, and meteorological correlation, which are compatible with the CALPUFF model input format; The diffusion simulation unit simulates the pollutant transport process within each 500m×500m grid hourly based on the Gaussian smoke flow model algorithm of the CALPUFF model and the Lagrangian particle tracking method. The Lagrangian particle tracking method is used to calculate the pollutant concentration distribution. The calculation formula is as follows: ; Where C(x,y,z,t) is the pollutant concentration at coordinate (x,y,z) at time t; Q i is the emission rate of the ith virtual point source; (x i ,y i ,z i ) is the virtual point source coordinate; x ,ơ y ,ơ z are the diffusion coefficients in the x, y, and z directions, respectively; In addition, the diffusion simulation unit of the present invention is developed based on the commercial CALPUFF model. Its built-in dry / wet deposition, chemical conversion and terrain influence modules can be automatically activated by inputting meteorological parameters, terrain data and VOCs reaction parameters generated by this system, without the need to modify the core algorithm of the model. Therefore, the document does not impose additional restrictions on the processing of environmental factors.

[0026] The output calibration unit compares the simulated concentration field with the measured data of the multi-source data acquisition module, and uses the same gradient descent algorithm as the traceability result verification module to calibrate the diffusion coefficient ơ x ,ơ y ,ơ z Perform adaptive iterative adjustments; the calibrated concentration field needs to be re-input into the pollution source tracing inversion module, and trigger the source tracing result verification module to calculate the characteristic pollutant spectrum matching degree, until the deviation between the simulated concentration and the measured data at the 500m×500m scale is controlled within the preset threshold, and finally output hourly high temporal and spatial resolution data containing the concentration of each VOCs component to provide model input for the pollution source tracing inversion module, where the iterative calibration trigger condition is that the spectrum matching degree is less than the preset threshold.

[0027] The pollution source tracing inversion module includes a data fusion unit, an inverse operation inversion unit, a contribution rate calculation unit and a spatial positioning unit; The data fusion unit is responsible for performing spatiotemporal alignment processing on the hourly pollutant concentration field data output by the micro-scale CALPUFF modeling module and the measured data of the multi-source data acquisition and fusion module; specifically, the Kalman filter algorithm is used to eliminate the deviation of the two types of data and generate a fused data set C model (x,y,z,t) and C obs (x, y, z, t); where the model concentration field data C model (x, y, z, t) comes from the simulation results of the CALPUFF modeling module based on the virtual point source parameters and the WRF meteorological field, and the measured concentration data C obs (x, y, z, t) Fixed-point monitoring and cruise monitoring equipment from the multi-source data acquisition module, including VOCs concentration, cruise trajectory, and characteristic component spectrum information; The inverse operation inversion unit uses an improved regularized inversion algorithm to solve the inverse trajectory of pollution transmission based on the data set generated by the data fusion unit. Specifically, by minimizing the mean square error between the measured concentration and the model-predicted concentration and introducing a regularized constraint term, the optimal solution of the virtual point source emission rate vector q is obtained. The calculation formula is as follows: ; Among them, C obs is the measured value of the multi-source data acquisition module, C model(q) is the simulated value based on q by the CALPUFF modeling module; q is the virtual point source emission rate vector. The initial value of q is generated by the area source-point source conversion module by mapping the measured area source emission data using the law of conservation of mass, and is subsequently iteratively adjusted as an optimization variable. The regularization parameter λ is determined through cross-validation based on historical monitoring data, and the constraint matrix L presets a priori constraints on the spatial distribution of pollution sources. The contribution rate calculation unit calculates the pollution contribution rate of each virtual point source to the fixed monitoring point receptor point in the park using the source allocation receptor model based on the virtual point source emission parameters solved by the inverse operation inversion unit; the calculation formula is as follows: ; Among them, Q i is the emission rate of the i-th virtual point source obtained by inversion, C contrib,i is the contribution of the point source to the receptor concentration simulated by the CALPUFF model, f i is the contribution rate of the ith virtual point source, Q j is the pollutant emission rate of the jth virtual point source, which is obtained by the inverse operation inversion unit of the pollution source tracing inversion module through the regularized inversion algorithm. Its initial value comes from the emission parameter mapping unit of the surface source to point source conversion module, which is generated by the mass conservation law mapping based on the measured emission data of the surface source; C contrib,j The jth virtual point source is simulated by the CALPUFF model to simulate the pollutant concentration contribution to the receptor point, which represents the concentration contribution value of the pollutant emitted by the point source at the receptor point. It is generated by the diffusion simulation unit of the microscale CALPUFF modeling module based on the virtual point source coordinates, emission rate Q j and WRF meteorological field data, obtained by simulating the pollutant diffusion process through the Gaussian smoke flow model and Lagrangian particle tracking method; The spatial positioning unit spatially matches the coordinates of virtual point sources whose contribution rates exceed a threshold with a virtual point source standardized parameter set and the surface source area division result generated by the surface source-point source conversion module, and generates a pollution source heat map in combination with a GIS geographic information system. The virtual point source coordinates are determined by the virtual point source generation unit of the surface source-point source conversion module, and the virtual point source standardized parameter set includes the sub-region geometric center coordinates output by the surface source area division unit, the dominant wind direction vector, and the emission rate and concentration parameters generated by the emission parameter mapping unit. Through spatial matching and visualization processing, the pollution source is located at a 500m×500m grid scale.

[0028] The traceability result verification module includes a spectrum data acquisition unit, a characteristic spectrum extraction unit, a matching degree calculation unit, a parameter iterative correction unit and a verification report generation unit; The spectrum data acquisition unit establishes a spatiotemporally matched verification data set by synchronously acquiring the virtual point source characteristic spectrum data output by the contribution rate calculation unit in the pollution source tracing and inversion module and the cruise monitoring spectrum data of the real-time data acquisition unit of the multi-source data acquisition and fusion module; wherein the virtual point source characteristic spectrum is the VOCs component concentration distribution vector of each virtual point source, which is generated by diffusion simulation of the micro-scale CALPUFF modeling module; the cruise monitoring spectrum is acquired by a flight mass spectrometer with a time resolution of 1 hour, and the spatial positioning is matched with the 500m×500m grid of the park; The characteristic spectrum extraction unit performs wavelet transform denoising on the original spectrum data to eliminate instrument noise and environmental interference, completes baseline correction through cubic polynomial fitting, then matches characteristic peaks based on the NIST mass spectrum database, extracts pollutant concentration values, and generates normalized characteristic pollutant spectrum vectors to provide standardized data input for subsequent matching degree calculations; The matching degree calculation unit uses the weighted cosine similarity algorithm to calculate the matching degree index WeightefdSimilarity between the virtual point source spectrum and the cruise monitoring spectrum to quantify the accuracy of the traceability result. The calculation formula is as follows: ; Among them, s model,i is the concentration of the i-th characteristic pollutant in the virtual point source spectrum, which is simulated by the CALPUFF model; s obs,i is the concentration of the i-th characteristic pollutant in the cruise monitoring spectrum; w i is the weight coefficient of the i-th pollutant, which is set according to the degree of harm of the pollutant in the industry standard; n is the number of characteristic pollutant components; When the parameter iteration correction unit detects that the matching degree is lower than a preset threshold, the unit triggers the inverse operation inversion unit of the pollution source tracing inversion module to perform parameter iteration optimization; The verification report generation unit integrates the matching degree calculation results and parameter correction records to generate a visual verification report including a matching degree time series, parameter iteration trajectory, GIS comparison diagram of pollution hotspot areas before and after correction, and optimization suggestions; the matching degree time series reflects the traceability accuracy of each time period, the parameter trajectory records the iteration process of q and λ, and the GIS comparison diagram intuitively displays the differences in the location of pollution sources.

[0029] The parameter iteration correction unit further includes the following contents: When the matching index between the virtual point source spectrum and the cruise monitoring spectrum is lower than the preset threshold, the error function Loss = 1 - Weighted Similarity is used as the optimization target to convert the spectrum matching into an optimizable error index. The gradient descent algorithm is used to iteratively update the virtual point source emission rate vector q and the regularization parameter λ until the matching index meets the standard or the number of iterations reaches the upper limit. The calculation formula is as follows: ; Among them, q k is the virtual point source emission rate vector of the kth iteration, and its initial value is generated by the area source-point source conversion module through the mass conservation law mapping; k+1 is the emission rate vector of the k+1th iteration, which gradually approaches the optimal solution through iteration; α is the learning rate, which controls the iteration step size; Loss is the gradient of the error function with respect to the emission rate vector q, which represents the direction and magnitude of the impact of the change in q on the error; λ k is the regularization parameter of the kth iteration, and its initial value is determined by cross-validation of historical monitoring data from the multi-source data acquisition module; k+1 The regularization parameter for the k+1th iteration; β is the learning rate, and the parameter update step size is less than α; The gradient of the Loss error function with respect to the regularization parameter λ represents the effect of the change of λ on the error; The iterated q and λ are fed back to the inverse operation inversion unit of the pollution source tracing inversion module to recalculate the pollution contribution rate of the virtual point source to the receptor point until the matching degree between the tracing result and the measured data meets the 500m×500m scale requirement.

[0030] A method for monitoring VOCs pollution in an industrial park, comprising the following steps: S1. The system first collects the park's VOCs online monitoring data, meteorological data, pollution source emission parameters and navigation observation trajectory data in real time through the multi-source data acquisition module, and simultaneously performs dynamic cleaning, spatiotemporal calibration and feature extraction on the data to form a standardized data set; S2. Use the area source to point source conversion module to discretize the area source pollution area of ​​the industrial park into virtual point sources. The location of the virtual point source is determined by combining the geometric center coordinates of the sub-area, the dominant wind direction vector, and the terrain undulation correction coefficient. The emission equivalence relationship between area source and point source is established through the law of conservation of mass. S3. Build a microscale CALPUFF model based on a 500m×500m grid precision, integrating parameters such as the coordinates of the virtual point source, emission rate, and pollutant composition with the wind field, temperature and humidity field, and pressure field data generated by WRF to simulate the diffusion and transmission process of VOCs in a small-scale space and output hourly concentration distribution data. S4, the pollution source inversion module integrates the concentration field output by the CALPUFF model with the measured data of VOCs concentration monitored at fixed points and characteristic component spectra of cruise monitoring. It solves the reverse trajectory of pollution transmission through a regularized inversion algorithm, calculates the pollution contribution rate of each virtual point source to the receptor point, and locates the spatial distribution of pollution sources. S5. The traceability result verification module obtains the VOCs component concentration distribution vector of the virtual point source and the dynamic spectrum data of the cruise monitoring. After preprocessing, the weighted cosine similarity algorithm is used to calculate the matching degree. When the matching degree is lower than the threshold, the virtual point source emission rate and regularization parameter are iteratively corrected. S6. Finally, based on the calibrated traceability results and combined with the pollution source list, a GIS heat map of the pollution hotspot area, a matching time series curve, and a visual verification report of the parameter iteration trajectory are generated to provide a decision-making basis for pollution control in the park.

[0031] A 6 km x 6 km chemical park, located in the southwestern suburbs of Hohhot at coordinates 111.68°E, 40.79°N, contains three chemical workshops (area sources), two chimneys (30 m / 20 m high), and a storage tank area (volume source). The terrain elevation is 1000-1050 m, with a prevailing northwesterly wind direction (315°, 28% frequency). During the monitoring period from 8:00 AM to 6:00 PM on June 15, 2025, the average wind speed was 5.2 m / s, the temperature was 22-28°C, and the atmospheric stability varied between CD and CT. The system consists of 10 GC-2030 gas chromatographs (detection limit 0.1 ppm), five weather stations, and a TOF-MS vehicle. Data is stored in HDF5 format with a spatial resolution of 500 m x 500 m and a temporal resolution of 1 hour.

[0032] Online monitoring of Workshop A shows benzene emissions of 1.2 kg / h and toluene of 0.8 kg / h from 8:00 to 9:00, stored in VOCs_Emission.csv; the UAV measured benzene at 52 ppm and toluene at 35 ppm at 8:30 at grid (2,3) (geographic coordinates 1000 m, 1500 m), stored in Airsampling_TOFMS.h5; meteorological station #1 (coordinates (1,1)) recorded wind speed of 5.1 m / s, wind direction of 312°, and temperature of 23°C at 8:00, stored in Meteorology_Station1.txt.

[0033] WRF obtains 1-hour resolution data from the WRF model through the API. The wind speed at the park center from 8:00 to 9:00 was 4.9 m / s and the wind direction was 310°. The original 1 km × 1 km grid was resampled to 500 m × 500 m through bilinear interpolation. After grid (2,2) interpolation, the wind speed was 4.85 m / s and the wind direction was 311°. The data was converted to CALPUFF format and stored in WRF_CALPUFF.h5.

[0034] The outlier value of 1.8 kg / h at 8:15 in the benzene emission rate series of workshop A (mean 1.2 kg / h, standard deviation 0.15 kg / h) was corrected to 1.3 kg / h using the 3σ principle. The missing data from workshop B between 9:00 and 10:00 were supplemented with a benzene concentration of 48 ppm and an emission rate of 20,000 m³ / h using the forward filling method. The cleaned data are stored in Cleaned_Data.h5. The online equipment clock offset is + 5 minutes synchronized via NTP, the navigation trajectory timestamps are aligned with the weather station (error ≤ ±30 seconds), and the geographic coordinates (1000 m, 1500 m) are converted to the grid (1, 2).

[0035] The benzene spectrum was decomposed and denoised using a three-layer DB4 wavelet decomposition. The signal-to-noise ratio (SNR) was improved from 15 dB to 25 dB. Benzene and toluene were extracted as characteristic components using principal component analysis (with a cumulative variance contribution of 85%). A feature matrix with a benzene content of 49.5 ppm and a toluene content of 33.2 ppm was generated and stored in Feature_Spectra.csv.

[0036] The surface source area division is based on the Delaunay triangulation algorithm, which divides the pollution source distribution points in the park (workshop A (1,2), B (2,3), C (3,1)) into 4 sub-areas. The vertices of sub-area 1 are (1,2), (2,2), and (2,3). The geometric center is calculated as: C i = ({1+2+2} / {3}, {2+2+3} / {3}=(1.667, 2.333), (grid coordinates); The virtual point source generates the dominant wind direction 315° and converts it into the unit vector W i = (0.707, -0.707), sub-area 1 elevations 1020m, 1030m, 1025m, calculated standard deviation σ = 4.08m, range 10m, normalized terrain correction coefficient: T i =0.608 The initial virtual point source coordinates are calculated as: P i = (44.667, -40.667); Because it is beyond the park area, T is revised i =0.2 and we get P i=(15.807, -11.807), the actual coordinates are (8403.5m, -5403.5m). Emission parameter mapping and verification sub-area 1 Total emission rate: benzene 1.2+1.0+0.8=3.0kg / h, toluene 0.8+0.6+0.5=1.9kg / h; Gaussian weight function w j = exp(-d_j² / (2ε²), sub-area diagonal ε=500√2 / 3≈235.7m, weight 50m near the source: virtual point source benzene concentration weighted calculation: C=49.0 ppm input CALPUFF simulation 8:30 receptor point (2,3) concentration C mode l=0.115, {mg / m³}, and the measured C obs =0.123{mg / m³}, with a deviation of 6.5%. After adjusting the discharge height to 12m, the deviation is reduced to 4%. The calibration parameters are stored in Calibrated_Source.h5.

[0037] Double grid nesting, with an outer layer of 2000m×2000m (3×3 grid) and an inner layer of 500m×500m (4×4 grid) covering the core area; WRF data were preprocessed by CALMET and combined with an elevation of 1020m to generate a meteorological input field of stability class C. The diffusion parameter was precalculated as: x =105,ơ y =85,ơ z =55, virtual point source integration and diffusion simulation: The virtual point source coordinates are transformed to the inner grid (2,2), the emission direction offset angle is 0°, and linear interpolation is used to generate a 3.0 kg / h emission matrix for the period 9:00-10:00; the concentration at the receptor point (3,3) is calculated as: C = 1.46 × 10 -18 ,{kg / m³}The error function between the measured concentration of 0.10mg / m³ and the simulated value of 0.09mg / m³ is Loss=(0.10-0.09)²=0.0001, adjust ơ y =85m→80m, after that, the Loss drops to 0.000025.

[0038] During the 8:30 monitoring period, the system first used Kalman filtering to fuse the model concentration field with the measured data. The model concentration field Cmodel = 0.11 mg / m³ comes from the CALPUFF simulation, and the measured value Cobs = 0.12 mg / m³ is taken from the cruise monitoring. Iterative calculations are performed based on the state-space equation, where the process noise covariance Q = 0.01 and the observation noise covariance R = 0.02. The final filtering results in the true concentration estimate x = 0.116 mg / m³, effectively eliminating the deviation between the two types of data. The inverse operation inversion unit starts the regularized inversion with the initial emission rate vector q = (3.0, 1.9) kg / h (benzene, toluene), and the objective function is , Where L is the second-order difference matrix. After 5 iterations, q converges to (2.92, 1.83) kg / h, and Loss = 0.001, which meets the inversion accuracy requirements.

[0039] Based on the inversion results, the contribution calculation unit determined that virtual point source 1 contributed 0.08 mg / m³ to the concentration at the receptor site, accounting for 80% of the total contribution of 0.10 mg / m³. Virtual point source 2 contributed 20%. The spatial positioning unit matched the coordinates (2,2) of the virtual point source with a contribution exceeding 70% with the GIS layer and found that it accurately corresponded to the location of workshop A. In the generated heat map, the red area (contribution rate > 70%) completely covers the workshop, and the yellow area (50-70%) is distributed on the surrounding roads. The virtual point source spectrum showed benzene concentrations of 49.0 ppm and toluene concentrations of 32.0 ppm. The cruise spectrum measured 50 ppm of benzene and 33 ppm of toluene at 8:30. The characteristic spectrum extraction unit performed a cubic polynomial baseline correction on the toluene spectrum, correcting the concentration to 32.5 ppm. This was then matched against the NIST database for benzene (m / z = 78, 98% match) and toluene (m / z = 92, 95% match), generating a normalized characteristic vector (0.8, 0.2).

[0040] The matching degree was calculated using the weighted cosine similarity algorithm with a weight coefficient of wi = (0.6, 0.4). Substituting this into the formula yields a numerator of 1470 + 422.4 = 1892.4 and a denominator of 3596. The final matching degree was approximately 0.526, below the preset threshold of 0.9, triggering parameter iteration. The parameter iteration correction unit optimized the target of Loss = 1 - 0.526 = 0.474 and adjusted q to (2.95, 1.85) kg / h and λ to 0.06 using a gradient descent algorithm. After 10 iterations, the matching degree improved to 0.92.

[0041] The verification report shows that the matching degree time series from 8:00 to 9:00 is 0.85→0.92→0.89, and the parameter trajectory record q converges from (3.0, 1.9) to (2.95, 1.85). The GIS comparison map clearly shows that the pollution hotspot in workshop A has been corrected from (2, 3) to (2, 2). Ultimately, it is recommended that the workshop increase its exhaust gas collection efficiency to above 90%.

[0042] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An industrial park VOCs pollution monitoring system, characterized by: It includes a multi-source data acquisition and fusion module, a surface source-point source conversion module, a micro-scale CALPUFF modeling module, a pollution source tracing inversion module and a tracing result verification module; the multi-source data acquisition and fusion module is responsible for real-time collection of VOCs online monitoring, meteorological, pollution source emissions and cruise observation data in the park, and simultaneously completes dynamic cleaning, spatiotemporal calibration and feature extraction to form a standardized data set for subsequent modeling and analysis; the surface source-point source conversion module adopts a virtual point source processing algorithm to discretize the surface source pollution area of ​​the industrial park into regularly distributed virtual point sources, establishes an equivalent relationship between surface sources and point sources of pollutant emissions, and provides input parameters for CALPUFF modeling; the micro-scale CALPUFF modeling ... The model module constructs the CALPUFF model based on the 500m×500m grid accuracy, integrates the virtual point source parameters and the three-dimensional meteorological field generated by WRF, simulates the diffusion and transmission process of VOCs in a small-scale space, and outputs hourly concentration distribution data; the pollution source tracing and inversion module combines the measured data of the monitoring points with the model concentration field, matches the pollution transmission path through inverse operations, calculates the pollution contribution rate of each virtual point source to the receptor point, and locates the spatial distribution of the pollution source; the tracing result verification module verifies the accuracy of the virtual point source tracing through the matching degree analysis of the characteristic pollutant spectra of the cruise monitoring data and the model inversion results, and iteratively corrects the model parameters to improve the tracing accuracy at the 500m×500m scale.

2. The industrial park VOCs pollution monitoring system according to claim 1, characterized in that: The multi-source data acquisition and fusion module includes a real-time data acquisition unit, a WRF data integration unit, a dynamic data cleaning unit, a time-space calibration unit and a feature extraction unit; The real-time data acquisition unit collects VOCs concentration, pollution source emission parameters, three-dimensional meteorological field data and cruise monitoring trajectory data in real time through online monitoring equipment, cruise vehicles and weather stations deployed in the park, and transmits the acquired multi-dimensional raw data to the dynamic data cleaning unit; The WRF data integration unit obtains the three-dimensional meteorological field data output by the mesoscale meteorological model WRF through an API interface, including wind field, temperature and humidity field, and pressure field. The temporal resolution of the data is not less than 1 hour, and the spatial resolution is resampled by a bilinear interpolation algorithm until it matches the spatial resolution of the park's 500m×500m grid; The dynamic data cleaning unit uses an adaptive filtering algorithm to remove outliers and noise from the collected multi-source heterogeneous data, and at the same time fills in missing data through a data integrity verification mechanism to form a preliminary standardized data set; The time-space calibration unit unifies the timestamps and maps the spatial coordinates of data from different sources based on the park's 500m×500m grid coordinate system, eliminating clock bias and spatial positioning errors of monitoring equipment. The feature extraction unit uses wavelet transform to remove spectral noise, uses principal component analysis algorithm to extract meteorological factor coupling characteristics, extracts VOCs characteristic component spectra, pollution plume diffusion characteristics and meteorological factor coupling relationships from the cleaned and calibrated data, and generates a high-temporal and spatial resolution data set suitable for virtual point source modeling. The high-temporal and spatial resolution data set includes a VOCs characteristic spectrum matrix and a meteorological factor load matrix, providing data input for the subsequent surface source-point source conversion module.

3. The industrial park VOCs pollution monitoring system according to claim 1, characterized in that: The surface source-point source conversion module includes a surface source area division unit, a virtual point source generation unit, an emission parameter mapping unit and a model adaptation verification unit; The non-point source area division unit divides the non-point source pollution area of ​​the industrial park into a preset number of sub-areas with regular shapes and similar areas using the Delaunay triangulation algorithm based on the park geographic information, land use type and pollution source distribution data provided by the multi-source data acquisition and fusion module; The virtual point source generation unit determines the location of the virtual point source in each sub-area using a virtual point source placement algorithm based on pollutant diffusion characteristics. The algorithm comprehensively considers the geometric center of the sub-area, the dominant wind direction, and the terrain undulation to determine the coordinates of the virtual point source. The calculation formula is as follows: ； Among them, P i is the coordinate of the virtual point source in the i-th sub-area, C i is the geometric center coordinate of the sub-region, W i is the dominant wind direction vector in the area, T i is the terrain relief correction coefficient; the geometric center coordinates of the sub-area C i After dividing the non-point source pollution area of ​​the industrial park by the non-point source area division unit using the Delaunay triangulation algorithm, the geometric center is calculated based on the coordinates of the sub-area polygon vertices; Dominant wind direction vector W i The park meteorological data collected by the multi-source data acquisition and fusion module is statistically analyzed by calculating the wind direction and speed data within the preset time period, and the wind direction frequency and wind speed weighted average are calculated to determine the direction and intensity of the dominant wind direction in the area and represent them in vector form; the terrain undulation correction coefficient T i Based on the digital elevation model data of the park, the standard deviation and range of terrain elevation in the sub-area are calculated. The range of terrain elevation is the difference between the highest elevation and the lowest elevation in the sub-area. After normalizing the standard deviation and the range of terrain elevation, the coefficient used to correct the position of the virtual point source is obtained. The emission parameter mapping unit maps the emission parameters of each pollutant component of the non-point source to the corresponding virtual point source based on the emission data of each pollution source in the non-point source sub-area through the law of conservation of mass and the spatial weight distribution method, establishes an equivalent relationship between the emission parameters of the non-point source and the virtual point source, and enables the virtual point source to reflect the intensity of non-point source pollution; wherein the parameters of the non-point source include but are not limited to the pollutant emission rate and concentration; specifically: According to the law of conservation of mass, the total pollution emissions of the surface source sub-area are calculated by the addition method, so that the total emissions of the virtual point source are consistent with the surface source, and then the Gaussian distance attenuation function w is used. j = exp(-d j 2 / 2ε 2 ) calculate the weight, where d j is the Euclidean distance from the pollution source to the virtual point source, ε is the characteristic scale of the sub-region, which is 1 / 3 of the length of the sub-region diagonal. The closer the pollution source is to the virtual point source, the higher the weight. Then, the emission parameters of each pollutant component of the non-point source are weighted and summed according to the spatial weight to obtain the spatial weighted mapping of the concentration parameters, and generate the standardized emission parameter set of the virtual point source, which includes: Quantitative parameters: emission rate and emission concentration of each pollutant component; Spatial parameters: emission height and emission direction, which are determined by the average emission height and dominant wind direction vector of the surface source sub-area; Time parameters: emission period distribution, directly inheriting the measured emission period data of non-point source pollution sources; The parameter set is stored in HDF5 format, including the coordinate mapping table, component concentration matrix and time series table, and is output to the model adaptation verification unit; The model adaptation verification unit inputs the virtual point source parameters generated by the emission parameter mapping unit into the small microscale CALPUFF model and verifies the rationality of the parameters through the following process: The CALPUFF model is used to simulate the pollution diffusion process of the virtual point source and output the hourly concentration field. The concentration deviation E of the receptor point is calculated by comparing it with the cruise monitoring and fixed-point monitoring data synchronized with the simulation period in the multi-source data acquisition module. The formula is as follows: ； Among them, C obs is the measured value of the multi-source data acquisition module, C model is the pollutant concentration at the receptor site simulated by the CALPUFF model; If E exceeds the preset threshold, the virtual point source position and emission parameters are adaptively adjusted through the error back propagation algorithm; for position adjustment: based on the terrain elevation range and standard deviation, the virtual point source coordinates P are corrected. i = f(C i , W i , T i ); For emission parameter adjustment: remap the emission rate and concentration of the surface source sub-area, and update the virtual point source parameter set; Iterative adjustments were made until the spatial resolution of 500m×500m was achieved, and the deviation between the model simulation results and the actual monitoring data was controlled within the preset threshold range, providing calibrated point source input data for the subsequent small-scale CALPUFF modeling module.

4. The industrial park VOCs pollution monitoring system according to claim 1, characterized in that: The microscale CALPUFF modeling module includes a grid division unit, a meteorological field processing unit, a virtual point source integration unit, a diffusion simulation unit and an output calibration unit; The grid division unit is based on the 500m×500m spatial resolution requirement and uses a double nested grid technology to divide the industrial park and the peripheral buffer area. The peripheral buffer area refers to the outer grid coverage area surrounding the core simulation area, providing boundary meteorological conditions for the inner high-resolution grid. The outer grid provides boundary conditions for the inner grid and controls the accuracy of small-scale simulation. The inner grid covers the core contaminated area, and the grid spacing is limited to 500m×500m; The meteorological field processing unit receives the WRF model output results provided by the multi-source data acquisition and fusion module, pre-processes the three-dimensional meteorological field data through the CALMET program module, and generates a meteorological input field suitable for the CALPUFF model by combining the park terrain elevation data and land use type; The virtual point source integration unit performs spatiotemporal matching between the virtual point source parameters of the area source-point source conversion module calibrated by the model adaptation and verification unit and the WRF meteorological field data of the multi-source data acquisition and fusion module. The double nested grid technology is used to first convert the virtual point source coordinates from the park geographic coordinate system to the CALPUFF model grid coordinate system through coordinate transformation, and the emission direction parameters are converted into the wind direction offset angle of the CALPUFF source attribute. Then, linear interpolation is performed on the intermittent emission parameters and the WRF meteorological field to generate an hourly emission rate matrix. Subsequently, the virtual point source, the park fixed point source and the volume source are integrated to generate an HDF5 format mixed source emission inventory containing coordinates, emission parameters and source types. The inventory data structure is SourceInventory = (x, y, z, t, Q i , C i , Type, H s , θ), where H s is the emission height, θ is the emission direction, and Type identifies the source type. At the same time, the diffusion coefficient is pre-calculated based on the corresponding grid meteorological parameters extracted from the virtual point source coordinates. The diffusion coefficient ơ is pre-calculated according to the Pasquill-Gifford classification method based on the emission height Hs and the meteorological stability level from WRF. x ,ơ y ,ơ z , ultimately providing spatiotemporally consistent source term input for the diffusion simulation of the microscale CALPUFF modeling module; the HDF5 format contains three levels of data sets, namely coordinate mapping, emission parameters, and meteorological correlation, which are compatible with the CALPUFF model input format; The diffusion simulation unit simulates the pollutant transport process within each 500m×500m grid hourly based on the Gaussian smoke flow model algorithm of the CALPUFF model and the Lagrangian particle tracking method. The Lagrangian particle tracking method is used to calculate the pollutant concentration distribution. The calculation formula is as follows: ； Where C(x,y,z,t) is the pollutant concentration at coordinate (x,y,z) at time t; Q i is the emission rate of the ith virtual point source; (x i ,y i ,z i ) is the virtual point source coordinate; x ,ơ y ,ơ z are the diffusion coefficients in the x, y, and z directions, respectively; The output calibration unit compares the simulated concentration field with the measured data of the multi-source data acquisition module, and uses the same gradient descent algorithm as the traceability result verification module to calibrate the diffusion coefficient ơ x ,ơ y ,ơ z Perform adaptive iterative adjustments; the calibrated concentration field needs to be re-input into the pollution source tracing inversion module, and trigger the source tracing result verification module to calculate the characteristic pollutant spectrum matching degree, until the deviation between the simulated concentration and the measured data at the 500m×500m scale is controlled within the preset threshold, and finally output hourly high temporal and spatial resolution data containing the concentration of each VOCs component to provide model input for the pollution source tracing inversion module, where the iterative calibration trigger condition is that the spectrum matching degree is less than the preset threshold.

5. The industrial park VOCs pollution monitoring system according to claim 1, characterized in that: The pollution source tracing inversion module includes a data fusion unit, an inverse operation inversion unit, a contribution rate calculation unit and a spatial positioning unit; The data fusion unit is responsible for performing spatiotemporal alignment processing on the hourly pollutant concentration field data output by the micro-scale CALPUFF modeling module and the measured data of the multi-source data acquisition and fusion module; specifically, the Kalman filter algorithm is used to eliminate the deviation of the two types of data and generate a fused data set C model (x,y,z,t) and C obs (x, y, z, t); where the model concentration field data C model (x, y, z, t) comes from the simulation results of the CALPUFF modeling module based on the virtual point source parameters and the WRF meteorological field, and the measured concentration data C obs (x, y, z, t) Fixed-point monitoring and cruise monitoring equipment from the multi-source data acquisition module, including VOCs concentration, cruise trajectory, and characteristic component spectrum information; The inverse operation inversion unit uses an improved regularized inversion algorithm to solve the inverse trajectory of pollution transmission based on the data set generated by the data fusion unit. Specifically, by minimizing the mean square error between the measured concentration and the model-predicted concentration and introducing a regularized constraint term, the optimal solution of the virtual point source emission rate vector q is obtained. The calculation formula is as follows: ； Among them, C obs is the measured value of the multi-source data acquisition module, C model (q) is the simulated value based on q by the CALPUFF modeling module; q is the virtual point source emission rate vector. The initial value of q is generated by the area source-point source conversion module by mapping the measured area source emission data using the law of conservation of mass, and is subsequently iteratively adjusted as an optimization variable. The regularization parameter λ is determined through cross-validation based on historical monitoring data, and the constraint matrix L presets a priori constraints on the spatial distribution of pollution sources. The contribution rate calculation unit calculates the pollution contribution rate of each virtual point source to the fixed monitoring point receptor point in the park using the source allocation receptor model based on the virtual point source emission parameters solved by the inverse operation inversion unit; the calculation formula is as follows: ； Among them, Q i is the emission rate of the i-th virtual point source obtained by inversion, C contrib,i is the contribution of the point source to the receptor concentration simulated by the CALPUFF model, f i is the contribution rate of the ith virtual point source, Q j is the pollutant emission rate of the jth virtual point source, which is obtained by the inverse operation inversion unit of the pollution source tracing inversion module through the regularized inversion algorithm. Its initial value comes from the emission parameter mapping unit of the surface source to point source conversion module, which is generated by the mass conservation law mapping based on the measured emission data of the surface source; C contrib,j The jth virtual point source is simulated by the CALPUFF model to simulate the pollutant concentration contribution to the receptor point, which represents the concentration contribution value of the pollutant emitted by the point source at the receptor point. It is generated by the diffusion simulation unit of the microscale CALPUFF modeling module based on the virtual point source coordinates, emission rate Q j and WRF meteorological field data, obtained by simulating the pollutant diffusion process through the Gaussian smoke flow model and Lagrangian particle tracking method; The spatial positioning unit spatially matches the coordinates of virtual point sources whose contribution rates exceed a threshold with a virtual point source standardized parameter set and the surface source area division result generated by the surface source-point source conversion module, and generates a pollution source heat map in combination with a GIS geographic information system. The virtual point source coordinates are determined by the virtual point source generation unit of the surface source-point source conversion module, and the virtual point source standardized parameter set includes the sub-region geometric center coordinates output by the surface source area division unit, the dominant wind direction vector, and the emission rate and concentration parameters generated by the emission parameter mapping unit. Through spatial matching and visualization processing, the pollution source is located at a 500m×500m grid scale.

6. The industrial park VOCs pollution monitoring system according to claim 1, characterized in that: The traceability result verification module includes a spectrum data acquisition unit, a characteristic spectrum extraction unit, a matching degree calculation unit, a parameter iterative correction unit and a verification report generation unit; The spectrum data acquisition unit establishes a verification data set for spatiotemporal matching by synchronously acquiring the virtual point source characteristic spectrum data output by the contribution rate calculation unit in the pollution source tracing and inversion module and the cruise monitoring spectrum data of the real-time data acquisition unit of the multi-source data acquisition and fusion module; The virtual point source characteristic spectrum is the VOCs component concentration distribution vector of each virtual point source, generated by diffusion simulation of the microscale CALPUFF modeling module; the flight monitoring spectrum is obtained by the flight mass spectrometer with a time resolution of 1 hour and spatial positioning matching the 500m×500m grid of the park; The characteristic spectrum extraction unit performs wavelet transform denoising on the original spectrum data to eliminate instrument noise and environmental interference, completes baseline correction through cubic polynomial fitting, then matches characteristic peaks based on the NIST mass spectrum database, extracts pollutant concentration values, and generates normalized characteristic pollutant spectrum vectors to provide standardized data input for subsequent matching degree calculations; The matching degree calculation unit uses the weighted cosine similarity algorithm to calculate the matching degree index WeightefdSimilarity between the virtual point source spectrum and the cruise monitoring spectrum to quantify the accuracy of the traceability result. The calculation formula is as follows: ； Among them, s model,i is the concentration of the i-th characteristic pollutant in the virtual point source spectrum, which is simulated by the CALPUFF model; s obs,i is the concentration of the i-th characteristic pollutant in the cruise monitoring spectrum; w i is the weight coefficient of the i-th pollutant, which is set according to the degree of harm of the pollutant in the industry standard; n is the number of characteristic pollutant components; When the parameter iteration correction unit detects that the matching degree is lower than a preset threshold, the unit triggers the inverse operation inversion unit of the pollution source tracing inversion module to perform parameter iteration optimization; The verification report generation unit integrates the matching degree calculation results and the parameter correction records to generate a visual verification report including a matching degree time series, parameter iteration trajectory, GIS comparison diagrams of pollution hotspot areas before and after correction, and optimization suggestions; The matching degree time series reflects the traceability accuracy in each time period, the parameter trajectory records the iterative process of q and λ, and the GIS comparison chart intuitively shows the differences in pollution source positioning.

7. The industrial park VOCs pollution monitoring system according to claim 6, characterized in that: The parameter iteration correction unit further includes the following contents: When the matching index between the virtual point source spectrum and the cruise monitoring spectrum is lower than the preset threshold, the error function Loss = 1- Weighted Similarity is used as the optimization target to convert the spectrum matching into an optimizable error index. The virtual point source emission rate vector q and the regularization parameter λ are iteratively updated using the gradient descent algorithm until the matching index meets the standard or the number of iterations reaches the upper limit. The calculation formula is as follows: ； Among them, q k is the virtual point source emission rate vector of the kth iteration, and its initial value is generated by the area source-point source conversion module through the mass conservation law mapping; k+1 is the emission rate vector of the k+1th iteration, which gradually approaches the optimal solution through iteration; α is the learning rate, which controls the iteration step size; Loss is the gradient of the error function with respect to the emission rate vector q, which represents the direction and magnitude of the impact of the change in q on the error; λ k is the regularization parameter of the kth iteration, and its initial value is determined by cross-validation of historical monitoring data from the multi-source data acquisition module; k+1 The regularization parameter for the k+1th iteration; β is the learning rate, and the parameter update step size is less than α; The gradient of the Loss error function with respect to the regularization parameter λ represents the effect of the change of λ on the error; The iterated q and λ are fed back to the inverse operation inversion unit of the pollution source tracing inversion module to recalculate the pollution contribution rate of the virtual point source to the receptor point until the matching degree between the tracing result and the measured data meets the 500m×500m scale requirement.

8. A method for monitoring VOCs pollution in an industrial park, applied to an industrial park VOCs pollution monitoring system according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1. The system first collects the park's VOCs online monitoring data, meteorological data, pollution source emission parameters and navigation observation trajectory data in real time through the multi-source data acquisition module, and simultaneously performs dynamic cleaning, spatiotemporal calibration and feature extraction on the data to form a standardized data set; S2. Use the area source to point source conversion module to discretize the area source pollution area of ​​the industrial park into virtual point sources. The location of the virtual point source is determined by combining the geometric center coordinates of the sub-area, the dominant wind direction vector, and the terrain undulation correction coefficient. The emission equivalence relationship between area source and point source is established through the law of conservation of mass. S3. Build a microscale CALPUFF model based on a 500m×500m grid precision, integrating parameters such as the coordinates of the virtual point source, emission rate, and pollutant composition with the wind field, temperature and humidity field, and pressure field data generated by WRF to simulate the diffusion and transmission process of VOCs in a small-scale space and output hourly concentration distribution data. S4, the pollution source inversion module integrates the concentration field output by the CALPUFF model with the measured data of VOCs concentration monitored at fixed points and characteristic component spectra of cruise monitoring. It solves the reverse trajectory of pollution transmission through a regularized inversion algorithm, calculates the pollution contribution rate of each virtual point source to the receptor point, and locates the spatial distribution of pollution sources. S5. The traceability result verification module obtains the VOCs component concentration distribution vector of the virtual point source and the dynamic spectrum data of the cruise monitoring. After preprocessing, the weighted cosine similarity algorithm is used to calculate the matching degree. When the matching degree is lower than the threshold, the virtual point source emission rate and regularization parameter are iteratively corrected. S6. Finally, based on the calibrated traceability results and combined with the pollution source list, a GIS heat map of the pollution hotspot area, a matching time series curve, and a visual verification report of the parameter iteration trajectory are generated to provide a decision-making basis for pollution control in the park.

Images