Air quality prediction and tracing method and system based on coupling of multi-scale models

Abstract

The application discloses an air quality prediction and source tracing method and system based on multi-scale model coupling, and belongs to the technical field of air quality prediction and evaluation. The method comprises the following steps: obtaining a pollution source emission inventory and meteorological field data, driving a regional scale model to perform regional air quality prediction, and obtaining a regional scale concentration field. The regional scale concentration field is taken as a background concentration field, and a local scale model and a small scale model are synchronously driven to perform pollution source contribution concentration prediction, and a local scale contribution concentration field and a small scale contribution concentration field are obtained respectively. Each scale concentration field is fused to generate a comprehensive concentration distribution map and perform source tracing analysis, and air quality influence is dynamically evaluated according to the source tracing analysis result. The application optimizes the accuracy of the regional scale model in small scale range simulation and solves the problem of the lack of a local scale concentration background field, and improves the accuracy of prediction and source tracing.

Description

Technical Field

[0001] This application belongs to the field of air quality prediction and assessment technology, and specifically relates to an air quality prediction and source tracing method and system based on multi-scale model coupling. Background Technology

[0002] With the rapid development of computer and atmospheric chemical analysis technologies, atmospheric chemical and physical processes described by numerical models are closer to the real atmosphere. Currently, these air quality numerical models based on the coupling of atmospheric physical and chemical processes consider the chemical interactions of the gas phase and aerosols in the atmosphere, as well as the physical and dynamic interactions of the atmospheric boundary layer. Their typical characteristic is the coupling of mesoscale atmospheric models and atmospheric chemical models, making them the primary means of studying air pollution prediction and addressing air pollution issues. In the past decade or so, my country has faced relatively serious air pollution problems, exhibiting distinct regional overall characteristics. Through continuous pollution control and prevention, the overall air quality nationwide has improved, and the degree of regional overall pollution has lessened. However, regional pollution problems still exist, and localized characteristic pollution problems are gradually becoming more prominent. With the increasing demand for refined air quality management, higher accuracy and resolution requirements have been placed on models in air quality management in recent years. Early computer technology and the development level of regional air quality models were limited, resulting in insufficient computing power to support large-scale, refined resolution calculations. On the other hand, early air quality models had some problems in multi-model co-operation. For example, the inability to unify model parameter settings, input parameters, and output parameters across different scales, as well as significant differences in model resolution, made model co-operation difficult. Specifically: 1) Local-scale models typically use meteorological parameters from nearby meteorological monitoring stations, resulting in incomplete parameter monitoring types and insufficient representativeness. 2) Local-scale models only simulate the contribution concentration of pollution point sources and cannot simulate the background air quality concentration; they usually use the monitoring concentration from nearby monitoring points as a substitute. 3) Regional-scale models simulate large areas with coarse grids, lacking sufficient refinement. Therefore, they cannot meet the urgent needs of precise pollution source identification and dynamic evaluation of emission reduction measures in current refined air quality management. Summary of the Invention

[0003] To address the aforementioned issues, this application provides an air quality prediction and source tracing method and system based on multi-scale model coupling. This addresses the problems of incomplete and insufficient representative meteorological parameters used in existing local-scale models, their inability to simulate background air quality concentrations, and the insufficient accuracy of regional-scale models in simulating pollution point sources.

[0004] An air quality prediction and source tracing method based on multi-scale model coupling includes: Obtain the pollution source emission inventory and meteorological field data for the simulated area; Based on pollution source emission inventories and meteorological field data, a regional-scale model is driven to predict regional air quality and obtain a regional-scale concentration field. Using the regional-scale concentration field as the background concentration field, the local-scale model and the small-scale model are simultaneously driven to predict the pollution source contribution concentration, and the local-scale contribution concentration field and the small-scale contribution concentration field are obtained respectively. The local-scale and small-scale contribution concentration fields are fused with the background concentration field to generate a comprehensive concentration distribution map. Source tracing analysis is conducted based on the comprehensive concentration distribution map, and the impact on air quality is dynamically assessed based on the source tracing analysis results.

[0005] According to a specific embodiment of the present invention, obtaining the pollution source emission inventory and meteorological field data of the simulated area specifically includes: Obtain the emission inventory of pollution sources within the simulation area and drive the meteorological model to generate real-time simulation and future preset time period meteorological field data covering the simulation area. The meteorological field data includes wind speed, wind direction, temperature, humidity and pressure data.

[0006] According to a specific embodiment of the present invention, after obtaining the pollution source emission inventory and meteorological field data of the simulated area, the method further includes: Preprocessing of pollution source emission inventories and meteorological field data yields preprocessed data.

[0007] According to a specific embodiment of the present invention, preprocessing of pollution source emission inventories and meteorological field data to obtain preprocessed data further includes: The pollution source emission inventory and meteorological field data are mapped to grid area sources and point sources according to three levels: regional scale, local scale and small scale, respectively, to obtain gridded data. The gridded data is subjected to spatiotemporal resolution matching and format standardization to obtain preprocessed data.

[0008] According to a specific embodiment of the present invention, based on pollution source emission inventories and meteorological field data, a regional-scale model is driven to predict regional air quality, and the resulting regional-scale concentration field further includes: Based on pollution source emission inventories and meteorological field data, the regional scale model is driven to predict regional pollutant concentrations, regional air quality (AQI) and regional particulate matter concentrations, respectively, and outputs regional pollutant concentration prediction data, regional air quality (AQI) prediction data and regional particulate matter concentration prediction data, and generates a regional scale concentration field.

[0009] According to a specific embodiment of the present invention, the regional pollutant concentration prediction data includes SO2, NO2, CO, PM2.5, PM10, and O3 concentration data for a future preset time period.

[0010] According to a specific embodiment of the present invention, the regional particulate matter component concentration prediction data includes prediction data for sulfate, nitrate, ammonium salt, organic matter, and primary particulate matter for a future preset time period.

[0011] According to a specific embodiment of the present invention, using a regional-scale concentration field as the background concentration field, and simultaneously driving a local-scale model and a small-scale model to predict the pollution source contribution concentration, the resulting local-scale contribution concentration field and small-scale contribution concentration field further include: Based on pollution source emission inventories and meteorological field data, a local-scale model is driven by a regional-scale concentration field as the background concentration field to predict the emission concentration of enterprises in a local area, outputting the predicted emission concentration data of enterprises in the local area and generating a local-scale contribution concentration field. Simultaneously, it drives the small-scale model to predict the emission concentration of enterprise emission outlets, outputs the emission concentration prediction data of each enterprise emission outlet, and generates a small-scale contribution concentration field.

[0012] According to a specific embodiment of the present invention, the predicted emission concentration data of enterprises in a local area includes the predicted hourly concentration data of SO2, NOx, VOCs and PM of enterprises in the local area for a future preset period.

[0013] According to a specific embodiment of the present invention, the emission concentration prediction data of enterprise emission outlets includes the hourly predicted concentration data of SO2, NOx, VOCs and PM emitted by each enterprise emission outlet in a future preset period. According to a specific embodiment of the present invention, fusing the local-scale contribution concentration field and the small-scale contribution concentration field with the background concentration field to generate a comprehensive concentration distribution map further includes: Based on the local-scale and small-scale contribution concentration fields, the grid concentration values ​​of the regional-scale concentration field in the same simulation area are covered, while the grid concentration values ​​of the regional-scale concentration field in the uncovered area are retained to generate a comprehensive concentration distribution map.

[0014] According to a specific embodiment of the present invention, source tracing analysis based on a comprehensive concentration distribution map and dynamic assessment of the impact on air quality based on the source tracing analysis results further include: Based on the comprehensive concentration distribution map, we conducted analysis of the contributions of regional pollution sources, industry pollution sources, enterprises, and enterprise emission outlets, respectively, and obtained the corresponding concentration values ​​of regional pollution source contributions, industry pollution source contributions, enterprise contributions, and enterprise emission outlet contributions. Based on the analysis results, the emission reduction effects of industrial enterprises and polluting industries were dynamically evaluated.

[0015] According to a specific embodiment of the present invention, based on the comprehensive concentration distribution map, regional pollution source contribution analysis, industry pollution source contribution analysis, enterprise contribution analysis, and enterprise emission outlet contribution analysis are performed respectively, and the corresponding regional pollution source contribution concentration value, industry pollution source contribution concentration value, enterprise contribution concentration value, and enterprise emission outlet contribution concentration value are further included: Based on the comprehensive concentration distribution map, a regional-scale model was used to analyze the regional pollution source contributions of SO2, NO2, PM2.5, PM10, and O3 concentrations at receptor sites, yielding the corresponding regional pollution source contribution concentration values. Based on the comprehensive concentration distribution map, a regional-scale model was used to analyze the industry pollution source contributions of SO2, NO2, PM2.5, PM10, and O3 concentrations at receptor sites, yielding the corresponding industry pollution source contribution concentration values. Based on the comprehensive concentration distribution map, a local-scale model was used to analyze the corporate contribution of SO2, NOx, PM, and VOCs concentrations at receptor sites, yielding corporate contribution concentration values. Based on the comprehensive concentration distribution map, a small-scale model is used to analyze the contribution of SO2, NOx, PM, and VOCs concentrations at the receptor site to enterprise emission outlets, and obtain the enterprise emission outlet contribution concentration values.

[0016] According to a specific embodiment of the present invention, the dynamic evaluation of the emission reduction effects of industrial enterprises and polluting industries based on the analysis results further includes: Based on the regional and industry-specific pollution source contribution concentrations, corresponding emergency measures are dynamically obtained using an input emission reduction ratio method. The reduction in SO2, NO2, PM2.5, PM10, and O3 concentrations of regional industrial enterprises after implementing these emergency measures is dynamically assessed. Based on the enterprise's contribution concentration value, corresponding emergency measures are dynamically obtained using an input emission reduction ratio method. The reduction in SO2, NOx, PM, and VOCs concentrations emitted by the enterprise after implementing these emergency measures is dynamically assessed. Based on the concentration values ​​contributed by the enterprise's emission outlets, the corresponding emergency measures are dynamically obtained using the input emission reduction ratio method, and the reduction in SO2, NOx, PM, and VOCs concentrations emitted by the enterprise's emission outlets after the implementation of emergency measures is dynamically assessed.

[0017] An air quality prediction and source tracing system based on multi-scale model coupling includes: The data acquisition module is used to obtain the pollution source emission inventory and meteorological field data of the simulated area; Regional-scale models are used to predict regional air quality based on pollution source emission inventories and meteorological field data, and to obtain regional-scale concentration fields. The local scale model is used to predict the concentration contribution of enterprise pollution sources against the background concentration field of the regional scale concentration field, and to obtain the local scale contribution concentration field. Small-scale models are used to predict the contribution concentration of pollution sources from enterprise emission outlets against a regional-scale concentration field as a background concentration field, and to obtain a small-scale contribution concentration field. The fusion module is used to fuse the local-scale contribution concentration field and the small-scale contribution concentration field with the background concentration field to generate a comprehensive concentration distribution map. The source analysis and assessment module is used to conduct source analysis based on the comprehensive concentration distribution map and to dynamically assess the impact on air quality based on the source analysis results.

[0018] According to a specific embodiment of the present invention, it further includes: The data processing module is used to preprocess the pollution source emission inventory and meteorological field data to obtain preprocessed data.

[0019] According to a specific embodiment of the present invention, the source tracing analysis and evaluation module further includes: The source analysis module is used to analyze the contribution of regional pollution sources, industry pollution sources, enterprises, and enterprise emission outlets based on the comprehensive concentration distribution map, and obtain the corresponding regional pollution source contribution concentration value, industry pollution source contribution concentration value, enterprise contribution concentration value, and enterprise emission outlet contribution concentration value. The evaluation module is used to dynamically evaluate the emission reduction effects of industrial enterprises and polluting industries based on the analysis results.

[0020] According to a specific embodiment of the present invention, the source tracing analysis module further includes: The first source analysis module is used to drive a regional-scale model to analyze the regional pollution source contribution of SO2, NO2, PM2.5, PM10, and O3 concentrations at receptor sites based on a comprehensive concentration distribution map, and obtain the corresponding regional pollution source contribution concentration values. The second source analysis module is used to drive a regional scale model based on the comprehensive concentration distribution map to analyze the industry pollution source contribution of SO2, NO2, PM2.5, PM10 and O3 concentrations at the receptor site, and obtain the corresponding industry pollution source contribution concentration values. The third source analysis module is used to analyze the corporate contribution of SO2, NOx, PM, and VOCs concentrations at receptor sites based on the comprehensive concentration distribution map and drive the local scale model to obtain the corporate contribution concentration value. The fourth source analysis module is used to drive a small-scale model based on the comprehensive concentration distribution map to analyze the contribution of SO2, NOx, PM, and VOCs concentrations at the receptor site to the enterprise emission outlets, and obtain the enterprise emission outlet contribution concentration value.

[0021] According to a specific embodiment of the present invention, the evaluation module further includes: The first assessment module is used to dynamically obtain corresponding emergency measures based on the regional pollution source contribution concentration value and the industry pollution source contribution concentration value, using the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NO2, PM2.5, PM10 and O3 concentrations of regional industrial enterprises after the implementation of emergency measures. The second assessment module is used to dynamically obtain corresponding emergency measures based on the enterprise's contribution concentration value and the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NOx, PM and VOCs concentrations emitted by the enterprise after taking emergency measures. The third assessment module is used to dynamically obtain corresponding emergency measures based on the enterprise's emission outlet contribution concentration value and the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NOx, PM and VOCs concentrations emitted by the enterprise's emission outlet after the emergency measures are taken.

[0022] An electronic device includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the above-described air quality prediction and source tracing method based on multi-scale model coupling.

[0023] A computer-readable storage medium storing a computer program, which is loaded and executed by a processor to implement the above-described air quality prediction and source tracing method based on multi-scale model coupling.

[0024] Compared with the prior art, this application has the following advantages: (1) Since the meteorological parameters used in traditional local scale models are monitoring data from nearby meteorological monitoring stations, there are problems such as incomplete parameter monitoring types and insufficient representativeness. This invention uses meteorological data calculated by meteorological models to drive local scale models, which can further improve the accuracy of simulated atmospheric pollutants.

[0025] (2) Since the local scale model only simulates the contribution concentration of pollution point sources and cannot simulate the background concentration of air quality, this invention simulates the concentration field of the regional scale model to provide the background concentration field for the local scale and small scale models, thereby improving the accuracy of the local scale model forecast.

[0026] (3) To address the issues of traditional regional-scale models simulating large areas with coarse grids and insufficient detail, this invention utilizes regional-local-small-scale model integration and automation technology. By coupling and combining regional-local-small-scale models, automated air quality prediction and source tracing are conducted, significantly improving the spatiotemporal accuracy and reliability of air quality forecasting and pollution source tracing. The system can accurately reflect the large-scale pollution transmission situation and finely depict the impact of individual enterprises or even emission outlets on the surrounding environment, improving the accuracy of simulating the contribution concentration of pollution point sources.

[0027] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of an air quality prediction and source tracing method based on multi-scale model coupling according to an embodiment of the present invention.

[0030] Figure 2 This is a flowchart of a method for preprocessing pollution source emission inventories and meteorological field data according to an embodiment of the present invention.

[0031] Figure 3 This is a flowchart of a method for predicting pollution source contribution concentrations by using a regional-scale concentration field as the background concentration field and simultaneously driving a local-scale model and a small-scale model, according to an embodiment of the present invention.

[0032] Figure 4 This is a flowchart of a method for performing source tracing analysis based on a comprehensive concentration distribution map and dynamically assessing the impact on air quality based on the source tracing analysis results, according to an embodiment of the present invention.

[0033] Figure 5 This is a flowchart of a method for analyzing the contribution of regional pollution sources, industry pollution sources, enterprises, and enterprise emission outlets based on a comprehensive concentration distribution map, according to an embodiment of the present invention.

[0034] Figure 6This is a flowchart of a method for dynamically evaluating the emission reduction effects of industrial enterprises and polluting industries based on the analysis results, according to an embodiment of the present invention.

[0035] Figure 7 This is a structural diagram of an air quality prediction and source tracing system based on multi-scale model coupling, provided according to an embodiment of the present invention.

[0036] Figure 8 This is a structural diagram of a traceability analysis and evaluation module provided according to an embodiment of the present invention.

[0037] Figure 9 This is a structural diagram of a traceability analysis module provided according to an embodiment of the present invention.

[0038] Figure 10 This is a structural diagram of an evaluation module provided according to an embodiment of the present invention.

[0039] Figure 11 This is a schematic diagram of a computer device structure according to an embodiment of the present invention.

[0040] Figure label: 01-Data Acquisition Module; 02-Data Processing Module; 03-Regional Scale Model; 04-Local Scale Model; 05-Small Scale Model; 06-Fusion Module; 07-Source Analysis and Evaluation Module; 071 - Source tracing analysis module; 072 - Evaluation module; 0711 - First source tracing analysis module; 0712 - Second source tracing analysis module; 0713 - Third source tracing analysis module; 0714 - Fourth source tracing analysis module; 0721 - First evaluation module; 0722 - Second evaluation module; 0723 - Third evaluation module. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] Example 1 Additional aspects and advantages of embodiments of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of the invention. Figures 1-6This invention provides an air quality prediction and source tracing method based on multi-scale model coupling, comprising: S1: Obtain the pollution source emission inventory and meteorological field data for the simulated area.

[0043] S2: Preprocess the pollution source emission inventory and meteorological field data to obtain preprocessed data.

[0044] S3: Based on the pollution source emission inventory and meteorological field data, drive the regional scale model to predict regional air quality and obtain the regional scale concentration field.

[0045] S4: Using the regional-scale concentration field as the background concentration field, the local-scale model and the small-scale model are simultaneously driven to predict the pollution source contribution concentration, and the local-scale contribution concentration field and the small-scale contribution concentration field are obtained respectively.

[0046] S5: The local-scale contribution concentration field and the small-scale contribution concentration field are fused with the background concentration field to generate a comprehensive concentration distribution map.

[0047] S6: Conduct source tracing analysis based on the comprehensive concentration distribution map and dynamically assess the impact on air quality based on the source tracing analysis results.

[0048] Specifically, step S1, obtaining the pollution source emission inventory and meteorological field data for the simulated area, includes: Obtain the emission inventory of pollution sources within the simulation area and drive the meteorological model to generate real-time simulation and future preset time period meteorological field data covering the simulation area. The meteorological field data includes wind speed, wind direction, temperature, humidity and pressure data.

[0049] In a specific embodiment of the present invention, a multi-level pollution source emission inventory covering the simulated area is obtained, including a regional-scale area source inventory, a local-scale point source inventory (enterprise-level), and a small-scale point source inventory (discharge outlet level). The regional-scale area source inventory uses a regular macro-grid (e.g., 3km × 3km) as the unit, and calculates the total emissions of major pollutants such as SO2, NOx, PM, and VOCs from all industrial, residential, road traffic, and agricultural sources within the grid. This data is typically estimated based on energy statistics, activity level data, and emission factor methods. The local-scale point source inventory (enterprise-level) is accurate to the geographical latitude and longitude coordinates of each polluting enterprise, and calculates the total pollutant emissions from enterprises within that area. This data comes from enterprise self-monitoring or environmental statistics databases. The small-scale point source inventory (discharge outlet level) is for large and complex plant areas, further refined to each independent production unit or specific emission outlet, obtaining more precise source strength information and achieving accurate location of emission sources.

[0050] In one specific embodiment of the present invention, a mesoscale meteorological model (e.g., a WRF model) is used for numerical simulation. This meteorological model simulates hourly meteorological fields in real time and for a preset future period (e.g., the next 3-5 days), outputting meteorological field data covering the entire target area for the preset future period, including but not limited to mesoscale meteorological field data such as wind speed, wind direction, temperature, humidity, and pressure. By reading the hourly prediction results, hourly parameters of wind speed, wind direction, temperature, pressure, and humidity in various regions across the country can be visualized. The resulting meteorological field data has a horizontal grid resolution that matches the regional scale model, providing comprehensive meteorological support for more refined local and small-scale models.

[0051] Specifically, step S2 preprocesses the pollution source emission inventory and meteorological field data to obtain preprocessed data, which further includes: S21: Map the pollution source emission inventory and meteorological field data to gridded area sources and point sources at three levels: regional scale, local scale, and small scale, respectively, to obtain gridded data.

[0052] S22: Perform spatiotemporal resolution matching and format standardization on the gridded data to obtain preprocessed data.

[0053] In a specific embodiment of the present invention, to ensure that the multi-source heterogeneous data obtained in step S1 can be correctly and collaboratively used by downstream models, the acquired raw heterogeneous data needs to be transformed into standardized inputs that are spatiotemporally aligned, formatted uniformly, and can directly drive models at various scales. First, the regional-scale area source list is mapped to a relatively coarse grid system (e.g., a 3km × 3km grid) used by the regional-scale model based on its geographical boundaries, with each grid assigned the sum of emissions from all pollution sources within that grid. Meteorological field data generated by the mesoscale meteorological model is directly mapped to or matched to the regional model's grid using conservative interpolation methods, ensuring that each grid is equipped with complete meteorological parameters. Each point source in the local-scale point source (enterprise-level) and small-scale point source (discharge outlet-level) lists is precisely located in the grid system used by the local-scale and small-scale models based on its accurate latitude and longitude coordinates. The meteorological field data of the covered area is downscaled from the coarser regional model grid to the finer local-scale and small-scale model grids, providing high-resolution meteorological drivers for refined simulations. Then, for the pollution source emission inventory, data with different time frequencies (such as annual emissions and monthly emissions) are uniformly allocated to the hourly level. This ensures that the processed hourly emission data of pollution sources completely corresponds to the hourly timestamps of the meteorological field data, achieving synchronization in the time dimension. Next, the gridded and time-matched data is uniformly converted from its original, diverse data formats and output into the standard data format required by the model. This achieves the mapping of source data with different spatial precisions onto a standardized spatial grid corresponding to its scale, thereby providing standardized input parameters for regional-scale models, local-scale models, and small-scale models.

[0054] Specifically, step S3, based on the pollution source emission inventory and meteorological field data, drives a regional-scale model to predict regional air quality, and the resulting regional-scale concentration field further includes: Based on pollution source emission inventories and meteorological field data, a regional-scale model is driven to predict regional pollutant concentrations, regional air quality index (AQI), and regional particulate matter concentrations, respectively. The model outputs corresponding regional pollutant concentration prediction data, regional AQI prediction data, and regional particulate matter concentration prediction data, and generates a regional-scale concentration field. The regional pollutant concentration prediction data includes SO2, NO2, CO, PM2.5, PM10, and O3 concentrations for a pre-defined future time period. The regional particulate matter concentration prediction data includes predictions for sulfate, nitrate, ammonium salts, organic matter, and primary particulate matter for a pre-defined future time period.

[0055] In a specific embodiment of the present invention, after obtaining the pollution source emission inventory and meteorological field data, a regional-scale model is driven to predict regional air quality, including the prediction of SO2, NO2, CO, PM2.5, PM10, and O3 concentrations for the next 3-5 days, the prediction of the Air Quality Index (AQI) for the next 3-5 days, and the prediction of atmospheric particulate matter component concentrations (sulfate, nitrate, ammonium salt, organic matter, and primary particulate matter) for the next 3-5 days. Specifically, the present invention uses a regional-scale photochemical grid model CAMx to simulate the national pollutant concentration output. It can predict the national gridded pollutant concentrations for the next 3-5 days daily, including the hourly concentrations of conventional pollutants and the hourly concentrations of pollutant components. By reading the hourly prediction results, the hourly parameters of SO2, NO2, CO, PM2.5, PM10, and O3 concentrations, regional air quality AQI, and atmospheric particulate matter components such as sulfate, nitrate, ammonium salt, organic matter, and primary particulate matter can be visualized for various regions across the country.

[0056] Specifically, step S4 uses the regional-scale concentration field as the background concentration field, and simultaneously drives the local-scale model and the small-scale model to predict the pollution source contribution concentration, obtaining the local-scale contribution concentration field and the small-scale contribution concentration field respectively. This further includes: S41: Based on pollution source emission inventories and meteorological field data, a local-scale model is used to predict the emission concentrations of local enterprises, driven by a regional-scale concentration field as the background concentration field. The model outputs predicted emission concentration data for local enterprises and generates a local-scale contribution concentration field. The predicted emission concentration data for local enterprises includes hourly predicted concentrations of SO2, NOx, VOCs, and PM2.5 for a predetermined period in the future.

[0057] S42: Simultaneously drive the small-scale model to predict emission concentrations at enterprise emission outlets, outputting predicted emission concentration data for each enterprise's emission outlet and generating a small-scale contribution concentration field. The predicted emission concentration data for enterprise emission outlets includes the hourly predicted concentrations of SO2, NOx, VOCs, and PM emitted by each enterprise's emission outlets in a pre-defined future time period.

[0058] In a specific embodiment of the present invention, after the regional-scale model simulates the regional air quality concentration, a local-scale model and a small-scale model are simultaneously activated to simulate and predict the pollution point source contribution down to the enterprise and emission outlet levels. The regional-scale concentration field obtained by the regional-scale model is used as the background field for pollutant concentrations in the local-scale and small-scale models, respectively, to predict the local pollution source contribution concentration and the small-scale pollution source contribution concentration. Specifically, the present invention uses a local-scale model (e.g., the CALPUFF model) to simulate and predict the impact of industrial enterprise emissions on sensitive locations in a small area over the next 3-5 days. Combined with the meteorological field and pollution source emission inventory provided by the WRF meteorological model, the invention simulates and outputs the hourly predicted concentrations of SO2, NOx, VOCs, and PM at the receptor sites of industrial enterprises that require key monitoring over the next 3-5 days, and visualizes the hourly parameters. Simultaneously, small-scale models (such as the AERMOD model) are used to simulate and predict the impact of enterprise emissions on sensitive sites over the next 3-5 days. Combined with the meteorological field provided by the WRF meteorological model and the emission inventory of industrial enterprises refined to each emission outlet, the simulation outputs the hourly predicted concentrations of SO2, NOx, VOCs, and PM at the receptor sites over the next 3-5 days, and the hourly parameters are visualized.

[0059] Specifically, step S5, which fuses the local-scale contribution concentration field and the small-scale contribution concentration field with the background concentration field to generate a comprehensive concentration distribution map, further includes: Based on the local-scale and small-scale contribution concentration fields, the grid concentration values ​​of the regional-scale concentration field in the same simulation area are covered, while the grid concentration values ​​of the regional-scale concentration field in the uncovered area are retained to generate a comprehensive concentration distribution map.

[0060] In a specific embodiment of the present invention, after the local-scale model and the small-scale model simulations are completed, the simulated concentration of the simulated area by the local-scale model will cover the grid concentration of the same simulated area by the regional-scale model. For areas not covered by the local-scale model, the simulation results of the regional-scale model are still retained, thereby improving the accuracy of the simulation. By fusing the local-scale contribution concentration field and the small-scale contribution concentration field with the regional-scale concentration field (background concentration field), a unified comprehensive concentration distribution map is finally generated. This effectively overcomes the limitations of a single-scale model, generating an air quality distribution image that is closer to reality, providing a reliable data foundation for subsequent accurate source tracing and dynamic assessment.

[0061] Specifically, step S6, which involves source tracing analysis based on the comprehensive concentration distribution map and dynamic assessment of the impact on air quality based on the source tracing analysis results, further includes: S61: Based on the comprehensive concentration distribution map, analyze the contributions of regional pollution sources, industry pollution sources, enterprises, and enterprise emission outlets respectively, to obtain the corresponding concentration values ​​of regional pollution source contributions, industry pollution source contributions, enterprise contributions, and enterprise emission outlet contributions, further including: S611: Based on the comprehensive concentration distribution map, the regional scale model is driven to analyze the regional pollution source contribution of SO2, NO2, PM2.5, PM10 and O3 concentrations at the receptor site, and the corresponding regional pollution source contribution concentration values ​​are obtained.

[0062] S612: Based on the comprehensive concentration distribution map, the regional scale model is driven to analyze the industry pollution source contribution of SO2, NO2, PM2.5, PM10 and O3 concentrations at the receptor site, and the corresponding industry pollution source contribution concentration values ​​are obtained.

[0063] S613: Based on the comprehensive concentration distribution map, the local scale model is driven to analyze the enterprise contribution of SO2, NOx, PM and VOCs concentrations at the receptor site, and obtain the enterprise contribution concentration value.

[0064] S614: Based on the comprehensive concentration distribution map, a small-scale model is driven to analyze the contribution of SO2, NOx, PM and VOCs concentrations at the receptor site to the enterprise emission outlets, and obtain the enterprise emission outlet contribution concentration values.

[0065] In a specific embodiment of the present invention, based on the comprehensive concentration distribution map, the regional pollution source contribution of SO2, NO2, PM2.5, PM10, and O3 concentrations at the receiver site is first analyzed using the particulate matter and ozone source tracing functions in the regional scale model CAMx, and the industry (e.g., industrial, civil, transportation, and power source) pollution source contribution of SO2, NO2, PM2.5, PM10, and O3 concentrations at the receiver site is also analyzed. Correspondingly, regional pollution source contribution concentration values ​​and industry pollution source contribution concentration values ​​are obtained. Then, the contribution of industrial enterprise emissions to the SO2, NOx, PM, and VOCs concentrations at the receiver site is analyzed using local scale and small scale models, respectively. This can be further refined to the contribution of emissions from each enterprise and each emission outlet to the SO2, NOx, PM, and VOCs concentrations at the receiver site, resulting in corresponding enterprise contribution concentration values ​​and enterprise emission outlet contribution concentration values. This invention utilizes a regional-scale model to obtain the contributions of regional and local industries to receptor sites, combined with contributions from individual enterprises and their emission outlets obtained through local-scale and small-scale models. This forms a three-tiered receptor site contribution structure: "regional industry - enterprise - enterprise emission outlet." This structure can be used for dynamic assessment of the impact of regional industry and enterprise emission reductions on air quality. Source tracing analysis of regional industrial pollution sources enhances the regulatory capacity for regional emissions and polluting enterprises. By simulating air quality concentrations at receptor sites and analyzing characteristic pollutants from enterprises, the source of pollution can be traced and analyzed. The results of this dynamic source tracing analysis provide supplementary support for law enforcement agencies.

[0066] S62: Based on the analysis results, conduct dynamic evaluations of the emission reduction effects of industrial enterprises and polluting industries, respectively, further including: S621: Based on the regional pollution source contribution concentration values ​​and industry pollution source contribution concentration values, adopt the input emission reduction ratio method to dynamically obtain corresponding emergency measures, and dynamically assess the reduction in SO2, NO2, PM2.5, PM10, and O3 concentrations of regional industrial enterprises after taking emergency measures.

[0067] S622: Based on the enterprise's contribution concentration value, the corresponding emergency measures are dynamically obtained using the input emission reduction ratio method, and the reduction in SO2, NOx, PM, and VOCs concentrations emitted by the enterprise after taking emergency measures is dynamically assessed.

[0068] S623: Based on the concentration contribution value of the enterprise's emission outlet, the corresponding emergency measures are dynamically obtained by using the input emission reduction ratio method, and the reduction in SO2, NOx, PM and VOCs concentrations emitted from the enterprise's emission outlet after the emergency measures are taken is dynamically assessed.

[0069] In a specific embodiment of the present invention, the present invention dynamically assesses the impact of emission reductions by various industries in a region on the concentration of air pollutants based on the results of source tracing analysis. The assessment primarily focuses on the effectiveness of emergency measures for polluted weather, while the impact of industrial emission reductions on receiver sites is evaluated based on the effectiveness of more refined emission reduction measures. Specifically, after obtaining the contribution concentration of local industries to receiver sites (including the contribution concentration values ​​of regional pollution sources and industry pollution sources), the emergency measures of local industries (industrial, residential, transportation, and power sources) are dynamically obtained by inputting emission reduction ratios. The reduction in SO2, NO2, PM2.5, PM10, and O3 concentrations of regional industrial enterprises after the implementation of emergency measures is dynamically assessed to evaluate the effectiveness of regional control policies and industry-level emission reduction plans. After obtaining the contribution concentration of each enterprise's emissions to the receiver site (enterprise contribution concentration value), the corresponding emergency measures are dynamically obtained by inputting the emission reduction ratio. The reduction in SO2, NOx, PM, and VOCs concentrations of enterprises after taking emergency measures is dynamically assessed. This is used to evaluate the effect of taking mandatory emission reduction measures on individual key polluting enterprises. The assessment results can accurately quantify the pollution contribution and emission reduction potential of each key enterprise, providing a scientific basis for issuing precise emission reduction instructions for implementing differentiated management and control of "one policy for one factory", and greatly improving the pertinence and efficiency of supervision. After obtaining the contribution concentration of each enterprise's emission outlet to the receiver site (enterprise emission outlet contribution concentration value), the corresponding emergency measures are dynamically obtained by inputting the emission reduction ratio. The reduction in SO2, NOx, PM, and VOCs concentrations emitted by the enterprise's emission outlet after taking emergency measures is dynamically evaluated for refined management within the enterprise. This allows control measures to be positioned at specific production stages. The evaluation results can guide enterprises to achieve optimal emission reduction while minimizing production impact. At the same time, it accurately assesses the liability for violations of different production facilities. This invention dynamically evaluates the emission reduction effects of industrial enterprises and polluting industries through steps S621 to S623, transforming the traditional "post-event monitoring" management model into an intelligent management model of "pre-event prediction, in-event simulation, and post-event verification." By pre-evaluating the air quality improvement effects that emission reduction measures of different intensities and ranges may bring, the most cost-effective pollution control plan can be formulated, realizing the scientific, precise, and intelligent management of the environment.

[0070] Example 2 Based on the above method, embodiments of the present invention also provide an air quality prediction and source tracing system based on multi-scale model coupling, such as... Figures 7-10 As shown, it includes: Data acquisition module 01 is used to acquire the pollution source emission inventory and meteorological field data of the simulated area.

[0071] Data processing module 02 is used to preprocess the pollution source emission inventory and meteorological field data to obtain preprocessed data.

[0072] Regional-scale model 03 is used to predict regional air quality based on pollution source emission inventories and meteorological field data, and obtain regional-scale concentration fields.

[0073] Local-scale model 04 is used to predict the concentration contribution of enterprise pollution sources against the background concentration field of the regional-scale concentration field, and obtain the local-scale contribution concentration field.

[0074] Small-scale model 05 is used to predict the contribution concentration of pollution sources from enterprise emission outlets using a regional-scale concentration field as the background concentration field, and obtains a small-scale contribution concentration field.

[0075] The fusion module 06 is used to fuse the local-scale contribution concentration field and the small-scale contribution concentration field with the background concentration field to generate a comprehensive concentration distribution map.

[0076] The Source Analysis and Assessment Module 07 is used to conduct source analysis based on the comprehensive concentration distribution map and to dynamically assess the impact on air quality based on the source analysis results.

[0077] Specifically, the present invention obtains the pollution source emission list within the simulated area through the data acquisition module 01, and drives the meteorological model to generate real-time simulation and future preset time period meteorological field data covering the simulated area, wherein the meteorological field data includes wind speed, wind direction, temperature, humidity and pressure data.

[0078] In a specific embodiment of the present invention, a multi-level pollution source emission inventory covering the simulated area is obtained through the data acquisition module 01. This inventory includes a regional-scale area source inventory, a local-scale point source inventory (enterprise-level), and a small-scale point source inventory (discharge outlet level). The regional-scale area source inventory uses a regular macro-grid (e.g., 3km × 3km) as the unit, and calculates the total emissions of major pollutants such as SO2, NOx, PM, and VOCs from all industrial, residential, road traffic, and agricultural sources within the grid. This data is typically estimated based on energy statistics, activity level data, and emission factor methods. The local-scale point source inventory (enterprise-level) is accurate to the geographical latitude and longitude coordinates of each polluting enterprise, and calculates the total pollutant emissions from enterprises within that area. This data comes from enterprise self-monitoring or environmental statistics databases. The small-scale point source inventory (discharge outlet level) is for large and complex factory areas, further refined to each independent production unit or specific emission outlet, obtaining more accurate source strength information and achieving precise location of emission sources.

[0079] In one specific embodiment of the present invention, the meteorological field data is obtained through numerical simulation using a mesoscale meteorological model (e.g., a WRF model). This meteorological model simulates hourly meteorological fields in real-time and for a preset future period (e.g., the next 3-5 days), outputting meteorological field data covering the entire target area for the preset future period, including but not limited to mesoscale meteorological field data such as wind speed, wind direction, temperature, humidity, and pressure. By reading the hourly prediction results, hourly parameters of wind speed, wind direction, temperature, pressure, and humidity in various regions across the country can be visualized. The resulting meteorological field data has a horizontal grid resolution that matches the regional scale model, providing comprehensive meteorological support for more refined local and small-scale models.

[0080] Specifically, data processing module 02 maps the pollution source emission inventory and meteorological field data to gridded area sources and point sources at three levels: regional scale, local scale, and small scale, respectively, to obtain gridded data. Then, spatiotemporal resolution matching and format standardization are performed on the gridded data to obtain preprocessed data.

[0081] In a specific embodiment of the present invention, to ensure that the acquired multi-source heterogeneous data can be correctly and collaboratively used by downstream models, the acquired raw heterogeneous data needs to be transformed into standardized inputs that are spatiotemporally aligned, formatted uniformly, and can directly drive models at various scales. First, the regional-scale area source list is mapped onto a relatively coarse grid system (e.g., a 3km × 3km grid) used by the regional-scale model according to its geographical boundaries. Each grid is assigned the sum of emissions from all pollution sources within that grid. Meteorological field data generated by the mesoscale meteorological model is directly mapped to or matched to the regional model's grid using conservative interpolation methods, ensuring that each grid is equipped with complete meteorological parameters. Each point source in the local-scale point source (enterprise-level) and small-scale point source (discharge outlet-level) lists is precisely located in the grid system used by the local-scale and small-scale models based on its accurate latitude and longitude coordinates. The meteorological field data of the covered area is downscaled from the coarser regional model grid to the finer local-scale and small-scale model grids, providing high-resolution meteorological drivers for refined simulations. Then, for the pollution source emission inventory, data with different time frequencies (such as annual emissions and monthly emissions) are uniformly allocated to the hourly level. This ensures that the processed hourly emission data of pollution sources completely corresponds to the hourly timestamps of the meteorological field data, achieving synchronization in the time dimension. Next, the gridded and time-matched data is uniformly converted from its original, diverse data formats and output into the standard data format required by the model. This achieves the mapping of source data with different spatial precisions onto a standardized spatial grid corresponding to its scale, thereby providing standardized input parameters for regional-scale models, local-scale models, and small-scale models.

[0082] Specifically, based on pollution source emission inventories and meteorological field data, the regional-scale model 03 is driven to predict regional pollutant concentrations, regional air quality index (AQI), and regional particulate matter concentrations, respectively. The corresponding outputs are regional pollutant concentration prediction data, regional AQI prediction data, and regional particulate matter concentration prediction data, and a regional-scale concentration field is generated. The regional pollutant concentration prediction data includes SO2, NO2, CO, PM2.5, PM10, and O3 concentrations for a preset future time period. The regional particulate matter concentration prediction data includes predictions for sulfate, nitrate, ammonium salts, organic matter, and primary particulate matter for a preset future time period.

[0083] In a specific embodiment of the present invention, after obtaining the pollution source emission inventory and meteorological field data, the regional scale model 03 is driven to predict regional air quality, including the prediction of SO2, NO2, CO, PM2.5, PM10, and O3 concentrations for the next 3-5 days, the prediction of the air quality index (AQI) for the next 3-5 days, and the prediction of atmospheric particulate matter component concentrations (sulfate, nitrate, ammonium salt, organic matter, and primary particulate matter) for the next 3-5 days. Specifically, the present invention uses the regional scale photochemical grid model CAMx to simulate the national pollutant concentration output. It can predict the national gridded pollutant concentrations for the next 3-5 days every day, including the hourly concentrations of conventional pollutants and the hourly concentrations of pollutant components. By reading the hourly prediction results, the hourly parameters of SO2, NO2, CO, PM2.5, PM10, and O3 concentrations, regional air quality index (AQI), and atmospheric particulate matter components such as sulfate, nitrate, ammonium salt, organic matter, and primary particulate matter can be visualized for various regions across the country.

[0084] Specifically, based on pollution source emission inventories and meteorological field data, a local-scale model 04 is used as the background concentration field to predict the emission concentrations of enterprises in a local area. This model outputs predicted emission concentration data for enterprises in the local area and generates a local-scale contribution concentration field. The predicted emission concentration data for enterprises in the local area includes the hourly predicted concentrations of SO2, NOx, VOCs, and PM2.5 for enterprises in the local area over a predetermined time period. Simultaneously, a small-scale model 05 is used to predict the emission concentrations of enterprise emission outlets. This model outputs predicted emission concentration data for each enterprise emission outlet and generates a small-scale contribution concentration field. The predicted emission concentration data for enterprise emission outlets includes the hourly predicted concentrations of SO2, NOx, VOCs, and PM2.5 emitted by each enterprise emission outlet over a predetermined time period.

[0085] In a specific embodiment of the present invention, after the regional-scale model 03 completes the simulation of regional air quality concentration, the local-scale model 04 and the small-scale model 05 are simultaneously activated to perform refined simulation and prediction of pollution point source contributions down to the enterprise and discharge outlet levels. The regional-scale concentration field obtained by the regional-scale model 03 is used as the background field for pollutant concentrations in the local-scale model 04 and the small-scale model 05, respectively, to predict the concentration contributions of local pollution sources and the concentration contributions of small-scale pollution sources. Specifically, the present invention uses the local-scale model 04 to simulate and predict the impact of industrial enterprise emissions on sensitive points in a small area over the next 3-5 days. Combined with the meteorological field and pollution source emission inventory provided by the WRF meteorological model, the invention simulates and outputs the hourly predicted concentrations of SO2, NOx, VOCs, and PM at the receptor points of industrial enterprises that require key monitoring over the next 3-5 days, and visualizes the hourly parameters. Simultaneously, the small-scale model 05 is used to simulate and predict the impact of enterprise emissions on sensitive sites in the next 3-5 days. Combined with the meteorological field provided by the WRF meteorological model and the emission inventory of industrial enterprises refined to each emission outlet, the simulation outputs the hourly predicted concentrations of SO2, NOx, VOCs, and PM at the receptor sites for 3-5 days, and the hourly parameters are visualized.

[0086] Specifically, the fusion module 06 is used to cover the grid concentration values ​​of the regional scale concentration field in the same simulation area based on the local scale contribution concentration field and the small scale contribution concentration field, while retaining the grid concentration values ​​of the regional scale concentration field in the uncovered area, to generate a comprehensive concentration distribution map.

[0087] In a specific embodiment of the present invention, after the simulations of the local-scale model 04 and the small-scale model 05 are completed, the simulated concentration of the simulated area by the local-scale model 04 will cover the grid concentration of the same simulated area by the regional-scale model 03. For areas not covered by the local-scale model, the simulation results of the regional-scale model 03 are still retained, thereby improving the accuracy of the simulation. By fusing the local-scale contribution concentration field and the small-scale contribution concentration field with the regional-scale concentration field (background concentration field), a unified comprehensive concentration distribution map is finally generated. This effectively overcomes the limitations of a single-scale model, generating an air quality distribution image that is closer to reality, providing a reliable data foundation for subsequent accurate source tracing and dynamic assessment.

[0088] Specifically, the source tracing analysis and evaluation module 07 also includes: The source analysis module 071 is used to analyze the contribution of regional pollution sources, industry pollution sources, enterprises, and enterprise emission outlets based on the comprehensive concentration distribution map, and obtain the corresponding regional pollution source contribution concentration value, industry pollution source contribution concentration value, enterprise contribution concentration value, and enterprise emission outlet contribution concentration value.

[0089] The source tracing analysis module 071 also includes: The first source analysis module 0711 is used to drive a regional scale model to analyze the regional pollution source contribution of SO2, NO2, PM2.5, PM10 and O3 concentrations at receptor sites based on the comprehensive concentration distribution map, and obtain the corresponding regional pollution source contribution concentration values.

[0090] The second source analysis module 0712 is used to drive a regional scale model based on the comprehensive concentration distribution map to analyze the industry pollution source contribution of SO2, NO2, PM2.5, PM10 and O3 concentrations at the receptor site, and obtain the corresponding industry pollution source contribution concentration values.

[0091] The third source analysis module 0713 is used to analyze the corporate contribution of SO2, NOx, PM, and VOCs concentrations at receptor sites based on the comprehensive concentration distribution map and drive the local scale model to obtain the corporate contribution concentration value.

[0092] The fourth source analysis module 0714 is used to drive a small-scale model based on the comprehensive concentration distribution map to analyze the contribution of SO2, NOx, PM and VOCs concentrations at the receptor site to the enterprise emission outlets, and obtain the enterprise emission outlet contribution concentration value.

[0093] In a specific embodiment of the present invention, based on a comprehensive concentration distribution map, the regional pollution source contribution of SO2, NO2, PM2.5, PM10, and O3 concentrations at the receiver site is first analyzed using the first source analysis module 0711, and the industry (e.g., industrial, civil, transportation, and power source) pollution source contribution of SO2, NO2, PM2.5, PM10, and O3 concentrations at the receiver site is analyzed using the second source analysis module 0712. Correspondingly, regional pollution source contribution concentration values ​​and industry pollution source contribution concentration values ​​are obtained. Then, the contribution of industrial enterprise emissions to the SO2, NOx, PM, and VOCs concentrations at the receiver site is analyzed using the third source analysis module 0713 and the fourth source analysis module 0714, respectively. This analysis can be further refined to the contribution of each enterprise's emissions from each emission outlet to the SO2, NOx, PM, and VOCs concentrations at the receiver site, resulting in corresponding enterprise contribution concentration values ​​and enterprise emission outlet contribution concentration values. This invention utilizes the contributions of regional and local industries to the recipient sites obtained through the first and second source tracing analysis modules 0711 and 0712, combined with the contributions of each enterprise and each enterprise's emission outlet to the recipient sites obtained through the third and fourth source tracing analysis modules 0713 and 0714, to form a three-tiered recipient site contribution structure: "regional industry - enterprise - enterprise emission outlet." This structure can be used for dynamic assessment of the impact of regional industry emission reduction and enterprise emission reduction on air quality. Source tracing analysis of regional industrial enterprise pollution sources enhances the regulatory capacity for regional emissions and polluting enterprises. By simulating air quality concentrations at recipient sites and analyzing enterprise-specific pollutant factors, the source of pollution can be traced and analyzed. The results of dynamic source tracing analysis provide supplementary support for law enforcement agencies.

[0094] The assessment module 072 is used to dynamically assess the emission reduction effects of industrial enterprises and polluting industries based on the analysis results.

[0095] Evaluation module 072 also includes: The first assessment module 0721 is used to dynamically obtain corresponding emergency measures based on the regional pollution source contribution concentration value and the industry pollution source contribution concentration value, using the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NO2, PM2.5, PM10 and O3 concentrations of regional industrial enterprises after the implementation of emergency measures.

[0096] The second assessment module 0722 is used to dynamically obtain corresponding emergency measures based on the enterprise's contribution concentration value and the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NOx, PM and VOCs concentrations emitted by the enterprise after taking emergency measures.

[0097] The third assessment module 0723 is used to dynamically obtain corresponding emergency measures based on the enterprise's emission outlet contribution concentration value and the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NOx, PM and VOCs concentrations emitted by the enterprise's emission outlet after the emergency measures are taken.

[0098] In a specific embodiment of the present invention, the present invention employs an assessment module 072 to dynamically assess the impact of emission reductions by various industries in a region on the concentration of air pollutants. This assessment primarily focuses on evaluating the effectiveness of emergency measures for polluted weather. The impact of industrial emission reductions on receiver sites is assessed by evaluating the effectiveness of more refined emission reduction measures. Specifically, after obtaining the contribution concentration of local industries to receiver sites (including the contribution concentration values ​​of regional pollution sources and industry pollution sources), the first assessment module 0721 dynamically acquires the emergency measures of local industries (industrial, residential, transportation, and power sources), and dynamically assesses the reduction in SO2, NO2, PM2.5, PM10, and O3 concentrations of regional industrial enterprises after the implementation of emergency measures. This assessment is used to evaluate the effectiveness of regional control policies and industry-level emission reduction plans. After obtaining the contribution concentration of each enterprise's emissions to the receiver site (enterprise contribution concentration value), the second assessment module 0722 is used to dynamically acquire the corresponding emergency measures and dynamically assess the reduction in SO2, NOx, PM, and VOCs concentrations of the enterprise after taking emergency measures. This is used to evaluate the effect of taking mandatory emission reduction measures on individual key polluting enterprises. The assessment results can accurately quantify the pollution contribution and emission reduction potential of each key enterprise, providing a scientific basis for issuing precise emission reduction instructions for implementing differentiated management and control of "one policy for one factory", and greatly improving the pertinence and efficiency of supervision. After obtaining the contribution concentration of each enterprise's emission outlet to the receiver site (enterprise emission outlet contribution concentration value), the third assessment module 0723 dynamically acquires the corresponding emergency measures and dynamically assesses the reduction in SO2, NOx, PM, and VOCs concentrations emitted from the enterprise's emission outlets after taking emergency measures. This assessment is used for refined management within the enterprise, positioning control measures at specific production stages. The assessment results can guide enterprises to achieve optimal emission reduction while minimizing production impact, and accurately assess the liability for violations of different production facilities. This invention transforms the traditional "post-event monitoring" management model into an intelligent management model of "pre-event prediction, in-event simulation, and post-event verification" by dynamically assessing the emission reduction effects of industrial enterprises and polluting industries. By pre-assessing the potential air quality improvement effects of emission reduction measures of different intensities and ranges, the invention formulates the most cost-effective pollution control plan, realizing the scientific, precise, and intelligent management of the environment.

[0099] Example 3 like Figure 11As shown, this embodiment of the invention also provides an electronic device, including a processor and a memory. The memory stores a computer program, which is loaded and executed by the processor to implement the aforementioned air quality prediction and source tracing method based on multi-scale model coupling. The device in this invention can be a server, PC, PAD, mobile phone, etc.

[0100] Furthermore, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which is loaded and executed by a processor to implement the above-described air quality prediction and source tracing method based on multi-scale model coupling.

[0101] In summary, the present invention provides the following advantages: (1) Since the meteorological parameters used in traditional local scale models are monitoring data from nearby meteorological monitoring stations, there are problems such as incomplete parameter monitoring types and insufficient representativeness. This invention uses meteorological data calculated by meteorological models to drive local scale models, which can further improve the accuracy of simulated atmospheric pollutants.

[0102] (2) Since the local scale model only simulates the contribution concentration of pollution point sources and cannot simulate the background concentration of air quality, this invention simulates the concentration field of the regional scale model to provide the background concentration field for the local scale and small scale models, thereby improving the accuracy of the local scale model forecast.

[0103] (3) To address the issues of traditional regional-scale models simulating large areas with coarse grids and insufficient detail, this invention utilizes regional-local-small-scale model integration and automation technology. By coupling and combining regional-local-small-scale models, automated air quality prediction and source tracing are conducted, significantly improving the spatiotemporal accuracy and reliability of air quality forecasting and pollution source tracing. The system can accurately reflect the large-scale pollution transmission situation and finely depict the impact of individual enterprises or even emission outlets on the surrounding environment, improving the accuracy of simulating the contribution concentration of pollution point sources.

[0104] Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for air quality prediction and source tracing based on multi-scale model coupling, characterized in that, include: Obtain the pollution source emission inventory and meteorological field data for the simulated area; Based on pollution source emission inventories and meteorological field data, a regional-scale model is driven to predict regional air quality and obtain a regional-scale concentration field. Using the regional-scale concentration field as the background concentration field, the local-scale model and the small-scale model are simultaneously driven to predict the pollution source contribution concentration, and the local-scale contribution concentration field and the small-scale contribution concentration field are obtained respectively. The local-scale contribution concentration field and the small-scale contribution concentration field are fused with the background concentration field to generate a comprehensive concentration distribution map; Source tracing analysis is conducted based on the comprehensive concentration distribution map, and the impact on air quality is dynamically assessed based on the source tracing analysis results.

2. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 1, characterized in that, The acquisition of the pollution source emission inventory and meteorological field data for the simulated area specifically includes: Obtain a pollution source emission inventory within the simulated area and drive the meteorological model to generate real-time simulation and future preset time period meteorological field data covering the simulated area. The meteorological field data includes wind speed, wind direction, temperature, humidity and pressure data.

3. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 1, characterized in that, After obtaining the pollution source emission inventory and meteorological field data for the simulated area, the process also includes: Preprocessing of pollution source emission inventories and meteorological field data yields preprocessed data.

4. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 3, characterized in that, The preprocessing of the pollution source emission inventory and meteorological field data to obtain preprocessed data further includes: The pollution source emission inventory and meteorological field data are mapped to grid area sources and point sources according to three levels: regional scale, local scale and small scale, respectively, to obtain gridded data. The gridded data is subjected to spatiotemporal resolution matching and format standardization to obtain preprocessed data.

5. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 1, characterized in that, The process of using pollution source emission inventories and meteorological field data to drive a regional-scale model for regional air quality prediction, resulting in a regional-scale concentration field, further includes: Based on pollution source emission inventories and meteorological field data, the regional scale model is driven to predict regional pollutant concentrations, regional air quality (AQI) and regional particulate matter concentrations, respectively, and outputs regional pollutant concentration prediction data, regional air quality (AQI) prediction data and regional particulate matter concentration prediction data, and generates a regional scale concentration field.

6. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 5, characterized in that, The predicted pollutant concentration data for the region includes SO2, NO2, CO, PM2.5, PM10, and O3 concentration data for a future preset time period.

7. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 5, characterized in that, The predicted concentration data of particulate matter components in the region includes predicted data of sulfate, nitrate, ammonium salt, organic matter and primary particulate matter for a future preset time period.

8. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 1, characterized in that, The process of using a regional-scale concentration field as the background concentration field and simultaneously driving local-scale and small-scale models to predict pollution source contribution concentrations, thereby obtaining local-scale and small-scale contribution concentration fields respectively, further includes: Based on pollution source emission inventories and meteorological field data, a local-scale model is driven by a regional-scale concentration field as the background concentration field to predict the emission concentration of enterprises in a local area, outputting the predicted emission concentration data of enterprises in the local area and generating a local-scale contribution concentration field. Simultaneously, it drives the small-scale model to predict the emission concentration of enterprise emission outlets, outputs the emission concentration prediction data of each enterprise emission outlet, and generates a small-scale contribution concentration field.

9. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 8, characterized in that, The predicted emission concentration data for local enterprises includes SO2 and NO emissions from local enterprises in a pre-defined future time period. x Hourly predicted concentration data for VOCs and PM2.

5.

10. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 8, characterized in that, The predicted emission concentration data for the enterprise emission outlets includes the SO2 and NO emissions from each enterprise emission outlets in a preset future time period. x Hourly predicted concentration data for VOCs and PM2.

5.

11. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 1, characterized in that, The step of fusing the local-scale contribution concentration field and the small-scale contribution concentration field with the background concentration field to generate a comprehensive concentration distribution map further includes: Based on the local-scale and small-scale contribution concentration fields, the grid concentration values ​​of the regional-scale concentration field in the same simulation area are covered, while the grid concentration values ​​of the regional-scale concentration field in the uncovered area are retained to generate a comprehensive concentration distribution map.

12. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 1, characterized in that, The process of conducting source tracing analysis based on comprehensive concentration distribution maps and dynamically assessing the impact on air quality based on the source tracing analysis results further includes: Based on the comprehensive concentration distribution map, we conducted analysis of the contributions of regional pollution sources, industry pollution sources, enterprises, and enterprise emission outlets, respectively, and obtained the corresponding concentration values ​​of regional pollution source contributions, industry pollution source contributions, enterprise contributions, and enterprise emission outlet contributions. Based on the analysis results, the emission reduction effects of industrial enterprises and polluting industries were dynamically evaluated.

13. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 12, characterized in that, The analysis of regional pollution source contributions, industry pollution source contributions, enterprise contributions, and enterprise emission outlet contributions based on the comprehensive concentration distribution map, respectively, further includes the following: Based on the comprehensive concentration distribution map, a regional-scale model was used to analyze the regional pollution source contributions of SO2, NO2, PM2.5, PM10, and O3 concentrations at receptor sites, yielding the corresponding regional pollution source contribution concentration values. Based on the comprehensive concentration distribution map, a regional-scale model was used to analyze the industry pollution source contributions of SO2, NO2, PM2.5, PM10, and O3 concentrations at receptor sites, yielding the corresponding industry pollution source contribution concentration values. Based on the comprehensive concentration distribution map, a local-scale model was used to analyze the corporate contribution of SO2, NOx, PM, and VOCs concentrations at receptor sites, yielding corporate contribution concentration values. Based on the comprehensive concentration distribution map, a small-scale model is used to analyze the contribution of SO2, NOx, PM, and VOCs concentrations at the receptor site to enterprise emission outlets, and obtain the enterprise emission outlet contribution concentration values.

14. The air quality prediction and source tracing method based on multi-scale model coupling according to claim 12, characterized in that, The dynamic evaluation of the emission reduction effects of industrial enterprises and polluting industries based on the analysis results further includes: Based on the regional and industry-specific pollution source contribution concentrations, corresponding emergency measures are dynamically obtained using an input emission reduction ratio method. The reduction in SO2, NO2, PM2.5, PM10, and O3 concentrations of regional industrial enterprises after implementing these emergency measures is dynamically assessed. Based on the enterprise's contribution concentration value, corresponding emergency measures are dynamically obtained using an input emission reduction ratio method. The reduction in SO2, NOx, PM, and VOCs concentrations emitted by the enterprise after implementing these emergency measures is dynamically assessed. Based on the concentration values ​​contributed by the enterprise's emission outlets, the corresponding emergency measures are dynamically obtained using the input emission reduction ratio method, and the reduction in SO2, NOx, PM, and VOCs concentrations emitted by the enterprise's emission outlets after the implementation of emergency measures is dynamically assessed.

15. An air quality prediction and source tracing system based on multi-scale model coupling, characterized in that, include: The data acquisition module is used to obtain the pollution source emission inventory and meteorological field data of the simulated area; Regional-scale models are used to predict regional air quality based on pollution source emission inventories and meteorological field data, and to obtain regional-scale concentration fields. The local scale model is used to predict the concentration contribution of enterprise pollution sources against the background concentration field of the regional scale concentration field, and to obtain the local scale contribution concentration field. Small-scale models are used to predict the contribution concentration of pollution sources from enterprise emission outlets against a regional-scale concentration field as a background concentration field, and to obtain a small-scale contribution concentration field. The fusion module is used to fuse the local-scale contribution concentration field and the small-scale contribution concentration field with the background concentration field to generate a comprehensive concentration distribution map; The source analysis and assessment module is used to conduct source analysis based on the comprehensive concentration distribution map and to dynamically assess the impact on air quality based on the source analysis results.

16. The air quality prediction and source tracing system based on multi-scale model coupling according to claim 15, characterized in that, Also includes: The data processing module is used to preprocess the pollution source emission inventory and meteorological field data to obtain preprocessed data.

17. The air quality prediction and source tracing system based on multi-scale model coupling according to claim 15, characterized in that, The source tracing analysis and evaluation module also includes: The source analysis module is used to analyze the contribution of regional pollution sources, industry pollution sources, enterprises, and enterprise emission outlets based on the comprehensive concentration distribution map, and obtain the corresponding regional pollution source contribution concentration value, industry pollution source contribution concentration value, enterprise contribution concentration value, and enterprise emission outlet contribution concentration value. The evaluation module is used to dynamically evaluate the emission reduction effects of industrial enterprises and polluting industries based on the analysis results.

18. The air quality prediction and source tracing system based on multi-scale model coupling according to claim 17, characterized in that, The source tracing analysis module also includes: The first source analysis module is used to drive a regional-scale model to analyze the regional pollution source contribution of SO2, NO2, PM2.5, PM10, and O3 concentrations at receptor sites based on a comprehensive concentration distribution map, and obtain the corresponding regional pollution source contribution concentration values. The second source analysis module is used to drive a regional scale model based on the comprehensive concentration distribution map to analyze the industry pollution source contribution of SO2, NO2, PM2.5, PM10 and O3 concentrations at the receptor site, and obtain the corresponding industry pollution source contribution concentration values. The third source analysis module is used to analyze the corporate contribution of SO2, NOx, PM, and VOCs concentrations at receptor sites based on the comprehensive concentration distribution map and drive the local scale model to obtain the corporate contribution concentration value. The fourth source analysis module is used to drive a small-scale model based on the comprehensive concentration distribution map to analyze the contribution of SO2, NOx, PM, and VOCs concentrations at the receptor site to the enterprise emission outlets, and obtain the enterprise emission outlet contribution concentration value.

19. The air quality prediction and source tracing system based on multi-scale model coupling according to claim 17, characterized in that, The evaluation module also includes: The first assessment module is used to dynamically obtain corresponding emergency measures based on the regional pollution source contribution concentration value and the industry pollution source contribution concentration value, using the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NO2, PM2.5, PM10 and O3 concentrations of regional industrial enterprises after the implementation of emergency measures. The second assessment module is used to dynamically obtain corresponding emergency measures based on the enterprise's contribution concentration value and the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NOx, PM and VOCs concentrations emitted by the enterprise after taking emergency measures. The third assessment module is used to dynamically obtain corresponding emergency measures based on the enterprise's emission outlet contribution concentration value and the input emission reduction ratio method, and to dynamically assess the reduction in SO2, NOx, PM and VOCs concentrations emitted by the enterprise's emission outlet after the emergency measures are taken.

20. An electronic device, characterized in that, include: A processor and a memory, wherein the memory stores a computer program, which is loaded and executed by the processor to implement the air quality prediction and source tracing method based on multi-scale model coupling as described in any one of claims 1 to 14.

21. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the air quality prediction and source tracing method based on multi-scale model coupling as described in any one of claims 1 to 14.

Images