Pollution source quantitative tracing method and device and storage medium

Abstract

The invention relates to a pollution source quantitative traceability method, equipment and a storage medium. The method comprises the following steps: acquiring atmospheric component data and a gas-related enterprise emission list of a target detection area; analyzing the atmospheric component data by using a pre-constructed atmospheric pollutant characteristic detection model to obtain initial pollutant characteristics and potential pollutant areas, determining acquisition nodes, and performing pollutant concentration dynamic monitoring; performing cross validation on the dynamic monitoring data and industry data in the emission list of the gas-related enterprise by using an atmospheric pollutant characteristic detection model to obtain a pollutant thermodynamic analysis chart; further generating a real-time wind field diagram and an industry contribution thermodynamic diagram; and according to the industry contribution thermodynamic diagram and the real-time wind field diagram, pollution source quantitative traceability is carried out. According to the technical scheme disclosed by the invention, high precision and dynamic traceability of the pollution source are realized by combining thermodynamic analysis and the pollutant transmission path, the transmission path of the pollutant can be accurately simulated under various meteorological conditions, and the precision is high.

Description

Technical Field

[0001] The present invention relates to the technical field of pollution prevention and control, and particularly to a method, device and storage medium for quantifying and tracing pollution sources. Background Art

[0002] In recent years, with the acceleration of urbanization and industrialization processes, the air pollution problem has become increasingly severe. Harmful gases such as fine particulate matter (PM2.5), nitrogen oxides (NOx), sulfur dioxide (SO2), etc. have had extremely prominent negative impacts on the environment and human health. In the work of pollutant treatment, accurately identifying pollution sources and quantifying their contributions has become the core point for effectively controlling and improving air quality. However, there are many problems in the field of pollution source tracing in the existing technology. Traditional methods mostly rely on static monitoring data and are lacking in comprehensively analyzing aspects such as the spatio-temporal changes of pollution sources, meteorological factors, and transmission processes. This makes it difficult for pollution source tracing to meet the actual requirements in terms of accuracy and timeliness.

[0003] Currently, various pollution source identification and tracing technologies such as the Chemical Mass Balance method (CMB) and the Positive Matrix Factorization method (PMF) have been applied. However, as a traditional pollution source tracing method, the Chemical Mass Balance method has many drawbacks. It is based on pollutant concentration data at sampling points and calculates the contributions of pollution sources by combining chemical components, requiring a large amount of monitoring data, including pollutant concentrations in air samples, pollution source emission factors, and meteorological data, etc., and solving through establishing a system of linear equations. However, this method belongs to a static model and cannot consider the dynamic changes of pollutants in space and time, such as the differences in pollution source emission characteristics in different seasons or time periods; its spatio-temporal resolution is limited, mostly analyzing a single monitoring point, and it is difficult to jointly analyze pollution sources at multiple sites or in a large-scale area, and it does not consider the atmospheric transmission model and cannot accurately simulate the pollutant propagation path; it is highly dependent on monitoring data, data errors or omissions will lead to result deviations, and it is difficult to obtain accurate pollution source lists and source characteristic data in some areas; it also assumes that the influences of different pollution sources are independent and it is difficult to handle the cross-influence between multiple pollution sources in the actual environment, and the tracing accuracy is poor under complex meteorological conditions.

[0004] Although the forward trajectory method traces the pollution sources based on meteorological and pollutant concentration data and calculates the sources by simulating the pollutant transmission paths, it also has problems. Its accuracy highly depends on the quality of meteorological data. If the meteorological data is incorrect or incomplete, it will wrongly simulate the pollutant transmission paths and affect the reliability of the tracing results. In some areas, the meteorological monitoring data is also spatiotemporally discontinuous or has insufficient resolution; the spatial resolution is limited, and it is difficult to simulate the microscopic effects of terrain, buildings, etc. on pollutant propagation in complex geographical or urban environments; it often assumes that the pollutant emissions are continuous and stable, ignoring the changes in emission intensity and type over time in reality, such as being affected by factors like seasons, meteorology, and human activities; and it assumes that the contributions of pollution sources are independent and cannot effectively handle the interactions between different pollution sources, which may lead to inaccurate assessment of source contributions.

[0005] In summary, the existing technologies are difficult to cope with complex pollution source distributions and transmission paths, cannot integrate multi-dimensional data, and have insufficient accuracy. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a method, device, and storage medium for quantitative tracing of pollution sources to solve the problems that the existing technologies are difficult to cope with complex pollution source distributions and transmission paths, cannot integrate multi-dimensional data, and have insufficient accuracy.

[0007] According to the first aspect of the embodiments of the present invention, a method for quantitative tracing of pollution sources is provided, including: Obtaining atmospheric component data, emission inventories of gas-related enterprises, and global wind field simulation data for the target detection area; the atmospheric component data includes wind profiler radar data, pollutant concentration monitoring data, and relevant meteorological data; Analyzing the atmospheric component data by using a pre-constructed atmospheric pollutant feature detection model to obtain preliminary pollutant features and potential pollutant areas; Determining collection nodes according to the preliminary pollutant features and potential pollutant areas, and dynamically monitoring the pollutant concentration at the collection nodes to obtain dynamic monitoring data; the dynamic monitoring data includes the concentration level and time variation law of the pollutants; Using the atmospheric pollutant feature detection model to cross-validate the dynamic monitoring data with the industry data in the emission inventories of gas-related enterprises to obtain a pollutant thermal analysis map; Integrating the pollutant thermal analysis map and the wind profiler radar data into a real-time wind field map of the target detection area, and marking the spatial distribution of the target pollutants in the wind field in the real-time wind field map; according to the global wind field simulation data, constructing the migration and diffusion paths and spatio-temporal distribution laws of the pollutants in the target area in the real-time wind field map; Conducting quantitative analysis on the pollutant thermal analysis map and the industry contribution data in the emission inventories of gas-related enterprises to generate an industry contribution thermal map; Quantitatively trace the pollution sources based on the industry contribution heat map and the real-time wind field map.

[0008] Preferably, the atmospheric pollutant characteristic detection model includes: a BP neural network and an LSTM neural network constructed using a deep learning algorithm framework, and a thermal analysis method is embedded in the atmospheric pollutant characteristic detection model.

[0009] Preferably, analyze the atmospheric component data using a pre-constructed atmospheric pollutant characteristic detection model to obtain preliminary pollutant characteristics and potential pollutant regions, including: Analyze the atmospheric component data using the BP neural network and the LSTM neural network to obtain the pollutant migration path and the concentration distribution characteristics; Analyze the atmospheric component data using the thermal analysis method to obtain a pollutant concentration distribution trend map; Based on the pollutant migration path, the concentration distribution characteristics, and the pollutant concentration distribution trend map, obtain the preliminary pollutant characteristics and the potential pollutant regions.

[0010] Preferably, quantitatively tracing the pollution sources based on the industry contribution heat map and the real-time wind field map further includes: Construct a pollution source tracing path model for the target detection area based on the industry contribution heat map and the real-time wind field map; Run the pollution source tracing path model, query the emission data within the target detection area, and use the pollution source tracing path model to match the pollutant characteristics with the emission data to obtain the pollution source area.

[0011] Preferably, the pollution source quantitative tracing method further includes: Simulate the diffusion path of pollutants from the pollution source area to the target detection area according to the real-time wind field map; Optimize the diffusion path using a machine learning prediction model.

[0012] Preferably, the pollution source quantitative tracing method further includes: Quantitatively evaluate the industry contribution of pollutants according to the industry contribution heat map to obtain the contribution ratio of a specific industry in a pollution event.

[0013] According to the second aspect of the embodiments of the present invention, there is provided a pollution source quantitative tracing device, including: A main controller and a memory connected to the main controller; The memory stores program instructions; The main controller is configured to execute the program instructions stored in the memory to execute the method described in any one of the above.

[0014] According to a third aspect of the embodiments of the present invention, there is provided a computer-readable storage medium storing a computer program, which when executed by a processor, implements the method described in any one of the above.

[0015] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: It can be understood that the technical solutions shown in the present invention can obtain the atmospheric component data of the target detection area and the emission list of gas-related enterprises; analyze the atmospheric component data using a pre-constructed atmospheric pollutant feature detection model to obtain preliminary pollutant features and potential pollutant areas, determine the collection nodes, and perform dynamic monitoring of pollutant concentrations; use the atmospheric pollutant feature detection model to cross-validate the dynamic monitoring data with the industry data in the emission list of gas-related enterprises to obtain a pollutant thermal analysis map; and then generate a real-time wind field map and an industry contribution thermal map; according to the industry contribution thermal map and the real-time wind field map, perform quantitative source tracing of pollution sources. The technical solutions shown in the present invention achieve high-precision and dynamic source tracing by combining thermal analysis and pollutant transmission paths, and can accurately simulate the pollutant transmission paths under various meteorological conditions with high precision.

[0016] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present invention and used together with the specification to explain the principles of the present invention.

[0018] Figure 1 is a schematic diagram of the steps of a method for quantitative source tracing of pollution sources shown according to an exemplary embodiment; Figure 2 is a flowchart of a method for quantitative source tracing of pollution sources shown according to an exemplary embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are only examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0020] In one embodiment, there is provided a method for quantitative source tracing of pollution sources, including: Step S11: Obtain the atmospheric component data, emission inventory of gas-related enterprises, and global wind field simulation data for the target detection area; the atmospheric component data includes wind profiler radar data, pollutant concentration monitoring data, and relevant meteorological data.

[0021] Step S12: Analyze the atmospheric component data using a pre-constructed atmospheric pollutant characteristic detection model to obtain preliminary pollutant characteristics and potential pollutant areas.

[0022] Step S13: Determine the collection nodes based on the preliminary pollutant characteristics and potential pollutant areas, and conduct dynamic monitoring of pollutant concentrations at the collection nodes to obtain dynamic monitoring data; the dynamic monitoring data includes the concentration level and time variation law of pollutants.

[0023] Step S14: Use the atmospheric pollutant characteristic detection model to perform cross-validation on the dynamic monitoring data and industry data in the emission inventory of gas-related enterprises to obtain a pollutant thermal analysis map.

[0024] Step S15: Integrate the pollutant thermal analysis map and the wind profiler radar data into a real-time wind field map of the target detection area, and mark the spatial distribution of target pollutants in the wind field on the real-time wind field map; according to the global wind field simulation data, construct the migration and diffusion path and spatio-temporal distribution law of pollutants in the target area on the real-time wind field map.

[0025] Step S16: Perform quantitative analysis on the pollutant thermal analysis map and the industry contribution data in the emission inventory of gas-related enterprises to generate an industry contribution thermal map.

[0026] Step S17: Conduct quantitative source tracing of pollution sources based on the industry contribution thermal map and the real-time wind field map.

[0027] It can be understood that the technical solution shown in the present invention integrates multi-source data, realizes a complete process from data collection, analysis to pollution source tracing, and improves the accuracy and comprehensiveness of quantitative pollution source tracing. Through dynamic monitoring and cross-validation, anomalies and errors in the data can be detected in a timely manner, ensuring the reliability of the analysis results and providing strong data support for subsequent precise pollution control. By combining thermal analysis with pollutant transmission paths, high-precision and dynamic pollution source tracing are achieved, and the transmission paths of pollutants can be accurately simulated under various meteorological conditions with high precision.

[0028] In specific practice, refer to Figure 2 , in terms of data acquisition, obtain the atmospheric component data, emission inventory of gas-related enterprises, and global wind field simulation data, and the atmospheric component data includes wind profiler radar data, pollutant concentration monitoring data, and relevant meteorological data.

[0029] Preferably, after data acquisition, data cleaning and standardization processing can be performed on the data.

[0030] After data preprocessing, the data is input into the atmospheric pollutant characteristic detection model.

[0031] It should be noted that the atmospheric pollutant characteristic detection model includes: a BP neural network and an LSTM neural network constructed using a deep learning algorithm framework, and a thermal analysis method is embedded in the atmospheric pollutant characteristic detection model.

[0032] The model uses a BP neural network and an LSTM neural network to construct the target pollutant characteristic detection ability. The BP neural network and the LSTM neural network are used to analyze the atmospheric component data to obtain the pollutant migration path and concentration distribution characteristics.

[0033] The thermal analysis method is embedded in the atmospheric pollutant characteristic detection model. The thermal analysis method is used to analyze the atmospheric component data to obtain a pollutant concentration distribution trend map.

[0034] According to the pollutant migration path, concentration distribution characteristics, and pollutant concentration distribution trend map, preliminary pollutant characteristics and potential pollutant areas are obtained. At the same time, by analyzing the characteristics of pollutants in the target area through the model and combining the data of different pollutant components, an association relationship between pollutant characteristics and industry emission information is established.

[0035] Based on the preliminary pollutant characteristics and potential pollutant areas in the target detection area, the distribution range of the target pollutant is further confirmed, and the collection node positions in the detection area are selected. By deploying sensor devices and monitoring instruments, real-time concentration data of the target pollutant (such as PM2.5) at different nodes are collected. The collected monitoring data will include the concentration level and time variation law of the pollutant, and will be cross-validated with the industry data in the regional emission inventory to update the preliminary pollutant characteristics and potential pollutant areas, and obtain a pollutant thermal analysis map.

[0036] The pollutant thermal analysis map mainly presents the distribution trend of pollutant concentration, visually shows the high and low changes of pollutant concentration in different areas, and enables users to quickly understand the severity and distribution range of pollution. At the same time, it can also show the possible pollution source areas, help determine the location of the pollution source, and provide important clues for subsequent traceability work.

[0037] The pollutant thermal analysis map and the wind profiler radar data are integrated into a real-time wind field map of the target detection area, and the spatial distribution of the target pollutant in the wind field is marked in the real-time wind field map. By projecting the pollutant concentration data into the wind field model and combining the global wind field simulation data, the migration and diffusion path and spatio-temporal distribution law of the pollutant in the target area are constructed in the real-time wind field map.

[0038] Through the quantitative evaluation model of industry contribution based on PSCF + linear regression combined with the thermal analysis technology, the pollutant distribution data in the pollutant thermal analysis map is quantitatively analyzed with the industry contribution data in the regional emission inventory to generate a thermal distribution map, visually showing the possible industry contributions and regional distribution characteristics of pollution sources.

[0039] It should be noted that according to the industry contribution thermal map and the real-time wind field map, for the quantitative tracing of pollution sources, it also includes: Construct a pollution source tracing path model for the target detection area according to the industry contribution thermal map and the real-time wind field map; run the pollution source tracing path model, query the emission data within the scope of the target detection area, and match the pollutant characteristics with the emission data by using the pollution source tracing path model to obtain the pollution source area.

[0040] Simulate the diffusion path of pollutants from the pollution source area to the target detection area according to the real-time wind field map; optimize the diffusion path by using a machine learning prediction model.

[0041] Quantitatively evaluate the industry contribution of pollutants according to the industry contribution thermal map to obtain the contribution ratio of specific industries in pollution incidents.

[0042] In specific practice, during the operation of the model, query the emission inventory data within the target area, match the characteristics of the target pollutant with the emission data of relevant industries, and lock the potential pollution source area. Simulate the diffusion path of pollutants from the emission source to the target area through real-time wind field data, and at the same time, combined with a machine learning prediction model, optimize the tracing path and output the positioning result of the pollution source. At the same time, quantitatively evaluate the industry contribution of pollutants through thermal analysis to obtain the contribution ratio of specific industries in pollution incidents.

[0043] The present invention realizes the accurate prediction of PM2.5 concentration and the efficient positioning of pollution sources by combining thermal analysis, artificial intelligence prediction models and pollution tracing technologies. Not only is the prediction accuracy improved by about 30% compared with traditional methods, but also the pertinence and real-time performance of pollution control are significantly improved. At the same time, the system's dynamic visualization function and regional joint prevention and control ability effectively reduce the control cost and optimize energy consumption, providing scientific support for environmental management, public health protection and smart city construction.

[0044] The following is a specific embodiment to illustrate the technical solution of the present invention.

[0045] Suppose pollution source quantification and tracing work is carried out in an industrial city A. There are multiple industrial areas, transportation hubs and residential areas in this city, and the air pollution problem is relatively prominent, and the PM2.5 pollution situation is serious.

[0046] First, use a wind profiler radar to monitor the wind field over City A in real time and obtain data such as wind speed and wind direction at different altitude levels. At the same time, distribute multiple monitoring stations within the city to collect PM2.5 concentration monitoring data and collect meteorological data in the area, such as temperature, humidity, air pressure, etc. In addition, obtain the emission inventory of air-related enterprises to clarify information such as the location of each enterprise, the types of pollutants emitted, and the emission volume.

[0047] Clean the collected data to remove outliers and incorrect data. Then perform standardization processing to unify data of different magnitudes to the same scale for subsequent analysis.

[0048] Input the preprocessed atmospheric component data into a pre-built atmospheric pollutant feature detection model based on BP neural network and LSTM neural network constructed with the TensorFlow framework. Through deep learning algorithms, the model analyzes the input data, combines data of different pollutant components, and establishes the correlation between PM2.5 features and industry emission information. For example, if it is found that the content of a certain specific chemical component in PM2.5 in a certain area is relatively high, by comparing with the emission inventory of air-related enterprises, it is preliminarily judged to be related to the emissions of a certain chemical enterprise. The model outputs the prediction results of the migration path of PM2.5 and the concentration distribution characteristics. At the same time, using the embedded thermal analysis method, preliminarily estimate the distribution trend of PM2.5 concentration in City A and its possible pollution source areas, and obtain a preliminary PM2.5 concentration distribution trend map, which shows that the concentration is relatively high in the southeastern area of the city and there may be pollution sources.

[0049] According to the above results, further confirm the distribution range of PM2.5 and select the locations of multiple sampling nodes in the area with relatively high concentration in the southeastern part of the city. Install sensor devices and monitoring instruments at the selected nodes to collect real-time concentration data of PM2.5 at different nodes. These data include the concentration level of PM2.5 and its temporal variation law. Cross-validate the collected monitoring data with the industry data in the regional emission inventory to check whether the previously established correlation between pollutant characteristics and industry emission information is accurate. Update the thermal analysis map according to the verification results to make the thermal analysis results more accurate.

[0050] Integrate the PM2.5 index data in the thermal analysis map with the real-time wind profiler radar data to generate a real-time wind field map of City A and mark the spatial distribution of PM2.5 in the wind field. By projecting the PM2.5 concentration data in the thermal analysis map into the wind field model and combining the output of the global wind field simulation, construct the migration and diffusion path and spatio-temporal distribution law of PM2.5 in City A. It is found that under the action of the southeast wind, PM2.5 diffuses from the industrial area in the southeastern part of the city to the northwest direction, and the concentration increases near the residential area.

[0051] Quantitative analysis: Using a quantitative evaluation model for industry contributions based on PSCF + linear regression in combination with thermal analysis technology, quantitative analysis is performed on the PM2.5 distribution data and the industry contribution data in the regional emission inventory. An industry contribution heat map is generated to intuitively display the PM2.5 diffusion path, pollution source categories, and the quantitative results of industry contributions. The heat map shows that a certain chemical enterprise and several small factories in the vicinity made relatively large contributions to this PM2.5 pollution incident, and these areas are darker in color on the heat map.

[0052] Based on the industry contribution heat map and the real-time wind field map, a pollution source tracing path model for City A is constructed. During the operation of the model, the emission inventory data within the scope of City A is queried, and the characteristics of PM2.5 are matched with the emission data of relevant industries. The potential pollution source areas are locked as the chemical industrial park in the southeast of the city and the concentrated area of small factories in the vicinity. The diffusion path of PM2.5 from the emission source to the target area is simulated through real-time wind field data. At the same time, combined with the machine learning prediction model, the tracing path is optimized and the positioning results of the pollution sources are output. Finally, several major polluting enterprises are determined, and through thermal analysis, the contribution ratio of these enterprises in the pollution incident is quantitatively evaluated, and it is obtained that the contribution of the chemical enterprise is about 60%, the contribution of the surrounding small factories is about 30%, and the contribution of other factors is about 10%.

[0053] According to the second aspect of the embodiments of the present invention, there is provided a pollution source quantitative tracing device, including: A main controller, and a memory connected to the main controller; The memory, in which program instructions are stored; The main controller is used to execute the program instructions stored in the memory to execute the method described in any one of the above.

[0054] According to the third aspect of the embodiments of the present invention, there is provided a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method described in any one of the above is implemented.

[0055] It can be understood that the same or similar parts in the above embodiments can be referred to each other, and the content not detailed in some embodiments can be seen in the same or similar content in other embodiments.

[0056] It should be noted that in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0057] Any process or method description depicted in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code that includes one or more executable instructions for implementing a specific logical function or process. The scope of the preferred embodiments of the present invention includes additional implementations where functions may be performed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0058] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0059] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried out in implementing the above-described embodiment methods can be completed by instructing relevant hardware through a program. The said program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0060] Furthermore, in each embodiment of the present invention, the functional units can be integrated into one processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0061] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, or the like.

[0062] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0063] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for quantitatively tracing pollution sources, characterized in that: include: Obtain atmospheric composition data, emission inventories of air-related enterprises and global wind field simulation data in the target detection area; atmospheric composition data include wind profile radar data, pollutant concentration monitoring data and related meteorological data; Use the pre-built atmospheric pollutant characteristic detection model to analyze the atmospheric component data to obtain preliminary pollutant characteristics and potential pollutant areas; Determine the collection nodes according to the preliminary pollutant characteristics and potential pollutant areas, dynamically monitor the pollutant concentrations at the collection nodes, and obtain dynamic monitoring data; the dynamic monitoring data includes the concentration level and time variation pattern of the pollutants; The atmospheric pollutant characteristic detection model is used to cross-validate the dynamic monitoring data with the industry data in the emission inventory of air-related enterprises to obtain the pollutant thermodynamic analysis diagram; Integrate the pollutant thermodynamic analysis diagram and wind profile radar data into a real-time wind field map of the target detection area, and annotate the spatial distribution of the target pollutants in the wind field in the real-time wind field map; construct the migration and diffusion paths and temporal and spatial distribution patterns of pollutants in the target area in the real-time wind field map based on global wind field simulation data; Quantitatively analyze the pollutant thermodynamic analysis diagram and the industry contribution data in the emission inventory of air-related enterprises to generate an industry contribution thermodynamic diagram; Quantitatively trace the pollution sources based on the industry contribution heat map and real-time wind field map.

2. The pollution source quantitative tracing method according to claim 1 is characterized in that: The atmospheric pollutant characteristic detection model includes: a BP neural network and an LSTM neural network constructed using a deep learning algorithm framework, and a thermal analysis method is embedded in the atmospheric pollutant characteristic detection model.

3. The pollution source quantitative tracing method according to claim 2 is characterized in that: The pre-built atmospheric pollutant characteristic detection model is used to analyze the atmospheric component data to obtain preliminary pollutant characteristics and potential pollutant areas, including: BP neural network and LSTM neural network are used to analyze atmospheric component data to obtain pollutant migration paths and concentration distribution characteristics; The atmospheric component data are analyzed using the thermal analysis method to obtain the pollutant concentration distribution trend diagram; Based on the pollutant migration path, concentration distribution characteristics and pollutant concentration distribution trend chart, preliminary pollutant characteristics and potential pollutant areas are obtained.

4. The pollution source quantitative tracing method according to claim 1 is characterized in that: Based on the industry contribution heat map and real-time wind field map, the pollution source is quantitatively traced, including: Construct a pollution source tracing path model for the target detection area based on the industry contribution heat map and real-time wind field map; Run the pollution source tracing path model, query the emission data within the target detection area, use the pollution source tracing path model to match the pollutant characteristics with the emission data, and obtain the pollution source area.

5. The pollution source quantitative tracing method according to claim 4 is characterized in that: Also includes: Simulate the diffusion path of pollutants from the pollution source area to the target detection area based on the real-time wind field map; The diffusion pathway is optimized using a machine learning predictive model.

6. The method for quantitatively tracing pollution sources according to claim 5 is characterized in that: Also includes: The industry contribution to pollutants is quantitatively evaluated based on the industry contribution heat map, and the contribution proportion of specific industries in pollution incidents is obtained.

7. A pollution source quantitative tracing device, characterized in that: include: A main controller, and a memory connected to the main controller; The memory stores program instructions; The main controller is used to execute program instructions stored in the memory and execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images