A method, device and storage medium for predicting regional pollutant concentration

Abstract

The present invention relates to a method, device and storage medium for predicting regional pollutant concentrations, which are applied in the field of pollution monitoring technology. Specifically, the method comprises: obtaining historical pollutant concentration monitoring data and meteorological data of each single site, and predicting the future fine particulate matter concentration and chemical composition of pollutants at a single site based on the EnvMulti‑KAN model that integrates the KAN and Transformer architectures; further, based on the prediction results of each single site in the target area, a multi-site fine particulate matter diffusion model is used in combination with geographic information and the interaction between each single site to simulate particle diffusion and convergence, construct regional pollutant concentrations, accurately predict regional fine particulate matter concentrations, and break the limitations of the single-site field of view; this application accurately predicts regional fine particulate matter concentrations and simultaneously accurately predicts the main components of pollutants by coordinating the single-site model with the multi-site model, providing empirical evidence for theoretical research on atmospheric chemistry.

Description

Technical Field

[0001] The present invention belongs to the technical field of pollution monitoring, and in particular relates to a method, device and storage medium for predicting regional pollutant concentration. Background Art

[0002] At present, fine particulate matter from atmospheric pollution (such as PM2.5) has caused serious harm to the environment and human health, and has attracted widespread attention. In response to fine particulate matter pollution in the atmosphere, relevant monitoring technologies have made certain progress. For example, high-precision particulate matter mass concentration monitoring instruments are constantly emerging, and particulate matter size distribution measurement equipment based on principles such as laser scattering is also becoming increasingly popular. However, the component analysis of fine particulate matter from atmospheric pollution faces many challenges. Existing analysis methods often require complex sampling processes and laboratory analysis. For example, after using filter membrane sampling, the organic matter in the particulate matter is analyzed using technologies such as gas chromatography-mass spectrometry (GC-MS). Although this method is accurate, it has a time lag and cannot achieve instant analysis.

[0003] Existing technologies usually focus on using a single sensor technology to infer the composition of fine particulate matter in atmospheric pollution. For example, optical sensors are used to detect the absorption and scattering characteristics of particulate matter to light of a specific wavelength. Based on the known response patterns of different substances to light, a simple discrimination model is established. Taking the detection of heavy metal elements as an example, when light is irradiated on particulate matter containing heavy metals, an absorption peak of a specific wavelength is generated. After the sensor captures the signal, it determines whether the particulate matter contains heavy metals and the approximate content based on preset thresholds and characteristic spectra. However, this method can only identify a limited number of substances with obvious characteristics. For complex organic mixtures and the coexistence of multiple trace elements, it is difficult to accurately distinguish the components due to mutual interference of optical signals, and it is even more impossible to fully reconstruct the overall composition of fine particulate matter. Moreover, it is easily affected by factors such as ambient light noise and particulate matter agglomeration, and has poor stability. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a regional pollutant concentration prediction method, device and storage medium to solve the problem in the prior art that a single sensor technology is used to infer the composition of fine particulate matter in atmospheric pollution, and can only identify a limited number of substances with obvious characteristics. Moreover, due to the mutual interference of optical signals, it is difficult to accurately distinguish the various components, and it is even more impossible to fully reconstruct the overall picture of the composition of fine particulate matter. Moreover, it is easily affected by factors such as ambient light noise and particle agglomeration, and has poor stability.

[0005] According to a first aspect of an embodiment of the present invention, a method for predicting regional pollutant concentration is provided, the method comprising:

[0006] Acquiring historical pollutant concentration monitoring data and meteorological data for each single site within the target area for a preset time period, wherein the pollutant concentration monitoring data includes the concentration of each pollutant and the chemical composition of each pollutant;

[0007] The pollutant concentration monitoring data and meteorological data of each single site for a preset time period are input into the pre-trained EnvMulti-KAN model, and the EnvMulti-KAN model is used to predict the fine particulate matter concentration and fine particulate matter composition of the single site in the future.

[0008] Obtain geographic feature data and real-time meteorological data for each single site in the target area;

[0009] The fine particulate matter concentrations and composition of fine particulate matter, geographic characteristics data, and real-time meteorological data for all single sites over a period of time in the future are input into the pre-trained multi-site fine particulate matter diffusion model;

[0010] The multi-site fine particulate matter diffusion model outputs the predicted value of the pollutant concentration in the target area, as well as the contribution of meteorological factors, geographical factors and each single site to the concentration of each pollutant in the target area.

[0011] Preferably,

[0012] The EnvMulti-KAN model includes: a multi-scale feature fusion module, a dynamic environment interaction module and a multi-task prediction decoder;

[0013] The prediction of the fine particulate matter concentration and fine particulate matter composition of a single site in the future period of time by the EnvMulti-KAN model includes:

[0014] The multi-scale feature fusion module uses a fine-grained spline function to obtain the nonlinear relationship between meteorological factors and pollutant concentrations in the input data. The multi-scale feature fusion module also uses a coarse-grained spline function to obtain the long-term trend of meteorological factors and pollutant concentrations;

[0015] The multi-scale feature fusion module is further used to fuse the nonlinear relationship between the meteorological factors and the pollutant concentration and the long-term trend to obtain multi-scale spatiotemporal coding features;

[0016] The dynamic environment interaction module constructs a component association diagram of each pollutant component according to the multi-scale spatiotemporal coding feature, and the dynamic environment interaction module also generates a pollution component modulation coefficient according to the multi-scale spatiotemporal coding feature;

[0017] The multi-task prediction decoder outputs the fine particulate matter concentration and fine particulate matter composition of a single site within a future period of time according to the component association diagram of each pollutant component and the pollution component modulation coefficient.

[0018] Preferably,

[0019] The multi-site fine particulate matter diffusion model includes: a graph construction module, a spatiotemporal cross memory module, a physical constraint module, and a multi-task decoding module;

[0020] The pollutant concentration prediction value of the target area output by the multi-site fine particulate matter diffusion model, as well as the contribution ratio of meteorological factors, geographical factors, and each single site to the concentration of each pollutant in the target area include:

[0021] The graph construction module constructs a regional feature graph composed of multiple sites based on various dimensional features in the input data, wherein the various dimensional features in the input data include fine particulate matter concentration, fine particulate matter composition, geographical features, and real-time weather;

[0022] The regional characteristic map composed of multiple sites and the pollutant concentration monitoring data of all single sites in the historical preset time period are input into the spatiotemporal cross memory module. The spatiotemporal cross memory module corrects the regional characteristic map composed of multiple sites based on the pollutant concentration monitoring data in the historical preset time period. At the same time, the spatiotemporal dependency relationship between the pollutant concentration monitoring data in the historical preset time period and the regional characteristic map composed of multiple sites is obtained through the spatiotemporal memory unit to obtain the node features of the corrected regional characteristic map;

[0023] Inputting the modified regional feature map node features into a physical constraint module, wherein the physical constraint module is preset with a plurality of physical laws, performing physical law constraints on the input modified regional feature map node features, and outputting node features that satisfy the physical laws;

[0024] The node features that satisfy the physical laws are input into the multi-task decoding, and the predicted value of the pollutant concentration in the target area is output, as well as the contribution ratio of meteorological factors, geographical factors and each single site to the concentration of each pollutant in the target area.

[0025] Preferably, it also includes:

[0026] The pollutant concentration monitoring data and meteorological data of the historical preset time period of each single site are input into the pre-trained EnvMulti-KAN model as input data, and interference factors are identified. The abnormal data in the pollutant concentration monitoring data and meteorological data are removed according to the interference factor identification results.

[0027] Preferably,

[0028] The interference factor identification includes: abnormal meteorological disaster identification and temporary large-scale pollution source intrusion identification;

[0029] The abnormal meteorological disaster identification includes: judging whether any abnormal meteorological disaster has occurred according to the meteorological data of the historical preset time period based on a plurality of preset abnormal meteorological disaster judgment conditions;

[0030] The temporary large-scale pollution source intrusion identification includes: presetting temporary large-scale pollution source intrusion determination conditions, and judging whether a temporary large-scale pollution source intrusion occurs based on the pollutant concentration monitoring data of the historical preset time period;

[0031] The removing of abnormal data from the pollutant concentration monitoring data and meteorological data according to the interference factor identification result includes:

[0032] If an abnormal meteorological disaster is identified, the pollutant concentration monitoring data for the historical preset time period and the pollutant concentration monitoring data for the period when the abnormal meteorological disaster occurred and the meteorological data are deleted;

[0033] Obtain pollutant concentration monitoring data and meteorological data for the same historical period as the abnormal meteorological disaster occurred, but no abnormal meteorological disaster occurred, and fill in the deleted pollutant concentration monitoring data and meteorological data;

[0034] If a temporary large-scale pollution source intrusion is identified, the pollutant concentration monitoring data of the historical preset time period during the period of the temporary large-scale pollution source intrusion is deleted;

[0035] Obtain the pollutant concentration monitoring data of the historical same period when temporary large-scale pollution sources intrusion occurred and when no temporary large-scale pollution sources intruded, and fill in the deleted pollutant concentration monitoring data.

[0036] Preferably, it also includes:

[0037] Preprocessing the output of the EnvMulti-KAN model and the output of the multi-site fine particle diffusion model respectively so that the time step and spatial resolution of the output of the EnvMulti-KAN model are consistent with those of the multi-site fine particle diffusion model;

[0038] The output of the EnvMulti-KAN model with consistent spatiotemporal dimensions and the output of the multi-site fine particulate matter dispersion model are integrated to construct spatiotemporal four-dimensional data on the characteristics of time, space, pollutant concentration, and chemical composition of pollutants.

[0039] Based on the spatiotemporal four-dimensional data, a regional fine particulate matter component distribution map is generated using a GIS platform. The regional fine particulate matter component distribution map is used to dynamically present the concentration changes and spatial distribution of pollutant components at each site;

[0040] generating a concentration heat map based on the regional fine particulate matter component distribution map, and obtaining a diffusion path of pollutants based on the concentration heat map;

[0041] Based on the concentration heat map, the transmission path of the pollutants is obtained by combining the wind rose diagram and the geographical feature map in the meteorological data;

[0042] The output of the EnvMulti-KAN model and the output of the multi-site fine particulate matter diffusion model are decomposed using positive definite matrix factorization to obtain the contribution ratios of different pollution source types, and the pollution source with the greatest influence is obtained based on the contribution ratios of different pollution source types.

[0043] According to a second aspect of an embodiment of the present invention, a regional pollutant concentration prediction device is provided, the device comprising:

[0044] Historical data acquisition module: used to obtain historical pollutant concentration monitoring data and meteorological data for each single site in the target area during a preset time period. The pollutant concentration monitoring data includes the concentration of each pollutant and the chemical composition of each pollutant;

[0045] Single-site prediction module: This module is used to input the pollutant concentration monitoring data and meteorological data of each single site for a preset time period into the pre-trained EnvMulti-KAN model, and use the EnvMulti-KAN model to predict the fine particulate matter concentration and fine particulate matter composition of the single site in the future.

[0046] Real-time data acquisition module: used to obtain geographic feature data and real-time meteorological data of each single site in the target area;

[0047] Multi-site input module: used to input the fine particulate matter concentration and fine particulate matter composition, geographical feature data and real-time meteorological data of all single sites over a period of time into the pre-trained multi-site fine particulate matter diffusion model;

[0048] Regional prediction module: used to output the predicted value of pollutant concentration in the target area through the multi-site fine particulate matter diffusion model, as well as the contribution of meteorological factors, geographical factors and each single site to the concentration of each pollutant in the target area.

[0049] According to a third aspect of an embodiment of the present invention, a storage medium is provided, wherein the storage medium stores a computer program, and when the computer program is executed by a host controller, each step in the above method is implemented.

[0050] The technical solutions provided by the embodiments of the present invention may have the following beneficial effects:

[0051] This application obtains historical pollutant concentration monitoring data and meteorological data from each single site, and predicts the future fine particulate matter concentration and pollutant chemical composition of the single site based on the EnvMulti-KAN model that integrates the KAN and Transformer architectures; further, based on the prediction results of each single site in the target area, a multi-site fine particulate matter diffusion model is used in combination with geographic information and the interaction between each single site to simulate particle diffusion and convergence, construct regional pollutant concentrations, and accurately predict regional fine particulate matter concentrations, breaking the limitations of the single-site field of view; this application accurately predicts regional fine particulate matter concentrations and simultaneously accurately predicts the main components of pollutants by coordinating the single-site model with the multi-site model, providing empirical evidence for theoretical research on atmospheric chemistry.

[0052] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0054] Figure 1 is a flow chart of a method for predicting regional pollutant concentration according to an exemplary embodiment;

[0055] Figure 2 is a system schematic diagram of a regional pollutant concentration prediction device according to another exemplary embodiment;

[0056] In the attached figure: 1-historical data acquisition module, 2-single site prediction module, 3-real-time data acquisition module, 4-multi-site input module, 5-regional prediction module. DETAILED DESCRIPTION

[0057] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0058] Example 1

[0059] Figure 1 FIG. 1 is a flow chart of a method for predicting regional pollutant concentration according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0060] S1, obtaining historical pollutant concentration monitoring data and meteorological data for each single site in the target area for a preset time period, wherein the pollutant concentration monitoring data includes the concentration of each pollutant and the chemical composition of each pollutant;

[0061] S2, inputting the pollutant concentration monitoring data and meteorological data of each single site for a preset time period into the pre-trained EnvMulti-KAN model, and using the EnvMulti-KAN model to predict the fine particulate matter concentration and fine particulate matter composition of the single site for a future period of time;

[0062] S3, obtains geographic feature data and real-time meteorological data of each single site in the target area;

[0063] S4: Input the fine particulate matter concentrations and fine particulate matter components, geographic feature data, and real-time meteorological data of all single sites over a period of time into the pre-trained multi-site fine particulate matter diffusion model;

[0064] S5, outputting the predicted pollutant concentration value of the target area through the multi-site fine particulate matter diffusion model, as well as the contribution ratio of meteorological factors, geographical factors, and each single site to the concentration of each pollutant in the target area;

[0065] It is understood that this application obtains high-resolution data of the past three years from each single site in the target area, including pollutant concentration monitoring data, chemical composition of pollutants, and meteorological data, and inputs the obtained historical data into the EnvMulti-KAN model. The EnvMulti-KAN model is based on the Kolmogorov-Arnold The KAN and Transformer architecture includes a multi-scale feature fusion module, a dynamic environment interaction module, and a multi-task prediction decoder. The multi-scale feature fusion module uses a fine-grained spline function to capture the nonlinear relationship between meteorological factors and pollutant concentrations in the input data, and a coarse-grained spline function to obtain the long-term trend of meteorological factors and pollutant concentrations. The two are fused to obtain multi-scale spatiotemporal coding features (hourly / daily / weekly embedding). The multi-scale spatiotemporal coding features are input into the dynamic environment interaction module, which is used to input meteorological variables (temperature, humidity, wind speed) and pollutant concentrations into the dynamic KAN branch to generate nonlinear modulation coefficients of meteorological conditions on pollution components and construct correlation maps of various pollution components. The nonlinear modulation coefficients and correlation maps of various pollution components are input into the multi-task prediction decoder to output the fine particulate matter concentration and fine particulate matter composition of a single site for a period of time in the future.

[0066] The target area is usually composed of various single-site areas. The prediction results of each single site have been obtained through the above-mentioned EnvMulti-KAN model. Next, the geographic data and real-time meteorological data of each single site are obtained, and the prediction results of each single site are combined and input into the multi-site fine particulate matter diffusion model. The multi-site fine particulate matter diffusion model includes: a graph construction module, a spatiotemporal cross-memory module, a physical constraint module and a multi-task decoding module; the graph construction module constructs a regional feature map composed of multiple sites according to the various dimensional features in the input data, and the various dimensional features include fine particulate matter concentration, fine particulate matter composition, geographic features and real-time meteorology; the input of the spatiotemporal cross-memory module consists of two parts, one is the regional feature map output by the graph construction module, and the other is The spatiotemporal cross-memory module uses the pollutant concentration monitoring data of a single site for a preset historical time period to correct the regional characteristic map composed of multiple sites, and at the same time obtains the spatiotemporal dependency between the two through the spatiotemporal memory unit to obtain the node features of the corrected regional characteristic map; the node features of the corrected regional characteristic map are input into the physical constraint module, which has a variety of physical laws preset in it, such as Newton's three laws, the law of conservation of momentum, etc., and outputs node features that satisfy the physical laws; the node features that satisfy the physical laws are input into the multi-task decoding, and the predicted pollutant concentration value of the target area is output, as well as the contribution ratio of each factor (output meteorological factors, geographical factors, and each single site) to the concentration of each pollutant in the target area;

[0067] To further improve the accuracy of prediction, this embodiment also discloses a method for removing external factors, namely, identifying abnormal meteorological disasters and temporary large-scale pollution sources, including:

[0068] Definition of types and determination conditions of abnormal meteorological disasters:

[0069] (1) Sandstorm: wind speed ≥15m / s, visibility ≤1km, PM10 concentration (instantaneous ≥500μg / m 3 );

[0070] (2) Extreme precipitation: 24-hour precipitation ≥ 100 mm (refer to the red warning standard for heavy rain issued by the Meteorological Bureau);

[0071] (3) Continuous high temperature inversion: the inversion layer height is ≤ 200 m and the daily maximum temperature is ≥ 35°C for three consecutive days;

[0072] (4) Calm weather: wind speed ≤ 1 m / s for more than 48 hours, boundary layer height ≤ 500 m;

[0073] Time window verification: The abnormal state must last for at least 2 hours (to avoid misjudgment due to instantaneous interference);

[0074] According to the historical meteorological data, it is judged whether any period meets any of the above abnormal meteorological disasters. Since abnormal meteorological disasters will affect the concentration of pollutants, after judging the occurrence of any abnormal meteorological disaster, it is necessary to delete the pollutant monitoring data and meteorological data of the abnormal meteorological disaster period in the historical input data at the same time, and use the pollutant monitoring data and meteorological data of the historical period of the abnormal meteorological disaster period when no abnormal meteorological disaster occurred to fill in the data. For example: in the historical data of 2016, extreme precipitation occurred from September 15th to September 16th, then the meteorological data and pollutant monitoring data from September 15th to September 16th in the input data will be deleted. The measured data is deleted, and then the meteorological data and pollutant monitoring data from September 15 to September 16, 2015 are obtained. If no abnormal meteorological disaster occurs from September 15 to September 16, 2015, the pollutant monitoring data and meteorological data from September 15 to September 16, 2015 are used to fill the deleted pollutant monitoring data and meteorological data from September 15 to September 16 of 2016. If an abnormal meteorological disaster also occurs from September 15 to September 16, 2015, the pollutant monitoring data and meteorological data from September 15 to September 16, 2014 are continued to be obtained;

[0075] Definition of temporary large pollution source intrusion and judgment conditions:

[0076] (1) Industrial accident: SO or NOx concentration increases by more than 300% within 1 hour, and there is a factory coordinate matching in the adjacent area;

[0077] (2) Traffic emergencies: a sharp increase in CO concentration (hourly average ≥ 5 ppm) accompanied by abnormal traffic monitoring data;

[0078] (3) Open-air burning: The proportion of potassium ions (K+) in PM2.5 increases suddenly (>10%) and matches the satellite fire monitoring;

[0079] The spatiotemporal causal relationship between pollution sources and monitoring data is confirmed through spatiotemporal correlation analysis. First, through time matching, when the monitoring station detects an abnormal increase in pollutant concentration, the time axis is traced back to find possible pollution events. Through spatial matching, the pollution source coordinates and the monitoring station location are combined to calculate the pollution diffusion path. For example, if a chemical plant leaks and a station within 5 kilometers downwind detects an increase in VOCs concentration after 30 minutes, it is determined to be a related event. Through diffusion model verification, the Gaussian diffusion model or Lagrangian particle model is used to simulate the pollutant transmission path to verify whether the concentration peak conforms to the expected distribution. In terms of technical implementation, a dynamic time window (such as ±1 hour) is set to match pollution events and concentration changes, and the potential impact area of ​​the pollution source is delineated based on meteorological data (wind direction, wind speed), and only the monitoring stations within the area are associated; data source It includes emission inventories, traffic monitoring data, satellite remote sensing fire point data and meteorological data. Meteorological data verifies the conditions for pollutant diffusion. For example, local emissions accumulate in calm weather. If PM2.5 at a certain station increases but there is no external transmission, it is attributed to the local source. In terms of technical implementation, Kriging interpolation is used to align the satellite fire point and traffic data with the monitoring station locations, and the Apriori algorithm is used to mine frequent item sets of multi-source data. For example, if "traffic congestion + sudden increase in CO2" occur at the same time, it is considered that a temporary large-scale pollution source has intruded due to a traffic emergency. Because the sudden increase in CO2 concentration at this time has a strong interference with the prediction results, it is necessary to delete the CO2 concentration monitoring data during the period when the temporary large-scale pollution source intrudes from the model input data and then replace it with the CO2 concentration monitoring data of the same historical period. The replacement method is the same as that of abnormal meteorological disaster data.

[0080] This embodiment also provides a post-processing and feedback mechanism, which stores the time, type, affected site, and treatment measures in the interference event record library for iterative model optimization and supports environmental management personnel to manually confirm or correct the automatic judgment results;

[0081] This embodiment also provides further analysis applications based on single-site output results and multi-site output results, including:

[0082] Data alignment technology is used to unify the time steps (e.g., hourly forecasts) and spatial resolutions (e.g., 1km grids) of the outputs of single-site and multi-site models, ensuring consistency in the spatiotemporal dimensions of meteorological conditions (e.g., wind direction, terrain barriers) and pollutant diffusion pathways, thus avoiding prediction biases caused by differences in data scales. The outputs of the single and multi-site models are then deeply integrated to construct spatiotemporal four-dimensional data encompassing time, space, pollutant concentrations, and characteristics of more than 20 components. Based on this integrated spatiotemporal four-dimensional data, a GIS platform is used to generate regional fine particulate matter component distribution maps, dynamically presenting the concentration changes and spatial distribution of pollutant components (e.g., nitrates, sulfates, and organic matter) at each site.

[0083] Generate a heat map from the regional fine particulate matter component distribution map, mark high-concentration areas with the heat map, and simulate the diffusion path of pollutants in combination with the streamline map;

[0084] Based on the heat map, combined with the wind rose diagram in the meteorological data and the geographic data map, the dominant wind field is analyzed, and the diffusion barriers (such as mountains and building complexes) are identified using terrain elevation data. The transmission path of pollutants is then traced back. For example, if the peak nitrate concentration in a certain area is strongly correlated with the northwest wind direction and there is an industrial park upwind, it can be determined that the industrial park is the main contributing source. Based on the reconstructed component data, positive definite matrix factorization (PMF) is used to obtain the contribution ratio of different pollution source types. Based on the contribution ratio of different pollution source types, the pollution source with the greatest influence is determined.

[0085] After obtaining single-point and multi-point prediction results, this embodiment integrates the fine particulate matter component data predicted by single sites and multiple sites. Based on the concentrations of the main components of fine particulate matter predicted by single sites and the regional fine particulate matter concentration distribution predicted by multiple sites, the spatial distribution of fine particulate matter components is established in combination with geographic datasets and meteorological condition data. Advanced data analysis algorithms are used to analyze the concentration differences of the predicted fine particulate matter components between different sites and regions. The influence of factors such as atmospheric flow and topography on the transmission and distribution of different components is considered, and the evolution law of fine particulate matter components during the transmission process is reversed. This provides comprehensive, timely, and high-precision data support for air pollution control from source prevention and control to process monitoring and effect evaluation.

[0086] Example 2:

[0087] Figure 2 1 is a system schematic diagram of a regional pollutant concentration prediction device according to another exemplary embodiment, the device comprising:

[0088] Historical data acquisition module 1: used to obtain historical pollutant concentration monitoring data and meteorological data for each single site in the target area during a preset time period, wherein the pollutant concentration monitoring data includes the concentration of each pollutant and the chemical composition of each pollutant;

[0089] Single-site prediction module 2: used to input the historical pollutant concentration monitoring data and meteorological data of each single site for a preset time period as input data into the pre-trained EnvMulti-KAN model, and use the EnvMulti-KAN model to predict the fine particulate matter concentration and fine particulate matter composition of the single site in the future;

[0090] Real-time data acquisition module 3: used to obtain geographic feature data and real-time meteorological data of each single site in the target area;

[0091] Multi-site input module 4: used to input the fine particulate matter concentration and fine particulate matter composition, geographical feature data and real-time meteorological data of all single sites in the future period into the pre-trained multi-site fine particulate matter diffusion model;

[0092] Regional prediction module 5: used to output the predicted value of pollutant concentration in the target area through the multi-site fine particulate matter diffusion model, as well as the contribution ratio of meteorological factors, geographical factors and each single site to the concentration of each pollutant in the target area.

[0093] Example 3:

[0094] This embodiment provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a host controller, each step in the above method is implemented;

[0095] It is understandable that the storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0096] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0097] It should be noted that, in the description of the present invention, the terms "first," "second," etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, "a small number of sparsely distributed" means at least two.

[0098] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or less sparsely distributed executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0099] 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-described embodiment, a small number of sparsely distributed 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 of the following technologies known in the art or a combination thereof can be used to implement it: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0100] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the 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.

[0101] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0102] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0103] Throughout this specification, references to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or a small number of sparsely distributed embodiments or examples.

[0104] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for predicting regional pollutant concentration, characterized in that: The method comprises: Acquiring historical pollutant concentration monitoring data and meteorological data for each single site within the target area for a preset time period, wherein the pollutant concentration monitoring data includes the concentration of each pollutant and the chemical composition of each pollutant; The pollutant concentration monitoring data and meteorological data of each single site for a preset time period are input into the pre-trained EnvMulti-KAN model, and the EnvMulti-KAN model is used to predict the fine particulate matter concentration and fine particulate matter composition of the single site in the future. The EnvMulti-KAN model includes: a multi-scale feature fusion module, a dynamic environment interaction module and a multi-task prediction decoder; The prediction of the fine particulate matter concentration and fine particulate matter composition of a single site in the future period of time by the EnvMulti-KAN model includes: The multi-scale feature fusion module uses a fine-grained spline function to obtain the nonlinear relationship between meteorological factors and pollutant concentrations in the input data. The multi-scale feature fusion module also uses a coarse-grained spline function to obtain the long-term trend of meteorological factors and pollutant concentrations; Obtain geographic feature data and real-time meteorological data for each single site in the target area; The fine particulate matter concentrations and composition of fine particulate matter, geographic characteristics data, and real-time meteorological data for all single sites over a period of time in the future are input into the pre-trained multi-site fine particulate matter diffusion model; Outputting the predicted values ​​of pollutant concentrations in the target area through the multi-site fine particulate matter diffusion model, as well as the contribution ratios of meteorological factors, geographical factors, and each single site to the concentrations of each pollutant in the target area; Preprocessing the output of the EnvMulti-KAN model and the output of the multi-site fine particle diffusion model respectively so that the time step and spatial resolution of the output of the EnvMulti-KAN model are consistent with those of the multi-site fine particle diffusion model; The output of the EnvMulti-KAN model with consistent spatiotemporal dimensions and the output of the multi-site fine particulate matter dispersion model are integrated to construct spatiotemporal four-dimensional data on the characteristics of time, space, pollutant concentration, and chemical composition of pollutants. Based on the spatiotemporal four-dimensional data, a regional fine particulate matter component distribution map is generated using a GIS platform. The regional fine particulate matter component distribution map is used to dynamically present the concentration changes and spatial distribution of pollutant components at each site; generating a concentration heat map based on the regional fine particulate matter component distribution map, and obtaining a diffusion path of pollutants based on the concentration heat map; Based on the concentration heat map, the transmission path of the pollutants is obtained by combining the wind rose diagram and the geographical feature map in the meteorological data; The output of the EnvMulti-KAN model and the output of the multi-site fine particulate matter diffusion model are decomposed using positive definite matrix factorization to obtain the contribution ratios of different pollution source types, and the pollution source with the greatest influence is obtained based on the contribution ratios of different pollution source types.

2. The method according to claim 1, characterized in that The multi-scale feature fusion module is further used to fuse the nonlinear relationship between the meteorological factors and the pollutant concentration and the long-term trend to obtain multi-scale spatiotemporal coding features; The dynamic environment interaction module constructs a component association diagram of each pollutant component according to the multi-scale spatiotemporal coding feature, and the dynamic environment interaction module also generates a pollution component modulation coefficient according to the multi-scale spatiotemporal coding feature; The multi-task prediction decoder outputs the fine particulate matter concentration and fine particulate matter composition of a single site within a future period of time according to the component association diagram of each pollutant component and the pollution component modulation coefficient.

3. The method according to claim 2, characterized in that The multi-site fine particulate matter diffusion model includes: a graph construction module, a spatiotemporal cross memory module, a physical constraint module, and a multi-task decoding module; The pollutant concentration prediction value of the target area output by the multi-site fine particulate matter diffusion model, as well as the contribution ratio of meteorological factors, geographical factors, and each single site to the concentration of each pollutant in the target area include: The graph construction module constructs a regional feature graph composed of multiple sites based on various dimensional features in the input data, wherein the various dimensional features in the input data include fine particulate matter concentration, fine particulate matter composition, geographical features, and real-time weather; The regional characteristic map composed of multiple sites and the pollutant concentration monitoring data of all single sites in the historical preset time period are input into the spatiotemporal cross memory module. The spatiotemporal cross memory module corrects the regional characteristic map composed of multiple sites based on the pollutant concentration monitoring data in the historical preset time period. At the same time, the spatiotemporal dependency relationship between the pollutant concentration monitoring data in the historical preset time period and the regional characteristic map composed of multiple sites is obtained through the spatiotemporal memory unit to obtain the node features of the corrected regional characteristic map; Inputting the modified regional feature map node features into a physical constraint module, wherein the physical constraint module is preset with a plurality of physical laws, performing physical law constraints on the input modified regional feature map node features, and outputting node features that satisfy the physical laws; The node features that satisfy the physical laws are input into the multi-task decoding, and the predicted value of the pollutant concentration in the target area is output, as well as the contribution ratio of meteorological factors, geographical factors and each single site to the concentration of each pollutant in the target area.

4. The method according to claim 3, characterized in that Also includes: The pollutant concentration monitoring data and meteorological data of the historical preset time period of each single site are input into the pre-trained EnvMulti-KAN model as input data, and interference factors are identified. The abnormal data in the pollutant concentration monitoring data and meteorological data are removed according to the interference factor identification results.

5. The method according to claim 4, characterized in that The interference factor identification includes: abnormal meteorological disaster identification and temporary large-scale pollution source intrusion identification; The abnormal meteorological disaster identification includes: judging whether any abnormal meteorological disaster has occurred according to the meteorological data of the historical preset time period based on a plurality of preset abnormal meteorological disaster judgment conditions; The temporary large-scale pollution source intrusion identification includes: presetting temporary large-scale pollution source intrusion determination conditions, and judging whether a temporary large-scale pollution source intrusion occurs based on the pollutant concentration monitoring data of the historical preset time period; The removing of abnormal data from the pollutant concentration monitoring data and meteorological data according to the interference factor identification result includes: If an abnormal meteorological disaster is identified, the pollutant concentration monitoring data for the historical preset time period and the pollutant concentration monitoring data for the period when the abnormal meteorological disaster occurred and the meteorological data are deleted; Obtain pollutant concentration monitoring data and meteorological data for the same historical period as the abnormal meteorological disaster occurred, but no abnormal meteorological disaster occurred, and fill in the deleted pollutant concentration monitoring data and meteorological data; If a temporary large-scale pollution source intrusion is identified, the pollutant concentration monitoring data of the historical preset time period during the period of the temporary large-scale pollution source intrusion is deleted; Obtain the pollutant concentration monitoring data of the historical same period when temporary large-scale pollution sources intrusion occurred and when no temporary large-scale pollution sources intruded, and fill in the deleted pollutant concentration monitoring data.

6. A regional pollutant concentration prediction device, characterized in that: The device comprises: Historical data acquisition module: used to obtain historical pollutant concentration monitoring data and meteorological data for each single site in the target area during a preset time period. The pollutant concentration monitoring data includes the concentration of each pollutant and the chemical composition of each pollutant; Single-site prediction module: This module is used to input the pollutant concentration monitoring data and meteorological data of each single site for a preset time period into the pre-trained EnvMulti-KAN model, and use the EnvMulti-KAN model to predict the fine particulate matter concentration and fine particulate matter composition of the single site in the future. The EnvMulti-KAN model includes: a multi-scale feature fusion module, a dynamic environment interaction module and a multi-task prediction decoder; The prediction of the fine particulate matter concentration and fine particulate matter composition of a single site in the future period of time by the EnvMulti-KAN model includes: The multi-scale feature fusion module uses a fine-grained spline function to obtain the nonlinear relationship between meteorological factors and pollutant concentrations in the input data. The multi-scale feature fusion module also uses a coarse-grained spline function to obtain the long-term trend of meteorological factors and pollutant concentrations; Real-time data acquisition module: used to obtain geographic feature data and real-time meteorological data of each single site in the target area; Multi-site input module: used to input the fine particulate matter concentration and fine particulate matter composition, geographical feature data and real-time meteorological data of all single sites over a period of time into the pre-trained multi-site fine particulate matter diffusion model; Regional prediction module: used to output the predicted value of pollutant concentration in the target area through the multi-site fine particulate matter diffusion model, as well as the contribution of meteorological factors, geographical factors and each single site to the concentration of each pollutant in the target area; Preprocessing the output of the EnvMulti-KAN model and the output of the multi-site fine particle diffusion model respectively so that the time step and spatial resolution of the output of the EnvMulti-KAN model are consistent with those of the multi-site fine particle diffusion model; The output of the EnvMulti-KAN model with consistent spatiotemporal dimensions and the output of the multi-site fine particulate matter dispersion model are integrated to construct spatiotemporal four-dimensional data on the characteristics of time, space, pollutant concentration, and chemical composition of pollutants. Based on the spatiotemporal four-dimensional data, a regional fine particulate matter component distribution map is generated using a GIS platform. The regional fine particulate matter component distribution map is used to dynamically present the concentration changes and spatial distribution of pollutant components at each site; generating a concentration heat map based on the regional fine particulate matter component distribution map, and obtaining a diffusion path of pollutants based on the concentration heat map; Based on the concentration heat map, the transmission path of the pollutants is obtained by combining the wind rose diagram and the geographical feature map in the meteorological data; The output of the EnvMulti-KAN model and the output of the multi-site fine particulate matter diffusion model are decomposed using positive definite matrix factorization to obtain the contribution ratios of different pollution source types, and the pollution source with the greatest influence is obtained based on the contribution ratios of different pollution source types.

7. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the main controller, each step of the method for predicting regional pollutant concentration according to any one of claims 1 to 5 is implemented.

Images