A multi-stage optimized space-time tracing method, device and storage medium

Abstract

The present invention relates to a multi-stage optimized spatiotemporal tracing method, device and storage medium, which are applied to the field of pollution monitoring technology. Specifically, it includes: building a closed-loop pollution source contribution optimization and tracing system by combining the AERMOD model with the deep learning model, combining the simulation output of AERMOD with actual monitoring data, meteorological data and geographic spatial information, correcting it through the deep learning model, and further decomposing the optimized total concentration contribution into the contribution proportion of individual pollutants. Multi-feature input integrates multi-dimensional features such as historical contribution weights, meteorological data, and geographic information. The output results of the model are dynamically fed back to the front-layer structure, and by continuously optimizing the pollution source contribution weights, a closed-loop linkage from data input to strategy output is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of pollution monitoring, and specifically relates to a multi-stage optimized spatiotemporal tracing method, device and storage medium. Background Art

[0002] Accurately assessing the contribution of emission sources to pollutant concentrations at monitoring points and analyzing the contribution of pollutant concentrations at different times have become key tasks for precise governance and air quality management. Traditional pollution source contribution identification and concentration prediction methods mainly rely on diffusion models and simple statistical analysis.

[0003] In the existing technology, there is a method based on combining traditional diffusion models with monitoring station data. The technical solution includes the following steps: collecting emission inventories of major pollution sources in the region, including pollutant emission intensity, emission source location, and emission time, as basic data input for pollutant diffusion simulation, using a diffusion model to numerically simulate the spatial diffusion trajectory of pollutants. The model calculates the concentration contribution value of pollutants diffused from the emission source to each monitoring point based on the physical formula of pollutant diffusion. Combined with the actual observed concentration data of pollutants at regional monitoring stations, the simulation results are calibrated and adjusted to reduce the deviation between the simulated concentration and the actual concentration. Through model calculation, the pollutant concentration at the monitoring point is decomposed into the contribution ratio of different emission sources, providing a basis for pollution control;

[0004] Although these methods can provide a theoretical basis for pollutant diffusion and emission source contributions, they still have some shortcomings, such as insufficient spatial feature modeling, limited ability to capture temporal features, and insufficient real-time and accuracy of contribution decomposition. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a multi-stage optimized spatiotemporal tracing method, device and storage medium to solve the problems in the prior art, such as insufficient spatial feature modeling, limited temporal feature capture capability, and insufficient real-time and accuracy of contribution decomposition.

[0006] According to a first aspect of an embodiment of the present invention, a multi-stage optimized spatiotemporal tracing method is provided, the method comprising:

[0007] Obtain meteorological data for a specified historical period within a preset range of the target site, obtain topographic data within a preset range of the target site, and obtain monitoring data for various pollutant monitoring points within a specified historical period within the preset range of the target site;

[0008] Preprocessing the meteorological data, topographic data, and monitoring data of various pollutant monitoring points to obtain first input data;

[0009] Input the first input data into the pre-built AERMOD model to obtain the total concentration contribution of the pollution source at the target site;

[0010] Obtaining real-time monitoring data from various pollutant monitoring points, and obtaining initial weights of pollution source contributions based on the proportion of each pollution source in the real-time monitoring data from various pollutant monitoring points;

[0011] Based on the total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, topographic data, and the initial weight of the pollution source contribution, the pre-built first spatiotemporal deep learning model DCRNN is used to obtain the corrected total pollution source concentration contribution of the target site;

[0012] Based on the total pollution source concentration contribution of the corrected target site, the real-time monitoring data of various pollutant monitoring points, real-time meteorological data, and the initial weight of the pollution source contribution, the pre-built second spatiotemporal deep learning model DCRNN is used to obtain the corrected individual pollutant concentration contribution value;

[0013] Based on the corrected individual pollutant concentration contribution value, real-time monitoring data of various pollutant monitoring points, topographic data, and the total concentration contribution of pollution sources at the target site, the contribution weight of the pollution source is obtained through the pre-built third spatiotemporal deep learning model DCRNN;

[0014] The initial weight of the pollution source contribution is updated by the contribution weight of the pollution source, and the corrected single pollutant concentration contribution value is obtained by the updated contribution weight of the pollution source.

[0015] Preferably,

[0016] The total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, the topographic data, and the initial weight of the pollution source contribution are obtained by using the pre-built first spatiotemporal deep learning model DCRNN to obtain the corrected total pollution source concentration contribution of the target site, including:

[0017] Preprocessing the total concentration contribution of pollution sources at the target site, real-time monitoring data of various pollutant monitoring points, topographic data, and initial weights of pollution source contributions to obtain second input data;

[0018] A first node feature matrix is ​​constructed based on the second input data to obtain graph structure information; the first node feature matrix and graph structure information are input into a pre-built first spatiotemporal deep learning model DCRNN to obtain the corrected total concentration contribution of the pollution source of the target site.

[0019] Preferably,

[0020] The method of obtaining the corrected individual pollutant concentration contribution value based on the corrected total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, the real-time meteorological data, and the initial weight of the pollution source contribution through the pre-built second spatiotemporal deep learning model DCRNN includes:

[0021] Obtaining a corrected pollutant contribution based on the corrected total pollution source concentration contribution of the target site and the initial weight of the pollution source contribution; and obtaining an error evaluation of the first spatiotemporal deep learning model DCRNN based on the corrected total pollution source concentration contribution of the target site and the total pollution source concentration contribution of the target site;

[0022] Acquiring real-time meteorological data within a preset range of the target site, preprocessing the real-time monitoring data of each pollutant monitoring point, the real-time meteorological data, the initial weights of pollution source contributions, the corrected total pollution source concentration contribution of the target site, the corrected pollutant contribution, and the error assessment to obtain third input data;

[0023] A second node feature matrix is ​​constructed based on the third input data, and the second node feature matrix and graph structure information are input into a pre-built second spatiotemporal deep learning model DCRNN to obtain the corrected single pollutant concentration contribution value.

[0024] Preferably,

[0025] The contribution weights of the pollution sources are obtained by using the pre-built third spatiotemporal deep learning model DCRNN based on the corrected individual pollutant concentration contribution values, real-time monitoring data of various pollutant monitoring points, topographic data, and the total pollution source concentration contribution of the target site. The weights include:

[0026] Preprocessing the corrected individual pollutant concentration contribution values, real-time monitoring data of various pollutant monitoring points, topographic data, and the total concentration contribution of pollution sources at the target site to obtain fourth input data;

[0027] A third node feature matrix is ​​constructed based on the fourth input data, and the third node feature matrix and graph structure information are input into a pre-built third spatiotemporal deep learning model DCRNN to obtain the contribution weight of the pollution source.

[0028] Preferably,

[0029] The obtaining of graph structure information includes:

[0030] Graph structure information is obtained based on the spatial distribution of the total concentration contribution of the pollution sources of the target site within a preset range of the target site.

[0031] Preferably,

[0032] The first spatiotemporal deep learning model DCRNN, the second spatiotemporal deep learning model DCRNN, and the third spatiotemporal deep learning model DCRNN all include: a graph convolution module, a recurrent neural network module, and a decoding module;

[0033] The graph convolution module is used to extract the spatial features of the input features, the recurrent neural network module is used to extract the temporal features of the input features, and the decoding module is used to fuse the extracted spatial features and temporal features.

[0034] According to a second aspect of an embodiment of the present invention, a multi-stage optimized spatiotemporal tracing device is provided, the device comprising:

[0035] The first data acquisition module is used to obtain meteorological data for a specified historical period within a preset range of the target site, obtain topographic data within a preset range of the target site, and obtain monitoring data of various pollutant monitoring points within a specified historical period within the preset range of the target site;

[0036] A first input module is used to pre-process the meteorological data, topographic data, and monitoring data of various pollutant monitoring points to obtain first input data;

[0037] AERMOD module: used to input the first input data into the pre-built AERMOD model to obtain the total concentration contribution of the pollution source of the target site;

[0038] The second data acquisition module is used to obtain real-time monitoring data of various pollutant monitoring points, and obtain the initial weight of the pollution source contribution according to the proportion of each pollution source in the real-time monitoring data of various pollutant monitoring points;

[0039] The first deep learning module is used to obtain the corrected total pollution source concentration contribution of the target site through a pre-built first spatiotemporal deep learning model DCRNN based on the total pollution source concentration contribution of the target site, real-time monitoring data of various pollutant monitoring points, topographic data, and initial weights of pollution source contributions;

[0040] The second deep learning module is used to obtain the corrected individual pollutant concentration contribution value through the pre-built second spatiotemporal deep learning model DCRNN based on the total pollution source concentration contribution of the corrected target site, the real-time monitoring data of various pollutant monitoring points, the real-time meteorological data, and the initial weight of the pollution source contribution;

[0041] The third deep learning module is used to obtain the contribution weight of the pollution source through the pre-built third spatiotemporal deep learning model DCRNN based on the corrected single pollutant concentration contribution value, real-time monitoring data of various pollutant monitoring points, topographic data, and the total concentration contribution of the pollution source at the target site;

[0042] Circular update module: used to update the initial weight of the pollution source contribution through the contribution weight of the pollution source, and obtain the corrected single pollutant concentration contribution value through the updated contribution weight of the pollution source.

[0043] 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.

[0044] The technical solutions provided by the embodiments of the present invention may have the following beneficial effects:

[0045] This application combines the AERMOD model with a deep learning model to build a closed-loop pollution source contribution optimization and tracing system, combining the simulation output of AERMOD with actual monitoring data, meteorological data and geographic spatial information, and correcting it through a deep learning model, so as to further decompose the optimized total concentration contribution into the contribution ratio of individual pollutants. Multi-feature input integrates multi-dimensional features such as historical contribution weights, meteorological data, and geographic information. The deep learning model can capture complex spatiotemporal dependencies and ensure that the decomposition results are more in line with reality. The output results of the model can be dynamically fed back to the front-layer structure, and a closed-loop linkage from data input to strategy output is achieved by continuously optimizing the pollution source contribution weights. This application introduces deep learning methods such as DCRNN into the pollution source tracing closed-loop system, combines the advantages of traditional simulation models, realizes the deep integration of data-driven and physical models, and realizes the linkage between AERMOD and deep learning models through a dynamic feedback mechanism to improve accuracy.

[0046] 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

[0047] 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.

[0048] Figure 1 is a flow chart illustrating a multi-stage optimized spatiotemporal tracing method according to an exemplary embodiment;

[0049] Figure 2 is a schematic diagram of an overall implementation process framework according to another exemplary embodiment;

[0050] Figure 3 is a schematic diagram of optimized total concentration contribution according to another exemplary embodiment;

[0051] Figure 4is a schematic diagram showing the corrected concentration contribution of a single pollutant according to another exemplary embodiment;

[0052] Figure 5 is a schematic diagram of pollution source contribution weights according to another exemplary embodiment;

[0053] Figure 6 is a system schematic diagram of a multi-stage optimized space-time tracing device according to another exemplary embodiment;

[0054] In the accompanying figure: 1-first data acquisition module, 2-first input module, 3-AERMOD module, 4-second data acquisition module, 5-first deep learning module, 6-second deep learning module, 7-third deep learning module, 8-loop update module. DETAILED DESCRIPTION

[0055] 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.

[0056] Example 1

[0057] Figure 1 is a flow chart of a multi-stage optimized spatiotemporal tracing method according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0058] S1, obtaining meteorological data for a specified historical period within a preset range of the target site, obtaining topographic data within a preset range of the target site, and obtaining monitoring data of various pollutant monitoring points within a specified historical period within the preset range of the target site;

[0059] S2, preprocessing the meteorological data, topographic data, and monitoring data of various pollutant monitoring points to obtain first input data;

[0060] S3, inputting the first input data into a pre-built AERMOD model to obtain the total concentration contribution of the pollution source at the target site;

[0061] S4, obtaining real-time monitoring data of various pollutant monitoring points, and obtaining initial weights of pollution source contributions based on the proportions of various pollution sources in the real-time monitoring data of various pollutant monitoring points;

[0062] S5, based on the total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, the topographic data, and the initial weights of the pollution source contributions, the pre-built first spatiotemporal deep learning model DCRNN is used to obtain the corrected total pollution source concentration contribution of the target site;

[0063] S6, based on the total pollution source concentration contribution of the corrected target site, the real-time monitoring data of various pollutant monitoring points, the real-time meteorological data, and the initial weights of the pollution source contributions, obtain the corrected individual pollutant concentration contribution value through the pre-built second spatiotemporal deep learning model DCRNN;

[0064] S7, based on the corrected individual pollutant concentration contribution value, the real-time monitoring data of various pollutant monitoring points, the topographic data, and the total concentration contribution of the pollution source at the target site, the contribution weight of the pollution source is obtained through the pre-built third spatiotemporal deep learning model DCRNN;

[0065] S8, updating the initial weight of the pollution source contribution according to the contribution weight of the pollution source, and obtaining a corrected single pollutant concentration contribution value according to the updated contribution weight of the pollution source;

[0066] It is understandable that Figure 2 As shown, this embodiment first needs to determine the emission source of the target site, then obtain meteorological data for a specified historical period within a preset range of the target site (usually a circular area with a radius of Y and centered on the target site), obtain geographic spatial information (usually topographic data) of the target site within the preset range, and obtain monitoring data of various pollutant monitoring points for a specified historical period within the preset range of the target site; preprocess the above data, which generally involves deleting outliers, filling missing values, and standardizing data from different sources and formats, to obtain first input data;

[0067] The first input data is fed into the pre-built AERMOD model, which outputs the total contribution of pollution source concentration at the target site. The AERMOD model is based on the Gaussian steady-state plume diffusion model and is generally used to predict the concentration distribution of pollutants in the atmosphere. By considering meteorological conditions and topographic factors, the model can provide relatively accurate concentration prediction data. It is applicable to emissions from a variety of emission sources (including point sources, surface sources, and volume sources), and is suitable for simulating and predicting a variety of emission diffusion scenarios, including rural and urban environments, flat and complex terrain, and ground and elevated sources.

[0068] Then, the current meteorological data and real-time monitoring data of various pollutant monitoring points are obtained, and the initial weight of the contribution of the pollution source is obtained according to the proportion of each pollution source in the real-time monitoring data of various pollutant monitoring points; the second input data is constructed according to the output of the AERMOD model, the current meteorological data, the real-time monitoring data of various pollutant monitoring points, the initial weight of the contribution of the pollution source and other information; the first node feature matrix X is constructed according to the second input data, that is, the multi-feature data set of multiple sites; at the same time, the graph structure information is obtained according to the spatial distribution of the total concentration contribution of the pollution source of the target site within the preset range of the target site; the first node feature matrix X and the graph structure information are used together as the input of the pre-built first spatiotemporal deep learning model DCRNN, and the corrected total concentration contribution of the pollution source of the target site is obtained through the first spatiotemporal deep learning model DCRNN, as shown in the attached figure. Figure 3 As shown;

[0069] Among them, the spatiotemporal deep learning model DCRNN, namely the diffuse convolutional recursive neural network, uses diffuse convolution to capture spatial features with the topological structure of the graph (the relationship between nodes and edges). Diffused convolution is a convolution operation based on a directed graph. It captures the propagation of information from one node to other nodes by simulating random walks. It uses an extended version based on GRU to construct a decoding module to accept the spatiotemporal features of the hidden state and map it to the target variable. Simply put, the spatiotemporal deep learning model DCRNN includes a graph convolution module, a recurrent neural network module and a decoding module. The graph convolution module extracts the spatial features of the input features, the recurrent neural network module is used to extract the temporal features of the input features, and the decoding module fuses the spatial features and the temporal features to obtain the spatiotemporal features.

[0070] According to the corrected total concentration contribution of the pollution source of the target site and the initial weight of the pollution source contribution, the corrected pollutant contribution and the error assessment are obtained; real-time meteorological data are obtained, and the real-time monitoring data of various pollutant monitoring points, real-time meteorological data, the initial weight of the pollution source contribution, the corrected total concentration contribution of the pollution source of the target site, the corrected pollutant contribution and the error assessment are preprocessed to ensure the consistency of the input time series and spatial characteristics, and obtain the third input data; based on the third input data, a second node feature matrix X is constructed, and the second node feature matrix and the graph structure information are input into the pre-built second spatiotemporal deep learning model DCRNN to obtain the corrected single pollutant concentration contribution value, as shown in the attached figure. Figure 4 As shown;

[0071] Then, the corrected single pollutant concentration contribution value, the real-time monitoring data of various pollutant monitoring points, the topographic data and the total concentration contribution of the pollution source of the target site are preprocessed to obtain the fourth input data; the third node feature matrix is ​​constructed based on the fourth input data, and the third node feature matrix and the graph structure information are input into the pre-built third spatiotemporal deep learning model DCRNN to obtain the contribution weight of the pollution source, as shown in the attached figure. Figure 5 As shown;

[0072] The contribution weights are updated to the above-mentioned optimization process. The updated pollution source contribution weights can more accurately reflect the impact of pollution sources on target sites. By continuously optimizing the pollution source contribution weights, a closed-loop linkage from data input to strategy output can be achieved.

[0073] Example 2:

[0074] Figure 6 1 is a system diagram illustrating a multi-stage optimized spatiotemporal tracing device according to another exemplary embodiment, the device comprising:

[0075] The first data acquisition module 1 is used to obtain meteorological data of a specified historical period within a preset range of the target site, obtain topographic data within a preset range of the target site, and obtain monitoring data of various pollutant monitoring points within a specified historical period within the preset range of the target site;

[0076] First input module 2: used for pre-processing the meteorological data, topographic data and monitoring data of various pollutant monitoring points to obtain first input data;

[0077] AERMOD module 3: used to input the first input data into the pre-built AERMOD model to obtain the total concentration contribution of the pollution source of the target site;

[0078] Second data acquisition module 4: used to obtain real-time monitoring data of various pollutant monitoring points, and obtain the initial weight of pollution source contribution according to the proportion of each pollution source in the real-time monitoring data of various pollutant monitoring points;

[0079] The first deep learning module 5 is used to obtain the corrected total pollution source concentration contribution of the target site through a pre-built first spatiotemporal deep learning model DCRNN based on the total pollution source concentration contribution of the target site, real-time monitoring data of various pollutant monitoring points, topographic data, and initial weights of pollution source contributions;

[0080] The second deep learning module 6 is used to obtain the corrected individual pollutant concentration contribution value through a pre-built second spatiotemporal deep learning model DCRNN based on the corrected total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, the real-time meteorological data, and the initial weight of the pollution source contribution;

[0081] The third deep learning module 7 is used to obtain the contribution weight of the pollution source through the pre-built third spatiotemporal deep learning model DCRNN based on the corrected single pollutant concentration contribution value, the real-time monitoring data of various pollutant monitoring points, the topographic data, and the total concentration contribution of the pollution source at the target site;

[0082] Circular updating module 8 is used to update the initial weight of the pollution source contribution by the contribution weight of the pollution source, and obtain the corrected single pollutant concentration contribution value by the updated pollution source contribution weight.

[0083] Example 3:

[0084] 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;

[0085] It is understandable that the storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0086] 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.

[0087] 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.

[0088] 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.

[0089] 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.

[0090] 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.

[0091] 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.

[0092] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0093] 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.

[0094] 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 multi-stage optimized spatiotemporal tracing method, characterized in that: The method comprises: Obtain meteorological data for a specified historical period within a preset range of the target site, obtain topographic data within a preset range of the target site, and obtain monitoring data for various pollutant monitoring points within a specified historical period within the preset range of the target site; Preprocessing the meteorological data, topographic data, and monitoring data of various pollutant monitoring points to obtain first input data; Input the first input data into the pre-built AERMOD model to obtain the total concentration contribution of the pollution source at the target site; Obtaining real-time monitoring data from various pollutant monitoring points, and obtaining initial weights of pollution source contributions based on the proportion of each pollution source in the real-time monitoring data from various pollutant monitoring points; Based on the total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, the topographic data and the initial weight of the pollution source contribution, the pre-built first spatiotemporal deep learning model DCRNN is used to obtain the corrected total pollution source concentration contribution of the target site; Based on the total pollution source concentration contribution of the corrected target site, the real-time monitoring data of various pollutant monitoring points, the real-time meteorological data, and the initial weight of the pollution source contribution, the pre-built second spatiotemporal deep learning model DCRNN is used to obtain the corrected single pollutant concentration contribution value; Based on the corrected individual pollutant concentration contribution value, real-time monitoring data of various pollutant monitoring points, topographic data, and the total concentration contribution of pollution sources at the target site, the contribution weight of the pollution source is obtained through the pre-built third spatiotemporal deep learning model DCRNN; The initial weight of the pollution source contribution is updated by the contribution weight of the pollution source, and the corrected single pollutant concentration contribution value is obtained by the updated contribution weight of the pollution source.

2. The method according to claim 1, characterized in that The total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, the topographic data, and the initial weight of the pollution source contribution are obtained by using the pre-built first spatiotemporal deep learning model DCRNN to obtain the corrected total pollution source concentration contribution of the target site, including: Preprocessing the total concentration contribution of pollution sources at the target site, real-time monitoring data of various pollutant monitoring points, topographic data, and initial weights of pollution source contributions to obtain second input data; A first node feature matrix is ​​constructed based on the second input data to obtain graph structure information; the first node feature matrix and graph structure information are input into a pre-built first spatiotemporal deep learning model DCRNN to obtain the corrected total concentration contribution of the pollution source of the target site.

3. The method according to claim 2, characterized in that The method of obtaining the corrected single pollutant concentration contribution value based on the corrected total pollution source concentration contribution of the target site, the real-time monitoring data of various pollutant monitoring points, the real-time meteorological data, and the initial weight of the pollution source contribution through the pre-built second spatiotemporal deep learning model DCRNN includes: Obtaining a corrected pollutant contribution based on the corrected total pollution source concentration contribution of the target site and the initial weight of the pollution source contribution, and obtaining an error evaluation of the first spatiotemporal deep learning model DCRNN based on the corrected total pollution source concentration contribution of the target site and the total pollution source concentration contribution of the target site; Acquiring real-time meteorological data within a preset range of the target site, preprocessing the real-time monitoring data of each pollutant monitoring point, the real-time meteorological data, the initial weights of pollution source contributions, the corrected total pollution source concentration contribution of the target site, the corrected pollutant contribution, and the error assessment to obtain third input data; A second node feature matrix is ​​constructed based on the third input data, and the second node feature matrix and graph structure information are input into a pre-built second spatiotemporal deep learning model DCRNN to obtain the corrected single pollutant concentration contribution value.

4. The method according to claim 3, characterized in that The contribution weight of the pollution source is obtained by using the pre-built third spatiotemporal deep learning model DCRNN based on the corrected single pollutant concentration contribution value, the real-time monitoring data of various pollutant monitoring points, the topographic data, and the total concentration contribution of the pollution source at the target site. The method includes: Preprocessing the corrected individual pollutant concentration contribution values, real-time monitoring data of various pollutant monitoring points, topographic data, and the total concentration contribution of pollution sources at the target site to obtain fourth input data; A third node feature matrix is ​​constructed based on the fourth input data, and the third node feature matrix and graph structure information are input into a pre-built third spatiotemporal deep learning model DCRNN to obtain the contribution weight of the pollution source.

5. The method according to claim 2, characterized in that The obtaining of graph structure information includes: Graph structure information is obtained based on the spatial distribution of the total concentration contribution of the pollution sources of the target site within a preset range of the target site.

6. The method according to claim 5, characterized in that The first spatiotemporal deep learning model DCRNN, the second spatiotemporal deep learning model DCRNN, and the third spatiotemporal deep learning model DCRNN all include: a graph convolution module, a recurrent neural network module, and a decoding module; The graph convolution module is used to extract the spatial features of the input features, the recurrent neural network module is used to extract the temporal features of the input features, and the decoding module is used to fuse the extracted spatial features and temporal features.

7. A multi-stage optimized space-time tracing device, characterized in that: The device comprises: The first data acquisition module is used to obtain meteorological data for a specified historical period within a preset range of the target site, obtain topographic data within a preset range of the target site, and obtain monitoring data of various pollutant monitoring points within a specified historical period within the preset range of the target site; A first input module is used to pre-process the meteorological data, topographic data, and monitoring data of various pollutant monitoring points to obtain first input data; AERMOD module: used to input the first input data into the pre-built AERMOD model to obtain the total concentration contribution of the pollution source of the target site; The second data acquisition module is used to obtain real-time monitoring data of various pollutant monitoring points, and obtain the initial weight of the pollution source contribution according to the proportion of each pollution source in the real-time monitoring data of various pollutant monitoring points; The first deep learning module is used to obtain the corrected total pollution source concentration contribution of the target site through a pre-built first spatiotemporal deep learning model DCRNN based on the total pollution source concentration contribution of the target site, real-time monitoring data of various pollutant monitoring points, topographic data, and initial weights of pollution source contributions; The second deep learning module is used to obtain the corrected individual pollutant concentration contribution value through the pre-built second spatiotemporal deep learning model DCRNN based on the total pollution source concentration contribution of the corrected target site, the real-time monitoring data of various pollutant monitoring points, the real-time meteorological data, and the initial weight of the pollution source contribution; The third deep learning module is used to obtain the contribution weight of the pollution source through the pre-built third spatiotemporal deep learning model DCRNN based on the corrected single pollutant concentration contribution value, real-time monitoring data of various pollutant monitoring points, topographic data, and the total concentration contribution of the pollution source at the target site; Circular update module: used to update the initial weight of the pollution source contribution through the contribution weight of the pollution source, and obtain the corrected single pollutant concentration contribution value through the updated pollution source contribution weight.

8. 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 in the multi-stage optimized spatiotemporal tracing method according to any one of claims 1 to 6 is implemented.

Images