Atmospheric pollutant monitoring method and system based on GIS platform

Abstract

The invention relates to an atmospheric pollutant monitoring method and system based on a GIS platform, and relates to the technical field of atmospheric pollutant monitoring. The method comprises the following steps of: integrating position information of each monitoring station and a time sequence of atmospheric pollutants collected by pollutant monitoring equipment of each monitoring station into a GIS (Geographic Information System) database; performing sulfur dioxide concentration time sequence mode feature extraction on the time sequence of the atmospheric pollutants collected by the pollutant monitoring equipment of each monitoring station; using a position encoder to encode the position information of each monitoring station; and carrying out feature vector sequence merging and information transfer reasoning on the sequence of the monitoring station position coding vectors and the sequence of the sulfur dioxide time sequence correlation feature vectors to obtain a pollution source spatial reasoning semantic feature vector so as to determine the predicted position information of the pollution source, and spatial reasoning of the pollution source can be carried out. Therefore, the spatial position information of the pollution source is predicted, and the potential source of the pollutant is traced.

Description

Technical Field

[0001] The present disclosure relates to the technical field of atmospheric pollutant monitoring, and in particular, to a method and system for monitoring atmospheric pollutants based on a GIS platform. Background Art

[0002] Atmospheric pollution refers to the phenomenon that harmful or excessive substances existing in the atmosphere cause harm to human health, ecosystems, and the environment. Sulfur dioxide is an irritating gas, and inhalation can lead to respiratory inflammation, causing respiratory problems such as coughing, sore throat, and asthma. Therefore, monitoring the concentration and distribution of sulfur dioxide atmospheric pollutants is an important means to protect the environment and human health.

[0003] However, traditional methods for monitoring atmospheric pollutants are usually implemented through fixed monitoring stations and sensors. These monitoring stations are usually located in the city center or industrial areas and cannot fully cover all areas of the city and rural areas. This results in insufficient spatial resolution of the monitoring data and cannot accurately reflect the pollution situation of the entire city or region. In addition, traditional monitoring schemes usually only provide pollutant concentration data and lack the ability to trace the pollution source. That is to say, traditional atmospheric pollutant monitoring schemes cannot accurately determine the source and propagation path of pollutants, which will affect the investigation and handling capabilities of relevant departments for pollution incidents.

[0004] Therefore, an optimized atmospheric pollutant monitoring scheme is desired. Summary of the Invention

[0005] This Summary of the Invention section is provided to introduce concepts in a brief form, which will be described in detail in the following Detailed Implementation section. This Summary of the Invention section is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to be used to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present disclosure provides a method for monitoring atmospheric pollutants based on a GIS platform, the method comprising:

[0007] Obtaining a time series of atmospheric pollutants collected by pollutant monitoring devices deployed at each monitoring station, wherein the atmospheric pollutant is a sulfur dioxide concentration value;

[0008] Obtaining the location information of each monitoring station;

[0009] Integrating the location information of each monitoring station and the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring station into a GIS database;

[0010] Extract the time series pattern features of sulfur dioxide concentration from the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors;

[0011] Use a position encoder to encode the location information of each monitoring site to obtain a sequence of monitoring site location encoding vectors;

[0012] Merge the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide time series correlation feature vectors, and perform information transfer reasoning to obtain a pollution source spatial reasoning semantic feature vector;

[0013] Based on the pollution source spatial reasoning semantic feature vector, determine the location information of the predicted pollution source.

[0014] Optionally, extracting the time series pattern features of sulfur dioxide concentration from the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors includes: arranging the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site along the time dimension respectively to obtain a sequence of sulfur dioxide time series input vectors; passing each sulfur dioxide time series input vector in the sequence of sulfur dioxide time series input vectors through a sulfur dioxide concentration time series pattern feature extractor based on a one-dimensional convolutional layer to obtain the sequence of sulfur dioxide time series correlation feature vectors.

[0015] Optionally, merging the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide time series correlation feature vectors, and performing information transfer reasoning to obtain a pollution source spatial reasoning semantic feature vector includes: merging the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide time series correlation feature vectors to obtain a sequence of sulfur dioxide time series correlation feature vectors containing location information; passing the sequence of sulfur dioxide time series correlation feature vectors containing location information through a feature vector sequence transfer reasoning module to obtain the pollution source spatial reasoning semantic feature vector.

[0016] Optionally, passing the sequence of the sulfur dioxide time-series correlation feature vectors containing location information through a feature vector sequence transfer inference module to obtain the pollution source spatial inference semantic feature vector, including: calculating the weighted sum of the first s sulfur dioxide time-series correlation feature vectors containing location information in the sequence of the sulfur dioxide time-series correlation feature vectors containing location information to obtain a neighbor semantic fusion feature vector; performing weighted processing on each location feature value in the s-th sulfur dioxide time-series correlation feature vector containing location information with the sum of a trainable preset hyperparameter and one as the weighting coefficient to obtain a weighted optimized sulfur dioxide time-series correlation feature vector containing location information; performing position-wise addition on the neighbor semantic fusion feature vector and the weighted optimized sulfur dioxide time-series correlation feature vector containing location information and then processing through a multi-layer perceptron to obtain the pollution source inference semantic feature vector corresponding to the s-th sulfur dioxide time-series correlation feature vector containing location information; calculating the position-wise addition between the pollution source inference semantic feature vectors corresponding to each sulfur dioxide time-series correlation feature vector containing location information in the sequence of the sulfur dioxide time-series correlation feature vectors containing location information to obtain the pollution source spatial inference semantic feature vector.

[0017] Optionally, based on the pollution source spatial inference semantic feature vector, determining the location information of the predicted pollution source, including: passing the pollution source spatial inference semantic feature vector through an inference device based on a decoder to obtain a decoding result, where the decoding result is the location information of the predicted pollution source.

[0018] Optionally, it further includes a training step: for training the sulfur dioxide concentration time-series pattern feature extractor based on a one-dimensional convolutional layer, the position encoder, the feature vector sequence transfer inference module, and the inference device based on a decoder.

[0019] Optionally, the training step includes: obtaining a time series of training atmospheric pollutants collected by pollutant monitoring devices deployed at each monitoring site, where the training atmospheric pollutant is a training sulfur dioxide concentration value; obtaining the training location information of each monitoring site; integrating the training location information of each monitoring site and the time series of training atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site into a GIS database; extracting sulfur dioxide concentration time series pattern features from the time series of training atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of training sulfur dioxide time series correlation feature vectors; encoding the training location information of each monitoring site using a position encoder to obtain a sequence of training monitoring site location encoding vectors; performing feature vector sequence merging and information transfer reasoning on the sequence of training monitoring site location encoding vectors and the sequence of training sulfur dioxide time series correlation feature vectors to obtain a training pollution source spatial reasoning semantic feature vector; performing feature optimization on the training pollution source spatial reasoning semantic feature vector to obtain an optimized training pollution source spatial reasoning semantic feature vector; passing the optimized training pollution source spatial reasoning semantic feature vector through the decoder-based inference engine to obtain a decoding loss function value; and training the sulfur dioxide concentration time series pattern feature extractor based on a one-dimensional convolutional layer, the position encoder, the feature vector sequence transfer reasoning module, and the decoder-based inference engine based on the decoding loss function value.

[0020] In a second aspect, the present disclosure provides an atmospheric pollutant monitoring system based on a GIS platform. The system includes:

[0021] An atmospheric pollutant acquisition module for obtaining a time series of atmospheric pollutants collected by pollutant monitoring devices deployed at each monitoring site, where the atmospheric pollutant is a sulfur dioxide concentration value;

[0022] A location information acquisition module for obtaining the location information of each monitoring site;

[0023] A GIS database integration module for integrating the location information of each monitoring site and the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site into a GIS database;

[0024] A sulfur dioxide concentration time series pattern feature extraction module for extracting sulfur dioxide concentration time series pattern features from the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors;

[0025] A location encoding module for encoding the location information of each monitoring site using a position encoder to obtain a sequence of monitoring site location encoding vectors;

[0026] A feature vector sequence merging and information transfer inference module is configured to perform feature vector sequence merging and information transfer inference on the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide time-series correlation feature vectors to obtain a pollution source spatial inference semantic feature vector.

[0027] A predicted pollution source location information determination module is configured to determine the predicted pollution source location information based on the pollution source spatial inference semantic feature vector.

[0028] Optionally, the sulfur dioxide concentration time-series pattern feature extraction module includes: a vector arrangement unit configured to arrange the time series of the air pollutants collected by the pollutant monitoring devices of each monitoring site along the time dimension respectively to obtain a sequence of sulfur dioxide time-series input vectors; a feature extraction unit configured to respectively pass each sulfur dioxide time-series input vector in the sequence of the sulfur dioxide time-series input vectors through a sulfur dioxide concentration time-series pattern feature extractor based on a one-dimensional convolutional layer to obtain a sequence of the sulfur dioxide time-series correlation feature vectors.

[0029] Optionally, the feature vector sequence merging and information transfer inference module includes: a sequence merging unit configured to perform feature vector sequence merging on the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide time-series correlation feature vectors to obtain a sequence of sulfur dioxide time-series correlation feature vectors containing location information; a feature vector sequence transfer inference unit configured to pass the sequence of the sulfur dioxide time-series correlation feature vectors containing location information through a feature vector sequence transfer inference module to obtain the pollution source spatial inference semantic feature vector.

[0030] By adopting the above technical solution, the location information of each monitoring site and the time series of the air pollutants collected by the pollutant monitoring devices of each monitoring site are integrated into the GIS database; sulfur dioxide concentration time-series pattern feature extraction is performed on the time series of the air pollutants collected by the pollutant monitoring devices of each monitoring site; the location information of each monitoring site is encoded using a location encoder; the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide time-series correlation feature vectors are subjected to feature vector sequence merging and information transfer inference to obtain a pollution source spatial inference semantic feature vector, so as to determine the predicted pollution source location information, and spatial inference of the pollution source can be performed, thereby predicting the spatial location information of the pollution source to trace the potential sources of pollutants.

[0031] Other features and advantages of the present disclosure will be described in detail in the subsequent specific implementation section. Description of the Drawings

[0032] In conjunction with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic and that the original and the elements are not necessarily drawn to scale. In the drawings:

[0033] Figure 1 is a flowchart of a method for monitoring atmospheric pollutants based on a GIS platform shown according to an exemplary embodiment.

[0034] Figure 2 is a block diagram of a system for monitoring atmospheric pollutants based on a GIS platform shown according to an exemplary embodiment.

[0035] Figure 3 is a block diagram of an electronic device shown according to an exemplary embodiment.

[0036] Figure 4 is an application scenario diagram of a method for monitoring atmospheric pollutants based on a GIS platform shown according to an exemplary embodiment. Specific Embodiments

[0037] The embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0038] It should be understood that the various steps recited in the method embodiments of the present disclosure can be executed in a different order and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0039] As used herein, the term "including" and its variations are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0040] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules, or units, and are not used to limit the order of the functions performed by these devices, modules, or units or their interdependent relationships.

[0041] It should be noted that the modifications of "one" and "multiple" mentioned in this disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0042] The names of the messages or information exchanged between multiple devices in the embodiments of this disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0043] To solve the above problems, this disclosure provides an air pollutant monitoring method and system based on a GIS platform. By integrating the location information of each monitoring site and the time series of air pollutants collected by the pollutant monitoring devices at each monitoring site into a GIS database; extracting the time series pattern features of sulfur dioxide concentration from the time series of air pollutants collected by the pollutant monitoring devices at each monitoring site; encoding the location information of each monitoring site using a location encoder; merging the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time series correlation feature vectors and performing information transfer reasoning to obtain the spatial reasoning semantic feature vector of the pollution source, so as to determine the location information of the predicted pollution source, and spatial reasoning of the pollution source can be carried out, thereby predicting the spatial location information of the pollution source to trace the potential sources of pollutants.

[0044] The following will detail the specific embodiments of this disclosure with reference to the accompanying drawings.

[0045] Figure 1 is a flowchart of an air pollutant monitoring method based on a GIS platform shown according to an exemplary embodiment, as Figure 1 shown, the method includes:

[0046] S101. Obtain the time series of air pollutants collected by the pollutant monitoring devices deployed at each monitoring site, where the air pollutant is the sulfur dioxide concentration value;

[0047] S102. Obtain the location information of each monitoring site;

[0048] S103. Integrate the location information of each monitoring site and the time series of air pollutants collected by the pollutant monitoring devices at each monitoring site into a GIS database;

[0049] S104. Extract the time series pattern features of sulfur dioxide concentration from the time series of air pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors;

[0050] S105. Encode the location information of each monitoring site using a location encoder to obtain a sequence of monitoring site location encoding vectors;

[0051] S106. Merge the sequences of the monitoring site location coding vectors and the sequences of the sulfur dioxide time-series correlation feature vectors, and perform feature vector sequence merging and information transfer reasoning to obtain the pollution source spatial reasoning semantic feature vectors.

[0052] S107. Based on the pollution source spatial reasoning semantic feature vectors, determine the location information of the predicted pollution sources.

[0053] It should be understood that the GIS platform can integrate different spatial data sources, including satellite remote sensing data, air quality monitoring data, geographical information data, etc., to achieve the fusion analysis of multi-source data, which helps to achieve the comprehensive monitoring and analysis of air pollutants, thus facilitating the determination of the location and influence range of pollution sources, so as to formulate targeted pollution prevention and control measures.

[0054] Based on this, in the technical solution of the present application, an air pollutant monitoring scheme based on the GIS platform is proposed. It can collect the concentration values of the air pollutant sulfur dioxide through the pollutant monitoring devices deployed at each monitoring site, and obtain the location information of each monitoring site. Further, the GIS platform is used to integrate the time-series data of the sulfur dioxide concentration collected at each monitoring site and the location information of each monitoring site. Then, an artificial intelligence-based data processing and analysis algorithm is introduced at the backend to perform the collaborative correlation analysis of the time-series data of the sulfur dioxide concentration values monitored at multiple monitoring sites and the location information of multiple monitoring sites, so as to perform the spatial reasoning of pollution sources, thereby predicting the spatial location information of pollution sources to trace the potential sources of pollutants.

[0055] Specifically, in the technical solution of the present application, first, obtain the time series of air pollutants collected by the pollutant monitoring devices deployed at each monitoring site, where the air pollutant is the sulfur dioxide concentration value, and obtain the location information of each monitoring site. Further, integrate the location information of each monitoring site and the time series of air pollutants collected by the pollutant monitoring devices at each monitoring site into the GIS database.

[0056] Then, considering that there are dynamic change laws of time series in the atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site, that is, the sulfur dioxide concentration values, which means that the sulfur dioxide concentrations monitored at each monitoring site will change continuously over time. Therefore, in order to better perform time series analysis on the atmospheric pollutants monitored at each monitoring site to facilitate revealing the change laws of sulfur dioxide concentration over time, including time series patterns and trend characteristics such as seasonal changes and trend changes, in the technical solution of this application, it is necessary to arrange the time series of the atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site according to the time dimension respectively to obtain a sequence of sulfur dioxide time series input vectors. By arranging the time series of the atmospheric pollutants monitored at each monitoring site according to the time dimension, the time series data of the sulfur dioxide concentrations monitored at different monitoring sites can be regularized according to the time dimension, thereby retaining the time series information of the sulfur dioxide concentration, which helps to extract the time series patterns and change characteristics of the sulfur dioxide concentration at each site and helps to evaluate the distribution of pollutants in different regions.

[0057] In an embodiment of the present disclosure, performing sulfur dioxide concentration time series pattern feature extraction on the time series of the atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors includes: arranging the time series of the atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site according to the time dimension respectively to obtain a sequence of sulfur dioxide time series input vectors; respectively passing each sulfur dioxide time series input vector in the sequence of sulfur dioxide time series input vectors through a sulfur dioxide concentration time series pattern feature extractor based on a one-dimensional convolutional layer to obtain the sequence of sulfur dioxide time series correlation feature vectors.

[0058] Furthermore, in order to capture the time series change patterns and characteristics of the sulfur dioxide concentrations at different monitoring sites to identify the sulfur dioxide concentration distribution in different regions, which is beneficial for subsequent determination of the source location, in the technical solution of this application, further perform feature mining on each sulfur dioxide time series input vector in the sequence of sulfur dioxide time series input vectors through a sulfur dioxide concentration time series pattern feature extractor based on a one-dimensional convolutional layer to respectively extract the time series patterns and dynamic feature information of the sulfur dioxide concentrations at each monitoring site in the time dimension, thereby obtaining a sequence of sulfur dioxide time series correlation feature vectors, which can help to identify the time series patterns and change laws in the time series data of the atmospheric pollutants monitored at each monitoring site, thereby discovering important patterns in the sulfur dioxide concentration time series changes, such as periodic changes, trend changes, and time series fluctuations, etc., providing a basis for subsequent pattern recognition and source location prediction.

[0059] Next, since the location information of each monitoring site is an important element in spatial data, which can affect the spatial characteristics and correlations of the monitoring data, and is also beneficial for judging the spatial distribution of atmospheric pollutants and predicting the location of pollution sources. Therefore, it is necessary to incorporate the location information of each monitoring site into the monitoring data of atmospheric pollutants. Based on this, in the technical solution of this application, a position encoder is further used to encode the location information of each monitoring site to obtain a sequence of monitoring site location encoding vectors. Through the processing of the position encoder, the location information of each monitoring site can be encoded into a vector form, which helps to subsequently embed the location information of each site into the model, enabling the model to better understand the spatial relationships between different monitoring sites and perform spatial location inference of pollution sources.

[0060] Subsequently, in order to incorporate the location information of the site into the temporal characteristics of the sulfur dioxide concentration monitored at each monitoring site, so as to more fully understand the spatial distribution of the sulfur dioxide concentration, which helps to improve the subsequent spatial location inference of pollution sources. Based on this, in the technical solution of this application, it is necessary to further merge the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide temporal correlation feature vectors to obtain a sequence of sulfur dioxide temporal correlation feature vectors containing location information. By merging the location encoding features and the temporal characteristics of the sulfur dioxide concentration of each monitoring site, it is possible to help establish the correlation relationship between space and time, thus better capturing the spatio-temporal correlation patterns in the monitoring data. This helps the model to better understand the spatio-temporal correlation in the monitoring data, facilitating the detection of the distribution of sulfur dioxide concentration in the time dimension and space dimension, and thus more accurately determining the location of pollution sources.

[0061] In an embodiment of the present disclosure, merging the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide temporal correlation feature vectors for feature vector sequence merging and information transfer inference to obtain a pollution source spatial inference semantic feature vector includes: merging the sequence of the monitoring site location encoding vectors and the sequence of the sulfur dioxide temporal correlation feature vectors to obtain a sequence of sulfur dioxide temporal correlation feature vectors containing location information; passing the sequence of the sulfur dioxide temporal correlation feature vectors containing location information through a feature vector sequence transfer inference module to obtain the pollution source spatial inference semantic feature vector.

[0062] It should be understood that considering that each of the sulfur dioxide time-series correlation feature vectors containing location information in the sequence of sulfur dioxide time-series correlation feature vectors reflects the fusion semantics between the time-series characteristics of the sulfur dioxide monitoring concentration and the site spatial location characteristics of each monitoring site. Therefore, in order to be able to perform spatial location inference of pollution sources based on the correlation relationship between each site, in the technical solution of this application, the sequence of the sulfur dioxide time-series correlation feature vectors containing location information is further passed through a feature vector sequence transfer inference module to obtain a pollution source spatial inference semantic feature vector. It should be understood that through the processing of the feature vector sequence transfer inference module, the sulfur dioxide time-series correlation feature vectors containing location information of each monitoring site can be information-transferred, so as to utilize the time-series characteristics of the sulfur dioxide concentration containing location information of each monitoring site for spatial inference, thereby obtaining the spatial inference semantics of the pollution source. That is to say, the transfer inference module can help analyze the spatial correlation of the time-series characteristics of sulfur dioxide concentration between each monitoring site, reveal the spatial interaction and influence between different monitoring sites, which helps to identify the potential pollution source locations and spatial distribution characteristics, and improve the performance of the model in pollution source identification and spatial analysis.

[0063] In an embodiment of the present disclosure, passing the sequence of the sulfur dioxide time-series correlation feature vectors containing location information through a feature vector sequence transfer inference module to obtain the pollution source spatial inference semantic feature vector includes: calculating the weighted sum of the first s sulfur dioxide time-series correlation feature vectors containing location information in the sequence of the sulfur dioxide time-series correlation feature vectors containing location information to obtain a neighbor semantic fusion feature vector; using the sum of a trainable preset hyperparameter and one as a weighting coefficient to weight each position feature value in the s-th sulfur dioxide time-series correlation feature vector containing location information to obtain a weighted and optimized sulfur dioxide time-series correlation feature vector containing location information; performing position-wise summation on the neighbor semantic fusion feature vector and the weighted and optimized sulfur dioxide time-series correlation feature vector containing location information and then processing through a multi-layer perceptron to obtain a pollution source inference semantic feature vector corresponding to the s-th sulfur dioxide time-series correlation feature vector containing location information; calculating the position-wise summation between the pollution source inference semantic feature vectors corresponding to each of the sulfur dioxide time-series correlation feature vectors containing location information in the sequence of the sulfur dioxide time-series correlation feature vectors containing location information to obtain the pollution source spatial inference semantic feature vector.

[0064] In particular, in a specific example of this application, the sequence of the sulfur dioxide time-series correlation feature vectors containing location information is processed through the feature vector sequence transfer inference module according to the following information transfer inference formula to obtain the pollution source spatial inference semantic feature vector;

[0065] Among them, the information transfer inference formula is as follows:

[0066]

[0067] Among them, V s is the s-th sulfur dioxide time-series correlation feature vector containing location information in the sequence of the sulfur dioxide time-series correlation feature vectors containing location information, V k is the k-th sulfur dioxide time-series correlation feature vector containing location information in the sequence of the sulfur dioxide time-series correlation feature vectors containing location information, ∈ is a trainable preset hyperparameter, MLP(·) represents a multi-layer perceptron, n is the number of feature vectors in the sequence of the sulfur dioxide time-series correlation feature vectors containing location information, and V is the pollution source space inference semantic feature vector.

[0068] Subsequently, the pollution source space inference semantic feature vector is passed through an inference device based on a decoder to obtain a decoding result, and the decoding result is the predicted location information of the pollution source. That is to say, the spatial inference semantic features of the pollution source are used for decoding regression to predict the spatial location information of the pollution source so as to trace the potential source of the pollutant.

[0069] In an embodiment of the present disclosure, determining the predicted location information of the pollution source based on the pollution source space inference semantic feature vector includes: passing the pollution source space inference semantic feature vector through an inference device based on a decoder to obtain a decoding result, and the decoding result is the predicted location information of the pollution source.

[0070] Further, in an embodiment of the present disclosure, the method for monitoring atmospheric pollutants based on the GIS platform further includes a training step: for training the sulfur dioxide concentration time series pattern feature extractor based on the one-dimensional convolutional layer, the position encoder, the feature vector sequence transfer inference module, and the decoder-based inference engine. The training step includes: obtaining the time series of training atmospheric pollutants collected by pollutant monitoring devices deployed at each monitoring site, where the training atmospheric pollutants are training sulfur dioxide concentration values; obtaining the training location information of each monitoring site; integrating the training location information of each monitoring site and the time series of training atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site into the GIS database; performing sulfur dioxide concentration time series pattern feature extraction on the time series of training atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of training sulfur dioxide time series correlation feature vectors; using the position encoder to encode the training location information of each monitoring site to obtain a sequence of training monitoring site position encoding vectors; performing feature vector sequence merging and information transfer inference on the sequence of training monitoring site position encoding vectors and the sequence of training sulfur dioxide time series correlation feature vectors to obtain a training pollution source spatial inference semantic feature vector; performing feature optimization on the training pollution source spatial inference semantic feature vector to obtain an optimized training pollution source spatial inference semantic feature vector; passing the optimized training pollution source spatial inference semantic feature vector through the decoder-based inference engine to obtain a decoding loss function value; and training the sulfur dioxide concentration time series pattern feature extractor based on the one-dimensional convolutional layer, the position encoder, the feature vector sequence transfer inference module, and the decoder-based inference engine based on the decoding loss function value.

[0071] In a preferred example of the above technical solution, passing the training pollution source spatial inference semantic feature vector through the decoder-based inference engine to obtain a decoding loss function value specifically includes:

[0072] Dividing each eigenvalue of the training pollution source spatial inference semantic feature vector by the maximum eigenvalue of the training pollution source spatial inference semantic feature vector to obtain a pollution source spatial inference semantic interaction representation vector;

[0073] Dividing the mean value of the eigenvalues of the training pollution source spatial inference semantic feature vector by the standard deviation of the eigenvalues of the training pollution source spatial inference semantic feature vector to obtain a statistical dimension interaction value corresponding to the training pollution source spatial inference semantic feature vector;

[0074] Subtract the statistical dimension interaction value from each eigenvalue of the training pollution source spatial inference semantic interaction representation vector, and then calculate the logarithm value at each position to obtain the training pollution source spatial inference semantic interaction information representation vector;

[0075] Add the statistical dimension interaction value to each eigenvalue of the training pollution source spatial inference semantic interaction representation vector, and then multiply by a predetermined weight hyperparameter to obtain the training pollution source spatial inference semantic interaction information representation vector;

[0076] Dot the training pollution source spatial inference semantic interaction information representation vector with the training pollution source spatial inference semantic interaction information representation vector to obtain an optimized training pollution source spatial inference semantic feature vector; and

[0077] Pass the optimized training pollution source spatial inference semantic feature vector through a decoder-based inference engine to obtain the decoding loss function value.

[0078] To enhance the feature expression effect of the cross-time domain inference transfer feature of the training pollution source spatial inference semantic feature vector based on the temporal correlation feature of atmospheric pollutants and position encoding information, in the above steps, a short sequence including a dimension interaction representation containing statistical features represented by the mean and standard values and a distribution interaction representation of the eigenvalue and the maximum eigenvalue as latent variable features is used as a sub-manifold latent motif under the complex manifold network of the training pollution source spatial inference semantic feature vector. Thus, the potential motif feature information pattern and feature distribution pattern of the training pollution source spatial inference semantic feature vector based on the training pollution source spatial inference semantic interaction information representation vector and the training pollution source spatial inference semantic interaction information representation vector are used as its global structure inference unit. Then, based on connection, the complex manifold structure of the training pollution source spatial inference semantic feature vector is reconstructed in the form of a global structure latent motif dictionary, so as to enhance the model's understanding ability of the generation and evolution of the manifold structure corresponding to the features under the complex feature representation dimension during the iterative process, improve the feature expression effect of the complex feature representation dimension of the training pollution source spatial inference semantic feature vector during the model iteration process, and thus improve the accuracy of the decoding result obtained by passing the training pollution source spatial inference semantic feature vector through a decoder-based inference engine. In this way, the spatial inference of the pollution source can be carried out more accurately, so as to predict the spatial position information of the pollution source and trace the potential source of the pollutant.

[0079] In summary, adopting the above solution, the concentration value of sulfur dioxide, an air pollutant, is collected by pollutant monitoring devices deployed at each monitoring site, and the location information of each monitoring site is obtained. Further, the time-series data of sulfur dioxide concentration collected by each monitoring site and the location information of each monitoring site are integrated through a GIS platform. Then, an artificial intelligence-based data processing and analysis algorithm is introduced at the backend to perform collaborative correlation analysis on the time-series data of sulfur dioxide concentration values monitored at multiple monitoring sites and the location information of multiple monitoring sites, so as to perform spatial reasoning of pollution sources, thereby predicting the spatial location information of pollution sources to trace the potential sources of pollutants.

[0080] Figure 2 is a block diagram of an air pollutant monitoring system based on a GIS platform shown according to an exemplary embodiment. As Figure 2 shown, the system 200 includes:

[0081] An air pollutant acquisition module 201, configured to acquire the time series of air pollutants collected by pollutant monitoring devices deployed at each monitoring site, where the air pollutant is the sulfur dioxide concentration value;

[0082] A location information acquisition module 202, configured to acquire the location information of each monitoring site;

[0083] A GIS database integration module 203, configured to integrate the location information of each monitoring site and the time series of air pollutants collected by the pollutant monitoring devices of each monitoring site into a GIS database;

[0084] A sulfur dioxide concentration time-series pattern feature extraction module 204, configured to perform sulfur dioxide concentration time-series pattern feature extraction on the time series of air pollutants collected by the pollutant monitoring devices of each monitoring site to obtain a sequence of sulfur dioxide time-series correlation feature vectors;

[0085] A location encoding module 205, configured to encode the location information of each monitoring site using a location encoder to obtain a sequence of monitoring site location encoding vectors;

[0086] A feature vector sequence merging and information transfer reasoning module 206, configured to perform feature vector sequence merging and information transfer reasoning on the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time-series correlation feature vectors to obtain a pollution source spatial reasoning semantic feature vector;

[0087] A predicted pollution source location information determination module 207, configured to determine the predicted pollution source location information based on the pollution source spatial reasoning semantic feature vector.

[0088] In one embodiment of the present disclosure, the sulfur dioxide concentration time series pattern feature extraction module includes: a vector arrangement unit configured to arrange the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site along the time dimension respectively to obtain a sequence of sulfur dioxide time series input vectors; and a feature extraction unit configured to respectively pass each sulfur dioxide time series input vector in the sequence of sulfur dioxide time series input vectors through a sulfur dioxide concentration time series pattern feature extractor based on a one-dimensional convolutional layer to obtain a sequence of sulfur dioxide time series correlation feature vectors.

[0089] In one embodiment of the present disclosure, the feature vector sequence merging and information transfer inference module includes: a sequence merging unit configured to merge the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time series correlation feature vectors to obtain a sequence of sulfur dioxide time series correlation feature vectors containing location information; and a feature vector sequence transfer inference unit configured to pass the sequence of sulfur dioxide time series correlation feature vectors containing location information through a feature vector sequence transfer inference module to obtain the pollution source space inference semantic feature vector.

[0090] Reference is made below to Figure 3 , which shows a schematic structural diagram of an electronic device 600 suitable for implementing the embodiments of the present disclosure. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functions and usage scope of the embodiments of the present disclosure.

[0091] As Figure 3 shown, the electronic device 600 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 601, which may perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0092] Typically, the following devices can be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 can allow the electronic device 600 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 3 an electronic device 600 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. More or fewer devices can be alternatively implemented or had.

[0093] In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above functions defined in the methods of the embodiments of the present disclosure are performed.

[0094] It should be noted that the computer-readable medium described above in the present disclosure may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0095] In some embodiments, the client and the server can communicate using any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0096] The above computer-readable medium may be included in the above electronic device; or it may exist separately without being assembled into the electronic device.

[0097] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages or combinations thereof. The programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0098] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0099] The modules described in the embodiments of the present disclosure may be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the module itself in some cases. For example, the test parameter acquisition module may also be described as "the module for acquiring the device test parameters corresponding to the target device".

[0100] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, by way of non-limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGA), application specific integrated circuits (ASIC), application specific standard products (ASSP), system on a chip (SOC), complex programmable logic devices (CPLD), and so on.

[0101] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0102] Figure 4 is an application scenario diagram of an air pollutant monitoring method based on a GIS platform shown according to an exemplary embodiment. As Figure 4 shown, in this application scenario, first, a time series of air pollutants collected by pollutant monitoring devices deployed at each monitoring site is obtained (for example, C1 as illustrated in Figure 4 ); the location information of each monitoring site is obtained (for example, C2 as illustrated in Figure 4 ); then, the obtained time series of air pollutants and location information are input into a server (for example, S as illustrated in Figure 4 ) deployed with an air pollutant monitoring algorithm based on a GIS platform, where the server can process the time series of air pollutants and the location information based on the air pollutant monitoring algorithm of the GIS platform to determine the location information of the predicted pollution source.

[0103] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, technical solutions formed by mutually replacing the above features with (but not limited to) technical features having similar functions disclosed in the present disclosure.

[0104] Moreover, although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in a sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the foregoing discussion, these should not be construed as limitations on the scope of the present disclosure. Certain features that are described in the context of separate embodiments may also be implemented combinatorially in a single embodiment. Conversely, the various features that are described in the context of a single embodiment may also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0105] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims. With regard to the apparatus in the foregoing embodiments, the specific manner in which each module performs an operation has been described in detail in the embodiments related to the method and will not be elaborated herein.

Claims

1. An air pollutant monitoring method based on a GIS platform, characterized in that, Including: Obtain the time series of atmospheric pollutants collected by pollutant monitoring devices deployed at each monitoring site, where the atmospheric pollutant is the sulfur dioxide concentration value; Obtain the location information of each monitoring site; Integrate the location information of each monitoring site and the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site into a GIS database; Extract the time series pattern features of sulfur dioxide concentration from the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors; Use a position encoder to encode the location information of each monitoring site to obtain a sequence of monitoring site location encoding vectors; Perform feature vector sequence merging and information transfer reasoning on the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time series correlation feature vectors to obtain a pollution source spatial reasoning semantic feature vector; Based on the pollution source spatial reasoning semantic feature vector, determine the location information of the predicted pollution source.

2. The method for monitoring air pollutants based on the GIS platform according to claim 1, wherein Extract the time series pattern features of sulfur dioxide concentration from the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors, including: Arrange the time series of atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site respectively according to the time dimension to obtain a sequence of sulfur dioxide time series input vectors; Pass each sulfur dioxide time series input vector in the sequence of sulfur dioxide time series input vectors through a sulfur dioxide concentration time series pattern feature extractor based on a one-dimensional convolutional layer to obtain the sequence of sulfur dioxide time series correlation feature vectors.

3. The method for monitoring atmospheric pollutants based on the GIS platform according to claim 2, wherein Perform feature vector sequence merging and information transfer reasoning on the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time series correlation feature vectors to obtain a pollution source spatial reasoning semantic feature vector, including: Merge the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time series correlation feature vectors to obtain a sequence of sulfur dioxide time series correlation feature vectors containing location information; Pass the sequence of sulfur dioxide time series correlation feature vectors containing location information through a feature vector sequence transfer reasoning module to obtain the pollution source spatial reasoning semantic feature vector.

4. The method for monitoring air pollutants based on the GIS platform according to claim 3, characterized in that, Pass the sequence of sulfur dioxide time series correlation feature vectors containing location information through a feature vector sequence transfer reasoning module to obtain the pollution source spatial reasoning semantic feature vector, including: Calculate the weighted sum of the first s sulfur dioxide time series correlation feature vectors containing location information in the sequence of sulfur dioxide time series correlation feature vectors containing location information to obtain a neighbor semantic fusion feature vector; Perform weighted processing on each location feature value in the s-th sulfur dioxide time series correlation feature vector containing location information using the sum of a trainable preset hyperparameter and one as the weighting coefficient to obtain a weighted optimized sulfur dioxide time series correlation feature vector containing location information; The neighbor semantic fusion feature vector and the weighted and optimized sulfur dioxide time-series correlation feature vector including position information are summed by position and then processed by a multi-layer perceptron to obtain the pollution source inference semantic feature vector corresponding to the s-th sulfur dioxide time-series correlation feature vector including position information; Calculate the sum by position between the pollution source inference semantic feature vectors corresponding to each sulfur dioxide time-series correlation feature vector including position information in the sequence of sulfur dioxide time-series correlation feature vectors including position information to obtain the pollution source spatial inference semantic feature vector.

5. The method for monitoring air pollutants based on the GIS platform according to claim 4, characterized in that, Based on the pollution source spatial inference semantic feature vector, determine the position information of the predicted pollution source, including: passing the pollution source spatial inference semantic feature vector through an inference device based on a decoder to obtain a decoding result, and the decoding result is the position information of the predicted pollution source.

6. The method for monitoring air pollutants based on the GIS platform according to claim 5, wherein It also includes a training step: used to train the sulfur dioxide concentration time-series pattern feature extractor based on a one-dimensional convolutional layer, the position encoder, the feature vector sequence transfer inference module, and the inference device based on a decoder.

7. The method for monitoring air pollutants based on the GIS platform according to claim 6, characterized in that, The training step includes: Obtain the time series of training atmospheric pollutants collected by pollutant monitoring devices deployed at each monitoring site, where the training atmospheric pollutant is the training sulfur dioxide concentration value; Obtain the training position information of each monitoring site; Integrate the training position information of each monitoring site and the time series of training atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site into a GIS database; Extract the sulfur dioxide concentration time-series pattern features from the time series of training atmospheric pollutants collected by the pollutant monitoring devices at each monitoring site to obtain a sequence of training sulfur dioxide time-series correlation feature vectors; Use a position encoder to encode the training position information of each monitoring site to obtain a sequence of training monitoring site position encoding vectors; Perform feature vector sequence merging and information transfer inference on the sequence of training monitoring site position encoding vectors and the sequence of training sulfur dioxide time-series correlation feature vectors to obtain a training pollution source spatial inference semantic feature vector; Perform feature optimization on the training pollution source spatial inference semantic feature vector to obtain an optimized training pollution source spatial inference semantic feature vector; Pass the optimized training pollution source spatial inference semantic feature vector through the inference device based on a decoder to obtain a decoding loss function value; Based on the decoding loss function value, train the sulfur dioxide concentration time-series pattern feature extractor based on a one-dimensional convolutional layer, the position encoder, the feature vector sequence transfer inference module, and the inference device based on a decoder.

8. An air pollutant monitoring system based on a GIS platform, characterized in that, It includes: An atmospheric pollutant acquisition module, used to obtain the time series of atmospheric pollutants collected by pollutant monitoring devices deployed at each monitoring site, where the atmospheric pollutant is the sulfur dioxide concentration value; A position information acquisition module, used to obtain the position information of each monitoring site; GIS database integration module, which is used to integrate the location information of each monitoring site and the time series of atmospheric pollutants collected by the pollutant monitoring equipment at each monitoring site into the GIS database; Sulfur dioxide concentration time series pattern feature extraction module, which is used to extract the sulfur dioxide concentration time series pattern features from the time series of atmospheric pollutants collected by the pollutant monitoring equipment at each monitoring site to obtain a sequence of sulfur dioxide time series correlation feature vectors; Location encoding module, which is used to encode the location information of each monitoring site using a location encoder to obtain a sequence of monitoring site location encoding vectors; Feature vector sequence merging and information transfer inference module, which is used to perform feature vector sequence merging and information transfer inference on the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time series correlation feature vectors to obtain a pollution source spatial inference semantic feature vector; Predicted pollution source location information determination module, which is used to determine the location information of the predicted pollution source based on the pollution source spatial inference semantic feature vector.

9. The air pollutant monitoring system based on the GIS platform according to claim 8, characterized in that, The sulfur dioxide concentration time series pattern feature extraction module includes: Vector arrangement unit, which is used to arrange the time series of atmospheric pollutants collected by the pollutant monitoring equipment at each monitoring site respectively according to the time dimension to obtain a sequence of sulfur dioxide time series input vectors; Feature extraction unit, which is used to pass each sulfur dioxide time series input vector in the sequence of sulfur dioxide time series input vectors through a sulfur dioxide concentration time series pattern feature extractor based on a one-dimensional convolutional layer to obtain a sequence of sulfur dioxide time series correlation feature vectors.

10. The air pollutant monitoring system based on the GIS platform according to claim 9, characterized in that, The feature vector sequence merging and information transfer inference module includes: Sequence merging unit, which is used to perform feature vector sequence merging on the sequence of monitoring site location encoding vectors and the sequence of sulfur dioxide time series correlation feature vectors to obtain a sequence of sulfur dioxide time series correlation feature vectors containing location information; Feature vector sequence transfer inference unit, which is used to pass the sequence of sulfur dioxide time series correlation feature vectors containing location information through a feature vector sequence transfer inference module to obtain the pollution source spatial inference semantic feature vector.

Images