A method for processing atmospheric pollutant emission data based on time series graph neural network
By using a data processing method based on time-series graph neural networks, atmospheric pollutant emission data is analyzed, solving the problem that existing technologies cannot accurately determine the trend of pollutant changes, and realizing accurate judgment and timely response to the trend of pollutant changes.
Patent Information
- Application Number
- CN202411146785.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-21
AI Technical Summary
Existing technologies fail to effectively utilize time-series neural networks to analyze pollutant change trends when analyzing air pollutant emission data, resulting in an inability to accurately determine the changes in pollutants.
An atmospheric pollutant emission data processing method based on time-series graph neural networks is adopted. By collecting atmospheric emission data at multiple time points in the target space, and combining the data with time-series graph neural networks to analyze pollutant emission data, the similarity between the current cycle and the pollution situation in the same cycle of the previous year is determined. The user terminal receives data similarity or abnormality signals and then takes action.
It enables accurate judgment of pollutant change trends, and user terminals can perform corresponding operations based on the signals, improving the accuracy and timeliness of pollutant change analysis.
Smart Images

Figure CN118656768B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of data processing, and in particular is a method for processing atmospheric pollutant emission data based on a time series graph neural network. Background Art
[0002] In order to effectively manage and reduce air pollution, it is crucial to monitor and analyze the emission data of air pollutants. Traditional analysis methods often use statistical models or regression models to model and predict air pollutant emission data; however, these data usually exist in the form of time series data, including pollutant emissions at different time points, so it is necessary to analyze pollutant emission data through time series graph neural networks.
[0003] At present, when analyzing atmospheric pollutant emission data, usually only the impact of pollutant concentration on the environment is analyzed, and the change of pollutants is not analyzed based on the time series graph neural network, so it is impossible to accurately judge the change trend of pollutants;
[0004] To this end, we propose a method for processing atmospheric pollutant emission data based on temporal graph neural networks. Summary of the invention
[0005] The purpose of the present invention is to provide a method for processing atmospheric pollutant emission data based on a time series graph neural network to solve the problems raised in the above background technology.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] A method for processing air pollutant emission data based on a time series graph neural network. The method for processing air pollutant emission data is as follows:
[0008] Step S1, collecting atmospheric emission data sets at multiple time points in the target space;
[0009] Step S2, preliminarily screening the atmospheric pollutant emission data in the atmospheric emission data set corresponding to the target space to obtain preliminary screening pollutant emission data;
[0010] Step S3, initially screening the pollutant emission data and combining it with the time series graph neural network to determine the similarity between the air pollution situation in the current collection period of the target space and the air pollution situation corresponding to the same period of the previous year;
[0011] Step S4: the user terminal receives a data similarity signal or a data anomaly signal and performs corresponding operations according to the signal.
[0012] Furthermore, the process of setting the atmospheric emission data set in step S1 is specifically as follows:
[0013] Step S11, comparing the coverage range of the target space with the standard range, if the coverage range of the target space is greater than the first standard range, setting a first preset number of pollution data collection points for the target space;
[0014] Step S12: if the coverage range of the target space is less than or equal to the first standard range and greater than the second standard range, a second preset number of pollution data collection points is set for the target space;
[0015] Step S13, if the coverage range of the target space is less than or equal to the second standard range, a third preset number of pollution data collection points is set for the target space; wherein the first preset number is greater than the second preset number, the second preset number is greater than the third preset number; and the first standard range is greater than the second standard range;
[0016] Step S14, install corresponding air sensors at positions corresponding to pollution data collection points in the target space, and start the air sensors according to the preset collection cycle to obtain atmospheric pollutant emission data corresponding to multiple time points within the collection cycle, and summarize and integrate the atmospheric pollutant emission data at positions corresponding to different pollution data collection points at the same time point to obtain an atmospheric emission data set.
[0017] Furthermore, the atmospheric emission data set at each time point includes atmospheric pollutant emission data corresponding to multiple pollution data collection points, and the atmospheric pollutant emission data of each pollution data collection point includes the sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration at the location of the corresponding pollution data collection point.
[0018] Furthermore, the step S2 includes the following sub-steps:
[0019] Step S21, acquiring atmospheric pollutant emission data of multiple pollution data collection points in the atmospheric emission data set, and obtaining sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration of each pollution data collection point in the atmospheric emission data set;
[0020] Step S22, analyzing all the atmospheric pollutant emission data in the atmospheric emission data set, defining the pollution data collection point currently being analyzed as a selected collection point, and calculating the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value, and particulate matter concentration difference value between the selected collection point and adjacent collection points;
[0021] Step S23, comparing the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value of the selected collection points with corresponding thresholds respectively;
[0022] Step S24: if any one of the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value corresponding to the selected collection point is greater than the corresponding threshold value, the air pollution emission data corresponding to the selected collection point is discarded, and an abnormal signal is generated and sent to the user terminal;
[0023] If the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value of the selected collection point are all less than the corresponding threshold value, the atmospheric pollution emission data of the corresponding selected collection point will be retained without any operation;
[0024] Step S25, executing steps S22 to S24 for the remaining data collection nodes in the same atmospheric emission data set, recording the pollution data collection points retained in the same atmospheric emission data set as preliminary screening collection points, and recording the atmospheric pollution emission data corresponding to the preliminary screening collection points as preliminary screening pollutant emission data;
[0025] After the initial screening of all data collection nodes in the atmospheric emission data set corresponding to the current time point is completed, the initial screening of the data collection nodes in the next atmospheric emission data set is completed according to the order of the time points.
[0026] Furthermore, the step S22 further includes the following sub-steps:
[0027] Step S221, drawing a radiation area corresponding to the selected collection point with the actual location of the selected collection point as the center and the estimated radiation length as the radius;
[0028] Step S222: If there is one pollution data collection point or no pollution data collection point in the radiation area, the selected collection point is not included in the calculation, and the pollution data collection point closest to the selected collection point is recorded as a neighboring collection point. The sulfur dioxide concentration difference between the corresponding selected collection point and the neighboring collection point is calculated by the formula. The specific formula is as follows:
[0029] ELC=λ×|ELX-ELL|; where ELX is the sulfur dioxide concentration value of the selected collection point, ELL is the sulfur dioxide concentration value corresponding to the adjacent collection point, and λ is the distance correction coefficient. If the adjacent collection point is outside the radiation area, 0<λ<1; if the adjacent collection point is within the radiation area, 1≤λ<1.7. The value of λ is inversely proportional to the distance between the selected collection point and the adjacent collection point, that is, the greater the distance between the selected collection point and the adjacent collection point, the smaller the value of λ.
[0030] Similarly, the nitrogen oxide concentration difference between the selected collection point and the adjacent collection point is calculated by the formula DYC=λ×|DYX-DYL|, and the particulate matter concentration difference between the selected collection point and the adjacent collection point is calculated by the formula KLC=λ×|KLX-KLL|; where DYX is the nitrogen oxide concentration value corresponding to the selected collection point, DYL is the nitrogen oxide concentration value corresponding to the adjacent collection point, KLX is the particulate matter concentration value corresponding to the selected collection point, and KLL is the particulate matter concentration corresponding to the adjacent collection point;
[0031] Step S223: if there is more than one contaminated data collection point in the radiation area, the selected collection point is not included in it, then the radiation area is divided into four radiation partitions, and if there is a contaminated data collection point in the radiation partition, the contaminated data collection point in the radiation partition and the closest to the selected collection point is selected as the adjacent collection point.
[0032] Furthermore, the step S223 further includes the following sub-steps:
[0033] Step S2231, if there is only one adjacent collection point, the sulfur dioxide concentration difference value is obtained by subtracting the sulfur dioxide concentration corresponding to the adjacent collection point from the sulfur dioxide concentration corresponding to the selected collection point and taking the absolute value. Similarly, the nitrogen oxide concentration difference value is obtained by subtracting the nitrogen oxide concentration corresponding to the adjacent collection point from the nitrogen oxide concentration corresponding to the selected collection point and taking the absolute value. The corresponding particulate matter concentration difference value is obtained by subtracting the particulate matter concentration corresponding to the adjacent collection point from the particulate matter concentration corresponding to the selected collection point and taking the absolute value.
[0034] Step S2232: if there are two or more adjacent collection points, the sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration of the multiple adjacent collection points are added and averaged to obtain the average sulfur dioxide concentration, average nitrogen oxide concentration and average particulate matter concentration of the adjacent collection points;
[0035] Step S2233, the sulfur dioxide concentration corresponding to the selected collection point is subtracted from the average sulfur dioxide concentration of the adjacent collection points to obtain the corresponding sulfur dioxide difference value;
[0036] Similarly, the nitrogen oxide concentration corresponding to the collection point is subtracted from the average nitrogen oxide concentration of the adjacent collection points and the corresponding nitrogen oxide difference value is obtained by taking the absolute value; the corresponding particulate matter difference value is obtained by subtracting the average particulate matter concentration of the adjacent collection points from the particulate matter concentration corresponding to the collection point and the absolute value.
[0037] Furthermore, the step S3 includes the following sub-steps:
[0038] Step S31, obtaining the primary screening pollutant emission data collected by the primary screening collection points in the atmospheric emission data set corresponding to different time points, and then forming a real-time sulfur dioxide sequence with the sulfur dioxide concentrations collected by different primary screening collection points in the same atmospheric emission data set. Similarly, forming a real-time nitrogen oxide sequence with the nitrogen oxide concentrations collected by different primary screening collection points in the same atmospheric emission data set, and forming a real-time particulate matter sequence with the particulate matter concentrations collected by different primary screening collection points in the same atmospheric emission data set;
[0039] Step S32, obtaining multiple groups of historical sulfur dioxide sequences, multiple groups of historical nitrogen oxide sequences, and multiple groups of historical particulate matter sequences corresponding to the current time point;
[0040] Step S33, the sulfur dioxide concentration collected by each preliminary screening collection point in the real-time sulfur dioxide sequence corresponding to the current time point is regarded as a real-time node, and the sulfur dioxide concentration collected by each preliminary screening collection point in multiple groups of historical sulfur dioxide sequences corresponding to the current time point is regarded as a historical node.
[0041] Furthermore, the step S3 further includes the following sub-steps:
[0042] Step S34, sequentially calculating the sulfur dioxide node similarity between each real-time node in the real-time sulfur dioxide sequence corresponding to the current time point and the corresponding historical node in the historical sulfur dioxide sequence;
[0043] Step S35, adding and averaging the sulfur dioxide node similarities between each real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence, to obtain the sulfur dioxide sequence similarity between the real-time sulfur dioxide sequence and the corresponding historical sulfur dioxide sequence at the current time point;
[0044] Similarly, the sulfur dioxide sequence similarity between the real-time sulfur dioxide sequence and the corresponding historical sulfur dioxide sequence at different time points can be obtained;
[0045] Step S36, repeatedly performing the above steps to obtain the nitrogen oxide sequence similarity between the real-time nitrogen oxide sequence at different time points and the corresponding historical nitrogen oxide sequence, and to obtain the particle matter sequence similarity between the real-time particulate matter sequence at different time points and the corresponding historical particulate matter sequence;
[0046] Step S37, if the sulfur dioxide sequence similarity of the real-time sulfur dioxide sequence, the nitrogen oxide sequence similarity of the real-time nitrogen oxide sequence, and the particle matter sequence similarity of the real-time particle matter sequence at all time points are less than or equal to the corresponding preset similarity, a data similarity signal is generated and sent to the user terminal;
[0047] Step S38: If the sulfur dioxide sequence similarity of any real-time sulfur dioxide sequence, the nitrogen oxide sequence similarity of any real-time nitrogen oxide sequence, or the particulate matter sequence similarity of any real-time particulate matter sequence at any time point is greater than the corresponding preset similarity, a data anomaly signal is generated and sent to the user terminal.
[0048] Furthermore, the similarity distance analysis process in step S34 includes the following sub-steps:
[0049] Step S341, comparing the sulfur dioxide concentration corresponding to the real-time node in the real-time sulfur dioxide sequence with the sulfur dioxide concentration corresponding to the historical node in the multiple groups of historical sulfur dioxide sequences;
[0050] Step S342, calculating the difference between the sulfur dioxide concentration difference corresponding to the real-time node and the sulfur dioxide concentration difference corresponding to multiple historical nodes and taking the absolute value to obtain multiple groups of sulfur dioxide concentration difference values;
[0051] Step S343, if there is a mode among the multiple groups of sulfur dioxide concentration difference values, the sulfur dioxide concentration difference value corresponding to the mode is used as the similarity distance between the real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence;
[0052] Step S344: if there is no mode among the multiple groups of sulfur dioxide concentration differences, then the multiple groups of sulfur dioxide concentration differences are added and averaged to obtain an average sulfur dioxide concentration difference, and the average sulfur dioxide concentration difference is used as the similarity distance between the real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence;
[0053] Step S345, repeat the above steps to obtain the sulfur dioxide node similarity between the remaining real-time nodes in the real-time sulfur dioxide sequence and the corresponding historical nodes in the historical sulfur dioxide sequence at the same current time point.
[0054] Furthermore, the process of performing corresponding operations according to the signal in step S4 is specifically as follows:
[0055] If a data similarity signal is received, no action is performed;
[0056] If a data anomaly signal is received, staff will be assigned to conduct a field investigation.
[0057] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:
[0058] The present invention first obtains an atmospheric emission data set at multiple time points in a target space, and preliminarily screens the atmospheric pollutant emission data in the atmospheric emission data set corresponding to the target space to obtain preliminary screened pollutant emission data, and then combines the preliminary screened pollutant emission data with a time series graph neural network to determine the similarity between the atmospheric pollution conditions in the current acquisition period of the target space and the atmospheric pollution conditions corresponding to the same period of the previous year. The user terminal performs corresponding operations based on the received data similarity signal or data abnormality signal. The present invention analyzes the changes in pollutants, thereby accurately determining the changing trends of pollutants. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.
[0060] Figure 1 is a flow chart of the method of the present invention;
[0061] Figure 2 It is a structural schematic diagram of selecting collection points in the present invention;
[0062] Figure 3 It is another structural schematic diagram of selecting collection points in the present invention;
[0063] Figure 4 It is a schematic diagram of the structure of the computer device in the present invention. DETAILED DESCRIPTION
[0064] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0065] Example 1: Please refer to Figure 1-Figure 3 As shown, the technical solution provided by the present invention is: a method for processing atmospheric pollutant emission data based on a time series graph neural network, comprising:
[0066] Step S1, obtaining atmospheric emission data sets at multiple time points in the target space;
[0067] The coverage of the target space is set by the user, and different numbers of pollution data collection points are set according to the coverage of the target space;
[0068] In step S1, the process of setting the atmospheric emission data set is as follows:
[0069] Step S11, comparing the coverage range of the target space with the standard range, if the coverage range of the target space is greater than the first standard range, setting a first preset number of pollution data collection points for the target space;
[0070] Among them, the coverage range and standard range of the target space can be set according to the previous atmospheric environment monitoring range;
[0071] Step S12: if the coverage range of the target space is less than or equal to the first standard range and greater than the second standard range, a second preset number of pollution data collection points is set for the target space;
[0072] Step S13, if the coverage range of the target space is less than or equal to the second standard range, a third preset number of pollution data collection points is set for the target space; wherein the first preset number is greater than the second preset number, the second preset number is greater than the third preset number; and the first standard range is greater than the second standard range;
[0073] Step S14, installing corresponding air sensors at positions corresponding to pollution data collection points in the target space, and starting the air sensors according to a preset collection cycle, obtaining atmospheric pollutant emission data corresponding to multiple time points within the collection cycle, and summarizing and integrating atmospheric pollutant emission data corresponding to positions of different pollution data collection points at the same time point to obtain an atmospheric emission data set;
[0074] It should be emphasized that the atmospheric emission data set at each time point contains atmospheric pollutant emission data corresponding to multiple pollution data collection points. The atmospheric pollutant emission data of each pollution data collection point specifically includes the sulfur dioxide concentration, nitrogen oxide concentration, particulate matter (PM2.5 and PM10) concentration and volatile organic compound concentration at the location of the corresponding pollution data collection point;
[0075] Step S2, preliminarily screening the atmospheric pollutant emission data in the atmospheric emission data set corresponding to the target space to obtain preliminary screening pollutant emission data;
[0076] The step S2 further comprises the following sub-steps:
[0077] Step S21, acquiring atmospheric pollutant emission data of multiple pollution data collection points in the atmospheric emission data set, and obtaining sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration of each pollution data collection point in the atmospheric emission data set;
[0078] Step S22, see Figure 2-Figure 3 , analyze all the atmospheric pollutant emission data in the atmospheric emission data set, define the pollution data collection point currently being analyzed as the selected collection point, and calculate the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value between the selected collection point and the adjacent collection points;
[0079] The step S22 further includes the following sub-steps:
[0080] Step S221, drawing a radiation area corresponding to the selected collection point with the actual location of the selected collection point as the center and the estimated radiation length as the radius;
[0081] Step S222, as Figure 2 As shown in the figure, if there is one pollution data collection point or no pollution data collection point in the radiation area, the selected collection point is not included in it, then the pollution data collection point closest to the selected collection point is recorded as the adjacent collection point, and the sulfur dioxide concentration difference between the corresponding selected collection point and the adjacent collection point is calculated by the formula. The specific formula is as follows:
[0082] ELC=λ×|ELX-ELL|; where ELX is the sulfur dioxide concentration value of the selected collection point, ELL is the sulfur dioxide concentration value corresponding to the adjacent collection point, and λ is the distance correction coefficient. If the adjacent collection point is outside the radiation area, 0<λ<1; if the adjacent collection point is within the radiation area, 1≤λ<1.7. The value of λ is inversely proportional to the distance between the selected collection point and the adjacent collection point, that is, the greater the distance between the selected collection point and the adjacent collection point, the smaller the value of λ.
[0083] Similarly, the nitrogen oxide concentration difference between the selected collection point and the adjacent collection point is calculated by the formula DYC=λ×|DYX-DYL|, and the particulate matter concentration difference between the selected collection point and the adjacent collection point is calculated by the formula KLC=λ×|KLX-KLL|; where DYX is the nitrogen oxide concentration value corresponding to the selected collection point, DYL is the nitrogen oxide concentration value corresponding to the adjacent collection point, KLX is the particulate matter concentration value corresponding to the selected collection point, and KLL is the particulate matter concentration corresponding to the adjacent collection point;
[0084] Step S223, as Figure 3 As shown, if there is more than one contaminated data collection point in the radiation area, the selected collection point is not included in it, then the radiation area is equally divided into four radiation partitions at 90°, if there is a contaminated data collection point in the radiation partition, then the contaminated data collection point in the radiation partition and the closest distance to the selected collection point is selected as the adjacent collection point; it should be noted that the number of adjacent collection points may be one or more, so the number of adjacent collection points is within the interval [1, R], and R is a positive integer;
[0085] Step S2231, if there is only one adjacent collection point, the sulfur dioxide concentration difference value is obtained by directly selecting the sulfur dioxide concentration corresponding to the collection point and subtracting the sulfur dioxide concentration corresponding to the adjacent collection point and taking the absolute value. Similarly, the nitrogen oxide concentration difference value is obtained by selecting the nitrogen oxide concentration corresponding to the collection point and subtracting the nitrogen oxide concentration corresponding to the adjacent collection point and taking the absolute value. The corresponding particulate matter concentration difference value is obtained by selecting the particulate matter concentration corresponding to the collection point and subtracting the particulate matter concentration corresponding to the adjacent collection point and taking the absolute value.
[0086] Step S2232: if there are two or more adjacent collection points, the sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration of the multiple adjacent collection points are added and averaged to obtain the average sulfur dioxide concentration, average nitrogen oxide concentration and average particulate matter concentration of the adjacent collection points;
[0087] Step S2233, select the sulfur dioxide concentration corresponding to the collection point and subtract the average sulfur dioxide concentration of the adjacent collection points to obtain the corresponding sulfur dioxide difference value; similarly, select the nitrogen oxide concentration corresponding to the collection point and subtract the average nitrogen oxide concentration of the adjacent collection points to obtain the corresponding nitrogen oxide difference value; select the particulate matter concentration corresponding to the collection point and subtract the average particulate matter concentration of the adjacent collection points to obtain the corresponding particulate matter difference value;
[0088] Step S23, comparing the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value of the selected collection points with corresponding thresholds respectively;
[0089] Step S24: if any one of the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value corresponding to the selected collection point is greater than the corresponding threshold value, the air pollution emission data corresponding to the selected collection point is discarded, and an abnormal signal is generated and sent to the user terminal;
[0090] If the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value of the selected collection point are all less than the corresponding threshold value, the atmospheric pollution emission data of the corresponding selected collection point will be retained without any operation;
[0091] Step S25, executing steps S22 to S24 for the remaining data collection nodes in the same atmospheric emission data set, recording the pollution data collection points retained in the same atmospheric emission data set as preliminary screening collection points, and recording the atmospheric pollution emission data corresponding to the preliminary screening collection points as preliminary screening pollutant emission data;
[0092] After the initial screening of all data collection nodes in the atmospheric emission data set corresponding to the current time point is completed, the initial screening of the data collection nodes in the next atmospheric emission data set is completed according to the order of the time points; one time point corresponds to a group of atmospheric emission data sets, and one atmospheric emission data set contains multiple initial screening collection points;
[0093] Step S3, initially screening the pollutant emission data and combining it with the time series graph neural network to determine the similarity between the air pollution situation in the current collection period of the target space and the air pollution situation corresponding to the same period of the previous year;
[0094] The step S3 also includes the following sub-steps:
[0095] Step S31, obtaining the primary screening pollutant emission data collected by the primary screening collection points in the atmospheric emission data set corresponding to different time points, and then forming a real-time sulfur dioxide sequence with the sulfur dioxide concentrations collected by different primary screening collection points in the same atmospheric emission data set. Similarly, forming a real-time nitrogen oxide sequence with the nitrogen oxide concentrations collected by different primary screening collection points in the same atmospheric emission data set, and forming a real-time particulate matter sequence with the particulate matter concentrations collected by different primary screening collection points in the same atmospheric emission data set;
[0096] Step S32, obtaining multiple groups of historical sulfur dioxide sequences, multiple groups of historical nitrogen oxide sequences, and multiple groups of historical particulate matter sequences corresponding to the current time point;
[0097] Among them, the historical sulfur dioxide series, historical nitrogen oxide series and historical particulate matter series can be atmospheric pollutant emission data collected at the same time point in the past few weeks and obtained after data analysis and data screening;
[0098] Step S33, treating the sulfur dioxide concentration collected at each preliminary screening collection point in the real-time sulfur dioxide sequence corresponding to the current time point as a real-time node, and treating the sulfur dioxide concentration collected at each preliminary screening collection point in the multiple groups of historical sulfur dioxide sequences corresponding to the current time point as a historical node;
[0099] Step S34, sequentially calculating the sulfur dioxide node similarity between each real-time node in the real-time sulfur dioxide sequence corresponding to the current time point and the corresponding historical node in the historical sulfur dioxide sequence;
[0100] The similarity distance analysis process in step S34 includes the following sub-steps:
[0101] Step S341, comparing the sulfur dioxide concentration corresponding to the real-time node in the real-time sulfur dioxide sequence with the sulfur dioxide concentration corresponding to the historical node in the multiple groups of historical sulfur dioxide sequences;
[0102] Step S342, calculating the difference between the sulfur dioxide concentration difference corresponding to the real-time node and the sulfur dioxide concentration difference corresponding to multiple historical nodes and taking the absolute value to obtain multiple groups of sulfur dioxide concentration difference values;
[0103] Step S343, if there is a mode among the multiple groups of sulfur dioxide concentration difference values, the sulfur dioxide concentration difference value corresponding to the mode is used as the similarity distance between the real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence;
[0104] Step S344: if there is no mode among the multiple groups of sulfur dioxide concentration differences, then the multiple groups of sulfur dioxide concentration differences are added and averaged to obtain an average sulfur dioxide concentration difference, and the average sulfur dioxide concentration difference is used as the similarity distance between the real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence;
[0105] Step S345, repeatedly executing the above steps S341 to S344, obtaining the sulfur dioxide node similarity between the remaining real-time nodes in the real-time sulfur dioxide sequence and the corresponding historical nodes in the historical sulfur dioxide sequence at the same current time point;
[0106] Step S35, adding and averaging the sulfur dioxide node similarities between each real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence, to obtain the sulfur dioxide sequence similarity between the real-time sulfur dioxide sequence and the corresponding historical sulfur dioxide sequence at the current time point;
[0107] Similarly, the sulfur dioxide sequence similarity between the real-time sulfur dioxide sequence and the corresponding historical sulfur dioxide sequence at different time points can be obtained;
[0108] Step S36, according to steps S31 to S35, obtaining the nitrogen oxide sequence similarity between the real-time nitrogen oxide sequence at different time points and the corresponding historical nitrogen oxide sequence, and obtaining the particle matter sequence similarity between the real-time particulate matter sequence at different time points and the corresponding historical particulate matter sequence;
[0109] Step S37, if the sulfur dioxide sequence similarity of the real-time sulfur dioxide sequence, the nitrogen oxide sequence similarity of the real-time nitrogen oxide sequence, and the particle matter sequence similarity of the real-time particle matter sequence at all time points are less than or equal to the corresponding preset similarity, a data similarity signal is generated and sent to the user terminal;
[0110] Step S38, if the sulfur dioxide sequence similarity of any real-time sulfur dioxide sequence, the nitrogen oxide sequence similarity of any real-time nitrogen oxide sequence, or the particulate matter sequence similarity of any real-time particulate matter sequence at any time point is greater than the corresponding preset similarity, a data anomaly signal is generated and sent to the user terminal;
[0111] Step S4: the user terminal receives a data similarity signal or a data anomaly signal and performs corresponding operations according to the signal, specifically:
[0112] If a data similarity signal is received, no operation is performed, which means that the air pollution situation in the current collection period of the user is similar to that of the previous year;
[0113] If a data anomaly signal is received, staff will be assigned to conduct a field investigation, which means that the air pollution situation in the user's current collection period is significantly different from that in the previous year;
[0114] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.
[0115] Embodiment 2: Figure 4 The structural diagram of a computer device is illustrated, and the computer device may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The processor may call the logic instructions in the memory to execute a method for processing atmospheric pollutant emission data based on a time-series graph neural network, and the method includes: obtaining an atmospheric emission data set at multiple time points in a target space; performing a preliminary screening of atmospheric pollutant emission data in the atmospheric emission data set corresponding to the target space to obtain preliminary screening pollutant emission data; combining the preliminary screening pollutant emission data with the time-series graph neural network to determine the similarity between the atmospheric pollution situation in the current acquisition period of the target space and the atmospheric pollution situation corresponding to the same period of the previous year; the user terminal receives a data similarity signal or a data abnormality signal, and performs corresponding operations according to the signal.
[0116] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.
[0117] On the other hand, the present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute a method for processing atmospheric pollutant emission data based on a time series graph neural network provided by the above methods, the method including: obtaining an atmospheric emission data set at multiple time points in a target space; performing a preliminary screening of the atmospheric pollutant emission data in the atmospheric emission data set corresponding to the target space to obtain preliminary screened pollutant emission data; combining the preliminary screened pollutant emission data with the time series graph neural network to determine the similarity between the atmospheric pollution conditions in the target space during the current collection period and the atmospheric pollution conditions corresponding to the same period in the previous year; the user terminal receives a data similarity signal or a data abnormality signal, and performs corresponding operations based on the signal.
[0118] On the other hand, the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the above-mentioned method for processing atmospheric pollutant emission data based on a time series graph neural network, the method comprising: obtaining an atmospheric emission data set at multiple time points in a target space; performing a preliminary screening of the atmospheric pollutant emission data in the atmospheric emission data set corresponding to the target space to obtain preliminary screened pollutant emission data; combining the preliminary screened pollutant emission data with the time series graph neural network to determine the similarity between the atmospheric pollution conditions in the target space during the current collection period and the atmospheric pollution conditions corresponding to the same period in the previous year; the user terminal receives a data similarity signal or a data abnormality signal, and performs corresponding operations based on the signal.
[0119] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0120] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A method for processing atmospheric pollutant emission data based on a time series graph neural network, characterized in that: The specific method for processing air pollutant emission data is as follows: Step S1, obtaining atmospheric emission data sets at multiple time points in the target space; The atmospheric emission data set at each time point includes atmospheric pollutant emission data corresponding to multiple pollution data collection points, and the atmospheric pollutant emission data of each pollution data collection point includes the sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration at the location of the corresponding pollution data collection point; Step S2, preliminarily screening the atmospheric pollutant emission data in the atmospheric emission data set corresponding to the target space to obtain preliminary screening pollutant emission data; Step S3, initially screening the pollutant emission data and combining it with the time series graph neural network to determine the similarity between the air pollution situation in the current collection period of the target space and the air pollution situation corresponding to the same period of the previous year; Step S4, the user terminal receives the data similarity signal or the data abnormality signal, and performs corresponding operations according to the signal; The step S3 includes the following sub-steps: Step S31, obtaining the primary screening pollutant emission data collected by the primary screening collection points in the atmospheric emission data set corresponding to different time points, and then forming a real-time sulfur dioxide sequence with the sulfur dioxide concentrations collected by different primary screening collection points in the same atmospheric emission data set. Similarly, forming a real-time nitrogen oxide sequence with the nitrogen oxide concentrations collected by different primary screening collection points in the same atmospheric emission data set, and forming a real-time particulate matter sequence with the particulate matter concentrations collected by different primary screening collection points in the same atmospheric emission data set; Step S32, obtaining multiple groups of historical sulfur dioxide sequences, multiple groups of historical nitrogen oxide sequences, and multiple groups of historical particulate matter sequences corresponding to the current time point; Step S33, treating the sulfur dioxide concentration collected at each preliminary screening collection point in the real-time sulfur dioxide sequence corresponding to the current time point as a real-time node, and treating the sulfur dioxide concentration collected at each preliminary screening collection point in the multiple groups of historical sulfur dioxide sequences corresponding to the current time point as a historical node; Step S34, sequentially calculating the sulfur dioxide node similarity between each real-time node in the real-time sulfur dioxide sequence corresponding to the current time point and the corresponding historical node in the historical sulfur dioxide sequence; Step S35, adding and averaging the sulfur dioxide node similarities between each real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence, to obtain the sulfur dioxide sequence similarity between the real-time sulfur dioxide sequence and the corresponding historical sulfur dioxide sequence at the current time point; Similarly, the sulfur dioxide sequence similarity between the real-time sulfur dioxide sequence and the corresponding historical sulfur dioxide sequence at different time points can be obtained; Step S36, repeatedly performing the above steps to obtain the nitrogen oxide sequence similarity between the real-time nitrogen oxide sequence at different time points and the corresponding historical nitrogen oxide sequence, and to obtain the particle matter sequence similarity between the real-time particulate matter sequence at different time points and the corresponding historical particulate matter sequence; Step S37, if the sulfur dioxide sequence similarity of the real-time sulfur dioxide sequence, the nitrogen oxide sequence similarity of the real-time nitrogen oxide sequence, and the particle matter sequence similarity of the real-time particle matter sequence at all time points are less than or equal to the corresponding preset similarity, a data similarity signal is generated and sent to the user terminal; Step S38, if the sulfur dioxide sequence similarity of any real-time sulfur dioxide sequence, the nitrogen oxide sequence similarity of any real-time nitrogen oxide sequence, or the particulate matter sequence similarity of any real-time particulate matter sequence at any time point is greater than the corresponding preset similarity, a data anomaly signal is generated and sent to the user terminal; The similarity distance analysis process in step S34 includes the following sub-steps: Step S341, comparing the sulfur dioxide concentration corresponding to the real-time node in the real-time sulfur dioxide sequence with the sulfur dioxide concentration corresponding to the historical node in the multiple groups of historical sulfur dioxide sequences; Step S342, calculating the difference between the sulfur dioxide concentration difference corresponding to the real-time node and the sulfur dioxide concentration difference corresponding to multiple historical nodes and taking the absolute value to obtain multiple groups of sulfur dioxide concentration difference values; Step S343, if there is a mode among the multiple groups of sulfur dioxide concentration difference values, the sulfur dioxide concentration difference value corresponding to the mode is used as the similarity distance between the real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence; Step S344: if there is no mode among the multiple groups of sulfur dioxide concentration differences, then the multiple groups of sulfur dioxide concentration differences are added and averaged to obtain an average sulfur dioxide concentration difference, and the average sulfur dioxide concentration difference is used as the similarity distance between the real-time node in the real-time sulfur dioxide sequence and the corresponding historical node in the historical sulfur dioxide sequence; Step S345, repeat the above steps to obtain the sulfur dioxide node similarity between the remaining real-time nodes in the real-time sulfur dioxide sequence and the corresponding historical nodes in the historical sulfur dioxide sequence at the same current time point.
2. The method for processing air pollutant emission data based on a time series graph neural network according to claim 1 is characterized in that: The process of setting the atmospheric emission data set in step S1 is specifically as follows: Step S11, comparing the coverage range of the target space with the standard range, if the coverage range of the target space is greater than the first standard range, setting a first preset number of pollution data collection points for the target space; Step S12: if the coverage range of the target space is less than or equal to the first standard range and greater than the second standard range, a second preset number of pollution data collection points is set for the target space; Step S13, if the coverage range of the target space is less than or equal to the second standard range, a third preset number of pollution data collection points is set for the target space; wherein the first preset number is greater than the second preset number, the second preset number is greater than the third preset number; and the first standard range is greater than the second standard range; Step S14, install corresponding air sensors at positions corresponding to pollution data collection points in the target space, and start the air sensors according to the preset collection cycle to obtain atmospheric pollutant emission data corresponding to multiple time points within the collection cycle, and summarize and integrate the atmospheric pollutant emission data at positions corresponding to different pollution data collection points at the same time point to obtain an atmospheric emission data set.
3. The method for processing air pollutant emission data based on a time series graph neural network according to claim 1 is characterized in that: The step S2 includes the following sub-steps: Step S21, acquiring atmospheric pollutant emission data of multiple pollution data collection points in the atmospheric emission data set, and obtaining sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration of each pollution data collection point in the atmospheric emission data set; Step S22, analyzing all the atmospheric pollutant emission data in the atmospheric emission data set, defining the pollution data collection point currently being analyzed as a selected collection point, and calculating the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value, and particulate matter concentration difference value between the selected collection point and adjacent collection points; Step S23, comparing the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value of the selected collection points with corresponding thresholds respectively; Step S24: if any one of the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value corresponding to the selected collection point is greater than the corresponding threshold value, the air pollution emission data corresponding to the selected collection point is discarded, and an abnormal signal is generated and sent to the user terminal; If the sulfur dioxide concentration difference value, nitrogen oxide concentration difference value and particulate matter concentration difference value of the selected collection point are all less than the corresponding threshold value, the atmospheric pollution emission data of the corresponding selected collection point will be retained without any operation; Step S25, executing steps S22 to S24 for the remaining data collection nodes in the same atmospheric emission data set, recording the pollution data collection points retained in the same atmospheric emission data set as preliminary screening collection points, and recording the atmospheric pollution emission data corresponding to the preliminary screening collection points as preliminary screening pollutant emission data; After the initial screening of all data collection nodes in the atmospheric emission data set corresponding to the current time point is completed, the initial screening of the data collection nodes in the next atmospheric emission data set is completed according to the order of the time points.
4. The method for processing air pollutant emission data based on a time series graph neural network according to claim 3 is characterized in that: The step S22 further includes the following sub-steps: Step S221, drawing a radiation area corresponding to the selected collection point with the actual location of the selected collection point as the center and the estimated radiation length as the radius; Step S222: if there is one pollution data collection point or no pollution data collection point in the radiation area, the selected collection point is not included in the calculation, and the pollution data collection point closest to the selected collection point is recorded as a neighboring collection point, and the sulfur dioxide concentration difference value between the corresponding selected collection point and the neighboring collection point is calculated. Similarly, the nitrogen oxide concentration difference value and particulate matter concentration difference value between the selected collection point and the neighboring collection point can be calculated; Step S223: if there is more than one contaminated data collection point in the radiation area, the selected collection point is not included in it, then the radiation area is divided into four radiation partitions, and if there is a contaminated data collection point in the radiation partition, the contaminated data collection point in the radiation partition and the closest to the selected collection point is selected as the adjacent collection point.
5. The method for processing air pollutant emission data based on a time series graph neural network according to claim 4 is characterized in that: The step S223 further includes the following sub-steps: Step S2231, if there is only one adjacent collection point, the sulfur dioxide concentration difference value is obtained by subtracting the sulfur dioxide concentration corresponding to the adjacent collection point from the sulfur dioxide concentration corresponding to the selected collection point and taking the absolute value. Similarly, the nitrogen oxide concentration difference value is obtained by subtracting the nitrogen oxide concentration corresponding to the adjacent collection point from the nitrogen oxide concentration corresponding to the selected collection point and taking the absolute value. The corresponding particulate matter concentration difference value is obtained by subtracting the particulate matter concentration corresponding to the adjacent collection point from the particulate matter concentration corresponding to the selected collection point and taking the absolute value. Step S2232: if there are two or more adjacent collection points, the sulfur dioxide concentration, nitrogen oxide concentration and particulate matter concentration of the multiple adjacent collection points are added and averaged to obtain the average sulfur dioxide concentration, average nitrogen oxide concentration and average particulate matter concentration of the adjacent collection points; Step S2233, the sulfur dioxide concentration corresponding to the selected collection point is subtracted from the average sulfur dioxide concentration of the adjacent collection points to obtain the corresponding sulfur dioxide difference value; Similarly, the nitrogen oxide concentration corresponding to the collection point is subtracted from the average nitrogen oxide concentration of the adjacent collection points and the corresponding nitrogen oxide difference value is obtained by taking the absolute value; the corresponding particulate matter difference value is obtained by subtracting the average particulate matter concentration of the adjacent collection points from the particulate matter concentration corresponding to the collection point and the absolute value.
6. The method for processing air pollutant emission data based on a time series graph neural network according to claim 1, characterized in that: The process of performing corresponding operations according to the signal in step S4 is specifically as follows: If a data similarity signal is received, no action is performed; If a data anomaly signal is received, staff will be assigned to conduct a field investigation.
Citation Information
Patent Citations
Air particulate matter analysis and early warning method and device based on artificial intelligence
CN114023399A
Atmospheric pollution multi-granularity accurate cause analysis method and device
CN114822709A
Air quality prediction method based on adaptive dynamic graph neural network
CN114881358A
Atmospheric pollution data analysis method and device and readable storage medium
CN115455256A
Atmospheric pollution analysis and prediction method and system based on machine learning
CN115759438A
