Air quality prediction management method and system for air pollution prevention and control

Abstract

The invention relates to an air quality prediction management method and system for air pollution prevention and control, and relates to the technical field of environment monitoring. The method mainly comprises the following steps: respectively calculating air quality prediction data of a target city and adjacent cities according to industrial emission data, traffic emission data and real-time pollutant concentration monitoring data of the target city and the adjacent cities; updating the air quality prediction data of the target city through the meteorological prediction data and geographic information data of the target city and the adjacent cities thereof and the air quality prediction data of the adjacent cities to obtain final air quality prediction data of the target city; determining whether the final air quality prediction data of the target city is greater than a preset abnormal value; if the air quality is larger than the preset abnormal value, industrial emission and traffic emission of the target city or adjacent cities are managed and controlled according to the generation reason of the final air quality prediction data.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and particularly to an air quality prediction management method and system for preventing and controlling air pollution. Background Art

[0002] Air pollution refers to the phenomenon that when the content of some substances in the atmosphere reaches a certain level, it will harm the natural and social environment as well as human health. Human activities such as industrial production, transportation, urban dust, and fuel combustion will all cause air pollution. With the increasing development of industrialization and urbanization, the problem of air pollution has become a challenge faced by all countries.

[0003] In recent years, air pollution has occurred frequently, and people's attention to air quality has been increasing day by day. Users can only query relevant air quality information on the website, which is difficult to find and inaccurate. Therefore, it is necessary to develop a monitoring system for air quality that can monitor the air in real time, is simple and practical, and has a relatively high cost performance. Summary of the Invention

[0004] The present invention aims to provide an air quality prediction management method and system for preventing and controlling air pollution to solve the deficiencies in the prior art. The technical problems to be solved by the present invention are achieved through the following technical solutions.

[0005] An embodiment of the present invention provides an air quality prediction management method for preventing and controlling air pollution, and the method includes: Obtain meteorological prediction data, industrial emission data, traffic emission data, geographic information data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities; Calculate the air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and the adjacent cities; Update the air quality prediction data of the target city through the meteorological prediction data, geographic information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city; Determine whether the final air quality prediction data of the target city is greater than a preset abnormal value; If it is greater than the preset abnormal value, determine the reason for the generation of the final air quality prediction data; Manage and control the industrial emissions and traffic emissions of the target city or its adjacent cities according to the reason for the generation of the final air quality prediction data.

[0006] In an alternative embodiment, calculating the air quality prediction data of the target city and the adjacent city respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and the adjacent city includes: Preprocess the industrial emission data, the traffic emission data, and the real-time pollutant concentration monitoring data, where the preprocessing includes abnormal data filtering and data normalization; Convert the preprocessed industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data into a feature data matrix in the time dimension; Input the feature data matrices corresponding to the target city and the adjacent city respectively into the air quality prediction model to obtain the air quality prediction data corresponding to the target city and the adjacent city.

[0007] In an alternative embodiment, updating the air quality prediction data of the target city by using the meteorological prediction data, geographical information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent city to obtain the final air quality prediction data of the target city includes: Calculate the static attenuation value through the geographical information data of the target city and its adjacent cities; calculate the wind direction correction value and wind speed correction value through the meteorological prediction data of the target city and its adjacent cities; Calculate the urban pollution weight value between the target city and its adjacent cities according to the static attenuation value, the wind direction correction value, and the wind speed correction value; Update the air quality prediction data of the target city by using the urban pollution weight value, the air quality prediction data, and the meteorological prediction data of the target city and its adjacent cities to obtain the final air quality prediction data of the target city.

[0008] In an alternative embodiment, calculating the urban pollution weight value between the target city and its adjacent cities according to the static attenuation value, the wind direction correction value, and the wind speed correction value includes: Determine the product result of the static attenuation value, the wind direction correction value, and the wind speed correction value as the urban pollution weight value between the target city and its adjacent cities. In an alternative embodiment, updating the air quality prediction data of the target city by using the urban pollution weight value, the air quality prediction data, and the meteorological prediction data of the target city and its adjacent cities to obtain the final air quality prediction data of the target city includes: Convert the urban pollution weight value, the air quality prediction data, and the meteorological prediction data of the target city and its adjacent cities into a feature vector matrix; Input the feature vector matrix into the spatio-temporal convolution model to obtain the final air quality prediction data of the target city.

[0009] In an optional embodiment, before calculating the wind direction correction value and the wind speed correction value through the meteorological prediction data of the target city and its adjacent cities, the method further includes: Obtain the historical meteorological data of the target city, and perform wavelet transform on the historical meteorological data to extract the monsoon cycle; Determine the monsoon dominant wind direction and the monsoon average wind speed in each monsoon cycle; Calculate the wind direction correction value and the wind speed correction value through the monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities.

[0010] In an optional embodiment, calculating the wind direction correction value and the wind speed correction value through the monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities includes: Perform weighted calculation on the monsoon dominant wind direction and the monsoon average wind speed in all lengths of monsoon cycles to update the monsoon dominant wind direction and the monsoon average wind speed; Calculate the wind direction correction value and the wind speed correction value through the updated monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities.

[0011] In an optional embodiment, after inputting the feature vector matrix into the spatio-temporal convolution model to obtain the final air quality prediction data of the target city, the method further includes: Obtain the day-night diffusion index of the target city; Update the final air quality prediction data of the target city according to the day-night diffusion index of the target city.

[0012] In an optional embodiment, obtaining the day-night diffusion index of the target city includes: Obtain the daytime diffusion coefficient and the nighttime diffusion coefficient of the target city; Calculate the ratio of the daytime diffusion coefficient and the nighttime diffusion coefficient of the target city to obtain the initial day-night diffusion index; Calculate the day-night diffusion index of the target city according to the greening coverage rate of the target city and the initial day-night diffusion index of the target city.

[0013] An embodiment of the present invention provides an air quality prediction management system for air pollution prevention and control, and the system includes: An acquisition module, configured to acquire meteorological prediction data, industrial emission data, traffic emission data, geographical information data, and real-time pollutant concentration monitoring data of a target city and its adjacent cities; A calculation module, configured to calculate air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities; An update module, configured to update the air quality prediction data of the target city through the meteorological prediction data, geographical information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city; A determination module, configured to determine whether the final air quality prediction data of the target city is greater than a preset abnormal value; The determination module is further configured to, if it is greater than the preset abnormal value, determine the cause of the final air quality prediction data; A management module, configured to manage and control industrial emissions and traffic emissions of the target city or its adjacent cities according to the cause of the final air quality prediction data.

[0014] The embodiments of the present invention have the following advantages: A method and system for air quality prediction management for air pollution prevention and control provided by the embodiments of the present invention first acquire meteorological prediction data, industrial emission data, traffic emission data, geographical information data, and real-time pollutant concentration monitoring data of a target city and its adjacent cities, and then calculate the air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities; then update the air quality prediction data of the target city through the meteorological prediction data, geographical information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city; determine whether the final air quality prediction data of the target city is greater than a preset abnormal value; if it is greater than the preset abnormal value, determine the cause of the final air quality prediction data; finally, manage and control industrial emissions and traffic emissions of the target city or its adjacent cities according to the cause of the final air quality prediction data. Compared with the prior art that can only query relevant air quality information on a website, this application calculates air quality prediction data based on various data between cities, and then updates the air quality prediction data of the target city according to the air quality prediction data of adjacent cities, thereby improving the accuracy of air quality prediction control of the target city and further improving the effectiveness of air quality management of the target city. Description of the Drawings

[0015] Figure 1It is a flowchart of an air quality prediction management method for preventing and controlling air pollution provided by an embodiment of the present invention; Figure 2 It is a flowchart for determining the final air quality prediction data of a target city provided by an embodiment of the present invention; Figure 3 It is a schematic structural diagram of an air quality prediction management system for preventing and controlling air pollution provided by an embodiment of the present invention. Detailed implementation manners

[0016] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0017] Please refer to Figure 1 , which is an air quality prediction management method for preventing and controlling air pollution provided by an embodiment of the present invention. The method specifically includes S101 - S106: S101, Obtain the meteorological prediction data, industrial emission data, traffic emission data, geographical information data, and real - time pollutant concentration monitoring data of the target city and its adjacent cities.

[0018] Among them, the meteorological prediction data may include: temperature, humidity, wind speed, wind direction, air pressure, boundary layer height, etc.; the industrial emission data includes industrial emissions, dust, straw burning, etc.; the traffic emission data is the tail gas emissions of vehicles in the city; the geographical information data includes: city longitude and latitude, DEM elevation, land use type, vegetation coverage index, etc.; the real - time pollutant concentration monitoring data is the pollutant concentration data such as PM2.5, SO2, O3 detected by national - controlled / provincial - controlled stations.

[0019] It should be noted that the real - time pollutant concentration monitoring data can be obtained through multiple monitoring points set in the city, and then the average value of the obtained real - time pollutant concentration monitoring data is calculated to obtain the final real - time pollutant concentration monitoring data.

[0020] S102, Calculate the air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real - time pollutant concentration monitoring data of the target city and its adjacent cities.

[0021] In an optional embodiment provided by the present application, the calculating the air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real - time pollutant concentration monitoring data of the target city and its adjacent cities includes: S1021, Pre - process the industrial emission data, the traffic emission data, and the real - time pollutant concentration monitoring data. The pre - processing includes abnormal data filtering and data normalization.

[0022] S1022. Convert the pre - processed industrial emission data, traffic emission data, and real - time pollutant concentration monitoring data into a feature data matrix in the time dimension.

[0023] Specifically, the feature data matrix includes feature data of a certain time length. For example, in units of days, this time length can be 7 days. That is, each row of data in the feature data matrix represents the industrial emission data, traffic emission data, and real - time pollution concentration detection data of one day.

[0024] S1023. Input the feature data matrices corresponding to the target city and the adjacent cities into the air quality prediction model to obtain the air quality prediction data corresponding to the target city and the adjacent cities.

[0025] Among them, the air quality prediction model in this embodiment is a pre - trained time prediction model. After obtaining the feature data matrices corresponding to the target city and its adjacent cities respectively, input the feature data matrix of the target city into the air quality prediction model to obtain the control quality prediction data of the target city; input the feature data matrix of the adjacent city into the air quality prediction model to obtain the control quality prediction data of the adjacent city.

[0026] S103. Update the air quality prediction data of the target city through the meteorological prediction data, geographical information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city.

[0027] It should be noted that the air quality prediction data of the target city and the adjacent cities calculated in step S102 only represent the air quality affected by themselves locally. However, the control quality will be affected by other cities. For example, if there is a sandstorm in a certain city, its adjacent cities will be affected by the sandstorm of that city due to the effect of the wind direction. For the above reasons, this embodiment provides a method for the final air quality prediction data of the target city to solve the above problems.

[0028] Specifically, as Figure 2 shown, this embodiment provides a flowchart for determining the final air quality prediction data of the target city. The determination process of the final air quality prediction data of the target city includes: S1031. Calculate the static attenuation value through the geographical information data of the target city and its adjacent cities; calculate the wind direction correction value and wind speed correction value through the meteorological prediction data of the target city and its adjacent cities.

[0029] Specifically, the static attenuation value can be calculated by the formula where, is the static attenuation value, is the geographical distance between the target city i and the adjacent city j, is the attenuation radius (usually taken as 200 - 500 km, adjusted according to the regional scale). For example: The distance from Beijing (39.9°N, 116.4°E) to Tianjin (39.1°N, 117.2°E) is approximately 110 km. If d0 = 200 km, then w base ≈0.58.

[0030] It should be noted that before calculating the wind direction correction value and the wind speed correction value through the meteorological prediction data of the target city and its adjacent cities, the method further includes: obtaining the historical meteorological data of the target city, and performing wavelet transform on the historical meteorological data to extract the monsoon cycle; determining the monsoon dominant wind direction and the monsoon average wind speed in each monsoon cycle; calculating the wind direction correction value and the wind speed correction value through the monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities.

[0031] Apply the Morlet wavelet transform to the annual wind speed data to extract the annual cycle (12 months) and the seasonal cycle (3 months), and small scales (such as hourly level). For example, perform wavelet analysis on the 10-year wind speed data of a certain city and find that the power spectra of the 12-month cycle (annual cycle) and the 3-month cycle (seasonal cycle) exceed the 95% confidence level and are determined as significant cycles. Reconstruct the dominant wind direction component (such as the northwest wind in winter and the southeast wind in summer). Apply the Daubechies wavelet to the hourly temperature data to extract the diurnal pattern of the boundary layer height change. For example, after reconstructing the winter monsoon cycle in Beijing, the northwest wind dominant period is from December 1st to February 28th, and the average wind speed is 4.2 m / s. The summer diurnal cycle in Shanghai shows that the boundary layer height drops below 200 meters at night, and PM2.5 is prone to accumulate.

[0032] Specifically, calculating the wind direction correction value and the wind speed correction value through the monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities includes: performing weighted calculation on the monsoon dominant wind direction and the monsoon average wind speed in all lengths of monsoon cycles respectively to update the monsoon dominant wind direction and the monsoon average wind speed; calculating the wind direction correction value and the wind speed correction value through the updated monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities.

[0033] In this embodiment, the weighted calculation is performed through the formula to update the monsoon dominant wind direction and the monsoon average wind speed. Among them, the weight coefficients α, β, γ can be determined by historical data regression. Then, calculate the wind direction correction value and the wind speed correction value through the updated monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities.

[0034] Specifically, in this embodiment, the wind direction correction value can be calculated by the formula and the wind speed correction value can be calculated by the formula . Among them, is the wind direction correction coefficient (0.2 - 0.5, fitted by historical data), is the included angle between the connecting line direction from the target city i to the adjacent city j and the monsoon dominant wind direction (0° means completely downwind, 180° means upwind), is the average monsoon wind speed (unit: m / s), is the reference wind speed (such as 3 m / s).

[0035] For example, the monsoon dominant wind direction: such as the northwest wind in winter (azimuth angle of 315°), the city connection direction: the azimuth angle from Beijing to Shanghai is in the southeast direction (120°), and the included angle = |315° - 120°| = 195°, and actually take the minimum included angle of 165° (i.e., 360° - 195°). cos = cos165° ≈ -0.966. When cos > 0: downwind, enhance the transmission weight; when cos < 0: upwind, inhibit the transmission weight. If α = 0.3, then the correction factor 1 + 0.3×(-0.966) ≈ 0.711 + 0.3×(-0.966) ≈ 0.71. If the average northwest wind speed in winter = 4.5 m / s and the reference wind speed = 3 m / s, then the correction factor: 4.5 / 3 = 1.5.

[0036] S1032. Calculate the urban pollution weight value between the target city and its adjacent cities according to the static attenuation value, the wind direction correction value, and the wind speed correction value.

[0037] In this embodiment, the cities can be converted into a graph structure. The nodes in the graph structure represent cities (such as Beijing, Shanghai, Guangzhou, etc.), and each node contains a feature vector (such as PM2.5 concentration, industrial emissions, meteorological data, etc.). The edges in the graph structure are used to represent the pollution transmission relationship between cities (represented by the urban pollution weight value). Specifically, calculating the urban pollution weight value between the target city and its adjacent cities according to the static attenuation value, the wind direction correction value, and the wind speed correction value includes: determining the product result of the static attenuation value, the wind direction correction value, and the wind speed correction value as the urban pollution weight value between the target city and its adjacent cities.

[0038] That is, the urban pollution weight value can be calculated by the following formula: ; For example, during the winter northwest wind period, the calculation process of the urban pollution weight value from Beijing to Tianjin is as follows: Distance: 110 km → basic weight e -110 / 200 ≈0.58.

[0039] Wind direction angle: 0° (completely downwind) → w wind direction = 1 + 0.3×1 = 1.3.

[0040] Wind speed correction: 4.5 m / s → w wind speed = 4.5 / 3 = 1.5.

[0041] Urban pollution weight value: 0.58×1.3×1.5≈1.13.

[0042] S1033. Update the air quality prediction data of the target city through the urban pollution weight value, the air quality prediction data of the target city and its adjacent cities, and the meteorological prediction data to obtain the final air quality prediction data of the target city.

[0043] In this embodiment, updating the air quality prediction data of the target city through the urban pollution weight value, the air quality prediction data of the target city and its adjacent cities, and the meteorological prediction data to obtain the final air quality prediction data of the target city includes: converting the urban pollution weight value, the air quality prediction data of the target city and its adjacent cities, and the meteorological prediction data into a feature vector matrix; inputting the feature vector matrix into a spatio-temporal convolutional model to obtain the final air quality prediction data of the target city.

[0044] In this embodiment, the spatio-temporal convolutional model processes the transmission relationship between cities, and the propagation formula for each layer is: ; where, is the activation function, is the dynamic adjacency matrix determined according to the urban pollution weight value, which is updated every hour; is the feature vector matrix converted from the air quality prediction data and meteorological prediction data of the target city and its adjacent cities, is the trainable weight matrix.

[0045] It should be noted that the loss function of the spatio-temporal convolutional model in this embodiment consists of the prediction error loss and the physical constraint loss , that is, the loss function . When the weighted value of the two loss values is less than a predetermined value, the training of the spatio-temporal convolutional model is completed, where the calculation formula of the prediction error loss is: ; Among them, is the predicted concentration of city i at time t + 1 by the model, is the true monitored concentration of city i at time t + 1. For example, if the model predicts that the PM2.5 in Beijing is 85 μg / m 3 , and the true value is 90 μg / m 3 , then the single-point error is (85 - 90)2 = 25. After averaging for 13 cities, = 15.6.

[0046] The formula for the physical constraint loss is: ; Among them, is the concentration change amount, indicating the accumulation or dissipation of pollutants over time. The transport amount is the net input / output amount between cities, which is calculated by the formula . Among them, is the total amount of pollutant transport from upstream city i to city j, is the urban pollution weight value from city i to city j, is the concentration gradient, driving the transport of pollutants from the high-concentration area (i) to the low-concentration area (j).

[0047] For example, the model predicts the PM2.5 concentrations of 13 cities in the Beijing-Tianjin-Hebei region in the next 6 hours and calculates the pollution transport path. True concentration: [90, 120, 85,...] T (unit: μg / m 3 ); Predicted concentration: [85, 115, 80,...] T ; Transport amount calculation: According to the dynamic adjacency matrix, the total net input is +25 μg / m 3 ·h. Concentration change: P M 2 . 5 t + 1 − P M 2 . 5 t = [ + 5 , − 1 0 , + 3 , … ] T ; ; If = 0.5, then ; It should be noted that the in this embodiment is used to control the contribution weight of the physical constraint term to the total loss. If it is too large, the model overly satisfies the physical constraints and may sacrifice the prediction accuracy; If it is too small, the physical law constraints are insufficient, and the model may output unreasonable results (such as pollutants being generated out of thin air). Based on this, this embodiment selects through cross-validation. Usually, is between 0.1 - 1.0.

[0048] In an alternative embodiment provided by the present application, after inputting the feature vector matrix into the spatio-temporal convolution model to obtain the final air quality prediction data of the target city, the method further includes: obtaining the day-night diffusion index of the target city; updating the final air quality prediction data of the target city according to the day-night diffusion index of the target city.

[0049] Wherein, obtaining the day-night diffusion index of the target city includes: obtaining the daytime diffusion coefficient and the nighttime diffusion coefficient of the target city; calculating the ratio of the daytime diffusion coefficient and the nighttime diffusion coefficient of the target city to obtain the initial day-night diffusion index; calculating the day-night diffusion index of the target city according to the greening coverage rate of the target city and the initial day-night diffusion index of the target city.

[0050] In this embodiment, the calculation formula of the day-night diffusion index is: ; ; ; ; Wherein, is the initial day-night diffusion index, is the daytime diffusion coefficient, which is positively correlated with the boundary layer height and the wind speed; is the nighttime diffusion coefficient, which is affected by the thickness of the inversion layer and the stability. is the daytime boundary layer height (which can be estimated through meteorological sounding data or models), is the daytime average wind speed, is the adjustment coefficient (usually taken as 0.1 - 0.3), is the normalized difference vegetation index (reflecting the greening coverage rate). is the nighttime boundary layer height (usually 10% - 30% of the daytime), is the nighttime average wind speed, is the temperature difference between the ground surface and the bottom of the inversion layer. The greater the temperature difference, the weaker the diffusion ability, is the attenuation coefficient (empirical value is about 0.05 - 0.1). is the day-night diffusion index of the target city, is the greening adjustment coefficient (which needs to be locally calibrated), is the greening coverage rate of the target city.

[0051] In this embodiment, updating the final air quality prediction data of the target city according to the day-night diffusion index of the target city can adopt the following formula: ; Wherein, For updated air quality prediction data, For the final air quality prediction data obtained through the spatio-temporal convolution model, For the historical average DNI, For the current calculated real-time , For the sensitivity coefficient (usually taken as 0.3 - 0.6, fitted through local data).

[0052] It should be noted that when > : The diffusion ability is better than normal, and the predicted concentration is lowered; when < : The diffusion ability deteriorates, and the predicted concentration is raised.

[0053] Aiming at the fact that traditional models often underestimate the pollution accumulation caused by nocturnal inversion, the air quality prediction data can be corrected through this embodiment. For example, the predicted PM2.5 at night in Shijiazhuang is 80 μg / m³, = 8 ( = 15), and the concentration can be increased by correcting through the correction formula in this embodiment, that is , and the actual observed value is 108 μg / m³. Therefore, the error can be reduced from 37.5% to 1.9% through the correction formula in this embodiment.

[0054] And for the increase in the urban green space rate, the DNI increases due to the enhanced nocturnal diffusion. For example, when the green space rate increases from 15% to 20%, increases from 12 to 13.5; the concentration can be decreased by correcting through the correction formula in this embodiment, that is .

[0055] S104, Determine whether the final air quality prediction data of the target city is greater than the preset outlier.

[0056] Among them, the preset outlier can be set according to actual needs.

[0057] S105, If it is greater than the preset outlier, then determine the cause of the final air quality prediction data.

[0058] In this embodiment, the process of determining the cause of the final air quality prediction data can be: calculating the pollution contribution rate, and then determining the cause of the final air quality prediction data according to the pollution contribution rate.

[0059] The calculation formula of the pollution contribution rate is: , .

[0060] Among them, The total amount of pollutant transfer from upstream city i to city j is the urban pollution weight value from city i to city j is the concentration gradient, driving the pollutant to transfer from the high-concentration area (i) to the low-concentration area (j). is the total pollutant concentration of city j at time t + 1, usually the monitoring value or model prediction value of a single spatio-temporal node.

[0061] S106, according to the reason for generating the final air quality prediction data, manage and control the industrial emissions and traffic emissions of the target city or its adjacent cities.

[0062] For example, the PM2.5 concentration in Tianjin at t + 1 , if the sum of the transfer fluxes from all upstream cities to Tianjin , .

[0063] Through this contribution rate, it can be determined that 60% of the PM2.5 pollution in Tianjin comes from external cities (such as Beijing, Shijiazhuang, etc.), and 40% comes from local emissions. That is, according to the reason that the pollution comes from external cities, steel production in Shijiazhuang can be limited by 50% to reduce local emissions; or Beijing can strengthen the inspection of the emissions of inbound trucks to reduce the transfer flux, etc. This embodiment does not make specific limitations on this.

[0064] For the reasons within the city itself, the industrial emissions and traffic emissions of the target city can be managed through the content in Table 1. Among them, DNI in the table is calculated by the formula calculated.

[0065] Table 1 DNI range Control level Example of measures DNI > 15 Yellow warning Strengthen road sprinkling at night and restrict the operation of dump trucks 5 ≤ DNI ≤ 15 Orange warning Shut down 30% of high-emission enterprises and suspend outdoor classes in primary and secondary schools DNI < 5 Red warning Implement odd-even license plate restrictions throughout the region and limit the production of heavy industries by 50% This embodiment provides an air quality prediction management method for air pollution prevention and control. First, meteorological prediction data, industrial emission data, traffic emission data, geographical information data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities are obtained. Then, according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and the adjacent cities, the air quality prediction data of the target city and the adjacent cities are calculated respectively. After that, through the meteorological prediction data, geographical information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities, the air quality prediction data of the target city are updated to obtain the final air quality prediction data of the target city. It is determined whether the final air quality prediction data of the target city is greater than a preset abnormal value. If it is greater than the preset abnormal value, the reason for the generation of the final air quality prediction data is determined. Finally, according to the reason for the generation of the final air quality prediction data, the industrial emissions and traffic emissions of the target city or the adjacent cities are managed and controlled. Compared with the prior art that can only query relevant air quality information on the website, this application calculates the air quality prediction data based on various data between cities, and then updates the air quality prediction data of the target city according to the air quality prediction data of the adjacent cities, thereby improving the accuracy of the target city's air quality prediction and further improving the effectiveness of the target city's air quality management.

[0066] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0067] In one embodiment, an air quality prediction management system for air pollution prevention and control is provided. As Figure 3 shown, the detailed descriptions of the functional modules of the air quality prediction management system for air pollution prevention and control are as follows: An acquisition module 31, configured to acquire meteorological prediction data, industrial emission data, traffic emission data, geographical information data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities; A calculation module 32, configured to calculate the air quality prediction data of the target city and the adjacent cities respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and the adjacent cities; An update module 33, configured to update the air quality prediction data of the target city through the meteorological prediction data, geographical information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city; A determination module 34, configured to determine whether the final air quality prediction data of the target city is greater than a preset abnormal value; The determining module 34 is further configured to determine the cause of the generation of the final air quality prediction data if it is greater than a preset outlier value; The management module 35 is configured to manage and control industrial emissions and traffic emissions in the target city or its adjacent cities according to the cause of the generation of the final air quality prediction data.

[0068] In an optional embodiment, the calculation module 32 is specifically configured to: Preprocess the industrial emission data, the traffic emission data, and the real-time pollutant concentration monitoring data, where the preprocessing includes abnormal data filtering and data normalization; Convert the preprocessed industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data into a feature data matrix in the time dimension; Input the feature data matrices corresponding to the target city and the adjacent cities into an air quality prediction model to obtain the air quality prediction data corresponding to the target city and the adjacent cities.

[0069] In an optional embodiment, the updating module 33 is specifically configured to: Calculate a static attenuation value through the geographic information data of the target city and its adjacent cities; calculate a wind direction correction value and a wind speed correction value through the meteorological prediction data of the target city and its adjacent cities; Calculate the urban pollution weight value between the target city and its adjacent cities according to the static attenuation value, the wind direction correction value, and the wind speed correction value; Update the air quality prediction data of the target city through the urban pollution weight value, the air quality prediction data, and the meteorological prediction data of the target city and its adjacent cities to obtain the final air quality prediction data of the target city.

[0070] In an optional embodiment, the updating module 33 is specifically configured to: Determine the product result of the static attenuation value, the wind direction correction value, and the wind speed correction value as the urban pollution weight value between the target city and its adjacent cities.

[0071] In an optional embodiment, the updating module 33 is specifically configured to: Convert the urban pollution weight value, the air quality prediction data, and the meteorological prediction data of the target city and its adjacent cities into a feature vector matrix; Input the feature vector matrix into a time-space convolutional model to obtain the final air quality prediction data of the target city.

[0072] In an optional embodiment, the acquisition module 31 is further configured to acquire the historical meteorological data of the target city, and perform wavelet transform on the historical meteorological data to extract the monsoon cycle; The determination module 34 is further configured to determine the monsoon dominant wind direction and the monsoon average wind speed in each monsoon cycle; The calculation module 32 is further configured to calculate a wind direction correction value and a wind speed correction value based on the monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities.

[0073] In an optional embodiment, the calculation module 32 is specifically configured to: Perform weighted calculation on the monsoon dominant wind direction and the monsoon average wind speed in monsoon cycles of all lengths respectively to update the monsoon dominant wind direction and the monsoon average wind speed; Calculate a wind direction correction value and a wind speed correction value based on the updated monsoon dominant wind direction, the monsoon average wind speed, and the meteorological prediction data of the target city and its adjacent cities.

[0074] In an optional embodiment, the acquisition module 31 is further configured to acquire the day-night diffusion index of the target city; The update module 33 is further configured to update the final air quality prediction data of the target city according to the day-night diffusion index of the target city.

[0075] In an optional embodiment, the acquisition module 31 is specifically configured to: Acquire the daytime diffusion coefficient and the nighttime diffusion coefficient of the target city; Calculate the ratio of the daytime diffusion coefficient and the nighttime diffusion coefficient of the target city to obtain the initial day-night diffusion index; Calculate the day-night diffusion index of the target city based on the greening coverage rate of the target city and the initial day-night diffusion index of the target city.

[0076] It should be noted that the above detailed description is exemplary and is intended to provide further illustration of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs.

[0077] For the specific limitations on an air quality prediction management system for air pollution prevention, reference may be made to the limitations on an air quality prediction management method for air pollution prevention in the above text, which will not be elaborated here. Each module in the above device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0078] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0079] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An air quality prediction management method for preventing and controlling air pollution, characterized in that, The method includes: Obtaining meteorological prediction data, industrial emission data, traffic emission data, geographic information data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities; Calculating the air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities; Updating the air quality prediction data of the target city through the meteorological prediction data, geographic information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city; Determining whether the final air quality prediction data of the target city is greater than a preset outlier; If it is greater than the preset outlier, determining the cause of the generation of the final air quality prediction data; Managing and controlling the industrial emissions and traffic emissions of the target city or its adjacent cities according to the cause of the generation of the final air quality prediction data.

2. The method according to claim 1, characterized in that, The calculating the air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities includes: Performing preprocessing on the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data, and the preprocessing includes abnormal data filtering and data normalization; Converting the preprocessed industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data into a feature data matrix in the time dimension; Inputting the feature data matrices corresponding to the target city and its adjacent cities into an air quality prediction model to obtain the air quality prediction data corresponding to the target city and its adjacent cities.

3. The method according to claim 2, wherein The updating the air quality prediction data of the target city through the meteorological prediction data, geographic information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city includes: Calculating a static attenuation value through the geographic information data of the target city and its adjacent cities; calculating a wind direction correction value and a wind speed correction value through the meteorological prediction data of the target city and its adjacent cities; Calculating the urban pollution weight value between the target city and its adjacent cities according to the static attenuation value, the wind direction correction value, and the wind speed correction value; Updating the air quality prediction data of the target city through the urban pollution weight value, the air quality prediction data, and the meteorological prediction data of the target city and its adjacent cities to obtain the final air quality prediction data of the target city.

4. The method according to claim 3, wherein The calculating the urban pollution weight value between the target city and its adjacent cities according to the static attenuation value, the wind direction correction value, and the wind speed correction value includes: Determining the product result of the static attenuation value, the wind direction correction value, and the wind speed correction value as the urban pollution weight value between the target city and its adjacent cities.

5. The method according to claim 4, characterized in that, Updating the air quality prediction data of the target city through the urban pollution weight value, the air quality prediction data and meteorological prediction data of the target city and its adjacent cities to obtain the final air quality prediction data of the target city, including: Converting the urban pollution weight value, the air quality prediction data and meteorological prediction data of the target city and its adjacent cities into a feature vector matrix; Inputting the feature vector matrix into a spatio-temporal convolutional model to obtain the final air quality prediction data of the target city.

6. The method according to claim 3, wherein Before calculating the wind direction correction value and wind speed correction value through the meteorological prediction data of the target city and its adjacent cities, the method further includes: Obtaining the historical meteorological data of the target city and performing wavelet transform on the historical meteorological data to extract the monsoon cycle; Determining the monsoon dominant wind direction and monsoon average wind speed in each monsoon cycle; Calculating the wind direction correction value and wind speed correction value through the monsoon dominant wind direction, the monsoon average wind speed and the meteorological prediction data of the target city and its adjacent cities.

7. The method according to claim 6, characterized in that, Calculating the wind direction correction value and wind speed correction value through the monsoon dominant wind direction, the monsoon average wind speed and the meteorological prediction data of the target city and its adjacent cities includes: Performing weighted calculation on the monsoon dominant wind direction and monsoon average wind speed in monsoon cycles of all lengths to update the monsoon dominant wind direction and monsoon average wind speed; Calculating the wind direction correction value and wind speed correction value through the updated monsoon dominant wind direction, monsoon average wind speed and the meteorological prediction data of the target city and its adjacent cities.

8. The method according to claim 5, wherein After inputting the feature vector matrix into the spatio-temporal convolutional model to obtain the final air quality prediction data of the target city, the method further includes: Obtaining the day-night diffusion index of the target city; Updating the final air quality prediction data of the target city according to the day-night diffusion index of the target city.

9. The method according to claim 8, characterized in that Obtaining the day-night diffusion index of the target city includes: Obtaining the daytime diffusion coefficient and nighttime diffusion coefficient of the target city; Calculating the ratio of the daytime diffusion coefficient and nighttime diffusion coefficient of the target city to obtain the initial day-night diffusion index; Calculating the day-night diffusion index of the target city according to the greening coverage rate of the target city and the initial day-night diffusion index of the target city.

10. An air quality prediction and management system for preventing and controlling air pollution, characterized in that, The system includes: An acquisition module, configured to acquire the meteorological prediction data, industrial emission data, traffic emission data, geographic information data, and real-time pollutant concentration monitoring data of the target city and its adjacent cities; A calculation module, configured to calculate the air quality prediction data of the target city and its adjacent cities respectively according to the industrial emission data, traffic emission data, and real-time pollutant concentration monitoring data of the target city and the adjacent cities; An update module, configured to update the air quality prediction data of the target city through the meteorological prediction data, geographic information data of the target city and its adjacent cities, and the air quality prediction data of the adjacent cities to obtain the final air quality prediction data of the target city; A determination module, configured to determine whether the final air quality prediction data of the target city is greater than a preset outlier value; The determination module is further configured to determine the cause of the generation of the final air quality prediction data if it is greater than the preset outlier value; A management module, configured to manage and control industrial emissions and traffic emissions in the target city or its adjacent cities according to the cause of the generation of the final air quality prediction data.

Images