An intelligent early warning system for excessive concentration of air pollutants

By introducing edge computing and dynamic feedback mechanisms in the air pollution monitoring system, abnormal data are screened out and the average and estimated value of pollutant concentrations are calculated, and dynamic thresholds are set, which solves the large amount of calculation and early warning of air pollution monitoring in the existing technology. Accurate problems achieve efficient and accurate air pollution warning and governance strategy support.

CN118761884BActive Publication Date: 2024-11-22SOUTH CHINA INST OF ENVIRONMENTAL SCI MEP
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411209325.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-22
Estimated Expiration
2044-08-30

AI Technical Summary

Technical Problem

The existing air pollution monitoring technology has large amounts of calculations and data not preprocessed, resulting in deviations in the calculation results, and the lack of effective early warnings. It lacks a third-party feedback mechanism, making it difficult to discover and solve data acquisition problems in a timely manner.

Method used

An intelligent early warning system for air pollutant concentration exceeding the standard was designed. An abnormal data was screened through edge computing equipment, the average value of pollutant concentration and estimated concentration values ​​were calculated, dynamic thresholds were set, and early warning judgment and optimization were performed in combination with dynamic feedback modules to ensure early warning. accuracy and timeliness.

Benefits of technology

It improves the accuracy and timeliness of air pollution warnings, avoids invalid warnings, ensures the integrity and reliability of data, and provides targeted governance directions and effective pollution control strategies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118761884B_ABST
    Figure CN118761884B_ABST
Patent Text Reader

Abstract

The present invention discloses an intelligent early warning system for excessive concentration of air pollutants, which relates to the field of air pollution monitoring. The main scheme is: using edge computing devices to screen out abnormal data and duplicate data to reduce the difficulty of data processing, and then comparing the average concentration value #imgabs0# of different pollutants with a dynamic threshold value #imgabs1# to determine whether to issue an alarm; comparing the estimated concentration value #imgabs2# of different pollutants with a threshold set of different pollutants to determine whether to issue an early warning, thereby further improving the accuracy of the early warning; the pollutant concentration diffusion model of the cause analysis module helps to quickly and accurately determine the location of the pollution source, providing direction for targeted governance, and the dynamic feedback module can continuously optimize the system, improve the accuracy and reliability of the monitoring data, and avoid invalid early warnings triggered by equipment failures.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of air pollution monitoring, and in particular to an intelligent early warning system for excessive concentration of air pollutants. Background Art

[0002] In order to effectively deal with the problem of air pollution, cities have established air pollution monitoring networks and developed corresponding monitoring and early warning technologies. However, the existing technologies still have the following shortcomings:

[0003] In the real-time monitoring and data collection stage, all data will be uniformly transmitted to the processing platform, which will increase the amount of calculation on the processing platform, and because the data is not pre-processed, the calculation results will be biased;

[0004] It can only judge the environmental conditions at the time of data collection, but cannot effectively predict the situation of pollutants and cannot provide accurate early warning;

[0005] There is a lack of a third-party feedback mechanism. When problems arise during the data acquisition phase, it is difficult to discover the problems in a timely manner and there is a lack of remedial measures. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the shortcomings of the prior art, the present invention provides an intelligent early warning system for excessive concentration of air pollutants. By receiving data from the data transmission module, the edge computing device screens out abnormal data and duplicate data, and can pre-process the data to reduce the difficulty of data processing. Then, the average concentration of different pollutants is calculated. With dynamic threshold Compare and determine whether to issue an alarm; compare the estimated concentration values ​​of different pollutants With dynamic threshold The matrix F0 is compared to determine whether to issue an early warning, which further improves the accuracy of the early warning. The dynamic feedback module can also provide third-party feedback to avoid invalid early warnings caused by faults.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent early warning system for excessive concentration of air pollutants, comprising:

[0010] The data acquisition module is used to obtain air pollution data at different monitoring points in the monitoring area; air pollution data includes concentration data of multiple pollutants; and the edge computing device is used to filter out abnormal data and duplicate data to generate a basic data set;

[0011] The data transmission module is used for data transmission of normal data sets. When a network interruption occurs during the transmission process, the air pollution data will be temporarily stored in the cache area of ​​the local data processing center and will be resumed after the network is restored.

[0012] Data fusion and processing module: Receives data from the data transmission module and calculates the average concentration of different pollutants and estimated concentrations ;

[0013] Threshold setting module, used to set dynamic thresholds for different pollutants , and preset threshold sets for different pollutants;

[0014] The early warning trigger module calculates the average concentration of different pollutants With dynamic threshold Compare and determine whether to issue an alarm; compare the estimated concentration values ​​of different pollutants Compare with the threshold set of different pollutants to determine whether to issue an early warning;

[0015] Cause analysis module, used to establish a pollutant concentration diffusion model and determine the location of pollution sources through data analysis;

[0016] Dynamic feedback module, which receives air quality information and third-party monitoring data from the public through mobile applications, compares and verifies it with system monitoring data, and assists in judging the accuracy of collected data;

[0017] The decision support module provides rectification suggestions based on the analysis results of the early warning trigger module and the cause analysis module.

[0018] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentrations of air pollutants, the monitoring points within the monitoring area are arranged in a grid pattern, and detection equipment is additionally installed in key pollution source areas, densely populated areas, and transportation hubs; the detection equipment includes a variety of different types of detection devices, which perform multi-dimensional detection of different pollutants, and are used to obtain multi-dimensional concentration data of multiple pollutants.

[0019] In the preferred solution of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the method for edge computing devices to screen out abnormal data and duplicate data is as follows:

[0020] Normal range thresholds for different pollutants are preset, and the real-time data of different pollutants are compared with the normal range thresholds. The data within the normal range thresholds are used to generate a normal data set; the concentration data of the same pollutant at adjacent time points in the normal data set are compared and deduplicated. The deduplication judgment formula is as follows:

[0021] ;

[0022] in: is the concentration data of the pollutant with serial number q at time point i, is the pollutant with serial number q at time point ( ) concentration data, time point i and time point ( ) are adjacent time points; It is the error range of the preset pollutant concentration data;

[0023] Generate the basic data set with the retained data.

[0024] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the average concentration of different pollutants is calculated. , the formula is as follows:

[0025] ;

[0026] Where: the value of i is (1, 2, 3...k), k is a positive integer representing the total number of time points; The serial number representing different pollutants is a positive integer; The pollutant numbers are of pollutants at a time point The concentration value at .

[0027] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the estimated concentration values ​​of different pollutants are calculated. The specific method is as follows:

[0028] Predict the estimated concentration values ​​of different pollutants based on the long short-term memory network LSTM model ; The formula is as follows:

[0029] ;

[0030] in: is the concentration value of the pollutant with serial number q at the current moment; It's the past The concentration value of the pollutant at the time sequence number q, The value is a positive integer; estimated concentration value is the concentration value of the pollutant with serial number q at one moment in the future, is the weight parameter of the LSTM model.

[0031] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the dynamic thresholds of different pollutants are calculated. The method is:

[0032] Obtaining basic thresholds for different pollutants , Current season parameters , Average seasonal impact index 、Current meteorological conditions parameters , Average meteorological conditions impact value , Current time period parameters , Average time period impact value ;

[0033] The dynamic thresholds of different pollutants are calculated using the above data , the formula is as follows:

[0034] ;

[0035] in: Current season parameters Average seasonal impact index The weight coefficient of the difference is 0.2~0.4; is the current meteorological condition parameter Impact value of average meteorological conditions The weight coefficient of the difference is 0.1~0.3; Is the current time period parameter Average time period impact value The weight coefficient of the difference is 0.2~0.4; and ;

[0036] The threshold set includes a normal impact threshold F1, a low impact threshold F2, a medium impact threshold F3, a high state threshold F4 and an abnormal state threshold F5, and F1<F2<F3<F4<F5.

[0037] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the method for determining whether to issue an alarm is:

[0038] The average concentration of different pollutants Dynamic thresholds for different pollutants When the average concentration is met, ≥ Dynamic Threshold If the condition is met, an alarm is issued; otherwise, it is judged to be normal;

[0039] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the method for determining whether to issue an early warning is:

[0040] The estimated concentration values ​​of different pollutants Compared with the threshold sets of different pollutants, ≤F1, it is judged as normal and no warning action is issued; when F1< ≤F2, it is judged as a low pollution state and a first-level warning is issued; when F2< ≤F3, it is judged as a moderate pollution state and a second-level warning is issued; when F3< ≤F4, it is judged as a highly polluted state and a level 3 warning is issued; when F4< When ≤F5, it is judged as an abnormal pollution state and an abnormal state warning is issued.

[0041] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the working steps of the cause analysis module are as follows:

[0042] Obtain meteorological data from different monitoring points within the monitoring area, establish a pollutant concentration diffusion model based on meteorological data and air pollution data, and determine the diffusion trend of pollutant concentration and the direction of pollution sources;

[0043] Use online map services as the base map to accurately mark the locations of different monitoring points in the monitoring area on the map; use color gradients or heat maps to display the concentration distribution on the map based on the pollutant concentration data of the monitoring points; determine the location of the pollution source through data analysis, mark it with icons or markers on the map, and display relevant pollution source information.

[0044] In the preferred embodiment of the above-mentioned intelligent early warning system for excessive concentration of air pollutants, the decision support module obtains the early warning information analyzed by the early warning trigger module and the pollution source information analyzed by the cause analysis module, generates a pollution analysis report, and generates rectification suggestions based on the pollution analysis report.

[0045] (III) Beneficial effects

[0046] The present invention provides an intelligent early warning system for excessive concentration of air pollutants, which has the following beneficial effects:

[0047] (1) By acquiring air pollution data from different monitoring points within the monitoring area, comprehensive acquisition of multi-pollutant concentration data within the monitoring area is achieved, ensuring the integrity and diversity of the data and providing a basis for accurate analysis and early warning;

[0048] (2) The breakpoint-resume function of the data transmission module ensures the continuity and integrity of air pollution data and avoids data loss due to network interruption;

[0049] (3) The data fusion and processing module can calculate the average concentration and estimated concentration value, providing more valuable data for subsequent comparison and analysis;

[0050] (4) The dynamic threshold setting of the threshold setting module can be flexibly adjusted according to the actual situation, which improves the accuracy and timeliness of the early warning. Combined with the early warning trigger module, by comparing the average concentration and the estimated concentration value with the corresponding threshold, early warning and alarm can be issued in advance, which helps to take timely response measures and reduce pollution hazards;

[0051] (5) The pollutant concentration diffusion model of the cause analysis module helps to quickly and accurately determine the location of the pollution source, providing direction for targeted governance;

[0052] (6) The dynamic feedback module can continuously optimize the system, improve the accuracy and reliability of monitoring data, and avoid invalid warnings triggered by equipment failures;

[0053] (7) The decision support module provides rectification suggestions based on comprehensive analysis results, providing strong support for relevant departments to formulate effective pollution control strategies and promote the improvement of air quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A schematic diagram of the system composition of the intelligent early warning system for excessive concentration of air pollutants of the present invention;

[0055] Figure 2 Schematic diagram of the steps for generating a basic data set in the intelligent early warning system for excessive concentration of air pollutants of the present invention;

[0056] Figure 3 A schematic diagram of determining whether to issue an alarm in the intelligent early warning system for excessive concentration of air pollutants of the present invention;

[0057] Figure 4 It is a schematic diagram for judging whether to issue a warning in the intelligent early warning system for excessive concentration of air pollutants of the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not 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.

[0059] See also Figure 1-4 The present invention provides an intelligent early warning system for excessive concentration of air pollutants, comprising:

[0060] The data acquisition module is used to obtain air pollution data at different monitoring points in the monitoring area; air pollution data includes concentration data of various pollutants, such as sulfur dioxide SO2, nitrogen oxides NO x , inhalable particulate matter PM2.5, PM10 , CO2, ozone O3 and other air pollutant concentration data; and filter out abnormal data and duplicate data through edge computing devices to generate basic data sets; edge computing devices can be edge servers or smart gateways with certain computing capabilities, which can perform preliminary screening of data collected by detection equipment to reduce the amount and difficulty of subsequent calculations;

[0061] Specifically, the monitoring points within the monitoring area are arranged in a grid pattern to reflect the air quality conditions in different locations in an all-round way, and additional detection equipment is installed in key pollution source areas, densely populated areas and transportation hubs. The detection equipment includes a variety of different types of detection devices to perform multi-dimensional detection of different pollutants, so as to obtain multi-dimensional concentration data of multiple pollutants and realize multiple measurements and mutual verification of the same pollutant. In addition to fixed detection equipment, mobile monitoring equipment is also provided, such as portable monitoring instruments installed on monitoring vehicles, which are used to conduct flexible and dynamic monitoring of specific areas or routes, supplement the deficiencies of fixed monitoring points and further ensure the accuracy of the data.

[0062] Specifically, the method by which edge computing devices filter out abnormal and duplicate data is as follows:

[0063] Normal range thresholds for different pollutants are preset, and the real-time data of different pollutants are compared with the normal range thresholds. The data within the normal range thresholds are used to generate a normal data set. The normal range thresholds for different air pollutants usually range from extremely low concentrations to higher concentrations, such as the normal range threshold for PM2.5 is 0~500μg / m³, the normal range threshold for CO2 is 0~2000ppm, and the normal range threshold for SO2 is 10~500μg / m³. Generally, values ​​beyond this range are generally caused by equipment abnormalities and are not real data.

[0064] The concentration data of the same pollutant at adjacent time points in the normal data set are compared and deduplicated. The deduplication judgment formula is as follows:

[0065] ;

[0066] in: is the concentration data of the pollutant with serial number q at time point i, is the pollutant with serial number q at time point ( ) concentration data, time point i and time point ( ) are adjacent time points; It is the preset error range of pollutant concentration data, which can be taken as about 1% to 5% of the concentration measurement value; and the retained data will be used to generate the basic data set.

[0067] In the above scheme, by obtaining air pollution data from different monitoring points in the monitoring area, comprehensive acquisition of multi-pollutant concentration data in the monitoring area is achieved, ensuring the integrity and diversity of the data, providing a basis for accurate analysis and early warning, and being able to screen out abnormal and duplicate data, improve data quality, and reduce the difficulty and amount of calculation of the overall data.

[0068] The data transmission module is used for data transmission of normal data sets. When a network interruption occurs during the transmission process, the air pollution data will be temporarily stored in the cache area of ​​the local data processing center and will be resumed after the network is restored.

[0069] Specifically, build a high-speed, stable and dedicated data transmission network, such as using 5G technology combined with satellite communications, to ensure that data can be transmitted to the data processing center in real time and without delay, and ensure data continuity even in remote areas with weak network signals; in order to cope with possible network failures or data loss, use data caching and breakpoint resumption technology, where the sensor stores data for a certain period of time locally, and automatically uploads the cached data when the network is restored; at the same time, encrypt the transmitted data and use advanced encryption algorithms to ensure the security and confidentiality of the data during transmission, and prevent data from being tampered with or stolen; ensure data integrity and avoid data loss due to network interruptions.

[0070] Data fusion and processing module: Receives data from the data transmission module and calculates the average concentration of different pollutants and estimated concentrations ;

[0071] Specifically, the concentration values ​​of pollutants at different time points within a time period are summed and divided by the number of time points to obtain the average value, and the average concentration value of different pollutants is calculated. , the formula is as follows:

[0072] ;

[0073] Where: the value of i is (1, 2, 3...k), k is a positive integer representing the total number of time points; The serial number representing different pollutants is a positive integer;

[0074] Calculate the estimated concentration values ​​of different pollutants The specific method is as follows:

[0075] Based on the long short-term memory network LSTM model, the estimated concentration values ​​of different pollutants are predicted based on the pollutant concentration data over the past period of time. ; The formula is as follows:

[0076] ;

[0077] in: The current time sequence number is The concentration of pollutants; It's the past The concentration value of the pollutant at the time sequence number q, The value is a positive integer; estimated concentration value is the concentration value of the pollutant with serial number q at one moment in the future; for the current moment; It's the past a moment; It's the past a moment; A moment in the future. It is the weight parameter of the LSTM model and can be obtained through model training.

[0078] Threshold setting module, used to set dynamic thresholds for different pollutants , and preset threshold sets for different pollutants;

[0079] Specifically, dynamic thresholds for different pollutants are calculated The method is:

[0080] Obtaining basic thresholds for different pollutants , Current season parameters , Average seasonal impact index 、Current meteorological conditions parameters , Average meteorological conditions impact value , Current time period parameters , Average time period impact value ;

[0081] The dynamic thresholds of different pollutants are calculated using the above data , the formula is as follows:

[0082] ;

[0083] in: Current season parameters Average seasonal impact index The weight coefficient of the difference is 0.2~0.4; is the current meteorological condition parameter Impact value of average meteorological conditions The weight coefficient of the difference is 0.1~0.3; Is the current time period parameter Average time period impact value The weight coefficient of the difference is 0.2~0.4; and ;

[0084] It should be noted that the basic thresholds for different pollutants It is a fixed threshold determined based on existing air quality standards and related research results at home and abroad. It represents the pollutant concentration level that is generally considered safe or acceptable. It comes from authoritative environmental quality standard documents, scientific research reports and relevant laws and regulations. For example, the World Health Organization's recommended value for PM2.5, or the limit value specified in the air quality standard formulated by the national environmental protection department; for example, the basic threshold of PM2.5 can be selected between 35 and 75 μg / m³; if according to my country's air quality standards, the 24-hour average PM2.5 concentration limit in the secondary standard is 75 μg / m³;

[0085] Current season parameters It is used to indicate the current season. It uses simple numerical values ​​to distinguish the impact of different seasons on pollutant concentration thresholds. Different seasons can be mapped to different numerical values, which are determined by the system's built-in season judgment module based on time and local climate characteristics. For example, the season can be judged based on the month or local climate characteristics such as temperature, precipitation, etc., such as 1 for winter and 0 for summer; it is usually a discrete value, such as 1 for winter and 0 for summer. Spring and autumn can be set to 0.5 or other appropriate values ​​according to specific circumstances. Assuming that it is winter, then ;

[0086] Average seasonal impact index It is the average value of the impact of different seasons on pollutant concentrations obtained through statistical analysis of historical data; it is obtained by statistically calculating the pollutant concentration data of the same season in the past years. For example, the pollutant concentration data of winter in the past five years are statistically analyzed to calculate the average value of the impact of winter on pollutant concentrations. The specific value depends on the statistical results of historical data. Assuming that according to historical data, the average impact of winter on pollutant concentration causes the threshold to increase by 10μg / m³, then ;

[0087] Current weather conditions parameters Indicates the current specific meteorological conditions, such as wind speed, humidity, etc. These conditions will affect the diffusion and concentration of pollutants. Through the data obtained in real time by meteorological monitoring equipment, the wind speed is usually in the range of zero to tens of meters per second, and the humidity is usually between 0%-100%. For example, the current wind speed is 5 meters per second and the humidity is 60%. These values ​​can be combined into a parameter ;

[0088] Average meteorological conditions impact value It is the impact value of average meteorological conditions on pollutant concentrations obtained through correlation analysis of historical meteorological data and corresponding pollutant concentration data; it is calculated through comprehensive analysis of long-term meteorological data and pollutant concentration data of the same period; it depends on specific meteorological factors and the analysis results of historical data. For example, on average, the impact of meteorological conditions with a wind speed of 5 meters per second and a humidity of 60% on pollutant concentrations reduces the threshold by 5μg / m³, then ;

[0089] Current time period parameters It is used to distinguish the impact of different time periods, such as weekdays and weekends, on the pollutant concentration threshold. It is determined according to the system's time judgment module and mapped to corresponding values, such as 1 for weekdays and 0 for weekends. Assuming that the current time is a weekday, ;

[0090] Average time period impact value The pollutant concentration of different time periods in historical data is analyzed to obtain the average time period's impact on pollutant concentration; the pollutant concentration data of different time periods in the past are statistically calculated, and the specific value depends on the statistical results of historical data. For example, if the statistics show that the pollutant concentration on weekdays is 8μg / m³ higher than that on weekends, then ;

[0091] , and It is used to adjust the impact of factors such as seasons, meteorological conditions and time periods on the threshold. It is determined through experiments and data analysis, and is usually optimized based on regression analysis of historical data or machine learning algorithms.

[0092] A threshold set of different pollutants is preset, and the threshold set of different pollutants includes a normal impact threshold F1, a low impact threshold F2, a medium impact threshold F3, a high state threshold F4 and an abnormal state threshold F5, and F1<F2<F3<F4<F5.

[0093] The threshold set for different pollutants is based on dynamic thresholds. Based on dynamic thresholds Set the normal impact threshold F1, and then set the dynamic threshold according to the concentration of different pollutants exceeding the dynamic threshold The low impact threshold F2, medium impact threshold F3, high state threshold F4 and abnormal state threshold F5 are set according to the degree of harm. The difference between two adjacent thresholds can be adjusted according to the impact of the increase in pollutant concentration on the environment and pedestrians in historical data.

[0094] The early warning trigger module calculates the average concentration of different pollutants With dynamic threshold Compare and determine whether to issue an alarm; compare the estimated concentration values ​​of different pollutants Compare with the threshold set of different pollutants to determine whether to issue an early warning;

[0095] Specifically, the method for determining whether to issue an early warning is:

[0096] The estimated concentration values ​​of different pollutants Compared with the threshold sets of different pollutants, ≤F1, it is judged as normal and no warning action is issued; when F1< ≤F2, it is judged as a low pollution state and a first-level warning is issued; when F2< ≤F3, it is judged as a moderate pollution state and a second-level warning is issued; when F3< ≤F4, it is judged as a highly polluted state and a level 3 warning is issued; when F4< When ≤F5, it is judged as abnormal pollution state and an abnormal state warning is issued;

[0097] Among them, during the first-level warning, the impact on public health and the environment is relatively small. The warning method uses mobile phone application push notifications to push brief warning information to users, informing them of the type of pollutants and the situation of slight exceeding the standard, and sending emails to send detailed pollutant concentration data and related instructions; during the second-level warning, there may be a certain impact on public health and the environment. The warning method uses mobile phone application push notifications to push more urgent warning information to users to remind the public to pay attention to protection, and send emails to send more detailed pollution situation analysis and health suggestions. At the same time, concise and clear warning content, including protective measures, is sent through text messages; during the third-level warning, it poses a greater threat to public health and the environment. The warning method uses mobile phone application push notifications to frequently push emergency warning information, emphasize the importance of hazards and protection, send detailed emergency guidelines and protective measures through emails, send warning content in a timely manner through text messages, and voice calls to notify key targets by phone. Intelligent voice broadcasts are broadcast in a loop through the public broadcasting system; during abnormal status warnings, it poses a serious threat to public health and the environment. All the above-mentioned warning channels are activated at the same time, and warning information is sent at a high frequency; the emergency command system is fully activated to coordinate resources from all parties to respond.

[0098] In the above scheme, the dynamic threshold setting of the threshold setting module can be flexibly adjusted according to actual conditions, which improves the accuracy and timeliness of the early warning; combined with the early warning trigger module, by comparing the average concentration value and the estimated concentration value with the corresponding threshold, early warning and alarm can be issued in advance, which helps to take timely response measures and reduce pollution hazards.

[0099] Cause analysis module, used to establish a pollutant concentration diffusion model and determine the location of pollution sources through data analysis;

[0100] Specifically, the working steps of the cause analysis module are:

[0101] Obtain meteorological data from different monitoring points within the monitoring area, establish a pollutant concentration diffusion model based on meteorological data and air pollution data, and determine the diffusion trend of pollutant concentration and the direction of pollution sources;

[0102] Use online map services as base maps to accurately mark the locations of different monitoring points within the monitoring area on the map; use color gradients or heat maps to display the concentration distribution on the map based on the pollutant concentration data at the monitoring points; determine the location of the pollution source through data analysis, mark it with icons or markers on the map, and display relevant pollution source information;

[0103] It should be noted that the method of establishing the pollutant concentration diffusion model is to select a suitable atmospheric diffusion model, such as the Gaussian diffusion model, the Lagrangian particle diffusion model, etc.; select a suitable model according to the characteristics of the city and the pollutant emissions; collect and analyze urban terrain data, obtain the city's digital elevation model, including terrain undulations, mountains, rivers, buildings and other information, analyze the terrain's blocking, guiding and accelerating effects on airflow, determine the parameters affecting the terrain on pollutant diffusion, input the terrain, meteorological and pollution source parameters into the selected diffusion model, perform parameterization settings, and use historical pollutant concentration monitoring data to verify the model. By adjusting some uncertain parameters in the model, such as the diffusion coefficient, turbulence intensity, etc., the simulation results of the model are made to be as consistent as possible with the monitoring data to improve the accuracy of the model.

[0104] In the above scheme, online map services such as Amap and Baidu Map can be used as base maps to accurately mark the locations of urban areas and monitoring points on the map. According to the pollutant concentration data of the monitoring points, the concentration distribution can be intuitively displayed on the map using color gradients or heat maps. The higher the concentration, the darker or brighter the color. The diffusion process of pollutant concentration is displayed in real time in the form of animation, allowing users to clearly see the concentration change trend and diffusion direction. The possible pollution source locations determined by data analysis are marked on the map with special icons or marks, and related pollution source information such as company name, emission type, etc. are displayed, which helps to quickly and accurately determine the location of the pollution source and provide direction for targeted governance.

[0105] Dynamic feedback module, which receives air quality information and third-party monitoring data from the public through mobile applications, compares and verifies it with system monitoring data, and assists in judging the accuracy of collected data;

[0106] In the above scheme, the public can provide feedback on their air quality experience through mobile phone applications, and third-party monitoring data can also be integrated and compared with the system monitoring data for verification; for example, if the public reports that the actual air quality in a certain area is worse than that monitored by the system, the detection equipment in that area may be faulty or improperly distributed. After verification, the sensor calibration parameters in that area can be adjusted or the number of sensors can be increased; vice versa.

[0107] In the above scheme, the dynamic feedback module can continuously optimize the system, improve the accuracy and reliability of monitoring data, and avoid invalid warnings triggered by equipment failures.

[0108] The decision support module provides rectification suggestions based on the analysis results of the early warning trigger module and the cause analysis module;

[0109] Specifically, the decision support module obtains the warning information analyzed by the warning trigger module and the pollution source information analyzed by the cause analysis module, generates a pollution analysis report, and generates rectification suggestions based on the pollution analysis report. It can provide strong support for relevant departments to formulate effective pollution control strategies and promote the improvement of air quality.

[0110] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware or in combination with computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0111] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0112] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An intelligent early warning system for excessive concentration of air pollutants, characterized by: include: The data acquisition module is used to obtain air pollution data from different monitoring points in the monitoring area; the monitoring points in the monitoring area are arranged in a grid, and detection equipment is added in key pollution source areas, densely populated areas and transportation hubs; the detection equipment includes a variety of different types of detection devices, which perform multi-dimensional detection of different pollutants to obtain multi-dimensional concentration data of multiple pollutants; the air pollution data includes concentration data of multiple pollutants; and the edge computing device is used to screen out abnormal data and duplicate data to generate a basic data set; the method for the edge computing device to screen out abnormal data and duplicate data is as follows: Normal range thresholds for different pollutants are preset, and the real-time data of different pollutants are compared with the normal range thresholds. The data within the normal range thresholds are used to generate a normal data set; the concentration data of the same pollutant at adjacent time points in the normal data set are compared and deduplicated. The deduplication judgment formula is as follows: ; in: Is the serial number q of pollutants at a time point i The concentration data at Is the serial number q The pollutants at the time point ( ) at the time point i and time point ( ) are adjacent time points; It is the error range of the preset pollutant concentration data; the retained data is used to generate the basic data set; The data transmission module is used for data transmission of normal data sets. When a network interruption occurs during the transmission process, the air pollution data will be temporarily stored in the cache area of ​​the local data processing center and will be resumed after the network is restored. Data fusion and processing module: Receives data from the data transmission module and calculates the average concentration of different pollutants and estimated concentrations ; Among them, the average concentration of different pollutants is calculated , the formula is as follows: ; in: The value of is 1, 2, 3... k , k is a positive integer, representing the total number of time points; The serial number representing different pollutants is a positive integer; Calculate the estimated concentration values ​​of different pollutants The specific method is as follows: Predict the estimated concentration values ​​of different pollutants based on the long short-term memory network LSTM model ; The formula is as follows: ; in: The current time sequence number is q The concentration of pollutants; It's the past The time sequence number is q The concentration of pollutants, The value is a positive integer; estimated concentration value When it is a future time, the sequence number is q The concentration of pollutants; is the weight parameter of the LSTM model; Threshold setting module, used to set dynamic thresholds for different pollutants , and preset threshold sets for different pollutants; specifically: obtain the basic thresholds for different pollutants , Current season parameters , Average seasonal impact index 、Current meteorological conditions parameters , Average meteorological conditions impact value , Current time period parameters , Average time period impact value ; Calculate dynamic thresholds for different pollutants , the formula is as follows: ; in: Current season parameters Average seasonal impact index The weight coefficient of the difference is 0.2~0.4; is the current meteorological condition parameter Impact value of average meteorological conditions The weight coefficient of the difference is 0.1~0.3; Is the current time period parameter Average time period impact value The weight coefficient of the difference is 0.2~0.4; and ; The threshold set includes a normal impact threshold F1, a low impact threshold F2, a medium impact threshold F3, a high state threshold F4 and an abnormal state threshold F5, and F1<F2<F3<F4<F5; The early warning trigger module calculates the average concentration of different pollutants With dynamic threshold Compare and determine whether to issue an alarm; compare the estimated concentration values ​​of different pollutants Compare with the threshold set of different pollutants to determine whether to issue an early warning; specifically: the average concentration of different pollutants Dynamic thresholds for different pollutants When the average concentration is met, ≥ Dynamic Threshold If the conditions are met, an alarm is issued; otherwise, it is judged to be normal; the estimated concentration values ​​of different pollutants are Compared with the threshold sets of different pollutants, ≤F1, it is judged as normal and no warning action is issued; when F1< ≤F2, it is judged as a low pollution state and a first-level warning is issued; when F2< ≤F3, it is judged as a moderate pollution state and a second-level warning is issued; when F3< ≤F4, it is judged as a highly polluted state and a level 3 warning is issued; when F4< When ≤F5, it is judged as abnormal pollution state and an abnormal state warning is issued; Cause analysis module, used to establish a pollutant concentration diffusion model and determine the location of pollution sources through data analysis; Dynamic feedback module, which receives air quality information and third-party monitoring data from the public through mobile applications, compares and verifies it with system monitoring data, and assists in judging the accuracy of collected data; The decision support module provides rectification suggestions based on the analysis results of the early warning trigger module and the cause analysis module.

2. The intelligent early warning system for excessive concentration of air pollutants according to claim 1 is characterized by: The working steps of the cause analysis module are: Obtain meteorological data from different monitoring points within the monitoring area, establish a pollutant concentration diffusion model based on meteorological data and air pollution data, and determine the diffusion trend of pollutant concentration and the direction of pollution sources; Use online map services as the base map to accurately mark the locations of different monitoring points in the monitoring area on the map; use color gradients or heat maps to display the concentration distribution on the map based on the pollutant concentration data of the monitoring points; determine the location of the pollution source through data analysis, mark it with icons or markers on the map, and display relevant pollution source information.

3. The intelligent early warning system for excessive concentration of air pollutants according to claim 2 is characterized by: The decision support module obtains the warning information analyzed by the warning trigger module and the pollution source information analyzed by the cause analysis module, generates a pollution analysis report, and generates rectification suggestions based on the pollution analysis report.

Citation Information

Patent Citations

  • Pollutant monitoring method and device, terminal equipment and storage medium

    CN111768038A

  • Real-time intelligent soil pollution monitoring system and method

    CN117538503A

  • Sewage system management scheduling method for drainage basin treatment

    CN118036995A

  • Water pollution monitoring and early warning method and electronic equipment

    CN118376745A