Air quality management method and platform based on the integration of AIoT and big data services

Through the air quality management method integrated with AIoT and big data services, air quality and meteorological data are integrated, pollution characteristics and periodic pollution sources are identified, and space-time dynamic trajectories are constructed, which solves the problem that traditional monitoring methods cannot monitor and locate pollution events in a timely manner, and realizes the problem of the failure of traditional monitoring methods to monitor and locate pollution events in a timely manner. Accurate traceability and effective control of pollution sources.

CN118761890BActive Publication Date: 2024-12-03WUHAN LEITE TECH CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411253467.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-12-03
Estimated Expiration
2044-09-09

AI Technical Summary

Technical Problem

Traditional air quality monitoring methods cannot timely monitor and locate pollution events, especially sudden pollution events, and it is difficult to trace the source of pollution, resulting in the inability to take targeted governance measures.

Method used

Adopt the air quality management method based on the integration of AIoT and big data services, integrate air quality and meteorological data through the big data analysis module, identify the main pollution characteristics, determine the causal relationship, conduct Fourier transform to determine the periodic pollution source, and build the pollutant Space-time dynamic trajectory, identifying polluted hot spots.

Benefits of technology

Real-time monitoring of pollution incidents and accurate traceability of pollution sources have been achieved, the accuracy of pollution source positioning and emergency response efficiency have been improved, pollution control measures have been taken in a timely manner to improve air quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118761890B_ABST
    Figure CN118761890B_ABST
Patent Text Reader

Abstract

The present application discloses an air quality management method and platform based on the integration of AIoT and big data services, which relates to the field of cloud platform technology. The method includes: in response to receiving air quality monitoring data and meteorological parameters of a monitoring area, using a big data analysis module to integrate the air quality monitoring data and meteorological parameters to obtain an air quality data set; performing feature recognition on the air quality monitoring data to obtain main pollution features; determining the causal relationship between the main pollution features and the meteorological parameters, and determining the target meteorological parameters corresponding to the main pollution features based on the causal relationship; performing Fourier transform on the air quality monitoring data to determine periodic pollution sources; determining the spatiotemporal dynamic trajectory of pollutants based on the main pollution features, target meteorological parameters and periodic pollution sources, the spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of pollutants; and identifying pollution hotspots through the spatiotemporal dynamic trajectory.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of cloud platform technology, and in particular to an air quality management method and platform based on the integration of AIoT and big data services. Background Art

[0002] Accurate air quality data is the basis for formulating and implementing environmental protection policies. Through real-time monitoring and early warning of air quality, preventive measures can be taken in advance to reduce the harm of air pollution to health and the environment. For traditional air quality monitoring methods, air quality data is usually collected at fixed time intervals (such as once an hour) and uploaded to a central server for processing to monitor air quality.

[0003] For traditional air quality monitoring methods, for sudden pollution events, that is, if a pollution event occurs in a certain area between two sampling time points, it cannot be transmitted to the central server for processing in time. At this time, it is impossible to promptly know that a pollution event has occurred and locate the area corresponding to the pollution event.

[0004] For the above-mentioned traditional air quality monitoring methods, even if real-time monitoring of pollution events can be achieved by collecting air quality data in real time, it is still impossible to locate the area corresponding to the pollution event, and it is difficult to trace the source of pollution, resulting in difficulty in taking targeted governance measures, thus missing the best time for pollution control.

[0005] Therefore, there is an urgent need for a method that can not only monitor pollution events in real time, but also accurately trace the source of pollution. Summary of the invention

[0006] The embodiments of the present application provide an air quality management method and platform based on the integration of AIoT and big data services, which can not only monitor pollution events in real time, but also accurately trace the source of pollution.

[0007] To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

[0008] In a first aspect, an air quality management method based on the integration of AIoT and big data services is provided, which is applied to a cloud platform, and the method includes:

[0009] In response to receiving the air quality monitoring data and meteorological parameters of the monitoring area in real time, integrating the air quality monitoring data and the meteorological parameters using a big data analysis module to obtain an air quality data set;

[0010] Performing feature recognition on the air quality monitoring data to obtain main pollution features;

[0011] Determining a causal relationship between the main pollution characteristics and the meteorological parameters, and determining a target meteorological parameter corresponding to the main pollution characteristics according to the causal relationship;

[0012] Performing Fourier transformation on the air quality monitoring data to determine periodic pollution sources;

[0013] Determine the spatiotemporal dynamic trajectory of pollutants according to the main pollution characteristics, the target meteorological parameters and the periodic pollution sources, wherein the spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of the pollutants;

[0014] Pollution hotspots are identified through the spatiotemporal dynamic trajectory.

[0015] In a possible implementation manner of the first aspect, determining the causal relationship between the main pollution feature and the meteorological parameter, and determining the target meteorological parameter corresponding to the main pollution feature according to the causal relationship includes:

[0016] Determining a first lag term of each of the meteorological parameters and a second lag term of each of the main pollution characteristics, wherein the first lag term is used to characterize the meteorological parameters of the previous n days, and the second lag term is used to characterize the main pollution characteristics of the previous n days;

[0017] Autoregression analysis is performed on the second lag term of the main pollution characteristics to obtain the constrained residual sum of squares;

[0018] Performing unconstrained regression analysis on the first lag term of the meteorological parameter to obtain an unconstrained residual sum of squares;

[0019] Using a preset significance calculation formula, the causal significance value of the meteorological parameter on the main pollution characteristic is calculated according to the constrained residual sum of squares and the unconstrained residual sum of squares;

[0020] When the causal significance value is greater than a preset critical value, the corresponding meteorological parameter is determined to be a target meteorological parameter corresponding to the main pollution feature.

[0021] In another possible implementation manner of the first aspect, performing Fourier transform on the air quality monitoring data to determine the periodic pollution source includes:

[0022] Performing a discrete Fourier transform on the air quality monitoring data to convert the air quality monitoring data in the time domain into frequency components and corresponding amplitudes in the frequency domain;

[0023] Generate a frequency spectrum according to the frequency components and corresponding amplitudes, determine significant frequency components based on the frequency spectrum, and calculate the period of the significant frequency components;

[0024] A periodic pollution source is determined according to the significant frequency component and the period of the significant frequency component.

[0025] In another possible implementation manner of the first aspect, determining the periodic pollution source according to the significant frequency component and the period of the significant frequency component includes:

[0026] In response to acquiring the background information of the monitoring area, determining the periodic characteristics of the significant frequency component according to the significant frequency component and the period of the significant frequency component;

[0027] A periodic pollution source is determined according to the background information and the periodic characteristics.

[0028] In another possible implementation manner of the first aspect, the background information includes industrial activity data and traffic flow data, and determining the periodic pollution source according to the background information and the periodic characteristics includes:

[0029] Associating and matching the industrial activity data and the traffic flow data with the periodic features respectively to obtain a matching result; the matching result includes the periodic features, and the industrial activity data and / or traffic flow data corresponding to the periodic features;

[0030] Determining a periodic pollution source according to the matching result;

[0031] Among them, if the matching result is that the periodic characteristics match the industrial activity data, it is determined that the periodic pollution source comes from the industrial activity data; if the matching result is that the periodic characteristics match the traffic flow data, it is determined that the periodic pollution source comes from the traffic flow data; if the matching result is that the periodic characteristics match both the industrial activity data and the traffic flow data, it is determined that the periodic pollution source comes from the industrial activity data and the traffic flow data.

[0032] In another possible implementation of the first aspect, the background information further includes GIS data, and determining the spatiotemporal dynamic trajectory of pollutants according to the main pollution characteristics, the target meteorological parameters, and the periodic pollution sources includes:

[0033] Mining the spatiotemporal association rules of the main pollution characteristics, the target meteorological parameters and the periodic pollution sources to obtain association rules;

[0034] Determining the target meteorological parameter and the periodic pollution source that are strongly associated with the main pollution feature according to the association rule;

[0035] Constructing a pollutant generation model based on the target meteorological parameters strongly associated with the main pollution characteristics and the periodic pollution sources;

[0036] Constructing a pollutant transport and diffusion model based on the GIS data and real-time acquired wind speed data, wind direction data, and terrain data;

[0037] The pollutant generation model is combined with the pollutant transport and diffusion model to generate the spatiotemporal dynamic trajectory of the pollutant.

[0038] In another possible implementation manner of the first aspect, combining the pollutant generation model with the pollutant transport and diffusion model to generate the spatiotemporal dynamic trajectory of the pollutant includes:

[0039] Dividing the monitoring area into a plurality of identical spatial grids;

[0040] According to the pollutant generation model, the amount of pollutants generated by the pollution source in each preset time step and the generation location and generation time of the corresponding pollutants are calculated;

[0041] The generation location of the pollutant, the generation time, the GIS data, the wind speed data, the wind direction data and the terrain data are input into the pollutant transmission and diffusion model to obtain the spatiotemporal dynamic trajectory of the pollutant; wherein the pollutant transmission and diffusion model is used to calculate the pollutant concentration of the pollutant in each of the spatial grids to generate the spatiotemporal dynamic trajectory of the pollutant.

[0042] In another possible implementation manner of the first aspect, identifying a pollution hotspot area through the spatiotemporal dynamic trajectory includes:

[0043] When the pollutant concentration in the spatial grid is greater than a preset concentration threshold, the spatial grid is determined to be a hot spot grid, and the pollutant hot spot area is a collection of the hot spot grids.

[0044] In a second aspect, the present application provides a cloud platform, which is applied to the above-mentioned air quality management method based on the integration of AIoT and big data services, and the cloud platform includes:

[0045] A big data analysis module, configured to respond to receiving air quality monitoring data and meteorological parameters of a monitoring area, and integrate the air quality monitoring data and the meteorological parameters using the big data analysis module to obtain an air quality data set;

[0046] A feature recognition module is used to perform feature recognition on the air quality monitoring data to obtain main pollution features;

[0047] a causal relationship determination module, used to determine the causal relationship between the main pollution characteristics and the meteorological parameters, and determine the target meteorological parameters corresponding to the main pollution characteristics according to the causal relationship;

[0048] A calculation module, used for performing Fourier transform on the air quality monitoring data to determine periodic pollution sources;

[0049] A spatiotemporal dynamic trajectory generation module, used to determine the spatiotemporal dynamic trajectory of pollutants according to the main pollution characteristics, the target meteorological parameters and the periodic pollution sources, wherein the spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of the pollutants;

[0050] The identification and display module is used to identify pollution hot spots through the spatiotemporal dynamic trajectory.

[0051] In a third aspect, the present application provides an electronic device for deploying the above-mentioned cloud platform, the device comprising:

[0052] a memory configured to store instructions; and

[0053] The processor is configured to call the instructions from the memory and to implement the above-mentioned air quality management method based on the integration of AIoT and big data services when executing the instructions.

[0054] Through the above technical solution, the big data analysis module can be used to integrate the real-time acquired air quality monitoring data and meteorological parameters to achieve real-time monitoring of pollution events. In addition, the causal relationship between the main pollution characteristics and meteorological parameters can be determined, and the target meteorological parameters corresponding to the main pollution characteristics can be determined based on the causal relationship to determine the association between the main pollution characteristics and meteorological conditions, which helps to more accurately analyze the causes of pollution and improve the accuracy of locating pollution sources. By performing Fourier transform on air quality monitoring data, determining periodic pollution sources, and determining the spatiotemporal dynamic trajectory of pollutants based on the main pollution characteristics, target meteorological parameters and periodic pollution sources, the propagation path and source of pollutants can be accurately tracked to achieve dynamic monitoring and tracing of pollutants. Finally, the pollution hotspots can be identified through the spatiotemporal dynamic trajectory to accurately locate the pollution hotspots, facilitate timely measures, and improve the efficiency of emergency response and the effectiveness of pollution control.

[0055] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flowchart of an air quality management method based on the integration of AIoT and big data services provided in an embodiment of the present application;

[0057] Figure 2 This is an interface display diagram of a cloud platform provided in an embodiment of the present application;

[0058] Figure 3 This is an interface display diagram of a cloud platform provided in an embodiment of the present application;

[0059] Figure 4 This is a diagram showing one of the interfaces of a cloud platform provided in an embodiment of the present application. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0061] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), such directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0062] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0063] Figure 1 The following schematically shows a flow chart of an air quality management method based on the integration of AIoT and big data services according to an embodiment of the present application. Figure 1 As shown, an embodiment of the present application provides an air quality management method based on the integration of AIoT and big data services, which is applied to a cloud platform. The method may include the following steps.

[0064] S110, in response to receiving the air quality monitoring data and meteorological parameters of the monitoring area in real time, integrating the air quality monitoring data and meteorological parameters using a big data analysis module to obtain an air quality data set;

[0065] S120, performing feature recognition on the air quality monitoring data to obtain main pollution features;

[0066] S130, determining the causal relationship between the main pollution characteristics and the meteorological parameters, and determining the target meteorological parameters corresponding to the main pollution characteristics according to the causal relationship;

[0067] S140, performing Fourier transform on the air quality monitoring data to determine the periodic pollution source;

[0068] S150, determining the spatiotemporal dynamic trajectory of pollutants based on the main pollution characteristics, target meteorological parameters and periodic pollution sources, the spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of pollutants;

[0069] S160. Identify pollution hotspots through spatiotemporal dynamic trajectories.

[0070] Figure 2 One of the interface display diagrams of a cloud platform provided in this embodiment is shown. Figure 2 As shown in the interface display diagram, it can be seen that the air quality monitoring data can include the concentrations of PM2.5, PM10, carbon monoxide (CO), nitrogen dioxide (NO2), sulfur dioxide (SO2) and ozone (O3).

[0071] The meteorological parameters of this embodiment may be temperature, humidity, wind speed, wind direction, rainfall, etc. Data integration refers to time synchronization and spatial matching of air quality monitoring data and meteorological parameters to integrate them into a unified data set. Specifically, the air quality monitoring data and meteorological parameters can be integrated according to the time and space dimensions through the big data analysis module to form a multi-dimensional air quality data set. The air quality data set not only contains the pollutant concentration at each time point, but also includes the corresponding meteorological parameters.

[0072] The main pollution characteristics refer to the characteristic indicators of key pollutants or pollutant combinations that can represent the pollution situation in air quality monitoring, such as Figure 4As shown in the primary pollutant analysis chart, in October 2015, NO2 can be determined as the main pollution feature by the proportion of NO2. In this embodiment, the underlying principle is to calculate the concentration proportion of each type of pollutant and the change trend of each type of pollutant in different time periods through statistical analysis methods to obtain the law of pollutant concentration change. Then, machine learning algorithms, such as principal component analysis, can be used to reduce the dimension of air quality monitoring data of multiple pollutants, extract the main pollution features that can represent the overall pollution situation, and display the change trend of the main pollution features in the cloud platform.

[0073] Causation refers to how meteorological parameters affect changes in key pollution characteristics. For example, high humidity may increase PM2.5 concentrations because particles condense more easily when humidity is high, or wind speed and direction determine the dispersion and transport of pollutants.

[0074] Specifically, Granger causality test can be used to analyze the time series data between the main pollution characteristics and various meteorological parameters to determine whether the meteorological parameters have a significant impact on the pollution characteristics. For example, if it is found that the high humidity and low wind speed in the previous few days have a significant causal relationship with the current PM2.5 concentration peak, then humidity and wind speed are determined as target meteorological parameters.

[0075] Fourier transform is used to convert time domain signals into frequency domain signals. Through Fourier transform, periodic change patterns in air quality monitoring data can be identified. In specific implementation, the air quality monitoring data is first subjected to discrete Fourier transform to obtain the amplitude and phase information of each frequency component. Then, by analyzing the spectrum graph, significant periodic components can be identified. For example, if a frequency component is found to have a large amplitude in the spectrum graph, and its period coincides with the period of certain industrial emission activities, it can be inferred that the pollution source corresponding to the frequency component has periodic emission characteristics. In this way, through Fourier transform, periodic pollution sources can be effectively identified.

[0076] According to the main pollution characteristics, target meteorological parameters and periodic pollution sources, the spatiotemporal dynamic trajectory of pollutants is determined. The spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of pollutants.

[0077] In the first embodiment, the diffusion process of pollutants in the atmosphere can be simulated based on the main pollution characteristics and target meteorological parameters. For example, the Lagrangian particle tracking method can be used to treat pollutants as a series of tiny particles and track their movement trajectories in the airflow to simulate the transmission path of pollutants. Then, combined with periodic pollution sources, the simulation results are further corrected to obtain a more accurate spatiotemporal dynamic trajectory of pollutants. In this way, the whole process of pollutants from source generation, atmospheric transmission, and final diffusion to various regions can be described in detail.

[0078] In the second embodiment, the main pollution characteristics, target meteorological parameters and periodic pollution sources can be firstly correlated and analyzed, and a pollutant generation model can be constructed based on the results of the correlation analysis. Then, a current pollutant transmission and diffusion model can be constructed based on the geographic data of the monitoring area and the meteorological parameters obtained in real time. The spatiotemporal dynamic trajectory of the pollutants can be obtained based on the pollutant generation model and the pollutant transmission and diffusion model. For details, please refer to the description of S610-S650 below.

[0079] Pollution hotspots refer to areas with high pollutant concentrations, usually where pollution sources are concentrated or where atmospheric conditions are unfavorable for pollutant diffusion. In specific implementation, pollutant concentrations can be determined based on spatiotemporal dynamic trajectories, and pollution hotspots can be determined based on pollutant concentrations. Figure 2 In the schematic diagram of the spatiotemporal dynamic trajectory of pollutants shown, the location of the black five-pointed star is the pollution hotspot.

[0080] For the cloud platform provided in the embodiment of the present application, the following can also be generated through the air quality monitoring data obtained in real time: Figure 2 The air quality trend graph shown is used to intuitively display the air quality trend in the monitoring area. The air quality details can also be generated through the real-time air quality monitoring data, where the air quality details can include air quality index, air quality and air quality level, and different cities in the monitoring area in the air quality details can be intuitively displayed through the air quality map.

[0081] for Figure 2 As for the air quality shown, it can also be Figure 3 The pie chart in the figure is used for intuitive display.

[0082] This embodiment can use a big data analysis module to integrate the real-time acquired air quality monitoring data and meteorological parameters to achieve real-time monitoring of pollution events. In addition, the causal relationship between the main pollution characteristics and meteorological parameters can be determined, and the target meteorological parameters corresponding to the main pollution characteristics can be determined based on the causal relationship to determine the association between the main pollution characteristics and meteorological conditions, which helps to more accurately analyze the causes of pollution and improve the accuracy of locating pollution sources. By performing Fourier transform on air quality monitoring data, determining periodic pollution sources, and determining the spatiotemporal dynamic trajectory of pollutants based on the main pollution characteristics, target meteorological parameters and periodic pollution sources, the propagation path and source of pollutants can be accurately tracked, and dynamic monitoring and tracing of pollutants can be achieved. Finally, the pollution hotspots can be identified through the spatiotemporal dynamic trajectory to accurately locate the pollution hotspots, facilitate timely measures, and improve the efficiency of emergency response and the effectiveness of pollution control.

[0083] In one implementation of this embodiment, determining the causal relationship between the main pollution characteristics and the meteorological parameters, and determining the target meteorological parameters corresponding to the main pollution characteristics according to the causal relationship, includes the following steps:

[0084] S210, determining a first lag term of each meteorological parameter and a second lag term of each main pollution characteristic, the first lag term is used to characterize the meteorological parameter of the previous n days, and the second lag term is used to characterize the main pollution characteristics of the previous n days;

[0085] S220, perform autoregression analysis on the second lag term of the main pollution characteristics to obtain the constrained residual sum of squares;

[0086] S230, performing unconstrained regression analysis on the first lag term of the meteorological parameter to obtain an unconstrained residual sum of squares;

[0087] S240, using a preset significance calculation formula, and calculating the causal significance value of the meteorological parameter on the main pollution characteristics according to the constrained residual sum of squares and the unconstrained residual sum of squares;

[0088] S250. When the causal significance value is greater than a preset critical value, determine that the corresponding meteorological parameter is a target meteorological parameter corresponding to the main pollution characteristic.

[0089] Determine the first lag term of each meteorological parameter and the second lag term of each main pollution feature. The first lag term is used to characterize the meteorological parameters of the previous n days, and the second lag term is used to characterize the main pollution features of the previous n days. In specific implementation, it is assumed that there are several meteorological parameters such as temperature, humidity, wind speed, etc., as well as several main pollution features such as PM2.5, PM10, etc. For each meteorological parameter, calculate its value in the previous n days. For example, if today is the tth day, then the first lag term is the meteorological parameter value on the t-1th day, the t-2th day... the tnth day. Similarly, for each main pollution feature, calculate its value in the previous n days. For example, if today is the tth day, then the second lag term is the pollution feature value on the t-1th day, the t-2th day... the tnth day.

[0090] Autoregression analysis is a statistical method used to analyze the autocorrelation in time series data. In specific implementation, assuming that there is a main pollution feature, such as the second lag data of PM2.5, when performing autoregression analysis, first build an autoregression model and estimate the model parameters by the least squares method to obtain a fitted model. Then, calculate the residual of the model, that is, the difference between the actual value and the fitted value. Next, calculate the sum of squares of the residuals, that is, sum each residual after squaring it to obtain a constrained residual sum of squares. The constrained residual sum of squares is used to measure the fitting effect of the autoregression model. The smaller the value, the better the model fitting effect, which can be used to evaluate the autocorrelation of the main pollution features.

[0091] Unconstrained regression analysis is a statistical method used to analyze the relationship between independent variables and dependent variables. In specific implementation, it is assumed that there are first lag terms of several meteorological parameters, such as temperature, humidity, wind speed, etc., and a second lag term of a major pollution characteristic. When performing unconstrained regression analysis, first establish a multivariate regression model, such as Y=β0+β1X1+β2X2+…+β n X n , where Y represents the second lag of the main pollution characteristics, X1, X2, ..., X n represents the first lag term of meteorological parameters, β0,β1,...,β n Represents the model parameters. The model parameters are estimated by the least squares method to obtain the fitted model. Then, the residual of the model is calculated, that is, the difference between the actual value and the fitted value. Next, the sum of squares of the residuals is calculated, that is, each residual is squared and summed to obtain the unconstrained residual sum of squares. The unconstrained residual sum of squares is used to measure the fitting effect of the unconstrained regression model. The smaller the value, the better the model fitting effect. It is used to evaluate the impact of meteorological parameters on the main pollution characteristics.

[0092] The significance calculation formula is preset and includes:

[0093] ;

[0094] Where q is the number of first lag terms of meteorological parameters, n is the sample size (e.g. 365 days), k is the number of parameters to be estimated in the unconstrained regression (the number of meteorological parameters + the number of main pollution characteristics + the number of first lag terms + the number of second lag terms), R is the constrained residual sum of squares, and UR is the unconstrained residual sum of squares.

[0095] By calculating the F value, we can evaluate the significance of meteorological parameters on the main pollution characteristics. The larger the significance value, the more significant the impact of meteorological parameters on the main pollution characteristics.

[0096] When the causal significance value is greater than the preset critical value, the corresponding meteorological parameter is determined to be the target meteorological parameter corresponding to the main pollution characteristics. In specific implementation, the preset critical value can be determined according to the significance level. For example, when the significance level is 0.05, the corresponding critical value can be obtained by consulting the F distribution table. Assuming that the calculated causal significance value is greater than the preset critical value, it indicates that the meteorological parameter has a significant impact on the main pollution characteristics. At this time, the meteorological parameter can be determined as the target meteorological parameter corresponding to the main pollution characteristics. For example, assuming that the causal significance values ​​of temperature, humidity and wind speed are all greater than the preset critical value, it can be determined that temperature, humidity and wind speed are the target meteorological parameters corresponding to PM2.5 concentration.

[0097] The determination of the first lag term and the second lag term in this implementation provides the basis of historical data; the autoregression analysis and unconstrained regression analysis reveal the autocorrelation of pollution characteristics and the influence of meteorological parameters; the application of the significance calculation formula quantifies the causal significance of meteorological parameters. Finally, the target meteorological parameters are determined by judging the preset critical value, which can effectively improve the scientificity and accuracy of pollution analysis.

[0098] In one implementation of this embodiment, Fourier transform is performed on air quality monitoring data to determine periodic pollution sources, including the following steps:

[0099] S310, performing discrete Fourier transform on the air quality monitoring data to convert the air quality monitoring data in the time domain into frequency components and corresponding amplitudes in the frequency domain;

[0100] S320, generating a frequency spectrum according to the frequency components and the corresponding amplitudes, determining significant frequency components based on the frequency spectrum, and calculating the period of the significant frequency components;

[0101] S330. Determine a periodic pollution source based on a significant frequency component and a period of the significant frequency component.

[0102] Discrete Fourier Transform (DFT) is a mathematical transformation used to analyze the frequency components of discrete time signals. In specific implementation, assume that there is a set of air quality monitoring data, such as PM2.5 concentration values, which records changes over a period of time. Through DFT, the time domain data can be converted into frequency domain data to obtain a series of frequency components and their corresponding amplitudes.

[0103] The discrete Fourier transform formula of this embodiment includes:

[0104] ;

[0105] In the formula, x(n) represents the air quality monitoring data at the nth time point, X(k) represents the kth frequency component, N represents the total number of data points, e is the base of the natural logarithm, and j is the imaginary unit. By calculating the amplitude of each frequency component, the frequency distribution in the frequency domain can be obtained.

[0106] The spectrum graph is used to display frequency components and their corresponding amplitudes. In specific implementation, the frequency components and amplitudes obtained by DFT are plotted into a spectrum graph, with the horizontal axis representing the frequency and the vertical axis representing the amplitude. By observing the spectrum graph, components with large amplitudes at certain frequencies, namely significant frequency components, can be identified. Significant frequency components indicate that at these frequencies, the air quality monitoring data has a strong periodicity. To calculate the period of the significant frequency components, the following formula can be used:

[0107] ;

[0108] Where T represents the period and f represents the frequency of the significant frequency component. For example, if the frequency of a frequency component is found to be 0.1 Hz on the spectrum graph, the corresponding period is 10 seconds.

[0109] In this embodiment, by analyzing the period of significant frequency components, the periodic characteristics of the pollution source can be inferred. For example, if a significant frequency component with a period of 24 hours is found in the air quality monitoring data, it can be inferred that there may be a source that regularly discharges pollutants every day. Other information, such as the geographical location of the pollution source, the time of emission, etc., can be combined to further determine the specific pollution source. For example, by analyzing the data of multiple monitoring points, the emission activities of a certain industrial area within a certain time period every day can be located to effectively identify periodic pollution sources.

[0110] In this embodiment, the discrete Fourier transform converts the time domain data into the frequency domain data, making the periodic characteristics more obvious; the calculation of the significant frequency component period and the determination of the pollution source are used to identify the specific periodic pollution source, which effectively improves the scientificity and accuracy of the pollution source analysis. By identifying the periodic pollution source, targeted measures can be taken to reduce the emission of pollutants and improve air quality.

[0111] In one implementation of this embodiment, determining a periodic pollution source according to a significant frequency component and a period of the significant frequency component includes the following steps:

[0112] S410, in response to obtaining background information of the monitoring area, determining the periodic characteristics of the significant frequency component according to the significant frequency component and the period of the significant frequency component;

[0113] S420. Determine periodic pollution sources based on background information and periodic characteristics.

[0114] In this embodiment, the background information includes industrial activity data, traffic flow data and GIS data. Specifically, GIS data refers to the acquisition of geographic information of the monitoring area; industrial activity data refers to the working hours and pollutant emission patterns of industrial activities (construction, etc.) in the monitoring area; traffic flow data refers to road traffic flow and peak hours.

[0115] The periodic feature may be a 24 hour period, a 7 day period or other period.

[0116] According to the background information and periodic characteristics, the periodic pollution sources are determined. In specific implementation, the background information can be analyzed to identify possible pollution sources. For example, in industrial areas, the emission time and emission amount of the factory are monitored; in areas with heavy traffic, traffic flow and exhaust emissions are monitored. Combined with the periodic characteristics of the significant frequency components determined previously, the scope of the pollution sources can be further narrowed. For example, if a significant frequency component with a period of 8 hours is found in the air quality monitoring data of an industrial area, it can be inferred that the frequency component may be related to the shift system or scheduled emission activities of a certain factory. In order to verify this inference, the production and emission records of the factory can be used to determine whether there are emission activities consistent with the period of the significant frequency component.

[0117] In another embodiment, a spatial analysis method can also be used to spatially match the periodic characteristics of significant frequency components with the distribution of pollution sources in the monitoring area to determine the specific location of the pollution source. In other words, a geographically weighted regression model can be first constructed, and the pollutant concentration in the air quality monitoring data is set as the dependent variable, and the characteristics of potential pollution sources in the monitoring area (such as factory emissions, traffic flow, etc.) are set as independent variables. The weighted regression model is used to calculate the local regression results of each geographical location in the monitoring area, and based on the local regression results, it is determined which pollution source characteristics have a significant impact on air quality at different locations.

[0118] Then, the spatial autocorrelation index of the air quality monitoring data is calculated to determine whether there is spatial autocorrelation in pollutant concentrations. If the spatial autocorrelation index is significant, the high and low value clustering areas of pollutant concentrations are further identified. Finally, the local regression results of the weighted regression model are combined with the high and low value clustering areas of pollutant concentrations to determine the specific location of the pollution source.

[0119] In response to obtaining background information of the monitoring area, this embodiment determines the periodic characteristics of the significant frequency components according to the significant frequency components and the period of the significant frequency components, so that the periodic characteristics are combined with the actual situation and more accurately reflect the periodic characteristics of the pollution source. According to the background information and periodic characteristics, the periodic pollution source is determined, the scope of the pollution source is further narrowed, and the specific location of the pollution source is determined, which effectively improves the accuracy of the pollution source analysis. By identifying the periodic pollution source, targeted measures can be taken to reduce the emission of pollutants and improve air quality.

[0120] In one implementation of this embodiment, the background information includes industrial activity data and traffic flow data. According to the background information and the periodic characteristics, determining the periodic pollution source includes the following steps:

[0121] S510, associating and matching the industrial activity data and the traffic flow data with the periodic characteristics respectively to obtain a matching result; the matching result includes the periodic characteristics, and the industrial activity data and / or traffic flow data corresponding to the periodic characteristics;

[0122] S520, determining a periodic pollution source according to the matching result;

[0123] Among them, if the matching result is that the periodic characteristics match the industrial activity data, then it is determined that the periodic pollution source comes from the industrial activity data; if the matching result is that the periodic characteristics match the traffic flow data, then it is determined that the periodic pollution source comes from the traffic flow data; if the matching result is that the periodic characteristics match both the industrial activity data and the traffic flow data, then it is determined that the periodic pollution source comes from both the industrial activity data and the traffic flow data.

[0124] In specific implementation, industrial activity data includes information such as the factory's production schedule, emission records, and equipment operating hours; traffic flow data includes information such as vehicle volume on the road, peak traffic hours, and vehicle types. By matching the industrial activity data and traffic flow data with the significant frequency components and their periodic characteristics previously obtained through spectrum analysis, it is possible to determine which industrial activities are consistent with the periodic characteristics. For example, if a significant frequency component with a period of 12 hours is found in the air quality monitoring data, the 12-hour periodic characteristics can be compared with the industrial activity data and traffic flow data to determine whether there are any factory emission activities or traffic flows that change in the same period, and the correlation coefficient between the period of the significant frequency component and the factory emission records or traffic flow data can be calculated to determine the degree of association between them.

[0125] The matching results are used to determine the matching of periodic features with industrial activity data or traffic flow data. If the matching results show that the periodic features match the industrial activity data, it can be determined that the periodic pollution source comes from industrial activities. For example, if the period of a significant frequency component is found to be consistent with the emission period of a certain factory, it can be inferred that the factory is a periodic pollution source. In order to further verify the above inference, other information, such as the geographical location of the factory, the type of emissions, etc., can be combined to ensure the accuracy of the matching results.

[0126] If the matching results show that the periodic features match the traffic flow data, it can be determined that the periodic pollution source comes from the traffic flow. For example, if the period of a significant frequency component is found to be consistent with the changes during the peak traffic period, it can be inferred that the traffic flow is a periodic pollution source. Similarly, other information, such as road type, vehicle type, etc., can be combined to further verify the above inference.

[0127] If the matching results show that the periodic characteristics match both the industrial activity data and the traffic flow data, it can be determined that the periodic pollution source comes from the joint effect of industrial activities and traffic flow. For example, if a significant frequency component is found to have a period that is consistent with both the factory emission cycle and the changes in traffic peak hours, it can be inferred that these two factors jointly cause periodic changes in air quality.

[0128] This implementation method combines industrial activity data and traffic flow data to determine specific periodic pollution sources. The industrial activity data and traffic flow data are respectively associated and matched with the periodic characteristics to obtain matching results, so that the periodic characteristics are combined with actual activities to more accurately reflect the periodic characteristics of the pollution source. According to the matching results, the periodic pollution source is determined, the scope of the pollution source is further narrowed, and the specific location of the pollution source is determined. It not only improves the accuracy of pollution source analysis, but also provides an important decision-making basis for pollution control. By identifying periodic pollution sources, targeted measures can be taken to reduce the emission of pollutants and improve air quality.

[0129] In one implementation of this embodiment, the background information also includes GIS data, and the spatiotemporal dynamic trajectory of pollutants is determined according to the main pollution characteristics, target meteorological parameters and periodic pollution sources, including the following steps:

[0130] S610, mining the spatiotemporal association rules of the main pollution characteristics, target meteorological parameters and periodic pollution sources to obtain association rules;

[0131] S620, determining target meteorological parameters and periodic pollution sources that are strongly associated with the main pollution characteristics according to the association rules;

[0132] S630, constructing a pollutant generation model based on target meteorological parameters and periodic pollution sources that are strongly associated with major pollution characteristics;

[0133] S640, build pollutant transport and diffusion models based on GIS data and real-time wind speed data, wind direction data, and terrain data;

[0134] S650, combining the pollutant generation model with the pollutant transport and diffusion model to generate the spatiotemporal dynamic trajectory of the pollutants.

[0135] First, the main pollution characteristics, target meteorological parameters and periodic pollution sources are mined through spatiotemporal association rules to obtain association rules. In specific implementation, the main pollution characteristics include information such as the type, concentration, and distribution of pollutants; the target meteorological parameters include meteorological data such as temperature, humidity, wind speed, and wind direction; and the periodic pollution sources include industrial activity data and traffic flow data. Through spatiotemporal association rule mining, the temporal and spatial associations between the main pollution characteristics, target meteorological parameters, and periodic pollution sources can be determined. The Apriori algorithm can be used to mine the data to determine the patterns and rules that appear frequently.

[0136] Assuming that in a certain area, the peak of pollutant concentration is significantly associated with a specific wind direction and the emission time of a certain factory, the association relationship can be extracted through association rule mining. The specific mining process includes steps such as data preprocessing, rule generation and rule screening. In the data preprocessing stage, the original data needs to be cleaned, standardized and formatted to make it suitable for the input requirements of the mining algorithm; in the rule generation stage, the association rule mining algorithm is used to generate preliminary association rules; in the rule screening stage, significant association rules are screened out based on indicators such as support and confidence.

[0137] According to the association rules, target meteorological parameters and periodic pollution sources that are strongly associated with the main pollution characteristics are determined. In specific implementation, it is necessary to analyze the association rules to determine the meteorological parameters and pollution sources that are highly correlated with the main pollution characteristics. For example, if an association rule shows that under certain wind direction and temperature conditions, the emissions of a factory will cause a significant increase in pollutant concentrations, it can be determined that the meteorological parameters and pollution sources are strongly associated with the main pollution characteristics.

[0138] In this embodiment, in order to verify the association rules, correlation analysis can be used to further quantify the relationship between the target meteorological parameters that are strongly associated with the main pollution characteristics and the periodic pollution sources. For example, a regression model can be established to analyze the influence of meteorological parameters and pollution sources on pollutant concentrations and determine which factors have a significant impact on the change in pollutant concentrations.

[0139] Based on the target meteorological parameters and periodic pollution sources that are strongly associated with the main pollution characteristics, a pollutant generation model is constructed. In specific implementation, the target meteorological parameters and periodic pollution sources that are strongly associated with the main pollution characteristics can be used as independent variables, and the pollutant concentration can be used as the dependent variable to establish a pollutant generation model. For example, the pollutant generation model can be a machine learning algorithm, which is a nonlinear model. The model training stage requires the use of historical data for training. After the model is trained, a pollutant generation model that can accurately predict changes in pollutant concentrations can be constructed.

[0140] In specific implementation, GIS data includes information such as the geographical location, topography, and land use of the monitoring area; real-time wind speed data and wind direction data can be obtained through meteorological stations or sensors. The pollutant transmission and diffusion model needs to consider the transmission and diffusion process of pollutants in the air. The pollutant transmission and diffusion model can be a Gaussian diffusion model, a Lagrangian particle model, etc. For example, the Gaussian diffusion model assumes that the diffusion of pollutants in the air follows a Gaussian distribution. By inputting parameters such as the location of the pollution source, emissions, wind speed, and wind direction, the concentration distribution of pollutants at different locations and times can be calculated. The Lagrangian particle model simulates the movement trajectory of pollutant particles, considers the influence of factors such as wind speed, wind direction, and terrain, and obtains the spatiotemporal distribution of pollutants. In the model construction stage, it is necessary to combine GIS data with real-time meteorological data, input the location and emissions of pollution sources, and calculate the transmission and diffusion process of pollutants in the air. In the model verification stage, monitoring data needs to be used for verification to evaluate the prediction performance of the model and ensure that the model can accurately reflect the transmission and diffusion laws of pollutants.

[0141] Finally, the pollutant generation model is combined with the pollutant transmission and diffusion model to generate the spatiotemporal dynamic trajectory of pollutants. In specific implementation, the pollutant concentration predicted by the generation model can be used as input, combined with the transmission and diffusion model to calculate the concentration changes of pollutants at different locations and times. For example, the generation model predicts the emissions of a factory in a specific time period, and the transmission and diffusion model calculates the diffusion process of pollutants from the factory emission point to the surrounding area based on meteorological parameters such as wind speed and wind direction, and obtains the concentration distribution of pollutants at different locations and times. In the above way, the spatiotemporal dynamic trajectory of pollutants can be generated to reflect the transmission and diffusion process of pollutants in the air.

[0142] This implementation method combines the main pollution characteristics, target meteorological parameters, periodic pollution sources and GIS data to generate the spatiotemporal dynamic trajectory of pollutants. The main pollution characteristics, target meteorological parameters and periodic pollution sources are mined for spatiotemporal association rules to obtain association rules so that the pollutant generation model can accurately reflect the changing pattern of pollutant concentrations. Based on GIS data and real-time wind speed data, wind direction data, and terrain data, a pollutant transmission and diffusion model is constructed to ensure that the model can accurately reflect the transmission and diffusion process of pollutants. The pollutant generation model is combined with the transmission and diffusion model to generate the spatiotemporal dynamic trajectory of pollutants, providing a scientific basis for pollution control. By identifying the spatiotemporal dynamic trajectory of pollutants, targeted measures can be taken to reduce pollutant emissions and improve air quality.

[0143] In one implementation of this embodiment, the pollutant generation model is combined with the pollutant transmission and diffusion model to generate the spatiotemporal dynamic trajectory of the pollutant, including the following steps:

[0144] S710, dividing the monitoring area into a plurality of identical spatial grids;

[0145] S720, calculating the amount of pollutants generated by the pollution source in each preset time step and the generation location and generation time of the corresponding pollutants according to the pollutant generation model;

[0146] S730. Input the generation location, generation time, GIS data, wind speed data, wind direction data and terrain data of the pollutants into the pollutant transmission and diffusion model to obtain the spatiotemporal dynamic trajectory of the pollutants; wherein the pollutant transmission and diffusion model is used to calculate the pollutant concentration of the pollutants in each spatial grid to generate the spatiotemporal dynamic trajectory of the pollutants.

[0147] This embodiment can divide the entire area into several grid units of the same size according to the geographical scope of the monitoring area and the required resolution. Each grid unit represents a small geographical area, usually divided in the form of a square or rectangle. The size of the divided grid can be adjusted according to actual needs. For example, a grid with a side length of 100 meters, 500 meters or 1 kilometer can be selected. Dividing the grid can refine the entire monitoring area, making subsequent pollutant transmission and diffusion calculations more accurate and convenient.

[0148] Specifically, GIS software or programming languages ​​(such as the Geopandas library in Python) can be used for grid division. The specific steps include: first determine the boundaries of the monitoring area, and then divide it into equal intervals according to the preset grid size to generate multiple grid cells. Each grid cell will have a unique number and geographic coordinates for subsequent pollutant concentration calculations and spatiotemporal dynamic trajectory generation. The monitoring area can be refined into multiple small areas, so that subsequent pollutant transport and diffusion calculations can be performed at a finer spatial scale, improving the accuracy and resolution of the calculation.

[0149] According to the pollutant generation model, the amount of pollutants generated by the pollution source in each preset time step and the generation location and generation time of the corresponding pollutants are calculated. In specific implementation, the pollutant generation model has determined the emission law of the pollution source and the change law of the pollutant concentration. The preset time step can be selected according to actual needs. For example, every hour, every half hour or every minute can be selected as the time step. In each time step, the emission of the pollution source is calculated according to the pollutant generation model. For example, assuming that a factory emits 100 kilograms of sulfur dioxide per hour, the sulfur dioxide emissions of the factory are 100 kilograms in each hour time step. At the same time, the generation location and generation time of the pollutant are recorded. For example, in a certain time step, the location of the factory is (X1, Y1) and the generation time is T1. In this way, the amount of pollutants generated by the pollution source and the corresponding generation location and generation time in each time step can be obtained.

[0150] The generation location, generation time, GIS data, wind speed data, wind direction data and terrain data of pollutants are input into the pollutant transmission and diffusion model to obtain the spatiotemporal dynamic trajectory of pollutants. In specific implementation, the pollutant transmission and diffusion model needs to consider the transmission and diffusion process of pollutants in the air. Commonly used transmission and diffusion models include Gaussian diffusion model, Lagrangian particle model, etc. For example, assuming that within a certain time step, the wind speed is 5 meters per second and the wind direction is north wind, the concentration distribution of pollutants at different locations can be calculated according to the Gaussian diffusion model.

[0151] The pollutant transport and diffusion model is used to calculate the pollutant concentration in each spatial grid to generate the spatiotemporal dynamic trajectory of the pollutants. In specific implementation, the pollutant transport and diffusion model needs to divide the monitoring area into multiple spatial grids and calculate the pollutant concentration in each grid cell within each time step. For example, the Gaussian diffusion model can be used to calculate the pollutant concentration in each grid cell. Assuming that the center coordinates of a grid cell are (X2, Y2), the pollutant concentration in the grid cell can be calculated based on parameters such as the location of the pollution source, emissions, wind speed, and wind direction. The calculation formula can be expressed as:

[0152] ;

[0153] Among them, C(x,y,z,t) represents the pollutant concentration at the location (x,y,z) and time t, Q represents the emission of the pollution source, and Respectively represent the lateral and vertical diffusion coefficients, and u represents the wind speed. In this way, the pollutant concentration in each grid unit can be calculated to generate the spatiotemporal dynamic trajectory of the pollutants. In other words, this embodiment can simulate the transmission and diffusion process of pollutants in detail in space and time to obtain the change of pollutant concentration in each grid unit in the monitoring area.

[0154] Comprehensive effect: Through the above steps, the generation and transmission process of pollutants can be systematically simulated, and the spatiotemporal dynamic trajectory of pollutants can be generated.

[0155] This implementation method first divides the monitoring area into multiple identical spatial grids so that pollutant transmission and diffusion calculations can be performed on a finer spatial scale. Then, according to the pollutant generation model, the amount of pollutants generated by the pollution source in each preset time step and the corresponding generation location and generation time are calculated to provide detailed input data for subsequent transmission and diffusion calculations. Next, the generation location, generation time, GIS data, wind speed data, wind direction data and terrain data of the pollutants are input into the pollutant transmission and diffusion model, the pollutant concentration in each spatial grid is calculated, and the spatiotemporal dynamic trajectory of the pollutants is generated, so that the spatiotemporal dynamic changes of pollutants in the monitoring area can be obtained, which provides a scientific basis for pollution control, helps to determine effective pollution control measures, and improves air quality.

[0156] In one implementation of this embodiment, identifying pollution hotspots through spatiotemporal dynamic trajectories includes the following steps:

[0157] S810: When the pollutant concentration in the spatial grid is greater than a preset concentration threshold, the spatial grid is determined to be a hotspot grid, and the pollutant hotspot area is a collection of hotspot grids.

[0158] In this embodiment, different pollutants have different concentration thresholds. Assuming that within a certain time step, the pollutant concentration of a certain grid unit is 0.6 mg / m3, which exceeds the preset threshold of 0.5 mg / m3, the grid unit is marked as a hotspot grid, so that all hotspot grids exceeding the preset concentration threshold can be identified within each time step. The cloud platform of this embodiment can automatically traverse all grid units and compare concentrations to identify hotspot grids.

[0159] All cells identified as hotspot grids in any time step can be set operated to obtain pollutant hotspot areas. For example, assuming that in a certain time step, grid cells A, B, and C are marked as hotspot grids, and in another time step, grid cells B, C, and D are marked as hotspot grids, these hotspot grids can be set operated to obtain the pollutant hotspot area as the set of grid cells A, B, C, and D. In order to more intuitively display the pollutant hotspot areas, the above hotspot grids can be visualized on the GIS map. In specific operations, GIS software can be used to draw the hotspot grids on the map to generate a visualization of the pollutant hotspot areas. For example, the hotspot grids can be marked in red and superimposed on the basic map of the monitoring area to intuitively display the spatial distribution of the pollutant hotspot areas, such as Figure 2 The spatiotemporal dynamic trajectories of pollutants are shown.

[0160] This implementation method first sets a reasonable concentration threshold and uses the spatiotemporal dynamic trajectory data of pollutants to identify the pollution hotspot grids in the monitoring area. Then, all cells identified as hotspot grids within any time step are set to obtain the pollutant hotspot area. Finally, a GIS map is used for visualization to intuitively display the spatial distribution of pollutant hotspot areas. In this way, areas with high pollutant concentrations can be accurately identified and located.

[0161] The embodiment of the present application further provides a cloud platform, which is applied to the above-mentioned air quality management method based on the integration of AIoT and big data services, and the cloud platform includes:

[0162] A big data analysis module, for responding to receiving the air quality monitoring data and meteorological parameters of the monitoring area, integrating the air quality monitoring data and meteorological parameters using the big data analysis module to obtain an air quality data set;

[0163] The feature recognition module is used to perform feature recognition on air quality monitoring data to obtain the main pollution features;

[0164] A causal relationship determination module is used to determine the causal relationship between the main pollution characteristics and the meteorological parameters, and determine the target meteorological parameters corresponding to the main pollution characteristics based on the causal relationship;

[0165] A calculation module for performing Fourier transform on air quality monitoring data to determine periodic pollution sources;

[0166] The spatiotemporal dynamic trajectory generation module is used to determine the spatiotemporal dynamic trajectory of pollutants based on the main pollution characteristics, target meteorological parameters and periodic pollution sources. The spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of pollutants;

[0167] The identification and display module is used to identify pollution hot spots through spatiotemporal dynamic trajectories.

[0168] The embodiment of the present application further provides an electronic device for deploying the above-mentioned cloud platform, the device comprising:

[0169] a memory configured to store instructions; and

[0170] The processor is configured to call instructions from the memory and implement the above-mentioned air quality management method based on the integration of AIoT and big data services when executing the instructions.

[0171] An embodiment of the present application also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the above-mentioned air quality management method based on the integration of AIoT and big data services.

[0172] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0173] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0174] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0175] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0176] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0177] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0178] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0179] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0180] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. An air quality management method based on the integration of AIoT and big data services, characterized in that: Applied to a cloud platform, the method includes: In response to receiving air quality monitoring data and meteorological parameters of the monitoring area in real time, a big data analysis module is used to fuse the air quality monitoring data and the meteorological parameters according to time and space dimensions to obtain a multi-dimensional air quality data set, wherein the air quality data set includes pollutant concentrations and corresponding meteorological parameters at each time point; Performing feature recognition on the air quality monitoring data to obtain main pollution features, wherein the main pollution features are used to characterize key pollutants or combinations of pollutants representing pollution conditions; Determining a causal relationship between the main pollution characteristics and the meteorological parameters, and determining a target meteorological parameter corresponding to the main pollution characteristics according to the causal relationship; Performing Fourier transformation on the air quality monitoring data to determine periodic pollution sources; Determine the spatiotemporal dynamic trajectory of pollutants according to the main pollution characteristics, the target meteorological parameters and the periodic pollution sources, wherein the spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of the pollutants; Through the spatiotemporal dynamic trajectory, pollution hotspots are identified; The performing Fourier transform on the air quality monitoring data to determine the periodic pollution source includes: Performing a discrete Fourier transform on the air quality monitoring data to convert the air quality monitoring data in the time domain into frequency components and corresponding amplitudes in the frequency domain; Generate a frequency spectrum according to the frequency components and corresponding amplitudes, determine significant frequency components based on the frequency spectrum, and calculate the period of the significant frequency components; Determining a periodic pollution source according to the significant frequency component and the period of the significant frequency component; The step of determining the periodic pollution source according to the significant frequency component and the period of the significant frequency component includes: In response to acquiring the background information of the monitoring area, determining the periodic characteristics of the significant frequency component according to the significant frequency component and the period of the significant frequency component; Determining a periodic pollution source according to the background information and the periodic characteristics; The background information includes industrial activity data and traffic flow data, and determining the periodic pollution source according to the background information and the periodic characteristics includes: Associating and matching the industrial activity data and the traffic flow data with the periodic features respectively to obtain a matching result; the matching result includes the periodic features, and the industrial activity data and / or traffic flow data corresponding to the periodic features; Determining a periodic pollution source according to the matching result; Wherein, if the matching result is that the periodic feature matches the industrial activity data, it is determined that the periodic pollution source comes from the industrial activity data; if the matching result is that the periodic feature matches the traffic flow data, it is determined that the periodic pollution source comes from the traffic flow data; if the matching result is that the periodic feature matches both the industrial activity data and the traffic flow data, it is determined that the periodic pollution source comes from both the industrial activity data and the traffic flow data; The background information also includes GIS data. The temporal and spatial dynamic trajectories of pollutants are determined based on the main pollution characteristics, the target meteorological parameters and the periodic pollution sources, including: Mining the spatiotemporal association rules of the main pollution characteristics, the target meteorological parameters and the periodic pollution sources to obtain association rules; Determining the target meteorological parameter and the periodic pollution source that are strongly associated with the main pollution feature according to the association rule; Constructing a pollutant generation model based on the target meteorological parameters strongly associated with the main pollution characteristics and the periodic pollution sources; Constructing a pollutant transport and diffusion model based on the GIS data and real-time acquired wind speed data, wind direction data, and terrain data; The pollutant generation model is combined with the pollutant transport and diffusion model to generate the spatiotemporal dynamic trajectory of the pollutant.

2. The method according to claim 1, characterized in that The determining of the causal relationship between the main pollution feature and the meteorological parameter, and determining the target meteorological parameter corresponding to the main pollution feature according to the causal relationship, includes: Determining a first lag term of each of the meteorological parameters and a second lag term of each of the main pollution characteristics, wherein the first lag term is used to characterize the meteorological parameters of the previous n days, and the second lag term is used to characterize the main pollution characteristics of the previous n days; Autoregression analysis is performed on the second lag term of the main pollution characteristics to obtain the constrained residual sum of squares; Performing unconstrained regression analysis on the first lag term of the meteorological parameter to obtain an unconstrained residual sum of squares; Using a preset significance calculation formula, the causal significance value of the meteorological parameter on the main pollution feature is calculated according to the constrained residual sum of squares and the unconstrained residual sum of squares; When the causal significance value is greater than a preset critical value, the corresponding meteorological parameter is determined to be a target meteorological parameter corresponding to the main pollution feature.

3. The method according to claim 1, characterized in that The step of combining the pollutant generation model with the pollutant transport and diffusion model to generate the spatiotemporal dynamic trajectory of the pollutant includes: Dividing the monitoring area into a plurality of identical spatial grids; According to the pollutant generation model, the amount of pollutants generated by the pollution source in each preset time step and the generation location and generation time of the corresponding pollutants are calculated; The generation location of the pollutant, the generation time, the GIS data, the wind speed data, the wind direction data and the terrain data are input into the pollutant transmission and diffusion model to obtain the spatiotemporal dynamic trajectory of the pollutant; wherein the pollutant transmission and diffusion model is used to calculate the pollutant concentration of the pollutant in each of the spatial grids to generate the spatiotemporal dynamic trajectory of the pollutant.

4. The method according to claim 3, characterized in that The method of identifying the pollution hotspot area through the spatiotemporal dynamic trajectory includes: When the pollutant concentration in the spatial grid is greater than a preset concentration threshold, the spatial grid is determined to be a hot spot grid, and the pollution hot spot area is a collection of the hot spot grids.

5. A cloud platform, characterized in that: The cloud platform includes: A big data analysis module, configured to respond to receiving air quality monitoring data and meteorological parameters of a monitoring area, and integrate the air quality monitoring data and the meteorological parameters using the big data analysis module to obtain an air quality data set; A feature recognition module is used to perform feature recognition on the air quality monitoring data to obtain main pollution features; a causal relationship determination module, used to determine the causal relationship between the main pollution characteristics and the meteorological parameters, and determine the target meteorological parameters corresponding to the main pollution characteristics according to the causal relationship; A calculation module, used for performing Fourier transform on the air quality monitoring data to determine periodic pollution sources; A spatiotemporal dynamic trajectory generation module, used to determine the spatiotemporal dynamic trajectory of pollutants according to the main pollution characteristics, the target meteorological parameters and the periodic pollution sources, wherein the spatiotemporal dynamic trajectory is used to characterize the generation, transmission and diffusion process of the pollutants; An identification and display module, used to identify pollution hotspot areas through the spatiotemporal dynamic trajectory; Wherein, the calculation module includes: A conversion module, used for performing a discrete Fourier transform on the air quality monitoring data to convert the air quality monitoring data in the time domain into frequency components and corresponding amplitudes in the frequency domain; A significant frequency component determination module, used to generate a frequency spectrum according to the frequency components and corresponding amplitudes, determine the significant frequency components based on the frequency spectrum, and calculate the period of the significant frequency components; A first determination module, configured to determine a periodic pollution source according to the significant frequency component and the period of the significant frequency component; The first determining module includes: A background information acquisition module, configured to determine the periodicity characteristics of the significant frequency component according to the significant frequency component and the period of the significant frequency component in response to acquiring the background information of the monitoring area; A periodic pollution source determination module, used to determine the periodic pollution source according to the background information and the periodic characteristics; The background information includes industrial activity data and traffic flow data, and the periodic pollution source determination module includes: An association matching unit, configured to associate and match the industrial activity data and the traffic flow data with the periodic feature respectively to obtain a matching result; the matching result includes the periodic feature, and the industrial activity data and / or traffic flow data corresponding to the periodic feature; A periodic pollution source determination unit, used to determine the periodic pollution source according to the matching result; Wherein, if the matching result is that the periodic feature matches the industrial activity data, it is determined that the periodic pollution source comes from the industrial activity data; if the matching result is that the periodic feature matches the traffic flow data, it is determined that the periodic pollution source comes from the traffic flow data; if the matching result is that the periodic feature matches both the industrial activity data and the traffic flow data, it is determined that the periodic pollution source comes from both the industrial activity data and the traffic flow data; The background information also includes GIS data, and the spatiotemporal dynamic trajectory generation module includes: An association rule mining module is used to perform spatiotemporal association rule mining on the main pollution characteristics, the target meteorological parameters and the periodic pollution sources to obtain association rules; A second determination module is used to determine the target meteorological parameter and the periodic pollution source that are strongly associated with the main pollution characteristics according to the association rule; A pollutant generation model building module, used to build a pollutant generation model based on the target meteorological parameters strongly associated with the main pollution characteristics and the periodic pollution sources; A pollutant transport and diffusion model building module, used to build a pollutant transport and diffusion model based on the GIS data and real-time acquired wind speed data, wind direction data, and terrain data; The model combining module is used to combine the pollutant generation model with the pollutant transmission and diffusion model to generate the spatiotemporal dynamic trajectory of the pollutant.

6. An electronic device, characterized in that: The device comprises: a memory configured to store instructions; and A processor is configured to call the instructions from the memory and to implement the air quality management method based on the integration of AIoT and big data services according to any one of claims 1 to 4 when executing the instructions.

Citation Information

Patent Citations

  • Water circulation treatment method for synergistically removing ferro-manganese, ammonia nitrogen and organic matters in underground water

    CN111439889A

  • Urban atmospheric pollution early warning supervision and decision analysis method

    CN112651553A

  • Atmospheric pollution cause judgment method based on convergence cross mapping algorithm

    CN116978490A

  • Diffusion prediction method and device for pollutants in atmosphere, equipment and medium

    CN117711518A

  • Water area pollution prediction method based on multiple scales and multiple dimensions

    CN118228923A