Methods, devices, equipment, media and products for identifying dominant meteorological factors of ozone pollution

Abstract

This application discloses a method, apparatus, equipment, medium, and product for identifying dominant meteorological factors of ozone pollution, relating to the field of ozone pollution analysis. The method comprises preprocessing ozone pollutant concentration data and meteorological factors to obtain daily ozone averages and meteorological factors matching the pollutant concentration data, and performing time series decomposition using a KZ filter to obtain baseline components of ozone and meteorological factors; decomposing the baseline components of ozone and meteorological factors using a first machine learning model and a KZ filter to obtain long-term components related to ozone emissions; determining long-term components related to ozone meteorology based on the long-term components related to ozone emissions; and determining dominant meteorological factors based on the long-term components related to ozone meteorology and meteorological factors using a second machine learning model and a SHAP algorithm. This application can reduce computational costs and improve identification efficiency.

Description

Technical Field

[0001] The present application relates to the field of ozone pollution analysis, and in particular to a method, device, equipment, medium and product for identifying dominant meteorological factors of ozone pollution. Background Art

[0002] Ozone (O3) as a secondary pollutant has attracted widespread attention for its impact on the environment, public health and climate change. The generation of O3 mainly depends on complex photochemical reactions, which involve volatile organic compounds (VOCs), nitrogen oxides (NO x ) and solar radiation. In this reaction, when NO reacts with O3 (NO + O3 → NO2 + O2), it is interfered with by peroxyalkyl radicals (RO2) and peroxyhydroxyl radicals (HO2) produced by the oxidation of VOCs and carbon monoxide (CO). NO is oxidized to NO2, leading to the accumulation of O3. This chemical process leads to a net increase in O3 concentration, thereby exacerbating photochemical pollution. In addition, the occurrence of O3 pollution is closely related to meteorological conditions. Weather factors such as high temperature, strong solar radiation, and light wind are favorable conditions for photochemical reactions. They regulate O3 concentration by accelerating the photochemical reaction rate and affecting atmospheric transport and dry and wet deposition processes.

[0003] Analyzing the drivers of O₃ pollution and clarifying the relative contributions of meteorological conditions and emissions are key to a deeper understanding of pollution mechanisms. By distinguishing the different impacts of human and meteorological factors on O₃ concentration changes, we can provide a scientific basis for developing more effective pollution control strategies, thereby protecting the ecological environment and improving public health. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, equipment, medium and product for identifying the dominant meteorological factors of ozone pollution, which can reduce computing costs and improve identification efficiency.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for identifying dominant meteorological factors of ozone pollution, comprising:

[0007] Obtain ozone pollutant concentration data and meteorological factors within a set time in a set area;

[0008] Preprocess the ozone pollutant concentration data and meteorological factors to obtain the daily average ozone value and meteorological factors that match the pollutant concentration data;

[0009] Performing time series decomposition using KZ filtering based on the daily average ozone value and the meteorological factors matching the pollutant concentration data to obtain a baseline component of ozone and a baseline component of the meteorological factors;

[0010] Decomposing the baseline component of ozone and the baseline component of meteorological factors using a first machine learning model and KZ filtering to obtain a long-term component related to ozone emissions;

[0011] Determine the long-term component related to ozone meteorology based on the long-term component related to ozone emissions;

[0012] The dominant meteorological factor is determined based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using the second machine learning model and the SHAP algorithm; wherein the long-term component of the meteorological factor is determined based on the baseline component of the meteorological factor.

[0013] In one embodiment, ozone pollutant concentration data and meteorological factors are preprocessed to obtain daily average ozone values ​​and meteorological factors that match the pollutant concentration data, specifically including:

[0014] Preprocess the ozone pollutant concentration data to calculate the daily average ozone value; the daily average ozone value is the maximum 8-hour average ozone value;

[0015] The meteorological factors were spatially matched with the ozone pollutant concentration data using the bilinear interpolation method to obtain meteorological factors that matched the pollutant concentration data.

[0016] In one embodiment, the meteorological factors specifically include boundary layer height, downward surface solar radiation, 2m temperature, 2m dew point temperature, mean sea level pressure, east-west wind speed at a set height, north-south wind speed at a set height, total cloud cover, and total precipitation.

[0017] In one embodiment, a first machine learning model and KZ filter decomposition are used to obtain a long-term component related to ozone emissions based on the baseline component of ozone and the baseline component of meteorological factors, specifically including:

[0018] constructing a first machine learning model based on the baseline component of ozone and the baseline component of meteorological factors;

[0019] The first residual in the first machine learning model is decomposed using KZ filtering to obtain a long-term component related to ozone emissions.

[0020] In one embodiment, the dominant meteorological factor is determined using a second machine learning model and a SHAP algorithm based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor, specifically including:

[0021] Constructing a second machine learning model based on the long-term components related to ozone meteorology and the long-term components of meteorological factors;

[0022] The second machine learning model is summarized using the SHAP algorithm to determine the dominant meteorological factors.

[0023] In one embodiment, the first machine learning model is a multiple linear regression model, a random forest model, LightGBM or XGBoost; the second machine learning model is a multiple linear regression model, a random forest model, LightGBM or XGBoost.

[0024] In a second aspect, the present application provides a device for identifying dominant meteorological factors of ozone pollution, comprising:

[0025] An acquisition module is used to obtain ozone pollutant concentration data and meteorological factors in a set area within a set time;

[0026] A preprocessing module is used to preprocess the ozone pollutant concentration data and meteorological factors to obtain the daily average ozone value and the meteorological factors that match the pollutant concentration data;

[0027] A time series decomposition module is used to perform time series decomposition using KZ filtering based on the daily average ozone value and the meteorological factors matching the pollutant concentration data to obtain a baseline component of ozone and a baseline component of the meteorological factors;

[0028] a long-term component decomposition module related to ozone emissions, configured to decompose the baseline component of ozone and the baseline component of meteorological factors using a first machine learning model and KZ filtering to obtain a long-term component related to ozone emissions;

[0029] An ozone meteorology-related long-term component determination module, used for determining an ozone meteorology-related long-term component based on an ozone emission-related long-term component;

[0030] The dominant meteorological factor determination module is used to determine the dominant meteorological factor based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using the second machine learning model and the SHAP algorithm; wherein the long-term component of the meteorological factor is determined based on the baseline component of the meteorological factor.

[0031] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for identifying the dominant meteorological factors of ozone pollution.

[0032] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for identifying the dominant meteorological factors of ozone pollution.

[0033] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the method for identifying the dominant meteorological factors of ozone pollution.

[0034] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0035] The present application provides a method, device, equipment, medium and product for identifying dominant meteorological factors of ozone pollution, which pre-processes ozone pollutant concentration data and meteorological factors to obtain daily average ozone values ​​and meteorological factors that match the pollutant concentration data; performs time series decomposition using KZ filtering based on the daily average ozone values ​​and the meteorological factors that match the pollutant concentration data, respectively, to obtain a baseline component of ozone and a baseline component of the meteorological factor; decomposes the baseline component of ozone and the baseline component of the meteorological factor using a first machine learning model and KZ filtering to obtain a long-term component related to ozone emissions; determines a long-term component related to ozone meteorology based on the long-term component related to ozone emissions; determines the dominant meteorological factor based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using a second machine learning model and SHAP algorithm; wherein the long-term component of the meteorological factor is determined based on the baseline component of the meteorological factor. By performing time series decomposition on the daily average ozone values ​​and the meteorological factors that match the pollutant concentration data using KZ filtering, and finally processing them using the first machine learning model and the second machine learning model, computing costs can be reduced and computing efficiency improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 This is an application environment diagram of a method for identifying dominant meteorological factors of ozone pollution in one embodiment of the present application.

[0038] Figure 2 A flowchart of a method for identifying dominant meteorological factors of ozone pollution provided in one embodiment of the present application.

[0039] Figure 3 A schematic diagram of a method for identifying dominant meteorological factors of ozone pollution provided in one embodiment of the present application.

[0040] Figure 4 A schematic diagram of the functional modules of a device for identifying dominant meteorological factors of ozone pollution provided in another embodiment of the present application.

[0041] Figure 5 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0043] Traditional analyses of ozone pollution have focused on single factors or simple meteorological models, lacking in-depth analysis of interannual variations in meteorological conditions and the combined impacts of multiple meteorological factors. Accurately assessing air quality changes requires comprehensive consideration of the dual effects of meteorological and human factors. Therefore, conducting multi-factor, comprehensive studies, combined with machine learning techniques to explore the relationship between meteorological factors and O3 concentrations, has become a key research direction.

[0044] In summary, this application proposes to use the KZ filtering method, a multivariate linear regression model, and a machine learning interpretation algorithm to systematically evaluate the impact of different meteorological factors on O3 concentrations and provide scientific support for regional air pollution control policies. This research not only fills an existing academic gap but also provides important reference for other regions with similar geographical and climatic characteristics.

[0045] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0046] The method for identifying the dominant meteorological factors of ozone pollution provided in the embodiment of the present application can be applied to Figure 1In the application environment shown, the terminal 102 communicates with the server 104 via a network. The data storage system can store data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. Terminal 102 may send the ozone pollutant concentration data and meteorological factors to be processed to server 104. After receiving the ozone pollutant concentration data and meteorological factors, server 104 preprocesses the ozone pollutant concentration data and meteorological factors to obtain a daily average ozone value and a meteorological factor matching the pollutant concentration data. KZ filtering is used to perform time series decomposition based on the daily average ozone value and the meteorological factor matching the pollutant concentration data to obtain a baseline component of ozone and a baseline component of the meteorological factor. A first machine learning model and KZ filtering are used to decompose the baseline components of ozone and meteorological factors to obtain a long-term component related to ozone emissions. A long-term component related to ozone meteorology is determined based on the long-term component related to ozone emissions. A dominant meteorological factor is determined based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using a second machine learning model and a SHAP algorithm. Server 104 may provide feedback of the obtained dominant meteorological factor to terminal 102. In addition, in some embodiments, the method for identifying the dominant meteorological factors of ozone pollution can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly identify the dominant meteorological factors of ozone pollution based on the ozone pollutant concentration data and meteorological factors to be processed, or the server 104 can obtain the ozone pollutant concentration data and meteorological factors to be processed from the data storage system, and identify the dominant meteorological factors of ozone pollution based on the ozone pollutant concentration data and meteorological factors to be processed.

[0047] Terminal 102 may include, but is not limited to, various desktop computers, laptops, smartphones, tablet computers, IoT devices, and portable wearable devices. IoT devices may include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices may include smart watches, smart bracelets, and head-mounted devices. Server 104 may be implemented as a standalone server or a server cluster consisting of multiple servers, or may be a cloud server.

[0048] In an exemplary embodiment, Figure 2 and Figure 3 As shown, a method for identifying the dominant meteorological factors of ozone pollution is provided. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is applied to Figure 1The server 104 in the example is used for explanation, including the following steps 201 to 206.

[0049] Step 201: Obtain ozone pollutant concentration data and meteorological factors in a set area within a set time.

[0050] Step 202: Preprocess the ozone pollutant concentration data and meteorological factors to obtain the daily average ozone value and meteorological factors that match the pollutant concentration data.

[0051] Step 203: performing time series decomposition using KZ filtering based on the daily average ozone value and the meteorological factors matching the pollutant concentration data, respectively, to obtain a baseline component of ozone and a baseline component of the meteorological factors.

[0052] Step 204: Decompose the baseline component of ozone and the baseline component of meteorological factors using a first machine learning model and KZ filtering to obtain a long-term component related to ozone emissions.

[0053] Step 205: Determine the long-term component related to ozone meteorology based on the long-term component related to ozone emissions.

[0054] Step 206: Determine the dominant meteorological factor based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using the second machine learning model and the SHAP algorithm; wherein the long-term component of the meteorological factor is determined based on the baseline component of the meteorological factor.

[0055] Implementing the above steps 201 to 206 can reduce computing costs and improve recognition efficiency.

[0056] In an exemplary embodiment, ozone pollutant concentration data and meteorological factors are preprocessed to obtain daily ozone average values ​​and meteorological factors that match the pollutant concentration data, specifically including: preprocessing the ozone pollutant concentration data to calculate the daily ozone average value; the daily ozone average value is the maximum 8-hour average value of the ozone day; and using bilinear interpolation to spatially match the meteorological factors with the ozone pollutant concentration data to obtain meteorological factors that match the pollutant concentration data.

[0057] Specifically, ozone pollutant concentration data and meteorological factors were collected in the study area within a certain period of time. After collection, the data were preprocessed to calculate the maximum daily 8-hour average of ozone (MDA8), and the meteorological factors were matched with neighboring pollutant observation sites.

[0058] The meteorological factors used include:

[0059] Boundary layer height (BLH, m), downward surface solar radiation (SSR, Jm-1 )、2m temperature (T2M, K), 2m dew point temperature (D2M, K), mean sea level pressure (MSP, m), u10 wind (ms -1 ), v10 wind (ms -1 ), total cloud cover (TCC, unitless), total precipitation (TP, mm).

[0060] According to the formula: , calculate relative humidity (RH, unitless).

[0061] The data were obtained from the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-5 reanalysis dataset (https: / / cds.climate.copernicus.eu / datasets / reanalysis-era5-single-levels?tab=overview). Bilinear interpolation was used to match meteorological factors with pollutant concentration data.

[0062] When the multiple linear regression model was subsequently established, the meteorological factors used included: T2M, RH, BLH, u10, v10, MSP, TCC, TP, and SSR.

[0063] In an exemplary embodiment, the meteorological factors specifically include boundary layer height, downward surface solar radiation, 2-meter temperature, 2-meter dew point temperature, mean sea level pressure, east-west wind speed at a set height, north-south wind speed at a set height, total cloud cover, and total precipitation. In practical applications, the set height is 10 meters, the east-west wind speed at 10 meters is represented by u10, and the north-south wind speed at 10 meters is represented by v10.

[0064] KZ filtering is used to decompose the time series based on the daily average ozone value and the meteorological factors that match the pollutant concentration data, respectively, to obtain the baseline component of ozone and the baseline component of the meteorological factors. Specifically, the KZ filter (Kolmogorov-Zurbenko filter) is used to decompose the time series scale of MDA8 O3. The daily time series of atmospheric pollutant concentrations can be mainly divided into long-term components, seasonal components, and short-term components. In this application, the short-term component is not considered. The sum of the seasonal and long-term components is the baseline component:

[0065] ;

[0066] ;

[0067] Where, X(t) is the time series of MDA8 during the study period; X LT (t) is the long-term component; X SN(t) is the seasonal component; X ST (t) is the short-term component; X BL (t) is the baseline component, where "X" indicates ozone-related quantities and "MET" indicates meteorological factor-related quantities. (m，p) , select the time window m = 15 days, the number of iterations p = 5 times, apply it to X(t), and pass KZ (15,5) After filtering out the daily fluctuations and random noise with a fluctuation period of less than 33 days, the obtained X BL (t) Long-term trends and seasonal variations will be preserved:

[0068] ;

[0069] The above three equations can be used to calculate the short-term component X ST (t):

[0070] ;

[0071] Further X BL (t) is decomposed into the recurring climate seasonal cycle during the study period and the sum of the first residual ε(t), KZ (365,3) The filter is applied to ε(t) to remove fluctuations with a time period shorter than 1.7 years to obtain the long-term component X LT (t):

[0072] ;

[0073] ;

[0074] Then the seasonal component X SN (t) can be expressed as:

[0075] .

[0076] In an exemplary embodiment, a first machine learning model and KZ filter decomposition are performed based on the baseline component of ozone and the baseline component of meteorological factors to obtain a long-term component related to ozone emissions, specifically comprising: constructing a first machine learning model based on the baseline component of ozone and the baseline component of meteorological factors; and decomposing the first residual in the first machine learning model using KZ filter decomposition to obtain a long-term component related to ozone emissions.

[0077] Specifically, the long-term changes in air pollutants may be affected by changes in emissions and meteorological conditions. LT (t) is the long-term component related to emissions [ Long-term components related to meteorology[ ], then XBL (t) can also be expressed as:

[0078] ;

[0079] To separate and , establishing MLR model is an effective method, using the baseline component MET of meteorological factors BL (t) and the concentration of ozone pollutants X BL (t) Construct the following MLR model:

[0080] ;

[0081] Where i is different meteorological factors; a0 is the intercept; a i is the regression coefficient of different meteorological factors; is the second residual. It not only includes the variability of O3 related to the long-term changes in atmospheric pollutant emissions, but also includes the small seasonal changes in O3 caused by meteorological effects that cannot be explained in the MLR model. (365,3) Removal Small seasonal changes can be separated , as shown below:

[0082] ;

[0083] You can also ask for :

[0084] .

[0085] In an exemplary embodiment, a second machine learning model and a SHAP algorithm are used to determine the dominant meteorological factor based on the long-term component related to the ozone meteorology and the long-term component of the meteorological factor, specifically including: constructing a second machine learning model based on the long-term component related to the ozone meteorology and the long-term component of the meteorological factor; summarizing the second machine learning model using the SHAP algorithm to determine the dominant meteorological factor.

[0086] A multivariate linear regression model was constructed using the long-term components of ozone meteorology and meteorological factors. The model results were interpreted using the SHAP (Shapley Additive ExPlanation) algorithm to identify the dominant meteorological factors. The summary results for each meteorological factor were obtained by calculating the mean of the absolute values ​​of the SHAP values ​​(mean |SHAP|). By analyzing the aggregated SHAP values, it is possible to identify the meteorological factors that have the greatest impact on the long-term trends of air pollutants, namely the variables with the largest mean |SHAP| values. The SHAP value calculation formula is:

[0087] ;

[0088] This formula is an explanation of the principle of the SHAP method. represents the SHAP value of the feature contribution of feature i; N is the set of all features; S is the subset that does not contain feature i; f(S) represents the model prediction value when only the feature subset is used.

[0089] The long-term component of meteorological factors is used as the dependent variable and the long-term component of meteorological factors is used as the independent variable to establish a multiple linear regression model, that is, ,in, is the intercept; is the regression coefficient; is the long-term component of the meteorological factor, The model residuals are then interpreted using the SHAP algorithm. The summary of SHAP values ​​provides an explanation of the overall impact of each meteorological factor on the model's dependent variable. Furthermore, the dominant meteorological factor can be identified using the mean of the absolute values ​​of the SHAP values ​​(mean|SHAP|) as an indicator.

[0090] Assessing the impact of anthropogenic emissions and meteorological changes on air quality is an important issue in environmental change. Changes in atmospheric pollution are generally determined by a combination of atmospheric pollutant emissions and meteorological conditions. However, meteorological conditions have a strong influence on changes in air quality, and their influence may mask the impact of changes in emission levels over time. Therefore, in order to ensure an accurate assessment of changes in air quality and further determine the key directions and specific governance measures for the prevention and control of atmospheric pollution, it is extremely important to study the impact of meteorological factors on air quality. Since this application chooses to establish a multivariate linear regression model, it has the advantages of simple implementation method, low computational cost and high efficiency. In addition, other methods can be selected when building the model, such as random forest, LightGBM, XGBoost and other machine learning models.

[0091] In an exemplary embodiment, the first machine learning model is a multiple linear regression model, a random forest model, LightGBM or XGBoost; the second machine learning model is a multiple linear regression model, a random forest model, LightGBM or XGBoost.

[0092] The present application also provides an application scenario, which applies the above-mentioned method for identifying the dominant meteorological factors of ozone pollution. Specifically: In the study, by collecting O3 concentration data and meteorological factors in the main urban area of ​​a certain city from 2015 to 2023, data preprocessing was performed to calculate MDA8 O3, and the KZ filtering method was used to decompose the original time series of MDA8 O3, decomposing it into short-term, seasonal and long-term components, and the same time series decomposition was performed on the meteorological factors. Afterwards, the baseline component of MDA8 O3 (the sum of the short-term component and the seasonal component) was used as the dependent variable, and the baseline component of the meteorological factor was used as the independent variable to establish a multiple linear regression equation, and the long-term components related to meteorology and emissions were further obtained through the equation results. Then, the long-term meteorological component was used as the dependent variable, and the long-term component of the meteorological factor was used as the independent variable. A multiple linear regression equation was established again and interpreted using the SHAP algorithm. The impact of meteorological factors on O3 pollution in the main urban area of ​​Chongqing under the long-term trend can be obtained, and the dominant meteorological factors can be identified. The meteorological contribution and emission contribution of O3 pollution can be clarified. Based on the research results, the key directions and specific governance measures for the prevention and control of O3 pollution in a certain city are proposed, and feasible emission reduction plans and policy recommendations are provided for areas with similar terrain and climate characteristics.

[0093] Based on the same inventive concept, embodiments of the present application also provide an ozone pollution dominant meteorological factor identification device for implementing the aforementioned ozone pollution dominant meteorological factor identification method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the ozone pollution dominant meteorological factor identification device provided below can be found in the above-mentioned limitations of the ozone pollution dominant meteorological factor identification method and will not be further elaborated here.

[0094] like Figure 4 As shown, in an exemplary embodiment, a device for identifying dominant meteorological factors of ozone pollution is provided, including the following modules.

[0095] The acquisition module 401 is used to obtain ozone pollutant concentration data and meteorological factors in a set area within a set time.

[0096] The pre-processing module 402 is used to pre-process the ozone pollutant concentration data and meteorological factors to obtain the daily average ozone value and the meteorological factors that match the pollutant concentration data.

[0097] The time series decomposition module 403 is used to perform time series decomposition using KZ filtering based on the daily average ozone value and the meteorological factors matching the pollutant concentration data to obtain a baseline component of ozone and a baseline component of the meteorological factors.

[0098] The long-term component decomposition module 404 related to ozone emission is used to decompose the baseline component of ozone and the baseline component of meteorological factors using a first machine learning model and KZ filtering to obtain the long-term component related to ozone emission.

[0099] The ozone meteorology-related long-term component determination module 405 is configured to determine the ozone meteorology-related long-term component based on the ozone emission-related long-term component.

[0100] The dominant meteorological factor determination module 406 is used to determine the dominant meteorological factor based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using the second machine learning model and the SHAP algorithm; wherein the long-term component of the meteorological factor is determined based on the baseline component of the meteorological factor.

[0101] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store identification data of the dominant meteorological factors of ozone pollution. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for identifying the dominant meteorological factors of ozone pollution is implemented.

[0102] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the above-mentioned method embodiments when executing the computer program.

[0103] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the above-mentioned method embodiments when executed by a processor.

[0104] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the above method embodiments are implemented.

[0105] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0106] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0107] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0108] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0109] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0110] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for identifying the dominant meteorological factors of ozone pollution, characterized in that: The method for identifying the dominant meteorological factors of ozone pollution includes: Obtain ozone pollutant concentration data and meteorological factors within a set time in a set area; Preprocess the ozone pollutant concentration data and meteorological factors to obtain the daily average ozone value and meteorological factors that match the pollutant concentration data; Performing time series decomposition using KZ filtering based on the daily average ozone value and the meteorological factors matching the pollutant concentration data to obtain a baseline component of ozone and a baseline component of the meteorological factors; Decomposing the baseline component of ozone and the baseline component of meteorological factors using a first machine learning model and KZ filtering to obtain a long-term component related to ozone emissions; Determine the long-term component related to ozone meteorology based on the long-term component related to ozone emissions; The dominant meteorological factor is determined based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using the second machine learning model and the SHAP algorithm; wherein the long-term component of the meteorological factor is determined based on the baseline component of the meteorological factor.

2. The method for identifying the dominant meteorological factors of ozone pollution according to claim 1, characterized in that: The ozone pollutant concentration data and meteorological factors are preprocessed to obtain the daily average ozone value and meteorological factors that match the pollutant concentration data, including: Preprocess the ozone pollutant concentration data to calculate the daily average ozone value; the daily average ozone value is the maximum 8-hour average ozone value; The meteorological factors were spatially matched with the ozone pollutant concentration data using the bilinear interpolation method to obtain meteorological factors that matched the pollutant concentration data.

3. The method for identifying the dominant meteorological factors of ozone pollution according to claim 1, characterized in that: The meteorological factors specifically include boundary layer height, downward surface solar radiation, 2m temperature, 2m dew point temperature, mean sea level pressure, east-west wind speed at a set height, north-south wind speed at a set height, total cloud cover and total precipitation.

4. The method for identifying the dominant meteorological factors of ozone pollution according to claim 1, characterized in that: Based on the baseline component of ozone and the baseline component of meteorological factors, a first machine learning model and KZ filter decomposition are used to obtain a long-term component related to ozone emissions, specifically including: constructing a first machine learning model based on the baseline component of ozone and the baseline component of meteorological factors; The first residual in the first machine learning model is decomposed using KZ filtering to obtain a long-term component related to ozone emissions.

5. The method for identifying the dominant meteorological factors of ozone pollution according to claim 1, characterized in that: The dominant meteorological factors are determined using the second machine learning model and the SHAP algorithm based on the long-term components related to ozone meteorology and meteorological factors. Specifically, the following are the main meteorological factors: Constructing a second machine learning model based on the long-term components related to ozone meteorology and the long-term components of meteorological factors; The second machine learning model is summarized using the SHAP algorithm to determine the dominant meteorological factors.

6. The method for identifying the dominant meteorological factors of ozone pollution according to claim 1, characterized in that: The first machine learning model is a multiple linear regression model, a random forest model, LightGBM or XGBoost; the second machine learning model is a multiple linear regression model, a random forest model, LightGBM or XGBoost.

7. A device for identifying dominant meteorological factors of ozone pollution, characterized in that: The ozone pollution dominant meteorological factor identification device includes: An acquisition module is used to obtain ozone pollutant concentration data and meteorological factors in a set area within a set time; A preprocessing module is used to preprocess the ozone pollutant concentration data and meteorological factors to obtain the daily average ozone value and the meteorological factors that match the pollutant concentration data; A time series decomposition module is used to perform time series decomposition using KZ filtering based on the daily average ozone value and the meteorological factors matching the pollutant concentration data to obtain a baseline component of ozone and a baseline component of the meteorological factors; a long-term component decomposition module related to ozone emissions, configured to decompose the baseline component of ozone and the baseline component of meteorological factors using a first machine learning model and KZ filtering to obtain a long-term component related to ozone emissions; An ozone meteorology-related long-term component determination module, used for determining an ozone meteorology-related long-term component based on an ozone emission-related long-term component; The dominant meteorological factor determination module is used to determine the dominant meteorological factor based on the long-term component related to ozone meteorology and the long-term component of the meteorological factor using the second machine learning model and the SHAP algorithm; wherein the long-term component of the meteorological factor is determined based on the baseline component of the meteorological factor.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for identifying the dominant meteorological factors of ozone pollution according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying the dominant meteorological factors of ozone pollution according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for identifying the dominant meteorological factors of ozone pollution according to any one of claims 1 to 6 is implemented.

Images