A method for predicting air pollutants based on big data

Abstract

The present invention relates to the field of air pollutant prediction, and discloses an air pollutant prediction method based on big data, including: obtaining the regional location information and event information of a target area, and determining a plurality of pollution sources within a preset range of the target area according to the regional location information; determining the activity pollution impact value according to the activity type and the degree of activity scale; determining the target historical time period according to the event holding time period; screening a plurality of alternative prediction models according to the prediction accuracy of the alternative prediction models to obtain a target prediction model; training the target prediction model, and using the trained target prediction model as an air pollutant prediction model; predicting the pollutant concentration of the target area based on the air pollutant prediction model to obtain an initial prediction result; and adjusting the initial prediction result according to the activity pollution impact value to obtain a standard prediction result. Thus, the prediction of air pollutants is realized.

Description

Technical Field

[0001] The present invention relates to the field of prediction of air pollutants, and particularly to a method for predicting air pollutants based on big data. Background Art

[0002] When holding large-scale events (such as fireworks shows), air quality is one of the important factors to ensure the smooth progress of the event. Through big data prediction technology, the air quality status during the event can be evaluated in advance, and corresponding safeguard measures can be formulated.

[0003] In the existing process of predicting air pollutants, the following technical problems often exist:

[0004] First, the prediction area range of conventional air pollutant prediction methods is relatively large, lacking local air pollutant prediction.

[0005] Second, since the holding time of large-scale events is fixed and cannot be changed, if the pollutants exceed the standard before the event starts, it will affect the smooth progress of the event. In order to ensure that the air quality meets the standard during the event, an effective emission limit scheme needs to be formulated. The existing technology lacks a targeted method for formulating emission limit schemes and often adopts a "one-size-fits-all" emission limit.

[0006] Third, when there are multiple malfunctioning air quality monitoring points, the staff lacks a clear maintenance policy, resulting in low maintenance efficiency, so that some urgently needed monitoring points cannot be processed in time, affecting the smooth progress of the event. Summary of the Invention

[0007] This part of the present invention is used to briefly introduce the concepts, which will be described in detail in the following detailed implementation part. This part of the present invention is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0008] The present invention proposes a method for predicting air pollutants based on big data to solve one or more of the technical problems mentioned in the above background art part.

[0009] The present invention provides a method for predicting air pollutants based on big data, including: obtaining the regional location information and event information of a target area, and determining a plurality of pollution sources within a preset range of the target area according to the regional location information; wherein, the event information includes the event time period, event type, and event scale; determining the event pollution impact value according to the event type and event scale; determining the target historical time period according to the event time period and obtaining the pollutant concentration information and meteorological information of each pollution source among the plurality of pollution sources within the target historical time period; obtaining a plurality of alternative prediction models and the prediction accuracy of each alternative prediction model among the plurality of alternative prediction models; screening the plurality of alternative prediction models according to the prediction accuracy of the alternative prediction models to obtain a target prediction model; training the target prediction model through the pollutant concentration information and meteorological information, and using the trained target prediction model as an air pollutant prediction model; predicting the pollutant concentration of the target area during the event time period based on the air pollutant prediction model to obtain an initial prediction result; and adjusting the initial prediction result according to the event pollution impact value to obtain a standard prediction result.

[0010] Optionally, the event pollution impact value is determined through the following steps:

[0011] Obtaining a plurality of historical event data, where the historical event data includes historical event types and historical event scales; matching according to the event type and event scale in the plurality of historical event data to obtain target historical event data; and obtaining the historical event pollution emissions of the target historical event data.

[0012] Obtaining the environmental capacity of the target area and a pre-established environmental quality standard table, and querying the environmental capacity in the environmental quality standard table to obtain the standard allowable emissions of the target area, where the environmental quality standard table includes different environmental capacities and the standard allowable emissions corresponding to each environmental capacity.

[0013] Calculating the ratio of the historical event pollution emissions to the standard allowable emissions to obtain the event pollution impact value.

[0014] Optionally, the standard prediction result is determined through the following steps:

[0015] Determining the adjustment value of the initial prediction result according to the event pollution impact value, and calculating the adjustment value and the initial prediction result to obtain the standard prediction result.

[0016] Optionally, each pollution source among the plurality of pollution sources has a corresponding pollution source type, and the corresponding pollution source type of each pollution source is determined through the following steps:

[0017] Obtaining the emission frequency of each pollution source within the target historical time period.

[0018] Determine the pollution source type of the pollution sources with fixed emission frequencies among multiple pollution sources as fixed pollution sources to obtain a fixed pollution source group; determine the pollution source type of the pollution sources with non-fixed emission frequencies among multiple pollution sources as temporary pollution sources.

[0019] Optionally, the fixed pollution sources have corresponding pollutant emission limit plans, and the pollutant emission limit plans are generated through the following steps:

[0020] Obtain the pollutant emission time periods of each fixed pollution source in the fixed pollution source group; determine the fixed pollution sources in the fixed pollution source group whose pollutant emission time periods coincide with the activity holding time periods as key pollution sources to obtain a key pollution source group;

[0021] Obtain the emissions of each key pollution source in the key pollution source group, calculate the emissions of each key pollution source, and obtain the total pollutant emissions corresponding to the key pollution source group;

[0022] Determine the degree of pollutant over-standard according to the total pollutant emissions; compare the degree of pollutant over-standard with the preset standard degree. If the degree of pollutant over-standard is greater than or equal to the preset standard degree, generate a pollutant emission limit plan, and the pollutant emission limit plan includes the restricted time and the standard emission amount; send the pollutant emission limit plan to the terminal devices corresponding to each key pollution source in the key pollution source group so that the management personnel corresponding to the terminal devices corresponding to each key pollution source perform the operation of restricting pollutant emissions;

[0023] Among them, the standard emission amount included in the pollutant emission limit plan is determined through the following steps:

[0024] Obtain the geographical location information of each key pollution source in the key pollution source group. According to the geographical location information of each key pollution source, determine the straight-line distance value between each key pollution source and the target area; generate the first standard emission amount for the key pollution sources in the key pollution source group whose straight-line distance values are greater than or equal to the preset distance value; generate the second standard emission amount for the key pollution sources in the key pollution source group whose straight-line distance values are less than the preset distance value, and the first standard emission amount is less than the second standard emission amount.

[0025] The present invention has the following beneficial effects:

[0026] 1. By obtaining the target area and the holding time of a large-scale event, and using the pollutant concentration information and meteorological information in the target historical time period corresponding to the event holding time period as training samples to train the target prediction model. Therefore, the trained target prediction model is an atmospheric pollutant prediction model for the target area during the event holding time period, enabling the atmospheric pollutant prediction model to have local and targeted predictions. The holding of a large-scale event itself will also affect the prediction of atmospheric pollutants. The type and scale of the event are the main factors affecting air pollution, and the environmental capacity, which is the ability of the environment to accommodate pollutants, plays a crucial role in determining the impact of a large-scale event on air pollution. Adjusting the initial prediction result with the event pollution impact value determined by the event type, scale, and the environmental capacity of the event area can more comprehensively consider the impact of the event on atmospheric pollutants, be more in line with the actual situation, thereby reducing the error of the initial prediction result and improving the reliability and accuracy of the prediction result;

[0027] 2. Due to different pollutant generation mechanisms and conditions, different types of pollution sources often have different emission frequencies. Therefore, the emission frequency is an important basis for distinguishing different types of pollution sources. Determining the type of pollution source by whether the emission frequency of the pollution source is fixed improves the accuracy of pollution source identification and helps formulate targeted pollution control plans. The decline in environmental quality may affect the smooth progress of the event. By restricting the emissions of fixed pollution sources during the event in advance, the smooth progress of the event can be ensured. Setting different pollutant emission limits for different key pollution sources can more precisely control the emissions of each pollution source and avoid a one-size-fits-all emission limit;

[0028] 3. According to the characteristics of the normal distribution, data points falling outside three standard deviations from the mean are considered extremely rare. Therefore, the three-sigma method is used to identify abnormal monitoring points in the pollutant concentration data set, which can improve the accuracy of determining abnormal monitoring points. Turning on the indicator light reminder and generating abnormal information for abnormal monitoring points. The indicator light can remind the staff, reduce the response time, enabling them to discover and respond to abnormal monitoring points in the first time without the need for the staff to actively judge, thus simplifying the troubleshooting process and saving troubleshooting manpower. The key areas of the event often gather a large number of people. Determining the maintenance priority based on the location attributes of each abnormal monitoring point can improve the timeliness of abnormal monitoring point maintenance and ensure the smooth progress of the event. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In combination with the drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the embodiments of the present invention will become more apparent. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic, and the elements and elements are not necessarily drawn to scale.

[0030] Figure 1 It is a flowchart of a method for predicting air pollutants based on big data according to the present invention. Specific embodiments

[0031] The present invention will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0032] In addition, it should be noted that for the convenience of description, only the parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0033] It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or mutual dependence relationship of the functions performed by these devices, modules or units.

[0034] It should be noted that the modifications of "one" and "plural" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0035] The names of the messages or information exchanged between multiple devices of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0036] The present invention will be described in detail below with reference to the drawings and in combination with embodiments.

[0037] As Figure 1 shown, it is a flowchart of a method for predicting air pollutants based on big data according to the present invention.

[0038] Step 101, obtain the regional location information and event information of the target area, and determine multiple pollution sources within the preset range of the target area according to the regional location information; wherein, the event information includes the event time period, event type, and event scale degree.

[0039] In some embodiments, the execution subject of a big data-based air pollutant prediction method of the present invention can be a background server. The target area is the area specified for research, which can be the entire activity area of a large fireworks show. The regional location information of the target area is obtained from an open-source electronic map interface. The regional location information of the target area can be No. 103, XX Street, XX District. Multiple pollution sources within the preset range of the target area are obtained through the regional location information of the target area. The preset range of the target area covers all pollution sources that may affect the air quality of the target area. The preset range of the target area can be multiple pollution sources within a radius of 10 kilometers centered on the target area. The multiple pollution sources can be factory emission sources, construction site dust, etc. The activity information of the target area is obtained from the activity terminal. The activity information includes the activity time period, activity type, and activity scale. For example, the activity time period in the activity information can be 18:00-21:00 on May 10, 2023, the activity type in the activity information can be a fireworks show, and the activity scale in the activity information can be a large-scale event.

[0040] Step 102: Determine the activity pollution impact value according to the activity type and activity scale.

[0041] Optionally, the activity pollution impact value is determined through the following steps:

[0042] Obtain multiple historical activity data, where the historical activity data includes historical activity types and historical activity scales; match according to the activity type and activity scale in the multiple historical activity data to obtain the target historical activity data; obtain the historical activity pollution emissions of the target historical activity data.

[0043] Obtain the environmental capacity of the target area and the pre-established environmental quality standard table. By querying the environmental capacity in the environmental quality standard table, obtain the standard allowable emissions of the target area. The environmental quality standard table includes different environmental capacities and the corresponding standard allowable emissions for each environmental capacity.

[0044] Calculate the ratio of the historical activity pollution emissions to the standard allowable emissions to obtain the activity pollution impact value.

[0045] In some embodiments, a plurality of historical activity data for the past year are obtained from a terminal. The historical activity data includes historical activity types and the scale degrees of historical activities. The historical activity types included in the historical activity data may be fireworks shows or music festivals, and the scale degrees of activities included in the historical activity data may be large-scale activities or medium-scale activities. The plurality of historical activity data may include historical activity data 1 and historical activity data 2. Historical activity data that matches the activity type and the scale degree of the activity information held in the target area is selected from the plurality of historical activity data as the target historical activity data. If both the activity type of historical activity data 2 and the activity held in the target area are fireworks shows and the scale degree of the activity is a large-scale activity, then historical activity data 2 is used as the target historical activity data. Since a pollutant concentration monitoring point is configured in the target historical activity, the pollutant concentration monitoring point is used to monitor the pollutant concentration data of the activity. Multiplying the obtained pollutant concentration data by the total duration of the target historical activity can obtain the historical activity pollution emissions corresponding to the target historical activity. The historical activity pollution emissions are uploaded to the terminal. Obtaining the historical activity pollution emissions of the target historical activity data from the terminal may be 0.5 tons. Environmental capacity refers to the maximum load value of pollutants that a certain environment can accommodate on the premise of ensuring that human survival and development are not harmed and the natural ecological balance is not damaged. The environmental capacity of the target area and a pre-formulated environmental quality standard table are obtained from the comprehensive database of the target area. The environmental capacity of the target area may be 400 tons. By querying the environmental capacity in the environmental quality standard table, the standard allowable emissions of the target area may be 0.7 tons. Among them, the environmental quality standard table includes different environmental capacities and the standard allowable emissions corresponding to each environmental capacity. The historical activity pollution emissions are compared with the standard allowable emissions to calculate the activity pollution impact value. Specifically, taking historical activity data 2 as an example, the historical activity pollution emissions of historical activity data 2, which is 0.5 tons, are compared with the standard allowable emissions of the target area, which is 0.7 tons, to calculate the activity pollution impact value of 0.714. The calculation process is 0.5 tons divided by 0.7 tons equals 0.714. If the activity pollution impact value is less than 1, it means that the pollution impact of the activity is within an acceptable range; if the activity pollution impact value is greater than or equal to 1, it means that the activity causes a relatively large pollution impact.

[0046] Step 103: Determine the target historical time period according to the activity holding time period, and obtain the pollutant concentration information and meteorological information of each pollutant source in the target historical time period; obtain the prediction demand information of the target area, and obtain a plurality of alternative prediction models and the prediction accuracy of each alternative prediction model in the plurality of alternative prediction models; screen the plurality of alternative prediction models according to the prediction accuracy of the alternative prediction models to obtain the target prediction model;

[0047] In some embodiments, if the activity period is from 18:00 to 21:00 on May 10, 2023, the target historical period is from 18:00 to 21:00 every day in the past year. Pollutant concentration information and meteorological information of each pollutant source during the period from 18:00 to 21:00 every day in the past year in the target area are obtained from the meteorological website interface. Specifically, the pollutant concentration information may be concentration data of pollutants such as PM2.5, PM10, SO2, NO2, etc., and the meteorological information may be temperature, humidity, wind speed, wind direction, etc. Meteorological information has an important impact on the diffusion, transmission, and transformation processes of pollutants. Therefore, meteorological conditions help to more accurately predict changes in pollutant concentrations. Multiple alternative prediction models are obtained from the model database. The multiple alternative prediction models may be alternative prediction model A (long short-term memory network model) with a prediction accuracy of 95%, alternative prediction model B (decision tree model) with a prediction accuracy of 68%, and alternative prediction model C (support vector machine model) with a prediction accuracy of 57%. The multiple alternative prediction models are screened, and the screening condition is to select the alternative prediction model with the highest prediction accuracy among the multiple alternative prediction models. Alternative prediction model A is the alternative prediction model with the highest prediction accuracy among the multiple alternative prediction models. Therefore, alternative prediction model A is used as the target prediction model.

[0048] Step 104, train the target prediction model with the pollutant concentration information and meteorological information, and use the trained target prediction model as the air pollutant prediction model;

[0049] In some embodiments, when training the target prediction model, the training process is as follows: Initialize the cell state and hidden state of the neurons in the long short-term memory network model, as well as other parameters in the long short-term memory network model; Input the pollutant concentration information and meteorological information during the period from 18:00 to 21:00 every day in the past year in the target area into the target prediction model (alternative prediction model A (long short-term memory network model)) to obtain the output result of the target prediction model. Compare the output result with the true value (the pollutant concentration information obtained during the period from 18:00 to 21:00 every day in the past year in the target area), calculate the loss value, and backpropagate the loss value layer by layer from the output layer to the input layer according to the gradient information of the loss function. Calculate the gradient value of each layer of parameters through the chain rule. During the parameter update process, use optimization algorithms such as gradient descent to update the parameters of each layer to reduce the value of the loss function. This process needs to be iterated multiple times until the performance of the target prediction model meets the preset requirements or reaches the maximum number of iterations, that is, the training of the target prediction model is completed, and the trained air pollutant prediction model is obtained.

[0050] Step 105: Predict the pollutant concentration in the target area during the event period based on the air pollutant prediction model to obtain the initial prediction result; adjust the initial prediction result according to the event pollution impact value to obtain the standard prediction result.

[0051] In some embodiments, the pollutant concentration in the target area during the event period (18:00 - 21:00 on May 10, 2023) is predicted by the trained air pollutant prediction model to obtain the initial prediction result. Taking the concentrations of PM2.5, PM10, and NO2 as examples, the initial prediction result may be that the PM2.5 concentration is 25 μg / m 3 ³, the PM10 concentration is 23 μg / m 3 ³, and the NO2 concentration is 27 μg / m 3 .

[0052] Optionally, the standard prediction result is determined through the following steps:

[0053] Determine the adjustment value of the initial prediction result according to the event pollution impact value, and calculate the adjustment value and the initial prediction result to obtain the standard prediction result.

[0054] In some embodiments, if the event pollution impact value is 0.714, the adjustment value of the initial prediction result is obtained by querying the pre - formulated pollution impact adjustment table. The pollution impact adjustment table includes multiple event pollution impact intervals and the corresponding adjustment values for each interval. The pollution impact adjustment table is obtained by converting the relationship between the event pollution impact value and the pollutant concentration. By matching the event pollution impact value of 0.714 in the pollution impact adjustment table, the corresponding adjustment value is 1.1. Multiply the adjustment value with the initial prediction result to obtain the standard prediction result. For example, for the initial prediction result with PM2.5 concentration of 25 μg / m 3 ³, PM10 concentration of 23 μg / m 3 ³, and NO2 concentration of 27 μg / m 3 respectively multiplied by the adjustment value, the standard prediction result may be that the PM2.5 concentration is 27.5 μg / m 3 ³, the PM10 concentration is 25.3 μg / m 3 ³, and the NO2 concentration is 29.7 μg / m 3 .

[0055] In some embodiments, by obtaining the target area and the holding time of a large-scale event, and using the pollutant concentration information and meteorological information of the target area during the target historical time period corresponding to the event holding time period as training samples to train the target prediction model. Therefore, the trained target prediction model is an atmospheric pollutant prediction model for the target area during the event holding time period, enabling the atmospheric pollutant prediction model to have local and targeted predictions. The holding of a large-scale event itself also affects the prediction of atmospheric pollutants. The type and scale of the event are the main factors affecting air pollution, and the environmental capacity, which is the ability of the environment to accommodate pollutants, plays a crucial role in determining the impact of a large-scale event on air pollution. By adjusting the initial prediction result with the activity pollution impact value determined by the event type, event scale, and the environmental capacity of the event area, the impact of the event on atmospheric pollutants can be considered more comprehensively, making it more in line with the actual situation, thereby reducing the error of the initial prediction result and improving the reliability and accuracy of the prediction result.

[0056] In some embodiments, in order to further solve Technical Problem 2 described in the background art section, that is, "since the holding time of a large-scale event is fixed and cannot be changed, if the pollutants exceed the standard before the event starts, it will affect the smooth progress of the event. In order to ensure that the air quality meets the standard during the event, an effective emission restriction plan needs to be formulated. The prior art lacks a targeted method for formulating an emission restriction plan and often adopts a 'one-size-fits-all' emission restriction", each pollution source among multiple pollution sources has a corresponding pollution source type, and the corresponding pollution source type of each pollution source is determined through the following steps:

[0057] Obtain the emission frequency of each pollution source during the target historical time period;

[0058] Determine the pollution source type of the pollution sources with a fixed emission frequency among the multiple pollution sources as fixed pollution sources to obtain a fixed pollution source group; determine the pollution source type of the pollution sources with a non-fixed emission frequency among the multiple pollution sources as temporary pollution sources.

[0059] In some embodiments, obtain the emission frequency of each pollution source during the target historical time period (18:00 - 21:00 every day in the past year) from the terminal of each pollution source. Assume that the multiple pollution sources include a first pollution source and a second pollution source. If the emission frequency of the first pollution source can be fixed at once at eight o'clock in the morning every day or twice a day, then the first pollution source is a fixed pollution source, such as factory emissions, power plant emissions, etc. If the emission frequency of the second pollution source is not fixed, it can be once at eight o'clock in the morning on Monday and once at nine o'clock in the morning on Monday, then the second pollution source is determined as a temporary pollution source, such as construction site dust, agricultural burning, etc.

[0060] There is a corresponding pollutant emission limit plan for stationary pollution sources, and the pollutant emission limit plan is generated through the following steps:

[0061] Step 1: Obtain the pollutant emission time periods of each stationary pollution source in the stationary pollution source group; determine the stationary pollution sources in the stationary pollution source group whose pollutant emission time periods overlap with the event time period as key pollution sources to obtain the key pollution source group;

[0062] In some embodiments, the pollutant emission time periods are recorded from the terminals of the stationary pollution sources, so the pollutant emission time periods of each stationary pollution source are obtained from the terminals of the stationary pollution sources. Suppose the stationary pollution source group includes stationary pollution source one, stationary pollution source two, and stationary pollution source three. Among them, stationary pollution source one can be a sewage treatment plant, and the pollutant emission time period is from 19:00 to 20:00 every day. Stationary pollution source two can be a waste incineration plant, and the pollutant emission time period is from 20:00 to 22:00 every day. Stationary pollution source three can be an industrial furnace and kiln plant, and the pollutant emission time period is from 6:00 to 7:00 every morning. When the event time period is from 18:00 to 21:00 on May 10, 2023, stationary pollution source one and stationary pollution source two overlap with the event time period, and stationary pollution source one and stationary pollution source two are determined as key pollution sources. The key pollution source group includes key pollution source one (stationary pollution source one) and key pollution source two (stationary pollution source two).

[0063] Step 2: Obtain the emissions of each key pollution source in the key pollution source group, calculate the emissions of each key pollution source, and obtain the total pollutant emissions corresponding to the key pollution source group;

[0064] Step 3: Determine the degree of pollutant overstandard according to the total pollutant emissions; compare the degree of pollutant overstandard with the preset standard degree. If the degree of pollutant overstandard is greater than or equal to the preset standard degree, generate a pollutant emission limit plan. The pollutant emission limit plan includes the restricted time and the standard emissions; send the pollutant emission limit plan to the terminal devices corresponding to each key pollution source in the key pollution source group so that the management personnel corresponding to the terminal devices corresponding to each key pollution source perform the operation of restricting pollutant emissions;

[0065] Among them, the standard emissions included in the pollutant emission limit plan are determined through the following steps:

[0066] Obtain the geographical location information of each key pollution source in the key pollution source group. According to the geographical location information of each key pollution source, determine the straight-line distance value between each key pollution source and the target area; Generate the first standard emission amount for the key pollution sources in the key pollution source group whose straight-line distance value is greater than or equal to the preset distance value; Generate the second standard emission amount for the key pollution sources in the key pollution source group whose straight-line distance value is less than the preset distance value, and the first standard emission amount is less than the second standard emission amount.

[0067] In some embodiments, obtain the emission amount of each key pollution source from the terminal of each pollution source. Add the emission amounts of key pollution source one and key pollution source two included in the key pollution source group to obtain that the total pollutant emission corresponding to the key pollution source group can be 800 kilograms. Obtain the pre-configured pollutant emission comparison table, which includes multiple pollutant emission amounts and the pollutant over-standard degree corresponding to each pollutant emission amount. By querying the total pollutant emission corresponding to the key pollution source group in the pollutant emission comparison table, the pollutant over-standard degree is obtained as 60%. Compare the pollutant over-standard degree with the preset standard degree (50%). If the pollutant over-standard degree is greater than or equal to the preset standard degree, generate a pollutant restricted emission plan, which includes the restricted time and the standard emission amount. Specifically, the restricted time included in the pollutant restricted emission plan can be from 18:00 to 21:00 on May 10, 2023 for restricted emission. The standard emission amount included in the pollutant restricted emission plan is determined through the following steps:

[0068] Obtain the geographical location information of each key pollution source in the key pollution source group through an open-source electronic map interface. By obtaining the geographical location information of each key pollution source and the regional location information of the target area through the open-source electronic map interface, the straight-line distance value between each key pollution source and the target area can be obtained, which can be five kilometers; Generate the first standard emission amount (obtain the first standard emission amount as 100 kilograms by querying the pre-established pollutant emission restriction table) for the key pollution sources in the key pollution source group whose straight-line distance value is greater than or equal to the preset distance value (three kilometers); Generate the second standard emission amount (obtain the second standard emission amount as 200 kilograms by querying the pre-established pollutant emission restriction table) for the key pollution sources in the key pollution source group whose straight-line distance value is less than the preset distance value, and the first standard emission amount is less than the second standard emission amount. Send the pollutant restricted emission plan to the terminal device corresponding to each key pollution source in the key pollution source group, so that the management personnel corresponding to the terminal device of each key pollution source perform the operation of restricting pollutant emissions. The terminal device can be the work computer of the management personnel corresponding to each key pollution source.

[0069] In some embodiments, due to different pollutant generation mechanisms and conditions, different types of pollution sources often have different emission frequencies. Therefore, the emission frequency is an important basis for distinguishing different types of pollution sources. Determining the type of pollution source based on whether the emission frequency of the pollution source is fixed improves the accuracy of pollution source identification and helps formulate targeted pollution control plans. The decline in environmental quality may affect the smooth progress of activities. By restricting the emissions of fixed pollution sources during activities in advance, the smooth progress of activities can be ensured. By formulating different pollutant emission limits for different key pollution sources, the emissions of each pollution source can be controlled more precisely, avoiding a one-size-fits-all emission limit.

[0070] In some embodiments, in order to further solve Technical Problem 3 described in the background art section, that is, "when there are multiple malfunctioning air quality monitoring points, the staff lacks a clear maintenance policy, resulting in low maintenance efficiency, so that some urgently needed monitoring points cannot be processed in time, affecting the smooth progress of activities", in some embodiments of the present invention, a target area is configured with multiple air quality monitoring points, and each air quality monitoring point among the multiple air quality monitoring points is configured with an indicator light. Each air quality monitoring point has a corresponding maintenance priority level, and the corresponding maintenance priority level of each air quality monitoring point is determined through the following steps:

[0071] Step 1, obtain the device identifier and device location information of each air quality monitoring point among the multiple air quality monitoring points in the target area; obtain the pollutant concentration data set of each air quality monitoring point.

[0072] Step 2, calculate the pollutant concentration data set of each air quality monitoring point to obtain the standard deviation of the pollutant concentration data set of each air quality monitoring point; for the multiple air quality monitoring points, determine the air quality monitoring points corresponding to the pollutant concentration data exceeding three times the standard deviation in the pollutant concentration data set of each air quality monitoring point as abnormal monitoring points, obtain a group of abnormal monitoring points, and generate the abnormal information corresponding to each abnormal monitoring point in the group of abnormal monitoring points and control the corresponding indicator light to turn on, to obtain a group of abnormal monitoring points, where the abnormal information includes the device identifier and device location information.

[0073] Step 3, obtain a group of key location points in the target area, and determine whether there are abnormal monitoring points within the preset range of the group of key location points. If so, determine the location attribute of the corresponding abnormal monitoring point as level 1. If not, determine the location attribute of the corresponding abnormal monitoring point as level 2; according to the location attribute of the abnormal monitoring point, score each abnormal monitoring point in the group of abnormal monitoring points to obtain the importance score of each abnormal monitoring point.

[0074] Step 4: Obtain the location information of the maintenance station. Based on the location information of the maintenance station, determine the maintenance distance from the maintenance station to each abnormal monitoring point in the abnormal monitoring point group; based on the maintenance distance of each abnormal monitoring point, determine the maintenance distance score corresponding to each abnormal monitoring point; configure weights for the importance score and the maintenance distance score corresponding to each abnormal monitoring point respectively, and perform weighted summation on the importance score and the maintenance distance score corresponding to each abnormal monitoring point through the weights to obtain the maintenance score of each abnormal monitoring point.

[0075] Step 5: Determine the maintenance priority of each abnormal monitoring point according to the maintenance score of each abnormal monitoring point.

[0076] In some embodiments, obtain the device identifier and device location information of each air quality monitoring point from the terminal. The air quality monitoring points are used to monitor the pollutant concentration data of activities. Among them, multiple air quality monitoring points may include Air Quality Monitoring Point 1, Air Quality Monitoring Point 2, and Air Quality Monitoring Point 3. For example, the device identifier of Air Quality Monitoring Point 1 is "1" and the device location information is Audience Area 1, and the device identifier of Air Quality Monitoring Point 2 is "2" and the device location information is Work Area 1. Obtain the corresponding pollutant concentration data from each air quality monitoring point to obtain the pollutant concentration data group of each air quality monitoring point. For example, the pollutant concentration data group of Air Quality Monitoring Point 1 may be the pollutant concentration data at 18:30, the pollutant concentration data at 18:40, and the pollutant concentration data at 18:50.

[0077] In some embodiments, the pollutant concentration data of Air Quality Monitoring Point 1 includes PM10, PM2.5, SO2, NO2, O3, etc. Taking the PM2.5 concentration data in Air Quality Monitoring Point 1 as an example, the PM2.5 concentration data group of Air Quality Monitoring Point 1 is {50, 55, 60, 65, 70, 250}, (unit: μg / m 3 ), first calculate the mean and standard deviation of this data group: the mean is: 83.33, and then calculate the standard deviation to get the standard deviation of 77.11. Among them, only 250 in the PM2.5 concentration data group exceeds three times the standard deviation. For example, 250 is greater than 3 multiplied by 77.11, then determine Air Quality Monitoring Point 1 as an abnormal monitoring point, obtain the abnormal monitoring point group, and generate the abnormal information corresponding to each abnormal monitoring point in the abnormal monitoring point group and control the corresponding indicator light to turn on to obtain the abnormal monitoring point group. Among them, the abnormal information includes the device identifier and the device location information. Since the pollutant concentration data of Air Quality Monitoring Point 1 includes PM10, PM2.5, SO2, NO2, O3, etc., as long as any one of the pollutants in the pollutant concentration data exceeds three times the standard deviation, the corresponding air quality monitoring point is determined as an abnormal monitoring point.

[0078] In some embodiments, key location point groups of a target area are obtained from a terminal. Each key location point in the key location point group includes a main passage or an activity center of an activity, where a large number of people often gather. Suppose the key location point group includes Key Location Point One and Key Location Point Two. The abnormal monitoring point group includes Abnormal Monitoring Point One and Abnormal Monitoring Point Two. When Abnormal Monitoring Point One is within the preset range (5 meters) of Key Location Point One in the key location point group, the location attribute of Abnormal Monitoring Point One is determined to be level one. If Abnormal Monitoring Point Two is outside the preset ranges of Key Location Point One and Key Location Point Two in the key location point group, the location attribute of Abnormal Monitoring Point Two is determined to be level two. The maintenance station is within the target area. The maintenance station location information of the maintenance station is obtained from the electronic map interface of the target area. Through the maintenance station location information, the maintenance distance from each abnormal monitoring point in the abnormal monitoring point group to the maintenance station is determined. For example, the maintenance distance from the maintenance station to Abnormal Monitoring Point One is 28 meters, and the maintenance distance from the maintenance station to Abnormal Monitoring Point Two is 25 meters. The maintenance comparison table is obtained from the database. The maintenance comparison table includes multiple location attributes, the scores corresponding to each location attribute, multiple maintenance distances, and the scores corresponding to each maintenance distance. By querying the location attribute and maintenance distance of each abnormal monitoring point in the maintenance comparison table, the importance score and the maintenance distance score of each abnormal monitoring point are obtained. Specifically, the importance score of Abnormal Monitoring Point One is 95 points, and the maintenance distance score is 90 points; the importance score of Abnormal Monitoring Point Two is 85 points, and the maintenance distance score is 92 points. A weight of 60% is configured for the importance score corresponding to each abnormal monitoring point, and a weight of 40% is configured for the maintenance distance score. The importance score and the maintenance distance score corresponding to each abnormal monitoring point are weighted and summed through the weights to obtain the maintenance score of each abnormal monitoring point. The maintenance score of Abnormal Monitoring Point One is 93 points (95×60% + 90×40% = 93), and the maintenance score of Abnormal Monitoring Point Two is 87.8 points (85×60% + 92×40% = 87.8). The maintenance score of Abnormal Monitoring Point One is greater than the maintenance score of Abnormal Monitoring Point Two. Therefore, the maintenance priority of Abnormal Monitoring Point One is high, and Abnormal Monitoring Point One is preferentially repaired.

[0079] In these embodiments, according to the characteristics of the normal distribution, data points falling outside three standard deviations from the mean are considered extremely rare. Therefore, the three - standard - deviation method is used to identify abnormal monitoring points in the pollutant concentration data group, which can improve the accuracy of determining abnormal monitoring points; an indicator light reminder is turned on for the abnormal monitoring points and abnormal information is generated. The indicator light can remind the staff, reduce the response time, enabling them to discover and respond to the abnormal monitoring points in the first time, and eliminating the need for the staff to make active judgments, thus simplifying the troubleshooting process and saving troubleshooting manpower; a large number of people often gather in the key areas of the activity. The maintenance priority is determined through the location attribute of each abnormal monitoring point, improving the timeliness of abnormal monitoring point maintenance and ensuring the smooth progress of the activity.

[0080] The above description is only some preferred embodiments of the present invention and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the invention involved in the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present invention.

Claims

1. A method for predicting air pollutants based on big data, characterized in that, Including: Obtain the regional location information and event information of the target area, and determine multiple pollution sources within the preset range of the target area according to the regional location information; The event information includes the event time period, event type, and event scale; Determine the event pollution impact value according to the event type and event scale, including: obtaining multiple historical event data, where the historical event data includes historical event types and historical event scales; matching according to the event type and event scale in the multiple historical event data to obtain target historical event data; obtaining the historical event pollution emissions of the target historical event data; obtaining the environmental capacity of the target area and the pre-established environmental quality standard table, and by querying the environmental capacity in the environmental quality standard table, obtaining the standard allowable emissions of the target area, where the environmental quality standard table includes different environmental capacities and the standard allowable emissions corresponding to each environmental capacity; calculating the ratio of the historical event pollution emissions to the standard allowable emissions to obtain the event pollution impact value; Determine the target historical time period according to the event time period and obtain the pollutant concentration information and meteorological information of each pollution source within the target historical time period; obtain multiple alternative prediction models and the prediction accuracy of each alternative prediction model; screen the multiple alternative prediction models according to the prediction accuracy of the alternative prediction models to obtain the target prediction model; Train the target prediction model through the pollutant concentration information and meteorological information, and use the trained target prediction model as the air pollutant prediction model; Predict the pollutant concentration of the target area during the event time period based on the air pollutant prediction model to obtain the initial prediction result; adjust the initial prediction result according to the event pollution impact value to obtain the standard prediction result, including: according to the event pollution impact value, obtain the adjustment value of the initial prediction result by querying the pre-established pollution impact adjustment table; calculate the adjustment value and the initial prediction result to obtain the standard prediction result.

2. The method for predicting air pollutants based on big data according to claim 1, wherein Each pollution source among the multiple pollution sources has a corresponding pollution source type, and the corresponding pollution source type of each pollution source is determined through the following steps: Obtain the emission frequency of each pollution source within the target historical time period; Determine the pollution source type of the pollution sources with fixed emission frequencies among the multiple pollution sources as fixed pollution sources to obtain a fixed pollution source group; determine the pollution source type of the pollution sources with non-fixed emission frequencies among the multiple pollution sources as temporary pollution sources.

3. The method for predicting air pollutants based on big data according to claim 2, wherein The fixed pollution sources have corresponding pollutant emission limit schemes, and the pollutant emission limit schemes are generated through the following steps: Obtain the pollutant emission time periods of each fixed pollution source in the fixed pollution source group; determine the fixed pollution sources in the fixed pollution source group whose pollutant emission time periods coincide with the event time period as key pollution sources to obtain a key pollution source group; Obtain the emissions of each key pollution source in the key pollution source group, and calculate the emissions of each key pollution source to obtain the total pollutant emissions corresponding to the key pollution source group; Determine the degree of pollutant over - standard according to the total amount of pollutant emissions; compare the degree of pollutant over - standard with the preset standard degree. If the degree of pollutant over - standard is greater than or equal to the preset standard degree, generate a pollutant restricted emission plan, and the pollutant restricted emission plan includes a restricted time and a standard emission amount; send the pollutant restricted emission plan to the terminal device corresponding to each key pollution source in the key pollution source group, so that the management personnel corresponding to the terminal device corresponding to each key pollution source perform the operation of restricting pollutant emissions; Among them, the standard emission amount included in the pollutant restricted emission plan is determined through the following steps: Obtain the geographical location information of each key pollution source in the key pollution source group, and determine the straight - line distance value between each key pollution source and the target area according to the geographical location information of each key pollution source; generate a first standard emission amount for the key pollution sources in the key pollution source group whose straight - line distance value is greater than or equal to the preset distance value; generate a second standard emission amount for the key pollution sources in the key pollution source group whose straight - line distance value is less than the preset distance value, and the first standard emission amount is less than the second standard emission amount.

Images