Air quality prediction and control method and system based on urban multidimensional data

Abstract

The present application provides an air quality prediction and control method and system based on urban multi-dimensional data, which relates to the field of intelligent control technology. The method includes: collecting meteorological factor characteristic values; collecting pollutant emission characteristic values; generating traffic flow characteristic values; generating air quality index prediction values; adjusting and optimizing traffic flow characteristic values ​​and pollutant emission characteristic values ​​to obtain recommended pollutant emission characteristic values ​​and recommended traffic flow characteristic values; and controlling quality according to the recommended pollutant emission characteristic values ​​and recommended traffic flow characteristic values. The present application can solve the technical problem that the accuracy of air quality prediction is insufficient due to the single consideration factor and the lack of fusion analysis of multi-dimensional data in the prior art. By fusing multi-dimensional data to perform air quality analysis and prediction, and then optimizing air quality control, the accuracy of air quality prediction can be improved, and the technical effect of precise air quality control can be achieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and in particular to an air quality prediction and control method and system based on urban multi-dimensional data. Background Art

[0002] With the acceleration of urbanization, air quality issues are becoming increasingly prominent. Accurate air quality prediction and effective control measures are crucial to improving air quality and protecting public health.

[0003] The deterioration of urban air quality is related to many factors. Traditional air quality prediction usually relies on meteorological forecasts and considers a single factor. However, due to information islands and the modal complexity of multidimensional data, it is difficult to achieve fusion analysis, which in turn affects the accuracy of air quality prediction and the effect of air control.

[0004] In summary, the existing technology has technical problems that the accuracy of air quality prediction is insufficient due to the single consideration of factors and the lack of fusion analysis of multi-dimensional data. Summary of the invention

[0005] The purpose of this application is to provide an air quality prediction and control method and system based on urban multi-dimensional data, so as to solve the technical problem in the prior art that the accuracy of air quality prediction is insufficient due to the single consideration factor and the lack of fusion analysis of multi-dimensional data.

[0006] In view of the above problems, the present application provides an air quality prediction and control method and system based on urban multi-dimensional data.

[0007] In the first aspect, the present application provides an air quality prediction and control method based on urban multidimensional data, and the method is implemented by an air quality prediction and control system based on urban multidimensional data, wherein the method includes: collecting meteorological factor characteristic values ​​of a preset city in a time zone to be predicted through the Internet of Things; collecting pollutant emission characteristic values ​​of the preset city in a time zone to be predicted through pollution disclosure information; mining historical data through a traffic flow statistics channel to generate traffic flow characteristic values ​​of the preset city in a time zone to be predicted; fitting through an air quality prediction network according to the meteorological factor characteristic values, the pollutant emission characteristic values, the traffic flow characteristic values ​​and the population density characteristic values ​​to generate an air quality index prediction value; when the air quality index prediction value is greater than or equal to the air quality index threshold, adjusting and optimizing the traffic flow characteristic value and the pollutant emission characteristic value to obtain recommended pollutant emission characteristic values ​​and recommended traffic flow characteristic values; and controlling quality according to the recommended pollutant emission characteristic values ​​and the recommended traffic flow characteristic values.

[0008] In the second aspect, the present application also provides an air quality prediction and control system based on urban multidimensional data, which is used to execute the air quality prediction and control method based on urban multidimensional data as described in the first aspect, wherein the system includes: a meteorological factor feature acquisition module, which is used to collect meteorological factor feature values ​​of a preset city to be predicted time zone through the Internet of Things; a pollutant emission feature acquisition module, which is used to collect pollutant emission feature values ​​of the preset city to be predicted time zone through pollution disclosure information; a traffic flow feature mining module, which is used to perform historical data mining through a traffic flow statistical channel to generate traffic flow feature values ​​of the preset city to be predicted time zone; an air quality fitting module, which is used to fit the meteorological factor feature values, the pollutant emission feature values, the traffic flow feature values ​​and the population density feature values ​​through an air quality prediction network to generate an air quality index prediction value; an air quality optimization module, which is used to adjust and optimize the traffic flow feature values ​​and the pollutant emission feature values ​​when the air quality index prediction value is greater than or equal to the air quality index threshold, and obtain recommended pollutant emission feature values ​​and recommended traffic flow feature values; an air quality control module, which is used to control quality control according to the recommended pollutant emission feature values ​​and the recommended traffic flow feature values.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] Through the Internet of Things, the meteorological factor characteristic values ​​of the preset city's predicted time zone are collected; through the pollution disclosure information, the pollutant emission characteristic values ​​of the preset city's predicted time zone are collected; through the traffic flow statistics channel, historical data mining is performed to generate the traffic flow characteristic values ​​of the preset city's predicted time zone; according to the meteorological factor characteristic values, the pollutant emission characteristic values, the traffic flow characteristic values ​​and the population density characteristic values, the air quality prediction network is fitted to generate the air quality index prediction value; when the air quality index prediction value is greater than or equal to the air quality index threshold, the traffic flow characteristic value and the pollutant emission characteristic value are adjusted and optimized to obtain the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value; according to the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value, the quality control is performed. In this way, air quality analysis and prediction are performed by integrating multi-dimensional data, and then the air quality control optimization is performed, so as to improve the accuracy of air quality prediction and achieve the technical effect of precise air quality control.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0013] Figure 1 This is a flow chart of the air quality prediction and control method based on urban multi-dimensional data in this application.

[0014] Figure 2 This is a structural diagram of the air quality prediction and control system based on urban multi-dimensional data in this application.

[0015] Explanation of the accompanying drawings: meteorological factor feature collection module 11, pollutant emission feature collection module 12, traffic flow feature mining module 13, air quality fitting module 14, air quality optimization module 15, air quality control module 16. DETAILED DESCRIPTION

[0016] This application provides an air quality prediction and control method and system based on urban multi-dimensional data, which solves the technical problem in the prior art that the accuracy of air quality prediction is insufficient due to the lack of fusion analysis of multi-dimensional data and the single consideration of factors. By fusing multi-dimensional data to perform air quality analysis and prediction, and then optimizing air quality control, the accuracy of air quality prediction is improved, thereby achieving the technical effect of precise air quality control.

[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] Embodiment 1

[0019] Please see attached Figure 1 The present application provides an air quality prediction and control method based on urban multidimensional data, wherein the method is applied to an air quality prediction and control system based on urban multidimensional data, and the method specifically comprises the following steps:

[0020] Step 1: Collect the characteristic values ​​of meteorological factors of the time zone to be predicted in the preset city through the Internet of Things;

[0021] Specifically, the preset city refers to any city for which air quality prediction and control is to be carried out, and the prediction time zone refers to the time interval for which air quality prediction and control is to be carried out, which needs to be determined in combination with actual conditions. The characteristic values ​​of meteorological factors include parameter values ​​of meteorological parameters such as temperature, humidity, wind speed, wind direction, air pressure, and precipitation. Specifically, sensors for monitoring meteorological factors can be selected and deployed in preset cities. Based on the existing Internet of Things technology, these sensors are connected to the data center using wireless communication technology to receive the data collected by the sensors and obtain the characteristic values ​​of meteorological factors.

[0022] Step 2: Collect the pollutant emission characteristic values ​​of the preset city in the predicted time zone through the pollution disclosure information;

[0023] Specifically, first, the pollution discharge public information of the preset city is extracted through the existing pollution discharge permit management information platform. The pollution discharge public information contains the pollutant emission information of different enterprises in the preset city at different times. The pollutant emission information of the time zone to be predicted is extracted from the pollution discharge public information, including the pollutant emission amount and pollutant type of different enterprises. Data analysis is carried out, and the pollutant emission characteristic values ​​are obtained after eliminating outliers.

[0024] Step 3: Perform historical data mining through the traffic flow statistics channel to generate traffic flow characteristic values ​​for the time zone to be predicted in the preset city;

[0025] Specifically, vehicle travel within the city is an important factor affecting urban air quality. The traffic flow statistics channel is used to mine the historical traffic flow data of the preset city and conduct traffic flow growth rate analysis, so as to predict the traffic flow in the predicted time zone and obtain the traffic flow characteristic value.

[0026] Step 4: According to the characteristic values ​​of the meteorological factors, the characteristic values ​​of the pollutant emissions, the characteristic values ​​of the traffic flow and the characteristic values ​​of the population density, an air quality prediction network is used to perform fitting to generate an air quality index prediction value;

[0027] Specifically, meteorological factors, pollutant emissions, traffic flow and population density are all key factors affecting air quality. Among them, the characteristic value of population density can be obtained based on the most recent census data of the preset city. By constructing an air quality prediction network, these characteristic values ​​can be effectively combined to generate an air quality index prediction value. The air quality prediction network can be constructed based on existing deep learning models, such as convolutional neural networks, recurrent neural networks, etc., and trained to convergence through sample data, so as to predict air quality based on the characteristic values ​​of meteorological factors, the characteristic values ​​of pollutant emissions, the characteristic values ​​of traffic flow and the characteristic values ​​of population density, and obtain an air quality index prediction value. The generated air quality index prediction value represents the air quality in the preset city in the predicted time zone, providing support for subsequent air quality control.

[0028] Step 5: When the air quality index prediction value is greater than or equal to the air quality index threshold, the traffic flow characteristic value and the pollutant emission characteristic value are adjusted and optimized to obtain a recommended pollutant emission characteristic value and a recommended traffic flow characteristic value;

[0029] Specifically, the air quality index is an indicator to measure the air pollution index. The larger the air quality index, the worse the air quality, that is, the more serious the pollution. The air quality index threshold is the maximum air quality index at which the air quality meets the specified requirements, which needs to be determined in combination with actual conditions. When the predicted air quality index value is greater than or equal to the air quality index threshold, it indicates that the air quality is at an unhealthy or dangerous level. The air quality can be improved and the air quality index can be reduced by adjusting and optimizing the traffic flow characteristic value and the pollutant emission characteristic value. That is, the traffic flow characteristic value and the pollutant emission characteristic value are used as optimization variables. Through optimization analysis, the air quality index is made less than the air quality index threshold. The traffic flow characteristic value and the pollutant emission characteristic value at this time are obtained as the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value.

[0030] Step six: performing control quality regulation based on the recommended pollutant emission characteristic values ​​and the recommended traffic flow characteristic values.

[0031] Specifically, the recommended pollutant emission characteristic values ​​and the recommended traffic flow characteristic values ​​are used as the pollutant emission standards and traffic flow standards for the preset time zone, and the pollutant emissions of enterprises and the traffic flow of the preset city are monitored and controlled to improve the air quality.

[0032] Furthermore, step 2 of this application also includes:

[0033] According to the pollution disclosure information, a pollutant emission type list and a pollutant emission amount list of the first enterprise's disclosure information in the preset city's predicted time zone are obtained; outlier analysis is performed through the pollutant emission amount list and the pollutant emission type list to obtain abnormal emission amount pollutant types; the abnormal emission amount pollutant types are sent to the first enterprise for correction, and the pollutant emission characteristic values ​​of the first enterprise are obtained, which are added to the pollutant emission characteristic values.

[0034] Furthermore, the present application also includes the following steps:

[0035] The pollutant emission list is traversed through the normalization function for processing to obtain a normalized characteristic value list, wherein the pollutant emission type list and the normalized characteristic value list correspond one to one; the first enterprise is publicly disclosed for retrospective pollution disclosure using the pollutant emission type list to obtain several enterprise public disclosure retrospective information; several pollutant retrospective emission lists of several enterprise public disclosure retrospective information are traversed through the normalization function for processing respectively to obtain several retrospective normalized characteristic value lists; according to the several retrospective normalized characteristic value lists, a normalized characteristic value constraint interval list of the pollutant emission type list is constructed; according to the normalized characteristic value constraint interval list, an abnormal value analysis is performed on the normalized characteristic value list to obtain the abnormal emission pollutant type; wherein the normalization function is:

[0036] ;

[0037] in, Characterize the normalized eigenvalues, Characterize the pollutant emissions, The minimum value of pollutant emissions that characterizes the pollutant emission list, The maximum value of a pollutant emission that represents a pollutant emission list.

[0038] Specifically, the process of collecting the pollutant emission characteristic values ​​of the preset city in the predicted time zone through the pollution disclosure information is as follows:

[0039] The pollution discharge public information includes the pollutant emission information of different enterprises in a preset city at different times. The first enterprise refers to any enterprise that emits pollutants. The pollutant emission information in the time zone to be predicted is extracted from the pollution discharge public information, including the pollutant emission type and the corresponding emission amount, such as the emission amount of sulfur dioxide, carbon monoxide, etc. The pollutant emission type list and pollutant emission amount list of the first enterprise public information are generated based on the different pollutant emission types and corresponding emission amounts corresponding to the first enterprise.

[0040] Further outlier analysis is performed through the pollutant emission list and the pollutant emission type list to obtain the pollutant type with abnormal emission, that is, to identify the pollutant type with abnormal emission, which may be a data collection error, and send it to the first enterprise for verification and correction. The specific process is as follows:

[0041] First, the pollutant emission list is traversed and processed by a normalization function to obtain a normalized eigenvalue list, wherein the normalization function is:

[0042] ;

[0043] in, Characterize the normalized eigenvalues, Characterizes the pollutant emissions, generally refers to the pollutant emissions corresponding to any pollutant emission type in the pollutant emission list, The minimum value of pollutant emissions that characterizes the pollutant emission list, The maximum value of the pollutant emission that characterizes the pollutant emission list can be calculated through a normalization function to obtain the normalized characteristic value corresponding to any pollutant emission type in the pollutant emission type list to form a normalized characteristic value list. It can be understood that the pollutant emission type list and the normalized characteristic value list correspond one to one.

[0044] Furthermore, the pollutant emission type list is used to conduct a retrospective pollutant emission disclosure of the first enterprise, that is, based on the pollutant emission disclosure information, the historical pollutant emissions of different enterprises in historical time are obtained to construct several retrospective pollutant emission lists, which are used as several retrospective enterprise disclosure information, and several retrospective pollutant emission lists of several enterprise disclosure information are traversed through a normalization function and processed separately to obtain several retrospective normalized characteristic value lists. According to the several retrospective normalized characteristic value lists, several retrospective normalized characteristic values ​​belonging to the same pollutant type in the several retrospective normalized characteristic value lists are counted, and a normalized characteristic value constraint interval is formed with the minimum and maximum values ​​of the several retrospective normalized characteristic values, thereby obtaining a normalized characteristic value constraint interval list corresponding to the pollutant emission type list.

[0045] According to the normalized eigenvalue constraint interval list, an outlier analysis is performed on the normalized eigenvalue list, that is, it is identified whether the normalized eigenvalue corresponding to any pollutant emission type in the normalized eigenvalue list is in the normalized eigenvalue constraint interval in the normalized eigenvalue constraint interval list, and the pollutant type that is not in the normalized eigenvalue constraint interval in the normalized eigenvalue constraint interval list is used as the abnormal emission pollutant type. Generally speaking, the pollutant emission type of an enterprise in the short term is a fixed type, so the amount of pollution emissions generally tends to be consistent, so the values ​​after normalization should tend to be consistent, thereby achieving rapid outlier analysis and ensuring the accuracy of subsequent analysis processes.

[0046] The abnormal emission pollutant type is sent to the first enterprise, which corrects the emission and uses the pollutant emission corresponding to the corrected abnormal emission pollutant type as the pollutant emission characteristic value of the first enterprise, and adds it to the pollutant emission characteristic value to realize the correction analysis of the pollutant emission characteristics and ensure the accuracy of subsequent air quality predictions.

[0047] Furthermore, step three of this application also includes:

[0048] Through the traffic flow statistics channel, historical data is backtracked to obtain several traffic flow month-on-month growth rates in preset cities; through the traffic flow statistics channel, historical data is backtracked to obtain several traffic flow month-on-month growth rates in preset cities; a central trend analysis is performed on the several traffic flow month-on-month growth rates to obtain a monthly month-on-month growth rate characteristic value; a central trend analysis is performed on the several traffic flow month-on-month growth rates to obtain a monthly year-on-year growth rate characteristic value; when the deviation between the monthly month-on-month growth rate characteristic value and the monthly year-on-year growth rate characteristic value is less than or equal to the growth rate deviation threshold, the traffic flow characteristic value is calculated according to the average of the monthly month-on-month growth rate characteristic value and the monthly year-on-year growth rate characteristic value; otherwise, the month before the month in the time zone to be predicted is taken as the cutoff month The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 2. The method comprises the following steps: 3. The method comprises the following steps: 4. The method comprises the following steps: 5. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 2. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 2. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 2. The method comprises the following steps: 3. The method comprises the following steps: 4. The method comprises the following steps: 5. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps: 1. The method comprises the following steps:

[0049] Specifically, the process of mining historical data through the traffic flow statistics channel to generate the traffic flow characteristic value of the time zone to be predicted in the preset city is as follows:

[0050] First, obtain the historical traffic flow data of the preset city in the past period of time, including the traffic flow of each month, the traffic flow during peak hours, the traffic flow of specific sections, etc. The monthly month-on-month growth rate of traffic flow is calculated for the historical traffic flow data through the traffic flow statistics channel. The monthly month-on-month growth rate of traffic flow refers to the relative change rate of traffic flow between two consecutive months. The traffic flow of the current month minus the traffic flow of the previous month is compared with the traffic flow of the previous month. The result is a number of monthly month-on-month growth rates of traffic flow. Through the traffic flow statistics channel, historical data is traced back to obtain several monthly year-on-year growth rates of traffic flow in the preset city. The monthly year-on-year growth rate of traffic flow refers to the relative change rate between a certain month this year and the same month last year, that is, the traffic flow of the current month minus the traffic flow of the same month last year is compared with the traffic flow of the same month last year. The result is a number of monthly year-on-year growth rates of traffic flow.

[0051] The purpose of conducting a central tendency analysis on the several month-on-month growth rates of traffic flow is to obtain the general level or central trend of the month-on-month growth rate of traffic flow. The central tendency analysis usually includes calculating statistics such as the mean, median, mode, or analyzing the central tendency through a box plot, which is a common technical means used by those skilled in the art. An arbitrary method is selected for central tendency analysis to obtain a monthly month-on-month growth rate characteristic value representing the general level or central trend of the month-on-month growth rate of traffic flow. Similarly, the same method is used to conduct a central tendency analysis on the several month-on-month year-on-year growth rates of traffic flow to obtain a month-on-year year-on-year growth rate characteristic value representing the general level or central trend of the month-on-month year-on-year growth rate of traffic flow.

[0052] A growth rate deviation threshold is further set. The growth rate deviation threshold is a smaller deviation value, that is, the deviation threshold that considers that the month-on-year growth rate and the month-on-month growth rate are tending to be consistent, and is set by those skilled in the art. When the deviation between the month-on-month growth rate characteristic value and the month-on-year growth rate characteristic value is less than or equal to the growth rate deviation threshold, the growth rate mean of the month-on-month growth rate characteristic value and the month-on-year growth rate characteristic value is calculated, and the traffic flow characteristic value is calculated with the growth rate mean, that is, the historical traffic flow in the time zone to be predicted is obtained, and the historical traffic flow is calculated according to the growth rate mean to obtain the traffic flow characteristic value.

[0053] When the deviation between the characteristic value of the month-on-month growth rate and the characteristic value of the month-on-year growth rate is greater than the growth rate deviation threshold, the characteristic values ​​of the month-on-month growth rate in previous years are traced back with the month before the month in the time zone to be predicted as the end month, and the traced years are at least 5 years, that is, the monthly month-on-month growth rate of traffic flow from the predicted time zone month to the past 5 years is calculated, and a central trend analysis is performed to obtain several characteristic values ​​of the month-on-month growth rate in previous years. At the same time, several month-on-month growth rates to be analyzed from the month before the month in the time zone to be predicted to the month in the time zone to be predicted in the traced year are calculated. For example, if the month in the time zone to be predicted is October, the monthly month-on-month growth rate between September and October of each year in the traced year is calculated, and a central trend analysis is performed to obtain several month-on-month growth rates to be analyzed.

[0054] Further, the deviations of the characteristic values ​​of the month-on-month growth rates in previous years and the month-on-month growth rates to be analyzed are calculated respectively, and the deviation values ​​are used as the abnormal sub-coefficients of the growth rates, and then the average of the abnormal sub-coefficients of the growth rates is calculated and set as the abnormal growth rate coefficient. The abnormal growth rate coefficient threshold is set by the professional and technical personnel in this field. When the abnormal growth rate coefficient is greater than or equal to the abnormal growth rate coefficient threshold, it means that the deviation between the characteristic values ​​of the month-on-month growth rate in previous years and the month-on-month growth rate to be analyzed is large, that is, the traffic flow in the time zone to be predicted may surge, for example, the time zone to be predicted may be a tourist peak season. At this time, according to the characteristic value of the month-on-month growth rate, the traffic flow in the same month of the previous year in the time zone to be predicted is calculated, and the result is the characteristic value of the traffic flow. When the abnormal growth rate coefficient is less than the abnormal growth rate coefficient threshold, according to the characteristic value of the month-on-month growth rate, the traffic flow in the previous month in the time zone to be predicted is calculated, and the result is the characteristic value of the traffic flow. Thus, by analyzing the month-on-month year-on-year growth rate and the month-on-month growth rate, the traffic flow is calculated, the traffic flow prediction accuracy is improved, and the accuracy of the subsequent processing steps is improved.

[0055] Furthermore, step 4 of this application also includes:

[0056] Collect a preset urban air quality monitoring data set, wherein the preset urban air quality monitoring data set includes meteorological factor record data, pollutant emission record data, traffic flow record data, population density record data and air quality index record data; construct a normalized matrix based on the meteorological factor record data, the pollutant emission record data, the traffic flow record data and the population density record data, wherein the meteorological factor record data, the pollutant emission record data, the traffic flow record data and the population density record data in the normalized matrix are distributed at preset element positions; configure the air quality prediction network using the normalized matrix as input data and the air quality index record data as output supervision data.

[0057] Specifically, according to the characteristic values ​​of the meteorological factors, the characteristic values ​​of the pollutant emissions, the characteristic values ​​of the traffic flow and the characteristic values ​​of the population density, the process of fitting through the air quality prediction network to generate the predicted value of the air quality index is as follows:

[0058] Collect the preset city air quality monitoring data set, which can be understood as the air quality monitoring data set of the preset city in the past period of time, including meteorological factor record data, pollutant emission record data, traffic flow record data, population density record data and air quality index record data. The meteorological factor record data, pollutant emission record data, traffic flow record data and population density record data with corresponding relationships in the preset city air quality monitoring data set are arranged according to the preset element positions to form a multidimensional data matrix, and the data of the multidimensional data matrix is ​​normalized so that it falls within the same numerical range, for example, using minimum-maximum normalization for normalization, which is a common technical means for those skilled in the art, and the normalized multidimensional data matrix is ​​used as the normalized matrix. The preset element position refers to the matrix element position predetermined by professional and technical personnel in this field. For example, assuming that the normalized matrix is ​​set to a 1×4 matrix, it can be predetermined where the meteorological factor record data, pollutant emission record data, traffic flow record data and population density record data are located, thereby obtaining the preset element position.

[0059] Furthermore, an existing machine learning model, such as a recurrent neural network, is selected to construct an air quality prediction network, and the normalized matrix is ​​used as input data, and the air quality index record data is used as output supervision data to train the air quality prediction network so that the air quality prediction network can learn the mapping relationship between the normalized matrix and the air quality index, thereby training the air quality prediction network to convergence. Finally, according to the preset element positions, the meteorological factor characteristic values, the pollutant emission characteristic values, the traffic flow characteristic values, and the population density characteristic values ​​are normalized to construct a normalized characteristic matrix, and the normalized characteristic matrix is ​​input into the air quality prediction network, and the air quality index prediction value is output. By fusing multidimensional data, accurate air quality prediction is achieved, which is convenient for subsequent air quality control.

[0060] Furthermore, step five of this application also includes:

[0061] Configure a list of pollutant emission constraint intervals and a traffic flow constraint interval; perform a random uniform distribution of a preset number of solutions according to the list of pollutant emission constraint intervals and the traffic flow constraint interval to generate an initial solution set; traverse the initial solution set, combine the meteorological factor record data and the population density record data, and obtain a number of first air quality index evaluation values ​​through the air quality prediction network; construct a convergence probability evaluation function:

[0062] ;

[0063] in, represents the convergence probability of the ith solution, represents the i-th solution, represents the air quality index evaluation value of the ith solution, Characterize the air quality index threshold, Characterization constant;

[0064] When any one of the convergence probability evaluation values ​​of the plurality of first air quality index evaluation values ​​is equal to 1, the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value are output; otherwise, the plurality of first air quality index evaluation values ​​are expanded and a loop is executed.

[0065] Furthermore, the present application also includes the following steps:

[0066] According to the convergence probability, the first number of head solutions are screened from the initial solution set from large to small, and the second number of tail solutions are screened from the initial solution set from small to large according to the convergence probability; a random one of the second number of tail solutions is used as the expansion starting point, and a random one of the first number of head solutions is used as the expansion end point to search a preset number of times to generate an initial expanded solution; according to the pollutant emission constraint interval list and the traffic flow constraint interval, the initial expanded solution is subjected to over-limit constraints to generate a target expanded solution, and a loop is executed.

[0067] Specifically, when the air quality index prediction value is greater than or equal to the air quality index threshold, the traffic flow characteristic value and the pollutant emission characteristic value are adjusted and optimized to obtain the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value.

[0068] First, configure the pollutant emission constraint interval list and traffic flow constraint interval, where the minimum value of the pollutant emission constraint interval and the traffic flow constraint interval is set by experts in this field based on experience, and the maximum value is the current pollutant emission characteristic value and traffic flow characteristic value. Randomly and uniformly distribute a preset number of solutions in the pollutant emission constraint interval list and the traffic flow constraint interval, where the preset number is set by professional and technical personnel in this field, such as 50, thereby obtaining the pollutant emission and traffic flow in the pollutant emission constraint interval list and the traffic flow constraint interval as the preset number of initial solutions to form an initial solution set.

[0069] Traverse the initial solution set, combine the meteorological factor record data and the population density record data, normalize the pollutant emissions and traffic flow in each initial solution with the meteorological factor record data and the population density record data, form a matrix and input it into the air quality prediction network, and output a number of first air quality index evaluation values. Construct a convergence probability evaluation function:

[0070] ;

[0071] in, represents the convergence probability of the ith solution, represents the i-th solution, represents the air quality index evaluation value of the ith solution, Characterize the air quality index threshold, Characterization constant. The convergence probability of the several first air quality index evaluation values ​​is calculated by the convergence probability evaluation function to obtain several convergence probabilities. When any one of the convergence probability evaluation values ​​of the several first air quality index evaluation values ​​is equal to 1, the pollutant emissions and traffic flow in the initial solution corresponding to the convergence probability are output as the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value. Otherwise, after expanding the several first air quality index evaluation values, a loop is executed. In this way, the optimization of pollutant emissions and traffic flow is achieved to reduce the degree of air pollution and improve air quality.

[0072] Among them, the process of executing the loop for expanding the several first air quality index evaluation values ​​is as follows: according to the order of convergence probability from large to small, the initial solution with the largest probability of an accident is extracted from the initial solution set as the first quantity head solution, and the initial solution with the smallest probability of convergence is extracted from the initial solution set as the second quantity tail solution according to the convergence probability. The first quantity head solution and the second quantity tail solution may contain multiple initial solutions. Taking a random one of the second quantity tail solutions as the expansion starting point and a random one of the first quantity head solutions as the expansion end point to search a preset number of times, that is, using the pollutant emissions and traffic flow in the expansion starting point, and the pollutant emissions and traffic flow in the expansion end point to establish a pollutant emissions interval and a traffic flow interval, and randomly searching and generating new pollutant emissions and traffic flow as the initial expansion solution in the pollutant emissions interval and the traffic flow interval according to the preset number of times, wherein the preset number of times is set by those skilled in the art.

[0073] Further, according to the pollutant emission constraint interval list and the traffic flow constraint interval, the initial expanded solution is subjected to over-limit constraints, that is, to ensure that the pollutant emission and traffic flow in the initial expanded solution are within the pollutant emission constraint interval list and the traffic flow constraint interval. If it exceeds the pollutant emission constraint interval list and the traffic flow constraint interval, it is adjusted to satisfy the pollutant emission constraint interval list and the traffic flow constraint interval, and the adjusted initial expanded solution is used as the target expanded solution. Then, the convergence probability of the target expanded solution is obtained through the convergence probability evaluation function. If the convergence probability of any solution in the target expanded solution is greater than 1, it is output as the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value. Otherwise, the target expanded solution is continued to be expanded until the convergence probability of any solution is greater than 1. In this way, the optimization of pollutant emissions and traffic flow is achieved, the control optimization of air quality is achieved, and air pollution is reduced.

[0074] In summary, the air quality prediction and control method based on urban multidimensional data provided by this application has the following technical effects:

[0075] Through the Internet of Things, the meteorological factor characteristic values ​​of the preset city's predicted time zone are collected; through the pollution disclosure information, the pollutant emission characteristic values ​​of the preset city's predicted time zone are collected; through the traffic flow statistics channel, historical data mining is performed to generate the traffic flow characteristic values ​​of the preset city's predicted time zone; according to the meteorological factor characteristic values, the pollutant emission characteristic values, the traffic flow characteristic values ​​and the population density characteristic values, the air quality prediction network is fitted to generate the air quality index prediction value; when the air quality index prediction value is greater than or equal to the air quality index threshold, the traffic flow characteristic value and the pollutant emission characteristic value are adjusted and optimized to obtain the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value; according to the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value, the quality control is performed. In this way, air quality analysis and prediction are performed by integrating multi-dimensional data, and then the air quality control optimization is performed, so as to improve the accuracy of air quality prediction and achieve the technical effect of precise air quality control.

[0076] Embodiment 2

[0077] Based on the same inventive concept as the method for predicting and controlling air quality based on urban multidimensional data in the aforementioned embodiment, the present application also provides an air quality prediction and control system based on urban multidimensional data, see Attachment 1. Figure 2 , the system comprising:

[0078] The meteorological factor characteristic collection module 11 is used to collect the meteorological factor characteristic values ​​of the time zone to be predicted in the preset city through the Internet of Things;

[0079] The pollutant emission characteristic collection module 12 is used to collect the pollutant emission characteristic values ​​of the preset city in the time zone to be predicted through the pollution discharge public information;

[0080] The traffic flow characteristic mining module 13 is used to mine historical data through the traffic flow statistical channel to generate the traffic flow characteristic value of the time zone to be predicted in the preset city;

[0081] The air quality fitting module 14 is used to generate an air quality index prediction value by fitting the meteorological factor characteristic value, the pollutant emission characteristic value, the traffic flow characteristic value and the population density characteristic value through an air quality prediction network;

[0082] The air quality optimization module 15 is used to adjust and optimize the traffic flow characteristic value and the pollutant emission characteristic value when the air quality index prediction value is greater than or equal to the air quality index threshold value, so as to obtain the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value;

[0083] The air quality control module 16 is used to control the air quality according to the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value.

[0084] Furthermore, the pollutant emission characteristic collection module 12 in the system is also used for:

[0085] According to the pollution discharge public information, obtain a pollutant emission type list and a pollutant emission amount list of the first enterprise public information in the time zone to be predicted in the preset city;

[0086] Performing outlier analysis on the pollutant emission amount list and the pollutant emission type list to obtain the pollutant type of abnormal emission amount;

[0087] The abnormal emission amount pollutant type is sent to the first enterprise for correction, and the pollutant emission characteristic value of the first enterprise is obtained and added to the pollutant emission characteristic value.

[0088] Furthermore, the pollutant emission characteristic collection module 12 in the system is also used for:

[0089] Traversing the pollutant emission amount list for processing through a normalization function to obtain a normalized characteristic value list, wherein the pollutant emission type list and the normalized characteristic value list correspond one to one;

[0090] Conducting pollution disclosure tracing back on the first enterprise using the pollutant emission type list to obtain disclosure tracing back information of several enterprises;

[0091] Through a normalization function, several pollutant retrospective emission lists of several enterprises' publicly disclosed retrospective information are processed respectively to obtain several retrospective normalized characteristic value lists;

[0092] Constructing a normalized characteristic value constraint interval list of the pollutant emission type list according to the plurality of retrospective normalized characteristic value lists;

[0093] According to the normalized eigenvalue constraint interval list, performing an outlier analysis on the normalized eigenvalue list to obtain the type of pollutant with abnormal emission amount;

[0094] Wherein, the normalization function is:

[0095] ;

[0096] in, Characterize the normalized eigenvalues, Characterize the pollutant emissions, The minimum value of pollutant emissions that characterizes the pollutant emission list, The maximum value of a pollutant emission that represents a pollutant emission list.

[0097] Furthermore, the traffic flow feature mining module 13 in the system is also used for:

[0098] Through the traffic flow statistics channel, historical data is traced back to obtain the monthly month-on-month growth rates of several traffic flows in the preset cities;

[0099] Through the traffic flow statistics channel, historical data is traced back to obtain several month-on-month year-on-year growth rates of traffic flows in preset cities;

[0100] Performing a central trend analysis on the several month-on-month growth rates of traffic flows to obtain a characteristic value of the month-on-month growth rate;

[0101] Performing a central trend analysis on the month-on-month year-on-year growth rates of the plurality of traffic flows to obtain a characteristic value of the month-on-month year-on-year growth rate;

[0102] When the deviation between the month-on-month growth rate characteristic value and the month-on-month year-on-year growth rate characteristic value is less than or equal to the growth rate deviation threshold, the traffic flow characteristic value is calculated according to the average of the month-on-month growth rate characteristic value and the month-on-year year-on-year growth rate characteristic value;

[0103] Otherwise, trace back several characteristic values ​​of the month-on-month growth rate of previous years with the month before the month in the time zone to be predicted as the end month, and calculate several month-on-month growth rates to be analyzed from the month before the month in the time zone to be predicted to the month in the time zone to be predicted in the traced year, where the traced year is at least 5 years;

[0104] Deviation calculations are performed on the characteristic values ​​of the month-on-month growth rates of the previous years and the month-on-month growth rates to be analyzed to obtain a number of growth rate anomaly sub-coefficients, and the average of the anomalous growth rate sub-coefficients is calculated and set as the growth rate anomaly coefficient;

[0105] When the growth rate abnormality coefficient is greater than or equal to the growth rate abnormality coefficient threshold, the traffic flow characteristic value is calculated according to the month-on-month year-on-year growth rate characteristic value;

[0106] When the growth rate abnormality coefficient is less than the growth rate abnormality coefficient threshold, the traffic flow characteristic value is calculated according to the month-on-month growth rate characteristic value.

[0107] Further, the air quality fitting module 14 in the system is also used for:

[0108] Collecting a preset urban air quality monitoring data set, wherein the preset urban air quality monitoring data set includes meteorological factor record data, pollutant emission record data, traffic flow record data, population density record data and air quality index record data;

[0109] A normalized matrix is ​​constructed according to the meteorological factor recorded data, the pollutant emission recorded data, the traffic flow recorded data, and the population density recorded data, wherein the meteorological factor recorded data, the pollutant emission recorded data, the traffic flow recorded data, and the population density recorded data in the normalized matrix are distributed at preset element positions;

[0110] The air quality prediction network is configured using the normalized matrix as input data and the air quality index record data as output supervision data.

[0111] Furthermore, the air quality optimization module 15 in the system is also used for:

[0112] Configure the pollutant emission constraint interval list and traffic flow constraint interval;

[0113] A preset number of solutions are randomly and evenly distributed according to the pollutant emission constraint interval list and the traffic flow constraint interval to generate an initial solution set;

[0114] Traversing the initial solution set, combining the meteorological factor record data and the population density record data, and obtaining a plurality of first air quality index evaluation values ​​through the air quality prediction network;

[0115] Construct a convergence probability evaluation function:

[0116] ;

[0117] in, represents the convergence probability of the ith solution, represents the i-th solution, represents the air quality index evaluation value of the ith solution, Characterize the air quality index threshold, Characterization constant;

[0118] When any one of the convergence probability evaluation values ​​of the plurality of first air quality index evaluation values ​​is equal to 1, outputting the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value;

[0119] Otherwise, the plurality of first air quality index evaluation values ​​are expanded and then a loop is executed.

[0120] Furthermore, the air quality optimization module 15 in the system is also used for:

[0121] Screening a first number of head solutions from the initial solution set from large to small according to the convergence probability, and screening a second number of tail solutions from the initial solution set from small to large according to the convergence probability;

[0122] Taking a random one of the second number of tail solutions as an expansion starting point and a random one of the first number of head solutions as an expansion end point, searching a preset number of times to generate an initial expansion solution;

[0123] According to the pollutant emission constraint interval list and the traffic flow constraint interval, the initial extended solution is subjected to over-limit constraints, a target extended solution is generated, and a loop is executed.

[0124] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The air quality prediction and control method based on urban multidimensional data and the specific examples in Example 1 are also applicable to the air quality prediction and control system based on urban multidimensional data in this embodiment. Through the above detailed description of the air quality prediction and control method based on urban multidimensional data, those skilled in the art can clearly know the air quality prediction and control system based on urban multidimensional data in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0125] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0126] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. The air quality prediction and control method based on urban multidimensional data is characterized by: include: Through the Internet of Things, the characteristic values ​​of meteorological factors in the time zone to be predicted in the preset city are collected; Collect pollutant emission characteristic values ​​of the preset city in the predicted time zone through pollution disclosure information; Perform historical data mining through the traffic flow statistics channel to generate traffic flow characteristic values ​​for the time zone to be predicted in the preset city; According to the characteristic values ​​of the meteorological factors, the characteristic values ​​of the pollutant emissions, the characteristic values ​​of the traffic flow and the characteristic values ​​of the population density, fitting is performed through an air quality prediction network to generate an air quality index prediction value; When the air quality index prediction value is greater than or equal to the air quality index threshold, the traffic flow characteristic value and the pollutant emission characteristic value are adjusted and optimized to obtain a recommended pollutant emission characteristic value and a recommended traffic flow characteristic value; Performing control quality regulation according to the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value; When the air quality index prediction value is greater than or equal to the air quality index threshold, the traffic flow characteristic value and the pollutant emission characteristic value are adjusted and optimized to obtain a recommended pollutant emission characteristic value and a recommended traffic flow characteristic value, including: Configure the pollutant emission constraint interval list and traffic flow constraint interval; A preset number of solutions are randomly and evenly distributed according to the pollutant emission constraint interval list and the traffic flow constraint interval to generate an initial solution set; Traversing the initial solution set, combining the meteorological factor record data and the population density record data, and obtaining a plurality of first air quality index evaluation values ​​through the air quality prediction network; Construct a convergence probability evaluation function: ； in, represents the convergence probability of the ith solution, represents the i-th solution, Characterizes the air quality index evaluation value of the i-th solution, Characterize the air quality index threshold, Characterization constant; When any one of the convergence probability evaluation values ​​of the plurality of first air quality index evaluation values ​​is equal to 1, outputting the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value; Otherwise, the plurality of first air quality index evaluation values ​​are expanded and then a loop is executed.

2. The method according to claim 1, characterized in that The pollutant emission characteristic values ​​of the preset city in the predicted time zone are collected through the pollution disclosure information, including: According to the pollution discharge public information, obtain a pollutant emission type list and a pollutant emission amount list of the first enterprise public information in the time zone to be predicted in the preset city; Performing outlier analysis on the pollutant emission amount list and the pollutant emission type list to obtain the pollutant type of abnormal emission amount; The abnormal emission amount pollutant type is sent to the first enterprise for correction, and the pollutant emission characteristic value of the first enterprise is obtained and added to the pollutant emission characteristic value.

3. The method according to claim 2, characterized in that The abnormal value analysis is performed through the pollutant emission amount list and the pollutant emission type list to obtain the abnormal emission amount pollutant type, including: Traversing the pollutant emission amount list for processing through a normalization function to obtain a normalized characteristic value list, wherein the pollutant emission type list and the normalized characteristic value list correspond one to one; Conducting pollution disclosure tracing back on the first enterprise using the pollutant emission type list to obtain disclosure tracing back information of several enterprises; Through a normalization function, several pollutant retrospective emission lists of several enterprises' publicly disclosed retrospective information are processed respectively to obtain several retrospective normalized characteristic value lists; Constructing a normalized characteristic value constraint interval list of the pollutant emission type list according to the plurality of retrospective normalized characteristic value lists; According to the normalized eigenvalue constraint interval list, performing an outlier analysis on the normalized eigenvalue list to obtain the type of pollutant with abnormal emission amount; Wherein, the normalization function is: ； in, Characterize the normalized eigenvalues, Characterize the pollutant emissions, The minimum value of pollutant emissions that characterizes the pollutant emission list, The maximum value of a pollutant emission that represents a pollutant emission list.

4. The method according to claim 1, characterized in that Through the traffic flow statistics channel, historical data mining is performed to generate the traffic flow characteristic value of the time zone to be predicted in the preset city, including: Through the traffic flow statistics channel, historical data is traced back to obtain the monthly month-on-month growth rates of several traffic flows in the preset cities; Through the traffic flow statistics channel, historical data is traced back to obtain several month-on-month year-on-year growth rates of traffic flows in preset cities; Performing a central trend analysis on the several month-on-month growth rates of traffic flows to obtain a characteristic value of the month-on-month growth rate; Performing a central trend analysis on the month-on-month year-on-year growth rates of the plurality of traffic flows to obtain a characteristic value of the month-on-month year-on-year growth rate; When the deviation between the month-on-month growth rate characteristic value and the month-on-month year-on-year growth rate characteristic value is less than or equal to the growth rate deviation threshold, the traffic flow characteristic value is calculated according to the average of the month-on-month growth rate characteristic value and the month-on-year year-on-year growth rate characteristic value; Otherwise, trace back several characteristic values ​​of the month-on-month growth rate of previous years with the month before the month in the time zone to be predicted as the end month, and calculate several month-on-month growth rates to be analyzed from the month before the month in the time zone to be predicted to the month in the time zone to be predicted in the traced year, where the traced year is at least 5 years; Deviation calculations are performed on the characteristic values ​​of the month-on-month growth rates of the previous years and the month-on-month growth rates to be analyzed to obtain a number of growth rate anomaly sub-coefficients, and the average of the anomalous growth rate sub-coefficients is calculated and set as the growth rate anomaly coefficient; When the growth rate abnormality coefficient is greater than or equal to the growth rate abnormality coefficient threshold, the traffic flow characteristic value is calculated according to the month-on-month year-on-year growth rate characteristic value; When the growth rate abnormality coefficient is less than the growth rate abnormality coefficient threshold, the traffic flow characteristic value is calculated according to the month-on-month growth rate characteristic value.

5. The method according to claim 1, characterized in that According to the characteristic values ​​of the meteorological factors, the characteristic values ​​of the pollutant emissions, the characteristic values ​​of the traffic flow and the characteristic values ​​of the population density, an air quality prediction network is used to perform fitting to generate an air quality index prediction value, including: Collecting a preset urban air quality monitoring data set, wherein the preset urban air quality monitoring data set includes meteorological factor record data, pollutant emission record data, traffic flow record data, population density record data and air quality index record data; A normalized matrix is ​​constructed according to the meteorological factor recorded data, the pollutant emission recorded data, the traffic flow recorded data, and the population density recorded data, wherein the meteorological factor recorded data, the pollutant emission recorded data, the traffic flow recorded data, and the population density recorded data in the normalized matrix are distributed at preset element positions; The air quality prediction network is configured using the normalized matrix as input data and the air quality index record data as output supervision data.

6. The method according to claim 1, characterized in that Expanding the plurality of first air quality index evaluation values ​​and executing a loop includes: Screening a first number of head solutions from the initial solution set from large to small according to the convergence probability, and screening a second number of tail solutions from the initial solution set from small to large according to the convergence probability; Taking a random one of the second number of tail solutions as an expansion starting point and a random one of the first number of head solutions as an expansion end point, searching a preset number of times to generate an initial expansion solution; According to the pollutant emission constraint interval list and the traffic flow constraint interval, the initial extended solution is subjected to over-limit constraints, a target extended solution is generated, and a loop is executed.

7. The air quality prediction and control system based on urban multi-dimensional data is characterized by: For implementing the steps of the method according to any one of claims 1 to 6, the system comprises: A meteorological factor characteristic collection module is used to collect the characteristic values ​​of meteorological factors of the time zone to be predicted in a preset city through the Internet of Things; A pollutant emission characteristic collection module, used to collect pollutant emission characteristic values ​​of the preset city in the predicted time zone through pollution disclosure information; A traffic flow feature mining module is used to mine historical data through a traffic flow statistical channel to generate a traffic flow feature value for the time zone to be predicted in the preset city; An air quality fitting module, used to generate an air quality index prediction value by fitting the meteorological factor characteristic value, the pollutant emission characteristic value, the traffic flow characteristic value and the population density characteristic value through an air quality prediction network; An air quality optimization module, used for adjusting and optimizing the traffic flow characteristic value and the pollutant emission characteristic value when the air quality index prediction value is greater than or equal to the air quality index threshold value, to obtain a recommended pollutant emission characteristic value and a recommended traffic flow characteristic value; An air quality control module, used for controlling the air quality according to the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value; The air quality optimization module is also used for: Configure the pollutant emission constraint interval list and traffic flow constraint interval; A preset number of solutions are randomly and evenly distributed according to the pollutant emission constraint interval list and the traffic flow constraint interval to generate an initial solution set; Traversing the initial solution set, combining the meteorological factor record data and the population density record data, and obtaining a plurality of first air quality index evaluation values ​​through the air quality prediction network; Construct a convergence probability evaluation function: ； in, represents the convergence probability of the ith solution, represents the i-th solution, Characterizes the air quality index evaluation value of the i-th solution, Characterize the air quality index threshold, Characterization constant; When any one of the convergence probability evaluation values ​​of the plurality of first air quality index evaluation values ​​is equal to 1, outputting the recommended pollutant emission characteristic value and the recommended traffic flow characteristic value; Otherwise, the plurality of first air quality index evaluation values ​​are expanded and then a loop is executed.

Images