Urban air pollutant variation range evaluation and ranking method and system

Abstract

The application belongs to the technical field of air pollution assessment, and relates to a city air pollutant change range assessment and ranking method and system, the method comprising: 1) obtaining multiple city air pollutant concentrations in different time dimensions and standardizing the air pollutant concentrations; 2) performing multi-dimensional time series analysis on the air pollutant concentrations to obtain multi-dimensional time series analysis results; 3) calculating the comprehensive score of each city according to the air pollutant concentrations and the multi-dimensional time series analysis results; 4) calculating the dynamic assessment score of each city and dynamically ranking the multiple cities according to the dynamic assessment score; and 5) assessing the change range of the pollutant concentration of each city and ranking the multiple cities according to the multi-dimensional time series analysis results and the change range to obtain the risk level of each city. The application is evaluated based on multi-dimensional time series analysis results and dynamic weights, and the evaluation results are comprehensive and accurate.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of air pollution assessment, and relates to a city air pollutant change range assessment and ranking method and system, in particular to a city air pollutant change range assessment and ranking method and system based on multi-dimensional time series analysis and dynamic weight. BACKGROUND

[0002] Air quality is an environmental public good, and its non-exclusivity and non-competitiveness determine that it is a basic public service that the government must provide. Air quality assessment and ranking are the main basis for discovering air quality exceeding problems and atmospheric pollution source emission management, and are also the basis for targeted management of different cities.

[0003] However, existing city air pollutant ranking assessment and methods are mostly based on static data at a single time point or simple time series statistics (such as mean value, maximum value) for assessment and ranking, lacking in-depth analysis of pollutant change trend, fluctuation range and multi-dimensional time characteristics. In addition, the traditional city air pollutant assessment and ranking method does not consider the dynamic weight distribution between different pollutants, resulting in an assessment result that is not comprehensive and accurate.

[0004] Therefore, in view of the defects in the prior art, there is an urgent need for a new city air pollutant change range assessment and ranking method. SUMMARY

[0005] In view of the defects of the prior art, the present application proposes a city air pollutant change range assessment and ranking method and system, which is based on multi-dimensional time series analysis and dynamic weight for assessment, and the assessment result is comprehensive and accurate.

[0006] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0007] A city air pollutant change range assessment and ranking method, characterized in that it comprises the following steps:

[0008] 1) Obtain multiple city air pollutant concentrations at different time dimensions and standardize them to obtain standardized air pollutant concentrations;

[0009] 2) Perform multi-dimensional time series analysis on the air pollutant concentrations to obtain multi-dimensional time series analysis results;

[0010] 3) Calculate the comprehensive score of each city according to the air pollutant concentrations and the multi-dimensional time series analysis results;

[0011] 4) Calculate the dynamic assessment score of each city based on the comprehensive score of each city and dynamically rank the multiple cities according to the dynamic assessment score.

[0012] 5) Evaluate the range of pollutant concentration of each city and evaluate multiple cities according to the results of multi-dimensional time series analysis and the range to obtain the risk level of each city.

[0013] Preferably, the step 1) specifically comprises:

[0014] 11) Obtain the concentration of multiple air pollutants in multiple cities in different time dimensions;

[0015] 12) Fill in the missing data of each air pollutant concentration by using time series interpolation method and identify and eliminate abnormal data in each air pollutant concentration based on 3σ principle or box plot method;

[0016] 13) Standardize each air pollutant concentration to eliminate dimensional difference and obtain the standardized air pollutant concentration z:

[0017]

[0018] Wherein, x is the original value of each air pollutant concentration, μ is the mean value of each air pollutant concentration, and σ is the standard deviation of each air pollutant concentration.

[0019] Preferably, the step 2) specifically comprises:

[0020] 21) Decompose the time series of each air pollutant concentration of each city into three parts: trend, season and residual by using time series decomposition method based on local weighted regression:

[0021] 22) For each air pollutant concentration of each city, calculate its standard deviation σ, coefficient of variation CV and maximum drawdown rate MDD;

[0022] 23) For each air pollutant concentration of each city, calculate its annualized rate of change AROC:

[0023]

[0024] Wherein, X initial and X final are the air pollutant concentrations at the initial time point and the end time point respectively, and n is the time span in years from the initial time point to the end time point.

[0025] Preferably, in the step 22), the calculation of the maximum drawdown rate MDD is specifically:

[0026] 221) For each time point t in the time series of the concentration of each air pollutant of each city, calculate the maximum air pollutant concentration from the starting time point to time point t, denoted as Max t :

[0027] Max t = max(X1,X2,...,X t );

[0028] wherein X t is the air pollutant concentration at time point t;

[0029] 222) Calculate the drawdown D t :

[0030]

[0031] 223) Calculate the maximum drawdown rate MDD:

[0032] MDD = max(D1,D2,...,D t ).

[0033] Preferably, in step 3), the comprehensive score Score of each city is:

[0034] Score = ω1 × z + ω2 × trend + ω3 × season + ω4 × residual + ω5 × σ + ω6 × CV + ω7 × MDD + ω8 × AROC,

[0035] wherein ω1, ω2, ω3, ω4, ω5, ω6, ω7 and ω8 are weighting coefficients.

[0036] Preferably, in step 3), the weighting coefficients ω1, ω2, ω3, ω4, ω5, ω6, ω7 and ω8 are automatically adjusted using a data-driven method.

[0037] Preferably, in step 4), the dynamic evaluation score of each city is calculated using a weighted moving average method.

[0038] Preferably, step 5) specifically comprises:

[0039] 51) Use the variation amplitude coefficient VAC to evaluate the variation amplitude of the pollutant concentration of each city:

[0040]

[0041] wherein Max(X t ) and Min(X t ) are the maximum and minimum values of the air pollutant concentration of the city from the starting time point to time point t, and Mean(Xt ) is an average value of the urban air pollutant concentration from the starting time point to the time point t;

[0042] 52) using a clustering algorithm, based on the standard deviation σ, the coefficient of variation CV and the maximum drawdown rate MDD and the variation amplitude coefficient VAC, a plurality of cities are clustered and analyzed to divide the plurality of cities into high, medium and low risk cities.

[0043] In addition, the present application also provides a city air pollutant variation amplitude evaluation and ranking system, characterized in that it comprises:

[0044] A data acquisition and preprocessing module is configured to acquire a plurality of air pollutant concentrations of a plurality of cities in different time dimensions and perform standardization processing thereon to obtain standardized air pollutant concentrations.

[0045] A multi-dimensional time series analysis module is configured to perform multi-dimensional time series analysis on the air pollutant concentrations to obtain multi-dimensional time series analysis results.

[0046] A comprehensive score calculation module is configured to calculate a comprehensive score of each city according to the air pollutant concentrations and the multi-dimensional time series analysis results.

[0047] A dynamic ranking module is configured to calculate a dynamic evaluation score of each city based on the comprehensive score of each city and to dynamically rank the plurality of cities according to the dynamic evaluation score.

[0048] A variation amplitude analysis and risk evaluation module is configured to evaluate the variation amplitude of the pollutant concentration of each city and to evaluate the plurality of cities according to the multi-dimensional time series analysis results and the variation amplitude to obtain a risk level of each city.

[0049] Finally, the present application also provides a city air pollutant variation amplitude evaluation and ranking device, characterized in that it comprises:

[0050] One or more processors;

[0051] A memory for storing one or more programs;

[0052] When the one or more programs are executed by the one or more processors, the one or more processors implement the city air pollutant variation amplitude evaluation and ranking method as described above.

[0053] Compared with the prior art, the city air pollutant variation amplitude evaluation and ranking method and system of the present application has one or more of the following beneficial technical effects:

[0054] 1、The present application can deeply analyze the change trend, fluctuation amplitude and multi-dimensional time characteristics of pollutants through multi-dimensional time series analysis, so as to deeply analyze the change rule and characteristics of pollutants.

[0055] 2、The present application considers the dynamic weight distribution between different pollutants, resulting in comprehensive and accurate evaluation results. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 is a flow chart of the urban air pollutant change amplitude evaluation and ranking method of the present application.

[0057] Figure 2 is a structural schematic diagram of the urban air pollutant change amplitude evaluation and ranking system of the present application. DETAILED DESCRIPTION

[0058] Before any embodiments of the present application are described in detail, it is to be understood that the present application is not limited in its applications to the details of construction and the arrangements of the components set forth in the following description or illustrated in the following drawings. The present application is capable of other embodiments and of being practiced or being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having" and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. Unless specified or limited otherwise, the terms "mounted," "connected," "supported," and "coupled" and variations thereof are used broadly and encompass both direct and indirect mountings, connections, supports, and couplings. Further, "connected" and "coupled" are not restricted to physical or mechanical connections or couplings.

[0059] Also, in the disclosure of the present application, the terms "longitudinal", "transverse", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate the orientation or positional relationship shown in the drawings, which are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the above terms cannot be understood as limiting the present application; secondly, the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of one element can be one, and in another embodiment, the number of the element can be multiple, the term "one" cannot be understood as limiting the number.

[0060] Figure 1 A flow chart of the urban air pollutant change amplitude evaluation and ranking method of the present application is shown. As shown in Figure 1 the urban air pollutant change amplitude evaluation and ranking method of the present application includes the following steps:

[0061] I. Data acquisition and preprocessing.

[0062] To evaluate and rank the change range of urban air pollution, the evaluation data must be collected first. In the present application, first, the concentration of multiple air pollutants in multiple cities at different time dimensions is obtained and standardized to obtain the standardized air pollutant concentration, which specifically includes:

[0063] 1. Data collection.

[0064] Obtain the monitoring value of multiple air pollutants (such as PM2.5, PM10, SO2, NO2, CO, O3) in multiple cities at different time dimensions (such as daily, weekly, monthly), i.e. pollutant concentration value.

[0065] 2. Data cleaning.

[0066] (1) Handle missing values: use time series interpolation method (such as linear interpolation or spline interpolation) to fill in the missing data of each air pollutant concentration.

[0067] (2) Remove outliers: identify and remove abnormal data of each air pollutant concentration based on 3σ principle or box plot method.

[0068] 3. Data standardization.

[0069] Z-score standardization is performed on each air pollutant concentration to eliminate dimensional differences and obtain the standardized air pollutant concentration z:

[0070]

[0071] Where x is the original value of each air pollutant concentration, μ is the mean of each air pollutant concentration, and σ is the standard deviation of each air pollutant concentration.

[0072] II. Multidimensional time series analysis.

[0073] In the present application, the standardized air pollutant concentration is subjected to multidimensional time series analysis to obtain the multidimensional time series analysis result, which specifically includes:

[0074] 1. Seasonal and trend decomposition.

[0075] The STL (Seasonal-Trend decomposition using Loess) method is used to decompose the time series of air pollutant concentrations in each city into three parts: the trend part, the seasonal part, and the residual part. Among them, the trend part trend reflects the long-term change trend of the concentration of air pollutants in the city over time; the seasonal part season can capture the periodic changes of years, months or seasons; the residual part residual removes the random fluctuations after the seasonality and trend.

[0076] The STL (Seasonal-Trend decomposition using Loess) method, also known as the Loess-based time series decomposition method, is a statistical method for decomposing time series data, which can decompose time series data into trend, seasonality and residual three main parts. This method was first proposed by Cleveland et al. in 1990, and is commonly used to identify and analyze different components in time series, helping to understand the basic structure and rules of data. Since the STL method is used for statistics and analysis, which is prior art, in order to simplify, the present invention does not describe it in detail.

[0077] 2. Volatility analysis.

[0078] Volatility is an important indicator to measure the fluctuation amplitude of air pollutant concentration. In this invention, for the concentration of pollutants in each city, the following indicators are used to quantify volatility:

[0079] (1) Standard deviation σ: Calculate the standard deviation σ of each pollutant concentration in each city, which represents the fluctuation amplitude of the pollutant concentration.

[0080] (2) Coefficient of Variation (CV): Calculate the coefficient of variation CV of each pollutant concentration in each city, which is the ratio of the standard deviation to the mean:

[0081]

[0082] The larger the value of the coefficient of variation CV, the stronger the volatility.

[0083] (3) Maximum Drawdown (MDD): Calculate the maximum drawdown amplitude of each pollutant concentration in each city to identify the risk of large decline in pollutant concentration. The specific calculation process is as follows:

[0084] First, calculate the historical maximum value Max of each time point t tFor each time point t in the time series, calculate the maximum pollutant concentration from the starting time point to time point t, denoted as Max t , which represents the highest pollution concentration up to the current time point t:

[0085] Max t = max(X1, X2,..., X t ).

[0086] Where X1 is the pollutant concentration at time point 1 (i.e., the starting time point), X2 is the pollutant concentration at time point 2,..., X t is the pollutant concentration at time t, and max(X1, X2,..., X t ) represents taking the maximum value.

[0087] Next, calculate the drawdown D t : The drawdown is the difference between the current (time point t) pollutant concentration and the historical maximum pollutant concentration, representing the decline in the current pollutant concentration from the historical maximum pollutant concentration:

[0088]

[0089] Finally, calculate the maximum drawdown rate MDD: The maximum drawdown rate MDD is the maximum value among the drawdowns at any one time in history, which represents the maximum decline in pollutant concentration, that is, the maximum proportional decline from a certain peak to the subsequent minimum value:

[0090] MDD = max(D1, D2,..., D t ).

[0091] D1 is the drawdown at time point 1, D2 is the drawdown at time point 2,..., D t is the drawdown at time point t.

[0092] 3. Dynamic trend evaluation.

[0093] In this invention, the Annualized Rate of Change (AROC) is used to calculate the long-term change trend of each pollutant concentration in each city, and the formula is as follows:

[0094]

[0095] Where X initial and X final are the pollutant concentrations at the initial and ending time points, respectively, and n is the time span (years) from the initial time point to the ending time point.

[0096] III. Comprehensive score calculation.

[0097] According to the air pollutant concentration and the multi-dimensional time series analysis result, the comprehensive score of each city is calculated.

[0098] In the present application, according to the air pollutant concentration after standardization processing and the multi-dimensional time series analysis result, a comprehensive score algorithm is designed, which considers the performance of each city in different pollutant concentration and time volatility, and the calculation of the comprehensive score Scroe is as follows:

[0099] Scroe = ω1 x z + ω2 x trend + ω3 x season + ω4 x residual + ω5 x σ + ω6 x CV + ω7 x MDD + ω8 x AROC.

[0100] Wherein, ω1, ω2, ω3, ω4, ω5, ω6, ω7 and ω8 are weighting coefficients.

[0101] Therefore, when calculating the comprehensive score, the present application comprehensively considers the pollutant concentration, the pollutant change trend, the pollutant fluctuation amplitude and the pollutant dynamic trend, so that it is more comprehensive.

[0102] Meanwhile, in the present application, a data-driven method is used to automatically adjust the weight of each index, i.e. the weighting coefficients ω1, ω2, ω3, ω4, ω5, ω6, ω7 and ω8, so that the evaluation is more comprehensive and accurate. Specifically, the principal component analysis (PCA) method, the random forest method or the Bayesian method can be used to automatically assign the weight according to the explanatory ability of each index to the city air quality.

[0103] Taking the principal component analysis method as an example, the principal component analysis (PCA) method is a dimension reduction technique, which finds the most important components in the data by calculating the covariance matrix of the original index, and sorts them according to the variance contribution. Each component represents a certain feature of the original data, and the most informative components can be effectively extracted by using PCA, and then the index is weighted. The specific steps are as follows:

[0104] PCA analysis is performed on the principal components of all cities, i.e. the pollutant concentration z and the related indexes (such as the trend part trend, the seasonal part season, the residual part residual, the standard deviation σ, the coefficient of variation CV, the maximum drawdown rate MDD and the annualized change rate AROC).

[0105] The variance explanation ratio of each principal component is calculated, and the greater the variance, the stronger the explanatory ability of the principal component to the data variation.

[0106] The variance contribution (i.e. the explanation variance ratio) of each principal component is taken as the weight. That is, the principal component that can explain the data variation has a larger weight.

[0107] In this way, the score of each city on each principal component (pollutant concentration and related indicators) can be obtained, and the comprehensive pollution index of each city can be calculated according to the scores.

[0108] IV. Dynamic ranking.

[0109] The dynamic evaluation score of each city is calculated based on the comprehensive score of each city, and the multiple cities are dynamically ranked according to the dynamic evaluation score.

[0110] In the present application, the Weighted Moving Average (WMA) method is used to dynamically rank the cities. When performing the weighted moving average, the length of the time window is set to n, that is, there are n time points in the time window. In each time window, the dynamic evaluation score WMA of the city is calculated t and sorted:

[0111]

[0112] where ω i is the weighted coefficient, representing the importance of the data point, and a linear decreasing weight is used, that is, the weight of the newer data point (closer to the current time t) is higher, and ω i The sum is 1. n represents the length of the time window, that is, the data of how many recent time points are considered to calculate the weighted average.

[0113] V. Change amplitude analysis and risk assessment

[0114] The change amplitude of the pollutant concentration of each city is evaluated, and the multiple cities are evaluated according to the results of multi-dimensional time series analysis and the change amplitude, to obtain the risk level of each city, which specifically includes:

[0115] 1. Change amplitude analysis.

[0116] In the present application, the change amplitude coefficient VAC is used to evaluate the change amplitude of the pollutant concentration of each city, that is, the fluctuation degree:

[0117]

[0118] where Max(X t ) and Min(X t ) are the maximum and minimum values of the air pollutant concentration of the city from the starting time point to the time point t, and Mean(X t ) is the average value of the air pollutant concentration of the city from the starting time point to the time point t.

[0119] 2. Risk assessment.

[0120] In the present application, a clustering algorithm, such as a clustering algorithm based on K-means or DBSCAN, is used to perform clustering analysis on multiple cities based on volatility analysis results, i.e., standard deviation σ, coefficient of variation CV, and maximum drawdown rate MDD, and variation amplitude coefficient VAC, to classify the multiple cities into high, medium, and low risk cities. Subsequently, cities classified as high risk can be given special attention.

[0121] Since clustering analysis using a clustering algorithm based on K-means or DBSCAN is a prior art, it is not described in detail here for simplicity.

[0122] Figure 2 The composition of the city air pollutant variation amplitude evaluation and ranking system of the present application is shown. As shown in Figure 2 The city air pollutant variation amplitude evaluation and ranking system of the present application comprises:

[0123] 1. Data acquisition and preprocessing module.

[0124] The data acquisition and preprocessing module is used to acquire multiple air pollutant concentrations of multiple cities at different time dimensions and to perform standardization processing to obtain standardized air pollutant concentrations z.

[0125] 2. Multidimensional time series analysis module.

[0126] The multidimensional time series analysis module is used to perform multidimensional time series analysis on the air pollutant concentration z to obtain multidimensional time series analysis results.

[0127] 3. Comprehensive score calculation module.

[0128] The comprehensive score calculation module is used to calculate the comprehensive score of each city according to the air pollutant concentration z and the multidimensional time series analysis results.

[0129] 4. Dynamic ranking module.

[0130] The dynamic ranking module is used to calculate the dynamic evaluation score of each city based on the comprehensive score of each city and to dynamically rank the multiple cities according to the dynamic evaluation score.

[0131] 5. Variation amplitude analysis and risk assessment module.

[0132] The variation amplitude analysis and risk assessment module is used to evaluate the variation amplitude of the pollutant concentration z of each city and to evaluate the multiple cities according to the multidimensional time series analysis results and the variation amplitude to obtain the risk level of each city.

[0133] Finally, the present application also provides a city air pollutant variation range evaluation and ranking device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the city air pollutant variation range evaluation and ranking method as described above.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not limited to the scope of protection of the present application. Those skilled in the art can modify or equivalently replace the technical solutions of the present application according to the idea of the present application without departing from the essence and scope of the technical solutions of the present application.

Claims

1. A method for assessing and ranking the magnitude of changes in urban air pollutants, characterized in that, Includes the following steps: 1) Obtain the concentrations of various air pollutants in multiple cities at different time dimensions and standardize them to obtain the standardized air pollutant concentrations; 2) Perform multidimensional time series analysis on air pollutant concentrations to obtain multidimensional time series analysis results; Specifically, this includes: 21) using a time series decomposition method based on local weighted regression to decompose the time series of each air pollutant concentration in each city into three parts: trend, seasonal, and residual; 22) calculating the standard deviation of each air pollutant concentration in each city. 23) Calculate the annualized rate of change (AROC) for the concentration of each air pollutant in each city: (The values ​​are: coefficient of variation (CV) and maximum drawdown (MDD); ,in, , where are the air pollutant concentrations at the initial and final time points, respectively, and n is the time span in years between the initial and final time points; 3) Calculate the comprehensive score for each city based on air pollutant concentrations and multidimensional time series analysis results; whereby the comprehensive score for each city... for: ， In the formula, , , , , , , and Here, z is the weighting coefficient, and z is the standardized concentration of air pollutants. 4) Calculate the dynamic evaluation score for each city based on its comprehensive score, and dynamically rank multiple cities according to their dynamic evaluation scores; 5) Assess the magnitude of pollutant concentration changes in each city and evaluate multiple cities based on multidimensional time series analysis results and magnitude of changes to obtain the risk level of each city. Specifically, this includes: 51) Utilizing the magnitude of change coefficient. To assess the magnitude of changes in pollutant concentrations in each city: In the formula, and These represent the maximum and minimum air pollutant concentrations in the city from the start time to time t, respectively. The average concentration of air pollutants in the city from the start time to time t; 52) Using a clustering algorithm, based on the standard deviation Coefficient of variation (CV), maximum drawdown (MDD), and magnitude of change coefficient Cluster analysis was performed on multiple cities to classify them into high-, medium-, and low-risk cities.

2. The method for assessing and ranking the variation range of urban air pollutants according to claim 1, characterized in that, Step 1) specifically includes: 11) Obtain the concentrations of various air pollutants in multiple cities at different time points; 12) Use time series interpolation to fill in the missing data for each air pollutant concentration and identify and remove outlier data for each air pollutant concentration based on the 3σ principle or box plot method; 13) Standardize the concentration of each air pollutant to eliminate dimensional differences and obtain the standardized air pollutant concentration z: ， Where x is the original value of the concentration of each air pollutant, μ is the mean of the concentration of each air pollutant, and σ is the standard deviation of the concentration of each air pollutant.

3. The method for assessing and ranking the variation range of urban air pollutants according to claim 1, characterized in that, In step 22), the calculation of the maximum drawdown rate (MDD) is specifically as follows: 221) For each time point t in the time series of each air pollutant concentration for each city, calculate the maximum air pollutant concentration from the start time point to time point t, denoted as . : ， in, It is the concentration of air pollutants at time point t; 222) Calculate the drawdown : ； 223) Calculate the maximum drawdown (MDD): 。 4. The method for assessing and ranking the variation range of urban air pollutants according to claim 1, characterized in that, In step 3), the weighting coefficients are automatically adjusted using a data-driven method. , , , , , , and .

5. The method for assessing and ranking the magnitude of changes in urban air pollutants according to any one of claims 1-4, characterized in that, In step 4), the weighted moving average method is used to calculate the dynamic evaluation score for each city.

6. A system for assessing and ranking the magnitude of changes in urban air pollutants, characterized in that, include: The data acquisition and preprocessing module is used to acquire the concentrations of various air pollutants in multiple cities at different time dimensions and to standardize them to obtain the standardized air pollutant concentrations. The multidimensional time series analysis module is used to perform multidimensional time series analysis on air pollutant concentrations to obtain multidimensional time series analysis results. Specifically, this includes: using a time series decomposition method based on local weighted regression to decompose the time series of each air pollutant concentration in each city into three parts: trend, seasonality, and residual; and calculating the standard deviation for each air pollutant concentration in each city. The coefficient of variation (CV) and maximum drawdown (MDD) were calculated; for each air pollutant concentration in each city, the annualized rate of change (AROC) was calculated separately. ,in, , where are the air pollutant concentrations at the initial and final time points, respectively, and n is the time span in years between the initial and final time points; The comprehensive score calculation module is used to calculate a comprehensive score for each city based on air pollutant concentrations and multidimensional time series analysis results; whereby the comprehensive score for each city... for: ， In the formula, , , , , , , and Here, z is the weighting coefficient, and z is the standardized concentration of air pollutants. The dynamic ranking module is used to calculate the dynamic evaluation score of each city based on its comprehensive score and to dynamically rank multiple cities based on their dynamic evaluation scores. The variation amplitude analysis and risk assessment module is used to assess the variation amplitude of pollutant concentrations in each city and evaluate multiple cities based on multidimensional time series analysis results and variation amplitudes to obtain the risk level of each city. Specifically, assessing the variation amplitude of pollutant concentrations in each city and evaluating multiple cities based on multidimensional time series analysis results and variation amplitudes to obtain the risk level of each city includes: using the variation amplitude coefficient... To assess the magnitude of changes in pollutant concentrations in each city: In the formula, and These represent the maximum and minimum air pollutant concentrations in the city from the start time to time t, respectively. The average concentration of air pollutants in the city from the start time to time t; using a clustering algorithm, based on the standard deviation... Coefficient of variation (CV), maximum drawdown (MDD), and magnitude of change coefficient Cluster analysis was performed on multiple cities to classify them into high-, medium-, and low-risk cities.

7. A device for assessing and ranking the variation range of urban air pollutants, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method for assessing and ranking changes in urban air pollutants as described in any one of claims 1-5.

Images