Air pollution control program evaluation method, device, equipment and readable storage medium

Abstract

The present invention provides an air pollution control scheme evaluation method, device, equipment and readable storage medium, which relates to the field of air pollution control technology, including obtaining first information, second information, third information, fourth information and fifth information; performing spatial clustering processing according to the fourth information and the fifth information to obtain a spatial distribution map of pollution concentration; performing correlation analysis processing according to the second information and the fifth information to obtain correlation results; performing system dynamics simulation processing according to the third information to obtain simulation results; constructing a control scheme effect prediction model based on the simulation results, the spatial distribution map and the correlation results; using the first information as the input value of the control scheme effect prediction model to perform prediction processing to obtain a comprehensive evaluation result. The present invention can comprehensively and systematically analyze and evaluate the feasibility and implementation effect of the control scheme by comprehensively considering the influence of multiple factors such as the distribution of air pollution sources, topography, meteorological conditions, etc. on the control effect.

Description

Technical Field

[0001] The present invention relates to the field of air pollution control technology, and in particular to an air pollution control scheme evaluation method, device, equipment and readable storage medium. Background Art

[0002] The current evaluation of air pollution control programs often has some limitations. Existing evaluation methods mainly rely on empirical analysis and simple models, which can only provide partial information and make it difficult to fully evaluate the control effect.

[0003] Based on the above-mentioned shortcomings of the prior art, there is an urgent need for an air pollution control program evaluation method, device, equipment and readable storage medium. Summary of the invention

[0004] The purpose of the present invention is to provide an air pollution control program evaluation method, device, equipment and readable storage medium to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] In the first aspect, the present application provides an air pollution control program evaluation method, comprising:

[0006] Acquire first information, second information, third information, fourth information and fifth information, wherein the first information includes the scheme to be evaluated, the second information includes real-time air pollution monitoring data of the target area, the third information includes historical governance measures and corresponding effects, the fourth information is map data of the target area, and the fifth information includes location information of at least one air pollutant emitting enterprise and corresponding air pollutant emission data;

[0007] Performing spatial clustering processing according to the fourth information and the fifth information, and calculating the pollutant concentrations of unmonitored points in the target area to obtain a spatial distribution map of the pollution concentrations;

[0008] Performing correlation analysis based on the second information and the fifth information, and obtaining a correlation result by establishing a mathematical model between the emission amount and the pollution index;

[0009] Performing system dynamics simulation processing according to the third information, and obtaining simulation results by simulating the impact of historical governance solutions on pollution levels;

[0010] A governance scheme effect prediction model is constructed based on the simulation results, the spatial distribution map and the correlation results;

[0011] The first information is used as the input value of the governance solution effect prediction model for prediction processing, and the scheme to be evaluated is simulated and evaluated in combination with Monte Carlo simulation to obtain a comprehensive evaluation result.

[0012] In a second aspect, the present application also provides an air pollution control scheme evaluation device, including:

[0013] an acquisition module, for acquiring first information, second information, third information, fourth information and fifth information, wherein the first information includes the scheme to be evaluated, the second information includes real-time air pollution monitoring data of the target area, the third information includes historical governance measures and corresponding effects, the fourth information is map data of the target area, and the fifth information includes location information of at least one air pollutant emitting enterprise and corresponding air pollutant emission data;

[0014] A clustering device, used for performing spatial clustering processing according to the fourth information and the fifth information, and calculating the pollutant concentration of unmonitored points in the target area to obtain a spatial distribution map of the pollution concentration;

[0015] An analysis device, used for performing correlation analysis processing according to the second information and the fifth information, and obtaining a correlation result by establishing a mathematical model between the emission amount and the pollution index;

[0016] A simulation device, used for performing system dynamics simulation processing according to the third information, and obtaining simulation results by simulating the impact of historical governance solutions on pollution levels;

[0017] A construction device is used to construct a governance scheme effect prediction model based on the simulation results, the spatial distribution map and the correlation results;

[0018] An evaluation device is used to use the first information as an input value of the governance solution effect prediction model for prediction processing, and simulate and evaluate the solution to be evaluated in combination with Monte Carlo simulation to obtain a comprehensive evaluation result.

[0019] In a third aspect, the present application also provides an air pollution control scheme evaluation device, comprising:

[0020] Memory for storing computer programs;

[0021] A processor is used to implement the steps of the air pollution control program evaluation method when executing the computer program.

[0022] In a fourth aspect, the present application also provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned air pollution control program evaluation method are implemented.

[0023] The beneficial effects of the present invention are:

[0024] The present invention comprehensively considers the influence of various factors on the control effect, such as the distribution of air pollution sources, topography, meteorological conditions, etc., and can comprehensively and systematically analyze and evaluate the feasibility and implementation effect of the control plan.

[0025] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 It is a schematic diagram of the process of the air pollution control scheme evaluation method described in an embodiment of the present invention;

[0028] Figure 2 It is a schematic diagram of the structure of the air pollution control scheme evaluation device described in an embodiment of the present invention;

[0029] Figure 3 Flow chart of the density clustering algorithm described in the embodiment of the present invention.

[0030] Markings in the figure: 800, air pollution control plan evaluation equipment; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, access component. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0032] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0033] Embodiment 1:

[0034] This embodiment provides an air pollution control program evaluation method.

[0035] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, step S400, step S500 and step S600.

[0036] Step S100, obtaining first information, second information, third information, fourth information and fifth information, wherein the first information includes the scheme to be evaluated, the second information includes real-time air pollution monitoring data of the target area, the third information includes historical governance measures and corresponding effects, the fourth information is map data of the target area, and the fifth information includes location information of at least one air pollutant emitting enterprise and corresponding air pollutant emission data;

[0037] It should be noted that the scheme to be evaluated, that is, the specific governance scheme whose effect needs to be evaluated, includes the specific measures, goals and expected effects of the governance scheme, involving various governance means, such as emission reduction measures, application of pollutant control technology or urban planning adjustment. Real-time air pollution monitoring data reflects the concentration level of various pollutants in the target area and the actual air quality. The monitoring data comes from various monitoring sites, including fixed monitoring stations and mobile monitoring equipment, which can cover data in different geographical locations and time periods. Historical governance measures and corresponding effects include measures in emission reduction policies, industrial upgrading, environmental protection investment and other aspects. By analyzing the effects of historical governance measures, a reference can be provided for the evaluation of the current governance scheme. The map data of the target area includes information such as the geographical location, topography, traffic roads, and distribution of pollution sources of the target area. The fifth information includes the location information of at least one air pollutant emitting enterprise and the corresponding air pollutant emission data, which can help understand the distribution and emission level of pollution sources and provide basic data for evaluating the effect of the governance scheme.

[0038] Step S200, performing spatial clustering processing according to the fourth information and the fifth information, and calculating the pollutant concentration of the unmonitored points in the target area to obtain a spatial distribution map of the pollution concentration;

[0039] It can be understood that the spatial distribution map of pollution concentration can intuitively display the pollution level and distribution of pollution sources in each area within the target area, providing a reference for subsequent governance plan evaluation and effect prediction.

[0040] Step S300: performing correlation analysis processing according to the second information and the fifth information, and obtaining correlation results by establishing a mathematical model between the emission amount and the pollution index;

[0041] Specifically, this step uses the emission of pollution sources as the independent variable and the pollution indicators (such as PM2.5, SO2, NO2, etc.) of the monitoring points as the dependent variable, and establishes the functional relationship between the emission and the pollution indicators by fitting these data. Such a mathematical model can help understand the impact of emissions on pollution indicators and predict the pollution level under different emissions.

[0042] Step S400, performing system dynamics simulation processing according to the third information, and obtaining simulation results by simulating the impact of historical governance solutions on pollution levels;

[0043] It can be understood that the system dynamics model is a mathematical model that describes the behavior of a dynamic system. It can take into account the interactions and feedback mechanisms between different factors, and then describe the impact of governance measures on pollution levels.

[0044] Step S500: constructing a governance scheme effect prediction model based on the simulation results, the spatial distribution map and the correlation results;

[0045] Specifically, the simulation results, spatial distribution maps and correlation results are first comprehensively analyzed to extract key features. These features include the changing trend of pollutant concentrations, the spatial distribution characteristics of pollutants, the correlation between emissions and pollution indicators, etc. These features are then used to build a governance scheme effect prediction model. This model is based on statistical methods, machine learning algorithms or other mathematical modeling techniques. In the process of model construction, factors such as the model's predictive ability, interpretability, and robustness need to be considered, and parameter tuning and model verification should be performed according to actual conditions. The final governance scheme effect prediction model can be used to predict and verify the scheme based on relevant information of the scheme to be evaluated, such as governance methods, expected effects, etc.

[0046] Step S600: Use the first information as the input value of the governance solution effect prediction model for prediction processing, and simulate and evaluate the solution to be evaluated in combination with Monte Carlo simulation to obtain a comprehensive evaluation result.

[0047] It is understandable that the governance methods and expected effects in the first information are first input into the constructed governance scheme effect prediction model, and the model is used to predict the effect of the scheme to be evaluated, and the possible impact of the scheme to be evaluated is preliminarily evaluated. Then the Monte Carlo simulation method is introduced to randomly sample factors such as governance effects and environmental conditions, simulate system behaviors under different circumstances, and obtain the effect distribution of the scheme to be evaluated under different circumstances by statistically analyzing a large number of simulation results. These effects include the average effect of the governance method, the range of variation of the expected effect, etc. This information can provide decision makers with a more comprehensive reference and better evaluate the risks and feasibility of the scheme to be evaluated.

[0048] It should be noted that step S200 includes step S210, step S220, step S230 and step S240.

[0049] Step S210, performing distance weighted calculation according to the fourth information and the fifth information, and obtaining a weighted distance set by combining the influence of the pollutant types and concentration data of each emission enterprise on the distance;

[0050] It should be noted that this step first considers the location information of the emission enterprise, that is, the location data of the atmospheric pollutant emission enterprise provided in the fifth information. For each monitoring missing location, calculate its distance from each emission enterprise, and take into account the impact of distance on pollutant transmission. In this process, the closer the distance is, the greater the impact of the emission enterprise on the monitoring missing location. Secondly, consider the type of pollutants and concentration data. Different types of pollutants have different transmission characteristics and influence ranges. Therefore, according to the type of pollutants and concentration data of each emission enterprise provided in the fifth information, the distance weighted calculation is adjusted to more accurately reflect the degree of influence of different types of pollutants on the monitoring missing location. Finally, the weighted distance set of each monitoring missing location and the surrounding emission enterprises is obtained through weighted calculation. In this set, each distance is assigned a weight, which reflects the degree of influence of the type and concentration of pollutants of the emission enterprise on the monitoring missing location.

[0051] Step S220, performing density clustering processing according to the weighted distance set, and obtaining a spatial clustering result by setting the distance threshold to a dynamic value related to a preset type of pollutant during the clustering process;

[0052] It can be understood that this step aims to divide the monitoring missing locations into cluster areas with similar pollution characteristics based on the dynamic change characteristics of different pollutant types and concentrations. In the clustering process, a dynamic distance threshold is used, and its value is related to the preset type of pollutant. This means that the clustering characteristics of different types of pollutants will be adjusted according to their dynamic changes to more accurately reflect the spatial distribution characteristics of different types of pollutants. Secondly, through density clustering processing, the monitoring missing locations are divided into multiple cluster areas with similar pollution characteristics. These areas are relatively concentrated in space and have similar pollutant concentration distributions. Figure 3 As shown in the figure, firstly, the monitoring missing positions and the positions of the known monitoring points are drawn on a two-dimensional plane. Points A, B, C, D and E in the figure represent the monitoring missing positions, and the remaining points are known monitoring points. The algorithm diffuses outward with all the monitoring missing points as the target points, and its neighborhood is defined as the boundary point. Subsequently, a circular area with a distance threshold as the radius is drawn with each monitoring missing position as the center. The range of these circular areas contains the points adjacent to the monitoring missing positions. In the figure, points in the same density clustering area are marked with different symbols or colors. These points are points within the distance threshold and adjacent to the monitoring missing positions. The density clustering area is usually continuous, that is, adjacent points belong to the same area. Finally, adjust the distance threshold as needed and redraw the density clustering area until a satisfactory clustering result is obtained. Such a process can intuitively show the spatial characteristics of the pollutant concentration distribution around the monitoring missing position.

[0053] In this way, the pollutant transport patterns and spatial distribution characteristics around the monitoring missing locations can be better captured, providing more accurate input data for subsequent spatial interpolation processing.

[0054] Step S230, performing spatial interpolation processing according to the spatial clustering result and the atmospheric pollutant concentration data to obtain an interpolation result;

[0055] It can be understood that this step can obtain pollutant concentration interpolation results of the monitoring missing positions through spatial interpolation processing, and these results fill in the pollutant concentration data of the monitoring missing positions.

[0056] Step S240: Perform pollution transmission simulation processing according to the interpolation result, and obtain a spatial distribution map of pollution concentration by using a preset Gaussian model to simulate the transmission and diffusion process of pollutants in the atmosphere.

[0057] Specifically, first, an initial pollutant concentration scenario is constructed based on the pollutant concentration data in the interpolation results. This scenario will reflect the initial pollutant concentration at each location in the monitoring area. Next, the preset Gaussian model will be used to simulate the transmission and diffusion process of pollutants in the atmosphere. The Gaussian model is a classic pollution transmission model that describes the propagation law of pollutants in the atmosphere based on the Gaussian function. Through this model, the transmission path, diffusion range and concentration distribution of pollutants in space can be predicted. In the simulation process, the influence of various factors on the transmission and diffusion of pollutants is taken into account, including wind speed, wind direction, topography, etc. These factors will be incorporated into the model to more accurately simulate the behavior of pollutants in the atmosphere. Finally, the spatial distribution map of pollutant concentration is obtained through pollution transmission simulation. This distribution map will show the pollutant concentration at different locations in the monitoring area, providing an important reference for evaluating the effectiveness of the governance plan.

[0058] It should be noted that step S230 includes step S231, step S232, step S233 and step S234.

[0059] Step S231: perform kernel density estimation processing based on the spatial clustering results and the atmospheric pollutant concentration data, and generate a heat map of the kernel density estimation on the map to obtain a heat distribution map. The areas in the heat distribution map that are not covered by the heat map are monitoring missing locations;

[0060] It can be understood that kernel density estimation is a non-parametric statistical method used to estimate the probability density function of a random variable. In this step, the probability density of the pollutant concentration around the monitoring point is calculated using the kernel density estimation method and represented as a heat map. When generating a heat distribution map, a corresponding heat map is drawn for each location in the monitoring area based on the results of the kernel density estimation. The color depth of the heat map will reflect the pollutant concentration level at that location, thus forming an intuitive pollutant distribution map.

[0061] Step S232: performing grid division processing according to the thermal distribution map, dividing the map into at least two levels of grids, and dynamically adjusting the size of each grid according to the distribution of pollution sources and terrain features on the map to obtain a grid map;

[0062] It should be noted that by using at least two levels of grid division, pollutant concentrations can be observed and analyzed at different resolutions. For example, the overall distribution of pollutants can be observed in high-level grids, while the pollution situation in local areas can be more carefully depicted in low-level grids. This multi-level grid division enables a comprehensive understanding and analysis of pollutant concentrations at different spatial scales. The size of each grid is dynamically adjusted according to the distribution of pollution sources and terrain characteristics on the map, which means that denser grid division is used around pollution sources or in areas with complex terrain to better reflect local pollution; while sparse grid division is used in areas far away from pollution sources or with flat terrain to save computing resources and improve computing efficiency. By combining heat maps with multi-level grids, pollutant concentrations in monitoring missing locations can be estimated more accurately. Heat maps can provide the overall distribution trend of pollutant concentrations, while multi-level grids can further refine this trend, allowing for a deeper understanding of the spatial variation of pollutant concentrations.

[0063] Step S233, interpolate the monitoring missing position in the grid map, calculate the distance between the monitoring missing position and the surrounding known monitoring points and the elevation information of the monitoring missing position by using the inverse distance weighted method to obtain a weight set, and obtain the interpolation result based on the weight set.

[0064] Specifically, the distance between the missing monitoring location and the surrounding known monitoring points is first calculated using the inverse distance weighting method. Distance weighting is a spatial interpolation method. Its basic idea is that the closer the monitoring point is, the greater the impact on the estimated location. However, since the transmission process of atmospheric pollutants is affected by factors such as terrain, considering only the distance may ignore the impact of terrain elevation on pollutant transmission. Therefore, the terrain elevation information of the missing monitoring location should also be considered to estimate the weight more accurately. Secondly, the weight set is calculated based on the calculated distance and terrain elevation information. The weight set here reflects the degree of influence of the monitoring point on the missing monitoring location, including the influence of both distance and terrain elevation. The closer the distance and the smaller the difference in terrain elevation, the higher the weight will be, while the farther the distance and the larger the difference in terrain elevation, the lower the weight will be. Such weight calculation can more accurately reflect the spatial relationship between monitoring points, thereby improving the accuracy of the interpolation results. Finally, the weight set is used to weighted average the pollutant concentrations of the surrounding known monitoring points, thereby obtaining an estimated value of the pollutant concentration at the missing monitoring location.

[0065] The weight calculation formula is:

[0066]

[0067] Where i is the serial number of the monitoring point; W iis the weight of the i-th monitoring point; Q i is the quality score of the i-th monitoring point; d i is the distance between the unknown point and the i-th monitoring point; h i is the elevation difference between the unknown point and the i monitoring points; p is the power parameter that controls the distance weight attenuation; q is the power parameter that adjusts the elevation effect; ∈ i is the adjustment factor.

[0068] The interpolation formula is:

[0069]

[0070] Where Z(x,y) is the predicted pollutant concentration at point (x,y); Z i is the measured pollutant concentration at the i-th monitoring point; n is the total number of monitoring points; i is the serial number of the monitoring point; d i is the distance between the unknown point and the i-th monitoring point; h i is the elevation difference between the unknown point and the i monitoring points; p is the power parameter that controls the distance weight attenuation; q is the power parameter that adjusts the elevation effect.

[0071] It should be noted that step S300 includes step S310, step S320, step S330 and step S340.

[0072] Step S310: performing data matching and format conversion processing according to the second information and the fifth information to obtain matching data;

[0073] It can be understood that this step aims to integrate the second information (real-time air pollution monitoring data of the target area) and the fifth information (location information of at least one air pollutant emitting enterprise and corresponding air pollutant emission data) to obtain a matching data set.

[0074] Step S320, performing correlation analysis processing according to the matching data, and obtaining the correlation analysis result by using the Spearman correlation coefficient to calculate the correlation coefficient between the emission of the emission enterprise and the pollution index of the monitoring point;

[0075] Specifically, the emission data of the emission enterprises and the pollution index data of the monitoring points are first sorted by size and assigned ranks to them respectively. Next, the difference between each rank is calculated and the square of the rank difference is calculated. Then, the Spearman correlation coefficient is calculated, and its calculation formula is as follows:

[0076]

[0077] Among them, r srepresents the Spearman correlation coefficient, d represents the rank difference between emissions and pollution indicators, and n represents the number of samples. By calculating the Spearman correlation coefficient, a value between -1 and 1 can be obtained, which reflects the strength and direction of the correlation between emissions and pollution indicators. If the correlation coefficient is close to 1, it means that the two are positively correlated; if the correlation coefficient is close to -1, it means that the two are negatively correlated; if the correlation coefficient is close to 0, it means that there is no obvious linear relationship between the two.

[0078] Step S330, performing fitting processing according to the correlation analysis results, and obtaining a correlation model by establishing a functional relationship between the emission amount and the pollution index using a polynomial regression algorithm;

[0079] It can be understood that polynomial regression is a regression analysis method that establishes the relationship between independent variables and dependent variables by fitting the polynomial function of the independent variable to the dependent variable. In this step, the pollution index is used as the dependent variable and the emission as the independent variable, and then the polynomial regression algorithm is used to fit the relationship between the two. The fitted model can describe the functional relationship between the emission and the pollution index, thus providing a basis for evaluating the air pollution control plan.

[0080] Step S340: Perform model verification processing according to the association model and the preset verification set to obtain the association result.

[0081] It can be understood that in this step, the data in the validation set is input into the established association model to obtain the corresponding predicted value. The predicted value is then compared with the actual observed value in the validation set, and the error index between them is calculated, such as the root mean square error (RMSE), mean absolute error (MAE) or correlation coefficient, etc. By comparing the difference between the model predicted value and the actual observed value, the accuracy and reliability of the association model are evaluated.

[0082] It should be noted that step S400 includes step S410, step S420, step S430 and step S440.

[0083] Step S410: Perform text analysis processing according to the third information, extract and process keywords, phrases and sentences of the text data by using a preset word embedding model and vectorize them to obtain text features;

[0084] Specifically, first, the third information will be processed, including pre-processing steps such as word segmentation and removal of stop words, so as to extract keywords, phrases and sentences. These key information involve descriptions of historical governance measures, evaluations of governance effects, descriptions of environmental changes, etc. Then, the extracted key information is vectorized using a pre-trained word embedding model. The word embedding model can map words or phrases to vector representations in a high-dimensional space, so that words with similar semantics are closer in the vector space. This processing can convert text information into numerical vectors that can be understood by computers, which is convenient for subsequent model building and calculation processing.

[0085] Step S420: performing vector decomposition processing according to text features, and obtaining a topic vector for each governance measure by subject-decomposing the description information of the governance measure;

[0086] Preferably, the latent Dirichlet Allocation (LDA) model is used to perform topic decomposition on the extracted text features. Latent Dirichlet Allocation is a text topic model that can represent documents as probability distributions of topics, and topics as probability distributions of words. Through the latent Dirichlet allocation model, the hidden topic structure in the text can be discovered, and the words in the document can be divided according to the topic. The descriptive information of the governance measures is then topic-decomposed to obtain the representation of each governance measure in the topic space. In this way, each governance measure can be represented by a topic vector, which reflects the distribution of the governance measure on different topics. Such a vector representation can better capture the semantic associations and similarities between governance measures.

[0087] Step S430: Based on a preset vector space model, a linear combination of the subject vector and the change in pollution level is constructed to obtain a historical scheme effect prediction model;

[0088] The linear combination in this step refers to the weighted summation of the topic vector of the governance measures and the change in the pollution level according to certain weights to obtain a comprehensive feature representation. Such a feature representation can comprehensively consider the semantic information of the governance measures and their impact on the pollution level, and provide a more comprehensive and accurate feature representation for the prediction of the effects of historical programs. The parameters of the prediction model are then learned through the training data set. During the training process, it is preferred to use supervised learning methods, such as linear regression or support vector machine algorithms, to fit the relationship between the topic vector of the governance measures and the change in the pollution level, so as to obtain a prediction model for the effect of the historical program. The final historical program effect prediction model can accept the topic vector of the governance measures as input and predict the corresponding change in the pollution level.

[0089] Step S440: Perform simulation calculations based on the historical solution effect prediction model to obtain simulation results.

[0090] It should be noted that step S500 includes step S510, step S520, step S530 and step S540.

[0091] Step S510: performing integration processing according to the simulation results, the spatial distribution map and the correlation results, and obtaining an integrated data set by integrating data from different sources into a unified data framework;

[0092] Preferably, during the integration process, the data is normalized to ensure that the data from different sources have the same scale and units. The data is then aligned in time and space dimensions for subsequent analysis and processing.

[0093] Step S520, performing feature extraction processing according to the integrated data set, obtaining a feature set through the pollutant concentration changes in the simulation results, the pollutant distribution characteristics in the spatial distribution map, and the relationship between the emission and the pollution index in the correlation results;

[0094] It should be noted that the change characteristics of pollutant concentrations are extracted from the simulation results. These characteristics include the change amplitude, change trend, and temporal and spatial distribution of pollutant concentrations before and after the implementation of the governance plan. The distribution characteristics of pollutants are extracted from the spatial distribution map. These characteristics reflect the spatial distribution of pollutants after the implementation of the governance plan, including the impact range of pollution sources, the degree of diffusion of pollutants, and the difference in pollutant concentrations between different regions. The relationship characteristics between emissions and pollution indicators are extracted from the correlation results. These characteristics include the degree of influence of emissions on pollutant concentrations, the mutual influence relationship between different pollution sources, and the correlation between different types of pollutants.

[0095] Step S530: Perform ridge regression modeling according to the feature set, standardize the features and establish a spatial weight matrix in combination with the spatial relationship between adjacent monitoring points to construct a mathematical relationship model between the effects of the governance plan and the features;

[0096] It should be noted that in this step, the feature set is standardized to eliminate the dimensional differences between different features and ensure that the weight of each feature on the model is relatively balanced. Preferably, the standardization can be achieved by subtracting the mean of each feature from its value and then dividing it by its standard deviation. A spatial weight matrix is ​​established in combination with the spatial relationship between adjacent monitoring points, and the degree of mutual influence between the monitoring points is determined by considering the geographical location and spatial distance between the monitoring points and possible terrain and other factors. The spatial weight matrix can reflect the spatial relationship between different monitoring points and help to more accurately describe the transmission and diffusion process of pollutants in space. Ridge regression is a regularized linear regression method that can effectively deal with collinearity and overfitting problems between features. By taking the effect of the governance plan as the dependent variable, the feature set and the spatial weight matrix as independent variables, a mathematical model that comprehensively considers the mutual relationship and spatial relationship between features can be established, so as to predict and evaluate the effect of the governance plan.

[0097] Step S540: Perform model verification and evaluation based on the mathematical relationship model to construct a governance solution effect prediction model.

[0098] Specifically, the mathematical relationship model will be verified using the existing data set, with part of the data used for model training and the other part of the data used for model verification. By comparing the difference between the model's prediction results for the verification data and the actual observed values, the model's fitting effect and prediction ability can be evaluated. Secondly, different evaluation indicators will be used to evaluate the performance of the model. Commonly used evaluation indicators include mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), etc. These indicators can help quantify the prediction accuracy and robustness of the model. Finally, the sensitivity analysis of the model will be carried out to evaluate the response of the model to changes in input parameters. Specifically, by introducing different parameter values ​​or disturbances into the model and observing the changes in the model output, it helps to understand the sensitivity of the model to parameter changes, thereby further improving the reliability and applicability of the model. The control scheme effect prediction model constructed in this step can provide a scientific basis and reference for air pollution control decision-making.

[0099] It should be noted that step S600 includes step S610, step S620, step S630 and step S640.

[0100] Step S610: extracting features and encoding the first information, vectorizing the governance methods and expected effects in the scheme to be evaluated, and combining the vectorized representations to obtain a combined vector;

[0101] Specifically, firstly, key features are extracted from the scheme to be evaluated, which include information on the type of governance method, implementation measures, expected effects, etc. By extracting and analyzing this information, it is possible to determine which features are the most critical and important for evaluating the effectiveness of the governance scheme. Next, the extracted features are encoded and converted into numerical or vector forms so that they can be processed and analyzed by computer programs, including converting information such as governance methods and implementation measures into corresponding digital codes or vector representations for subsequent calculations and simulations. Specifically, using technologies such as deep neural networks (DNNs) or convolutional neural networks (CNNs), the various features in the scheme to be evaluated are used as the input of the model, and are mapped to a high-dimensional feature space through multi-layer nonlinear transformations and feature extraction of the network. In this process, the network automatically learns the correlation and complex nonlinear relationships between features, so as to better characterize the features of the scheme to be evaluated. Through the feature representation learned by the deep learning model, a richer and more informative combination vector can be obtained, which contains various key features of the scheme to be evaluated. Such a combination vector can not only better reflect the characteristics of the scheme and the relationship between the features, but also improve the prediction and evaluation performance of the subsequent model.

[0102] Step S620: input the combined vector into the governance scheme effect prediction model for prediction processing to obtain an effect prediction value;

[0103] It is understandable that the effect prediction value can reflect the possible effects and impacts of the governance plan after its implementation.

[0104] Step S630: performing Monte Carlo simulation processing according to the first information, and obtaining Monte Carlo simulation results by randomly sampling different parameters of the governance scheme and combining different environmental conditions;

[0105] It should be noted that, first of all, according to the governance plan and expected effects provided in the first information, the parameters and environmental conditions that need to be simulated are determined. These parameters include different governance measures, the range of changes in pollutant emissions, and the time of implementation of the governance plan. At the same time, environmental conditions involve factors such as meteorological conditions, topography, and population density. Then these parameters and environmental conditions are randomly sampled to generate a large amount of sample data. For each sample data, based on the simulation results, spatial distribution map and correlation results, combined with the previously constructed governance plan effect prediction model, the effect of the governance plan in this case is predicted. By simulating a large amount of sample data, the prediction results of the governance plan effects under various different situations can be obtained. These results reflect the uncertainty and changes of the governance plan under different parameters and environmental conditions, and thus more accurately evaluate the risks and feasibility of the plan.

[0106] Step S640: perform evaluation processing according to the effect prediction value and the Monte Carlo simulation result, and obtain a comprehensive evaluation result by comparing the predicted effect value with the simulation results under different environmental conditions in the Monte Carlo simulation result.

[0107] It should be noted that the effect prediction value is the prediction result based on the governance solution effect prediction model, while the Monte Carlo simulation result is the effect prediction result under various possible situations obtained through random simulation. By comparing the consistency and difference between the two, the reliability and stability of the governance solution under different situations can be determined. Then, considering the different situations in the predicted effect value and the Monte Carlo simulation results, the advantages and disadvantages of the governance solution are comprehensively evaluated. By analyzing the simulation results under various situations, the risk level, feasibility and applicability of the governance solution can be determined, thereby providing a more comprehensive evaluation basis.

[0108] Embodiment 2:

[0109] This embodiment provides an air pollution control program evaluation device, the device comprising:

[0110] an acquisition module, for acquiring first information, second information, third information, fourth information and fifth information, wherein the first information includes the scheme to be evaluated, the second information includes real-time air pollution monitoring data of the target area, the third information includes historical governance measures and corresponding effects, the fourth information is map data of the target area, and the fifth information includes location information of at least one air pollutant emitting enterprise and corresponding air pollutant emission data;

[0111] A clustering device, used for performing spatial clustering processing according to the fourth information and the fifth information, and calculating the pollutant concentration of unmonitored points in the target area to obtain a spatial distribution map of the pollution concentration;

[0112] An analysis device, used for performing correlation analysis processing according to the second information and the fifth information, and obtaining a correlation result by establishing a mathematical model between the emission amount and the pollution index;

[0113] A simulation device, used for performing system dynamics simulation processing according to the third information, and obtaining simulation results by simulating the impact of historical governance solutions on pollution levels;

[0114] A construction device is used to construct a governance scheme effect prediction model based on the simulation results, the spatial distribution map and the correlation results;

[0115] An evaluation device is used to use the first information as an input value of the governance solution effect prediction model for prediction processing, and simulate and evaluate the solution to be evaluated in combination with Monte Carlo simulation to obtain a comprehensive evaluation result.

[0116] Embodiment 3:

[0117] Corresponding to the above method embodiment, this embodiment also provides an air pollution control scheme evaluation device. The air pollution control scheme evaluation device described below and the air pollution control scheme evaluation method described above can refer to each other.

[0118] Figure 2 FIG. 8 is a block diagram of an air pollution control program evaluation device 800 according to an exemplary embodiment. Figure 2 As shown, the air pollution control program evaluation device 800 may include: a processor 801 and a memory 802. The air pollution control program evaluation device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0119] The processor 801 is used to control the overall operation of the air pollution control program evaluation device 800 to complete all or part of the steps in the air pollution control program evaluation method described above. The memory 802 is used to store various types of data to support the operation of the air pollution control program evaluation device 800, and these data may include, for example, instructions for any application or method operating on the air pollution control program evaluation device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent through the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the air pollution control program evaluation device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0120] In an exemplary embodiment, the air pollution control program evaluation device 800 can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processor, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned air pollution control program evaluation method.

[0121] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned air pollution control program evaluation method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the air pollution control program evaluation device 800 to complete the above-mentioned air pollution control program evaluation method.

[0122] Embodiment 4:

[0123] Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the air pollution control program evaluation method described above can refer to each other.

[0124] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the air pollution control scheme evaluation method of the above method embodiment.

[0125] The readable storage medium can specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other readable storage media that can store program codes.

[0126] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0127] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for evaluating an air pollution control program, characterized in that: include: Acquire first information, second information, third information, fourth information and fifth information, wherein the first information includes the scheme to be evaluated, the second information includes real-time air pollution monitoring data of the target area, the third information includes historical governance measures and corresponding effects, the fourth information is map data of the target area, and the fifth information includes location information of at least one air pollutant emitting enterprise and corresponding air pollutant emission data; Performing spatial clustering processing according to the fourth information and the fifth information, and calculating the pollutant concentrations of unmonitored points in the target area to obtain a spatial distribution map of the pollution concentrations; Performing correlation analysis based on the second information and the fifth information, and obtaining a correlation result by establishing a mathematical model between the emission amount and the pollution index; Performing system dynamics simulation processing according to the third information, and obtaining simulation results by simulating the impact of historical governance solutions on pollution levels; A governance scheme effect prediction model is constructed based on the simulation results, the spatial distribution map and the correlation results; The first information is used as an input value of the governance scheme effect prediction model for prediction processing, and the scheme to be evaluated is simulated and evaluated in combination with Monte Carlo simulation to obtain a comprehensive evaluation result; The correlation analysis is performed according to the second information and the fifth information, and a correlation result is obtained by establishing a mathematical model between the emission amount and the pollution index, including: Perform data matching and format conversion processing according to the second information and the fifth information to obtain matching data; Performing correlation analysis on the matching data, and calculating the correlation coefficient between the emission of the emission enterprise and the pollution index of the monitoring point by using the Spearman correlation coefficient to obtain the correlation analysis result; Perform fitting processing according to the correlation analysis results, and obtain a correlation model by using a polynomial regression algorithm to establish a functional relationship between emissions and pollution indicators; Performing model verification processing according to the association model and a preset verification set to obtain an association result; The system dynamics simulation is performed according to the third information, and the simulation results are obtained by simulating the impact of historical governance solutions on the pollution level, including: Performing text analysis processing according to the third information, extracting and processing keywords, phrases, and sentences of the text data by using a preset word embedding model and vectorizing them to obtain text features; Performing vector decomposition processing according to the text features, and obtaining a topic vector for each governance measure by subject-decomposing the description information of the governance measure; Based on a preset vector space model, the subject vector and the change in pollution level are linearly combined to construct a historical scheme effect prediction model; The simulation result is obtained by performing simulation calculation processing according to the historical scheme effect prediction model.

2. The air pollution control program evaluation method according to claim 1, characterized in that: Performing spatial clustering processing according to the fourth information and the fifth information, obtaining a spatial distribution map of pollution concentration by grouping pollution sources and monitoring points according to spatial distribution characteristics and calculating pollutant concentrations of unmonitored points in the target area, including: Performing distance weighted calculation according to the fourth information and the fifth information, and obtaining a weighted distance set by combining the influence of the pollutant types and concentration data of each emission enterprise on the distance; Performing density clustering processing according to the weighted distance set, and obtaining spatial clustering results by setting the distance threshold to a dynamic value related to a preset type of pollutant during the clustering process; Performing spatial interpolation processing according to the spatial clustering result and the atmospheric pollutant concentration data to obtain an interpolation result; Pollution transmission simulation processing is performed according to the interpolation results, and a spatial distribution map of pollution concentration is obtained by simulating the transmission and diffusion process of pollutants in the atmosphere using a preset Gaussian model.

3. The air pollution control program evaluation method according to claim 2, characterized in that: Performing spatial interpolation processing according to the spatial clustering result and the atmospheric pollutant concentration data to obtain an interpolation result includes: Performing kernel density estimation processing according to the spatial clustering results and the atmospheric pollutant concentration data, and generating a thermal map of the kernel density estimation on a map to obtain a thermal distribution map, wherein the areas in the thermal distribution map that are not covered by the thermal map are monitoring missing locations; Performing grid division processing according to the thermal distribution map, dividing the map into at least two levels of grids, and dynamically adjusting the size of each grid according to the distribution of pollution sources and terrain features on the map to obtain a grid map; Interpolation processing is performed on the monitoring missing position in the grid map, and the weight set is obtained by calculating the distance between the monitoring missing position and the surrounding known monitoring points and the elevation information of the monitoring missing position using the inverse distance weighted method, and the interpolation result is calculated based on the weight set.

4. The air pollution control program evaluation method according to claim 1, characterized in that: A governance scheme effect prediction model is constructed based on the simulation results, the spatial distribution map and the correlation results, including: Performing integration processing according to the simulation results, the spatial distribution map and the correlation results, and obtaining an integrated data set by integrating data from different sources into a unified data framework; Performing feature extraction processing according to the integrated data set, obtaining a feature set through changes in pollutant concentrations in the simulation results, pollutant distribution characteristics in the spatial distribution map, and the relationship between emissions and pollution indicators in the correlation results; Perform ridge regression modeling based on the feature set, standardize the features and establish a spatial weight matrix based on the spatial relationship between adjacent monitoring points to construct a mathematical relationship model between the effects of the governance plan and the features; Model verification and evaluation are performed based on the mathematical relationship model to construct a prediction model for the effect of the governance plan.

5. The air pollution control program evaluation method according to claim 1, characterized in that: The first information is used as the input value of the governance scheme effect prediction model for prediction processing, and the scheme to be evaluated is simulated and evaluated in combination with Monte Carlo simulation to obtain a comprehensive evaluation result, including: Perform feature extraction and encoding processing based on the first information, vectorize the governance methods and expected effects in the scheme to be evaluated, and combine the vectorized representations to obtain a combined vector; Inputting the combined vector into the treatment scheme effect prediction model for prediction processing to obtain an effect prediction value; Performing Monte Carlo simulation processing according to the first information, obtaining Monte Carlo simulation results by randomly sampling different parameters of the governance scheme and combining different environmental conditions; An evaluation process is performed based on the effect prediction value and the Monte Carlo simulation result, and a comprehensive evaluation result is obtained by comparing the predicted effect value with the simulation results under different environmental conditions in the Monte Carlo simulation result.

6. An air pollution control program evaluation device, characterized in that: include: an acquisition module, for acquiring first information, second information, third information, fourth information and fifth information, wherein the first information includes the scheme to be evaluated, the second information includes real-time air pollution monitoring data of the target area, the third information includes historical governance measures and corresponding effects, the fourth information is map data of the target area, and the fifth information includes location information of at least one air pollutant emitting enterprise and corresponding air pollutant emission data; A clustering device, used for performing spatial clustering processing according to the fourth information and the fifth information, and calculating the pollutant concentration of unmonitored points in the target area to obtain a spatial distribution map of the pollution concentration; An analysis device, used for performing correlation analysis processing according to the second information and the fifth information, and obtaining a correlation result by establishing a mathematical model between the emission amount and the pollution index; A simulation device, used for performing system dynamics simulation processing according to the third information, and obtaining simulation results by simulating the impact of historical governance solutions on pollution levels; A construction device is used to construct a governance scheme effect prediction model based on the simulation results, the spatial distribution map and the correlation results; An evaluation device, used for performing prediction processing on the first information as an input value of the governance scheme effect prediction model, and simulating and evaluating the scheme to be evaluated in combination with Monte Carlo simulation to obtain a comprehensive evaluation result; Wherein, the analysis device comprises: A first matching unit, configured to perform data matching and format conversion processing according to the second information and the fifth information to obtain matching data; A first analysis unit is used to perform correlation analysis processing according to the matching data, and obtain a correlation analysis result by calculating the correlation coefficient between the emission of the emission enterprise and the pollution index of the monitoring point by using the Spearman correlation coefficient; A first fitting unit, used for performing fitting processing according to the correlation analysis result, and obtaining a correlation model by establishing a functional relationship between the emission amount and the pollution index by using a polynomial regression algorithm; A first verification unit, used to perform model verification processing according to the association model and a preset verification set to obtain an association result; Wherein, the simulation device comprises: A second analysis unit is used to perform text analysis processing according to the third information, and extract and process keywords, phrases and sentences of the text data by using a preset word embedding model and vectorize them to obtain text features; A first decomposition unit is used to perform vector decomposition processing according to the text features, and obtain a topic vector of each governance measure by subject decomposing the description information of the governance measure; A first construction unit, based on a preset vector space model, linearly combines the subject vector and the change in pollution level to obtain a historical scheme effect prediction model; The first simulation unit is used to perform simulation calculation processing according to the historical scheme effect prediction model to obtain a simulation result.

7. An air pollution control program evaluation device, characterized in that: include: Memory for storing computer programs; A processor, used to implement the steps of the air pollution control program evaluation method as described in any one of claims 1 to 5 when executing the computer program.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the air pollution control program evaluation method as described in any one of claims 1 to 5.

Images