A prediction method and related device for a pollution reduction-carbon reduction-greening-growth coupled system

Abstract

This application discloses a prediction method and related apparatus for a pollution reduction-carbon reduction-green expansion-growth coupled system, relating to the field of resource and environmental management. The method includes: constructing an evaluation index system for the pollution reduction-carbon reduction-green expansion-growth coupled system in a target area; using the AHP-entropy weight method to determine the combined weights of each index in the evaluation index system; determining the historical coupling coordination degree sequence based on the combined weights and an improved coupling coordination degree model; clustering the historical coupling coordination degree sequence using a SOM unsupervised neural network to obtain the evolution patterns of each target city; determining the scenario parameters of each target city based on several set development scenarios for its evolution pattern; and using an adaptive hybrid prediction model to predict the coupling coordination degree in future periods based on the data characteristics of the historical coupling coordination degree sequence and the scenario parameters of each target city.

Description

Technical Field

[0001] This application relates to the field of resource and environmental management, and in particular to a prediction method and related apparatus for a pollution reduction-carbon reduction-greening-growth coupled system. Background Technology

[0002] Currently, research on pollution reduction, carbon reduction, greening, and growth is mostly concentrated in single areas, such as pollutant emission assessment, carbon emission accounting, ecological environment quality assessment, and economic development level measurement. While these methods are relatively mature within their respective fields, they lack a holistic characterization of the synergistic relationships between multiple systems and fail to reflect the coupling and interaction mechanisms between subsystems.

[0003] To overcome the aforementioned limitations, some studies have introduced coupling coordination degree models to measure the level of coordinated development of multiple systems and constructed indicator systems by combining comprehensive evaluation methods such as the analytic hierarchy process (AHP) and entropy weight method. However, these methods mainly focus on static evaluation and ex-post analysis, with limited predictive ability for future development trends, making it difficult to provide forward-looking support for policy making. Furthermore, while existing forecasting methods (such as system dynamics models) can predict trends for some indicators, they are mostly limited to single-system or few-variable analysis, lacking organic integration with multi-system coupled evaluation systems, making it difficult to achieve integrated evaluation and forecasting analysis. At the same time, existing technologies generally lack the ability to optimize the design of multi-system coordinated development paths under resource and environmental constraints, making it difficult to form optimal control routes that balance pollution reduction, carbon reduction, greening, and economic growth.

[0004] Therefore, there is an urgent need to propose a comprehensive method that integrates multi-system evaluation, dynamic prediction and path optimization in order to achieve a systematic characterization of the coupling relationship between pollution reduction, carbon reduction, greening and growth and a scientific design of future development paths. Summary of the Invention

[0005] The purpose of this application is to provide a prediction method and related device for a pollution reduction-carbon reduction-green expansion-growth coupled system, which can predict the future trend of the coupled and coordinated development level of multiple systems of pollution reduction, carbon reduction, green expansion and growth.

[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a prediction method for a pollution reduction-carbon reduction-greening-growth coupled system, including: S1. Construct an evaluation index system for the pollution reduction-carbon reduction-greening-growth coupling system in the target area; the evaluation index system includes a pollution reduction subsystem, a carbon reduction subsystem, a greening subsystem, a growth subsystem, and several indicators under each subsystem; the target area includes several target cities; S2. The combined weights of each indicator in the evaluation index system are determined by using the AHP-entropy weight method. S3. Based on the combined weights and the improved coupling coordination degree model, determine the historical coupling coordination degree sequence; the improved coupling coordination degree model includes a coordination development degree index calculation module, a coupling degree index calculation module, and a coupling coordination degree index calculation module. S4. Cluster the historical coupling coordination degree sequence based on the SOM unsupervised neural network to obtain the evolution pattern of each target city; S5. For the evolutionary pattern of each target city, based on several set development scenarios, determine the scenario parameters of the target city; the scenario parameters are the simulated parameters of the indicators in the evaluation index system of the target city under each development scenario. S6. Based on the data characteristics of the historical coupling coordination degree sequence and the scenario parameters of each target city, an adaptive hybrid prediction model is used to predict the coupling coordination degree in future periods.

[0007] In a second aspect, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the prediction method for a pollution reduction-carbon reduction-greening-growth coupled system as described above.

[0008] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the prediction method for a pollution reduction-carbon reduction-greening-growth coupled system as described above.

[0009] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the prediction method for a pollution reduction-carbon reduction-greening-growth coupled system as described above.

[0010] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a prediction method and related apparatus for a pollution reduction-carbon reduction-green expansion-growth coupled system. By constructing a complete evaluation index system comprising four subsystems, it covers the coordinated development requirements of pollution reduction, carbon reduction, green expansion, and growth across these four dimensions. The method improves the rationality of index weights by combining weights using the AHP-entropy weight method. An improved coupling coordination degree model yields a historical coupling coordination degree sequence, which more accurately reflects the true coupling and interaction relationships between the four subsystems. An unsupervised SOM neural network is used to cluster the historical coupling coordination degree sequence to obtain the evolutionary patterns of each city, automatically identifying the development path characteristics of different cities and avoiding the subjective bias of manually dividing evolutionary patterns. Scenario settings are set for different evolutionary patterns, and adaptive hybrid prediction models are matched, automatically selecting appropriate prediction model combinations based on the characteristics of different historical data. This application solves the problem of existing technologies' difficulty in conducting integrated evaluation and prediction, and can provide scientific and forward-looking support for the formulation of regional green and low-carbon development policies. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is an application environment diagram of a prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to an embodiment of this application; Figure 2 A schematic flowchart illustrating a prediction method for a pollution reduction-carbon reduction-greening-growth coupled system provided in an embodiment of this application; Figure 3 A scatter plot of 110 cities in a certain economic belt of the Yangtze River Economic Belt provided as an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0015] The prediction method for the pollution reduction-carbon reduction-greening-growth coupled system provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server.

[0016] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0017] In one exemplary embodiment, such as Figure 2 As shown, a prediction method for a pollution reduction-carbon reduction-greening-growth coupled system is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the following steps are included: S1. Construct an evaluation index system for the pollution reduction-carbon reduction-greening-growth coupling system in the target area; the evaluation index system includes a pollution reduction subsystem, a carbon reduction subsystem, a greening subsystem, a growth subsystem, and several indicators under each subsystem; the target area includes several target cities; S2. The combined weights of each indicator in the evaluation index system are determined by using the AHP-entropy weight method. S3. Based on the combined weights and the improved coupling coordination degree model, determine the historical coupling coordination degree sequence; the improved coupling coordination degree model includes a coordination development degree index calculation module, a coupling degree index calculation module, and a coupling coordination degree index calculation module. S4. Cluster the historical coupling coordination degree sequence based on the SOM unsupervised neural network to obtain the evolution pattern of each target city; S5. For the evolutionary pattern of each target city, based on several set development scenarios, determine the scenario parameters of the target city; the scenario parameters are the simulated parameters of the indicators in the evaluation index system of the target city under each development scenario. S6. Based on the data characteristics of the historical coupling coordination degree sequence and the scenario parameters of each target city, an adaptive hybrid prediction model is used to predict the coupling coordination degree in future periods.

[0018] In an exemplary embodiment, when performing steps 201-206, the specific steps may be as follows: To promote the coordinated development of carbon reduction, pollution reduction, greening, and growth, a basic indicator library for PCGE evaluation was constructed. During indicator selection, based on the subsystem interaction mechanism, a comprehensive evaluation system covering 22 indicators across four subsystems—carbon reduction, pollution reduction, greening, and growth—was built, as shown in Table 1. The evaluation indicator system includes a system layer (4 primary indicators), a criterion layer (8 secondary indicators), and an indicator layer (22 tertiary indicators). The primary indicators include the pollution reduction subsystem, carbon reduction subsystem, greening subsystem, and growth subsystem; the secondary indicators include resource utilization, environmental governance, energy utilization, low-carbon construction, agriculture, shared development, development efficiency, and development quality; the tertiary indicators highlight the level of coordinated development of urban pollution reduction, carbon reduction, greening, and growth, as well as key areas, crucial links, and safeguard measures.

[0019] Table 1 Evaluation Index System for Pollution Reduction, Carbon Reduction, Greening, and Growth in a Certain River Economic Belt

[0020] Specifically, this embodiment takes 110 cities in a certain Yangtze River Economic Belt from 2004 to 2022 as the research object, and uses an improved coupling coordination model to calculate the coupling coordination degree among the four systems of pollution reduction, carbon reduction, greening expansion and growth, and deeply analyzes their spatial evolution characteristics: construct a PCGE evaluation index system for the Yangtze River Economic Belt, and use convergence analysis, Moran's index and standard deviation ellipse to evaluate and reveal the coupling coordination ability of the four and their spatial evolution characteristics.

[0021] This application uses City N as an example to demonstrate the historical data obtained, as shown in Table 2: Table 2 Historical Data

[0022] Specifically, when determining the combined weights of each indicator in the evaluation index system, the AHP-entropy weighting method is selected as follows: The Analytic Hierarchy Process (AHP) combines qualitative and quantitative methods to determine indicator weights and is widely used in management and decision-making research. Its specific calculation steps are as follows: ① Establish a comparison matrix ,in As factors u Compared to factorsv Importance quantification value The value range is 1-9.

[0023] ② Calculate the matrix U The largest eigenvector is obtained through the formula Obtain the matrix Maximum eigenvalue and its corresponding eigenvectors w A consistency check is then performed.

[0024] ③ If the random consistency ratio Therefore, it can be determined that the matrix has strong consistency, and then... w After normalization, the final weight vector is obtained. That is Weights of each indicator.

[0025] The entropy weight method is a method for calculating the weight of indicators under objective conditions. It is widely used in environmental assessment and management. The basic principle of this method is that the greater the difference between evaluation indicators, the more important that indicator is. In the indicator evaluation process, positive indicators generally show that a larger indicator value is more beneficial to the evaluation result, while the opposite is true for negative indicators; a smaller value is more beneficial to the evaluation result. The entropy weight method has advantages such as objective calculation process and high accuracy of results. Therefore, this embodiment uses the entropy weight method to determine the weight of each indicator in an urban-scale evaluation system. The specific calculation steps are as follows: ①Establish using raw data m × n Indicator Judgment Matrix R .

[0026] ② To eliminate the influence of different orders of magnitude and dimensions in the original data, the judgment matrix will be... R The standardization process yields a new standardization matrix. B In the specific standardization process, the indicators are divided into positive, negative, and moderate categories, and the specific calculation methods are as follows: (1); (2); (3); In the formula: For the first i The first indicator, the first j Each sample value for Standardized values, and The first i The maximum and minimum values ​​of each indicator.

[0027] ③ Calculate the indicator weights; see the formula for details: (4); (5); In the formula: Indicates the first i Class indicator weights, n Indicates the total number of indicator types. Indicates the first i Information entropy of each indicator.

[0028] Since the entropy weight method primarily assigns weights based on the degree of data dispersion, it may overlook some indicators that are theoretically significant but exhibit little regional variation. Therefore, relying solely on the entropy weight method cannot comprehensively reflect the importance of the indicators. To address this, this embodiment introduces a combined weighting method based on the AHP-entropy weighting approach. This reduces the possibility of a mismatch between subjective / objective weighting and the actual importance of the indicators, thus more accurately reflecting the weights of each evaluation indicator. After determining the subjective and objective weights, the combined weights of each indicator can be obtained, as shown in Table 3. The specific calculation formula is as follows: (6); Table 3. Evaluation Indicator Weights of Urban PCGE Coupling and Coordination Effect

[0029] The remaining indicators are processed as described above to obtain their weights. These weights are then normalized to obtain the weights of the indicators at the criterion layer, as shown in the formula: (7); The improved coupling coordination degree model can be specifically described as follows: (1) Coordinated Development Index Calculation Module: The comprehensive evaluation method is used to calculate the city's PCGE coordinated development index. The specific formula is as follows: (8); (9); (10); In the formula, U 1 , U 2 , U 3 , U 4 These represent the comprehensive development scores of the pollution reduction subsystem, carbon reduction subsystem, green expansion subsystem, and growth subsystem, respectively. The weight values ​​for each criterion layer of the pollution reduction subsystem, carbon reduction subsystem, green expansion subsystem, and growth subsystem are as follows: Gk This represents the evaluation results of each criterion level in the pollution reduction subsystem, carbon reduction subsystem, green expansion subsystem, and growth subsystem. T To coordinate the development index.

[0030] (2) The coupling index in the coupling index calculation module is typically used to characterize the interaction relationships between systems. When C When = 0, it indicates that the coupling between systems is extremely low, and the subsystems or internal elements of the system are in a relatively disordered state; when C When the coupling index is 1, it indicates that the coupling degree between systems has reached its maximum, and there is benign cooperation between subsystems or elements within the system. The system as a whole will evolve towards a new ordered structure. The coupling degree index is calculated as follows: (11); Generally, the coupling degree model of the four subsystems is calculated using the above-described standard form. However, in practical applications, there are often... C The model suffers from issues such as low validity of values. Therefore, it is necessary to improve the coupling degree model and enhance its effectiveness. C The value differentiation effect allows the calculation results to more reasonably reflect the coupling and coordination relationship between systems and their development level. The coupling degree classification level is shown in Table 4.

[0031] Table 4 Coupling Degree and Evaluation Level

[0032] when At that time, assuming , (12); In the formula: , C This is the coupling degree index.

[0033] (3) In the coupling coordination index calculation module, since coupling degree mainly reflects the coupling strength between systems, it is difficult to comprehensively reflect the overall coordination level of the composite system. In actual analysis, there may be situations where the comprehensive index values ​​of each subsystem are low, but the coupling degree is high. Therefore, it is necessary to introduce a coordination development degree index on the basis of coupling degree to construct a coupling coordination index. D This method aims to more comprehensively depict the coordination relationships between systems and serves as an important approach for analyzing system relationships and a standard for classifying coupling coordination degree levels. Table 5 shows the specific results, and Table 6 shows the evaluation results of the PCGE coupling coordination degree of a certain economic belt in Jiangxi Province from 2004 to 2022. The calculation method for the coupling coordination degree index is as follows: (13); Table 5 Coupling Coordination Degree and Evaluation Level

[0034] Table 6 Evaluation Results of PCGE Coupling Coordination in a Certain River Economic Belt from 2004 to 2022

[0035] After constructing the evaluation index system, a convergence analysis is performed.

[0036] Specifically, based on σ Convergence and absolute β The study analyzes the evolution characteristics of the PCGE coupling and coordination differences in the entire Yangtze River Economic Belt and its three major regions. σ Convergence indicates that the differences in PCGE coupling coordination effects across regions gradually decrease over time. When, it indicates that the area exists. σ Convergence. Using the coefficient of variation. σ The following formula is used for testing: (14); In the formula: n The number of cities in the region. for i City No. t Annual coupling coordination degree, for n The city t Average annual coupling coordination degree.

[0037] absolute β Convergence is based on the assumption that, given the fundamental characteristics of each region, the PCGE coupling coordination effect will gradually converge to the same level over time. When β If the value is less than 0 and passes the significance test, then an absolute value is considered to exist. β Convergence is indicated by convergence, while divergence is indicated by divergence, as shown in the following formula: (15); In the formula: for i Urban base-period coupling coordination degree α For constant terms, β This is the coupling coordination coefficient. This is the random error term.

[0038] As shown in Table 7, during the study period, the β value of the PCGE coupling coordination effect in a certain city of the Yangtze River Economic Belt was -0.4796, and it was significant at the 1% level, indicating that there is an absolute PCGE coupling coordination effect among the 110 cities. β Convergence means that, under basically similar development conditions, the differences in the coupling and coordination effects of PCGE (Planetary, Industrial, and Geological) will gradually decrease over time. From a regional perspective, the three major regions... βAll values ​​were negative and passed the 1% significance test, indicating that the PCGE coupling coordination effect among cities in the region will gradually converge to the same level over time.

[0039] Table 7. Absolute Synergistic Effect of Pollution Reduction, Carbon Reduction, Greening, and Growth Coupling in a Certain City in the Yangtze River Economic Belt β convergence

[0040] The global Moran index can be used to determine whether there is spatial autocorrelation in the PCGE coupling and coordination effect among cities in a certain Yangtze River Economic Belt. The formula is as follows: (16); In the formula: and Representing cities i and j PCGE coupling coordination degree; For variance; For elements of the economic geography matrix; The value ranges from [-1, 1]. A positive value indicates a positive correlation, a negative value indicates a negative correlation, and a value of 0 indicates a random distribution.

[0041] Calculating only the global Moran index will not yield specific characteristics; therefore, the local Moran index must be calculated next, as shown in the following formula: (17); The standard deviation ellipse and centroid migration are mainly used to analyze the directionality and changing trends of the spatial distribution of geographic features. Basic parameters include the center, major axis, minor axis, and azimuth. This example uses ArcGIS 10.8 software to analyze the centroid migration trajectory and spatial distribution characteristics of the PCGE coupling coordination effect in cities. The formula is as follows: Average center: ; (18); X-axis standard deviation: (19); Y-axis standard deviation: (20); In the formula: The spatial region under study; Spatial weights; These are the coordinates of the centroid of the ellipse; for X Shaft standard deviation; for Y Axis standard deviation.

[0042] Among them, such as Figure 3As shown in (a)-(d) of the figure, the spatial evolution characteristics of the PCGE coupling and coordination effect of 110 cities in a certain Yangtze River Economic Belt are explored using Moran scatter plots. 2006, 2011, 2016, and 2021 were selected as observation years, with coupling and coordination degree as the criterion. Z Using the horizontal axis as the x-axis and the spatial lag term WZ as the y-axis, cities are divided into four quadrants. The PCGE coupling and coordination effect of cities in a certain Yangtze River Economic Belt exhibits clustering characteristics, meaning that high-value and low-value areas have strong spatial correlations, and the influence between regions is also strong. The increase in the number of "high-high" and "low-low" clusters in 2011, 2016, and 2021 indicates strong spatial correlations between cities and a spatial promoting effect. In 2021, "high-high" clusters were mainly concentrated in the core area of ​​the Yangtze River Delta and some midstream cities. These cities have strong interactions and a high degree of synergy in economic development and green policies, indicating a significant spatial spillover effect. "Low-low" clusters are distributed in upstream areas; these cities have relatively weak economic foundations, greater ecological pressure, and limited driving effects between adjacent cities.

[0043] Specifically, when determining the evolutionary pattern of each target city, the following can be considered: 1) Calculate the average distance between each sample data point in the historical coupling coordination degree sequence and the data points within the same cluster, as well as the distance between the sample data points and the data points in neighboring clusters; 2) Based on the average distance and the distance between the sample data and the data points in the neighboring clusters, determine the contour score based on the tightness and similarity between each sample data in the historical coupling coordination degree sequence and its own cluster and neighboring clusters; 3) Based on the contour scores, adjust the preset number of clusters in the SOM unsupervised neural network until the contour scores reach their highest value, thus obtaining the final evolutionary pattern classification results for each target city. Specifically, based on PCGE coupling coordination degree time series data, an unsupervised SOM neural network is used to classify urban development patterns and output stable clustering results for differentiated prediction. A trend-corrected obstacle degree model is adopted, which integrates indicator weights, contribution, annual average obstacle degree, frequency of high-frequency obstacles, and interannual change rate to achieve dynamic and forward-looking obstacle factor identification.

[0044] Pattern Analysis and Clustering (SOM) learns from data in the input space to generate a low-dimensional, discrete mapping to adapt to changing and complex environments, and has been widely used in the field of hydrogeology. SOM is an unsupervised neural network algorithm proposed by Kohonen for pattern analysis and clustering. It clusters individuals based on distance calculations, grouping individuals with small distances into one class and those with large distances into another, thus achieving pattern recognition and classification of samples. The SOM network mainly consists of an input layer and a competition layer. The input layer receives external data and maps sample data to the best-matching neuron; the competition layer continuously adjusts the distance weights between neurons to find classification patterns, calculated using the following formula: (twenty one); In the formula: For neurons i The input vector, For input layer neurons i Competing layer neurons j The weight values ​​between them n This represents the number of vectors in the input layer. This embodiment uses the somtoolbox 2.0 toolbox in Matlab R2020b to implement sample modeling, training, and application.

[0045] This embodiment utilizes Matlab R2020b to import 2019 coupling coordination degree data from 110 cities in a certain Yangtze River Economic Belt. Using the SOM algorithm, a grid diagram is used to visualize the clustering results, allowing for intuitive observation of the cluster categories among different cities. The silhouette score calculates the tightness and similarity between each sample and its own cluster and neighboring clusters by measuring the average distance between a sample and data points within the same cluster, as well as the distance between a sample and data points in neighboring clusters. A higher silhouette score indicates tighter clustering and greater inter-cluster data differences, resulting in better model performance. The calculation formula is as follows: (twenty two); (twenty three); In the formula, a Indicates the average distance within a sample cluster; b The average distance between the nearest neighbor clusters of the representative sample; N The number of clusters. S The overall outline score. This represents the number of true categories.

[0046] Specifically, based on the SOM clustering results of PCGE system coupling coordination degree data of 110 cities from 2004 to 2022, the differences in the spatial distribution and temporal evolution of coupling coordination degree among cities in a certain Yangtze River Economic Belt can be clearly identified. On this basis, by selecting representative sample cities and analyzing their changing trends, 27 cities were finally selected for simulation and prediction. The specific city clustering results are shown in Table 8.

[0047] Table 8. Results of the selection of representative cities

[0048] The obstacle degree model reflects the degree to which each indicator restricts the evaluation level. A higher obstacle degree value indicates a more significant impact of the indicator on the evaluation results, and is more detrimental to the region's development. To identify key factors restricting the coupling and coordination of the PCGE system, this embodiment uses the obstacle degree model to assess the impact of each indicator on the four systems. This model combines the gap between the indicator and the target value with the indicator weights to identify major obstacle factors, providing a basis for policy optimization. The obstacle degree calculation formula is as follows: (twenty four); (25); In the formula: The weights of each indicator (the weight of each individual indicator in relation to the overall goal). Scoring for each indicator, , For the first j The degree to which each indicator hinders the level of coordinated development at its corresponding criterion level. For the first i The degree of obstacle to the collaborative development level of a system layer on the criterion layer to which it belongs.

[0049] Based on the above formula, the obstacle degree of each indicator and subsystem from 2004 to 2022 was calculated. By ranking them by magnitude, the main obstacle factors affecting the coupling and coordination of the PCGE systems in 110 cities were diagnosed. Furthermore, determining the key influencing factors requires multi-dimensional aggregation analysis to evaluate the indicator weights and obstacle degree ranking frequencies, thereby adjusting the key influencing factors affecting the evaluation results. This provides a scientific basis for formulating local plans and strategic measures for the coupling and coordination of pollution reduction, carbon reduction, greening, and growth. The specific steps are as follows: ① Calculate the annual average obstacle level for each indicator.

[0050] (26); ② Calculate the frequency with which each indicator becomes a "major obstacle". Based on the aforementioned steps and formulas, the raw data is first standardized, and then, using tertiary indicators as units, the obstacles affecting the coupled and coordinated development of the PCGE system are identified and ranked. Further, the obstacle degree for each city is calculated year by year and arranged in descending order. The top 5 key obstacle indicators are selected (considering the large number of tertiary indicators) to highlight the main influencing factors. The total number of times each indicator appears in the top 5 across all cities and all years is calculated: (27); (28); In the formula: I () is the indicator function (1 if the indicator ranks in the top K in the city in that year, otherwise 0).

[0051] ③ Calculate the interannual rate of change.

[0052] (29); (30); In the formula: It is an indicator j exist t Yearly obstacles; It is a regression cutoff term. It is the interannual rate of change. It is a random error term. A value greater than 0 indicates an increasing obstacle level year by year. If the value is less than 0, the barrier level decreases year by year.

[0053] ④ Construct a dynamically adjusted barrier degree. Embed the trend coefficient into the barrier degree calculation process.

[0054] (31); (32); In the formula: T j This is the normalized trend coefficient; This is the trend adjustment coefficient; This represents the barrier degree after trend correction.

[0055] ⑤ Calculate the overall score. The score is calculated by weighting the average obstacle level (40%), the frequency of Top K (40%), and the annual rate of change (20%) (the weights are preset by the model and reflect the importance of each dimension). The higher the score, the more critical the indicator.

[0056] (33); Specifically, this application calculates the comprehensive score of each indicator based on the principles of high comprehensive score, domain focus, and comprehensive selection, using an obstacle degree model. After sorting the indicators in descending order of score, the specific ranking is shown in Table 9. The top 30% of indicators are selected as key factors.

[0057] Table 9 Ranking of Key Barrier Factors

[0058] The various development scenarios set when determining the scenario parameters for the target city are as follows: This embodiment primarily simulates the PCGE (Plan-Do-Check-Act) coupling and coordinated development level of various cities in a certain Yangtze River Economic Belt from 2023 to 2030 based on the following five different development scenarios, thereby determining the optimal guarantee scheme and providing direction and theoretical basis for future PCGE coupling and coordinated development work in the Yangtze River Economic Belt. The scenario settings take into account the heterogeneity of cities. This embodiment sets the following five development scenarios for cities in the Yangtze River Economic Belt: (1) Status quo scenario. This scenario introduces no additional policy shocks, and all 10 key indicators grow at the historical average growth rate. (See the corresponding table.) X 2 , X 5 , X 7 , X 9 , X 10 , X 13 , X 15 , X 16 , X 20 , X 21 All values ​​are "&", meaning they are taken based on historical trends, without setting separate parameters for average annual growth or decline.

[0059] (2) Open source and cost reduction model. This scenario focuses on strengthening sewage collection and treatment and agricultural non-point source emission reduction. In principle, the centralized collection rate of domestic sewage in each city should be increased by more than 5 percentage points compared with 2020, and the comprehensive utilization rate of livestock and poultry manure should reach more than 80% by 2025. The sewage treatment rate is set to increase by about 2% to 4% per year, the comprehensive utilization rate of livestock and poultry manure should increase by about 1% to 3% per year, and the amount of chemical fertilizer applied should decrease by about 2% to 5% per year.

[0060] (3) Environmentally friendly type. This scenario emphasizes ecological benefits. The goal is to achieve a forest coverage rate of 24.1% and an air quality good or excellent rate of 87.5% in prefecture-level and above cities by 2025. The target is to increase the air quality good or excellent rate by about 2% to 4% per year and the forest coverage rate by about 0.2% to 0.8% per year.

[0061] (4) Economic Growth Model. This scenario emphasizes high-quality growth driven by the increasing proportion of the service sector. The average annual increase in the proportion of the tertiary industry is set at approximately 1% to 3%.

[0062] (5) Comprehensive and coordinated type. Combining the improvement measures of scenarios (2) to (4), the annual carbon emissions per unit of GDP are set to decrease by about 3% to 5%.

[0063] Based on the resource endowment, economic structure, and ecological positioning of the 27 cities, they were divided into five city clusters. Scenario parameters were set according to the differences among the cities, and the specific values ​​for the four subsystem indicators are shown in Table 10. Specifically, the indicators corresponding to the pollution reduction subsystem include water supply coverage rate, the proportion of administrative villages treating domestic sewage, and the centralized treatment rate of sewage treatment plants; the indicators corresponding to the carbon reduction subsystem include total energy consumption and carbon dioxide emissions per unit of GDP; the indicators corresponding to the greening subsystem include the decrease in chemical fertilizer application, the comprehensive utilization rate of livestock and poultry manure, and the forest coverage rate; and the indicators corresponding to the growth subsystem include the proportion of the tertiary industry's added value to GDP and the proportion of days with good or excellent air quality in the city.

[0064] Table 10 Simulation Parameter Calibration for Each Subsystem

[0065] In this process, based on the data characteristics of the historical coupling coordination degree sequence and the scenario parameters of each target city, an adaptive hybrid prediction model is used to predict the coupling coordination degree in future periods. This requires constructing a candidate model library for the adaptive hybrid prediction model. This candidate model library includes the ARIMA linear prediction model, the GM(1,1) grey prediction model, and the exponential smoothing model, covering linear, nonlinear, small-sample, and long-time-series scenarios.

[0066] (1) ARIMA linear prediction model: The ARIMA model is a classic time series forecasting method suitable for modeling and predicting data with time-evolutionary characteristics. Coupling and coordination, as a typical time series variable, can be used in this model for trend characterization and predictive analysis. ARIMA (… p , d , q In the model, p Indicates the order of the autoregressive term. d It is the difference order.q This represents the order of the moving average term. Essentially, this model, based on the autoregressive moving average model, transforms a non-stationary sequence into a stationary one through differencing, thereby improving prediction accuracy. When the original sequence is non-stationary, differencing is required until the stationarity requirement is met, and the order is determined accordingly. d The value is then used to determine the order of the autoregressive term that yields the optimal result, combined with model identification methods. p and the order of the moving average term q The general expression for the ARIMA model is: (34); In the formula: for t Difference sequences of order; for t Order noise sequence; for p Parameters for fitting the first-order model; for q Parameters for fitting the first-order model; r It is a constant.

[0067] Considering that the ARIMA model is more suitable for short-term forecasting, excessively long fitting and forecasting periods can lead to error accumulation. Therefore, this embodiment uses data from 2004 to 2019 as the training set and data from 2020 to 2022 as the test set to evaluate the model's fitting and predictive performance. The specific modeling steps are as follows: First, the stationarity of the series is determined using time series plots and the ADF unit root test; second, non-stationary series are differencing, and the processed series undergoes a white noise test; third, the model is identified and its order is determined by combining autocorrelation function and partial autocorrelation function plots; subsequently, the model parameters are estimated using the least squares method; then, the model's effectiveness is verified through residual white noise tests and parameter significance tests; finally, short-term predictive analysis is conducted after the model passes the tests.

[0068] (2) GM(1,1) Grey Prediction Model: This method is suitable for predicting the coupling coordination degree of urban PCGE systems with limited data volume, small sequence fluctuations, and clear trends. The GM (1,1) grey prediction model has been widely used in economic, social, and engineering fields. This embodiment specifically adopts the sequence prediction method in this model, that is, by quantitatively predicting the evolution of the indicator over time, the value of the indicator at future points in time is obtained. First, the original coupling coordination degree sequence is... Perform first-order accumulation to generate a sequence. Construct matrices B and Y n .

[0069] Then, the grey differential equation is constructed: (35); In the formula: β 1 To develop grayscale, To determine the endogenous control gray number; secondly, to set the vector of parameters to be estimated as... Solve using the least squares method ,have to Solving equation (35), we obtain the prediction model as follows: (36); In the formula: k = (1, 2, ..., n).

[0070] Finally, the accuracy of the prediction model is verified. If the verification results simultaneously satisfy... P >0.7 and C If the value is less than 0.65, then the model can obtain a reasonable and effective predicted value for the indicator through testing.

[0071] (3) Exponential smoothing model (Holt): The trend forecast of the exponential smoothing model is a weighted average of historical actual values; that is, the closer the actual value is to the forecast value, the higher the weight. During the weighting process, historical data are assigned a weight, which follows the principle of larger weights for closer data and smaller weights for farther data.

[0072] (37); In the formula: a , b , c These are the weight values; F 4 This is a predicted value; Y 1 , Y 2 , Y 3 This is the actual value.

[0073] The single-smoothing method, also known as the simple smoothing method, does not add any trend terms.

[0074] (38); In the formula: Y t yes t The actual value for the period; F t yes t The predicted value for the period, a It is a smoothing coefficient. It is selected by comparing the predicted value with the actual value to minimize the prediction error.

[0075] After constructing the candidate model library, the selection mechanism for the adaptive hybrid prediction model is as follows: based on input data features, such as stationarity, sample size, and nonlinearity intensity, the system automatically calculates MAE, RMSE, and R. 2 Select the optimal model. Model selection is automatic. Lower MAE and RMSE are better. R 2 The closer to 1, the better. Normalize the three indicators (eliminate dimensions) and set weights: RMSE (0.4), MAE (0.3), R... 2 (0.3). Calculate the overall score for each model, and the model with the highest score is the optimal model. To select the optimal prediction model, this application uses MAE, RMSE, and R to evaluate the ARIMA model, GM(1,1) model, and Holt exponential smoothing model. 2 The predictive performance of each model was comprehensively evaluated using various indicators. Table 11 below shows a comparison of the prediction results of the GM, ARIMA, and exponential smoothing (Holt) models. During the prediction period, the coupling and coordination degree of the Yangtze River Economic Belt showed a continuous positive trend.

[0076] Table 11 Prediction Results of Three Major Models for a Certain River Economic Belt

[0077] As shown in Table 12, the ARIMA model has the highest overall score of 0.2038, and its overall predictive performance is better than the GM(1,1) model and the Holt exponential smoothing model, indicating that it can more effectively characterize the time series evolution of PCGE coupling coordination degree in a certain Yangtze River Economic Belt. Therefore, this application ultimately selects the ARIMA model as the optimal prediction model, and on this basis, conducts simulation and prediction analysis of PCGE coupling coordination degree under different scenarios.

[0078] Table 12 Optimal Model Selection

[0079] The ARIMA linear forecasting model was used to predict the PCGE coupling coordination degree of 27 representative cities from 2023 to 2030, forecasting future time series trends. Based on the differences in the original time series data of each city, autocorrelation and partial autocorrelation analyses were used to select the parameters for the ARIMA model, ultimately determining the model parameters for simulation and prediction. The parameter selection for the ARIMA model is shown in Table 13.

[0080] Table 13 ARIMA Model Parameter Selection

[0081] The effectiveness of the predictions for 27 cities was evaluated, and the relative errors of the ARIMA model predictions for each region ranged from 1.33% to 3.24%. Table 14 shows the average prediction accuracy of the ARIMA model from 2004 to 2022. The average relative error represents the mean relative error of different prediction models for representative cities from 2004 to 2022. As shown in the table below, the prediction accuracy for each city is relatively ideal, meaning it can predict the coupling coordination degree of the study area from 2023 to 2030.

[0082] Table 14. Mean Relative Error of ARIMA Model, 2004–2022

[0083] Specifically, Tables 15-19 show the predicted coupling coordination degree of 27 cities under five scenarios.

[0084] Table 15 Coupling and Coordination Degree under the Scenario of Continued Development of the Current Status Quo

[0085] Table 16 Coupling and Coordination Degree under the Development Scenario of Open Source and Cost Reduction

[0086] Table 17 Coupling Coordination Degree under Environmentally Friendly Development Scenario

[0087] Table 18 Coupling Coordination Degree under Economic Growth Development Scenario

[0088] Table 19 Coupling Coordination Degree under Integrated Coordination Development Scenario

[0089] This application uses the ARIMA linear forecasting model to predict the period from 2023 to 2030. The model's relative error ranges from 1.33% to 3.24%, indicating reliable prediction results. The forecast trend shows that the coupling coordination level of most cities continues to improve, evolving from a state of barely coordinated development to a state of primary coordinated development, with regional disparities showing a convergence trend. In comparisons of different scenarios, the comprehensive coordination model performs best in improving the level of PCGE coupling coordination development, achieving stronger synergistic effects and demonstrating greater overall improvement than a single scenario. Therefore, in the future, while maintaining existing development, a comprehensive coordination path should be prioritized, emphasizing the linkage between addressing ecological shortcomings and green transformation, strengthening pollution control, and simultaneously promoting low-carbon transformation and industrial structure upgrading. This will encourage all subsystems to work in the same direction, continuously improving the level of PCGE coupling coordination development in the Yangtze River Economic Belt.

[0090] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4 As shown, the computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores the predictive structure of a pollution reduction-carbon reduction-greening-growth coupled system. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a predictive method for a pollution reduction-carbon reduction-greening-growth coupled system.

[0091] Figure 4 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0092] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0093] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0094] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with relevant regulations and be authorized by the owner of the corresponding device.

[0095] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0096] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0097] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0098] This embodiment uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application; at the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. In summary, the content of this specification should not be construed as a limitation of this application.

Claims

1. A prediction method for a pollution reduction-carbon reduction-greening-growth coupled system, characterized in that, include: S1. Construct an evaluation index system for the pollution reduction-carbon reduction-greening-growth coupling system in the target area; the evaluation index system includes a pollution reduction subsystem, a carbon reduction subsystem, a greening subsystem, a growth subsystem, and several indicators under each subsystem; the target area includes several target cities; S2. The combined weights of each indicator in the evaluation index system are determined by using the AHP-entropy weight method. S3. Based on the combined weights and the improved coupling coordination degree model, determine the historical coupling coordination degree sequence; the improved coupling coordination degree model includes a coordination development degree index calculation module, a coupling degree index calculation module, and a coupling coordination degree index calculation module. S4. Cluster the historical coupling coordination degree sequence based on the SOM unsupervised neural network to obtain the evolution pattern of each target city; S5. For the evolutionary pattern of each target city, based on several set development scenarios, determine the scenario parameters of the target city; the scenario parameters are the simulated parameters of the indicators in the evaluation index system of the target city under each development scenario. S6. Based on the data characteristics of the historical coupling coordination degree sequence and the scenario parameters of each target city, an adaptive hybrid prediction model is used to predict the coupling coordination degree in future periods.

2. The prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to claim 1, characterized in that, The coupling coordination degree sequence includes the pollution reduction subsystem index, carbon reduction subsystem index, greening subsystem index, growth subsystem index, coordinated development degree index, coupling degree index, and coupling coordination degree index.

3. The prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to claim 1, characterized in that, The formula expression for the coordinated development index calculation module is as follows: ； ； ； The formula for calculating the coupling index is as follows: ； The formula for calculating the coupling coordination index is as follows: ； In the formula, U 1 , U 2 , U 3 , U 4 These represent the comprehensive development scores of the pollution reduction subsystem, carbon reduction subsystem, green expansion subsystem, and growth subsystem, respectively. The weight values ​​for each criterion layer of the pollution reduction subsystem, carbon reduction subsystem, green expansion subsystem, and growth subsystem are as follows: G k This represents the evaluation results of each criterion level in the pollution reduction subsystem, carbon reduction subsystem, green expansion subsystem, and growth subsystem. T To coordinate the development index, C This is the coupling degree index. D This is the coupling coordination index.

4. The prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to claim 1, characterized in that, Clustering of the historical coupling coordination degree sequence based on the SOM unsupervised neural network yields the evolutionary patterns of each target city, specifically including: Calculate the average distance between each sample data point in the historical coupling coordination sequence and the data points within the same cluster, as well as the distance between the sample data points and the data points in neighboring clusters; Based on the average distance and the distance between the sample data and the data points in the neighboring clusters, the contour score is determined according to the tightness and similarity between each sample data in the historical coupling coordination degree sequence and its own cluster and neighboring clusters. Based on the contour scores, the preset number of clusters in the SOM unsupervised neural network is adjusted until the contour scores reach their highest value, thus obtaining the final evolutionary pattern classification results for each target city.

5. The prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to claim 4, characterized in that, The formula for calculating the contour score is: ； ； In the formula, a Indicates the average distance within a sample cluster; b The average distance between the nearest neighbor clusters of the representative sample; N The number of clusters. S The overall outline score. This represents the number of true categories.

6. The prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to claim 1, characterized in that, The candidate model library for the adaptive hybrid prediction model includes the ARIMA linear prediction model, the GM(1,1) grey prediction model, and the exponential smoothing model.

7. The prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to claim 6, characterized in that, The expression for the ARIMA linear prediction model is: ； In the formula: for t Difference sequences of order; for t Order noise sequence; for p Parameters for fitting the first-order model; for q Parameters for fitting the first-order model; r It is a constant.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement a prediction method for a pollution reduction-carbon reduction-greening-growth coupled system according to any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements a prediction method for a pollution reduction-carbon reduction-greening-growth coupled system as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements a prediction method for a pollution reduction-carbon reduction-greening-growth coupled system as described in any one of claims 1-7.

Images