A method for constructing a dynamic threshold system of county-level per capita carbon emissions

Abstract

This application relates to the field of low-carbon city technology and provides a method for constructing a dynamic threshold system for per capita carbon emissions at the county level. The method involves: determining the basic principles for selecting reference standards for dynamic thresholds of per capita carbon emissions at the county level; selecting preset countries as low-level, medium-level, and high-level reference standards based on these principles; collecting historical data on per capita carbon emissions and per capita GDP of the preset countries; constructing mathematical models of CO2 environmental Kuznets curves corresponding to the low-level, medium-level, and high-level reference standards based on the collected historical data; and matching the corresponding CO2 environmental Kuznets curve mathematical models to the historical economic development level of the target county with the national average, thereby calculating the dynamic thresholds for per capita carbon emissions at different development stages of the target county. This application provides a scientific basis for counties at different development levels to formulate low-carbon emission reduction targets.

Description

Technical Field

[0001] This application relates to the field of low-carbon city technology, and in particular to a method for constructing a dynamic threshold system for per capita carbon emissions in counties. Background Technology

[0002] Currently, my country's low-carbon pilot programs are mainly conducted at the provincial and municipal levels, generally setting only comprehensive emission reduction targets such as carbon emission intensity and total carbon emissions. No per capita carbon emission control reference standards have yet been established for the county level. Existing carbon emission threshold studies often directly reference static data from high-income developed regions that have already surpassed their per capita carbon emission peaks, failing to consider the economic development differences among counties at different stages of development. This results in thresholds that are either too high to effectively constrain emissions reduction or too low to restrict normal economic development in counties. Summary of the Invention

[0003] This application provides a method for constructing a dynamic threshold system for per capita carbon emissions in counties, aiming to solve the problems of existing methods that do not combine the inherent evolutionary laws of carbon emissions and economic development, cannot provide thresholds that are dynamically adjusted with the development of county economies, and have not established a hierarchical reference system adapted to counties with different economic levels, making it difficult to meet the needs of low-carbon transformation planning in my country's numerous and unevenly developed counties.

[0004] In a first aspect, embodiments of this application provide a method for constructing a dynamic threshold system for per capita carbon emissions in a county, the method comprising: The basic principles for selecting reference standards for the dynamic threshold of per capita carbon emissions in counties include benchmarking against advanced standards, dynamism, regularity of the CO2 environmental Kuznets curve, convergence of the CO2 environmental Kuznets curve, and economic sensitivity. Based on the basic selection principles, preset countries were selected as low-end, mid-end, and high-end reference standards, respectively. All preset countries were identified by preset green evaluation tools as having excellent low-carbon performance, and their per capita carbon emissions and per capita GDP both showed the carbon dioxide environmental Kuznets curve pattern. Collect historical data on per capita carbon emissions and per capita GDP of the target country; construct carbon dioxide environmental Kuznets curve mathematical models corresponding to low, medium and high reference standards based on the collected historical data; and calculate the dynamic threshold of per capita carbon emissions at different development stages of the target county by matching the corresponding carbon dioxide environmental Kuznets curve mathematical models with the historical economic development level of the target county with the national average level.

[0005] In some embodiments, the basic principles for selecting the reference standard for determining the dynamic threshold of per capita carbon emissions in counties include: setting the benchmark for advanced performance by selecting countries or regions that have been identified as having excellent low-carbon performance by green assessment tools; setting the dynamic nature by adjusting the selected reference standard according to changes in economic development level and per capita income; setting the regularity of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of first increasing and then decreasing per capita carbon emissions with per capita GDP; setting the convergence of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of convergence of per capita carbon emissions with increasing income; and setting economic sensitivity by setting differentiated per capita carbon emission benchmark thresholds based on different income levels.

[0006] In some embodiments, the preset countries are all identified as having excellent low-carbon performance by a preset green assessment tool, and their per capita carbon emissions and per capita GDP both exhibit the CO2 Environmental Kuznets Curve pattern. This includes: selecting countries from a global list that have been identified as having excellent low-carbon performance by a preset green assessment tool; selecting countries with more than 50 consecutive years of time series data on per capita carbon emissions and per capita GDP; performing an existence test of the CO2 Environmental Kuznets Curve on the selected countries; and retaining the countries that pass the existence test as preset countries.

[0007] In some embodiments, the step of selecting preset countries as low-end, mid-end, and high-end reference standards based on basic selection principles includes: calculating the historical peak per capita GDP of all preset countries; sorting the preset countries from low to high according to their historical peak per capita GDP; selecting the first third of the sorted countries as a low-end reference standard preparation set; selecting the middle third of the sorted countries as a mid-end reference standard preparation set; selecting the last third of the sorted countries as a high-end reference standard preparation set; and selecting one country from each of the three preparation sets as the reference standard for the corresponding level.

[0008] In some embodiments, the collection of historical per capita carbon emission data and historical per capita GDP data for a preset country includes: obtaining annual per capita carbon emission data for more than fifty consecutive years from a preset publicly authoritative database; obtaining annual nominal per capita GDP data for the corresponding year from the same database; converting the nominal per capita GDP data into real data expressed in US dollars at constant prices for the preset year; and performing time series alignment processing on the converted data to ensure that the per capita carbon emission data and the per capita GDP data correspond one-to-one.

[0009] In some embodiments, constructing mathematical models of carbon dioxide environmental Kuznets curves corresponding to low-end, mid-end, and high-end reference standards based on collected historical data includes: performing logarithmic transformations on the collected per capita carbon emission data and per capita GDP data; performing a unit root stationarity test on the transformed data to verify whether the data is a first-order integrated sequence; performing a cointegration test on the data that passes the stationarity test; performing regression analysis on the cointegrated variables using the least squares method; adding a first-order autocorrelation term to eliminate the autocorrelation of the regression residuals; performing a stationarity test on the corrected regression residual sequence; and determining the regression equation as the corresponding carbon dioxide environmental Kuznets curve mathematical model if the residual sequence is stationary.

[0010] In some embodiments, matching the target county's historical economic development level with the national average level to the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model includes: obtaining the target county's annual per capita GDP data for the past twenty years; obtaining the national annual per capita GDP data for the same period; comparing the target county's per capita GDP with the national per capita GDP year by year; counting the number of years in which the target county's per capita GDP exceeded the national per capita GDP; and matching the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model based on the number of years counted.

[0011] In some embodiments, the calculation of the dynamic threshold for per capita carbon emissions at different development stages of the target county includes: obtaining the per capita GDP forecast data for each planning year of the target county; converting the per capita GDP forecast data into a value expressed in US dollars at constant prices for the preset year; substituting the converted value into the matched carbon dioxide environmental Kuznets curve mathematical model; calculating the dynamic threshold for per capita carbon emissions for the corresponding planning year of the target county; and organizing the calculation results in chronological order to form a sequence of dynamic thresholds for per capita carbon emissions of the target county.

[0012] In some embodiments, the method further includes: using the isolated forest algorithm to detect outliers in the collected preset national historical data; removing the detected outliers and supplementing the missing data using linear interpolation; using a long short-term memory network model to predict the future per capita GDP of the target county under multiple scenarios; generating per capita GDP prediction sequences under optimistic, neutral, and pessimistic scenarios; substituting the prediction sequences under the three scenarios into the matched mathematical model respectively; and calculating the dynamic threshold sequence of per capita carbon emissions of the target county under the corresponding three scenarios.

[0013] In some embodiments, the method further includes: collecting actual per capita carbon emission data and actual per capita GDP data of the target county on an annual basis; comparing the actual per capita carbon emission data with the dynamic threshold of per capita carbon emission for the corresponding year; calculating the deviation between the actual value and the threshold; using a deep reinforcement learning algorithm to dynamically optimize the parameters of the carbon dioxide environmental Kuznets curve mathematical model based on the deviation; using the optimized mathematical model to update the dynamic threshold of per capita carbon emission for subsequent years in the target county; and generating an annual threshold adjustment report containing the threshold update content and the basis for adjustment.

[0014] This application establishes a dynamic per capita carbon emission threshold system adapted to the development characteristics of counties in my country, filling the gap in per capita carbon emission reference standards at the county level and providing a scientific basis for counties at different development levels to formulate low-carbon emission reduction targets. By setting five core selection principles, the application ensures that the reference standard is both advanced and applicable, avoiding the unreasonable thresholds caused by directly copying static data from high-income regions. A three-tiered reference standard system (low, medium, and high) is constructed, which can accurately match the emission reduction needs of counties at different economic development levels, balancing emission reduction fairness with the coordination of economic development. Based on the carbon dioxide environmental Kuznets curve, a dynamic model is constructed to achieve dynamic adjustment of the per capita carbon emission threshold according to the county's economic development level, solving the problem that static thresholds cannot adapt to changes in the county's development stage.

[0015] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart illustrating the steps of a method for constructing a dynamic threshold system for per capita carbon emissions in a county, as provided in an embodiment of this application. Figure 2 A flowchart illustrating one implementation method of the present invention for constructing a dynamic threshold system for per capita carbon emissions in a county. Figure 3 It serves as a major domestic and international carbon emission assessment indicator and standard system; Figure 4 A schematic diagram of the Kuznets curve representing per capita carbon emissions versus per capita GDP; Figure 5 The graph shows the relationship between Singapore's per capita carbon emissions and the logarithm of GDP per capita (EKC), along with the fitted curve. Figure 6 This is a schematic block diagram of a system for constructing a dynamic threshold system for per capita carbon emissions in a county, provided in one embodiment of this application. Figure 7 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0019] 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, 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.

[0020] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0021] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0022] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0023] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0024] Currently, my country's low-carbon pilot programs are mainly conducted at the provincial and municipal levels, generally setting only comprehensive emission reduction targets such as carbon emission intensity and total carbon emissions, without establishing per capita carbon emission control reference standards for the county level. Existing carbon emission threshold studies mostly directly refer to static data from high-income developed regions that have already surpassed the peak per capita carbon emissions, failing to consider the economic development differences among counties at different stages of development. This results in thresholds that are either too high to effectively constrain emissions reduction or too low to restrict normal economic development in counties. Furthermore, existing methods do not incorporate the inherent evolutionary laws of carbon emissions and economic development, failing to provide thresholds that can be dynamically adjusted with county-level economic development, and lacking a tiered reference system adapted to counties with different economic levels. This makes it difficult to meet the low-carbon transformation planning needs of my country's numerous and unevenly developed counties.

[0025] Please refer to Figure 1 This invention provides a method for constructing a dynamic threshold system for per capita carbon emissions at the county level. The method relies on computer equipment to complete the entire automated computation process. The computer equipment can be deployed in three ways: first, on a single physical server, running the entire computation program independently in a standalone hardware environment; second, on a distributed server cluster built with multiple servers, relying on the cluster's computing power to complete batch parallel data computation across multiple counties; and third, using a lightweight deployment mode, where the program is streamlined and installed on portable terminal hardware with data storage, data retrieval, and numerical computation capabilities, such as laptops, portable smart terminals, wearable smart devices, and smart service robots. All macroeconomic statistical data from various countries and domestic county-level economic and carbon emission statistical data required for this method's computation are collected in accordance with regulations and under the legal authorization of the data holding institutions and the local development and reform and environmental protection authorities. The data acquisition process strictly follows relevant standards, and there is no unauthorized capture of confidential information or illegal collection of user privacy data.

[0026] The provided method for constructing a dynamic threshold system for per capita carbon emissions at the county level includes steps S101 to S103. Details are as follows: Step S101. Determine the basic principles for selecting the reference standard for the dynamic threshold of per capita carbon emissions in the county. The basic principles include benchmarking against advanced standards, dynamism, regularity of the carbon dioxide environmental Kuznets curve, convergence of the carbon dioxide environmental Kuznets curve, and economic sensitivity.

[0027] Specifically, the computer pre-defines the judgment logic of five screening criteria at the program level. The five criteria are: benchmarking advancement, dynamism, regularity of the CO2 environmental Kuznets curve, convergence of the CO2 environmental Kuznets curve, and economic sensitivity. Each criterion is standardized and defined according to the general evaluation standards of low-carbon industries, the law of carbon emission and economic evolution, and empirical conclusions on the convergence of residents' income and carbon emissions. All judgment rules serve as prerequisite hard constraints for subsequent screening of international benchmark countries. All subsequent program calculations for country screening and tier classification are automatically executed according to the pre-defined criteria in this step.

[0028] Step S102. Based on the basic selection principles, preset countries are selected as low-end, mid-end, and high-end reference standards, respectively. All preset countries are identified as having excellent low-carbon performance by preset green evaluation tools, and their per capita carbon emissions and per capita GDP both show the carbon dioxide environmental Kuznets curve pattern.

[0029] Specifically, the first step involves the computer calling upon the five pre-set criteria of S101 to screen candidate countries from around the world that meet the green evaluation criteria, and whose per capita carbon emissions and per capita GDP conform to the evolution characteristics of the carbon dioxide environmental Kuznets curve. After screening, Portugal, Singapore, and Japan are selected as benchmark reference countries. The second step involves classifying the three countries into tiers according to their economic development levels, designating Portugal as the low-tier reference standard country, Singapore as the mid-tier reference standard country, and Japan as the high-tier reference standard country. The third step involves simultaneously transferring the national tier classification rules to the domestic county-level classification system. The domestic county-level classification is based on the comparison of per capita GDP over the past twenty years with the national average level during the same period, forming a three-tier benchmark framework with international benchmarking and domestic tiered correspondence.

[0030] Step S103. Collect historical data on per capita carbon emissions and per capita GDP of the preset countries; construct carbon dioxide environmental Kuznets curve mathematical models corresponding to low, medium and high reference standards based on the collected historical data; and calculate the dynamic threshold of per capita carbon emissions at different development stages of the target county by matching the corresponding carbon dioxide environmental Kuznets curve mathematical models with the comparison results of the historical economic development level of the target county with the national average level.

[0031] Specifically, this step consists of four sequential sub-operations: Sub-operation one: The computer connects to the World Bank's official authoritative database data interface to collect annual per capita carbon emission data and nominal per capita GDP data for Portugal, Singapore, and Japan in batches. The per capita GDP is uniformly converted according to the constant 2010 US dollar standard. The two types of data are pre-processed to align with the same year, and invalid entries with misaligned years or missing data are removed. Sub-operation two involves performing econometric verification and regression modeling operations on three national standardized time series data corresponding to low-end, medium-end, and high-end data, respectively, and sequentially completing data preprocessing, stationarity verification, cointegration verification, regression fitting, and residual verification, thereby constructing three independent mathematical models of carbon dioxide environmental Kuznets curves. Sub-operation three: retrieve the per capita GDP data of the target county over the past 20 years and the per capita GDP data of the country during the same period, compare the values ​​year by year and count the total number of years in which the county's per capita GDP is higher than the national average, and match the curve model of the corresponding level according to the classification rules. Taking Changxing County as an example, the county's per capita GDP in most years over the past 20 years is higher than the national average, so it is matched with the corresponding high-level benchmark model of Japan. Sub-operation four: Input the predicted per capita GDP for each planning year of the target county, complete the currency conversion and input the matched curve model, automatically solve the dynamic threshold of per capita carbon emissions for each year, and sort and summarize them according to the natural year to form a complete dynamic threshold sequence for the whole cycle of the county.

[0032] In some embodiments, the basic principles for selecting the reference standard for determining the dynamic threshold of per capita carbon emissions in counties include: setting the benchmark for advanced performance by selecting countries or regions that have been identified as having excellent low-carbon performance by green assessment tools; setting the dynamic nature by adjusting the selected reference standard according to changes in economic development level and per capita income; setting the regularity of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of first increasing and then decreasing per capita carbon emissions with per capita GDP; setting the convergence of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of convergence of per capita carbon emissions with increasing income; and setting economic sensitivity by setting differentiated per capita carbon emission benchmark thresholds based on different income levels.

[0033] This embodiment is used to complete the writing of quantitative judgment rules for five basic principles within a computer program, including: The first principle is to benchmark against advanced standards: The program has a built-in global mainstream green evaluation project directory, and only marks countries or regions that have obtained official green certifications such as the European Green Capital Award and the Asian Green City Index as meeting the advanced screening criteria. Article 2, Dynamism: The program sets threshold linkage correction logic, and the benchmark threshold is automatically adjusted synchronously with the changes in the values ​​of the target county's annual per capita disposable income and regional economic growth rate, thus abandoning the fixed static carbon emission standards throughout the process. The third point is the regularity of the carbon dioxide environmental Kuznets curve: the program has a built-in inverted U-shaped judgment logic. Only when the long-term data of the corresponding region verifies that the per capita carbon emissions show a change characteristic of first increasing and then falling along with the increase of per capita GDP, can this screening condition be met. Article 4, Convergence of the CO2 Environmental Kuznets Curve: The program verifies through convergence calculation logic that if the per capita carbon emission index in the sample area continues to converge toward the carbon emission range of developed economies as residents' income increases year by year, then the convergence requirement is met. Article 5, Economic Sensitivity: The program divides income into multiple tiers based on per capita GDP, and sets a differentiated per capita carbon emission benchmark limit for each income tier, so as to achieve differentiated control of thresholds based on income levels.

[0034] In some embodiments, the preset countries are all identified as having excellent low-carbon performance by a preset green assessment tool, and their per capita carbon emissions and per capita GDP both exhibit the CO2 Environmental Kuznets Curve pattern. This includes: selecting countries from a global list that have been identified as having excellent low-carbon performance by a preset green assessment tool; selecting countries with more than 50 consecutive years of time series data on per capita carbon emissions and per capita GDP; performing an existence test of the CO2 Environmental Kuznets Curve on the selected countries; and retaining the countries that pass the existence test as preset countries.

[0035] This embodiment enables a fully automated computer screening of eligible benchmark countries, performing four layers of screening operations sequentially: The first step is to retrieve the list of basic information of global sovereign states, and based on the benchmarking and advancement judgment rules in Example 1, to initially screen the countries that have been rated as having excellent low-carbon development by authoritative green evaluation tools, and generate the first round of candidate country list. The second step involves the program traversing all countries in the first round of the list, accessing the World Bank database to retrieve macroeconomic statistics for each country, retaining only countries with complete per capita carbon emissions and per capita GDP time-series data for 50 years or more, and removing entities with key data gaps of more than three years to form a second-round screening list. The third step is to invoke the rules for determining the regularity of the curve and conduct a quantitative test on the existence of the carbon dioxide environmental Kuznets curve for each country in the secondary list. The fourth step is to retain all countries that pass the curve pattern test as a preset country pool. After screening, this scheme finally determined Portugal, Singapore and Japan as the three benchmark countries.

[0036] In some embodiments, the step of selecting preset countries as low-end, mid-end, and high-end reference standards based on basic selection principles includes: calculating the historical peak per capita GDP of all preset countries; sorting the preset countries from low to high according to their historical peak per capita GDP; selecting the first third of the sorted countries as a low-end reference standard preparation set; selecting the middle third of the sorted countries as a mid-end reference standard preparation set; selecting the last third of the sorted countries as a high-end reference standard preparation set; and selecting one country from each of the three preparation sets as the reference standard for the corresponding level.

[0037] This embodiment completes the automatic classification of the base country's gear level, implemented step by step: The first step is for the computer to extract the peak GDP per capita data for each country in the preset country pool for the entire statistical period. The second step is to arrange all the preset countries in ascending order of their peak GDP per capita, from lowest to highest, to generate an ordered sequence of countries. The third step is to divide the ordered sequence into three equal intervals: the countries in the first third interval are assigned to the low-end candidate set, the middle third to the medium-end candidate set, and the last third to the high-end candidate set. The fourth step is to select Portugal as the low-end benchmark country from the low-end candidate set, Singapore as the mid-end benchmark country from the mid-end candidate set, and Japan as the high-end benchmark country from the high-end candidate set. The fifth step is to store the binding relationship between the three countries and their corresponding tiers in the program database, so that subsequent modeling and county-level matching can directly retrieve this binding relationship.

[0038] In some embodiments, the collection of historical per capita carbon emission data and historical per capita GDP data for a preset country includes: obtaining annual per capita carbon emission data for more than fifty consecutive years from a preset publicly authoritative database; obtaining annual nominal per capita GDP data for the corresponding year from the same database; converting the nominal per capita GDP data into real data expressed in US dollars at constant prices for the preset year; and performing time series alignment processing on the converted data to ensure that the per capita carbon emission data and the per capita GDP data correspond one-to-one.

[0039] This embodiment standardizes the complete process steps for raw data capture, currency conversion, and time series pairing, including: The first step involved the computer using an official data interface to retrieve raw statistics on per capita carbon emissions for Portugal, Singapore, and Japan from 1960 to 2016 from the World Bank database. The second step is to synchronously extract the raw data of nominal GDP per capita for the corresponding country and the corresponding statistical year from the same database. The third step involves using a built-in conversion factor for US dollars at constant prices in 2010. This factor is used to convert nominal GDP per capita into real GDP per capita in US dollars at constant prices in 2010, thus eliminating data discrepancies caused by inflation and exchange rate fluctuations. The fourth step involves using the natural calendar year as a unified index field to align per capita carbon emission data and converted per capita GDP data for the same country and year. Data entries with incorrect years or missing key indicators are then removed to generate a standardized panel dataset that can be directly used for modeling.

[0040] In some embodiments, constructing mathematical models of carbon dioxide environmental Kuznets curves corresponding to low-end, mid-end, and high-end reference standards based on collected historical data includes: performing logarithmic transformations on the collected per capita carbon emission data and per capita GDP data; performing a unit root stationarity test on the transformed data to verify whether the data is a first-order integrated sequence; performing a cointegration test on the data that passes the stationarity test; performing regression analysis on the cointegrated variables using the least squares method; adding a first-order autocorrelation term to eliminate the autocorrelation of the regression residuals; performing a stationarity test on the corrected regression residual sequence; and determining the regression equation as the corresponding carbon dioxide environmental Kuznets curve mathematical model if the residual sequence is stationary.

[0041] This embodiment relies on standardized panel data to complete the construction of three-tiered measurement models step by step, with the entire process automated by computer. The first step is to perform a logarithmic transformation on the two sets of raw data, namely per capita carbon emissions and per capita GDP, after preprocessing, to eliminate the interference of sample heteroscedasticity on the modeling accuracy. The second step is to use the unit root test algorithm to perform stationarity tests on the logarithmic data and retain only the valid data samples that satisfy the first-order integrated feature. The third step is to conduct cointegration tests on the variables that have passed the stationarity test and screen for combinations of variables that have long-term equilibrium relationships. The fourth step is to use the least squares method to perform regression fitting on the cointegrated variables and introduce a first-order autocorrelation term into the regression equation to eliminate the autocorrelation problem of the residual sequence. The fifth step is to perform a stationarity check on the model residual sequence after autocorrelation correction. If the residual sequence satisfies the stationarity characteristic at the 1% significance level, the current regression equation is determined to be valid. Step 6: Repeat the above operations to model panel data for Portugal, Singapore, and Japan respectively, and finally generate three independent and effective carbon dioxide environmental Kuznets curve benchmark models for low, medium and high levels.

[0042] In some embodiments, matching the target county's historical economic development level with the national average level to the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model includes: obtaining the target county's annual per capita GDP data for the past twenty years; obtaining the national annual per capita GDP data for the same period; comparing the target county's per capita GDP with the national per capita GDP year by year; counting the number of years in which the target county's per capita GDP exceeded the national per capita GDP; and matching the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model based on the number of years counted.

[0043] This embodiment achieves automatic identification and model binding of county-level economic levels, implemented step by step: The first step is for the computer to retrieve annual per capita GDP statistics for the target county for the past twenty years from the database authorized by the county statistics bureau. The second step is to simultaneously retrieve official statistics on per capita GDP for each year within the same statistical period nationwide. The third step is to compare the per capita GDP of the target county with the national per capita GDP for the same period year by year, and mark whether the county's value is higher or lower than the national average year by year. The fourth step is to count the total number of years in the past twenty years in which the per capita GDP of the target county was higher than the national average. The fifth step involves matching a low-end model to a county whose per capita GDP for more than 15 years is lower than the national average. If the per capita GDP for each year fluctuates similarly to the national average, a mid-end model is matched. If the per capita GDP for more than 15 years is higher than the national average, a high-end model is matched. Since Changxing County's per capita GDP has been significantly higher than the national average for many years, the program automatically matches the corresponding high-end benchmark model in Japan.

[0044] In some embodiments, the calculation of the dynamic threshold for per capita carbon emissions at different development stages of the target county includes: obtaining the per capita GDP forecast data for each planning year of the target county; converting the per capita GDP forecast data into a value expressed in US dollars at constant prices for the preset year; substituting the converted value into the matched carbon dioxide environmental Kuznets curve mathematical model; calculating the dynamic threshold for per capita carbon emissions for the corresponding planning year of the target county; and organizing the calculation results in chronological order to form a sequence of dynamic thresholds for per capita carbon emissions of the target county.

[0045] This embodiment calculates the full-cycle threshold in batches based on the matched benchmark model: The first step is to obtain the per capita GDP forecast data for each planned year of the target county from the local development and reform department; The second step is to call the built-in 2010 constant price USD conversion rules to uniformly convert the RMB-denominated county per capita GDP forecast to the 2010 constant price USD value. The third step is to substitute the per capita GDP forecast data after currency conversion into the tier benchmark model that has been matched in Example 6. The fourth step is to use the model to calculate and output the dynamic threshold for per capita carbon emissions in each county for the corresponding planning year. The fifth step involves the program systematically archiving and organizing all calculated thresholds in chronological order from the earliest to the latest year of the planning, generating a time series table of dynamic per capita carbon emission thresholds for the entire lifecycle of the county. Changxing County completed the threshold calculation for the years 2005 to 2019 through this step.

[0046] In some embodiments, the method further includes: using the isolated forest algorithm to detect outliers in the collected preset national historical data; removing the detected outliers and supplementing the missing data using linear interpolation; using a long short-term memory network model to predict the future per capita GDP of the target county under multiple scenarios; generating per capita GDP prediction sequences under optimistic, neutral, and pessimistic scenarios; substituting the prediction sequences under the three scenarios into the matched mathematical model respectively; and calculating the dynamic threshold sequence of per capita carbon emissions of the target county under the corresponding three scenarios.

[0047] This embodiment relies on artificial intelligence algorithms to optimize the quality of raw data and expand the dimensions of threshold measurement in multiple scenarios, serving as a supplementary optimization method to the basic threshold scheme: The first step involves the computer retrieving preprocessed historical panel data from three benchmark countries and using the isolated forest anomaly detection algorithm to automatically identify anomalous data entries within the dataset that deviate from the normal range. The second step is to remove the identified abnormal data and use linear interpolation to fill in the missing values ​​at the missing data positions, thus completing the secondary cleaning and optimization of the benchmark dataset. The third step is to build a long short-term memory network prediction model within a computer program and use cleaned cross-national macro-time series data to train the network model parameters. The fourth step involves generating three differentiated per capita GDP prediction sequences for the target county's future planning year, based on the converged neural network model after training, under three scenarios: optimistic, neutral, and pessimistic. The fifth step is to substitute all the per capita GDP forecast data under the three scenarios into the county-level benchmark model that has been matched respectively. The sixth step involves scenario-based calculations to obtain dynamic threshold sequences for per capita carbon emissions in counties under three scenarios: optimistic, neutral, and pessimistic. These three thresholds serve as alternative reference indicators for differentiated low-carbon planning in counties.

[0048] In some embodiments, the method further includes: collecting actual per capita carbon emission data and actual per capita GDP data of the target county on an annual basis; comparing the actual per capita carbon emission data with the dynamic threshold of per capita carbon emission for the corresponding year; calculating the deviation between the actual value and the threshold; using a deep reinforcement learning algorithm to dynamically optimize the parameters of the carbon dioxide environmental Kuznets curve mathematical model based on the deviation; using the optimized mathematical model to update the dynamic threshold of per capita carbon emission for subsequent years in the target county; and generating an annual threshold adjustment report containing the threshold update content and the basis for adjustment.

[0049] This embodiment establishes an annual data feedback and iteration mechanism, optimizing the model and thresholds year by year based on county-level measured data to form a closed-loop management system. The first step is to automatically collect the official per capita carbon emission data and annual per capita GDP statistics of the target county at the end of each natural year accounting cycle. The second step is for the program to retrieve the generated benchmark threshold data for the same year, compare the actual carbon emission values ​​with the corresponding year's calculated thresholds, and quantify the positive and negative deviations between the two sets of data. The third step is to build a deep reinforcement learning optimization algorithm module, using the annual measured deviation value as the algorithm reward and punishment feedback parameter, and dynamically fine-tuning the various coefficients inside the corresponding Kuznets curve model based on the reward and punishment parameter. The fourth step is to use the updated baseline model with optimized parameters to recalculate and correct the dynamic thresholds for per capita carbon emissions for all subsequent planning years in the target county. The fifth step involves the program automatically summarizing the annual data comparison results, model parameter adjustment details, and full-cycle threshold update content, compiling a standardized annual threshold dynamic adjustment report, and archiving it for future reference, thereby achieving a closed-loop iterative optimization of the model and threshold every year.

[0050] In some embodiments, such as Figure 2 , Figure 3 , Figure 4 and Figure 5 As shown, the basic principles for selecting the reference standard for the dynamic threshold of per capita carbon emissions in counties are as follows. These principles include: benchmarking against advanced practices, dynamism, EKC regularity, EKC convergence, and economic sensitivity.

[0051] Benchmarking against best practices: my country's carbon emission reduction targets and pilot programs are mostly based at the provincial and municipal levels. For counties that have not yet clearly defined low-carbon development emission reduction targets, they can refer to advanced regions at home and abroad that have achieved good low-carbon and green development, and study and learn from their advanced development experiences in achieving peak per capita carbon emissions. This will help to formulate and implement the best low-carbon transformation and development policies.

[0052] Dynamism: When selecting advanced regions as benchmarks, international cities have often already surpassed their per capita carbon emission peaks and decoupled from the benchmark. Therefore, using static data from the current point in time for benchmarking may lead to evaluation bias and impracticality. For counties that still prioritize economic development, their per capita carbon emission benchmark thresholds should change in accordance with their economic development level and per capita income.

[0053] CKC regularity: Based on the current status and stage of my country's per capita carbon emission development, the selected regions need to exhibit a CKC regularity for per capita carbon emissions, which will be beneficial for future county-level benchmarking to achieve similar results.

[0054] Further, CKC refers to the Environmental Kuznets Curve (EKC). The basic meaning of EKC is that in the early stages of regional economic development, rapid population growth, relatively backward industrial production technology, and severe resource waste lead to continuously aggravated environmental pollution. With economic development, the contribution of technologically advanced industries to the economy becomes increasingly prominent, and people's awareness and ability to control environmental pollution gradually increase, resulting in a gradual slowdown in pollutant emissions and an improvement in environmental quality. For carbon dioxide emissions, this hypothesis is called the CO2 Environmental Kuznets Curve (CKC). Figure 3 ).

[0055] CKC Convergence: Empirical studies on convergence show that increased income leads to more similar per capita emissions, implying that with economic development and increased income in counties, per capita carbon emissions can also converge and converge under certain conditions according to EKC. To enhance the practicality of the benchmark, counties are compared using CKC curve convergence.

[0056] Economic sensitivity: The convergence benchmark of the county-level CKC curve is mainly based on the CKC curve pattern of advanced regions, and a per capita carbon emission benchmark threshold with economic sensitivity is formulated according to different income levels.

[0057] Meanwhile, based on the above-mentioned basic principles, the benchmark regions for per capita carbon emission benchmark thresholds in low-carbon counties must meet three conditions: international advancement, data availability, and comparability. They must be regions that have been selected as having excellent performance through mature low-carbon and green evaluation tools or standards; long-term panel data of the region must be available, which has mathematical and statistical significance and can be verified through econometric modeling to determine whether the region has a CKC curve; and they must be comparable to counties in terms of scale, population density, industrial structure, and development path, which is beneficial for the benchmark regions to achieve the same results in the future.

[0058] The reasons for selecting countries as reference standards for S2 low, medium, and high per capita carbon emission thresholds are as follows: Certain cities in Portugal, Singapore, and Japan have been identified as performing well by the green assessment tool, and all three countries exhibit a CKC curve pattern in their per capita carbon emission thresholds and per capita GDP. Considering the economic development of these three countries, Portugal's CKC curve is used as the reference standard for low-end county-level per capita carbon emission thresholds, Singapore as the reference standard for medium-end county-level per capita carbon emission thresholds, and Japan as the reference standard for high-end county-level per capita carbon emission thresholds. Lisbon, the capital of Portugal, was the winner of the 2020 European Green Capital Award and is one of the less economically developed developed countries. Singapore is considered the greenest city in Asia in the Asia Green Cities Index. Since 1965, Singapore's industrial structure has undergone a transformation similar to my country's in terms of economy and energy, evolving from a developing to a developed post-industrial era. Tokyo, Yokohama, and Osaka in Japan have passed the Green Cities Index assessment. In the past two decades of rapid economic development, Japan has completed its industrial transformation and upgrading, ranking among the world's leaders in industry, manufacturing, and technology, with leading technological strength in some areas. After experiencing rapid economic growth, Japan has taken a solid step forward in developing a low-carbon economy through various measures.

[0059] The classic equation for the Kuznets EKC curve is as follows: Y it =α i +β 1 x it +β 2 x 2 it +β 3 x 3 it +β 4 z it +ε it (Formula 2) In the formula, Y it For environmental quality indicators, X it As an indicator of economic development, ε it For random error terms; subscript i It represents a country or region, and t refers to time. α i It is a constant. β k These are the coefficients of the explanatory variables, all of which are constants to be estimated. Based on... β kPresented differently X it and Y it The different relationships. Among them, if β 1>0, β 2<0, and β 3=0, X it and Y it The relationship between them follows an inverted U-shaped EKC curve, with the inflection point at - β 1 / 2 β 2. This model depicts the slowing growth of per capita carbon emissions in an economy as it transitions from rapid income growth to higher income levels, reflecting the convergence between income level and growth rate.

[179] To reduce data fluctuations and eliminate heteroscedasticity, most studies have adopted the method of simplifying the study by performing a logarithmic transformation on the relevant data.

[180] .

[0060] This invention examines the relationship between per capita carbon emissions (CEPC), a quality indicator of carbon emissions, and economic development. It selects per capita gross domestic product (GDPPC), which reflects the level of economic development per capita, to represent the level of economic development, and expresses this data in constant 2010 US dollars to eliminate the influence of population factors, exchange rates, and inflation. Similarly, the classic model is simplified by taking the logarithm, and the model is constructed as follows: lnCEPC it =α i +β 1 lnGDPPC it +β 2 lnGDPPC 2 it +β 3 lnGDPPC 3 it +ε it (Formula 3) In the formula, CEPC it Carbon emissions per capita, expressed as tCO2e / person. GDPPC it GDP per capita, expressed in US dollars per person, at constant 2010 prices. ε it For random error terms; subscript i Representing a country or region t Refers to time; α It is a constant. β kThese are the coefficients of the explanatory variables, all of which are constants to be estimated; the relevant calculations are performed in Eveiws 9.0.

[0061] The criteria for classifying county-level carbon emission thresholds into three levels (I, II, and III) can be found in Table 1.

[0062] Table 1 Basic data collection: per capita carbon emissions and per capita GDP data for Japan, Singapore, and Portugal.

[0063] The study used World Bank data on per capita GDP and per capita carbon emissions for Japan, Singapore, and Portugal from 1960 to 2017, as shown in Table 2.

[0064] Table 2. World Bank figures for GDP per capita and carbon emissions per capita in Portugal, Singapore, and Japan, 1960-2016. Unit root stationarity and cointegration tests were performed on the logarithm lnCEPC of Singapore's dependent variable per capita carbon emissions (CEPC) and the logarithms lnGDPPC, ln2GDPPC, and ln3GDPPC of the independent variable per capita gross domestic product (GDPPC). The stationarity test results are shown in Table 2.

[0065] Table 3. Results of the stationarity test of the logarithmic series of per capita carbon emissions and per capita GDP Note: c, t, and p represent the intercept, time trend term, and lag number, respectively; Δ represents the first difference. Table 3 shows that at the 1%, 5%, and 10% significance levels, lnCEPC, lnGDPPC, ln 2 GDPPC, ln 3 GDPPC All sequences are first-order integrated sequences, satisfying the condition for integration of the same order, and can be tested for cointegration.

[0066] Cointegration and EKC test lnCEPC Dependent variable lnGDPPC, ln 2 GDPPC Using the least squares regression method with ar(1) as the independent variable, we add the ar(1) term to eliminate autocorrelation and obtain the cointegration equation: lnCEPC=18.09388*lnGDPPC-0.911661*ln2GDPPC-87.06907+[ar(1)=0.450405]; R 2 =0.883017; (Formula 4) In the formula, CEPC (Carbon Emission Per Capita) represents carbon emissions per capita. GDPPC (GDPPerCapita) is the per capita gross domestic product (GDP per capita), expressed in US dollars at constant prices in 2010, and ar(1) is the residual, which can be regarded as the observed value of the error.

[0067] The stationarity of the residual sequence of the cointegration equation (Equation 4) was tested, and the results are shown in Table 3. It can be concluded that the residual sequence is stationary at the 1% significance level.

[0068] Table 4. Results of stationarity test for difference sequences therefore lnPCCE and lnPCGDP, ln 2 PCGDP A cointegration relationship exists. From the cointegration equation, it can be deduced that Singapore's per capita carbon emissions and economic growth (per capita GDP) exhibit an inverted U-shaped curve relationship. That is, from 1960 to 2014, Singapore's per capita carbon emissions initially increased and then decreased along with the growth of per capita GDP. Figure 4 As shown. Therefore, Singapore meets all the principles for the per capita carbon emission benchmark threshold of counties in my country, presents the EKC curve pattern, and has the characteristics of being advanced, comparable, dynamic, similar and complete in data. It can be used as the dynamic benchmark threshold for per capita carbon emissions of counties at different development stages. Similarly, the EKC equation of Portugal is: lnCEPC=18.09388*lnGDPPC-0.911661*ln2GDPPC-87.0; 6907+[ar(1)=0.450405]; R2=0.883017; The Japanese EKC equation is: lnCEPC=18.09388*lnGDPPC-0.911661*ln2GDPPC-87.06907+[ar(1)=0.450405];R2=0.883017.

[0069] Please see Figure 6 As shown, Figure 6 This is a schematic diagram of the structure of the county-level per capita carbon emission dynamic threshold system construction system 200 provided in this application embodiment. The county-level per capita carbon emission dynamic threshold system construction system 200 is used to execute the steps of the county-level per capita carbon emission dynamic threshold system construction method shown in the above embodiments. The county-level per capita carbon emission dynamic threshold system construction system 200 can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, laptop computer, wearable device, or robot.

[0070] like Figure 6 As shown, the system for constructing a dynamic threshold system for per capita carbon emissions at the county level includes: Unit 201, which is used to determine the basic principles for selecting the reference standard for the dynamic threshold of per capita carbon emissions in counties, includes benchmarking against advanced standards, dynamism, regularity of the carbon dioxide environmental Kuznets curve, convergence of the carbon dioxide environmental Kuznets curve, and economic sensitivity. The standard selection unit 202 is used to select preset countries as low-level reference standards, medium-level reference standards and high-level reference standards according to the basic selection principles. The preset countries are all identified by the preset green evaluation tool as having excellent low-carbon performance, and their per capita carbon emissions and per capita GDP both show the carbon dioxide environmental Kuznets curve pattern. The threshold calculation unit 203 is used to collect historical data on per capita carbon emissions and per capita GDP of the preset country; based on the collected historical data, it constructs mathematical models of carbon dioxide environmental Kuznets curves corresponding to low, medium and high reference standards; according to the comparison results between the historical economic development level of the target county and the national average level, it matches the corresponding level of carbon dioxide environmental Kuznets curve mathematical model and calculates the dynamic threshold of per capita carbon emissions at different development stages of the target county.

[0071] In some embodiments, the basic principles for selecting the reference standard for determining the dynamic threshold of per capita carbon emissions in counties include: setting the benchmark for advanced performance by selecting countries or regions that have been identified as having excellent low-carbon performance by green assessment tools; setting the dynamic nature by adjusting the selected reference standard according to changes in economic development level and per capita income; setting the regularity of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of first increasing and then decreasing per capita carbon emissions with per capita GDP; setting the convergence of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of convergence of per capita carbon emissions with increasing income; and setting economic sensitivity by setting differentiated per capita carbon emission benchmark thresholds based on different income levels.

[0072] In some embodiments, the preset countries are all identified as having excellent low-carbon performance by a preset green assessment tool, and their per capita carbon emissions and per capita GDP both exhibit the CO2 Environmental Kuznets Curve pattern. This includes: selecting countries from a global list that have been identified as having excellent low-carbon performance by a preset green assessment tool; selecting countries with more than 50 consecutive years of time series data on per capita carbon emissions and per capita GDP; performing an existence test of the CO2 Environmental Kuznets Curve on the selected countries; and retaining the countries that pass the existence test as preset countries.

[0073] In some embodiments, the step of selecting preset countries as low-end, mid-end, and high-end reference standards based on basic selection principles includes: calculating the historical peak per capita GDP of all preset countries; sorting the preset countries from low to high according to their historical peak per capita GDP; selecting the first third of the sorted countries as a low-end reference standard preparation set; selecting the middle third of the sorted countries as a mid-end reference standard preparation set; selecting the last third of the sorted countries as a high-end reference standard preparation set; and selecting one country from each of the three preparation sets as the reference standard for the corresponding level.

[0074] In some embodiments, the collection of historical per capita carbon emission data and historical per capita GDP data for a preset country includes: obtaining annual per capita carbon emission data for more than fifty consecutive years from a preset publicly authoritative database; obtaining annual nominal per capita GDP data for the corresponding year from the same database; converting the nominal per capita GDP data into real data expressed in US dollars at constant prices for the preset year; and performing time series alignment processing on the converted data to ensure that the per capita carbon emission data and the per capita GDP data correspond one-to-one.

[0075] In some embodiments, constructing mathematical models of carbon dioxide environmental Kuznets curves corresponding to low-end, mid-end, and high-end reference standards based on collected historical data includes: performing logarithmic transformations on the collected per capita carbon emission data and per capita GDP data; performing a unit root stationarity test on the transformed data to verify whether the data is a first-order integrated sequence; performing a cointegration test on the data that passes the stationarity test; performing regression analysis on the cointegrated variables using the least squares method; adding a first-order autocorrelation term to eliminate the autocorrelation of the regression residuals; performing a stationarity test on the corrected regression residual sequence; and determining the regression equation as the corresponding carbon dioxide environmental Kuznets curve mathematical model if the residual sequence is stationary.

[0076] In some embodiments, matching the target county's historical economic development level with the national average level to the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model includes: obtaining the target county's annual per capita GDP data for the past twenty years; obtaining the national annual per capita GDP data for the same period; comparing the target county's per capita GDP with the national per capita GDP year by year; counting the number of years in which the target county's per capita GDP exceeded the national per capita GDP; and matching the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model based on the number of years counted.

[0077] In some embodiments, the calculation of the dynamic threshold for per capita carbon emissions at different development stages of the target county includes: obtaining the per capita GDP forecast data for each planning year of the target county; converting the per capita GDP forecast data into a value expressed in US dollars at constant prices for the preset year; substituting the converted value into the matched carbon dioxide environmental Kuznets curve mathematical model; calculating the dynamic threshold for per capita carbon emissions for the corresponding planning year of the target county; and organizing the calculation results in chronological order to form a sequence of dynamic thresholds for per capita carbon emissions of the target county.

[0078] In some embodiments, the method further includes: using the isolated forest algorithm to detect outliers in the collected preset national historical data; removing the detected outliers and supplementing the missing data using linear interpolation; using a long short-term memory network model to predict the future per capita GDP of the target county under multiple scenarios; generating per capita GDP prediction sequences under optimistic, neutral, and pessimistic scenarios; substituting the prediction sequences under the three scenarios into the matched mathematical model respectively; and calculating the dynamic threshold sequence of per capita carbon emissions of the target county under the corresponding three scenarios.

[0079] In some embodiments, the method further includes: collecting actual per capita carbon emission data and actual per capita GDP data of the target county on an annual basis; comparing the actual per capita carbon emission data with the dynamic threshold of per capita carbon emission for the corresponding year; calculating the deviation between the actual value and the threshold; using a deep reinforcement learning algorithm to dynamically optimize the parameters of the carbon dioxide environmental Kuznets curve mathematical model based on the deviation; using the optimized mathematical model to update the dynamic threshold of per capita carbon emission for subsequent years in the target county; and generating an annual threshold adjustment report containing the threshold update content and the basis for adjustment.

[0080] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the above-described system for constructing the dynamic threshold system for per capita carbon emissions in counties and the specific working process of each module can be referred to the corresponding content in the various embodiments of the above-described method for constructing the dynamic threshold system for per capita carbon emissions in counties, and will not be repeated here.

[0081] The aforementioned method for constructing a dynamic threshold system for per capita carbon emissions at the county level can be implemented as a computer program, which can be used in various ways, such as... Figure 6 It runs on the system shown.

[0082] Please see Figure 7 , Figure 7 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0083] The storage medium can store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any method for constructing a dynamic threshold system for per capita carbon emissions at the county level.

[0084] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0085] The internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to execute any method for constructing a dynamic threshold system for per capita carbon emissions in a county.

[0086] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0087] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0088] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: Compare the population size corresponding to the resource, environment and ecological carrying capacity with the predicted population value to determine the population prediction scenario where the predicted population value is less than the population size corresponding to the resource, environment and ecological carrying capacity; use the inventory method to calculate the total carbon emissions of the county, and combine the analysis and prediction of the total carbon emissions of the county with the carbon emission driving factors. Based on the determined population projection scenario and the predicted total carbon emissions in the county, calculate and predict the per capita carbon emissions in the county; construct a correlation model between per capita carbon emissions and per capita GDP, and calculate the international advanced level threshold for per capita carbon emissions under the corresponding per capita GDP. The predicted per capita carbon emissions in the county are compared with the international advanced level threshold for per capita carbon emissions. If the predicted value is less than the threshold, the scheme with the minimum per capita carbon emissions is selected. If the predicted value is greater than the threshold, the carbon emission driving parameters are adjusted in multiple rounds to improve the situation until the adjusted per capita carbon emissions are less than the threshold, thus obtaining the optimal solution for the county's low-carbon target.

[0089] In some embodiments, the basic principles for selecting the reference standard for determining the dynamic threshold of per capita carbon emissions in counties include: setting the benchmark for advanced performance by selecting countries or regions that have been identified as having excellent low-carbon performance by green assessment tools; setting the dynamic nature by adjusting the selected reference standard according to changes in economic development level and per capita income; setting the regularity of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of first increasing and then decreasing per capita carbon emissions with per capita GDP; setting the convergence of the carbon dioxide environmental Kuznets curve by selecting countries or regions that show a pattern of convergence of per capita carbon emissions with increasing income; and setting economic sensitivity by setting differentiated per capita carbon emission benchmark thresholds based on different income levels.

[0090] In some embodiments, the preset countries are all identified as having excellent low-carbon performance by a preset green assessment tool, and their per capita carbon emissions and per capita GDP both exhibit the CO2 Environmental Kuznets Curve pattern. This includes: selecting countries from a global list that have been identified as having excellent low-carbon performance by a preset green assessment tool; selecting countries with more than 50 consecutive years of time series data on per capita carbon emissions and per capita GDP; performing an existence test of the CO2 Environmental Kuznets Curve on the selected countries; and retaining the countries that pass the existence test as preset countries.

[0091] In some embodiments, the step of selecting preset countries as low-end, mid-end, and high-end reference standards based on basic selection principles includes: calculating the historical peak per capita GDP of all preset countries; sorting the preset countries from low to high according to their historical peak per capita GDP; selecting the first third of the sorted countries as a low-end reference standard preparation set; selecting the middle third of the sorted countries as a mid-end reference standard preparation set; selecting the last third of the sorted countries as a high-end reference standard preparation set; and selecting one country from each of the three preparation sets as the reference standard for the corresponding level.

[0092] In some embodiments, the collection of historical per capita carbon emission data and historical per capita GDP data for a preset country includes: obtaining annual per capita carbon emission data for more than fifty consecutive years from a preset publicly authoritative database; obtaining annual nominal per capita GDP data for the corresponding year from the same database; converting the nominal per capita GDP data into real data expressed in US dollars at constant prices for the preset year; and performing time series alignment processing on the converted data to ensure that the per capita carbon emission data and the per capita GDP data correspond one-to-one.

[0093] In some embodiments, constructing mathematical models of carbon dioxide environmental Kuznets curves corresponding to low-end, mid-end, and high-end reference standards based on collected historical data includes: performing logarithmic transformations on the collected per capita carbon emission data and per capita GDP data; performing a unit root stationarity test on the transformed data to verify whether the data is a first-order integrated sequence; performing a cointegration test on the data that passes the stationarity test; performing regression analysis on the cointegrated variables using the least squares method; adding a first-order autocorrelation term to eliminate the autocorrelation of the regression residuals; performing a stationarity test on the corrected regression residual sequence; and determining the regression equation as the corresponding carbon dioxide environmental Kuznets curve mathematical model if the residual sequence is stationary.

[0094] In some embodiments, matching the target county's historical economic development level with the national average level to the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model includes: obtaining the target county's annual per capita GDP data for the past twenty years; obtaining the national annual per capita GDP data for the same period; comparing the target county's per capita GDP with the national per capita GDP year by year; counting the number of years in which the target county's per capita GDP exceeded the national per capita GDP; and matching the corresponding level of the carbon dioxide environmental Kuznets curve mathematical model based on the number of years counted.

[0095] In some embodiments, the calculation of the dynamic threshold for per capita carbon emissions at different development stages of the target county includes: obtaining the per capita GDP forecast data for each planning year of the target county; converting the per capita GDP forecast data into a value expressed in US dollars at constant prices for the preset year; substituting the converted value into the matched carbon dioxide environmental Kuznets curve mathematical model; calculating the dynamic threshold for per capita carbon emissions for the corresponding planning year of the target county; and organizing the calculation results in chronological order to form a sequence of dynamic thresholds for per capita carbon emissions of the target county.

[0096] In some embodiments, the method further includes: using the isolated forest algorithm to detect outliers in the collected preset national historical data; removing the detected outliers and supplementing the missing data using linear interpolation; using a long short-term memory network model to predict the future per capita GDP of the target county under multiple scenarios; generating per capita GDP prediction sequences under optimistic, neutral, and pessimistic scenarios; substituting the prediction sequences under the three scenarios into the matched mathematical model respectively; and calculating the dynamic threshold sequence of per capita carbon emissions of the target county under the corresponding three scenarios.

[0097] In some embodiments, the method further includes: collecting actual per capita carbon emission data and actual per capita GDP data of the target county on an annual basis; comparing the actual per capita carbon emission data with the dynamic threshold of per capita carbon emission for the corresponding year; calculating the deviation between the actual value and the threshold; using a deep reinforcement learning algorithm to dynamically optimize the parameters of the carbon dioxide environmental Kuznets curve mathematical model based on the deviation; using the optimized mathematical model to update the dynamic threshold of per capita carbon emission for subsequent years in the target county; and generating an annual threshold adjustment report containing the threshold update content and the basis for adjustment.

[0098] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the steps of the method for constructing a dynamic threshold system for per capita carbon emissions in a county as provided in any embodiment of this application.

[0099] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0100] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for constructing a dynamic threshold system for per capita carbon emissions in a county, characterized in that, include: The basic principles for selecting reference standards for the dynamic threshold of per capita carbon emissions in counties include benchmarking against advanced standards, dynamism, regularity of the CO2 environmental Kuznets curve, convergence of the CO2 environmental Kuznets curve, and economic sensitivity. Based on the basic selection principles, preset countries were selected as low-end, mid-end, and high-end reference standards, respectively. All preset countries were identified by preset green evaluation tools as having excellent low-carbon performance, and their per capita carbon emissions and per capita GDP both showed the carbon dioxide environmental Kuznets curve pattern. Collect historical data on per capita carbon emissions and per capita GDP of the target country; construct carbon dioxide environmental Kuznets curve mathematical models corresponding to low, medium and high reference standards based on the collected historical data; and calculate the dynamic threshold of per capita carbon emissions at different development stages of the target county by matching the corresponding carbon dioxide environmental Kuznets curve mathematical models with the historical economic development level of the target county with the national average level.

2. The method according to claim 1, characterized in that, The basic principles for selecting the reference standard for determining the dynamic threshold of per capita carbon emissions in counties include: The benchmarking standard is set at selecting countries or regions that have been identified as having excellent low-carbon performance by green assessment tools; The dynamic nature is set so that the selected reference standards are adjusted according to changes in the level of economic development and per capita income; The carbon dioxide environmental Kuznets curve is defined as follows: the selected countries or regions show a pattern in which per capita carbon emissions first increase and then decrease with per capita GDP. The convergence of the carbon dioxide environmental Kuznets curve is defined as the convergence of per capita carbon emissions with increasing income in the selected countries or regions. Economic sensitivity is set as a reference standard to formulate differentiated per capita carbon emission benchmark thresholds based on different income levels.

3. The method according to claim 1, characterized in that, The pre-defined countries have all been identified as having excellent low-carbon performance using a pre-defined green assessment tool, and both their per capita carbon emissions and per capita GDP exhibit the CO2 Environmental Kuznets Curve pattern, including: Select countries from a global list that have been identified as having excellent low-carbon performance using a pre-defined green assessment tool; Select countries that have continuous time-series data on per capita carbon emissions and per capita GDP for more than fifty years; The existence test of the carbon dioxide environmental Kuznets curve was conducted on the selected countries. Countries that pass the existence test are retained as preset countries.

4. The method according to claim 1, characterized in that, The selection of preset countries based on the basic principles is as follows: These countries are designated as low-end, mid-end, and high-end reference standards, respectively. Calculate the historical peak GDP per capita for all preset countries; The preset countries are sorted from lowest to highest based on their historical peak GDP per capita. The top third of the countries after sorting were selected as a low-level reference set. The middle third of the countries after sorting were selected as a mid-range reference set. The last third of the countries after sorting were selected as a high-level reference set. One country from each of the three candidate sets was selected as the reference standard for the corresponding level.

5. The method according to claim 1, characterized in that, The collection of historical data on per capita carbon emissions and per capita GDP for the preset countries includes: Obtain annual per capita carbon emission data for a predetermined country for more than fifty consecutive years from a pre-defined, publicly authoritative database; Obtain annual nominal GDP per capita data for the corresponding year from the same database; Convert nominal GDP per capita data into real data expressed in constant US dollars for a predetermined year; The transformed data is aligned to time series to ensure that the per capita carbon emission data and per capita GDP data correspond one-to-one.

6. The method according to claim 1, characterized in that, The mathematical models of the carbon dioxide environmental Kuznets curves corresponding to low-grade, medium-grade, and high-grade reference standards, constructed based on collected historical data, include: Logarithmic transformations were performed on the collected per capita carbon emissions data and per capita GDP data, respectively. Perform a unit root stationarity test on the transformed data to verify whether the data is a first-order integrated sequence. Perform a cointegration test on the data that passed the stationarity test; The least squares method was used to perform regression analysis on the cointegration variables; Adding a first-order autocorrelation term eliminates the autocorrelation of the regression residuals; The stationarity of the corrected regression residual series is tested. If the residual sequence is stationary, then the regression equation is determined to be the corresponding mathematical model of the carbon dioxide environment Kuznets curve.

7. The method according to claim 1, characterized in that, The process involves matching the target county's historical economic development level with the national average, and then using a corresponding carbon dioxide environmental Kuznets curve mathematical model, including: Obtain annual per capita GDP data for the target county over the past two decades; Obtain the national annual per capita GDP data for the same period; Compare the per capita GDP of the target county with the national per capita GDP year by year; The number of years in which the per capita GDP of the target county exceeded the national per capita GDP; The mathematical model of the carbon dioxide environmental Kuznets curve is matched with the number of years obtained from statistics.

8. The method according to claim 7, characterized in that, The calculation yields dynamic thresholds for per capita carbon emissions at different development stages in the target county, including: Obtain the per capita GDP forecast data for each planning year in the target county; Convert the per capita GDP forecast data into a value expressed in constant US dollars for the preset year; Substitute the converted values ​​into the matching mathematical model of the carbon dioxide environment Kuznets curve. The dynamic threshold for per capita carbon emissions in the target county for the corresponding planning year was calculated. The calculation results are organized in chronological order to form a dynamic threshold sequence of per capita carbon emissions for the target county.

9. The method according to claim 1, characterized in that, The method further includes: The Isolation Forest algorithm is used to detect outliers in the collected historical data of preset countries. Detected outlier data is removed and missing data is supplemented using linear interpolation. A long short-term memory network model was used to predict the future per capita GDP of the target county in multiple scenarios. Generate per capita GDP forecast series under three scenarios: optimistic, neutral, and pessimistic. Substitute the predicted sequences under the three scenarios into the matching mathematical model respectively; The dynamic threshold sequence of per capita carbon emissions for the target county under the corresponding three scenarios was calculated.

10. The method according to claim 1, characterized in that, The method further includes: Collect actual per capita carbon emissions data and actual per capita GDP data for the target counties annually; Compare the actual per capita carbon emission data with the dynamic threshold for per capita carbon emissions in the corresponding year; Calculate the deviation between the actual value and the threshold; A deep reinforcement learning algorithm is used to dynamically optimize the parameters of the mathematical model of the carbon dioxide environment Kuznets curve based on the deviation value; The optimized mathematical model is used to update the dynamic threshold of per capita carbon emissions for subsequent years in the target county. Generate an annual threshold adjustment report that includes the updated threshold information and the basis for the adjustment.

Images