Regional carbon peak reaching prediction method and system based on system dynamics

Abstract

The invention provides a regional carbon peak reaching prediction method and system based on system dynamics, and relates to the technical field of carbon emission prediction. The method comprises the steps that data are collected and preprocessed; a causal feedback relation covering economy, science and technology, population, energy and industrial subsystems is constructed, and measurement and calculation are carried out based on system dynamics; setting a core influence factor to construct a benchmark, and synthesizing and strengthening three types of scenes; carbon emission is predicted, all indexes obtained through measurement and calculation are collected and serve as an input variable set, and the carbon emission is obtained through calculation according to the steps that the output is calculated through electricity and carbon is calculated through the output; carbon peak reaching and carbon neutralization years are predicted, a prediction broken line graph is drawn according to carbon emission prediction values of regions and industries in the next few years, and theoretical carbon peak reaching years and carbon neutralization years under different scenes are obtained. According to the method, the prediction precision of the carbon peak arrival time point is improved, and the method can adapt to the carbon emission change conditions of a city in different development stages.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission prediction, and in particular to a method and system for regional carbon peak prediction based on system dynamics. Background Art

[0002] As global attention to climate change and environmental protection increases, countries have pledged to achieve carbon neutrality goals. Achieving carbon peak and carbon neutrality is a broad and profound systemic change in the economic and social system. The current period is the critical period and window period for carbon peak, and it is urgent to carry out carbon emission monitoring, calculation, and forecasting. In this context, the carbon peak prediction model at the city level has become particularly important, especially in cities with a high proportion of industry and a coal-based energy structure. Accurately predicting the time points of carbon peak and carbon neutrality can not only help cities optimize their carbon emission control strategies, but also provide strong decision-making support for achieving carbon neutrality.

[0003] At present, there are two main types of models for carbon peak and carbon neutrality prediction tasks, namely complex system models and machine learning models. Among them, the complex system model has weak prediction function and only reflects the mathematical relationship between policy or management and carbon emissions, and cannot achieve prediction on a time scale. Machine learning models make predictions based on historical data, and changes in the policy environment often have strong uncertainty, which reduces the accuracy of predictions.

[0004] It can be seen from this that the existing models cannot take into account both theoretical paths and strong predictive accuracy, complex system models and machine learning models cannot be effectively combined, and cannot take into account both explainability and accuracy. The carbon peak prediction model has great limitations in practical applications and it is difficult to provide policymakers with accurate and real-time decision-making references. Summary of the invention

[0005] The purpose of an embodiment of the present invention is to provide a regional carbon peak prediction method based on system dynamics, which improves the prediction accuracy of the carbon peak time point and can adapt to changes in carbon emissions in cities at different development stages.

[0006] In order to achieve the above object, the present invention provides a method for predicting regional carbon peak based on system dynamics, which comprises:

[0007] Collect data and pre-process the data using missing value filling, base period chain recursive energy consumption calculation method, breakpoint test and segmented regression algorithm, and influencing factor decomposition method;

[0008] Construct causal feedback relationships covering economic, technological, population, energy and industrial subsystems, and measure them based on system dynamics;

[0009] By setting core influencing factors, three scenarios are constructed: baseline, comprehensive and enhanced;

[0010] Carbon emission prediction: collect various indicators obtained by measurement and use them as input variable sets, and calculate carbon emissions according to the steps of calculating output by electricity and carbon by output;

[0011] Predict the years of carbon peak and carbon neutrality, draw a forecast line chart based on the predicted carbon emissions of regions and industries in the next few years, and obtain the theoretical carbon peak year and carbon neutrality year under different scenarios.

[0012] Preferably, the data preprocessing using the base period chain recursive energy consumption calculation method includes: according to statistical rules, using a five-year iterative calculation method, using real GDP instead of GDP at current prices combined with regional unit GDP energy consumption data to calculate the total energy consumption of prefecture-level cities;

[0013] Use breakpoint test and segmented regression algorithm to segment the data before and after the known breakpoint to establish a regression model, use Zou test to identify whether the main parameters of the model have changed significantly, and use the backtracking method to simulate and replace the data before the caliber change to eliminate the caliber change factor and make the series comparable before and after; segment according to formula (1),

[0014]

[0015] Among them, year is the year value of the caliber change, β i is a parameter, t is the main time variable, ε t is a random disturbance term.

[0016] Preferably, a causal feedback relationship covering economic, technological, population, energy and industrial subsystems is constructed. Based on the system dynamics module, starting from the microscopic system structure, the logical relationship between the elements is described through loops. At the same time, cause-and-effect diagrams and stock-flow diagrams are used to qualitatively display the relationship between the elements, and the quantitative relationship is described through equations.

[0017] Preferably, constructing the causal feedback relationship covering the economic, technological, population, energy and industrial subsystems also includes: combining the bottom-up energy system model MARKAL model with the top-down macroeconomic model MACRO model to maximize the energy discount utility and simulate the carbon emission path by optimizing reserves, investment and consumption decisions; wherein the core formula of the MACRO model is the Cobb-Douglas production function, as shown in formula (2),

[0018] Y=A·K ∝ ·L β ·E γ (2)

[0019] Among them, Y is the gross domestic product (GDP), A is the coefficient of technological progress, and total factor productivity (TFP); K is capital input, L is labor input, E is energy input, ∝, β, and γ are the output elasticities of capital, labor, and energy, respectively. If the sum of the coefficients is equal to 1, it indicates that the scale returns are constant;

[0020] The objective function of the MACRO model is to maximize the social utility U. Formula (3) is used as the utility function.

[0021]

[0022] Where C is consumption, θ is the risk aversion coefficient, ρ is the discount rate, t0 and T are the initial time and the end time respectively;

[0023] According to formula (4), the MARKAL model is used to determine the energy discounted utility maximization problem by minimizing the cost of the energy system.

[0024]

[0025] Where Z is the total cost, C t is the energy system cost in each year t, including investment, operation, maintenance, fuel and emission costs, and r is the discount rate;

[0026] The MARKAL model and the MACRO model are coupled, and the results of the energy system are fed back to the economic system according to formula (5). In addition, the economic variables affect the parameters of the energy system.

[0027]

[0028] Among them, C t is the energy system cost, U t It is the utility of the economic system.

[0029] Preferably, three scenarios are constructed by setting core influencing factors: baseline, comprehensive and enhanced, including:

[0030] Construct a baseline scenario to predict future carbon emissions based on historical economic, population, energy consumption structure and energy intensity trends;

[0031] Construct a comprehensive scenario based on national and local economic, energy and carbon emission policies, set some weight constraints, and predict future carbon emissions;

[0032] Construct an enhanced scenario based on national and local policies, set stricter constraints than the comprehensive scenario, and predict more conservative carbon emissions.

[0033] Preferably, the carbon emission forecast includes:

[0034] Calculate output based on electricity, train data such as electricity consumption, GDP, clean energy share, population, and energy consumption and industrial output data to build a regression analysis model, and extrapolate annual energy consumption and industrial output data year by year through the annual electricity data predicted under different scenarios;

[0035] Calculate carbon based on output and multiply energy consumption and industrial output data by relevant carbon emission factors to obtain carbon emissions.

[0036] Preferably, an electricity-carbon calculation model is constructed according to formula (6) to predict carbon emissions. The model adopts the autoregressive distributed lag model SARDL-ECM and introduces a seasonal fluctuation adjustment parameter and ECM as the training parameters of the model,

[0037]

[0038] in, Adjust the parameters for seasonal fluctuations, y t-i,r is the autoregressive term, x t-i,r is the lag term, z t-i,r is the adjustment parameter, u t,r is the random walk term, δECM t-1,r is the error correction term.

[0039] Another aspect of the present invention provides a regional carbon peak prediction system based on system dynamics, which runs the regional carbon peak prediction method based on system dynamics as described above. The regional carbon peak prediction system based on system dynamics includes:

[0040] Data collection and processing module, used to obtain historical data of the target area;

[0041] Complex system module, used to construct the cause-effect feedback relationship between economic, technological, population, energy and industrial subsystems;

[0042] The scenario setting module is used to design constraint variables according to the carbon peak target and set baseline, comprehensive and enhanced scenarios;

[0043] The carbon emission prediction module is used to predict the peak of carbon emissions based on the electricity-carbon calculation model.

[0044] Through the above technical scheme, the present invention constructs the causal feedback relationship between multiple subsystems of economy, energy, population, science and technology, and industry, so that the model can fully reflect the complex dynamic process that affects carbon emissions. Compared with the traditional single factor prediction method, the model can more accurately simulate the carbon emission path of prefecture-level cities under different scenarios, and improve the accuracy and credibility of the prediction results. At the same time, by setting different constraints and assumed variables, it is possible to flexibly respond to the carbon emission reduction needs under different stages and different policy objectives. This diversified scenario setting makes the model highly adaptable, can provide multi-angle reference for policy making, and help formulate more targeted and forward-looking emission reduction strategies. In addition, the present invention adopts a system dynamics modeling method, which can dynamically simulate the long-term evolution trend of carbon emissions, and optimize the feedback relationship between energy consumption and macroeconomics in combination with the MARKAL-MACRO model, so that the model can not only support long-term carbon peak and carbon neutrality planning, but also provide flexible prediction results according to short-term economic or energy policy adjustments, and provide decision support for policy makers.

[0045] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0047] Figure 1 It is an overall framework diagram of the carbon peak prediction model in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0048] Figure 2 It is a schematic diagram of the economic subsystem constructed in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0049] Figure 3 It is a schematic diagram of the scientific and technological subsystem constructed in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0050] Figure 4 It is a schematic diagram of a population subsystem constructed in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0051] Figure 5 It is a schematic diagram of the energy subsystem constructed in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0052] Figure 6It is a schematic diagram of an industrial subsystem constructed in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0053] Figure 7 It is the energy consumption carbon emission system indicator map;

[0054] Figure 8 It is the logic diagram of the complex system of carbon emission prediction;

[0055] Fig. 9 It is a schematic diagram of three types of scenarios constructed by influencing factors in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0056] Fig.10 It is a framework diagram of the electricity-carbon calculation model in the regional carbon peak prediction method based on system dynamics provided by the present invention;

[0057] Fig.11 This is a graph of the carbon peak prediction results for a certain place. DETAILED DESCRIPTION

[0058] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0059] See also Figure 1 The present invention provides a regional carbon peak prediction method based on system dynamics, the method comprising:

[0060] Step 1: Collect data, gather electricity-related data through data reporting, interface integration, and data dedicated lines, obtain guidance indicators and emission factor data through data authorization purchase and public channel download, and obtain industrial enterprises' gas consumption, heating, new energy power generation, output value, and output through research and investigation;

[0061] Step 2: preprocess the data by using missing value filling, chain recursive energy consumption calculation method, breakpoint test and segmented regression algorithm and influencing factor decomposition method to preprocess the data obtained in step 1;

[0062] Step 3: Construct causal feedback relationships covering economic, technological, population, energy and industrial subsystems, and calculate them based on system dynamics;

[0063] Step 4: Construct three scenarios: baseline, comprehensive and enhanced by setting core influencing factors;

[0064] Step 5: Carbon emission prediction: collect the various indicators calculated in steps 3 and 4 as the input variable set, and calculate the carbon emissions according to the steps of calculating output by electricity and calculating carbon by output;

[0065] Step 6, predict the years of carbon peak and carbon neutrality. After completing the model training and prediction with reference to steps 1 to 5, draw a prediction line chart based on the predicted carbon emissions values ​​of regions and industries in the next few years to obtain the theoretical carbon peak year and carbon neutrality year under different scenarios.

[0066] Specifically, the data source in step 1 can be macro data collected from national, provincial, and municipal statistical yearbooks, or data from power grid companies, report data from provincial and municipal statistical bureau websites, and supplementary economic data collected from national data websites.

[0067] Since the above data may be missing or incomparable due to changes in statistical caliber, it is necessary to process the data. Therefore, in step 2, missing value filling, base period chain recursive energy consumption calculation method, breakpoint test and segmented regression algorithm and influencing factor decomposition method are used to preprocess the data obtained in step 1;

[0068] The variable base period chain recursive energy consumption calculation method adopts an iterative calculation method every five years according to statistical rules, uses actual GDP instead of GDP at the current price, and combines regional unit GDP energy consumption data to calculate the total energy consumption of prefecture-level cities, so as to achieve consistency in the total energy consumption at the provincial and municipal levels and truthfully reflect the growth of output.

[0069] The breakpoint test and segmented regression algorithm establishes a regression model by segmenting the data before and after the known breakpoint, uses the Zou test to identify whether the main parameters of the model have changed significantly, and uses the backtracking method to simulate and replace the data before the caliber change to eliminate the caliber change factor, making the series comparable before and after, and solving the problem that the statistical caliber change directly affects the quality of the input data. The piecewise function formula is:

[0070]

[0071] Among them, year is the year value of the caliber change, β i is a parameter, t is the main time variable, ε t is a random disturbance term. i The historical data is used as the input of the model for training and determination. The model will train these parameters based on the least squares objective function, which is the square of the deviation between the set equation curve and the true value, in order to minimize the objective function.

[0072] In step 3, we build a causal feedback relationship covering the economic, technological, population, energy and industrial subsystems. Based on the system dynamics module, we start from the micro-system structure and describe the logical relationship between the elements through loops. At the same time, we use causal relationship diagrams and stock-flow diagrams to qualitatively display the relationship between the elements, and use equations to describe the quantitative relationship. We use system dynamics simulation tools to predict the carbon emission trends and peak times of various cities. Specifically:

[0073] Since the economic system directly affects electricity consumption and GDP, the economic subsystem is established based on the output value of the three major industries and the number of employees (see Figure 2 ), and add supplementary variables such as the output value of high-tech products to make the system more complete and three-dimensional:

[0074] GDP = GDP of the previous year × (GDP growth rate + 1) + output value of high-tech industries × β1;

[0075] Output value of primary industry = GDP × proportion of output value of primary industry;

[0076] Output value of secondary industry = GDP × proportion of output value of secondary industry;

[0077] Output value of tertiary industry = GDP × output value proportion of tertiary industry;

[0078] Electricity consumption of primary industry = f(output value of primary industry, number of employees in primary industry);

[0079] Electricity consumption of secondary industry = f (output value of secondary industry, number of employees in secondary industry);

[0080] Electricity consumption of the tertiary industry = f (output value of the tertiary industry, number of employees in the tertiary industry).

[0081] Technology Subsystem (see Figure 3 ) Promote economic growth, establish an objective function through R&D funds, number of patents, and number of R&D personnel, and calculate the output value of high-tech:

[0082] R&D expenditure = R&D expenditure in the previous year × (1 + R&D investment growth rate);

[0083] Number of patent applications = R&D expenditure × β1 + ε1;

[0084] Number of R&D personnel = number of R&D personnel in the previous year × (1 + growth rate of R&D personnel);

[0085] High-tech output value = f (number of patent applications, number of R&D personnel).

[0086] Population subsystem (see Figure 4 ) will not only affect the economic system, but also the electricity consumption of residents. The equation of the population subsystem is constructed by using the key indicators of the resident population and the number of employees in the three major industries, as well as per capita disposable income, resident population, and urbanization rate as variables:

[0087] Per capita disposable income = per capita disposable income of the previous year × (per capita income growth rate + 1);

[0088] Permanent population = permanent population of the previous year × (permanent population growth rate + 1);

[0089] Residential electricity consumption = f(per capita disposable income, permanent population, urbanization rate);

[0090] Employees = f(resident population, per capita disposable income);

[0091] Number of people in the primary industry = number of employees × β1 + ε1;

[0092] Number of people in the secondary industry = number of employees × β2 + ε2;

[0093] The number of people in the tertiary industry = number of employees × β3 + ε3.

[0094] Energy subsystem (see Figure 5 ) describes the changes in the production and consumption of different types of energy, and focuses on the consumption of various energy sources. The equation for calculating the total energy consumption through key indicators such as total social electricity consumption, GDP, proportion of renewable energy, and GDP energy intensity is:

[0095] Total energy consumption = f (total electricity consumption, GDP, renewable energy share, GDP energy intensity);

[0096] Crude oil consumption = total energy consumption × crude oil consumption share;

[0097] Coal consumption = total energy consumption × coal consumption proportion;

[0098] Natural gas consumption = total energy consumption × natural gas consumption share;

[0099] Carbon emissions from energy consumption = crude oil consumption × β1 + ε1 + coal consumption × β2 + ε2 + natural gas consumption × β3 + ε3.

[0100] Focusing on electricity consumption and carbon emissions, we use the industrial output value and electricity consumption ratio of different industries, and the output of products in different industries to establish a mapping relationship and build an industrial subsystem (see Figure 6 ):

[0101] Total industrial output value above designated size = output value of secondary industry × β1 + ε1;

[0102] Heavy industry output value = industrial output value above designated size × β2 + ε2;

[0103] Electricity consumption of the steel industry = f (heavy industry output value, electricity consumption of the secondary industry);

[0104] Aluminum industry electricity consumption = f (heavy industry output value, secondary industry electricity consumption);

[0105] Electricity consumption of cement industry = f(heavy industry output value, electricity consumption of secondary industry);

[0106] Carbon emissions from industrial processes = pig iron production × β3 + steel production × β4 + electrolytic aluminum production × β5 + cement production × β6.

[0107] The carbon emission prediction module is affected by the combined effects of the above subsystems and their internal elements, such as Figure 7 As shown, it shows the main indicators of the carbon emission peak prediction system for prefecture-level cities and explains the energy consumption and carbon emissions within the system.

[0108] At the same time, the method provided by the present invention also integrates the MARKAL-MACRO model, combining the bottom-up energy system model MARKAL model with the top-down macroeconomic model MACRO model to maximize the energy discount utility and simulate the carbon emission path by optimizing reserves, investment and consumption decisions; wherein, the core formula of the MACRO model is the Cobb-Douglas production function:

[0109] Y=A·K ∝ ·L β ·E γ

[0110] Among them, Y is gross domestic product (GDP), A is the coefficient of technological progress, and total factor productivity (TFP); K is capital input, L is labor input, E is energy input, ∝, β, and γ are the output elasticities of capital, labor, and energy, respectively (the sum of the coefficients is equal to 1, indicating that the scale returns remain constant);

[0111] The objective function of the MACRO model is to maximize the social utility U, using the utility function:

[0112]

[0113] Where C is consumption, θ is the risk aversion coefficient, ρ is the discount rate, t0 and T are the initial time and the end time respectively;

[0114] The MARKAL model determines the problem of maximizing the discounted energy utility by minimizing the cost of the energy system:

[0115]

[0116] Where Z is the total cost, C t is the energy system cost in each year t, including investment, operation, maintenance, fuel and emission costs, and r is the discount rate;

[0117] The coupling of the MARKAL model and the MACRO model feeds back the results of the energy system to the economic system and affects the parameters of the energy system through economic variables. The energy price and energy demand generated by the MARKAL model affect the production function in the MACRO model, and the economic growth in the MACRO model reacts to the energy consumption in the MARKAL model:

[0118]

[0119] Among them, C t is the energy system cost, U t It is the utility of the economic system.

[0120] In step 4, by setting core influencing factors, three scenarios, namely baseline, comprehensive and enhanced, are constructed to predict the peak value and time of carbon peak and the time to achieve carbon neutrality (such as Fig. 9 ). Among them,

[0121] The baseline scenario predicts future carbon emissions based on the historical economic, population, energy consumption structure and energy intensity trends;

[0122] The comprehensive scenario is based on national and local economic, energy and carbon emission-related policies, sets some weight constraints, and predicts future carbon emission scenarios;

[0123] The enhanced scenario is based on relevant national and local policies, sets stricter constraints than the comprehensive scenario, and predicts a more conservative carbon emissions scenario.

[0124] In step 5, the indicators calculated in steps 3 and 4 are collected as the electricity-carbon calculation model (see Fig.10 ) is used as an input variable set, and the carbon emissions are calculated by following the steps of calculating output by electricity and calculating carbon by output, where:

[0125] Calculate output based on electricity: Train data such as electricity consumption, GDP, clean energy share, population, and energy consumption and industrial output data to build a regression analysis model, and extrapolate annual energy consumption and industrial output data year by year based on the annual electricity data predicted under different scenarios;

[0126] Calculate carbon based on output: Multiply energy consumption and industrial output data by relevant carbon emission factors to obtain carbon emissions.

[0127] The electricity-carbon calculation model constructed in step 5 adopts the autoregressive distributed lag model SARDL-ECM. Based on the ARDL algorithm, the seasonal fluctuation adjustment parameter is introduced. As one of the training parameters of the model, it can better capture the trend changes of variables, remove the influence of seasonal cycle factors, and reduce the risk of drift. At the same time, the ECM (Error Correction Model) parameter is introduced to solve the multicollinearity problem that may exist in the input data of the electric energy (output) part. The specific formula is:

[0128]

[0129] in, Adjust the parameters for seasonal fluctuations, y t-i,r is the autoregressive term, x t-i,r is the lag term, z t-i,r is the adjustment parameter, u t,r is the random walk term, δECM t-1,r is the error correction term.

[0130] In step 6, after completing the model training and prediction with reference to steps 1 to 5, a forecast line chart is drawn based on the predicted carbon emissions of regions and industries in the next few years to obtain the theoretical carbon peak years and carbon neutral years under different scenarios (see Fig.11 ).

[0131] It can be seen that the present invention builds a causal feedback relationship between multiple subsystems such as economy, energy, population, technology and industry, so that the model can fully reflect the complex dynamic process that affects carbon emissions. Compared with the traditional single-factor prediction method, the model can more accurately simulate the carbon emission path of prefecture-level cities under different scenarios, improving the accuracy and credibility of the prediction results.

[0132] At the same time, by setting different constraints and hypothetical variables, it is possible to flexibly respond to carbon emission reduction needs at different stages and under different policy objectives. This diverse scenario setting makes the model highly adaptable, providing multi-angle references for policy making and helping to formulate more targeted and forward-looking emission reduction strategies.

[0133] In addition, the present invention adopts a system dynamics modeling method, which can dynamically simulate the long-term evolution trend of carbon emissions, and optimize the feedback relationship between energy consumption and the macroeconomy in combination with the MARKAL-MACRO model, so that the model can not only support long-term carbon peak and carbon neutrality planning, but also provide flexible forecasting results based on short-term economic or energy policy adjustments, providing decision-making support for policy makers.

[0134] Furthermore, the model provided by this method not only takes into account key factors such as macroeconomics, energy consumption, and population structure, but also includes specific indicators such as energy intensity and technological progress. These indicators influence each other through logical relationships within the system, so that the prediction results of the model can not only reflect the overall trend, but also explain the reasons behind the changes in carbon emissions through changes in specific indicators, which helps to improve the interpretability and transparency of the prediction results.

[0135] In addition, the present invention also provides a regional carbon peak prediction system based on system dynamics, which runs the regional carbon peak prediction method based on system dynamics as described above. The regional carbon peak prediction system based on system dynamics includes:

[0136] Data collection and processing module, used to obtain historical data of the target area;

[0137] Complex system module, used to construct the cause-effect feedback relationship between economic, technological, population, energy and industrial subsystems;

[0138] The scenario setting module is used to design constraint variables according to the carbon peak target and set baseline, comprehensive and enhanced scenarios;

[0139] The carbon emission prediction module is used to predict the peak of carbon emissions based on the electricity-carbon calculation model.

[0140] In this way, combined with system dynamics simulation technology, it is possible to accurately simulate complex carbon emission systems through computer programs. The introduction of this simulation technology not only verifies the feasibility of the model, but also ensures the operability of the model in practical applications. Through computer simulation, prediction results under different scenarios can be quickly obtained, providing a scientific basis for policy adjustments and carbon emission reduction path planning.

[0141] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0142] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0143] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0144] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0145] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0146] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0147] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0148] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0149] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A regional carbon peak prediction method based on system dynamics, characterized in that: The method comprises: Collect data and pre-process the data using missing value filling, base period chain recursive energy consumption calculation method, breakpoint test and segmented regression algorithm, and influencing factor decomposition method; Construct causal feedback relationships covering economic, technological, population, energy and industrial subsystems, and measure them based on system dynamics; By setting core influencing factors, three scenarios are constructed: baseline, comprehensive and enhanced; Carbon emission prediction: collect various indicators obtained by measurement and use them as input variable sets, and calculate carbon emissions according to the steps of calculating output by electricity and carbon by output; Predict the years of carbon peak and carbon neutrality, draw a forecast line chart based on the predicted carbon emissions of regions and industries in the next few years, and obtain the theoretical carbon peak year and carbon neutrality year under different scenarios.

2. The regional carbon peak prediction method based on system dynamics according to claim 1 is characterized in that: Data preprocessing using the variable base period chain recursive energy consumption calculation method includes: according to statistical rules, using the method of iterative calculation every five years, using real GDP instead of GDP at current prices combined with regional unit GDP energy consumption data to calculate the total energy consumption of prefecture-level cities; Use breakpoint test and segmented regression algorithm to segment the data before and after the known breakpoint to establish a regression model, use Zou test to identify whether the main parameters of the model have changed significantly, and use the backtracking method to simulate and replace the data before the caliber change to eliminate the caliber change factor and make the series comparable before and after; segment according to formula (1), Among them, year is the year value of the caliber change, β i is a parameter, t is the main time variable, ε t is a random disturbance term.

3. The regional carbon peak prediction method based on system dynamics according to claim 1 is characterized in that: Construct a causal feedback relationship covering economic, technological, population, energy and industrial subsystems. Based on the system dynamics module, start from the micro-system structure and describe the logical relationship between the elements through loops. At the same time, use causal relationship diagrams and stock-flow diagrams to qualitatively display the relationship between the elements, and describe the quantitative relationship through equations.

4. The regional carbon peak prediction method based on system dynamics according to claim 3 is characterized in that: Constructing the causal feedback relationship covering the economic, technological, population, energy and industrial subsystems also includes: combining the bottom-up energy system model MARKAL model with the top-down macroeconomic model MACRO model to maximize the energy discount utility and simulate the carbon emission path by optimizing reserves, investment and consumption decisions; the core formula of the MACRO model is the Cobb-Douglas production function, as shown in formula (2), Y=A·K ∝ ·L β ·AND γ (2) Among them, Y is the gross domestic product (GDP), A is the coefficient of technological progress, and total factor productivity (TFP); K is capital input, L is labor input, E is energy input, ∝, β, and γ are the output elasticities of capital, labor, and energy, respectively. If the sum of the coefficients is equal to 1, it indicates that the scale returns are constant; The objective function of the MACRO model is to maximize the social utility U. Formula (3) is used as the utility function. Where C is consumption, θ is the risk aversion coefficient, ρ is the discount rate, t0 and T are the initial time and the end time respectively; According to formula (4), the MARKAL model is used to determine the energy discounted utility maximization problem by minimizing the cost of the energy system. Where Z is the total cost, C t is the energy system cost in each year t, including investment, operation, maintenance, fuel and emission costs, and r is the discount rate; The MARKAL model and the MACRO model are coupled, and the results of the energy system are fed back to the economic system according to formula (5). In addition, the economic variables affect the parameters of the energy system. Among them, C t is the energy system cost, U t It is the utility of the economic system.

5. The regional carbon peak prediction method based on system dynamics according to claim 1 is characterized in that: By setting core influencing factors, three scenarios are constructed: baseline, comprehensive and enhanced, including: Construct a baseline scenario to predict future carbon emissions based on historical economic, population, energy consumption structure and energy intensity trends; Construct a comprehensive scenario based on national and local economic, energy and carbon emission policies, set some weight constraints, and predict future carbon emissions; Construct an enhanced scenario based on national and local policies, set stricter constraints than the comprehensive scenario, and predict more conservative carbon emissions.

6. The regional carbon peak prediction method based on system dynamics according to claim 1 is characterized in that: Carbon emissions forecasts include: Calculate output based on electricity, train data such as electricity consumption, GDP, clean energy share, population, and energy consumption and industrial output data to build a regression analysis model, and extrapolate annual energy consumption and industrial output data year by year through the annual electricity data predicted under different scenarios; Calculate carbon based on output and multiply energy consumption and industrial output data by relevant carbon emission factors to obtain carbon emissions.

7. The regional carbon peak prediction method based on system dynamics according to claim 6 is characterized in that: According to formula (6), an electricity-carbon calculation model is constructed to predict carbon emissions. The model adopts the autoregressive distributed lag model SARDL-ECM and introduces a seasonal fluctuation adjustment parameter and ECM as the training parameters of the model, in, Adjust the parameters for seasonal fluctuations, y t-i,r is the autoregressive term, x t-i,r is the lag term, z t-i,r is the adjustment parameter, u t,r is the random walk term, δECM t-1,r is the error correction term.

8. A regional carbon peak prediction system based on system dynamics, running the regional carbon peak prediction method based on system dynamics as described in any one of claims 1 to 7, characterized in that: The regional carbon peak prediction system based on system dynamics includes: Data collection and processing module, used to obtain historical data of the target area; Complex system module, used to construct the cause-effect feedback relationship between economic, technological, population, energy and industrial subsystems; The scenario setting module is used to design constraint variables according to the carbon peak target and set baseline, comprehensive and enhanced scenarios; The carbon emission prediction module is used to predict the peak of carbon emissions based on the electricity-carbon calculation model.

Images