A method for dynamic simulation and optimization decision of low-carbon policy in industrial park based on energy-carbon-economy synergy

Abstract

The present application relates to a kind of industrial park low carbon policy dynamic simulation and optimization decision method based on energy-carbon-economy coordination, it is related to industrial park low carbon policy formulation and evaluation technical field, including step park boundary identification and carbon emission responsibility unit division, park multi-source data acquisition and energy-carbon-economy coordination database construction, construct energy flow-carbon flow-economic flow dynamic coupling simulation model, construct energy-carbon-economy coordination evaluation model, construct policy net economic effect model, establish policy tool variable set and policy scenario library, construct multi-objective optimization model and solve Pareto optimal front etc..The present application establishes multi-dimensional coordinated evaluation index system and the whole life cycle policy net economic effect accounting model including system cost of power grid, combined with multi-objective optimization algorithm and policy sandbox intelligent recommendation mechanism, realize the whole process dynamic simulation of industrial park low carbon policy, multi-dimensional effect evaluation, multi-constraint compliance verification and global optimization decision.

Description

Technical Field

[0001] This invention relates to the field of low-carbon policy formulation and evaluation technology for industrial parks, and in particular to a dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy. Background Technology

[0002] Industrial parks, as core carriers of industrial agglomeration, energy consumption, and carbon emissions, are key scenarios for implementing regional green and low-carbon transformation. However, the current technical system for formulating and evaluating low-carbon policies for industrial parks still has significant shortcomings, specifically in the following aspects:

[0003] Firstly, most existing technologies focus on single indicators such as energy intensity, carbon emission intensity, or corporate electricity consumption for policy evaluation, lacking a unified characterization and quantification of the synergistic relationship between energy efficiency, carbon reduction effectiveness, and changes in economic benefits. For example, in the article "Does China's low-carbon city pilot intervention limit electricity consumption? An analysis of industrial energy efficiency using time-varying DID mode" in Volume 121 of *Energy Economics* in 2023, some scholars used corporate electricity consumption intensity as a starting point to assess the effectiveness of carbon reduction policies, but failed to incorporate economic impacts into a unified evaluation framework, nor did they conduct in-depth analysis of the actual impact of electricity consumption reduction on corporate profit levels and market competitiveness, resulting in a lack of dimensions in the evaluation system.

[0004] Secondly, existing policy evaluation methods are mostly static analysis or ex-post statistics, which make it difficult to accurately depict the temporal evolution of energy flow, carbon flow and economic flow in the park under the influence of policies. They are unable to effectively identify the lag effect, substitution effect and rebound effect in policy implementation, and it is difficult to quantify the real marginal effect brought about by policy intervention. As a result, the evaluation results deviate significantly from the actual operation of the policies.

[0005] Furthermore, existing methods have significant limitations in analyzing the economic effects of low-carbon policies, particularly lacking quantitative accounting for the additional system-side costs caused by renewable energy replacing traditional energy, including peak-shaving costs, reserve capacity costs, grid upgrading costs, and energy storage support costs. They only focus on the direct cost-benefit of policy implementation, ignoring the systemic implicit costs on the grid side, resulting in significant discrepancies between the assessment results of the policy's economic feasibility and the actual engineering situation.

[0006] At the same time, existing technologies are insufficient in their ability to combine and optimize different policy tools. They are unable to output the optimal policy implementation path that can be implemented based on the local characteristics of the park's industrial structure, energy structure, resource endowment, fiscal budget and grid absorption constraints. They cannot achieve precise policy design of "one policy for each type and one policy for each park". Most of them remain at the level of ex-post policy evaluation and cannot achieve proactive optimization and generation of policy solutions.

[0007] Finally, existing research also has significant technical limitations. In the aforementioned *Energy Economics* paper, scholars constructed a natural experimental scenario for low-carbon city pilot policies and used a time-varying difference-in-differences model to test the policy's effectiveness, confirming that low-carbon pilot policies can promote electricity reduction for enterprises through a dual path of investment in energy-saving equipment and green technology innovation. However, this research still has significant technical limitations: the economic impact is only mentioned as background information and is not included in a unified evaluation framework for policy effectiveness. It also fails to deeply analyze the transmission impact of policy-induced changes in electricity consumption on corporate profits and economic competitiveness. Furthermore, it lacks in-depth research on optimizing policy mix effects and designing policies for local adaptation, thus failing to provide full-process decision support for the formulation of low-carbon policies in industrial parks. Summary of the Invention

[0008] (a) The technical problem to be solved:

[0009] In view of the above-mentioned shortcomings and deficiencies of the existing technology, the present invention provides a dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy. By constructing a coupled dynamic model of energy flow, carbon flow and economic flow in industrial parks, a multi-dimensional collaborative evaluation index system and a full life-cycle policy net economic effect accounting model including the systemic cost of the power grid are established. Combined with multi-objective optimization algorithms and policy sandbox intelligent recommendation mechanisms, the method realizes full-process dynamic simulation, multi-dimensional effect evaluation, multi-constraint compliance verification and global optimization decision-making for low-carbon policies in industrial parks.

[0010] (II) Technical Solution:

[0011] To achieve the above objectives, the main technical solution adopted in this invention is a dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy, comprising the following steps:

[0012] Step 1, Park Boundary Identification and Carbon Emission Responsibility Unit Delineation: Through boundary delineation algorithms, the park's boundaries are accurately delineated in multiple dimensions to clarify the park's carbon emission accounting boundaries and responsibility attribution;

[0013] Step 2: Multi-source data collection and construction of energy-carbon-economy collaborative database in the park: Collect core micro data of the park from all dimensions, clean, correct, denoise, standardize and unify the time stamp of the collected raw data, remove invalid data and outliers, and construct the basic database of energy-carbon-economy collaborative database of the park through the fusion of multi-source heterogeneous data;

[0014] Step 3: Construct a dynamic coupling simulation model of energy flow, carbon flow, and economic flow: Through the full-dimensional micro-core data of the park's energy-carbon-economic synergy basic database, track the temporal evolution of energy flow, carbon flow, and economic flow in the park, construct a dynamic simulation model of policies, and quantify the marginal changes brought about by policy implementation;

[0015] Step 4: Construct an energy-carbon-economy synergistic evaluation model: Perform dimensionless processing on energy, carbon, and economic indicators under different policy scenarios, and obtain the park synergy index by weighted fusion calculation of weight coefficients to form a standardized policy comprehensive performance evaluation system;

[0016] Step 5: Construct a policy net economic effect model: Define the accounting boundaries and scope of the direct costs, indirect benefits and systemic costs of policy implementation. Based on the policy quantification model and the energy substitution elasticity model, introduce the full life cycle economic cost-benefit analysis method to quantify the comprehensive economic effect of the marginal substitution rate of renewable energy for thermal power, construct a policy net economic effect accounting model, and evaluate the economic feasibility of a single policy or a combination of policies.

[0017] Step 6: Establish a set of policy instrument variables and a policy scenario library: Based on the design requirements of the park's low-carbon policy, set policy instrument variables and clarify their reasonable value ranges, construct a standardized policy scenario library, and classify the data collected in Step 2 according to the policy scenario library to provide a standardized scenario basis for subsequent simulation analysis and optimization solutions;

[0018] Step 7: Construct a multi-objective optimization model to solve the Pareto optimal frontier: Integrate multi-objective optimization algorithms, simulate the Pareto optimal frontier under different policy combinations, and set multi-dimensional constraints to find the global optimal balance point among the three major objectives of improving energy efficiency, achieving carbon emission reduction targets, and stable economic growth.

[0019] Step 8: Establish a policy sandbox and intelligent recommendation mechanism: Develop a policy sandbox simulation environment that allows users to set their own constraints. Through machine learning prediction engines and reinforcement learning algorithms, it automatically recommends the optimal carbon reduction path that meets all constraints.

[0020] Step 9: Visualization of Results and Interpretation of Decisions: Based on the full-process results of policy simulation in Step 6, multi-objective optimization in Step 7, and intelligent recommendation in Step 8, generate multi-dimensional visualization outputs to intuitively display the comprehensive performance of policy solutions in the three dimensions of energy, carbon, and economy, and mark the deviation of each indicator from the baseline scenario.

[0021] The core data in step 2 includes enterprise energy consumption and cost data, distributed photovoltaic / wind power generation operation data, real-time carbon emission factor database of the power grid, enterprise financial statement data, and policy-related parameter data.

[0022] Step 3 specifically involves constructing three sub-models based on the park's energy-carbon-economic synergy database. These include an energy flow model that describes the entire process of energy purchase, conversion, transmission, distribution, consumption, and substitution in the park; a carbon flow model that depicts the mapping relationship between energy consumption and carbon emissions in the park, as well as the emission reductions generated by low-carbon substitution measures; and an economic flow model that quantifies the dynamic changes in investment, costs, benefits, profits, and industrial added value in the park before and after policy implementation.

[0023] By temporally coupling the energy flow model, carbon flow model, and economic flow model along a unified time axis, a dynamic policy simulation model is constructed to quantify the marginal changes brought about by policy implementation. The model expression is as follows:

[0024] ;

[0025] in: The park's energy-carbon-economy synergy index is defined by a higher value, which indicates better synergy in the park's energy efficiency improvement, carbon emission reduction effectiveness, and economic benefit guarantee. The change in the industrial added value of the park is a core economic output indicator used to measure the new value created by the park's production activities. It represents the change in the park's overall energy consumption, covering all types of energy, including electricity and heat, and has a more comprehensive coverage compared to a single indicator of electricity consumption intensity. The change in the total carbon emissions of the park is a core environmental impact indicator. The change in the total profit of enterprises in the park is a direct indicator of the park's economic vitality. This represents the change in the total cost of enterprises in the park, including energy costs, pollution control costs, and equipment investment depreciation. , and For policy weighting coefficients, satisfying The coefficient can be flexibly set by policymakers based on policy orientation.

[0026] In step 4, the weighting coefficients are determined using a combination of the Analytic Hierarchy Process (AHP) and the entropy weighting method. Subjective weights are determined using the AHP based on policy guidance and expert experience, while objective weights are calculated using the entropy weighting method based on historical operational data of the park. The final combined weighting formula is as follows:

[0027] ;

[0028] in: For the final combined weights; This is a subjective weighting adjustment coefficient, which can be flexibly adjusted according to the development stage of the park; These are the subjective weights obtained from the Analytic Hierarchy Process (AHP). The objective weights obtained by the entropy weight method;

[0029] By flexibly adjusting the weighting coefficients, the system can adapt to different regions and development stages of industrial parks, generating standardized evaluation results that meet localized needs.

[0030] The basic model expression for the net economic effect of the policy in step 5 is:

[0031] ;

[0032] in: For the net economic effect of the policy, if This indicates that the policy is economically feasible throughout its entire life cycle; Energy savings refer to the amount of energy cost savings resulting from improved energy efficiency. ,in For comprehensive energy prices; Carbon asset returns refer to the comprehensive benefits generated by carbon emission reduction, including revenue from the sale of surplus carbon emission allowances, savings in carbon tax / carbon emission compliance costs, green electricity premium revenue, and green special subsidies, which quantifies the economic value of carbon reduction behavior. New investment costs refer to the capital expenditures incurred by enterprises in response to policy requirements, such as purchasing energy-saving equipment, investing in photovoltaic / wind power projects, and constructing energy storage facilities. The systemic cost of the power grid is used to quantify the indirect system-side costs generated by the grid connection of renewable energy. Specifically, it includes the cost of additional standby thermal power unit capacity to absorb fluctuating renewable energy, the amortization cost of grid upgrades and renovations, and the economic losses caused by the imbalance between power supply and demand.

[0033] Furthermore, the model can be extended to a full-scope cost-benefit accounting form:

[0034] ;

[0035] in: For the special revenue from green electricity trading; For the benefits of special policy subsidies; To account for equipment operation and maintenance costs, and to achieve full-caliber accounting of all economic inflows and outflows throughout the entire policy lifecycle.

[0036] The policy tool variables in step 6 include carbon tax rate, carbon quota tightening ratio, green electricity subsidy amount, energy storage subsidy ratio, equipment retrofit subsidy ratio, time-of-use electricity price parameters, and demand response compensation parameters.

[0037] Step 6 includes the policy scenario library, which includes: the baseline scenario BAU with no new policies, single policy scenarios, and combined policy scenarios. The single policy scenarios include carbon quota tightening S1, green electricity subsidies / green electricity premiums S2, and energy efficiency improvement / equipment retrofit subsidies S4. The combined policy scenarios include carbon quotas + green electricity subsidies S5 and carbon quotas + green electricity subsidies + energy storage incentives S6.

[0038] The objective function of the multi-objective optimization problem in step 7 is:

[0039] ;

[0040] in: The core evaluation indicator in existing research is the minimization of comprehensive energy consumption intensity, with the optimization objective being the minimum comprehensive energy consumption intensity. Extend the indicators to the environmental dimension, with the optimization objective of minimizing carbon emission intensity; The core economic indicator is the maximization of return on assets as the optimization objective.

[0041] The multi-dimensional constraints include total investment budget constraints for the park, regional power grid absorption capacity constraints, minimum economic growth requirements, average return on assets threshold constraints for enterprises, rigid constraints on emission reduction targets, and investment payback period constraints.

[0042] The core constraint expression for the policy sandbox in step 8 is:

[0043] ;

[0044] in: The policy tool variable vector includes carbon emission intensity, carbon tax rate, green electricity subsidy amount, and energy storage electricity price policy; The feature vector for enterprises / parks includes park size, enterprise ownership, industry type, and geographical location, used to characterize the heterogeneity of policy transmission; To describe the complex functional relationship in which policies affect optimization objectives through enterprise / park characteristics; The vector of resource and economic constraints includes the total investment budget of the park, the regional power grid's absorption capacity, and the minimum economic growth requirement; and These are the lower and upper limits of the constraint conditions, respectively.

[0045] Step 9 provides multi-dimensional visualization outputs including an energy-carbon-economic synergy radar chart, a net economic effect change curve, a marginal emission reduction cost curve (MACC), a Pareto frontier distribution map, and a multi-scenario indicator comparison bar chart. Simultaneously, it outputs sensitivity analysis of key influencing factors, explanations of constraint fulfillment, and policy implementation priorities.

[0046] (III) Beneficial Effects:

[0047] The beneficial effects of this invention are:

[0048] This invention represents a comprehensive upgrade of low-carbon policies for industrial parks from static evaluation based on single indicators to multi-dimensional, collaborative, and dynamic assessment. It accurately quantifies the combined impact of different policies on the park's energy, carbon emissions, and economic benefits, fully reconstructing the actual dynamic process of policy implementation, resulting in assessments that are more realistic. It achieves precise accounting of the economic effects of low-carbon policies throughout their entire lifecycle, effectively identifying systemic costs of the power grid that are easily overlooked in traditional assessments, internalizing environmental externalities, and significantly improving the accuracy of policy economic feasibility assessments.

[0049] This invention achieves global intelligent optimization of low-carbon policy combinations under multiple constraints. It can directly output the optimal policy combination and carbon reduction implementation path that meets multiple constraints such as park budget, profit margin, emission reduction targets, and grid absorption. This significantly improves the feasibility of policy implementation and enhances the scientific nature and enforceability of policy formulation. Furthermore, it possesses strong scenario adaptability, enabling differentiated modeling based on the industrial park's industrial structure, energy structure, resource endowment, and development goals. This allows for tailored and precise policy implementation, fully adapting to the low-carbon transformation needs of different types of parks, including chemical, equipment manufacturing, and energy-intensive industries.

[0050] This invention presents the policy implementation effects and decision-making logic intuitively through multi-dimensional visualization output and interpretability analysis, significantly enhancing the transparency and credibility of policy formulation, and providing park managers and policy-making departments with an easy-to-use, efficient, and accurate decision support tool. Attached Figure Description

[0051] Figure 1 This is an overall flowchart of the dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economy synergy, as described in this invention.

[0052] Figure 2 A monthly aggregated energy-related time series linkage diagram under the BAU baseline scenario;

[0053] Figure 3 This is a diagram showing the observable rebound rate of the dynamic simulation method.

[0054] Figure 4 A comparison of the time offset effects between dynamic and static methods;

[0055] Figure 5 A comparison diagram of the structure transition between dynamic and static methods;

[0056] Figure 6 A life-cycle cost-benefit curve for different policy combinations;

[0057] Figure 7 A comparison chart of the frontier advantages and stability of different multi-objective optimization algorithms;

[0058] Figure 8 A two-dimensional Pareto scatter plot of carbon emission intensity versus profit;

[0059] Figure 9 A bar chart comparing carbon emission intensity indicators under different policy scenarios;

[0060] Figure 10 A bar chart comparing profit indicators under different policy scenarios;

[0061] Figure 11 A bar chart comparing the synergy index indicators under different policy scenarios;

[0062] Figure 12 A bar chart showing the marginal emission reduction costs of the industrial park under different policy scenarios;

[0063] Figure 13 A radar chart showing the energy-carbon-economic synergy between the BAU baseline and optimal scenarios. Detailed Implementation

[0064] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0065] This invention presents a dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy. Taking the eastern non-resource-based chemical industrial parks, where existing research has shown significant policy effects, as an example, this invention integrates the method described herein to conduct dynamic simulation and optimization decision-making for low-carbon policies. The flowchart of this embodiment is as follows: Figure 1 As shown, it includes the following steps:

[0066] Step 1, Park Boundary Identification and Carbon Emission Responsibility Unit Delineation: Through boundary delineation algorithms, the park's boundaries are accurately delineated in multiple dimensions to clarify the park's carbon emission accounting boundaries and responsibility attribution;

[0067] Specifically, the spatial geographical boundaries of the industrial park, the list of resident enterprises, the list of production equipment, the topology information of public auxiliary facilities, and the energy access and transmission relationships are obtained to complete the three-level division of the park's spatial boundaries, organizational boundaries, and energy boundaries. Based on the corresponding flow relationships of enterprises, production equipment, process links, and energy media, direct emission units, indirect emission units, and alternative emission reduction units within the park are identified to form a set of carbon emission responsibility units in the park, and the carbon emission accounting scope and responsible entities of each unit are clarified.

[0068] Step 2: Multi-source data collection and construction of energy-carbon-economy collaborative database for the park: Collect core micro-level data of the park from all dimensions, clean, correct, denoise, standardize and unify the time stamp of the collected raw data, remove invalid data and outliers, and construct the basic database of energy-carbon-economy collaborative database of the park through the fusion of multi-source heterogeneous data.

[0069] Specifically, the core data includes enterprise energy consumption and cost data, distributed photovoltaic / wind power generation operation data, real-time carbon emission factor database of the power grid, enterprise financial statement data, and policy-related parameter data.

[0070] The core data collected includes four main categories: energy consumption data for all categories, such as electricity, heat, and gas, consumed by enterprises on an hourly / daily basis; economic data for enterprises, including output, industrial added value, operating income, profit, and comprehensive costs; flexible resource operation data, such as distributed photovoltaic / wind power output, energy storage charging and discharging, and demand response; and policy-related basic parameters, such as carbon emission factors, time-of-use electricity prices, carbon trading prices, subsidy parameters, and carbon tax rates.

[0071] To meet the needs of dynamic simulation with high time-series accuracy, hourly time-series data such as meteorological data, irradiance data, photovoltaic output power, and electricity load can be collected for 8760 hours throughout the year in the park, providing full-time data support for subsequent dynamic simulation.

[0072] Step 3: Construct a dynamic coupling simulation model of energy flow, carbon flow, and economic flow: Through the full-dimensional micro-core data of the park's energy-carbon-economic synergy basic database, track the temporal evolution of energy flow, carbon flow, and economic flow in the park, construct a dynamic simulation model of policies, and quantify the marginal changes brought about by policy implementation.

[0073] Specifically, based on the park's energy-carbon-economic synergy database, three sub-models are constructed: an energy flow model to describe the entire process of energy purchase, conversion, transmission, distribution, consumption, and substitution in the park; a carbon flow model to characterize the mapping relationship between energy consumption and carbon emissions in the park, as well as the emission reductions formed by low-carbon substitution measures; and an economic flow model to quantify the dynamic changes in investment, costs, benefits, profits, and industrial added value in the park before and after policy implementation.

[0074] By temporally coupling the energy flow model, carbon flow model, and economic flow model along a unified time axis, a dynamic policy simulation model is constructed to quantify the marginal changes brought about by policy implementation. The model expression is as follows:

[0075] ;

[0076] in: The park's energy-carbon-economy synergy index is defined by a higher value, which indicates better synergy in the park's energy efficiency improvement, carbon emission reduction effectiveness, and economic benefit guarantee. The change in the industrial added value of the park is a core economic output indicator used to measure the new value created by the park's production activities. It represents the change in the park's overall energy consumption, covering all types of energy such as electricity and heat, and has a more comprehensive coverage compared to a single indicator of electricity consumption intensity. The change in the total carbon emissions of the park is a core environmental impact indicator. The change in the total profit of enterprises in the park is a direct indicator of the park's economic vitality. This represents the change in the total cost of enterprises in the park, including energy costs, pollution control costs, and equipment investment depreciation. , and For policy weighting coefficients, satisfying The coefficient can be flexibly set by the decision-maker; for example, it can be increased. Weighting, which can be increased when emphasizing stable economic growth. and The weight.

[0077] Step 4: Construct an energy-carbon-economy synergistic evaluation model: Perform dimensionless processing on energy, carbon, and economic indicators under different policy scenarios, and obtain the park synergy index by weighted fusion calculation of weight coefficients, forming a standardized policy comprehensive performance evaluation system.

[0078] The weighting coefficients are determined using a combination of the Analytic Hierarchy Process (AHP) and the entropy weighting method. Subjective weights are determined using the AHP based on policy guidance and expert experience, while objective weights are calculated using the entropy weighting method based on historical operational data of the park. The final combined weighting formula is as follows:

[0079] ;

[0080] in: For the final combined weights; This is a subjective weighting adjustment coefficient, which can be flexibly adjusted according to the development stage of the park; These are the subjective weights obtained from the Analytic Hierarchy Process (AHP). The objective weights are obtained using the entropy weight method.

[0081] By flexibly adjusting the weighting coefficients, the system can adapt to different regions and development stages of industrial parks, generating standardized evaluation results that meet localized needs.

[0082] Step 5: Construct a policy net economic effect model: Define the accounting boundaries and scope of the direct costs, indirect benefits and systemic costs of policy implementation. Based on the policy quantification model and the energy substitution elasticity model, introduce the full life cycle economic cost-benefit analysis method to quantify the comprehensive economic effect of the marginal substitution rate of renewable energy for thermal power, construct a policy net economic effect accounting model, and evaluate the economic feasibility of a single policy or a combination of policies.

[0083] Specifically, the basic model expression for the net economic effect of the policy is as follows:

[0084] ;

[0085] in: For the net economic effect of the policy, if This indicates that the policy is economically feasible throughout its entire life cycle; Energy savings refer to the amount of energy cost savings resulting from improved energy efficiency. ,in For comprehensive energy prices; Carbon asset returns refer to the comprehensive benefits generated by carbon emission reduction, including revenue from the sale of surplus carbon emission allowances, savings in carbon tax / carbon emission compliance costs, green electricity premium revenue, and green special subsidies, which quantifies the economic value of carbon reduction behavior. New investment costs refer to capital expenditures incurred by enterprises in response to policy requirements, such as purchasing energy-saving equipment, investing in photovoltaic / wind power projects, and constructing energy storage facilities. The systemic cost of the power grid is used to quantify the indirect system-side costs generated by the grid connection of renewable energy. Specifically, it includes the cost of additional standby thermal power unit capacity to absorb fluctuating renewable energy, the amortization cost of grid upgrades and renovations, and the economic losses caused by the imbalance between power supply and demand.

[0086] Furthermore, the model can be extended to a full-scope cost-benefit accounting form:

[0087] ;

[0088] in: For the special revenue from green electricity trading; For the benefits of special policy subsidies; To account for equipment operation and maintenance costs, and to achieve full-caliber accounting of all economic inflows and outflows throughout the entire policy lifecycle.

[0089] Step 6: Establish a set of policy instrument variables and a policy scenario library: Based on the design requirements of the park's low-carbon policy, set policy instrument variables and clarify their reasonable value ranges, and build a standardized policy scenario library to provide a standardized scenario foundation for subsequent simulation analysis and optimization solutions.

[0090] The policy tool variables include carbon tax rate, carbon quota tightening ratio, green electricity subsidy amount, energy storage subsidy ratio, equipment retrofit subsidy ratio, time-of-use electricity price parameters, and demand response compensation parameters. In this embodiment, the policy tool variable set is set as follows: four core policy tools are defined: carbon quota tightening ratio (range 0%-20%), green electricity subsidy amount (range 0-0.1 yuan / kWh), energy storage subsidy ratio (range 0%-30%), and equipment retrofit subsidy ratio (range 0%-40%). The reasonable value range and calculation step size for each variable are clearly defined.

[0091] Based on different combinations of policy instrument variables, six typical scenarios are constructed: the baseline scenario BAU without new policies, single policy scenarios, and combined policy scenarios. The single policy scenarios include carbon quota tightening S1, green electricity subsidies / green electricity premiums S2, and energy efficiency improvement / equipment retrofit subsidies S4. The combined policy scenarios include carbon quotas + green electricity subsidies S5 and carbon quotas + green electricity subsidies + energy storage incentives S6.

[0092] Based on six policy scenarios, the core operational data for each scenario in step 2 are shown in Tables 1 and 2 below. The monthly aggregated energy-related time series data under the BAU baseline scenario are as follows: Figure 2 As shown.

[0093] Table 1. Core energy operation data for each policy scenario:

[0094] Scene <![CDATA[Total annual carbon emissions (tCO2)]]> Total annual load (kWh) Annual PV self-consumption (kWh) Annual electricity purchase (kWh) Annual electricity transmitted to other regions (kWh) BAU 571991.5 1039490 256119.9 315871.6 783369.8 S1 554831.8 1039490 249624.4 305207.4 789865.3 S2 571991.5 1122649 259184 312807.5 863464.8 S4 526232.2 1039490 238678.2 287554 800811.4 S5 543391.9 1122649 248100.1 295291.8 874548.8 S6 543391.9 1143439 353166.5 190225.4 781356.7 ;

[0095] Table 2. Core efficiency and economic data for each policy scenario:

[0096] Scene PV self-use rate Load Coverage Annual total energy consumption (tce) Total annual profit (RMB) Average profit margin BAU 0.24639 0.447769 183.2056 593929.3 0.494941 S1 0.240141 0.44991 177.0203 534940 0.445783 S2 0.230868 0.453126 181.4283 544310.9 0.453592 S4 0.229611 0.453561 166.7813 468789.8 0.390658 S5 0.220995 0.456577 171.2693 479171.8 0.39931 S6 0.308864 0.64993 110.3308 404515 0.337096 .

[0097] Based on the six policy scenarios, the synergy index of each policy scenario in step 3 is shown in Table 3.

[0098] Table 3. Calculation results of the synergy index for each policy scenario:

[0099] Scene Synergy Index (CI) BAU 0.886960393 S1 0.889280185 S2 0.886960986 S4 0.89324888 S5 0.890852603 S6 0.890872911 .

[0100] Meanwhile, by tracking the temporal evolution of energy flow, carbon flow, and economic flow through dynamic simulation models, the rebound effect, temporal shift effect, and structural transfer effect after policy implementation can be clearly observed. Traditional static methods, however, can only output the final result and cannot capture these dynamic changes. A performance comparison of the two methods is provided in the example. Figure 3 , Figure 4 , Figure 5 As shown, the technical advantages of the dynamic coupling model of the present invention are fully verified.

[0101] The evaluation results show that scenario S4 has the highest synergy index and performs best in the energy efficiency improvement dimension; scenario S6 performs best in the carbon emission control dimension and significantly improves the self-sufficiency rate of new energy; scenario S1 performs most balanced in terms of carbon emission reduction and economic balance. Through dimensionless processing and weighted fusion, a standardized horizontal comparison of the comprehensive performance of different policy scenarios was achieved.

[0102] Based on six policy scenarios, the final net economic effect results obtained in step 5 are shown in Table 4, and the cost-benefit curves for different policy combinations are shown in Table 4. Figure 6 As shown.

[0103] Table 4. Calculation results of net economic effects for each policy scenario:

[0104] Scene Net economic effect of the policy NE (yuan) BAU 443929.2861 S1 384939.981 S2 394310.9163 S4 318789.825 S5 329171.8117 S6 254515.043 .

[0105] The calculation results show that all policy scenarios have achieved the core objectives of carbon emission reduction and energy efficiency improvement. Among them, the S6 scenario has the best emission reduction effect and has a significant long-term advantage in terms of cost-benefit performance throughout the entire life cycle, which verifies the accuracy of the net economic effect model of the present invention that includes systemic costs.

[0106] Step 7: Construct a multi-objective optimization model to solve the Pareto optimal frontier: Integrate multi-objective optimization algorithms, simulate the Pareto optimal frontier under different policy combinations, and set multi-dimensional constraints to find the global optimal balance point among the three major objectives of improving energy efficiency, achieving carbon emission reduction targets, and stable economic growth.

[0107] Specifically, the objective function of the multi-objective optimization problem is:

[0108] ;

[0109] in: The core evaluation indicator in existing research is the minimization of comprehensive energy consumption intensity, with the optimization objective being the minimum comprehensive energy consumption intensity. Extend the indicators to the environmental dimension, with the optimization objective of minimizing carbon emission intensity; The core economic indicator is the return on assets, with the optimization goal of maximizing the return on assets.

[0110] At the same time, multi-dimensional constraints are set, including total investment budget constraints for the park, regional power grid absorption capacity constraints, minimum economic growth requirements, average return on assets threshold constraints for enterprises, rigid constraints on emission reduction targets, and investment payback period constraints, to ensure the feasibility of engineering implementation of the optimization plan.

[0111] This embodiment employs constraint algorithms, MOEA / D, weighted sum optimization algorithms, and NSGA-II algorithms to solve the problem. The frontier advantage and stability of different algorithms are compared. Figure 7 As shown, the NSGA-II algorithm is finally used to output the Pareto optimal solution set. The calculation results of carbon emission intensity and asset return rate for each policy scenario are shown in Table 5, the verification results of the Pareto solution for multi-objective optimization are shown in Table 6, and the two-dimensional Pareto scatter plot of carbon emission intensity-profit is shown in Table 6. Figure 8 As shown.

[0112] Table 5 Core optimization target values ​​for each policy scenario:

[0113] Scene <![CDATA[Carbon emission intensity (tCO2 / 10,000 yuan of output value)]]> Return on assets BAU 0.47666 0.000153 S1 0.46236 0.000148 S2 0.47666 0.000151 S4 0.438527 0.000139 S5 0.452827 0.000143 S6 0.452827 9.19E-05 .

[0114] Table 6. Verification results of Pareto solutions for multi-objective optimization:

[0115] Scene Annual total energy consumption (tce) Total annual profit (RMB) Synergy Index (CI) Is it a Pareto solution? BAU 183.2056 593929.3 0.88696 true S1 177.0203 534940 0.88928 true S2 181.4283 544310.9 0.886961 true S4 166.7813 468789.8 0.893249 true S5 171.2693 479171.8 0.890853 true S6 110.3308 404515 0.890873 true .

[0116] The results show that scenarios BAU, S1, S2, S4, S5, and S6 in this embodiment are all Pareto efficient solutions, corresponding to the optimal solutions under different policy preferences.

[0117] Step 8: Establish a policy sandbox and intelligent recommendation mechanism: Develop a policy sandbox simulation environment that allows users to set their own constraints. Through machine learning prediction engines and reinforcement learning algorithms, it automatically recommends the optimal carbon reduction path that meets all constraints.

[0118] The core constraint expression of the policy sandbox is:

[0119] ;

[0120] in: The policy tool variable vector includes carbon emission intensity, carbon tax rate, green electricity subsidy amount, energy storage electricity price policy, etc. The feature vector for enterprises / parks includes park size, enterprise ownership, industry type, geographical location, etc., which is used to characterize the heterogeneous characteristics of policy transmission. To describe the complex functional relationship in which policies affect optimization objectives through enterprise / park characteristics; The vector of resource and economic constraints includes the total investment budget of the park, the regional power grid's absorption capacity, and the minimum economic growth rate requirement. and These are the lower and upper limits of the constraints, such as the average return on assets of enterprises not being lower than a preset threshold, the total amount of policy subsidies not exceeding the annual fiscal budget, and the reduction in carbon emissions not being lower than the rigid target value.

[0121] The candidate policy combination is rapidly simulated and evaluated by machine learning prediction models. Then, the candidate schemes that meet the constraints are screened and ranked in multiple dimensions by reinforcement learning or heuristic search algorithms. The optimal carbon reduction path that meets all constraints is output, and the second-best alternative policy combination is also output. This realizes the leap from ex-post policy evaluation to proactive generation of the optimal policy combination.

[0122] In this embodiment, a bar chart comparing carbon emission intensity indicators under different policy scenarios is shown below. Figure 9 As shown in the bar chart comparing profit indicators under different policy scenarios, see below. Figure 10 As shown in the bar chart comparing the synergy index indicators under different policy scenarios, see below. Figure 11 As shown. Simultaneously, bar charts of marginal emission reduction costs for the industrial park under different policy scenarios are generated, as shown below. Figure 12 As shown.

[0123] Step 9: Result Visualization and Decision Interpretation: Based on the full-process results of policy simulation in Step 6, multi-objective optimization in Step 7, and intelligent recommendation in Step 8, multi-dimensional visualization output is generated to intuitively display the comprehensive performance of the policy scheme in the three dimensions of energy, carbon, and economy, and to mark the deviation of each indicator from the baseline scenario. The multi-dimensional visualization output includes an energy-carbon-economy synergistic radar chart, a net economic effect change curve, a marginal emission reduction cost curve (MACC), a Pareto frontier distribution map, and a multi-scenario indicator comparison bar chart. Simultaneously, sensitivity analysis of key influencing factors, explanations of constraint fulfillment, and policy implementation priority suggestions are output, providing decision-makers with clear, interpretable, and implementable decision support. In the example, a synergistic radar chart of the BAU baseline scenario and the S6 optimal scenario is generated, as shown below. Figure 13 As shown.

[0124] This invention constructs a three-dimensional synergistic evaluation framework for energy, carbon, and economy, breaking through the limitations of existing technologies that evaluate policy effectiveness from only one dimension of energy consumption or carbon emissions. By weighted summation, it integrates the dynamic changes of energy efficiency, carbon emission reduction, and economic benefits into a synergistic index, forcing decision-makers to consider the balance among the three in policy evaluation. At the same time, it achieves the scientific determination of weight coefficients through a combination of subjective and objective weighting methods, which can flexibly adapt to policy orientations at different stages of development and has strong scenario adaptability and operability.

[0125] A dynamic coupling simulation model of energy flow, carbon flow, and economic flow was constructed, breaking through the technical bottleneck of traditional static analysis methods. It adopts a variable perspective to focus on the marginal effect of policy intervention, realizes the temporal coupling of energy flow, carbon flow, and economic flow under a unified time axis, and can accurately depict the temporal evolution of the three flows during policy implementation. It can effectively identify policy lag effects, rebound effects, and structural transfer effects, and fully restore the real dynamic process of policy implementation.

[0126] A full life-cycle policy net economic effect model that includes systemic costs was proposed, filling the technical gap of insufficient quantification of systemic implicit costs in existing technologies. For the first time, the systemic implicit costs of renewable energy grid connection, such as peak shaving, reserve, and grid transformation, were internalized into the decision-making and accounting system. The costs were broken down into quantifiable sub-cost accounting formulas, realizing full-caliber accounting of economic inflows and outflows throughout the policy life cycle, and solving the problem of distortion in the assessment of carbon reduction economic effects by traditional methods.

[0127] Simultaneously, it achieves a technological leap from ex-post policy evaluation to proactive generation of optimal policy combinations. It deeply integrates the improved multi-objective optimization algorithm with the policy sandbox intelligent recommendation mechanism, solves the Pareto optimal solution set with economic feasibility as a hard constraint, and incorporates the heterogeneous characteristics of parks and enterprises. It can generate precise policy recommendations for different types of parks, solving the core pain point of traditional methods that can only evaluate the effect of a single policy and cannot output the global optimal combination solution.

[0128] Finally, a complete technological closed loop covering the entire process of low-carbon policy formulation was formed, and a full-process technological architecture was constructed, which includes data collection, dynamic simulation, collaborative evaluation, economic accounting, scenario construction, multi-objective optimization, intelligent recommendation, and visualization output. This architecture can support the scientific formulation, precise implementation, and dynamic iterative optimization of low-carbon policies for industrial parks.

[0129] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy, characterized in that, Includes the following steps: Step 1, Park Boundary Identification and Carbon Emission Responsibility Unit Delineation: Through boundary delineation algorithms, the park's boundaries are accurately delineated in multiple dimensions to clarify the park's carbon emission accounting boundaries and responsibility attribution; Step 2: Multi-source data collection and construction of energy-carbon-economy collaborative database in the park: Collect core micro data of the park from all dimensions, clean, correct, denoise, standardize and unify the time stamp of the collected raw data, remove invalid data and outliers, and construct the basic database of energy-carbon-economy collaborative database of the park through the fusion of multi-source heterogeneous data; Step 3: Construct a dynamic coupling simulation model of energy flow, carbon flow, and economic flow: Through the full-dimensional micro-core data of the park's energy-carbon-economic synergy basic database, track the temporal evolution of energy flow, carbon flow, and economic flow in the park, construct a dynamic simulation model of policies, and quantify the marginal changes brought about by policy implementation; Specifically, based on the park's energy-carbon-economic synergy database, three sub-models are constructed: an energy flow model to describe the entire process of energy purchase, conversion, transmission and distribution, consumption and substitution in the park; a carbon flow model to characterize the mapping relationship between energy consumption and carbon emissions in the park, as well as the emission reductions formed by low-carbon substitution measures; and an economic flow model to quantify the dynamic changes in investment, costs, benefits, profits and industrial added value in the park before and after policy implementation. By temporally coupling the energy flow model, carbon flow model, and economic flow model along a unified time axis, a dynamic policy simulation model is constructed to quantify the marginal changes brought about by policy implementation. The model expression is as follows: ； in: The park's energy-carbon-economy synergy index is defined by a higher value, which indicates better synergy in the park's energy efficiency improvement, carbon emission reduction effectiveness, and economic benefit guarantee. The change in the industrial added value of the park is a core economic output indicator used to measure the new value created by the park's production activities. This represents the change in the park's overall energy consumption, covering all types of energy, including electricity and heat. The change in the total carbon emissions of the park is a core environmental impact indicator. The change in the total profit of enterprises in the park is a direct indicator of the park's economic vitality. This represents the change in the total cost of enterprises in the park, including energy costs, pollution control costs, and equipment investment depreciation. , and For policy weighting coefficients, satisfying The coefficient is flexibly set by policymakers based on policy guidance; Step 4: Construct an energy-carbon-economy synergistic evaluation model: Perform dimensionless processing on energy, carbon, and economic indicators under different policy scenarios, and obtain the park synergy index by weighted fusion calculation of weight coefficients to form a standardized policy comprehensive performance evaluation system; Step 5: Construct a policy net economic effect model: Define the accounting boundaries and scope of the direct costs, indirect benefits and systemic costs of policy implementation. Based on the policy quantification model and the energy substitution elasticity model, introduce the full life cycle economic cost-benefit analysis method to quantify the comprehensive economic effect of the marginal substitution rate of renewable energy for thermal power, construct a policy net economic effect accounting model, and evaluate the economic feasibility of a single policy or a combination of policies. Step 6: Establish a set of policy instrument variables and a policy scenario library: Based on the design requirements of the park's low-carbon policies, set policy instrument variables and clarify their reasonable value ranges, construct a standardized policy scenario library, and classify the data collected in Step 2 according to the policy scenario library to provide a standardized scenario foundation for subsequent simulation analysis and optimization solutions; The policy scenario library includes: the baseline scenario BAU without new policies, single policy scenarios, and combined policy scenarios. Among them, single policy scenarios include carbon quota tightening S1, green electricity subsidies / green electricity premiums S2, and energy efficiency improvement / equipment renovation subsidies S4; combined policy scenarios include carbon quota + green electricity subsidies S5, and carbon quota + green electricity subsidies + energy storage incentives S6; Step 7: Construct a multi-objective optimization model to solve the Pareto optimal frontier: Integrate multi-objective optimization algorithms, simulate the Pareto optimal frontier under different policy combinations, and set multi-dimensional constraints to find the global optimal balance point among the three major objectives of improving energy efficiency, achieving carbon emission reduction targets, and stable economic growth. Step 8: Establish a policy sandbox and intelligent recommendation mechanism: Develop a policy sandbox simulation environment that allows users to set their own constraints. Through machine learning prediction engines and reinforcement learning algorithms, it automatically recommends the optimal carbon reduction path that meets all constraints. Step 9: Visualization of Results and Interpretation of Decisions: Based on the full-process results of policy simulation in Step 6, multi-objective optimization in Step 7, and intelligent recommendation in Step 8, generate multi-dimensional visualization outputs to intuitively display the comprehensive performance of policy solutions in the three dimensions of energy, carbon, and economy, and mark the deviation of each indicator from the baseline scenario.

2. The dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy as described in claim 1, characterized in that: The core data in step 2 includes enterprise energy consumption and cost data, distributed photovoltaic / wind power generation operation data, real-time carbon emission factor database of the power grid, enterprise financial statement data, and policy-related parameter data.

3. The dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy as described in claim 1, characterized in that, In step 4, the weighting coefficients are determined using a combination of the Analytic Hierarchy Process (AHP) and the entropy weighting method. Subjective weights are determined using the AHP based on policy guidance and expert experience, while objective weights are calculated using the entropy weighting method based on historical operational data of the park. The final combined weighting formula is as follows: ； in: For the final combined weights; This is a subjective weighting adjustment coefficient, which can be flexibly adjusted according to the development stage of the park; These are the subjective weights obtained from the Analytic Hierarchy Process (AHP). The objective weights obtained by the entropy weight method; By flexibly adjusting the weighting coefficients, the system can adapt to different regions and development stages of industrial parks, generating standardized evaluation results that meet localized needs.

4. The dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy as described in claim 1, characterized in that, The basic model expression for the net economic effect of the policy in step 5 is: ； in: For the net economic effect of the policy, if This indicates that the policy is economically feasible throughout its entire life cycle; For energy-saving benefits, ,in For comprehensive energy prices; Carbon asset revenue includes income from the sale of surplus carbon emission allowances, savings from carbon tax / carbon emission compliance costs, green electricity premium revenue, and green special subsidies, quantifying the economic value of carbon reduction behavior; New investment costs refer to the capital expenditures incurred by enterprises in response to policy requirements, such as purchasing energy-saving equipment, investing in photovoltaic / wind power projects, and constructing energy storage facilities. The systemic cost of the power grid is used to quantify the indirect system-side costs generated by the grid connection of renewable energy. Specifically, it includes the cost of additional standby thermal power unit capacity to absorb fluctuating renewable energy, the amortization cost of grid upgrades and renovations, and the economic losses caused by the imbalance between power supply and demand. Furthermore, the model can be extended to a full-scope cost-benefit accounting form: ； in: For the special revenue from green electricity trading; For the benefits of special policy subsidies; To account for equipment operation and maintenance costs, and to achieve full-caliber accounting of all economic inflows and outflows throughout the entire policy lifecycle.

5. The dynamic simulation and optimization decision-making method for low-carbon policies in industrial parks based on energy-carbon-economic synergy as described in claim 1, characterized in that, The policy tool variables in step 6 include carbon tax rate, carbon quota tightening ratio, green electricity subsidy amount, energy storage subsidy ratio, equipment retrofit subsidy ratio, time-of-use electricity price parameters, and demand response compensation parameters.

6. The method for dynamic simulation and optimization decision-making of low-carbon policies in industrial parks based on energy-carbon-economic synergy as described in claim 1, characterized in that, The objective function of the multi-objective optimization problem in step 7 is: ； in: The core evaluation indicator in existing research is the minimization of comprehensive energy consumption intensity, with the optimization objective being the minimum comprehensive energy consumption intensity. Extend the indicators to the environmental dimension, with the optimization objective of minimizing carbon emission intensity; The core economic indicator is the maximization of return on assets as the optimization objective. The multi-dimensional constraints include total investment budget constraints for the park, regional power grid absorption capacity constraints, minimum economic growth requirements, average return on assets threshold constraints for enterprises, rigid constraints on emission reduction targets, and investment payback period constraints.

7. The method for dynamic simulation and optimization decision-making of low-carbon policies in industrial parks based on energy-carbon-economic synergy as described in claim 1, characterized in that, The core constraint expression for the policy sandbox in step 8 is: ； in: The policy tool variable vector includes carbon emission intensity, carbon tax rate, green electricity subsidy amount, and energy storage electricity price policy; The feature vector for enterprises / parks includes park size, enterprise ownership, industry type, and geographical location, used to characterize the heterogeneity of policy transmission; To describe the complex functional relationship in which policies affect optimization objectives through enterprise / park characteristics; The vector of resource and economic constraints includes the total investment budget of the park, the regional power grid's absorption capacity, and the minimum economic growth requirement; and These are the lower and upper limits of the constraint conditions, respectively.

8. The method for dynamic simulation and optimization decision-making of low-carbon policies in industrial parks based on energy-carbon-economic synergy as described in claim 1, characterized in that, Step 9 provides multi-dimensional visualization outputs including an energy-carbon-economic synergy radar chart, a net economic effect change curve, a marginal emission reduction cost bar chart, a Pareto frontier distribution chart, and a multi-scenario indicator comparison bar chart. Simultaneously, it outputs sensitivity analysis of key influencing factors, explanations of constraint fulfillment, and policy implementation priorities.

Images