A grid-side energy storage investment decision-making method and system driven by AI and carbon asset returns

Abstract

This invention provides a grid-side energy storage investment decision-making method and system driven by AI and carbon asset returns. The method includes: acquiring EMS operation data and electricity market data from energy storage power stations; establishing a four-dimensional electricity market return calculation model covering peak-valley arbitrage, capacity leasing, peak shaving and frequency regulation, and capacity pricing returns; using AI algorithms to perform joint probability prediction of carbon price fluctuations, policy changes, and filing success rates, and outputting confidence intervals; establishing an automatic verification method for carbon emission reductions by replacing thermal power for peak shaving, and generating carbon revenue cash flow; constructing a five-dimensional coupled cash flow model, using total investment IRR, equity IRR, and dynamic investment payback period as evaluation indicators; establishing a carbon revenue rights pledge credit enhancement method, and jointly determining investment decisions based on financing feasibility thresholds; and establishing an EMS and MRV data interface to sequentially execute data interaction at each stage. This invention achieves synergistic evaluation of electricity market returns and carbon asset returns, improving the scientific nature of investment decisions.

Description

Technical Field

[0001] This invention relates to the field of grid-side energy storage investment decision-making technology, and in particular to a grid-side energy storage investment decision-making method and system driven by AI and carbon asset returns. Background Technology

[0002] The revenue sources of grid-side independent energy storage power stations include electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue, and capacity tariff revenue. The carbon emission reductions generated by replacing thermal power peak shaving can be converted into carbon asset revenue. Existing investment decision-making methods are mostly aimed at photovoltaic or user-side energy storage, lacking an independent calculation model for the four-dimensional electricity market revenue of grid-side energy storage; carbon price forecasting mostly uses single time series models, failing to comprehensively consider uncertainties such as carbon market policy transitions and the success rate of carbon asset registration; carbon emission reduction verification relies on manual data collection and manual report preparation, resulting in long verification cycles and a high risk of errors.

[0003] Furthermore, existing technologies assess electricity market revenue and carbon asset revenue separately, without establishing a coupled cash flow model; investment decisions and financing credit enhancement are independent of each other, without considering the impact of carbon revenue rights pledging on financing feasibility; the data formats of energy storage power station EMS and carbon asset MRV systems are not unified, and data flow at each stage relies on manual import and export, forming data silos, which makes it difficult to coordinate the various links of investment decision-making, carbon asset development, operation monitoring, certification and trading and revenue collection. Summary of the Invention

[0004] The purpose of this invention is to provide a grid-side energy storage investment decision-making method and system based on AI and carbon asset returns, in order to solve the technical problems in the existing technology, such as the single dimension of grid-side energy storage electricity market revenue measurement, insufficient reliability of carbon asset return prediction, high labor cost of carbon emission reduction verification, separate evaluation of electricity market revenue and carbon asset revenue, disconnect between investment decision-making and financing credit enhancement, and data silos between EMS and MRV systems.

[0005] To achieve the above objectives, this invention provides a grid-side energy storage investment decision-making method driven by AI and carbon asset returns, comprising the following steps: S1. Obtain EMS operation data, electricity spot trading data, ancillary service settlement data, and capacity pricing policy data from energy storage power stations. Establish a four-dimensional electricity market revenue calculation model that covers electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue, and capacity pricing revenue. Output the total revenue of the four-dimensional electricity market. S2. Obtain carbon market MRV system data, and use AI algorithms to perform joint probability prediction on carbon price time-series fluctuations, carbon market policy switching and carbon asset registration success rate, and output carbon asset returns with confidence intervals; incorporate carbon asset returns as independent revenue cash flow into the investment evaluation model in advance, establish an automatic verification method for carbon emission reductions that replace thermal power peak shaving with energy storage, and generate carbon revenue cash flow. S3. Couple the total revenue of the four-dimensional electricity market with the carbon revenue cash flow to obtain a five-dimensional coupled cash flow model; use the total investment IRR, equity IRR and dynamic investment payback period as evaluation indicators to calculate the NPV and IRR evaluation results with confidence intervals. S4. Establish a carbon revenue rights pledge credit enhancement method based on carbon revenue cash flow and calculate the financing feasibility threshold; comprehensively judge the total investment IRR, equity IRR and dynamic investment payback period with the financing feasibility threshold, and output the judgment result of whether the investment decision is approved or not. S5. Output investment decision plan based on the judgment result, establish data interface between energy storage power station EMS operation data and carbon asset MRV system, and carry out multi-stage data interaction and processing based on the data interface.

[0006] Preferably, in S1, the electricity spot peak-valley arbitrage profit is calculated based on the spot price difference and charging / discharging efficiency during the charging and discharging periods of the energy storage power station, and the calculation formula is as follows: ; In the formula, The spot electricity price for discharge during period t. The spot electricity price for charging during period t. Where η is the rated energy storage capacity, and η is the charge / discharge efficiency. The duration is specified; capacity leasing revenue is calculated based on the rated energy storage capacity and the unit price of capacity leasing, using the following formula: ; In the formula, This is the unit price for capacity leasing. This refers to capacity availability.

[0007] Preferably, in S2, the AI ​​algorithm takes historical carbon market transaction data, carbon price time series data, policy text data, and historical data of energy storage project filings as input, and uses a joint probability prediction model to output predicted carbon asset returns and their confidence intervals under multiple scenarios.

[0008] Preferably, in S2, the automatic verification method for carbon emission reductions includes: Based on the EMS operation data of energy storage power stations, the discharge amount and marginal emission factor of energy storage to replace thermal power peak shaving are obtained. The carbon emission reduction is automatically calculated by AI and a verification report that meets the requirements of the MRV system is generated. The verified carbon emission reduction is converted into tradable carbon assets and discounted into carbon revenue cash flow.

[0009] Preferably, in S3, the five-dimensional coupled cash flow model aggregates the electricity spot peak-valley arbitrage income, capacity leasing income, peak-shaving and frequency regulation ancillary service income, capacity electricity price income and carbon asset income year by year according to the whole life cycle, and constructs an annual net cash flow sequence. The formula for calculating NPV is: ; In the formula, Let r be the net cash flow in year n, r be the weighted average cost of capital (WACC), and N be the project operating period. The formula for calculating the total investment IRR is: .

[0010] Preferably, in S4, the credit enhancement methods for carbon revenue rights pledging include: The carbon revenue cash flow during the forecast period is used as collateral to calculate the loan-to-value ratio and credit enhancement limit; the financing feasibility threshold is determined based on the capital ratio after credit enhancement, debt coverage ratio and minimum acceptable IRR.

[0011] Preferably, S5 includes: The investment decision-making stage determines the carbon asset development strategy; the construction stage deploys EMS and MRV data interfaces; the operation stage monitors energy storage charging and discharging data in real time and automatically verifies carbon emission reductions; the trading stage puts the verified carbon assets into the market for trading; and the revenue recovery stage feeds back the carbon revenue cash flow to the investment evaluation model for dynamic correction.

[0012] Preferably, in S5, multi-stage data interaction and processing based on the data interface includes: The system outputs investment plan data during the investment decision-making stage, carbon asset registration data during the carbon asset development stage, energy storage operation data during the operation monitoring stage, carbon asset verification data during the verification and trading stage, and carbon revenue data is fed back to the investment evaluation model during the revenue recovery stage.

[0013] Preferably, a grid-side energy storage investment decision-making system driven by AI and carbon asset returns includes: The data acquisition module is used to acquire EMS operation data of energy storage power stations, electricity spot trading data, ancillary service settlement data and capacity pricing policy data, and establish a four-dimensional electricity market revenue calculation model covering electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue and capacity pricing revenue, and output the total four-dimensional electricity market revenue. The joint forecasting module is used to acquire carbon market MRV system data, and based on AI algorithms, it performs joint probability forecasting of carbon price time-series fluctuations, carbon market policy switching and carbon asset registration success rate, and outputs carbon asset returns with confidence intervals; carbon asset returns are pre-incorporated as independent revenue cash flow into the investment evaluation model, and an automatic verification method for carbon emission reductions that replace thermal power peak shaving with energy storage is established to generate carbon revenue cash flow. The investment assessment module is used to couple the total revenue of the four-dimensional electricity market with the carbon revenue cash flow to obtain a five-dimensional coupled cash flow model; using the total investment IRR, equity IRR and dynamic investment payback period as evaluation indicators, the NPV and IRR assessment results with confidence intervals are calculated. The comprehensive judgment module is used to establish a carbon revenue rights pledge credit enhancement method based on carbon revenue cash flow and calculate the financing feasibility threshold; it comprehensively judges the total investment IRR, equity IRR and dynamic investment payback period with the financing feasibility threshold, and outputs the judgment result of whether the investment decision is approved or not. The decision generation module is used to output investment decision schemes based on the judgment results, establish a data interface between the EMS operation data of the energy storage power station and the MRV system of carbon assets, and perform multi-stage data interaction and processing based on the data interface.

[0014] The advantages and beneficial effects of this invention compared to the prior art are: 1. By establishing a four-dimensional electricity market revenue calculation model that covers electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue, and capacity electricity price revenue, the comprehensive quantification of grid-side energy storage electricity market revenue has been achieved, solving the problem of a single revenue dimension in existing technologies.

[0015] 2. By using AI algorithms to perform joint probability predictions on carbon price fluctuations, carbon market policy shifts, and carbon asset registration success rates, the system outputs carbon asset return predictions with confidence intervals, thereby improving the reliability of carbon asset return assessment.

[0016] 3. By establishing an automatic carbon emission reduction verification method for energy storage to replace thermal power peak shaving, the discharge amount and marginal emission factor are automatically obtained based on the EMS operation data of the energy storage power station, and a verification report that meets the requirements of the MRV system is generated, which shortens the verification cycle and reduces labor costs.

[0017] 4. By constructing a five-dimensional coupled cash flow model, the total revenue of the four-dimensional electricity market and the revenue of carbon assets are aggregated year by year according to the whole life cycle, realizing the coordinated evaluation of electricity market revenue and carbon asset revenue.

[0018] 5. By establishing a carbon revenue rights pledge credit enhancement method, the carbon revenue cash flow during the forecast period is used as the pledge target to calculate the pledge ratio and credit enhancement limit, and is comprehensively judged in conjunction with the total investment IRR, capital IRR and dynamic investment payback period, thus realizing the technical linkage between financing feasibility and investment returns.

[0019] 6. By establishing a data interface between the energy storage power station EMS operation data and the carbon asset MRV system, data interaction and processing are carried out in sequence at each stage, connecting the entire process of investment decision-making, carbon asset development, operation monitoring, certification and trading, and revenue collection.

[0020] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0021] Figure 1 This is a flowchart of a grid-side energy storage investment decision-making method driven by AI and carbon asset returns, as described in an embodiment of the present invention. Figure 2 This is a structural diagram of a grid-side energy storage investment decision-making system driven by AI and carbon asset returns, according to an embodiment of the present invention. Detailed Implementation

[0022] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use. They are used only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," and "connect" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0023] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] like Figure 1 As shown, this invention provides a grid-side energy storage investment decision-making method driven by AI and carbon asset returns, including the following steps: S1. Obtain EMS operation data, electricity spot trading data, ancillary service settlement data, and capacity pricing policy data from energy storage power stations. Establish a four-dimensional electricity market revenue calculation model that covers electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue, and capacity pricing revenue. Output the total revenue of the four-dimensional electricity market. S2. Obtain carbon market MRV system data, and use AI algorithms to perform joint probability prediction on carbon price time-series fluctuations, carbon market policy switching and carbon asset registration success rate, and output carbon asset returns with confidence intervals; incorporate carbon asset returns as independent revenue cash flow into the investment evaluation model in advance, establish an automatic verification method for carbon emission reductions that replace thermal power peak shaving with energy storage, and generate carbon revenue cash flow. S3. Couple the total revenue of the four-dimensional electricity market with the carbon revenue cash flow to obtain a five-dimensional coupled cash flow model; use the total investment IRR, equity IRR and dynamic investment payback period as evaluation indicators to calculate the NPV and IRR evaluation results with confidence intervals. S4. Establish a carbon revenue rights pledge credit enhancement method based on carbon revenue cash flow and calculate the financing feasibility threshold; comprehensively judge the total investment IRR, equity IRR and dynamic investment payback period with the financing feasibility threshold, and output the judgment result of whether the investment decision is approved or not. S5. Output investment decision plan based on the judgment result, establish data interface between energy storage power station EMS operation data and carbon asset MRV system, and carry out multi-stage data interaction and processing based on the data interface.

[0025] Preferably, in S1, the electricity spot peak-valley arbitrage profit is calculated based on the spot price difference and charging / discharging efficiency during the charging and discharging periods of the energy storage power station, and the calculation formula is as follows: ; In the formula, The spot electricity price for discharge during period t. The spot electricity price for charging during period t. Where η is the rated energy storage capacity, and η is the charge / discharge efficiency. The duration is specified; capacity leasing revenue is calculated based on the rated energy storage capacity and the unit price of capacity leasing, using the following formula: ; In the formula, This is the unit price for capacity leasing. This refers to capacity availability.

[0026] In one embodiment, in step S1, the electricity spot peak-valley arbitrage profit is calculated based on the spot price difference during the charging and discharging period of the energy storage power station, the rated capacity of the energy storage, the charging and discharging efficiency, and the duration of the period, and is accumulated on a time-by-time basis to obtain the daily or annual peak-valley arbitrage profit; the capacity leasing profit is calculated based on the rated capacity of the energy storage, the capacity leasing unit price, and the capacity availability rate, and the annual capacity leasing profit is calculated. The parameters in the above formulas are all obtained through EMS operating data or publicly available electricity market data, wherein the discharge spot price and the charging spot price are taken from the day-ahead or real-time market clearing price of the corresponding period, the rated capacity of the energy storage is taken from the power station design parameters, the charging and discharging efficiency is taken from the EMS measured statistical value, and the capacity availability rate is determined according to the ratio of the number of available hours in the historical operating data to the total number of hours in the statistical period.

[0027] Preferably, in S2, the AI ​​algorithm takes historical carbon market transaction data, carbon price time series data, policy text data, and historical data of energy storage project filings as input, and uses a joint probability prediction model to output predicted carbon asset returns and their confidence intervals under multiple scenarios.

[0028] In one embodiment, the AI ​​algorithm takes historical carbon market trading data, carbon price time-series data, policy text data, and historical energy storage project registration data as input. The historical carbon market trading data includes daily transaction prices and volumes in national and pilot carbon markets; the carbon price time-series data includes seasonal, cyclical, and trend characteristics within the target prediction period; the policy text data includes national and local carbon market policy documents, with policy strength, coverage, and switching probability characteristics extracted through natural language processing; and the historical energy storage project registration data includes the success rate of registration for similar projects, certification cycles, and emission reduction issuance records. A joint probabilistic prediction model is employed. First, macroeconomic features of the carbon market are extracted through a shared neural network layer. Then, the probability distributions of carbon price fluctuations, policy switching, and registration success rates are output separately for each task branch. After fusion through a Bayesian network, the predicted carbon asset returns and their confidence intervals under multiple scenarios are output.

[0029] Preferably, in S2, the automatic verification method for carbon emission reductions includes: Based on the EMS operation data of energy storage power stations, the discharge amount and marginal emission factor of energy storage to replace thermal power peak shaving are obtained. The carbon emission reduction is automatically calculated by AI and a verification report that meets the requirements of the MRV system is generated. The verified carbon emission reduction is converted into tradable carbon assets and discounted into carbon revenue cash flow.

[0030] In one embodiment, the automatic carbon emission reduction verification method obtains the discharge volume of energy storage replacing thermal power peak shaving based on EMS operation data of energy storage power stations, and combines it with the latest marginal emission factor of the regional power grid to automatically calculate the carbon emission reduction through AI. Specifically, the system extracts the discharge period, discharge volume, and operating efficiency data of energy storage power stations participating in grid peak shaving from the EMS database, matches it with the regional power grid marginal emission factor released by the grid dispatching department, and automatically calculates the carbon emission reduction generated by replacing thermal power peak shaving. Subsequently, the system automatically generates a verification report according to the MRV system requirements. The report includes project boundaries, baseline scenarios, emission reduction calculation formulas, data sources, and uncertainty analysis. The verified carbon emission reduction is converted into tradable carbon assets and discounted into carbon revenue cash flow based on the predicted carbon price, and incorporated into the investment model.

[0031] Preferably, in S3, the five-dimensional coupled cash flow model aggregates the electricity spot peak-valley arbitrage income, capacity leasing income, peak-shaving and frequency regulation ancillary service income, capacity electricity price income and carbon asset income year by year according to the whole life cycle, and constructs an annual net cash flow sequence. The formula for calculating NPV is: ; In the formula, Let r be the net cash flow in year n, r be the weighted average cost of capital (WACC), and N be the project operating period. The formula for calculating the total investment IRR is: .

[0032] In one embodiment, the five-dimensional coupled cash flow model aggregates electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue, capacity electricity price revenue, and carbon asset revenue annually over the project's entire lifecycle to construct an annual net cash flow sequence. The annual net cash flow is the sum of the aforementioned five-dimensional revenues for each year, minus the annual operation and maintenance costs, taxes, and principal and interest repayment expenses. Based on this annual net cash flow sequence, the project's net present value (NPV) is calculated by discounting and summing the annual NPVs using the weighted average cost of capital as the discount rate. An iterative method is then used to find the discount rate that results in a zero NPV, yielding the internal rate of return (IRR) for the entire investment. Simultaneously, the project's internal rate of return on equity and dynamic payback period are calculated, and the distribution of evaluation indicators under different confidence intervals is output.

[0033] Preferably, in S4, the credit enhancement methods for carbon revenue rights pledging include: The carbon revenue cash flow during the forecast period is used as collateral to calculate the loan-to-value ratio and credit enhancement limit; the financing feasibility threshold is determined based on the capital ratio after credit enhancement, debt coverage ratio and minimum acceptable IRR.

[0034] In one embodiment, the carbon revenue rights pledge credit enhancement method uses the carbon revenue cash flow during the forecast period as the pledged collateral, and calculates the credit enhancement limit based on the carbon asset liquidity, project credit rating, and the pledge ratio approved by the financial institution. The financing feasibility threshold is determined comprehensively based on the credit-enhanced capital ratio, debt coverage ratio, and minimum acceptable IRR. Specifically, the credit-enhanced capital ratio is the ratio of the original capital plus the credit enhancement limit to the total investment; the debt coverage ratio is the ratio of the annual cash flow available for principal and interest repayment to the annual principal and interest repayment amount; and the minimum acceptable IRR is set according to industry benchmarks and project risk level. The total investment IRR, capital IRR, and dynamic investment payback period are compared item by item with the above financing feasibility threshold. When all three indicators meet the threshold requirements, the investment decision is approved.

[0035] Preferably, S5 includes: determining the carbon asset development strategy in the investment decision-making stage; deploying EMS and MRV data interfaces in the construction stage; monitoring energy storage charging and discharging data in real time and automatically verifying carbon emission reductions in the operation stage; putting the verified carbon assets into the market for trading in the trading stage; and feeding back the carbon revenue cash flow to the investment evaluation model for dynamic correction in the revenue recovery stage.

[0036] In one embodiment, the investment decision-making stage determines a carbon asset development strategy based on the assessment results, including selecting CCER, local carbon incentives, or other carbon asset trading channels, and formulating a filing and certification plan. During the construction stage, EMS and MRV data interfaces are deployed in the existing monitoring system of the energy storage power station, and data transmission protocols and security encryption modules are configured. During the operation stage, energy storage charging and discharging data are collected in real-time or periodically through the data interfaces, and the system automatically triggers carbon emission reduction calculations and generates certification reports. During the trading stage, the certified carbon assets are listed on the market, and transaction settlement is completed. During the revenue recovery stage, the carbon revenue receipt information is automatically fed back to the investment evaluation model, triggering the five-dimensional coupled cash flow model to recalculate NPV and IRR, and comparing them with the original predicted values. When the deviation exceeds a set threshold, an early warning is issued, and subsequent annual revenue assumptions are revised.

[0037] Preferably, in S5, multi-stage data interaction and processing based on the data interface includes: The system outputs investment plan data during the investment decision-making stage, carbon asset registration data during the carbon asset development stage, energy storage operation data during the operation monitoring stage, carbon asset verification data during the verification and trading stage, and carbon revenue data is fed back to the investment evaluation model during the revenue recovery stage.

[0038] In one embodiment, multi-stage data interaction and processing based on a data interface includes: in the investment decision-making stage, outputting investment plan data, including project capacity, total investment, revenue calculation parameters, and carbon asset development strategy, to the project management platform; in the carbon asset development stage, submitting carbon asset registration data, including project design documents, baseline scenario descriptions, and emission reduction forecast tables, to the MRV system; in the operation monitoring stage, collecting energy storage operation data through standard communication protocols to obtain charging and discharging power, daily power generation, and equipment operating status; in the certification and trading stage, outputting the carbon asset certification report and trading certificate number to the trading settlement system; and in the revenue recovery stage, feeding back carbon revenue amount, trading timestamp, and carbon price data to the investment evaluation model, triggering the model to automatically recalculate and update the investment decision dashboard.

[0039] Preferred, such as Figure 2 As shown, a grid-side energy storage investment decision-making system driven by AI and carbon asset returns includes: The data acquisition module is used to acquire EMS operation data of energy storage power stations, electricity spot trading data, ancillary service settlement data and capacity pricing policy data, and establish a four-dimensional electricity market revenue calculation model covering electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue and capacity pricing revenue, and output the total four-dimensional electricity market revenue. The joint forecasting module is used to acquire carbon market MRV system data, and based on AI algorithms, it performs joint probability forecasting of carbon price time-series fluctuations, carbon market policy switching and carbon asset registration success rate, and outputs carbon asset returns with confidence intervals; carbon asset returns are pre-incorporated as independent revenue cash flow into the investment evaluation model, and an automatic verification method for carbon emission reductions that replace thermal power peak shaving with energy storage is established to generate carbon revenue cash flow. The investment assessment module is used to couple the total revenue of the four-dimensional electricity market with the carbon revenue cash flow to obtain a five-dimensional coupled cash flow model; using the total investment IRR, equity IRR and dynamic investment payback period as evaluation indicators, the NPV and IRR assessment results with confidence intervals are calculated. The comprehensive judgment module is used to establish a carbon revenue rights pledge credit enhancement method based on carbon revenue cash flow and calculate the financing feasibility threshold; it comprehensively judges the total investment IRR, equity IRR and dynamic investment payback period with the financing feasibility threshold, and outputs the judgment result of whether the investment decision is approved or not. The decision generation module is used to output investment decision schemes based on the judgment results, establish a data interface between the EMS operation data of the energy storage power station and the MRV system of carbon assets, and perform multi-stage data interaction and processing based on the data interface.

[0040] In one embodiment, the system comprises a data acquisition module, a joint forecasting module, an investment evaluation module, a comprehensive judgment module, and a decision generation module. The data acquisition module connects to the energy storage power station EMS and the electricity market trading platform via a standard industrial communication protocol, periodically acquiring operational data, spot electricity prices, ancillary service settlement data, and capacity pricing policy data to establish a four-dimensional electricity market revenue calculation model. The joint forecasting module calls a joint probabilistic forecasting model, inputs carbon market and policy data, outputs carbon asset revenue forecasts with confidence intervals, and executes an automatic carbon emission reduction verification process to generate carbon revenue cash flow. The investment evaluation module has a built-in five-dimensional coupled cash flow calculation engine, supporting annual net cash flow aggregation throughout the entire lifecycle and real-time NPV and IRR calculation. The comprehensive judgment module connects to the credit interface of financial institutions, dynamically acquiring the pledge ratio and benchmark interest rate, and calculating the financing feasibility threshold online. The decision generation module aggregates the outputs of all modules, generates an investment decision report, and pushes it to the investor management platform, achieving end-to-end data integration.

[0041] The present invention will be further described in detail below with reference to specific embodiments, so that those skilled in the art can understand it.

[0042] Taking a 100MW / 200MWh grid-side independent energy storage power station as an example, the project has an operating period of 20 years and a weighted average cost of capital (WACC) of 6.5%.

[0043] S1. Obtain the power station's EMS operation data, provincial electricity spot market transaction data, peak-shaving and frequency regulation ancillary service settlement data, and capacity pricing policy documents. Electricity spot market peak-valley arbitrage revenue is calculated based on a day-ahead market peak-valley price difference of 0.45 yuan / kWh, a rated capacity of 100MW, a charge-discharge efficiency of 92%, and one cycle per day, with 300 effective operating days per year, resulting in an annual peak-valley arbitrage revenue of approximately 248.4 million yuan. Capacity leasing revenue is calculated based on a leasing price of 300 yuan / kW·year and an availability rate of 95%, resulting in an annual revenue of approximately 28.5 million yuan. Peak-shaving and frequency regulation ancillary service revenue is calculated based on an annual peak-shaving mileage of 12,000 MW·h, a performance index of 1.2, and a market clearing price of 350 yuan / MW, resulting in an annual revenue of approximately 50.4 million yuan. Capacity pricing revenue is calculated based on an available capacity of 100MW and a compensation standard of 600 yuan / kW·year, resulting in an annual revenue of approximately 60 million yuan. The total annual revenue from the four-dimensional electricity market is approximately 387.3 million yuan.

[0044] S2. Obtain national carbon market MRV system data. The AI-powered probabilistic prediction model takes the carbon price time series data of the past 3 years, policy documents of the past 5 years, and the filing history of similar projects as input, and outputs an annual carbon asset return forecast of RMB 2.28 million (90% confidence interval [195,261] RMB 10,000). Based on EMS data, obtain an annual replacement of thermal power peak-shaving discharge of 50 million kWh. Using the regional grid marginal emission factor of 0.5703 tCO2 / MWh, the AI ​​automatically calculates the annual carbon emission reduction of 28,515 tCO2. After generating an MRV verification report, it is converted into CCER assets. Discounted at a carbon price of RMB 80 / ton, the annual carbon revenue cash flow is RMB 2.28 million.

[0045] S3. Construct a five-dimensional coupled cash flow model, with a net cash flow of approximately 320 million yuan in the first year. The calculated NPV of the project is approximately 185 million yuan; the calculated IRR of total investment is 12.8%, the IRR of equity is 18.5%, and the dynamic investment payback period is 6.8 years.

[0046] S4. The discounted cash flow of the projected carbon revenue over the 10-year period will be used as collateral. The bank has approved a loan-to-value ratio of 60%, with a credit enhancement limit of approximately RMB 9.9 million. After the credit enhancement, the equity ratio will increase from 25% to 31%, the debt coverage ratio will increase from 1.15% to 1.32%, and the minimum acceptable IRR threshold will be 8%. The total investment IRR, equity IRR, and payback period are all better than the thresholds, and the investment is deemed approved.

[0047] S5. Output investment decision plan, deploy EMS and MRV data interfaces, and execute data interaction at each stage in sequence: investment decision stage determines the development CCER strategy; construction stage completes interface deployment; operation stage automatically verifies monthly; transaction stage lists for trading quarterly; profit recovery stage automatically feeds back to the model after funds are received, triggering dynamic correction.

[0048] The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention is determined by the claims.

[0049] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. In case of any inconsistency, the meaning set forth in this specification or derived from the content described herein shall prevail. Furthermore, the terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit the scope of this application.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A grid-side energy storage investment decision-making method driven by AI and carbon asset returns, characterized in that, Includes the following steps: S1. Obtain EMS operation data, electricity spot trading data, ancillary service settlement data, and capacity pricing policy data from energy storage power stations. Establish a four-dimensional electricity market revenue calculation model that covers electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue, and capacity pricing revenue. Output the total revenue of the four-dimensional electricity market. S2. Obtain carbon market MRV system data, and use AI algorithms to perform joint probability prediction on carbon price time-series fluctuations, carbon market policy switching and carbon asset registration success rate, and output carbon asset returns with confidence intervals; incorporate carbon asset returns as independent revenue cash flow into the investment evaluation model in advance, establish an automatic verification method for carbon emission reductions that replace thermal power peak shaving with energy storage, and generate carbon revenue cash flow. S3. Couple the total revenue of the four-dimensional electricity market with the carbon revenue cash flow to obtain a five-dimensional coupled cash flow model; use the total investment IRR, equity IRR and dynamic investment payback period as evaluation indicators to calculate the NPV and IRR evaluation results with confidence intervals. S4. Establish a carbon revenue rights pledge credit enhancement method based on carbon revenue cash flow and calculate the financing feasibility threshold; comprehensively judge the total investment IRR, equity IRR and dynamic investment payback period with the financing feasibility threshold, and output the judgment result of whether the investment decision is approved or not. S5. Output investment decision plan based on the judgment result, establish data interface between energy storage power station EMS operation data and carbon asset MRV system, and carry out multi-stage data interaction and processing based on the data interface.

2. The grid-side energy storage investment decision-making method based on AI and carbon asset returns as described in claim 1, characterized in that, In S1, the electricity spot peak-valley arbitrage profit is calculated based on the spot price difference and charging / discharging efficiency during the charging and discharging periods of the energy storage power station, and the calculation formula is as follows: ； In the formula, The spot electricity price for discharge during period t. The spot electricity price for charging during period t. Where η is the rated energy storage capacity, and η is the charge / discharge efficiency. The duration is specified; capacity leasing revenue is calculated based on the rated energy storage capacity and the unit price of capacity leasing, using the following formula: ； In the formula, This is the unit price for capacity leasing. This refers to capacity availability.

3. The grid-side energy storage investment decision-making method based on AI and carbon asset returns as described in claim 2, characterized in that, In S2, the AI ​​algorithm takes historical carbon market transaction data, carbon price time series data, policy text data, and historical data of energy storage project filing as input, and uses a joint probability prediction model to output the predicted value of carbon asset returns under multiple scenarios and their confidence intervals.

4. The grid-side energy storage investment decision-making method based on AI and carbon asset returns as described in claim 3, characterized in that, In S2, the automatic verification method for carbon emission reductions includes: Based on the EMS operation data of energy storage power stations, the discharge amount and marginal emission factor of energy storage to replace thermal power peak shaving are obtained. The carbon emission reduction is automatically calculated by AI and a verification report that meets the requirements of the MRV system is generated. The verified carbon emission reduction is converted into tradable carbon assets and discounted into carbon revenue cash flow.

5. The grid-side energy storage investment decision-making method based on AI and carbon asset returns as described in claim 4, characterized in that, In S3, the five-dimensional coupled cash flow model aggregates the electricity spot peak-valley arbitrage income, capacity leasing income, peak-shaving and frequency regulation ancillary service income, capacity electricity price income and carbon asset income year by year according to the whole life cycle, and constructs an annual net cash flow sequence. The formula for calculating NPV is: ； In the formula, Let r be the net cash flow in year n, r be the weighted average cost of capital (WACC), and N be the project operating period. The formula for calculating the total investment IRR is: 。 6. The grid-side energy storage investment decision-making method based on AI and carbon asset returns as described in claim 5, characterized in that, In S4, the carbon revenue rights pledge credit enhancement method includes: The carbon revenue cash flow during the forecast period is used as collateral to calculate the pledge ratio and credit enhancement limit; the financing feasibility threshold is determined based on the enhanced capital ratio, debt coverage ratio and minimum acceptable IRR.

7. The grid-side energy storage investment decision-making method based on AI and carbon asset returns as described in claim 6, characterized in that, S5 includes: The investment decision-making stage determines the carbon asset development strategy; the construction stage deploys EMS and MRV data interfaces; the operation stage monitors energy storage charging and discharging data in real time and automatically verifies carbon emission reductions; the trading stage puts the verified carbon assets into the market for trading; and the revenue recovery stage feeds back the carbon revenue cash flow to the investment evaluation model for dynamic correction.

8. The grid-side energy storage investment decision-making method based on AI and carbon asset returns as described in claim 7, characterized in that, In step S5, multi-stage data interaction and processing based on the data interface includes: The system outputs investment plan data during the investment decision-making stage; carbon asset registration data during the carbon asset development stage; energy storage operation data during the operation monitoring stage; carbon asset verification data during the verification and trading stage; and carbon revenue data is fed back to the investment evaluation model during the revenue recovery stage.

9. A grid-side energy storage investment decision-making system driven by AI and carbon asset returns, characterized in that, include: The data acquisition module is used to acquire EMS operation data of energy storage power stations, electricity spot trading data, ancillary service settlement data and capacity pricing policy data, and establish a four-dimensional electricity market revenue calculation model covering electricity spot peak-valley arbitrage revenue, capacity leasing revenue, peak-shaving and frequency regulation ancillary service revenue and capacity pricing revenue, and output the total four-dimensional electricity market revenue. The joint forecasting module is used to acquire carbon market MRV system data, and based on AI algorithms, it performs joint probability forecasting of carbon price time-series fluctuations, carbon market policy switching and carbon asset registration success rate, and outputs carbon asset returns with confidence intervals; carbon asset returns are pre-incorporated as independent revenue cash flow into the investment evaluation model, and an automatic verification method for carbon emission reductions that replace thermal power peak shaving with energy storage is established to generate carbon revenue cash flow. The investment assessment module is used to couple the total revenue of the four-dimensional electricity market with the carbon revenue cash flow to obtain a five-dimensional coupled cash flow model; using the total investment IRR, equity IRR and dynamic investment payback period as evaluation indicators, the NPV and IRR assessment results with confidence intervals are calculated. The comprehensive judgment module is used to establish a carbon revenue rights pledge credit enhancement method based on carbon revenue cash flow and calculate the financing feasibility threshold; it comprehensively judges the total investment IRR, equity IRR and dynamic investment payback period with the financing feasibility threshold, and outputs the judgment result of whether the investment decision is approved or not. The decision generation module is used to output investment decision schemes based on the judgment results, establish a data interface between the EMS operation data of the energy storage power station and the MRV system of carbon assets, and perform multi-stage data interaction and processing based on the data interface.

Images