Battery full life cycle carbon reduction intelligent management and control method fusing peak-valley carbon price

Abstract

The application relates to the field of battery management and carbon emission reduction, and specifically discloses a battery full-life-cycle intelligent carbon-saving management and control method fusing peak-valley carbon prices, which comprises the following steps: establishing a dynamic carbon asset net value model, calculating the asset net value in real time based on residual carbon emission reduction potential and historical carbon cost allocation; generating a dynamic expected carbon cost curve fusing a risk premium coefficient, quantifying the upward risk of future carbon cost; taking maximizing the full-cycle carbon asset net value as a target, performing cross-period optimization solution by fusing a health state attenuation model, and generating a long-term optimal charging and discharging plan; and calibrating the model by using actual data through a rolling execution and closed-loop feedback mechanism. The application realizes a paradigm transition of battery management from energy consumption control to carbon asset operation, solves the problems of long-term carbon risk quantification and asset dynamic valuation, and significantly improves the carbon emission reduction benefit and economic value of the battery full-life-cycle.

Description

Technical Field

[0001] This invention relates to the fields of battery management and carbon emission reduction technology, specifically to a smart management and control method for carbon saving throughout the entire battery lifecycle that integrates peak and valley carbon prices. Background Technology

[0002] As battery energy storage systems play an increasingly crucial role in power system peak shaving and renewable energy consumption, their carbon emission reduction potential throughout their entire life cycle has attracted much attention. Existing battery management systems mainly focus on battery physical state management, safety protection, and peak-valley arbitrage optimization based on electricity prices. The core of their control logic lies in optimizing economic operating costs or delaying battery physical degradation.

[0003] While existing technologies have begun to incorporate carbon factors into energy storage scheduling—for example, some schemes consider adding fixed carbon allowance costs or simple real-time carbon price signals to optimization models—these methods have significant limitations. First, they typically treat carbon costs as static or short-term external parameters, failing to dynamically assess and quantify the carbon reduction value inherent in the battery itself from an asset perspective, and ignoring the accumulation and allocation of carbon costs at each stage of battery production and recycling. Second, existing methods lack consideration for the long-term volatility of the carbon market and policy uncertainties, and have not established a correlation model between the decline in the physical health of the battery and the reduction in its carbon asset value, leading to short-term control strategies and failing to achieve the preservation and appreciation of carbon asset value throughout its entire lifecycle. Finally, for cascaded-use batteries, where value assessment is particularly complex and carbon management is almost nonexistent, existing technologies lack a targeted management framework capable of transforming their low-value-added physical states into high-value carbon assets.

[0004] Therefore, the industry urgently needs an innovative management and control method that can deeply couple dynamic carbon price signals, the carbon footprint of the battery throughout its entire life cycle, and its health status. This would not only enable real-time carbon cost optimization but also facilitate long-term value management and risk mitigation of batteries as a carbon asset. Summary of the Invention

[0005] The purpose of this invention is to provide a smart management and control method for carbon saving throughout the entire life cycle of batteries that integrates peak and valley carbon prices, in order to solve the technical problems in the prior art that the carbon emission reduction potential cannot be maximized and the carbon economic benefits throughout the entire life cycle are low due to the lack of dynamic valuation of batteries as carbon asset units, quantification of long-term carbon market risks, and intertemporal value preservation and appreciation capabilities.

[0006] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution: A method for intelligent carbon-saving management and control of batteries throughout their entire lifecycle, integrating peak and valley carbon pricing, includes the following steps: S1. Establish a dynamic carbon asset net value model for batteries in the cascade utilization stage and calculate their carbon asset net value in real time; S2. Based on the dynamic carbon asset net value model and long-term market forecast data, generate a dynamic expected carbon cost curve covering the remaining battery life; S3. Using the dynamic expected carbon cost curve as the core input, and aiming to maximize the net value of carbon assets throughout the entire cycle, solve to generate an inter-period charging and discharging strategy; S4. Execute the intertemporal charge-discharge strategy and update the dynamic carbon asset net value model and dynamic expected carbon cost curve on a rolling basis based on actual operating data.

[0007] As a preferred embodiment of the present invention, S1 specifically includes: S11. Acquire and analyze the battery's carbon footprint at the factory, historical charge and discharge data during the main use phase, and corresponding carbon emission data. Use a preset carbon emission reduction accounting methodology model to assess and quantify the battery's remaining carbon emission reduction potential after the main use phase ends, and confirm the quantified value as the underlying asset of the dynamic carbon asset net value model. S12. Obtain the carbon cost of the battery during the production stage. Based on the expected remaining lifespan and available capacity decay curve of the secondary utilization scenario, the carbon cost of the production stage is allocated to the future operating cycle using a preset amortization rule to form historical carbon cost amortization liability. At the same time, based on the typical operating load and maintenance plan of the secondary utilization scenario, the future operation and maintenance carbon cost is predicted, which together with the historical carbon cost amortization liability constitutes the model liability. S13. Real-time access to the real-time carbon price data stream of the carbon trading market, calculate the difference between the revalued underlying asset value and liability value, and output the updated net carbon asset value of the battery in real time.

[0008] As a preferred embodiment of the present invention, S13 specifically includes: S131. Establish an interface with the carbon trading market data platform to obtain real-time carbon price data streams; perform timestamp alignment, outlier filtering, and validity verification on the obtained data to generate standardized, verified real-time carbon prices; S132. The verified real-time carbon price is used as the core value coefficient and input into the asset revaluation method and the liability revaluation method respectively; the asset revaluation method multiplies the verified real-time carbon price by the current achievable proportion of the remaining carbon emission reduction potential to calculate the revalued value of the underlying assets; the liability revaluation method multiplies the verified real-time carbon price by the current allocation of the model liability to calculate the revalued value of the liabilities. S133. Subtract the revalued liability value from the revalued underlying asset value to calculate the current carbon asset net value in real time; and output the carbon asset net value result to the dynamic database, which also serves as the current state value of the dynamic carbon asset net value model, providing input for step S2.

[0009] As a preferred embodiment of the present invention, S2 specifically includes: S21. Real-time acquisition of future multi-period grid carbon intensity forecast data from the grid dispatch system, clean energy power generation plan data from the power generation side, and long-term carbon price trend forecast data from the carbon trading market; alignment and fusion of the above multi-source data on a time scale to form a unified time series long-term carbon market forecast data covering several future dispatch cycles to the end of the expected battery life. S22. Based on the long-term carbon market forecast data and combined with the charge and discharge efficiency model of the cascaded battery, calculate the expected carbon emissions and expected carbon costs corresponding to a unit charge and discharge behavior in each future period, and generate an initial basic expected carbon cost sequence. S23. Access historical volatility data of the carbon market and calendar of major policy events. Based on the uncertainty of the long-term carbon market forecast data, dynamically calculate the risk premium coefficient for each future period through a risk assessment model. The risk assessment model is used to quantify the risk of rising carbon costs due to carbon market volatility, policy adjustments, or uncertainty in renewable energy output. S24. Add the expected carbon cost value for each period in the basic expected carbon cost sequence to the risk premium coefficient for the corresponding period to synthesize the final dynamic expected carbon cost curve including risk adjustment, and output it to the subsequent optimization decision-making process.

[0010] As a preferred embodiment of the present invention, S23 specifically includes: S231. Real-time collection of multi-dimensional risk data for quantifying uncertainty, including historical volatility data of carbon prices in the carbon trading market, statistical values ​​of renewable energy forecast deviations reflecting the uncertainty of clean energy output, and policy and event calendar data that identify major events that may affect the supply and demand of the carbon market; S232. Input the long-term carbon market forecast data and multi-dimensional risk data into a preset risk assessment model. The risk assessment model comprehensively evaluates the deviation of the forecast data from historical fluctuation patterns, the potential impact level of planned events during the forecast period, and the transmission effect of clean energy uncertainty on the carbon intensity of the power grid, and dynamically calculates the risk premium coefficient corresponding to each future period. S233. The calculated risk premium coefficients for each time period are organized according to the time index that is completely aligned with the basic expected carbon cost sequence to form a risk premium coefficient time series that can be directly used for algebraic superposition, and output to the subsequent S24 step for curve synthesis.

[0011] As a preferred embodiment of the present invention, S3 specifically includes: S31. With maximizing the cumulative net carbon asset value increment after discounting over the entire period from the current moment to the end of the battery's expected lifespan as the core optimization objective, an intertemporal decision function is constructed. The intertemporal decision function sets the decision variable for each future time period as the battery's charging and discharging power, and takes the carbon cost value of each time period in the dynamic expected carbon cost curve as the core economic parameter input. The total planned carbon cost is calculated by multiplying the planned charging and discharging behavior with the carbon cost value of the corresponding time period and summing them up. S32. Load the running constraints for the intertemporal decision function, and couple the pre-established battery health state degradation model with the intertemporal decision function. The degradation model predicts the degradation of battery health state and available capacity at each future time node, and integrates the degradation amount into the objective function. S33. Using dynamic programming or model predictive control algorithms, the intertemporal decision function coupled with the influence of health state decay is solved in a rolling or global manner under the premise of satisfying all constraints. The solution is a long-term optimal charge and discharge plan covering multiple future scheduling cycles. While ensuring real-time operational safety, the plan actively guides the battery to reduce discharge during high-cost periods and increase discharge or charging during low-cost periods in the dynamic expected carbon cost curve, thereby maximizing the net value of carbon assets throughout the entire cycle.

[0012] As a preferred embodiment of the present invention, S32 specifically includes: S321. Based on the physical parameters of the batteries used in the second phase and the requirements of the application scenarios, formally define a set of mathematical inequalities that constitute the constraints; the set of conditions shall include at least the upper and lower limits of charging and discharging power determined by the rated power of the battery, the upper and lower limits of the state of charge defined by the safe operating range, and the net discharge capacity requirement within the cycle determined by the load requirements of the second phase application scenario. S322. The battery health state degradation model is embedded into the optimization framework. The battery health state degradation model predicts the degradation trajectory of the battery health state with charge and discharge cycles based on the assumed future charge and discharge power sequence in the intertemporal decision function. A health state-asset value mapping function is established. The function maps the predicted health state degradation amount to the expected reduction in the future value of the underlying assets in the dynamic carbon asset net value model, and quantifies this expected reduction amount into a health state degradation penalty term that is proportional to the degradation amount. S323. The intertemporal decision function is taken as the core objective, the constraints are taken as hard boundaries, and the health state decay penalty term is taken as a negative benefit cost term and added to the core objective function. This constructs a complete coupled optimization model that integrates economic objectives, physical operation constraints and long-term asset preservation requirements, which is used to solve step S33.

[0013] As a preferred embodiment of the present invention, S4 specifically includes: S41. Send the charging and discharging power command corresponding to the current scheduling period in the long-term optimal charging and discharging plan to the battery BMS actuator for control; synchronously monitor the battery operating status data and the actual carbon intensity data from the power grid in real time, wherein the battery operating status data includes at least the actual state of charge, health status, temperature and actual throughput. S42. Using a preset fixed period as the rolling trigger condition, when the trigger condition is met, the latest monitored battery operating status data, the latest acquired actual carbon intensity data, and the updated long-term forecast data re-acquired from the external market are used as new inputs to repeat steps S2 and S3; by regenerating the dynamic expected carbon cost curve and solving the new inter-period decision function, the subsequent long-term optimal charge and discharge plan that has not yet been executed is rolled up and dynamically corrected. S43. After the completion of an accounting cycle synchronized with the carbon market accounting cycle, calculate the corresponding actual carbon market value and health status prediction deviation, and input them together into the dynamic carbon asset net value model in S1 to calibrate the model's basic asset value, liability cost parameters, and decay model parameters, thus completing the closed loop from strategy execution to value feedback.

[0014] As a preferred embodiment of the present invention, S43 specifically includes: S431. Based on the battery operating status data and actual carbon intensity data recorded during the accounting period, the same carbon emission reduction accounting methodology model as step S11 is used to calculate the actual carbon emission reduction generated by the battery due to optimized control during the period; the actual carbon emission reduction is multiplied by the average carbon price in the carbon trading market during the accounting period to calculate its corresponding actual carbon market value. S432. Obtain the actual measured value of the battery health status at the end of the accounting cycle, compare the actual measured value with the predicted value of the health status at the end of the cycle by the battery health status decay model at the beginning of the cycle in step S32, and calculate the health status prediction deviation. S433. The actual carbon market value is used as an asset-side calibration signal and input into the dynamic carbon asset net value model in S1 to positively correct the valuation parameters of the underlying assets of the model or verify its effectiveness; at the same time, the health status prediction deviation is used as a model performance feedback signal and input into the battery health status decay model in step S3.2 to reversely correct the internal parameters of the decay model in order to improve the accuracy of its subsequent predictions.

[0015] A battery lifecycle carbon-saving intelligent management and control system integrating peak and valley carbon pricing, used to implement a battery lifecycle carbon-saving intelligent management and control method integrating peak and valley carbon pricing, including: The carbon asset net value management module is used to build and update a dynamic carbon asset net value model in real time for batteries in the cascade utilization stage. The model takes the battery's remaining carbon emission reduction potential as the basic asset, the allocated historical carbon cost and operation and maintenance carbon cost as liabilities, and calculates and outputs the battery's carbon asset net value based on real-time carbon price fluctuations. The carbon cost prediction and risk assessment module is connected to the carbon asset net value management module. It is used to generate a dynamic expected carbon cost curve covering the expected remaining life of the battery based on the carbon asset net value, long-term market forecast data and risk data. The curve includes the expected carbon cost for each period after risk adjustment. The optimization decision and plan generation module is connected to the carbon asset net value management module and the carbon cost prediction and risk assessment module, respectively. It is used to maximize the carbon asset net value over the whole life cycle, with the dynamic expected carbon cost curve as the core input, and integrates the battery health state decay model to perform inter-period optimization solution to generate the long-term optimal charge and discharge plan. The execution and feedback control module, connected to the optimization decision and plan generation module, is used to execute the current instructions in the charge and discharge plan and monitor battery operation and grid carbon intensity data in real time. It also triggers at fixed intervals to drive the carbon cost prediction and risk assessment module and the optimization decision and plan generation module to perform rolling optimization using the latest data. At the same time, after a calculation cycle, it calculates the carbon emission reduction value and health status deviation based on actual data and feeds the results back to the carbon asset net value management module for model parameter calibration, forming a closed loop.

[0016] Compared with the prior art, the present invention has the following advantages: 1. By constructing a dynamic carbon asset net value model, batteries are defined as carbon assets that can be valued in real time. This allows the control objective of the BMS to shift from optimizing immediate energy consumption costs to maximizing the net value of carbon assets throughout the entire life cycle, achieving a fundamental shift from equipment control to asset operation.

[0017] 2. By generating a dynamic expected carbon cost curve with integrated risk premium coefficient and establishing a quantitative correlation between health status decline and asset impairment, the system is equipped with a forward-looking risk management capability, enabling it to proactively avoid future high carbon cost risks and balance the long-term asset preservation needs.

[0018] 3. Through rolling optimization and a dual feedback calibration mechanism based on actual data, the system can dynamically adjust its strategy according to changes in the market and battery status, and continuously self-correct and optimize the model, ensuring the real-time adaptability, continuous optimization and verification credibility of carbon asset management. Attached Figure Description

[0019] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the method described in Embodiment 1 of the present invention.

[0021] Figure 2 This is a framework diagram of the system described in Embodiment 2 of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The concepts involved in this application will first be described with reference to the accompanying drawings. It should be noted that the following descriptions of various concepts are only for the purpose of making the content of this application easier to understand and do not constitute a limitation on the scope of protection of this application; furthermore, the embodiments and features in the embodiments of this application can be combined with each other unless otherwise specified. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0024] Example 1

[0025] like Figure 1 As shown, this invention provides a method for intelligent carbon-saving management and control of the entire battery lifecycle that integrates peak and valley carbon prices, including the following steps: S1. Establish a dynamic net carbon asset value model for batteries in the secondary utilization stage, and calculate their net carbon asset value in real time; specifically including: S11. Basic asset valuation and quantification, specifically: S111. Obtain the battery's carbon footprint data at the time of manufacture. This data covers the greenhouse gas emissions throughout the entire process of battery production, from raw material acquisition, electrode preparation, cell assembly, module integration to the finished system. The aforementioned carbon footprint data is then structured and analyzed. Based on the product category rules of the greenhouse gas accounting system, carbon emissions at each stage are classified into direct emissions, energy-related indirect emissions, and other indirect emissions. A phased and hierarchical carbon footprint database is established as the benchmark dataset for subsequent full life-cycle carbon flow tracking.

[0026] S112. Acquire historical charge and discharge data during the primary usage phase. This data records the battery's electrical behavior under all operating conditions during its initial application scenario. Preprocess the historical charge and discharge data, including missing value imputation, smoothing of abnormal transition points, multi-source sensor data fusion calibration, and time series alignment.

[0027] S113. Based on preprocessed historical charge and discharge data, calculate the corresponding carbon emission data for the main usage phase. Specifically, based on the grid carbon emission factor or the zero-carbon attribute of on-site renewable energy generation during each charge and discharge period, convert electrical energy flow into carbon emission flow. For the charging process, calculate the implicit carbon emissions from external energy input based on the charging amount, charging period, and the grid marginal emission factor or green electricity traceability certificate for that period; for the discharging process, calculate the baseline scenario carbon emissions to be avoided based on the discharging amount and the emission intensity of the alternative benchmark energy system.

[0028] Attribution analysis was performed on the above carbon emission data to distinguish between inherent emissions caused by battery energy conversion efficiency loss, time-varying emissions caused by changes in power grid structure, and indirect emissions caused by fluctuations in renewable energy penetration. A mapping relationship between carbon emissions and operating conditions, energy structure, and environmental conditions was established to form a spatiotemporal distribution dataset of carbon emissions during the main use phase.

[0029] S114. Employ a pre-defined carbon emission reduction accounting methodology model to evaluate the cumulative carbon emission reduction benefits achieved by the battery during its primary use phase. This methodology model follows the basic principles of voluntary greenhouse gas emission reduction project methodologies, defining project boundaries, identifying baseline scenarios, quantifying project emissions and leaks, and calculating emission reductions.

[0030] The project boundary definition covers the battery system itself and its directly associated energy input and output interfaces; the baseline scenario is set as the carbon emission level of alternative technologies that meet the same energy service needs without the battery system; the project emission accounting covers the energy consumption of the battery charge and discharge cycle, the auxiliary energy consumption of thermal management, and the self-consumption of the battery management system; leakage considerations include potential emission escape during the battery retirement and disposal stage.

[0031] Based on the above methodological framework, the cumulative carbon emission reduction during the main use phase is calculated, which is the difference between baseline scenario emissions and actual project emissions and leakage, as the carbon emission reduction contribution already achieved by the battery.

[0032] S115. Combining the current state of battery health degradation and available capacity retention rate, estimate the remaining carbon emission reduction potential after the main use phase ends. Specifically, based on the health status assessment results determined by capacity testing, electrochemical impedance spectroscopy analysis, and differential capacity curve analysis at the end of the main use phase, predict the available capacity degradation trajectory and expected cycle life of the battery in the secondary use scenario.

[0033] Based on this degradation trajectory, the equivalent number of full cycles and total energy throughput that the battery can complete in the tiered utilization stage are estimated; combined with the expected grid carbon emission factor and renewable energy consumption ratio of the tiered utilization scenario, the carbon emissions of the benchmark scenario that this energy throughput can replace are predicted; after deducting the incremental emissions caused by the decline in operating efficiency, the increase in maintenance energy consumption and the accelerated performance degradation in the tiered utilization stage, the net remaining carbon emission reduction potential is obtained.

[0034] The quantified value of the remaining carbon emission reduction potential is identified as the underlying asset of the dynamic carbon asset net value model. This value represents the maximum carbon emission reduction contribution that the battery can theoretically achieve in the cascade utilization stage, providing a value benchmark for subsequent real-time carbon price revaluation and operation strategy optimization.

[0035] S12. Model liability construction, specifically: S121. Obtain the carbon cost of the battery production stage, which includes the carbon footprint data of each production stage in step S111, and the comprehensive monetary value after conversion by the carbon trading market pricing mechanism or social carbon cost parameters.

[0036] A structural analysis of carbon costs during the production phase is conducted, distinguishing between sunk carbon costs (one-time investments) and amortizable capitalized carbon costs. Sunk carbon costs include unrecoverable carbon emissions due to irreversible process losses, pilot production, and yield reductions; capitalized carbon costs include net carbon inputs that can be recovered through subsequent energy services, forming the basis for subsequent amortization calculations. Simultaneously, the temporal distribution characteristics of carbon costs during the production phase are identified, differentiating the heterogeneity of carbon costs across different production batches, different technological routes, and different supply chain sources, thus establishing a refined ledger of carbon costs during the production phase.

[0037] S122. Assess the expected remaining lifespan of the battery in the secondary use scenario, which characterizes the battery's sustainable service life from retirement to final disposal under the current application scenario. Specifically, based on the health status diagnosis results at the end of the main use phase, combined with the application characteristics of the secondary use scenario, such as shallow charge-discharge cycles in energy storage power stations, float charging standby mode of backup power supplies, or intermittent use of low-speed electric vehicles, predict the capacity decay kinetics of the battery under the current stress conditions.

[0038] This prediction employs a physics-based degradation mechanism model or a data-driven lifetime prediction algorithm, comprehensively considering the coupled effects of temperature stress, current rate, depth of discharge, and calendar aging to generate a probability distribution estimate of the expected remaining lifetime, typically quantified by the remaining equivalent full cycle count or remaining service years. Simultaneously, a dynamic correction mechanism for the expected remaining lifetime is established, periodically updating the lifetime prediction model parameters and adjusting the estimated expected remaining lifetime as operational data accumulates during the tiered utilization phase.

[0039] S123. Construct an available capacity decay curve for the secondary use scenario. This curve describes the degradation trajectory of the effective energy that the battery can release over its expected remaining lifespan over time or the number of cycles. Specifically, based on historical capacity decay data from the primary use phase, fit an empirical decay model, such as a double exponential decay model, a square root time model, or a thermal acceleration model based on the Arrhenius equation, and extrapolate it to the operating conditions of the secondary use scenario.

[0040] This curve needs to consider the differences in operating conditions between the secondary utilization scenario and the primary use stage. The primary use stage is typically a high-rate, deep-discharge power application, while the secondary utilization stage is typically a low-rate, shallow-discharge energy storage application. The change in operating conditions may lead to changes in the decay rate. By adjusting the stress coupling coefficient in the model, an available capacity decay curve adapted to the secondary utilization scenario is generated. This curve outputs the available capacity retention rate at each time point in discrete time point form or as a continuous function.

[0041] S124. Carbon costs from the production phase are allocated to future operating cycles using a pre-defined amortization rule. This amortization rule is determined based on the techno-economic characteristics of the cascade utilization scenario and carbon accounting policy requirements. Optional models include: Straight-line amortization method: The capitalized carbon cost is evenly distributed over the expected remaining lifespan, with equal amortization amounts in each period. It is suitable for scenarios with stable operating loads and uniform value consumption. Accelerated amortization method: The amortization ratio in the early stage is higher than that in the later stage, reflecting the characteristics of battery performance being higher in the early stage and lower in the later stage, and the energy service value being greater in the early stage. The double-declining balance method or the sum-of-the-years method can be selected. Production-based amortization: This method allocates energy based on the proportion of actual energy throughput to the expected total throughput in each cycle, and is suitable for scenarios with large fluctuations in operating load.

[0042] Based on the selected amortization rule and combined with the expected remaining lifespan and available capacity decay curves, the current period's amortization amount for each future operating cycle is calculated, forming a time series of historical carbon cost amortization liabilities. This liability represents the carbon cost burden that the battery needs to repay periodically during the tiered utilization stage due to historical production investments.

[0043] S125. Based on typical operating loads and maintenance plans for cascade utilization scenarios, predict future operation and maintenance carbon costs. Specifically: Based on typical operating loads, the inherent energy loss caused by the charge-discharge energy conversion efficiency being lower than the ideal value is calculated. This loss needs to be replenished by external energy, corresponding to indirect carbon emission costs. The energy consumption of the thermal management system that maintains the battery in a suitable temperature range is calculated, including cooling energy consumption and heating energy consumption. This energy consumption is affected by both environmental climate conditions and load intensity. The continuous self-discharge of the active balancing circuit of the battery management system, as well as the auxiliary energy consumption of communication, monitoring and safety protection functions, are calculated.

[0044] Based on the maintenance plan, predict the carbon emission costs of maintenance operations such as periodic inspections, capacity calibration, fault repair and module replacement, including the energy consumption of maintenance equipment, the implicit carbon emissions of replacement parts and the transportation energy consumption of maintenance personnel.

[0045] The above-mentioned maintenance carbon costs are summarized to form a future maintenance carbon cost prediction curve covering the expected remaining lifespan.

[0046] S126. Align the historical carbon cost allocation liability time series obtained in step S124 with the future operation and maintenance carbon cost prediction curve obtained in step S125 on the same time base, and accumulate them period by period to form a comprehensive model liability.

[0047] The model's liabilities are output in a structured manner, including total liabilities, current period allocation, remaining unequal allocation, and liability maturity structure. Attribute labels are also provided for each liability component, offering a detailed data foundation for the liability re-estimation in step S132. These model liabilities correspond to the underlying assets identified in step S115, together forming the balance sheet structure of the dynamic carbon asset net value model.

[0048] S13. Real-time carbon price revaluation and net worth calculation, specifically: S131. Establish a real-time data connection with the carbon trading market data platform, which is implemented through a standardized application programming interface or a dedicated data push protocol. Continuously receive carbon price data streams through polling or subscription push modes. Parse the received raw data streams using the protocol, extracting carbon price values, pricing units, trading instrument codes, data release timestamps, and data source identifiers, and convert them into an internally unified data structure. Perform timestamp alignment processing on the acquired carbon price data to ensure consistency with the time base of battery operation data. Employ statistical anomaly detection algorithms or preset threshold rules to identify and filter outomas caused by network transmission errors, abnormal market fluctuations, or data platform malfunctions. Through data integrity verification and source validity verification, exclude unauthorized or untrusted data sources and generate standardized, verified real-time carbon prices.

[0049] S132. Implementation of the Asset and Liability Revaluation Approach: a. The verified real-time carbon price is used as the core value coefficient and synchronously input into both the asset revaluation and liability revaluation methods. This carbon price coefficient acts as a bridge connecting physical carbon emissions and financial value, converting the physical quantities of underlying assets and model liabilities into economic value under current market conditions. A dynamic caching and version control mechanism for the carbon price coefficient is established to record the carbon price values, timestamps, and data sources used in each revaluation, supporting the traceability needs of subsequent accounting and auditing. Simultaneously, the fluctuation range of carbon prices is monitored. When price fluctuations exceed a preset threshold within a short period, an emergency response mode for revaluation is triggered, increasing the update frequency to capture market changes.

[0050] b. Perform an asset revaluation method, which multiplies the verified real-time carbon price by the current achievable proportion of remaining carbon emission reduction potential to calculate the revalued value of the underlying assets. Specifically, the current achievable proportion of remaining carbon emission reduction potential is calculated based on the total remaining carbon emission reduction potential assessed in step S115, combined with the expected remaining lifespan determined in step S122 and the current operational stage, to calculate the percentage of potential that can be actually converted into emission reduction benefits in the current period.

[0051] The determination of this ratio takes into account the following factors: the energy service contract constraints of batteries under the tiered utilization scenario, the proportion of dispatchable capacity to remaining available capacity in the current period; the time-specific characteristics of grid carbon emission factors, the impact of the current grid cleanliness on the realization of emission reduction benefits; and the technical feasibility constraints, the impact of the current battery health status on energy conversion efficiency. Multiplying the verified real-time carbon price by the above-mentioned current achievable ratio and the total remaining carbon emission reduction potential yields the revalued value of the underlying assets, reflecting the realizable value of the underlying assets under current market conditions.

[0052] c. Perform a liability revaluation method. This algorithm multiplies the verified real-time carbon price by the current period allocation of the model liability to calculate the revalued liability value. Specifically, the current period allocation of the model liability is determined based on the historical carbon cost allocation rules in step S124 and the future operation and maintenance carbon cost forecasts in step S125 to determine the amount of liability to be recognized in the current period.

[0053] For historical carbon cost amortization liabilities, the current amortization amount is directly calculated from the preset amortization rules; for future operation and maintenance carbon costs, the current amortization amount is determined based on the expected execution of typical operating loads and maintenance plans in the current period. Multiplying the verified real-time carbon price by the above-mentioned current amortization amount yields the revalued liability value reflecting the actual burden of the liability under current market conditions.

[0054] The attribute tags that distinguish the components of liabilities are used to calculate the revaluation value of historical amortized liabilities and the revaluation value of operation and maintenance forecast liabilities, supporting subsequent detailed analysis of the net value structure of carbon assets.

[0055] S133. Calculation and Output of Net Carbon Asset Value: a. Calculate the difference between the revalued underlying asset value and the revalued liability value. The result is the net carbon asset value of the battery at the current moment. This calculation follows the accounting equation: net carbon asset value equals the revalued value of the underlying asset minus the revalued value of the model liability.

[0056] The numerical characteristics of the calculation results are analyzed as follows: When the net carbon asset value is positive, it indicates that the value of the underlying assets exceeds the liability burden, the battery has a positive net carbon emission reduction contribution capacity in the cascade utilization stage, and its continued operation under the current market conditions is reasonable in terms of carbon benefits; when the net carbon asset value is negative, it indicates that the liabilities exceed the assets, the battery operation has generated a net carbon burden, and carbon performance needs to be improved by optimizing the operation strategy, extending the service life, or participating in the carbon credit mechanism; when the net carbon asset value approaches zero, it indicates that the battery is in a critical state of carbon benefits, and the rationality of continuing operation needs to be carefully assessed.

[0057] Simultaneously, the marginal rate of change of carbon asset net value is calculated, that is, the magnitude of net value change caused by a unit change in carbon price, which characterizes the sensitivity of carbon assets to market prices and provides a reference for risk management and hedging strategies.

[0058] b. The calculated net carbon asset value is written to a dynamic database for persistent storage. A data partitioning and indexing mechanism is established to support fast retrieval by multiple dimensions such as time range, battery identification, and tiered utilization scenario.

[0059] The calculated net carbon asset value is set as the current state value of the dynamic net carbon asset value model. This state value represents the instantaneous output of the model at the current moment and reflects the latest assessment results of the value of battery carbon assets.

[0060] The current state value is passed as the core input parameter to step S2 to generate the dynamic expected carbon cost curve.

[0061] S2. Based on a dynamic carbon asset net worth model and long-term market forecast data, generate a dynamic expected carbon cost curve covering the remaining battery life; specifically including: S21. Multi-source data acquisition and spatiotemporal fusion, specifically: S211. Real-time acquisition of future multi-period grid carbon intensity forecast data from the grid dispatching system. This data is calculated by the grid dispatching agency based on generation plans, load forecasts, and unit combination optimization results, representing the average CO2 emission intensity of the marginal generation combination of the grid in each future period. Acquisition channels include the application programming interface of the grid dispatching data platform, the electricity market information disclosure system, or authorized dispatching operation data sharing mechanisms.

[0062] S212. Real-time acquisition of clean energy power generation plan data from the power generation side. This data covers planned output curves for wind power, photovoltaic power, hydropower, nuclear power, and other non-fossil energy power generation, provided by clean energy power generation companies or centralized new energy power forecasting systems. Acquisition channels include new energy dispatch management systems, power generation company data interfaces, or new energy trading plan disclosures from power trading centers.

[0063] S213. Real-time acquisition of long-term carbon price trend forecast data from the carbon trading market. This data is generated by carbon market analysis agencies, financial institutions, or certified carbon price forecasting models. It forecasts future carbon price trends based on macroeconomic trends, total carbon emission control targets, quota allocation schemes, offsetting mechanism policies, and market supply and demand balance analysis. Acquisition channels include carbon trading market data service providers, research reports from carbon asset management institutions, or outputs from self-developed carbon price forecasting models.

[0064] S214. Align and fuse the above multi-source data on a time scale to form a unified time series. The final output is a unified time series long-term carbon market forecast data covering several future scheduling cycles until the end of the battery's expected lifespan. This dataset includes the grid carbon intensity forecast, clean energy output share, and carbon price forecast for each period.

[0065] S22. Calculation of the basic expected carbon cost sequence, specifically: S221. Call the charge-discharge efficiency model of the secondary battery, which characterizes the energy conversion loss characteristics of the battery in its current health state. Based on the health state assessment results in step S115 and the available capacity decay curve in step S123, this model establishes a mapping relationship between charge-discharge power, state of charge, temperature, and energy conversion efficiency.

[0066] The model inputs include: expected charge / discharge power levels, current state of charge range, ambient temperature conditions, and estimated battery core temperature; the model outputs include: DC-side charge / discharge efficiency, AC-DC conversion efficiency, thermal management system efficiency ratio, and overall energy conversion efficiency under this operating condition. This efficiency model needs to consider the efficiency degradation caused by increased battery internal resistance and intensified polarization during the secondary use stage, and should be corrected relative to the new battery state during the primary use stage.

[0067] S222. Combining the grid carbon intensity forecast from long-term carbon market forecast data with the proportion of clean energy output, calculate the expected carbon emissions corresponding to a unit charge-discharge behavior of the cascaded battery in each future period. Specifically, for charging behavior, calculate the implicit carbon emissions incurred due to obtaining electricity from the grid: divide the grid carbon intensity forecast for the period by the comprehensive energy conversion efficiency to obtain the upstream carbon emissions per unit of charging power; for discharging behavior, calculate the carbon emission reduction benefits achieved by replacing grid power supply: multiply the grid carbon intensity forecast for the period by the comprehensive energy conversion efficiency to obtain the replacement emission reduction per unit of discharging power.

[0068] Distinguish between zero-carbon and low-carbon electricity periods in clean energy power generation plan data. When charging is carried out during these periods, adjust the carbon emission calculation coefficient for charging based on clean energy consumption certificates or green electricity trading contracts to reflect the carbon benefits of prioritizing the consumption of clean energy.

[0069] S223. Combine the expected carbon emissions calculated in step S222 with the carbon price forecast from long-term carbon market data to calculate the expected carbon cost per unit of charging and discharging activity. Specifically, multiply the expected carbon emissions from charging by the corresponding carbon price forecast for that period to obtain the charging carbon cost; multiply the emission reduction from discharging by the corresponding carbon price forecast for that period to obtain the discharging carbon revenue, represented as a negative cost. This calculation considers the time dimension of the carbon price forecast; the long-term carbon cost uses the predicted carbon price discounted to the present value using an appropriate discount rate, forming a comparable carbon cost measurement across periods.

[0070] S224. Summarize the unit charge / discharge carbon costs for each future time period to generate an initial baseline expected carbon cost sequence. This sequence is plotted with the time index on the horizontal axis and the expected net carbon cost per unit of energy or power on the vertical axis, forming a baseline cost curve covering the forecast time range. This sequence only reflects the expected carbon cost based on deterministic forecasts and has not yet incorporated adjustments for uncertainty risks, serving as a benchmark for subsequent risk premium overlay.

[0071] S23. The risk premium coefficient is calculated dynamically, specifically as follows: S231. Real-time acquisition of multi-dimensional risk data for quantifying uncertainty: The collection of historical volatility data for carbon prices includes: acquiring historical trading data from the carbon trading market and calculating the logarithmic return series of carbon prices at different time scales; calculating historical volatility indicators based on this series, including standard deviation, mean absolute deviation, and realized volatility; and using a generalized autoregressive conditional heteroscedasticity model or a stochastic volatility model to predict future volatility trends and generate a volatility term structure.

[0072] The collection of statistical values ​​for renewable energy forecast deviations includes: obtaining historical forecast records and actual output records of clean energy power generation plans, calculating the forecast error sequence; analyzing the statistical distribution characteristics of the errors, such as mean, variance, skewness, kurtosis, and time series characteristics, such as autocorrelation and seasonality; identifying extreme forecast deviation events and their correlation with meteorological conditions, and establishing a quantitative model for forecast uncertainty.

[0073] The collection of policy and event calendar data includes: tracking the dynamics of carbon market-related policies, including adjustments to quota allocation schemes, offset mechanism reforms, market expansion plans, and progress in carbon tax legislation; establishing a calendar of major events, marking the dates of policy releases, market compliance deadlines, major meetings, and macroeconomic data releases; and assessing the potential market impact of each event to form an event risk matrix.

[0074] S232. Implementation of Risk Assessment Model and Calculation of Risk Premium Coefficient: Forecast Deviation Assessment: Compare the carbon price forecasts in long-term carbon market forecast data with the price distribution generated by random simulation based on historical volatility, and calculate the quantile position of the forecast value in the distribution; when the forecast value is in the upper tail region of the historical distribution, it indicates that there is an overestimation risk, and the risk premium is lowered; when it is in the lower tail region, it indicates that there is an underestimation risk of upside potential, and the risk premium is raised.

[0075] Event Impact Assessment: For each planned event identified in the policy and event calendar, based on the event type, market reactions to similar historical events, and current market sensitivity, the potential impact of the event on carbon prices is quantified; the impacts of each event are superimposed according to their time distribution to form an event-driven risk premium time structure.

[0076] Assessment of the transmission of uncertainty in clean energy: Based on the statistical value of renewable energy forecast deviation, a transmission model is established to the transmission of uncertainty in clean energy output to grid carbon intensity. When the output of clean energy is lower than expected, the grid needs to call up high-carbon emission fossil energy units to supplement it, which leads to an increase in grid carbon intensity. The strength and probability of this transmission effect are quantified and transformed into the risk of rising carbon costs.

[0077] Based on the combined results of the three assessments, the risk assessment model outputs a risk premium coefficient for each future period. This coefficient, measured in monetary units, represents the risk compensation that needs to be added on top of the expected carbon cost, covering the possibility that the actual carbon cost will be higher than expected.

[0078] S233. The calculated risk premium coefficients for each time period are organized according to time indices that are completely aligned with the basic expected carbon cost sequence, forming a risk premium coefficient time series that can be directly used for algebraic superposition, and output to the subsequent S24 step for curve synthesis. Specifically, ensure that the risk premium coefficients and the basic expected carbon cost sequence are completely consistent in time resolution, prediction start point, prediction end point, and time stamping method; perform calendar alignment conversion for potential time misalignments caused by the calculation characteristics of the risk assessment model; output the risk premium coefficient series with the same data structure as the basic series, supporting element-by-element addition operations.

[0079] S24. Synthesis of dynamic expected carbon cost curves, specifically including: S241. The expected carbon cost values ​​for each time period in the basic expected carbon cost sequence are algebraically added to the corresponding risk premium coefficient. This addition operation follows the vector addition rule, summing the values ​​at each time index position: the risk-adjusted carbon cost equals the basic expected carbon cost plus the risk premium coefficient. This summation operation considers the directionality of carbon costs: for charging costs—positive values, the cost increases further after adding the risk premium, reflecting the upside risk faced by charging behavior; for discharging benefits—negative values, the absolute value of the benefit may decrease or turn positive after adding the risk premium, reflecting the uncertainty risk of realizing discharging benefits.

[0080] S242. Organize the risk-adjusted carbon cost values ​​for each time period into structured curve data to form the final dynamic expected carbon cost curve. This curve includes the following attributes: time axis coverage, time resolution, risk-adjusted expected carbon cost values ​​for each time period, corresponding basic expected carbon cost values, risk premium coefficient values, and data quality and confidence level indicators. This curve is dynamic and continuously evolves with the updates of multi-source forecast data in step S21 and the adjustments of risk assessment parameters in step S23, supporting periodic recalculation under a rolling optimization framework.

[0081] S243. The dynamically projected carbon cost curve is output to the subsequent S3 step as the core input to the intertemporal decision function. Specifically, this curve provides the carbon cost coefficient of the objective function for the optimization problem constructed in the S3 step, influencing the spatiotemporal arrangement of charging and discharging plans; the risk premium structure of the curve provides a basis for setting constraints or robust optimization parameters in the S3 step; and the long-term trend characteristics of the curve provide data support for the full life-cycle optimization perspective of the S3 step. Through this output transmission, a closed-loop value chain is achieved from carbon market forecasting to operational strategy optimization.

[0082] S3. Using the dynamic expected carbon cost curve as the core input, and aiming to maximize the net value of carbon assets over the entire lifecycle, a cross-period charge-discharge strategy is generated; specifically including: S31. Construction of intertemporal decision functions, specifically: S311. The core optimization objective is defined as maximizing the cumulative net carbon asset value increment after discounting. This increment represents the total increase in net carbon asset value achieved through optimized charge-discharge operation strategies over the entire period from the current moment to the end of the battery's expected lifespan. A discount factor is introduced to handle intertemporal value comparisons, discounting the carbon costs and carbon benefits of future periods to their equivalent value at the current moment, reflecting time preference and capital cost.

[0083] The calculation of the increase in carbon asset net value is based on the current carbon asset net value output in step S133. By comparing the expected final value of carbon assets under the optimized operating strategy with the expected final value of carbon assets under the benchmark operating strategy, the increase in net value brought about by strategy optimization is determined. This increase comprehensively reflects the contributions of carbon cost savings, increased carbon emission reduction benefits, and asset preservation effects during the operating phase.

[0084] S312. Set the decision variable of the inter-period decision function to the battery charging and discharging power for each future period. This power variable can be a continuous or discrete variable, determined according to the scheduling accuracy requirements of the tiered utilization scenario; positive values ​​represent discharge power, negative values ​​represent charging power, and zero values ​​represent idle state. The time dimension of the decision variable covers the same time range as the dynamic expected carbon cost curve output in step S24, and the time resolution matches the curve's time resolution to ensure the correspondence between the time-period optimization decision and the carbon cost prediction.

[0085] S313. Use the risk-adjusted expected carbon cost values ​​for each period in the dynamic expected carbon cost curve output from step S24 as the core economic parameter input for the intertemporal decision function. This parameter directly determines the carbon economic evaluation of charging and discharging behavior in each period: discharging behavior in high carbon cost periods results in high carbon emission costs or low emission reduction benefits, with poor economic performance; discharging behavior in low carbon cost or even negative carbon cost periods results in low carbon emission costs or high emission reduction benefits, with better economic performance.

[0086] A mapping rule for calculating charging / discharging power and carbon cost is established: the planned charging / discharging power for each time period is multiplied by the expected carbon cost per unit power for that time period to obtain the planned carbon cost or carbon benefit for that time period; the results for each time period are accumulated along the time axis to obtain the total planned carbon cost. This total carbon cost serves as a core component of the objective function of the intertemporal decision function, and its minimization or the minimization of its discounted cumulative value is pursued.

[0087] S314. Based on the above factors, construct the mathematical expression of the intertemporal decision function. The objective function is to minimize the total planned carbon cost, or equivalently, to maximize the cumulative net value increment of carbon assets after discounting. This function uses the charging and discharging power in each future period as the decision variable, the dynamic expected carbon cost curve as the parameter input, and the discount rate as the time preference parameter, forming a standard mathematical optimization problem structure, laying the foundation for subsequent constraint loading and model solving.

[0088] S32. The constraint loading is coupled with the health state decay model, specifically as follows: S321. Based on the physical parameters and application scenario requirements of batteries for secondary use, formally define a set of mathematical inequalities constituting the constraints. This set of conditions must include at least: Charge / discharge power upper and lower limits: Determined by the battery's rated power and the current available capacity determined in step S123, these limits set an absolute upper limit for the charge / discharge power at each time period to prevent equipment damage or safety risks caused by over-power operation. This upper limit is dynamically adjusted as the battery's health deteriorates, reflecting the degradation of power capability during the secondary utilization phase.

[0089] State of charge (SOC) upper and lower limits: Defined by the battery's safe operating range, these limits set upper and lower thresholds for the SOC at various times to prevent irreversible electrochemical damage caused by overcharging or over-discharging. This safe range is determined based on the battery's chemical system characteristics, temperature conditions, and health status, and may tighten appropriately as the battery ages.

[0090] Net discharge requirement constraint within a cycle: Determined by the load demand of the secondary utilization scenario, a lower limit or upper / lower limit for the net discharge within a specific time period is set to ensure that the battery fulfills its energy service contract function and meets the power supply reliability requirements of the external load. This requirement constraint may vary depending on the scenario type: frequency regulation service scenarios emphasize power response capability, energy time-shift scenarios emphasize energy throughput, and backup power scenarios emphasize the maintenance of available capacity.

[0091] The above constraints are embedded into the optimization framework in the form of mathematical inequalities or equations to form the boundary of the feasible solution space of the decision variables.

[0092] S322. Embed the battery health state degradation model into the optimization framework and establish its coupling relationship with the intertemporal decision function. Based on the health state assessment results of step S115, this degradation model uses mechanism-driven or data-driven aging modeling methods to predict the degradation trajectory of battery health state with charge-discharge cycles.

[0093] The model input is the assumed future charge and discharge power sequence in the intertemporal decision function, including the power amplitude, state of charge change depth, and duration for each time period. The model output is the predicted battery health status at each future time point and the corresponding degradation of usable capacity. This prediction considers the combined effect of cycle aging and calendar aging, as well as the influence of accelerating aging factors such as temperature and charge / discharge rate.

[0094] A health status-asset value mapping function is established to map the predicted decline in health status to the expected reduction in the future value of the underlying assets in the dynamic carbon asset net value model. This mapping is based on the remaining carbon emission reduction potential assessment logic of step S115: a decline in health status leads to a reduction in available capacity, which in turn reduces the future achievable energy throughput and the corresponding carbon emission reduction potential, resulting in an expected reduction in the value of the underlying assets.

[0095] The expected reduction is quantified as a health status decline penalty term proportional to the decline amount, which serves as a negative benefit cost term in the optimization objective. The coefficient of this penalty term reflects the sensitivity of carbon asset net worth to health status and can be determined based on the carbon asset net worth structure analysis in step S133.

[0096] S323. Using the intertemporal decision function as the core objective, the constraints defined in step S321 as hard boundaries, and the health state decay penalty term determined in step S322 as a negative benefit cost term, a complete coupled optimization model is constructed.

[0097] This model integrates three objectives: economic efficiency is achieved by minimizing the total planned carbon cost; physical operational constraints are guaranteed by a set of constraints; and the long-term asset preservation objective is reflected by a health status decay penalty term. The model output is a comprehensive optimal solution that satisfies all constraints and balances short-term carbon costs and long-term asset value, used for the algorithm solution in step S33.

[0098] S33. Solving the coupled optimization model and generating the long-term optimal charge-discharge plan, specifically: S331. Select an appropriate solution algorithm based on the problem size, time urgency, and degree of predictability: Dynamic programming algorithms are suitable for scenarios with moderate state space dimensionality and high predictability. They discretize continuous time into stages and continuous state variables into a grid, and find the globally optimal strategy through backward recursion or forward search. However, the computational complexity increases exponentially with the state dimension, and the curse of dimensionality needs to be mitigated through state aggregation or approximate dynamic programming.

[0099] Model predictive control algorithm: suitable for scenarios with significant uncertainty and frequent response to new information; at each decision moment, based on the current state and the latest prediction data, it solves a finite time domain open-loop optimization problem, only executes the decision for the first time period, and then updates the prediction and state repeatedly; it suppresses the accumulation of prediction error through a feedback mechanism and adapts to the dynamically changing environment.

[0100] Configure algorithm parameters, including time discretization step size, state space grid accuracy, prediction time domain length, control time domain length, and solution termination condition, to balance solution accuracy and computational efficiency.

[0101] S332. Execute the selected algorithm to solve the coupled optimization model constructed in step S323. During the solution process, the carbon cost parameters are obtained in real time by calling the dynamic expected carbon cost curve from step S24, the battery health state degradation model from step S322 is called to evaluate the impact of the decision on the health state, and the constraint satisfaction of step S321 is verified.

[0102] The solution process iteratively explores the feasible solution space, searching for the charging and discharging power time series that optimizes the objective function. The comprehensive evaluation of the objective function includes: the cumulative sum of the charging and discharging power and the corresponding carbon cost for each time period, the cumulative sum of the health state decay and the penalty coefficient for each time period, and the intertemporal adjustment of the discount factor.

[0103] S333. The solution output is a long-term optimal charge and discharge plan covering multiple future scheduling cycles. This plan specifies the optimal charge and discharge power setpoints and the corresponding expected state of charge trajectory for each time period in the form of a time series table or continuous curve.

[0104] The core features of this plan are: proactively guiding batteries to reduce discharge or switch to charging during high-cost periods identified in the dynamic expected carbon cost curve, thereby avoiding the risks of high carbon emission costs or low emission reduction benefits; increasing discharge or deep charging and discharging during low-cost periods to capture low-carbon cost opportunities or maximize emission reduction benefits; and at the same time, through the constraint of the health status degradation penalty term, avoiding the adoption of aggressive strategies that accelerate aging and damage long-term asset value in pursuit of short-term carbon cost optimization.

[0105] The plan maximizes the net value of carbon assets over the entire lifecycle while meeting the real-time operational safety constraints defined in step S321, reflecting a strategic upgrade from immediate cost optimization to long-term value management.

[0106] S334. The long-term optimal charge / discharge plan is output to step S4 as a baseline plan for rolling optimization execution. Specifically, this plan provides step S4 with execution instructions for the current period and reference trajectories for subsequent periods; it provides a target benchmark for real-time monitoring and deviation correction in step S4; and it provides a basis for comparing planned expectations with actual performance for periodic calculations and model corrections in step S4. Through this output transmission, a closed-loop connection is achieved from strategy optimization to execution control.

[0107] S4. Implement the intertemporal charge-discharge strategy and continuously update the dynamic carbon asset net value model and the dynamic expected carbon cost curve based on actual operating data; specifically including: S41. Plan execution and real-time monitoring, specifically: S411. Extract and issue the charging / discharging power command corresponding to the current scheduling period from the long-term optimal charging / discharging plan generated in step S33. After receiving the command, the battery management system actuator performs power parsing and safety verification to confirm that the command value is within the upper and lower limits of the charging / discharging power defined in step S321; drives the power electronic devices to perform charging / discharging loop switching and power adjustment so that the actual battery operating power tracks the command set value; synchronously starts the closed-loop control algorithm, fine-tunes the duty cycle or modulation index based on real-time voltage and current feedback, suppresses disturbance deviations, and ensures power control accuracy.

[0108] S412. Synchronously monitors battery operating status data in real time, establishing a status perception system covering multiple dimensions such as electrical, thermal, and aging. Specific monitoring content includes: Actual state of charge: The percentage of the battery’s current remaining usable capacity relative to the total usable capacity is calculated in real time by integrating with a high-precision current sensor and calibrating with open-circuit voltage, or by using a state estimation algorithm based on an electrochemical model. This actual state of charge is compared with the expected state of charge trajectory in step S33 to evaluate the deviation from the plan.

[0109] Health status: The ratio of the battery's current maximum usable capacity to its rated initial capacity is evaluated in real time through online impedance spectroscopy monitoring, capacity increment analysis, or data-driven aging index extraction; this health status data is used for prediction deviation calculation in step S422 and model calibration in step S433.

[0110] Temperature: By using an array of temperature sensors distributed on the surface of individual battery cells and key nodes, the operating temperature distribution of the battery is monitored; abnormal temperature hotspots are identified, thermal management interventions are triggered, and temperature data is used as an important input parameter for the health status degradation model.

[0111] Actual throughput: The cumulative charging and discharging ampere-hour throughput and watt-hour energy throughput within the calculation period, distinguishing between the charging absorbed energy and the discharging working condition energy, and providing basic energy flow data for carbon emission reduction calculation in step S431.

[0112] The aforementioned status data is sampled at a fixed frequency, preprocessed, and then written into a real-time database, supporting status updates at the millisecond to second level.

[0113] S413. Synchronously monitor actual carbon intensity data from the power grid in real time. This data is obtained through the power grid dispatch data interface, the electricity market information disclosure platform, or third-party carbon emission monitoring services, and characterizes the marginal emission intensity of the current actual power generation mix of the power grid.

[0114] The actual carbon intensity data is compared with the predicted carbon intensity data of the power grid obtained in step S21 to evaluate the accuracy of the prediction; the actual carbon intensity is combined with the charging and discharging behavior executed in step S41 to calculate the actual carbon cost or carbon benefit of the current operation in real time, providing a basis for deviation judgment for the rolling optimization triggering in step S42.

[0115] S42. Rolling optimization triggering and dynamic plan correction, specifically: S421. A fixed period is preset as the rolling trigger condition, with a selectable range of 15 minutes to 24 hours. At each trigger moment, it is determined whether the rolling optimization start condition is met. When the trigger condition is met, a new round of repeated execution of steps S2 and S3 is started, the current moment is marked as the new optimization starting point, historical data of the executed time period is frozen, and the subsequent time period that has not yet been executed is taken as the new optimization time domain.

[0116] S422. Obtain three types of the latest data as new optimization inputs: the latest monitored battery operating status data, the latest acquired actual carbon intensity data, and the updated long-term forecast data.

[0117] S423. Using the updated long-term forecast data as input, repeat steps S22 to S24 to regenerate the dynamic expected carbon cost curve. This regeneration process considers the correction of carbon cost expectations by the latest market information, as well as the parameter updates of the risk assessment model in step S23 based on the latest forecast deviation statistics, and outputs an updated carbon cost curve that reflects current market perception and future risk assessment.

[0118] S424. Using the updated battery operating state data as initial conditions and the newly generated dynamic expected carbon cost curve as the core parameter, repeat steps S31 to S33 to resolve the intertemporal decision function. This resolution uses the same algorithm framework configured in step S331, but generates a new optimal strategy based on the updated input data.

[0119] The newly obtained long-term optimal charge / discharge plan is compared with the unexecuted portions of the original plan to identify strategy adjustments, such as power amplitude changes, time period shifts, and charge / discharge mode switching, forming a continuously updated subsequent execution plan. This dynamic correction mechanism suppresses the cumulative effects of model prediction errors and external disturbances, enabling the operating strategy to continuously adapt to changing environments and the latest information.

[0120] S43. Accounting cycle closed-loop correction, specifically: S431. Based on the battery operating status data monitored and recorded in step S412 and the actual carbon intensity data monitored in step S413 within the accounting cycle, the same carbon emission reduction accounting methodology model as in step S11 is used to calculate the actual carbon emission reduction generated by the battery due to optimized control within the cycle. Specifically: Based on the actual charge and discharge power and actual carbon intensity monitored for each period, the upstream carbon emissions during the charging process and the alternative baseline emissions during the discharging process are calculated within the cycle. These are then compared with the baseline emissions of alternative technologies that would meet the same energy service needs without the battery system to determine the net carbon reduction. This calculation follows the same methodology framework for voluntary greenhouse gas emission reduction projects as step S114, ensuring consistency in accounting boundaries, baseline settings, and leakage considerations.

[0121] The calculated actual carbon emission reductions are multiplied by the average actual transaction price in the carbon trading market during the accounting period to calculate the actual carbon market value corresponding to the actual carbon emission reductions in that period. This value represents the carbon asset appreciation achieved by the optimized operating strategy in this period, providing an asset-side verification signal for model calibration in step S433.

[0122] S432. Obtain the actual measured value of the battery health status at the end of the accounting cycle. This measurement value is obtained through dedicated capacity testing, electrochemical impedance spectroscopy, or offline diagnostic procedures, with higher accuracy than the online estimate in step S412. Compare this actual measured value with the predicted value of the battery health status at the end of the cycle by the battery health status degradation model in step S322, and calculate the health status prediction deviation. This deviation is defined as the actual measured value minus the predicted value; a positive value indicates that the actual aging degree is lower than expected, and a negative value indicates that the actual aging degree is higher than expected. Analyze the temporal distribution of the deviation and its influencing factors to identify systematic shifts in model parameters or unexpected changes in external stress.

[0123] S433. The actual carbon market value calculated in step S431 is used as an asset-side calibration signal and input into the dynamic carbon asset net value model in S1. If the actual carbon market value is consistent with and close to the carbon asset net value predicted in step S133, the validity of the basic asset valuation parameters is verified; if there is a significant deviation, the source of the deviation is analyzed, and the achievable proportion parameter in the remaining carbon emission reduction potential assessment in step S115 or the current achievable proportion in the asset revaluation method in step S132 is positively corrected to make the model prediction closer to the actual performance.

[0124] The health state prediction deviation calculated in step S432 is used as a model performance feedback signal and input into the battery health state degradation model in step S322. Based on the sign and magnitude of the deviation, parameter estimation algorithms such as recursive least squares, Kalman filtering, or Bayesian update are used to reverse-correct key parameters in the degradation model, including the cycle aging rate coefficient, temperature acceleration factor, depth of discharge sensitivity index, and calendar aging baseline rate. Through parameter correction, the accuracy of the degradation model in predicting subsequent time-domain health state trajectories is improved, the evaluation accuracy of the health state degradation penalty term in step S322 is enhanced, and the robustness of the long-term operation strategy is optimized.

[0125] Through the above calibration and correction, a closed-loop feedback loop is completed from strategy execution and performance monitoring to model updates, enabling the continuous evolution of the dynamic carbon asset net value model and the battery health state decay model, and supporting the optimization decision-making for the next accounting cycle.

[0126] Example 2

[0127] like Figure 2 As shown, a battery lifecycle carbon-saving intelligent management and control system integrating peak and valley carbon pricing is used to implement a battery lifecycle carbon-saving intelligent management and control method integrating peak and valley carbon pricing, including: The carbon asset net worth management module is used to perform carbon asset net worth modeling and initialization; specifically, it includes: The basic asset assessment unit is used to acquire and analyze the battery's carbon footprint at the factory, historical charge and discharge data and corresponding carbon emission data during the main use phase, and to assess and quantify the battery's remaining carbon emission reduction potential after the main use phase ends by using a pre-set carbon emission reduction accounting methodology model, and to confirm the quantified value as a basic asset. The debt cost modeling unit is used to obtain the carbon cost of the battery production stage. Based on the expected remaining life and available capacity decay curve of the secondary utilization scenario, the carbon cost of the production stage is allocated to the future operating cycle using a preset amortization rule to form historical carbon cost amortization liability. At the same time, based on the typical operating load and maintenance plan of the secondary utilization scenario, the future operation and maintenance carbon cost is predicted, which together with the historical carbon cost amortization liability constitutes the model liability. The net asset value dynamic calculation unit is used to access the real-time carbon price data stream of the carbon trading market in real time. It uses the verified real-time carbon price as a value coefficient to revalue the current value of the underlying assets and the liabilities of the model respectively, and outputs the updated carbon asset net value in real time by calculating the difference between the revalued asset and liability values.

[0128] The carbon cost prediction and risk assessment module, connected to the carbon asset net worth management module, is used to generate a dynamic expected carbon cost curve; specifically, it includes: The multi-source data fusion unit is used to acquire in real time the future multi-period grid carbon intensity forecast data from the grid dispatch system, the clean energy power generation plan data from the power generation side, and the long-term carbon price trend forecast data from the carbon trading market, and to align and fuse the above multi-source data on a time scale to form unified long-term carbon market forecast data. The basic cost calculation unit is used to calculate the expected carbon emissions and expected carbon costs corresponding to a unit charge and discharge behavior in various future periods based on long-term carbon market forecast data and combined with the charge and discharge efficiency model of the cascaded battery, and generate a basic expected carbon cost sequence. The risk assessment and premium loading unit is used to collect historical carbon price volatility data, renewable energy forecast deviation statistics, and policy and event calendar data in real time, and dynamically calculates the risk premium coefficient for each future period through a risk assessment model based on the uncertainty of long-term carbon market forecast data. The curve synthesis output unit is used to add the expected carbon cost value of each period in the basic expected carbon cost sequence to the risk premium coefficient of the corresponding period, synthesize the final dynamic expected carbon cost curve including risk adjustment, and output it.

[0129] The optimized decision-making and planning generation module is connected to the carbon asset net worth management module and the carbon cost prediction and risk assessment module, respectively, to generate the long-term optimal charge-discharge plan; specifically including: The objective function construction unit is used to construct an intertemporal decision function with the core optimization objective of maximizing the discounted cumulative net value increase of carbon assets over the entire period from the current moment to the end of the battery's expected lifespan, and to input the dynamic expected carbon cost curve as the core economic parameter. The constraint and decay coupling unit is used to load operational constraints, including upper and lower limits of battery power, safe range of state of charge, and scenario load requirements, onto the intertemporal decision function. At the same time, a pre-established battery health state decay model is coupled to the function. The decay model predicts the health state decay based on the planned charge and discharge power and throughput, and converts this decay into a penalty term for potential reduction in future carbon asset net value, which is then integrated into the objective function. The optimization solution unit is used to perform rolling or global solutions on the intertemporal decision function that couples the effects of health state decay with operational constraints using dynamic programming or model predictive control algorithms, and generate a long-term optimal charging and discharging plan covering multiple future scheduling cycles.

[0130] The execution and feedback control module, connected to the optimization decision-making and plan generation module, is used for plan execution, rolling optimization, and closed-loop feedback; specifically, it includes: The instruction execution and monitoring unit is used to send the charge and discharge power instruction corresponding to the current scheduling period in the long-term optimal charge and discharge plan to the battery BMS execution mechanism, and simultaneously monitor the battery operating status data and the actual carbon intensity data from the power grid in real time. The rolling triggering and updating unit is used to trigger the carbon cost prediction and risk assessment module and the optimization decision and plan generation module again by using the latest monitoring data and the newly acquired external market forecast data when triggered, so as to update and correct the subsequent charging and discharging plans that have not yet been executed. The periodic accounting and calibration unit is used to calculate the actual carbon emission reductions generated by the battery and its market value based on the actual data monitored during the period after the end of an accounting cycle synchronized with the carbon market accounting cycle. It also compares the actual decay value of the battery's health status with the predicted value. The deviation between the actual carbon market value and the predicted health status is used as a feedback signal and sent to the basic asset verification unit, the liability cost modeling unit, and the constraint and decay coupling unit of the optimization decision and plan generation module of the carbon asset net value management module, respectively, for calibrating the relevant model parameters.

[0131] As can be seen from the above description, the embodiments of the present invention achieve the following technical effects: This invention represents a paradigm shift in battery management, moving from energy-consuming equipment control to carbon asset investment and operation, and constructing a dynamic lifecycle carbon value management framework with financial attributes. The invention defines a battery as a carbon asset with clearly defined underlying assets and liabilities. Through real-time carbon price data streams, the system dynamically re-evaluates these assets and liabilities, continuously outputting the net carbon asset value. This transforms the battery's carbon reduction capacity into a measurable, estimable, and tradable digital asset; provides a credible value assessment basis for battery financing, pledging, and trading in the carbon market; and elevates the BMS's control objective from reducing instantaneous electricity costs or losses to maximizing the net carbon asset value over its entire remaining lifecycle, achieving a breakthrough in management paradigms.

[0132] This invention overcomes the challenges of quantifying the uncertainties of the long-term carbon market and the risks of battery physical degradation, endowing the system with forward-looking risk avoidance and intertemporal optimization capabilities. By introducing a risk premium coefficient, the uncertainty of carbon price fluctuations and policy events is quantified as a cost, generating a forward-looking cost curve that includes risk compensation. Simultaneously, by establishing a mapping relationship between health status degradation and carbon asset value reduction, physical degradation is transformed into an economic penalty term and incorporated into the optimization objective. The system can proactively identify and avoid future high-carbon-cost, high-risk operating periods while capturing low-carbon-cost opportunities. The control strategy achieves an optimal balance between pursuing current carbon gains and delaying the depreciation of battery carbon assets, realizing long-term asset preservation and appreciation. It significantly improves the economic predictability and reliability of investing in and operating battery energy storage projects in a complex and uncertain carbon market environment.

[0133] This invention establishes a data-driven closed loop of perception, decision-making, execution, and learning, ensuring the real-time adaptability, continuous optimization, and verifiable credibility of the carbon asset management strategy. Through fixed-period rolling triggering of re-prediction and re-optimization, the strategy can dynamically adjust to the latest information. A dual-feedback calibration mechanism based on the deviation between actual carbon market value and battery health status accurately and in a closed loop feeds the execution results back to the carbon asset net value model and the health status decay model. The system possesses self-learning and evolution capabilities, with model prediction accuracy and control effectiveness continuously improving over time. It ensures that the core indicator of carbon asset net value always closely reflects the real market value and actual battery status, eliminating model drift risk. The entire optimization decision-making process is based on real-time monitoring data, and all key carbon data streams are traceable and verifiable. The generated carbon emission reductions meet the verification requirements of the carbon trading market, greatly enhancing the system's credibility and industrialization capabilities.

[0134] The embodiments and / or implementation methods described above are merely preferred embodiments and / or implementation methods for implementing the technology of the present invention, and are not intended to limit the implementation methods of the technology of the present invention in any way. Any person skilled in the art can make some modifications or alterations to other equivalent embodiments without departing from the scope of the technical means disclosed in the present invention, but these should still be regarded as the technology or embodiments that are substantially the same as the present invention.

[0135] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. The above descriptions are only preferred embodiments of this application. It should be noted that due to the limitations of written expression, while there are objectively infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner. These improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of this application.

Claims

1. A method for intelligent carbon-saving management and control of the entire battery lifecycle that integrates peak and valley carbon pricing, characterized in that, include: Establish a dynamic carbon asset net value model for batteries in the secondary utilization stage and calculate their carbon asset net value in real time; Based on the aforementioned dynamic carbon asset net value model and long-term market forecast data, a dynamic expected carbon cost curve covering the remaining battery life is generated. Using the dynamic expected carbon cost curve as the core input and aiming to maximize the net value of carbon assets over the entire cycle, a cross-period charging and discharging strategy is generated. The intertemporal charge-discharge strategy is executed, and the dynamic carbon asset net value model and dynamic expected carbon cost curve are updated on a rolling basis based on actual operating data.

2. The intelligent carbon-saving management method for the entire life cycle of a battery that integrates peak and valley carbon prices, as described in claim 1, is characterized in that... The aforementioned method establishes a dynamic carbon asset net value model for batteries in the secondary utilization stage, and calculates their carbon asset net value in real time, specifically including: Acquire and analyze the battery's carbon footprint at the time of manufacture, historical charge and discharge data during the main use phase, and corresponding carbon emission data. Use a preset carbon emission reduction accounting methodology model to assess and quantify the battery's remaining carbon emission reduction potential after the main use phase ends, and confirm the quantified value as the underlying asset of the dynamic carbon asset net value model. The carbon cost of the battery production stage is obtained. Based on the expected remaining lifespan and available capacity decay curve of the secondary utilization scenario, the carbon cost of the production stage is allocated to the future operating cycle using a preset amortization rule to form historical carbon cost amortization liability. At the same time, based on the typical operating load and maintenance plan of the secondary utilization scenario, the future operation and maintenance carbon cost is predicted, which together with the historical carbon cost amortization liability constitutes the model liability. It accesses the real-time carbon price data stream of the carbon trading market, calculates the difference between the revalued underlying asset value and liability value, and outputs the updated net carbon asset value of the battery in real time.

3. The intelligent carbon-saving management method for the entire life cycle of a battery that integrates peak and valley carbon prices, as described in claim 2, is characterized in that... The real-time carbon price data stream accessed in the carbon trading market calculates the difference between the revalued underlying asset value and the liability value, and outputs the updated net carbon asset value of the battery in real time, specifically including: Establish an interface with the carbon trading market data platform to obtain real-time carbon price data streams; perform timestamp alignment, outlier filtering, and validity verification on the obtained data to generate standardized, verified real-time carbon prices; The verified real-time carbon price is used as the core value coefficient and input into the asset revaluation method and the liability revaluation method respectively. The asset revaluation method multiplies the verified real-time carbon price by the current achievable proportion of the remaining carbon emission reduction potential to calculate the revalued value of the underlying assets. The liability revaluation method multiplies the verified real-time carbon price by the current allocation of the model liability to calculate the revalued value of the liabilities. The revalued value of the underlying assets is subtracted from the revalued value of the liabilities to calculate the current net carbon asset value in real time; the result of the net carbon asset value is output to the dynamic database and serves as the current state value of the dynamic net carbon asset value model.

4. The intelligent carbon-saving management method for the entire life cycle of a battery that integrates peak and valley carbon prices, as described in claim 3, is characterized in that... Based on the aforementioned dynamic carbon asset net worth model and long-term market forecast data, a dynamic expected carbon cost curve covering the remaining battery life is generated, specifically including: Real-time acquisition of multi-period grid carbon intensity forecast data from the grid dispatch system, clean energy power generation plan data from the power generation side, and long-term carbon price trend forecast data from the carbon trading market; alignment and fusion of the above multi-source data on a time scale to form a unified time series long-term carbon market forecast data covering several future dispatch cycles to the end of the expected battery life. Based on the long-term carbon market forecast data and combined with the charge-discharge efficiency model of the cascaded battery, the expected carbon emissions and expected carbon costs corresponding to a unit charge-discharge behavior in each future period are calculated to generate an initial basic expected carbon cost sequence. By accessing historical volatility data of the carbon market and a calendar of major policy events, and based on the uncertainty of the long-term carbon market forecast data, the risk premium coefficient for each future period is dynamically calculated through a risk assessment model. The expected carbon cost value for each period in the basic expected carbon cost sequence is added to the risk premium coefficient for the corresponding period to synthesize the final dynamic expected carbon cost curve that includes risk adjustment.

5. The intelligent carbon-saving management method for the entire life cycle of a battery that integrates peak and valley carbon prices, as described in claim 4, is characterized in that... The method of dynamically calculating the risk premium coefficient for each future period using a risk assessment model specifically includes: Real-time collection of multi-dimensional risk data for quantifying uncertainty, including historical volatility data of carbon prices in the carbon trading market, statistical values ​​of renewable energy forecast deviations reflecting the uncertainty of clean energy output, and policy and event calendar data that identify major events that may affect the supply and demand of the carbon market; The long-term carbon market forecast data and multi-dimensional risk data are input into a preset risk assessment model. The risk assessment model comprehensively evaluates the deviation of the forecast data from historical fluctuation patterns, the potential impact level of planned events during the forecast period, and the transmission effect of clean energy uncertainty on the grid carbon intensity, and dynamically calculates the risk premium coefficient corresponding to each future period. The calculated risk premium coefficients for each time period are organized according to time indices that are perfectly aligned with the basic expected carbon cost sequence, forming a risk premium coefficient time series that can be directly used for algebraic superposition.

6. The intelligent carbon-saving management method for the entire life cycle of a battery, integrating peak and valley carbon prices, as described in claim 5, is characterized in that... Using the aforementioned dynamic expected carbon cost curve as the core input, and aiming to maximize the net value of carbon assets over the entire lifecycle, a cross-period charge-discharge strategy is generated, specifically including: With the core optimization objective of maximizing the incremental increase in the discounted cumulative net carbon asset value over the entire period from the current moment to the end of the battery's expected lifespan, an intertemporal decision function is constructed. The intertemporal decision function sets the decision variable for each future time period as the battery's charging and discharging power, and takes the carbon cost value of each time period in the dynamic expected carbon cost curve as the core economic parameter input. The total planned carbon cost is calculated by multiplying the planned charging and discharging behavior with the carbon cost value of the corresponding time period and summing them up. Load the running constraints for the intertemporal decision function, and couple the pre-established battery health state degradation model with the intertemporal decision function. The degradation model predicts the degradation of battery health state and available capacity at each future time point, and integrates the degradation amount into the objective function. Dynamic programming or model predictive control algorithms are used to solve the intertemporal decision function coupled with the influence of health state decay in a rolling or global manner, provided that all constraints are met. The solution is a long-term optimal charging and discharging plan that covers multiple future scheduling cycles.

7. The intelligent carbon-saving management method for the entire life cycle of a battery that integrates peak and valley carbon prices, as described in claim 6, is characterized in that... The intertemporal decision function is loaded with runtime constraints, and a pre-established battery health state degradation model is coupled with the intertemporal decision function. The degradation model predicts the degradation of battery health state and available capacity at future time points, and the degradation amount is integrated into the objective function, specifically including: Based on the physical parameters and application scenario requirements of batteries for secondary use, a set of mathematical inequalities constituting the constraints are formally defined. The battery health status degradation model is embedded into the optimization framework. Based on the assumed future charge and discharge power sequence in the intertemporal decision function, the battery health status degradation model predicts the degradation trajectory of the battery health status with charge and discharge cycles. A health status-asset value mapping function is established, which maps the predicted health status degradation amount to the expected reduction in the future value of the underlying asset in the dynamic carbon asset net value model, and quantifies this expected reduction amount into a health status degradation penalty term that is proportional to the degradation amount. By taking the intertemporal decision function as the core objective, the constraints as hard boundaries, and adding the health state decay penalty term as a negative benefit cost term to the core objective function, a coupled optimization model is constructed.

8. The intelligent carbon-saving management method for the entire life cycle of a battery, integrating peak and valley carbon prices, as described in claim 7, is characterized in that... The execution of the intertemporal charge-discharge strategy, and the rolling updates of the dynamic carbon asset net value model and the dynamic expected carbon cost curve based on actual operating data, specifically include: The charging and discharging power command corresponding to the current scheduling period in the long-term optimal charging and discharging plan is sent to the battery BMS actuator for control; the battery operating status data and the actual carbon intensity data from the power grid are monitored in real time simultaneously. Using a preset fixed period as the rolling trigger condition, when the trigger condition is met, the latest monitored battery operating status data, the latest acquired actual carbon intensity data, and the updated long-term forecast data re-acquired from the external market are used as new inputs. By regenerating the dynamic expected carbon cost curve and solving the new inter-period decision function, the subsequent long-term optimal charge and discharge plan that has not yet been executed is rolled up and dynamically corrected. After an accounting cycle synchronized with the carbon market accounting cycle ends, the corresponding actual carbon market value and health status prediction deviation are calculated and input into the dynamic carbon asset net value model to calibrate the model.

9. The intelligent carbon-saving management method for the entire life cycle of a battery that integrates peak and valley carbon prices, as described in claim 8, is characterized in that... After the completion of an accounting cycle synchronized with the carbon market accounting cycle, the corresponding actual carbon market value and health status prediction deviation are calculated and input into the dynamic carbon asset net value model for model calibration. This process specifically includes: Based on the battery operating status data and actual carbon intensity data recorded during the accounting period, a carbon emission reduction accounting methodology model is used to calculate the actual carbon emission reduction generated by the battery due to optimized control during the period; the actual carbon emission reduction is multiplied by the average carbon price in the carbon trading market during the accounting period to calculate its corresponding actual carbon market value. Obtain the actual measured value of the battery health status at the end of the accounting cycle, compare the actual measured value with the predicted value of the health status at the end of the cycle by the battery health status decay model at the beginning of the cycle, and calculate the health status prediction deviation. The actual carbon market value is used as an asset-side calibration signal and input into the dynamic carbon asset net value model to positively correct the valuation parameters of the underlying assets of the model or to verify its effectiveness; at the same time, the health status prediction deviation is used as a model performance feedback signal and input into the battery health status decay model to reversely correct the internal parameters of the decay model.

10. A battery lifecycle carbon-saving intelligent management and control system integrating peak and valley carbon pricing, characterized in that, A method for intelligent carbon-saving management and control of a battery throughout its entire lifecycle, incorporating peak-valley carbon pricing as described in any one of claims 1-9, includes: The carbon asset net value management module is used to build and update a dynamic carbon asset net value model in real time for batteries in the cascade utilization stage. The model takes the battery's remaining carbon emission reduction potential as the basic asset, the allocated historical carbon cost and operation and maintenance carbon cost as liabilities, and calculates and outputs the battery's carbon asset net value based on real-time carbon price fluctuations. The carbon cost prediction and risk assessment module is connected to the carbon asset net value management module. It is used to generate a dynamic expected carbon cost curve covering the expected remaining life of the battery based on the carbon asset net value, long-term market forecast data and risk data. The curve includes the expected carbon cost for each period after risk adjustment. The optimization decision and plan generation module is connected to the carbon asset net value management module and the carbon cost prediction and risk assessment module, respectively. It is used to maximize the carbon asset net value over the whole life cycle, with the dynamic expected carbon cost curve as the core input, and integrates the battery health state decay model to perform inter-period optimization solution to generate the long-term optimal charge and discharge plan. The execution and feedback control module, connected to the optimization decision and plan generation module, is used to execute the current instructions in the charge and discharge plan and monitor battery operation and grid carbon intensity data in real time. It also triggers at fixed intervals to drive the carbon cost prediction and risk assessment module and the optimization decision and plan generation module to perform rolling optimization using the latest data. At the same time, after a calculation cycle, it calculates the carbon emission reduction value and health status deviation based on actual data and feeds the results back to the carbon asset net value management module for model parameter calibration.

Images