Carbon asset and carbon checking management method and platform based on AI

Abstract

The invention provides an AI-based carbon asset and carbon inspection management method and platform, and belongs to the technical field of carbon emission management, and the method comprises the steps: inputting the operation activity data of a target object into a preset energy efficiency analysis model, and obtaining an energy efficiency analysis result; based on the energy efficiency analysis result, performing energy flow modeling and analysis on the operation activity data of the target object to obtain an energy flow analysis result; the energy flow analysis result is input into a preset energy efficiency balance optimization model, an optimized energy distribution scheme and carbon emission prediction values, predicted based on the energy distribution scheme, of all links are obtained, and the energy efficiency balance optimization model is configured as follows: under the condition that the production constraint condition of the target object is met, the energy distribution scheme is optimized; optimizing energy distribution by taking minimization of the total carbon emission as a target; comprehensively calculating the carbon emission predicted value of each link to obtain a target total carbon emission predicted value of the target object; and generating a corresponding target carbon asset management strategy according to the target total carbon emission prediction value, the carbon transaction market data and the carbon sink data.

Description

Technical Field

[0001] This application relates to the field of carbon emission management technology, and in particular to an AI-based carbon asset and carbon verification management method and platform. Background Art

[0002] With the advancement of global actions to combat climate change and the proposal of carbon peak and carbon neutrality goals, carbon asset management and carbon verification have become key links for enterprises to achieve sustainable development.

[0003] Currently, carbon verification and management methods rely on manual data collection and empirical analysis, which is not only inefficient and inaccurate, but also difficult to fully grasp the complex and changing dynamics of the carbon trading market and the carbon emissions in corporate operations. In addition, when processing operational activity data, carbon trading market data and carbon sink data, there are problems such as data integration difficulties and insufficient analysis depth, which makes it difficult to meet the needs of enterprises for optimizing energy utilization efficiency and precise control of carbon emissions.

[0004] Therefore, there is an urgent need for an AI-based carbon asset and carbon verification management method and platform to meet the needs of enterprises for optimizing energy utilization efficiency and accurately controlling carbon emissions. Summary of the Invention

[0005] In order to solve the above technical problems, this application provides an AI-based carbon asset and carbon verification management method and platform.

[0006] A first aspect of an embodiment of the present application provides an AI-based carbon asset and carbon verification management method, comprising:

[0007] Acquire multi-source data of a target object, wherein the multi-source data includes: operational activity data, carbon trading market data, and carbon sink data of the target object;

[0008] Inputting the operational activity data of the target object into a preset energy efficiency analysis model to obtain an energy efficiency analysis result representing the energy utilization efficiency of the target object;

[0009] Based on the energy efficiency analysis results, energy flow modeling and analysis are performed on the operational activity data of the target object to obtain energy flow analysis results that describe energy flow paths, conversion efficiency, and loss distribution;

[0010] Inputting the energy flow analysis results into a preset energy efficiency balance optimization model to obtain an optimized energy allocation plan and a predicted carbon emission value for each link based on the energy allocation plan, wherein the energy efficiency balance optimization model is configured to optimize energy allocation with the goal of minimizing total carbon emissions while satisfying the production constraints of the target object;

[0011] Comprehensively calculate the carbon emission forecast values ​​of each link to obtain the target total carbon emission forecast value of the target object;

[0012] A corresponding target carbon asset management strategy is generated according to the target total carbon emission forecast value, the carbon trading market data, and the carbon sink data.

[0013] In another possible implementation of the first aspect, the AI-based carbon asset and carbon verification management method further includes:

[0014] Obtain the actual carbon emissions verification report of the target entity during the actual compliance period;

[0015] Calculate the actual carbon emissions data of the target object based on the actual carbon emissions verification report;

[0016] Performing a deviation analysis on the actual total carbon emissions data and the target total carbon emissions forecast value to obtain a deviation analysis result;

[0017] Based on the deviation result, the parameters of the energy efficiency balance optimization model are iteratively optimized.

[0018] In another possible implementation of the first aspect, the iteratively optimizing the parameters of the energy efficiency balance optimization model based on the deviation result includes:

[0019] When the absolute deviation value in the deviation result is greater than a first threshold value or the relative deviation percentage is greater than a second preset threshold value, one or more key links that contribute the most to the deviation between the target total carbon emissions forecast value and the actual total carbon emissions data are identified;

[0020] Obtain operational activity data related to the key links and environmental parameter data that directly affect carbon emissions from the key links in the actual carbon emissions verification report;

[0021] The operational activity data, environmental parameter data and corresponding actual carbon emission data of the key links are used as new training samples;

[0022] The energy efficiency balance optimization model is trained using the newly added training samples to iteratively optimize its parameters.

[0023] In another possible implementation of the first aspect, generating a corresponding target carbon asset management strategy based on the target total carbon emissions forecast value, the carbon trading market data, and the carbon sink data includes:

[0024] According to the target total carbon emission forecast value, the carbon trading market data and the carbon sink data, query a preset carbon asset management strategy knowledge base, and determine a corresponding target carbon asset management strategy based on preset matching rules; or

[0025] Inputting the target total carbon emission forecast value, the carbon trading market data, and the carbon sink data into a carbon asset management strategy recommendation model to obtain a target carbon asset management strategy corresponding to the target object;

[0026] The target carbon asset management strategy includes: one or more of a carbon quota trading program, a carbon sink project offset path, and internal emission reduction measures.

[0027] In another possible implementation of the first aspect, inputting the target total carbon emissions forecast value, the carbon trading market data, and the carbon sink data into a preset carbon asset management strategy recommendation model to obtain a target carbon asset management strategy corresponding to the target object includes:

[0028] Determining an initial action space vector of a reinforcement learning algorithm corresponding to the target carbon asset management strategy;

[0029] Determine multiple standard carbon asset management strategies and determine the action space vector of the reinforcement learning algorithm corresponding to each standard carbon asset management strategy;

[0030] Determining a state space vector of a reinforcement learning algorithm based on the target total carbon emissions forecast value, the carbon trading market data, and the carbon sink data;

[0031] Determining a value function of the reinforcement learning algorithm based on the action space vector, the state space vector, and the reward function;

[0032] Determining an optimal action vector, i.e., the target carbon asset management strategy, based on the initial action space vector and a feedback value obtained by solving the value function;

[0033] The reward function is as follows: a positive reward is given when, after executing the action space vector, the predicted or actual total compliance cost is reduced and the production constraints are not violated; a negative reward is given when executing the action space vector causes the predicted or actual production to be interrupted, the total compliance cost to increase, or the constraints to be violated.

[0034] In another possible implementation of the first aspect, inputting the target total carbon emissions forecast value, the carbon trading market data, and the carbon sink data into a preset carbon asset management strategy recommendation model to obtain a target carbon asset management strategy corresponding to the target object further includes:

[0035] Determine penalty factors based on risk indicators in the carbon trading market;

[0036] Determining a value function of the reinforcement learning algorithm based on the action space vector, the state space vector, the reward function, and the penalty factor;

[0037] An optimal action vector, ie, the target carbon asset management strategy, is determined based on the initial action space vector and the feedback value obtained by solving the value function.

[0038] In another possible implementation of the first aspect, determining the penalty factor based on the risk indicator of the carbon trading market includes:

[0039] Obtain historical carbon price data, real-time market depth data, trading volume data and relevant policy texts from the carbon trading market data;

[0040] Calculate price volatility based on the standard deviation of historical carbon price data within a preset time window;

[0041] The liquidity index score is calculated based on the combination of market depth and trading volume decay rate according to the preset weight formula;

[0042] Analyze relevant policy texts using natural language processing technology, extract keywords, and determine policy risk levels based on preset rules or classification models;

[0043] The price volatility index, liquidity index score and policy risk level are input into a preset risk scoring function to obtain a penalty factor.

[0044] In another possible implementation of the first aspect, the method for calculating the predicted carbon emission values ​​of each link includes:

[0045] Analyze the energy allocation plan to obtain the energy type and energy consumption of each link;

[0046] Obtain emission factors corresponding to energy types in each link based on the carbon emission factor database;

[0047] For the direct emission links in each link, the emission factor method is used to calculate the direct carbon emissions of the corresponding link;

[0048] For the links involving chemical reactions in the above-mentioned links, the mass balance method is used to calculate the process carbon emissions of the corresponding links.

[0049] In another possible implementation of the first aspect, the comprehensively calculating the predicted carbon emissions of each link to obtain the predicted target total carbon emissions value of the target object includes:

[0050] Obtain the first carbon emission data of the suppliers used in each link during the production process, and obtain the second carbon emission data of the products used in each link during the transportation process;

[0051] The direct carbon emissions, process carbon emissions, corresponding first carbon emission data and second carbon emission data of each link are calculated according to a preset accounting standard to obtain a target total carbon emission forecast value of the target object.

[0052] In another possible implementation of the first aspect, the AI-based carbon asset and carbon verification management method further includes:

[0053] Generate a carbon emissions verification report according to a preset template based on the target total carbon emissions forecast value, the carbon emissions forecast value of each link, and the energy allocation plan;

[0054] Inputting the carbon emission verification report into a verification rule engine for compliance verification, wherein the verification rule engine is pre-installed with the carbon verification policy of the jurisdiction where the target entity is located and the ISO14064 standard clauses;

[0055] If the compliance check fails, the abnormal data link will be located based on the abnormal identification output by the rule engine;

[0056] If the compliance check passes, the carbon emission verification report is electronically signed and a carbon emission verification report with a digital signature is output.

[0057] A second aspect of the embodiments of the present application provides an AI-based carbon asset and carbon verification management platform, including:

[0058] A data acquisition module, which acquires multi-source data of a target object, wherein the multi-source data includes: operational activity data, carbon trading market data, and carbon sink data of the target object;

[0059] An energy efficiency analysis module, configured to input the operational activity data of the target object into a preset energy efficiency analysis model to obtain an energy efficiency analysis result characterizing the energy utilization efficiency of the target object;

[0060] An energy flow analysis module is used to perform energy flow modeling and analysis on the operational activity data of the target object based on the energy efficiency analysis results, and obtain energy flow analysis results that describe the energy flow path, conversion efficiency and loss distribution;

[0061] An energy efficiency balance optimization module is configured to input the energy flow analysis results into a preset energy efficiency balance optimization model to obtain an optimized energy allocation plan and a predicted carbon emission value for each link based on the energy allocation plan, wherein the energy efficiency balance optimization model is configured to optimize energy allocation with the goal of minimizing total carbon emissions while satisfying the production constraints of the target object;

[0062] A target carbon emission total amount prediction module is used to comprehensively calculate the carbon emission prediction values ​​of each link to obtain the target carbon emission total amount prediction value of the target object;

[0063] The carbon asset management strategy module is used to generate a corresponding target carbon asset management strategy based on the target total carbon emission forecast value, the carbon trading market data and the carbon sink data.

[0064] A third aspect of an embodiment of the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of the above-mentioned AI-based carbon asset and carbon verification management method are implemented.

[0065] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned AI-based carbon asset and carbon verification management method are implemented.

[0066] The beneficial effects of an AI-based carbon asset and carbon verification management method and platform provided by the embodiment of the present application are as follows: the present application uses energy efficiency analysis models and energy flow modeling to deeply analyze the energy utilization status, from efficiency evaluation to energy flow detail analysis, to fully grasp the energy usage of the target object; secondly, through the energy efficiency balance optimization model, while meeting production constraints, it optimizes energy distribution with the goal of minimizing carbon emissions, which can not only ensure production but also effectively reduce carbon emissions. Finally, the predicted values ​​of each link are combined to obtain the target total carbon emission predicted value, and the carbon asset management strategy is generated by combining carbon trading and carbon sink data, which meets the needs of enterprises for energy utilization efficiency optimization and precise control of carbon emissions. In addition, this method enables enterprises to reasonably plan transactions and utilize carbon sink resources in the carbon market, reduce costs, and increase the value of carbon assets. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A flowchart of an AI-based carbon asset and carbon verification management method provided in one embodiment of the present application;

[0068] Figure 2 This is a structural diagram of the AI-based carbon asset and carbon verification management platform provided in one embodiment of the present application;

[0069] Figure 3 A schematic block diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0070] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0071] To make the purpose, technical solutions and advantages of this application clearer, Figure 1-3 The following description will be given through specific examples.

[0072] Please refer to Figure 1 , Figure 1 A flowchart of an AI-based carbon asset and carbon verification management method provided in one embodiment of the present application includes:

[0073] S101: Acquire multi-source data of a target object, where the multi-source data includes: operational activity data, carbon trading market data, and carbon sink data of the target object.

[0074] In this embodiment, the target entity is an enterprise or organization requiring carbon asset and carbon verification management. This embodiment uses a pre-defined interface within the carbon asset and carbon verification management platform to obtain operational activity data from each device within the target entity in real time, providing data support for subsequent carbon asset management. This operational activity data includes device operating status, power consumption, operating hours, material consumption data, energy procurement records, and more.

[0075] The equipment of the target object of this embodiment includes: energy management equipment and production operation equipment, etc.; among them, energy management equipment includes: energy conversion equipment and energy control equipment; among them, production operation equipment includes: core production equipment and auxiliary production equipment; among them, energy conversion equipment includes: motors, boilers, heat exchangers, etc., which are energy conversion equipment that converts primary energy into secondary energy required for production; among them, energy control equipment includes: inverters and smart switches, etc., which are energy control equipment used to optimize energy distribution, improve energy utilization efficiency, and reduce energy waste by adjusting equipment operating power, start and stop time, etc.; among them, core production equipment is determined according to the industry attributes of the target object, such as production line equipment in the manufacturing industry, reactors in the chemical industry, etc. The operating status and energy consumption of these production equipment directly affect the operating activities and carbon emissions of the target object; among them, auxiliary production equipment includes: air compressors, ventilators and communication equipment, etc., which are equipment used to ensure the normal operation of core production equipment.

[0076] This embodiment can calculate the carbon emissions of each device through operational activity data. Specifically, the calculation formula for carbon emissions is usually based on the power consumption and operating time of the equipment, combined with the carbon emission coefficient. In addition, the carbon asset and carbon verification management platform obtains carbon trading market data and carbon sink data through the blockchain data interface and public data capture technology; among them, carbon trading market data includes carbon quota price fluctuations, transaction history data, policy dynamics, etc., and carbon sink data includes: forestry carbon sink project progress and soil carbon sequestration monitoring data, etc. In addition, after obtaining the multi-source data of the target object, this embodiment will use machine learning algorithms to clean, denoise, and normalize the obtained multi-source data, remove outliers and redundant information, and provide a high-quality data foundation for subsequent analysis.

[0077] S102: Inputting the operational activity data of the target object into a preset energy efficiency analysis model to obtain an energy efficiency analysis result representing the energy utilization efficiency of the target object.

[0078] In this embodiment, a preset energy efficiency analysis model is obtained by training a long short-term memory network or a convolutional neural network using historical energy efficiency data. The model can obtain energy efficiency analysis results that characterize the energy utilization efficiency of the target object through deep mining and feature extraction of operational activity data, and present them in the form of visual charts to intuitively display the energy utilization efficiency status of the target object. The energy efficiency analysis results include indicators such as the energy utilization rate of each production link and the equipment energy efficiency ratio, as well as a comparative analysis with the industry average level.

[0079] S103: Based on the energy efficiency analysis results, energy flow modeling and analysis are performed on the target object's operational activity data to obtain energy flow analysis results that describe the energy flow path, conversion efficiency, and loss distribution.

[0080] In this embodiment, based on the results of energy efficiency analysis, AI-driven energy flow modeling technology is used to perform energy flow modeling and analysis on the operational activity data of the target object. Specifically, a graph neural network is used to construct an energy flow network model, and the production system of the target object is abstracted into a network structure containing energy input nodes, conversion nodes, output nodes, and energy flow edges. This embodiment determines the direction, flow rate, and conversion efficiency of energy flow between nodes through analysis of operational activity data, simulates the flow path of energy in different links, and the distribution of energy losses during the conversion process, and provides an accurate and quantitative analysis basis for energy optimization.

[0081] S104: Input the energy flow analysis results into the preset energy efficiency balance optimization model to obtain the optimized energy allocation plan and the carbon emission prediction value of each link based on the energy allocation plan. Among them, the energy efficiency balance optimization model is configured as follows: while meeting the production constraints of the target object, optimize energy allocation with the goal of minimizing the total carbon emissions; (production constraints are: material balance constraints between processes; maximum and minimum equipment load rate constraints; energy supply stability constraints; total carbon emission threshold constraints).

[0082] In this embodiment, linear programming is used to construct an energy efficiency balance optimization model; wherein, the model is subject to the production constraints of the target object, wherein the production constraints include material balance constraints between processes (ensuring that the input and output of materials in the production process are balanced), equipment maximum and minimum load rate constraints (ensuring that the equipment operates within a reasonable operating range), energy supply stability constraints (maintaining the continuity and stability of energy supply), and total carbon emission threshold constraints (complying with national or industry-mandated carbon emission limits); energy allocation is optimized with minimizing total carbon emissions as the optimization goal. In this embodiment, the energy flow analysis results are input into the energy efficiency balance optimization model, and through iterative calculation and optimization solution of the model, an optimized energy allocation plan is obtained. At the same time, based on this energy allocation plan, the carbon emission forecast value of each link is predicted.

[0083] In one possible embodiment, the energy efficiency balance optimization model is:

[0084] Among them, the objective function (minimizing total carbon emissions):

[0085]

[0086] in, For the The first step of the allocation Energy amount, links, Energy types; For the The first step uses Carbon emission factors of various energy sources.

[0087] Constraints:

[0088] Material balance constraints between processes:

[0089]

[0090] in, To input process A collection of links; Output process A collection of links; For input The material conversion coefficient is determined by the production process; For the output link The material conversion coefficient is determined by the production process; For the The first input link consumes Amount of energy; For the The output link consumes Amount of energy.

[0091] Equipment load rate constraints:

[0092]

[0093] in, For the Minimum load rate of link equipment; For the Maximum load rate of link equipment; is the equipment load factor corresponding to unit energy.

[0094] Energy supply stability constraints:

[0095]

[0096] in, For the Minimum supply capacity of the energy source; For the The maximum supply capacity of this energy.

[0097] Total carbon emissions threshold constraints:

[0098]

[0099] in, A carbon emissions cap set by policy.

[0100] Use the simplex method to solve the linear programming problem and output:

[0101] Optimal energy allocation plan ;

[0102] Predicted carbon emissions at each stage: ;

[0103] Total carbon emissions forecast .

[0104] S105: Comprehensively calculate the predicted carbon emissions for each link to obtain the target total carbon emissions forecast for the target object. In this embodiment, the predicted carbon emissions for each link are combined and weighted to obtain the target total carbon emissions forecast for the target object. This embodiment assigns weights to different production links based on their contribution to carbon emissions, resulting in a more accurate calculation result.

[0105] S106: Generate a corresponding target carbon asset management strategy based on the target total carbon emission forecast value, carbon trading market data, and carbon sink data.

[0106] In this embodiment, there are multiple ways to determine the strategy based on the predicted target total carbon emissions, combined with carbon quota price trends and transaction costs in the carbon trading market data, and available carbon sink volume and carbon sink transaction prices in the carbon sink data.

[0107] The first method is to generate the corresponding target carbon asset management strategy based on the predicted value of the target total carbon emissions, combined with the carbon quota price trend and transaction costs in the carbon trading market data, as well as the available carbon sink volume and carbon sink transaction price in the carbon sink data, using decision analysis methods such as decision trees and Bayesian networks.

[0108] The second method involves querying a pre-defined carbon asset management strategy knowledge base, which contains numerous carbon asset management strategy examples across various industries and carbon emission scenarios. Based on pre-defined matching rules, the current data is matched against the examples in the knowledge base to select the most suitable target carbon asset management strategy. These matching rules include: lowest cost, lowest risk, and optimal offset ratio. For example, when the matching rule is lowest cost, the strategy that minimizes carbon management costs is prioritized.

[0109] The third method is to input the target total carbon emissions forecast value, carbon trading market data and carbon sink data into the carbon asset management strategy recommendation model; this model is based on the reinforcement learning algorithm and outputs the target carbon asset management strategy corresponding to the target object.

[0110] In this embodiment, the target carbon asset management strategy includes but is not limited to: formulating a carbon quota purchase plan when the total predicted carbon emissions exceed a threshold; planning the timing of carbon quota sales when there are excess carbon quotas; and achieving optimal allocation of carbon assets through the development or purchase of carbon sink projects to reduce the company's carbon management costs and carbon emission risks.

[0111] From the above, it can be concluded that this application uses energy efficiency analysis models and energy flow modeling to deeply analyze the energy utilization status, from efficiency evaluation to detailed analysis of energy flow, to fully grasp the energy usage of the target object; secondly, through the energy efficiency balance optimization model, while meeting production constraints, it optimizes energy distribution with the goal of minimizing carbon emissions, which can not only ensure production but also effectively reduce carbon emissions. Finally, the predicted values ​​of each link are combined to obtain the target total carbon emissions forecast value, and the carbon asset management strategy is generated by combining carbon trading and carbon sink data, meeting the company's needs for energy utilization efficiency optimization and precise control of carbon emissions. In addition, this method enables companies to rationally plan transactions and utilize carbon sink resources in the carbon market, reducing costs and increasing the value of carbon assets.

[0112] In one embodiment of the present application, the AI-based carbon asset and carbon verification management method further includes:

[0113] Obtain the actual carbon emissions verification report of the target entity during the actual compliance period;

[0114] Based on the actual carbon emission verification report, calculate the actual total carbon emission data of the target object;

[0115] Conduct deviation analysis on the actual total carbon emissions data and the target total carbon emissions forecast value to obtain the deviation analysis results;

[0116] Based on the deviation results, the parameters of the energy efficiency balance optimization model are iteratively optimized.

[0117] In this embodiment, blockchain technology is used to obtain the target entity's actual carbon emissions verification report for the actual compliance period. This report is uploaded to the blockchain by a third-party certification body. Simultaneously, this embodiment also performs an integrity check on the obtained actual carbon emissions verification report to verify that it contains key information such as basic enterprise information, verification scope, verification method, emission source identification, and data quality control. If any missing information is found, an early warning mechanism is triggered, prompting relevant personnel to complete it.

[0118] In this embodiment, the actual total carbon emissions data for the target entity is calculated based on the actual carbon emissions verification report. Specifically, the carbon emissions of each emission source are accurately calculated based on the calculation method for each emission source, combined with the activity data, emission factors, and other parameters provided in the actual carbon emissions verification report, in accordance with relevant standards such as the "Greenhouse Gas Emissions Accounting and Reporting Requirements." The carbon emissions of each emission source are aggregated through a weighted summation method, and the carbon offset amount in the actual carbon emissions verification report is calculated to ultimately obtain the actual total carbon emissions data for the target entity. The carbon offset amount in the actual carbon emissions verification report is offset through carbon sink projects or carbon quota trading.

[0119] In this embodiment, deviation analysis can be performed on the actual total carbon emissions data and the target total carbon emissions forecast value from multiple dimensions. At the numerical level, the absolute deviation and relative deviation between the actual total carbon emissions data and the target total carbon emissions forecast value are calculated. At the temporal dimension, the deviation trend over different time periods is compared to analyze whether the deviation exhibits cyclical characteristics.

[0120] This embodiment deeply analyzes the deviation of each emission source to determine the key emission sources that have a greater impact on the total deviation. Based on the energy efficiency analysis results, energy flow analysis results and the actual situation during production operations, the causes of the deviation are analyzed, such as fluctuations in production process parameters, changes in energy supply quality, and unreasonable model parameter settings.

[0121] In this embodiment, this embodiment can establish a parameter optimization rule base based on the deviation analysis results, and the parameter optimization rule base includes a mapping relationship between different deviation types and parameter adjustment strategies. For example, when it is found that the actual carbon emissions of a certain production link far exceed the predicted value and are not related to the unreasonable energy allocation, the adjustment rule for the energy allocation weight parameter of this link is triggered. This embodiment can also use optimization algorithms such as genetic algorithms to iteratively optimize the parameters of the energy efficiency balance optimization model with the objective function of reducing the deviation between the actual total carbon emissions and the predicted value; in the optimization process, set the iteration termination conditions, and the iteration termination conditions include reaching the maximum number of iterations, the objective function value converges to a certain accuracy range, etc.

[0122] In summary, this embodiment forms a closed loop of prediction-execution-feedback-optimization through deviation analysis between actual carbon emission data and predicted values, so that the energy efficiency balance optimization model dynamically adjusts the parameters of the energy efficiency balance optimization model according to actual operating conditions, thereby improving the prediction accuracy and the accuracy of carbon emission accounting.

[0123] In one embodiment of the present application, based on the deviation result, iteratively optimizing the parameters of the energy efficiency balance optimization model includes: when the absolute deviation value in the deviation result is greater than a first threshold value or the relative deviation percentage is greater than a second preset threshold value, identifying one or more key links that contribute the most to the deviation between the target total carbon emissions prediction value and the actual total carbon emissions data;

[0124] Obtain operational activity data related to key links and environmental parameter data that directly affect carbon emissions in key links from actual carbon emission verification reports;

[0125] The operational activity data, environmental parameter data and corresponding actual carbon emission data of key links are used as new training samples;

[0126] The energy efficiency balance optimization model is trained using the newly added training samples, and its parameters are iteratively optimized.

[0127] In this embodiment, the first threshold and the second preset threshold are derived from a statistical analysis of historical carbon emissions data for the target industry. Specifically, by collecting data on the deviation between the predicted and actual carbon emissions of a large number of companies within the industry, and applying statistical methods such as calculating the mean and standard deviation, combined with the industry's carbon emissions management accuracy requirements, a reasonable first threshold and second preset threshold are determined. For example, if the absolute deviation corresponding to the mean plus twice the standard deviation of the deviation data within the industry is X, the first threshold is set to a value slightly higher than X; if the 90th percentile of the relative deviation within the industry is Y%, the second preset threshold is set to Y%.

[0128] In this embodiment, when the absolute deviation value in the deviation result is greater than the first threshold or the relative deviation percentage is greater than the second preset threshold, the sensitivity analysis method is used to calculate the contribution of each production link to the total deviation; the parameters such as energy consumption and carbon emission factor of each link are perturbed, and the change range of the total deviation is observed. The larger the change range of the link, the higher its contribution to the total deviation, thereby identifying one or more key links that contribute the most to the deviation between the target total carbon emissions forecast value and the actual total carbon emissions data.

[0129] In this embodiment, a key link identification model based on association rule mining can be constructed to analyze the correlation between the target total carbon emissions prediction value and the actual total carbon emissions data in each production link and energy use process, and identify one or more key links that contribute the most to the deviation.

[0130] In this embodiment, the text content of the actual carbon emission verification report is parsed by natural language processing technology, combined with structured data extraction methods, to accurately obtain operational activity data related to key links. For example, equipment operating time, energy input and output flow, material ratio, etc. In this embodiment, environmental parameter data that directly affect carbon emissions in key links are collected through meteorological monitoring equipment and geographic information systems, including meteorological conditions such as temperature, humidity, wind speed, as well as external environmental data such as regional power grid energy structure and real-time policy changes in the carbon trading market. This embodiment preprocesses the acquired operational activity data, environmental parameter data and corresponding actual carbon emission data of key links; wherein the preprocessing includes: using data normalization methods to unify data of different magnitudes into the same interval; and using data encoding technology to convert non-numerical data into numerical forms that can be processed by the energy efficiency balance optimization model.

[0131] In summary, this embodiment focuses on the key links with the greatest contribution when the deviation exceeds the threshold, improving the efficiency and pertinence of model iteration. Secondly, by combining operational data and environmental parameters from these key links as additional training samples, the model more accurately captures the core factors influencing carbon emissions, improving local prediction accuracy.

[0132] In one embodiment of the present application, a corresponding target carbon asset management strategy is generated based on the target total carbon emission forecast value, carbon trading market data, and carbon sink data, including:

[0133] According to the target total carbon emission forecast value, carbon trading market data and carbon sink data, query the preset carbon asset management strategy knowledge base and determine the corresponding target carbon asset management strategy based on the preset matching rules; or

[0134] Input the target total carbon emission forecast value, carbon trading market data and carbon sink data into the carbon asset management strategy recommendation model to obtain the target carbon asset management strategy for the corresponding target object;

[0135] The target carbon asset management strategy includes one or more of: carbon quota trading schemes, carbon sink project offset paths, and internal emission reduction measures.

[0136] In this embodiment, the carbon asset management strategy knowledge base stores a large number of carbon asset management strategy cases based on different industries and carbon emission scenarios. Specifically, the carbon asset management strategy knowledge base includes a rule base, a case base, and a model base. The rule base stores deterministic knowledge such as carbon trading policies and regulations, and industry standards. The case base collects historical carbon asset management success and failure cases, each of which includes information such as enterprise characteristics, strategy selection, and implementation results. The model base includes models for carbon emission forecasting, carbon price fluctuation analysis, and carbon sink project evaluation.

[0137] In this embodiment, matching rules include: lowest cost, lowest risk, and optimal offset ratio. Specifically, if the lowest cost rule is selected, the costs of different strategy combinations are first calculated, including the cost of purchasing carbon quotas, the investment cost of carbon sink projects, and the cost of modifying internal emission reduction measures. Simultaneously, based on the potential cost changes brought about by price fluctuations in the carbon trading market, Monte Carlo simulation methods are used to predict future carbon prices, evaluate the expected costs of each strategy under different market scenarios, and select the target carbon asset management strategy with the lowest cost. If the lowest risk rule is selected, a risk assessment indicator system is constructed to score strategies based on market risk, policy risk, technical risk, and other dimensions, with priority given to strategies with the lowest overall risk scores.

[0138] In this example, a carbon asset management strategy recommendation model is constructed based on a deep Q-network reinforcement learning model. The defined state space includes information such as target carbon emissions forecasts, carbon trading market data, carbon sink data, and the company's own carbon asset reserves and production and operation status. The defined action space includes various carbon asset management strategy options, including carbon quota trading, carbon sink project selection, and implementation of internal emission reduction measures. The reward function is designed based on the actual results of strategy implementation. For example, positive results such as carbon emission reductions, carbon asset returns, and cost savings are rewarded positively, while negative results such as exceeding carbon emission targets and trading losses are rewarded negatively.

[0139] In this embodiment, the carbon quota trading scheme includes developing a carbon quota trading plan based on a comparison of the predicted target carbon emissions and the number of carbon quotas held by the enterprise. When the predicted total carbon emissions exceed the held quotas, price trends in the carbon trading market are analyzed to purchase the full quota during periods of low carbon prices. When there are surplus allowances, sales are planned based on market supply and demand and price fluctuations to increase the value of carbon assets. Furthermore, financial instruments such as hedging are used to mitigate trading risks associated with carbon price fluctuations.

[0140] The carbon sequestration project offsetting path involves assessing a company's own resources and conditions and selecting appropriate carbon sequestration project types, such as forestry carbon sequestration, wetland carbon sequestration, or carbon credits generated by renewable energy projects. For forestry carbon sequestration projects, planning is required regarding the afforestation area, tree species selection, planting scale, and maintenance plan. For renewable energy projects, feasibility and economic viability are analyzed to determine specific strategies for investment and construction or purchasing carbon credits. During implementation, strict adherence to relevant standards and certification processes ensures that the emission reductions from carbon sequestration projects are verifiable and tradable.

[0141] Internal emission reduction measures include: developing internal emission reduction measures in energy management, production process improvements, and equipment upgrades. In terms of energy management, we have introduced intelligent energy management systems to monitor energy consumption in real time and optimize energy distribution. In terms of production processes, we have developed or adopted low-carbon production technologies to reduce carbon emissions per unit of product. For high-energy-consuming equipment, we have developed equipment replacement plans, replacing it with energy-saving equipment to improve energy efficiency and reduce carbon emissions.

[0142] In this embodiment, enterprises can flexibly combine the above three strategies according to their actual conditions to form the most suitable target carbon asset management strategy to meet the management needs of different enterprises.

[0143] In one embodiment of the present application, the target total carbon emissions forecast value, carbon trading market data, and carbon sink data are input into a preset carbon asset management strategy recommendation model to obtain a target carbon asset management strategy for the target object, including:

[0144] Determine the initial action space vector of the reinforcement learning algorithm corresponding to the target carbon asset management strategy;

[0145] Determine multiple standard carbon asset management strategies and determine the action space vector of the reinforcement learning algorithm corresponding to each standard carbon asset management strategy;

[0146] Determine the state space vector of the reinforcement learning algorithm based on the target total carbon emissions forecast value, carbon trading market data, and carbon sink data;

[0147] Determine the value function of the reinforcement learning algorithm based on the action space vector, state space vector, and reward function;

[0148] Determine the optimal action vector, i.e., the target carbon asset management strategy, based on the initial action space vector and the feedback value obtained by solving the value function;

[0149] The reward function is: when the execution of the action space vector reduces the predicted or actual total compliance cost and does not violate the production constraints, a positive reward is given; when the execution of the action space vector leads to the predicted or actual production interruption, the total compliance cost increases, or the constraints are violated, a negative reward is given.

[0150] In this embodiment, the various optional actions in the carbon asset management strategy are encoded to form an initial action space vector. Action types such as carbon quota trading, carbon sink project selection, and implementation of internal emission reduction measures are encoded, with each action type corresponding to a dimension in the vector. For example, if there are n possible action types, the initial action space vector is an n-dimensional vector. The value of each position in the vector represents the probability or weight of executing the corresponding action. Initially, the weights of each action can be set equal or based on historical experience to set initial preferences.

[0151] In this embodiment, multiple standard carbon asset management strategies can be determined by analyzing typical successful cases in the industry; and the action space vector corresponding to each standard carbon asset management strategy can be statistically analyzed through expert knowledge or historical data.

[0152] In this example, a state space vector for a reinforcement learning algorithm is constructed based on the target carbon emissions forecast, carbon trading market data, and carbon sink data. The target carbon emissions forecast is normalized and mapped to the interval [0, 1], serving as one dimension of the state vector. For the carbon trading market data, features such as carbon price trends (e.g., the price change rate over three months), trading volume, and market liquidity are extracted and compressed into low-dimensional vectors using dimensionality reduction methods such as principal component analysis. For the carbon sink data, features such as available carbon sink volume, carbon sink price, and the rate of return of different types of carbon sink projects are extracted and similarly reduced. The processed feature vectors are then concatenated to form a complete state space vector.

[0153] In this embodiment, based on the initial action space vector and the feedback value obtained by solving the value function, a greedy strategy is adopted to explore the action space and find the action vector with the maximum value. During the exploration process, the exploration probability is dynamically adjusted according to the risk preference of the enterprise. Enterprises with higher risk preferences appropriately increase the exploration intensity to discover better strategies; enterprises with lower risk preferences are more inclined to choose known better strategies.

[0154] In one embodiment of the present application, the target total carbon emissions forecast value, carbon trading market data, and carbon sink data are input into a preset carbon asset management strategy recommendation model to obtain a target carbon asset management strategy for the target object, further comprising:

[0155] Determine penalty factors based on risk indicators in the carbon trading market;

[0156] Determine the value function of the reinforcement learning algorithm based on the action space vector, state space vector, reward function and penalty factor;

[0157] The optimal action vector, i.e., the target carbon asset management strategy, is determined based on the initial action space vector and the feedback value obtained by solving the value function.

[0158] In this embodiment, a penalty factor calculation model is constructed based on risk indicators in the carbon trading market. This model is constructed based on multiple factors, including market price volatility risk, liquidity risk, and policy risk. This embodiment quantifies and weights each risk indicator to generate a dynamically adjusted penalty factor. The penalty factor is used to impose additional penalties on high-risk strategies in the value function calculation of reinforcement learning, guiding the agent to learn the optimal strategy that balances risk and reward.

[0159] In this embodiment, by adding a risk penalty term to the value function, the dynamic risk of the carbon trading market is converted into a quantifiable decision constraint, thereby avoiding decision failures caused by market uncertainty and preventing a surge in compliance costs caused by sudden policy changes.

[0160] In one embodiment,

[0161] The value function is:

[0162]

[0163] The reward function is:

[0164]

[0165] in, is the state space vector; is the action space vector; A is the action space, including all possible actions; is the change in total compliance cost, and ; is the baseline total compliance cost, which is calculated using the compliance cost when the strategy was not implemented in historical data; Identify violations of production constraints; is the reward base, which is a constant; is the penalty factor; is the strategic risk exposure; is the discount factor, preset [0,1).

[0166] in,

[0167]

[0168] in, is the absolute value of carbon trading volume, is the current carbon price; is the price volatility; is the absolute value of carbon sink purchases; is the unit price of carbon sink; For policy risk.

[0169] In one embodiment of the present application, determining a penalty factor based on a risk indicator of the carbon trading market includes:

[0170] Obtain historical carbon price data, real-time market depth data, trading volume data and relevant policy texts from carbon trading market data;

[0171] Calculate price volatility based on the standard deviation of historical carbon price data within a preset time window;

[0172] The liquidity index score is calculated based on the combination of market depth and trading volume decay rate according to the preset weight formula;

[0173] Analyze relevant policy texts using natural language processing technology, extract keywords, and determine policy risk levels based on preset rules or classification models;

[0174] The price volatility index, liquidity index score and policy risk level are input into the preset risk scoring function to obtain the penalty factor.

[0175] In this embodiment, based on the historical carbon price data within a preset time window, the standard deviation of the price series is calculated and used as an indicator of price volatility; the above indicator is standardized and mapped to the interval [0,1] to obtain a quantitative score of price volatility. In this embodiment, a liquidity assessment model is constructed by combining market depth data and trading volume data; market depth is defined as the total number of current buy and sell orders, and the trading volume decay rate is the sensitivity of trading volume to price changes. In this embodiment, natural language processing technology is applied to analyze relevant policy texts. The steps include: first, text preprocessing, including word segmentation, stop word removal, lemma restoration and other operations; then, extracting keywords from the policy text; constructing a policy impact classification model, classifying the policy text, and determining the policy risk level; for example, high risk level, medium risk level and low risk level.

[0176] In this embodiment, the price volatility calculation formula is:

[0177]

[0178] in, is the preset time window length; is the time point serial number; The first a carbon price; is the average carbon price within the time window.

[0179] The liquidity score calculation formula is:

[0180]

[0181] in, Score liquidity; is the weight of market depth; is the weight of the transaction volume decay rate; is the market depth; is the transaction volume decay rate.

[0182] The formula for calculating the policy risk level is:

[0183]

[0184] in, For natural language processing models; Policy documents issued by governments / regulatory bodies.

[0185] The penalty factor is calculated as follows:

[0186] in, is the price volatility weight; is the liquidity score weight; are policy risk weights; these three weights can be obtained based on historical experience or set by users.

[0187] In one embodiment of the present application, a method for calculating the predicted carbon emission value of each link includes:

[0188] Analyze the energy allocation plan to obtain the energy type and energy consumption of each link;

[0189] Based on the carbon emission factor database, the emission factors corresponding to the energy types in each link are obtained;

[0190] For the direct emission links in each link, the emission factor method is used to calculate the direct carbon emissions of the corresponding link;

[0191] For each link where chemical reactions occur, the mass balance method is used to calculate the process carbon emissions of the corresponding link.

[0192] In this example, a structured analysis is performed on the optimized energy allocation plan to extract the energy type and energy consumption for each production link. Energy types include electricity, natural gas, coal, and biomass. A data extraction algorithm is used to identify key data fields from the energy allocation plan text or table, and a mapping table between energy type and consumption is established.

[0193] In this embodiment, the carbon emission factor database includes emission factor data for different energy types, different production processes, and different regions. Specifically, a hierarchical structure is used to design the database, wherein the first layer is classified according to energy type, the second layer is classified according to technology level, and the third layer is classified according to regional characteristics. When matching the emission factors corresponding to the energy types of each link, this embodiment uses a semantic matching algorithm to match the parsed energy types with the emission factors in the carbon emission factor database, giving priority to the emission factor data that is closest to actual production. For example, for electric energy, the corresponding grid emission factor is selected based on the grid structure of the region; for fossil fuels, the corresponding emission factor is selected based on the fuel quality and the type of combustion equipment.

[0194] In this embodiment, for direct emission links, the emission factor method is used to calculate carbon emissions; wherein, direct emission links include fuel combustion and waste treatment, etc. For links with chemical reactions, the mass balance method is used to calculate process carbon emissions; wherein, links with chemical reactions include cement production, chemical synthesis, etc. In this embodiment, for links with chemical reactions, a chemical reaction equation is first established to determine the conservation of matter relationship between raw material input and product output; then, the conversion path of carbon elements in the reaction process is analyzed to calculate the emission of carbon elements. For example, in cement production, according to the limestone decomposition reaction CaCO3→CaO+CO2, combined with the limestone input and calcium carbonate content, the CO2 emissions generated by the decomposition are calculated.

[0195] In one embodiment of the present application, the carbon emission prediction values ​​of each link are comprehensively calculated to obtain the target total carbon emission prediction value of the target object, including:

[0196] Obtain the first carbon emission data of the suppliers used in each link during the production process, and obtain the second carbon emission data of the products used in each link during the transportation process;

[0197] The direct carbon emissions, process carbon emissions, corresponding first carbon emission data and second carbon emission data of each link are calculated according to the preset accounting standards to obtain the target total carbon emission forecast value of the target object.

[0198] In this embodiment, primary carbon emissions data from the production process of suppliers of raw materials and intermediate products used in each link is obtained through a data interface or questionnaire survey. This data includes carbon emissions information from the entire process of raw material extraction, processing, and transportation to the enterprise. Simultaneously, using data provided by logistics management systems and transportation service providers, secondary carbon emissions data from the transportation of products used in each link is obtained, including information such as transportation mode (road, rail, water, air), transportation distance, and energy consumption of transportation vehicles.

[0199] In this embodiment, according to the preset accounting standards (such as ISO14067 product carbon footprint standard and PAS2050 product and service life cycle greenhouse gas emission assessment specification), the direct carbon emissions, process carbon emissions, corresponding first carbon emission data and second carbon emission data of each link are integrated and calculated.

[0200] In one embodiment of the present application, the AI-based carbon asset and carbon verification management method further includes:

[0201] Generate a carbon emissions verification report according to a preset template based on the target total carbon emissions forecast, the carbon emissions forecast for each link, and the energy allocation plan;

[0202] Input the carbon emission verification report into the verification rule engine for compliance verification. The verification rule engine is pre-installed with the carbon verification policy of the jurisdiction where the target entity is located and the ISO14064 standard clauses;

[0203] If the compliance check fails, the abnormal data link will be located based on the abnormal identification output by the rule engine;

[0204] If the compliance check passes, the carbon emission verification report will be electronically signed and a carbon emission verification report with a digital signature will be output.

[0205] In this embodiment, based on the target total carbon emissions forecast value, the carbon emissions forecast value of each link and the energy allocation plan, a carbon emissions verification report with clear logic and detailed content is automatically generated according to a preset template; wherein, the carbon emissions verification report includes the current status of the company's carbon emissions, forecast trends and emission reduction measures. This embodiment can use visual chart generation technology to convert carbon emissions data into various visual forms such as line charts, bar charts, pie charts, and Sankey diagrams to intuitively display information such as the proportion of carbon emissions in each link, the relationship between energy flow and carbon emissions, etc. This embodiment can also introduce a third-party platform to obtain the target total carbon emissions forecast value, the carbon emissions forecast value of each link and the energy allocation plan to generate a carbon emissions verification report with clear logic and detailed content.

[0206] In this embodiment, the verification rule engine utilizes a microservices architecture, splitting jurisdictional policy rules and ISO14064 standard rules into independent rule microservices. A dynamic monitoring mechanism is established for jurisdictional policy rules, using web crawler technology to capture real-time updates from policy and regulatory websites. Upon discovery of policy changes, the verification rule engine automatically triggers a rule update process. Using semantic matching and rule conversion techniques from natural language processing, the new policy terms are converted into executable logical expressions.

[0207] In this embodiment, when the verification rule engine outputs an anomaly indicator, an association analysis algorithm is used, combining the logical relationships between carbon emission data and business processes, to quickly locate the data anomaly. For example, by tracing the data source, calculation process, and related influencing factors, it can accurately determine whether the anomaly is caused by data entry errors, calculation model deviations, or policy misunderstandings.

[0208] In this embodiment, a distributed electronic signature technology based on blockchain is adopted, an asymmetric encryption algorithm is used to generate a unique digital signature, and the digital signature information is bound to the content of the carbon emission verification report to ensure that the content of the carbon emission verification report cannot be tampered with.

[0209] Corresponding to the AI-based carbon asset and carbon verification management method in the above embodiment, Figure 2This is a structural block diagram of an AI-based carbon asset and carbon verification management platform provided in one embodiment of the present application. For ease of explanation, only the parts related to the embodiment of the present application are shown. Figure 2 The AI-based carbon asset and carbon verification management platform 20 includes: a data acquisition module 21, an energy efficiency analysis module 22, an energy flow analysis module 23, an energy efficiency balance optimization module 24, a target total carbon emission prediction module 25 and a carbon asset management strategy module 26.

[0210] The data acquisition module 21 acquires multi-source data of the target object, including: operation activity data, carbon trading market data and carbon sink data of the target object;

[0211] Energy efficiency analysis module 22, used to input the target object's operational activity data into a preset energy efficiency analysis model to obtain energy efficiency analysis results that characterize the target object's energy utilization efficiency;

[0212] The energy flow analysis module 23 is used to perform energy flow modeling and analysis on the target object's operational activity data based on the energy efficiency analysis results, and obtain energy flow analysis results that describe the energy flow path, conversion efficiency, and loss distribution;

[0213] The energy efficiency balance optimization module 24 is configured to input the energy flow analysis results into a preset energy efficiency balance optimization model to obtain an optimized energy allocation plan and predicted carbon emissions for each link based on the energy allocation plan. The energy efficiency balance optimization model is configured to optimize energy allocation with the goal of minimizing total carbon emissions while satisfying the production constraints of the target object.

[0214] The target carbon emission total amount prediction module 25 is used to comprehensively calculate the carbon emission prediction value of each link to obtain the target carbon emission total amount prediction value of the target object;

[0215] The carbon asset management strategy module 26 is used to generate a corresponding target carbon asset management strategy based on the target total carbon emission forecast value, carbon trading market data and carbon sink data.

[0216] See also Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided in one embodiment of the present application. Figure 3The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of the modules in the above-mentioned device embodiments, such as Figure 2 The functions of the data acquisition module 21, energy efficiency analysis module 22, energy flow analysis module 23, energy efficiency balance optimization module 24, target carbon emission total amount prediction module 25 and carbon asset management strategy module 26 are shown.

[0217] It should be understood that in the embodiment of the present application, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0218] The input device 302 may include a touchpad, a fingerprint collection sensor (for collecting user fingerprint information and fingerprint direction information), a microphone, etc. The output device 303 may include a display (LCD, etc.), a speaker, etc.

[0219] The memory 304 may include a read-only memory and a random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store device type information.

[0220] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present application can execute the implementation method described in any embodiment of the AI-based carbon asset and carbon verification management method provided in the embodiments of the present application, and can also execute the implementation method of the electronic device described in the embodiments of the present application, which will not be repeated here.

[0221] In another embodiment of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, all or part of the process of the method in the above embodiment is implemented. The computer program can also be used to instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of each of the above method embodiments are implemented. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium.

[0222] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the computer-readable storage medium can include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or is about to be output.

[0223] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0224] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0225] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or can be an electrical, mechanical or other form of connection.

[0226] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0227] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0228] The above are only specific embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An AI-based carbon asset and carbon verification management method, characterized in that: include: Acquire multi-source data of a target object, wherein the multi-source data includes: operational activity data, carbon trading market data, and carbon sink data of the target object; Inputting the operational activity data of the target object into a preset energy efficiency analysis model to obtain an energy efficiency analysis result representing the energy utilization efficiency of the target object; Based on the energy efficiency analysis results, energy flow modeling and analysis are performed on the operational activity data of the target object to obtain energy flow analysis results that describe energy flow paths, conversion efficiency, and loss distribution; Inputting the energy flow analysis results into a preset energy efficiency balance optimization model to obtain an optimized energy allocation plan and a predicted carbon emission value for each link based on the energy allocation plan, wherein the energy efficiency balance optimization model is configured to optimize energy allocation with the goal of minimizing total carbon emissions while satisfying the production constraints of the target object; Comprehensively calculate the carbon emission forecast values ​​of each link to obtain the target total carbon emission forecast value of the target object; A corresponding target carbon asset management strategy is generated according to the target total carbon emission forecast value, the carbon trading market data, and the carbon sink data.

2. The AI-based carbon asset and carbon verification management method according to claim 1 is characterized in that: Also includes: Obtain the actual carbon emissions verification report of the target entity during the actual compliance period; Calculate the actual carbon emissions data of the target object based on the actual carbon emissions verification report; Performing a deviation analysis on the actual total carbon emissions data and the target total carbon emissions forecast value to obtain a deviation analysis result; Based on the deviation result, the parameters of the energy efficiency balance optimization model are iteratively optimized.

3. The AI-based carbon asset and carbon verification management method according to claim 2 is characterized in that: The iteratively optimizing the parameters of the energy efficiency balance optimization model based on the deviation result includes: When the absolute deviation value in the deviation result is greater than a first threshold value or the relative deviation percentage is greater than a second preset threshold value, one or more key links that contribute the most to the deviation between the target total carbon emissions forecast value and the actual total carbon emissions data are identified; Obtain operational activity data related to the key links and environmental parameter data that directly affect carbon emissions from the key links in the actual carbon emissions verification report; The operational activity data, environmental parameter data and corresponding actual carbon emission data of the key links are used as new training samples; The energy efficiency balance optimization model is trained using the newly added training samples to iteratively optimize its parameters.

4. The AI-based carbon asset and carbon verification management method according to claim 1, characterized in that: The generating of a corresponding target carbon asset management strategy according to the target total carbon emission forecast value, the carbon trading market data, and the carbon sink data includes: According to the target total carbon emission forecast value, the carbon trading market data and the carbon sink data, query a preset carbon asset management strategy knowledge base, and determine a corresponding target carbon asset management strategy based on preset matching rules; or Inputting the target total carbon emission forecast value, the carbon trading market data, and the carbon sink data into a carbon asset management strategy recommendation model to obtain a target carbon asset management strategy corresponding to the target object; The target carbon asset management strategy includes: one or more of a carbon quota trading program, a carbon sink project offset path, and internal emission reduction measures.

5. The AI-based carbon asset and carbon verification management method according to claim 4 is characterized in that: The step of inputting the target total carbon emission forecast value, the carbon trading market data, and the carbon sink data into a preset carbon asset management strategy recommendation model to obtain a target carbon asset management strategy corresponding to the target object includes: Determining an initial action space vector of a reinforcement learning algorithm corresponding to the target carbon asset management strategy; Determine multiple standard carbon asset management strategies and determine the action space vector of the reinforcement learning algorithm corresponding to each standard carbon asset management strategy; Determining a state space vector of a reinforcement learning algorithm based on the target total carbon emissions forecast value, the carbon trading market data, and the carbon sink data; Determining a value function of the reinforcement learning algorithm based on the action space vector, the state space vector, and the reward function; Determining an optimal action vector, i.e., the target carbon asset management strategy, based on the initial action space vector and a feedback value obtained by solving the value function; The reward function is as follows: a positive reward is given when, after executing the action space vector, the predicted or actual total compliance cost is reduced and the production constraints are not violated; a negative reward is given when executing the action space vector causes the predicted or actual production to be interrupted, the total compliance cost to increase, or the constraints to be violated.

6. The AI-based carbon asset and carbon verification management method according to claim 5 is characterized in that: Also includes: Determine penalty factors based on risk indicators in the carbon trading market; Determining a value function of the reinforcement learning algorithm based on the action space vector, the state space vector, the reward function, and the penalty factor; An optimal action vector, ie, the target carbon asset management strategy, is determined based on the initial action space vector and the feedback value obtained by solving the value function.

7. The AI-based carbon asset and carbon verification management method according to claim 6, characterized in that: The penalty factor determined based on the risk indicators of the carbon trading market includes: Obtain historical carbon price data, real-time market depth data, trading volume data and relevant policy texts from the carbon trading market data; Calculate price volatility based on the standard deviation of historical carbon price data within a preset time window; The liquidity index score is calculated based on the combination of market depth and trading volume decay rate according to the preset weight formula; Analyze relevant policy texts using natural language processing technology, extract keywords, and determine policy risk levels based on preset rules or classification models; The price volatility index, liquidity index score and policy risk level are input into a preset risk scoring function to obtain a penalty factor.

8. The AI-based carbon asset and carbon verification management method according to claim 1 is characterized in that: The calculation method of the carbon emission forecast value of each link includes: Analyze the energy allocation plan to obtain the energy type and energy consumption of each link; Obtain emission factors corresponding to the energy types of each link based on the carbon emission factor database; For the direct emission links in each link, the emission factor method is used to calculate the direct carbon emissions of the corresponding link; For the links involving chemical reactions in the above-mentioned links, the mass balance method is used to calculate the process carbon emissions of the corresponding links.

9. The AI-based carbon asset and carbon verification management method according to claim 1, characterized in that: Also includes: Generate a carbon emissions verification report according to a preset template based on the target total carbon emissions forecast value, the carbon emissions forecast value of each link, and the energy allocation plan; Inputting the carbon emission verification report into a verification rule engine for compliance verification, wherein the verification rule engine is pre-installed with the carbon verification policy of the jurisdiction where the target entity is located and the ISO14064 standard clauses; If the compliance check fails, the abnormal data link will be located based on the abnormal identification output by the rule engine; If the compliance check passes, the carbon emission verification report is electronically signed and a carbon emission verification report with a digital signature is output.

10. An AI-based carbon asset and carbon verification management platform, characterized by: The AI-based carbon asset and carbon verification management method according to any one of claims 1 to 9, wherein the AI-based carbon asset and carbon verification management platform comprises: A data acquisition module, which acquires multi-source data of a target object, wherein the multi-source data includes: operational activity data, carbon trading market data, and carbon sink data of the target object; An energy efficiency analysis module, configured to input the operational activity data of the target object into a preset energy efficiency analysis model to obtain an energy efficiency analysis result characterizing the energy utilization efficiency of the target object; An energy flow analysis module is used to perform energy flow modeling and analysis on the operational activity data of the target object based on the energy efficiency analysis results, and obtain energy flow analysis results that describe the energy flow path, conversion efficiency and loss distribution; An energy efficiency balance optimization module is configured to input the energy flow analysis results into a preset energy efficiency balance optimization model to obtain an optimized energy allocation plan and a predicted carbon emission value for each link based on the energy allocation plan, wherein the energy efficiency balance optimization model is configured to optimize energy allocation with the goal of minimizing total carbon emissions while satisfying the production constraints of the target object; A target carbon emission total amount prediction module is used to comprehensively calculate the carbon emission prediction values ​​of each link to obtain the target carbon emission total amount prediction value of the target object; The carbon asset management strategy module is used to generate a corresponding target carbon asset management strategy based on the target total carbon emission forecast value, the carbon trading market data and the carbon sink data.

Images