Multi-node carbon emission intelligent contract verification method based on block chain

Abstract

The invention relates to the technical field of data processing, provides a multi-node carbon emission intelligent contract verification method based on a block chain, and aims to solve the problems that data verification in an existing carbon asset transaction system is non-transparent, high in trust cost and difficult to adapt to policy changes. Distributed storage and dynamic verification of data are realized by constructing a hierarchical block chain architecture, the data are collected by using edge computing nodes, and third-party verification data are obtained through an oracle machine node cluster. The dynamic rule engine monitors data changes in real time, triggers a multi-stage verification process, and improves the data accuracy. And meanwhile, an improved Byzantine fault-tolerant mechanism is adopted to reach a consensus, and the system has the functions of strategy dynamic updating and security enhancement, so that the flexibility and security of the system are improved. According to the method, multi-party consensus and real-time verification of the carbon emission data are realized, trust cost of cooperation among enterprises is reduced, and healthy development of a carbon emission permit trading market is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a multi-node carbon emission intelligent contract verification method based on blockchain. Background Art

[0002] Existing carbon asset trading systems mostly adopt a centralized architecture and rely on technologies such as robotic process automation (RPA), data integration, and machine learning to improve trading efficiency and risk control capabilities. Existing carbon asset trading platforms achieve data evidence storage through blockchain technology, but their functions are mainly limited to data storage and traceability, and they fail to solve the problem of the lack of real-time verification and trust mechanism in multi-party collaboration.

[0003] Traditional systems rely on manual or centralized institutions to review carbon emission data, which is prone to verification delays due to single-point failures and difficult to meet the real-time processing requirements of massive data. Existing technologies lack a consensus mechanism for multi-party participation, and the data verification process is not transparent, resulting in high trust costs during collaboration between enterprises. Although some systems introduce blockchain technology, the intelligent contracts are only used for transaction execution and data evidence storage, and no rules are designed for multi-node dynamic verification of carbon emission data, making it difficult to adapt to complex policy changes and multi-dimensional verification requirements. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the present invention provides a multi-node carbon emission intelligent contract verification method based on blockchain, which is used to solve the problems that although some systems introduce blockchain technology, the existing blockchain carbon emission intelligent contracts are only used for transaction execution and data evidence storage, do not design rules for multi-node dynamic verification of carbon emission data, and lack a cross-verification mechanism for fuel-end accounting data and online monitoring data, making it difficult to adapt to complex policy changes and multi-dimensional carbon data quality control and verification automation requirements.

[0005] To solve the above technical problems, the specific technical solution of the present invention is as follows: The multi-node carbon emission intelligent contract verification method based on blockchain provided by the present invention includes: Step S1, constructing a hierarchical blockchain architecture, where the hierarchical blockchain architecture includes a data layer, a verification layer, and a contract layer: Step S2, based on the data layer of the hierarchical blockchain architecture, collecting enterprise-side fuel consumption, laboratory data, and carbon emission online monitoring data through edge computing nodes, and preprocessing the collected data to generate a standardized data packet carrying a hash chain data fingerprint generated by a secure hash algorithm; Step S3: Input the standardized data packets generated in Step S2 into the oracle node cluster. Through the implementation of two-way TLS authentication and zero-knowledge proof verification, establish a cross-chain trusted channel with the third-party data source, obtain the associated data verified by reputation scoring, and perform consistency comparison between the associated data and the standardized data packets. Step S4: When it is detected that the absolute value of the mutation of CO2 concentration exceeds 15% under non-start / stop conditions, the absolute value of the relative difference between the calculated emissions at the fuel end and the online monitoring data at the emission end exceeds 10%, or the policy document version is updated; When any one of the conditions that the change rate of production and emissions does not match occurs, the main verification contract is triggered to start a multi-level verification process. The online monitoring data at the emission end includes multi-dimensional carbon data quality control comparison data such as flue CO2 monitoring, and drone or satellite image inversion. The calculated emissions at the fuel end are used as the basis for carbon emission rights trading settlement. Step S5: The main verification contract distributes verification tasks to the verification layer nodes according to the trigger event type in Step S4, and obtains verification results. Step S6: Each verification node submits the verification results in Step S5 to the verification consensus module, and uses an improved Byzantine fault tolerance mechanism to perform weighted fusion on the verification results. After reaching a consensus, update the blockchain data status and generate a final data verification report. The weight coefficients for the weighted fusion are dynamically adjusted according to the historical accuracy and real-time status of the nodes.

[0006] Furthermore, for the multi-node carbon emission intelligent contract verification method based on blockchain of the present invention, the hierarchical blockchain architecture includes a data layer, a verification layer, and a contract layer, where: Data layer: Use a permissioned chain to store verification metadata, and save the encrypted original carbon emission data through an off-chain storage system. Verification layer: Deploy a dynamic rule engine and a verification network including regulatory nodes, audit nodes, and industry nodes. Contract layer: Load the main verification contract and the Merkle Patricia Tree policy library for storing policy files.

[0007] Furthermore, for the multi-node carbon emission intelligent contract verification method based on blockchain of the present invention, in Step S5, the main verification contract distributes verification tasks to the verification layer nodes according to the trigger event type in Step S4, and obtains verification results, where: Dispatch a compliance verification algorithm task to the regulatory node to verify whether the carbon emission data complies with the preset regulations; Dispatch a Monte Carlo simulation verification task to the audit node to generate a simulated emission data set and calculate the confidence interval; Dispatch a benchmark comparison task for similar enterprises to the industry node to generate an industry benchmark analysis result by calling the benchmark database.

[0008] Furthermore, the blockchain-based multi-node carbon emission smart contract verification method of the present invention further includes: Step S7, when the policy library is updated, the governance contract is used to simulate and run the new policy in the sandbox environment. After verification, the Merkle root of the policy library is updated, and the updated policy is synchronized to the dynamic rule engine through the contract layer. At the same time, a double-chain graph recording the version change relationship is constructed.

[0009] Furthermore, for the blockchain-based multi-node carbon emission smart contract verification method of the present invention, the off-chain storage system adopts horizontal sharding and vertical sharding strategies: Horizontal sharding distributes data in the IPFS node cluster according to industry types; Vertical sharding separates basic monitoring data and derivative verification indicators into different encrypted databases.

[0010] Furthermore, for the blockchain-based multi-node carbon emission smart contract verification method of the present invention, the dynamic rule engine includes: A rule parser that compiles a policy file in JSON or XML format into executable verification logic; A differential hash comparator that identifies version changes by comparing the hash values of policy files; An event correlation analyzer that establishes a dynamic mapping relationship among production data, fuel data, and online monitoring data of carbon emissions.

[0011] Furthermore, for the blockchain-based multi-node carbon emission smart contract verification method of the present invention, the distribution and verification task in step S5 includes phased processing: In the primary verification stage, a pre-compiled contract template is called to verify the data threshold with millisecond-level response. In the intermediate verification stage, a secure multi-party computing protocol is activated to achieve cross-verification of encrypted data among multiple nodes; In the in-depth verification stage, an LSTM prediction model of a long short-term memory network is loaded. The LSTM prediction model generates an emission trend prediction curve through time window sliding training. Its input features include the enterprise's historical fuel consumption, production volume, and policy adjustment factors. When the relative deviation between the actual emission data and the prediction curve exceeds the threshold of ±5%, an artificial review instruction is generated.

[0012] Furthermore, for the blockchain-based multi-node carbon emission smart contract verification method of the present invention, the intermediate verification stage includes: The regulatory node runs a compliance verification algorithm and outputs a regulatory compliance index; The auditing node assigns Monte Carlo simulation verification tasks and generates a simulated emission dataset through the following steps: According to the historical operation data of the equipment, the operating condition intervals are divided by ±5% of the load rate; within the same operating condition interval, the fuel consumption parameters of consecutive time windows are randomly selected; Latin hypercube sampling is performed on the carbon content parameters and equipment operating status parameters to generate a simulated emission dataset of the order of 10^5. The sampling of the fuel consumption parameters, carbon content parameters, and equipment operating status parameters needs to satisfy the operating condition interval division and parameter correlation constraints. The confidence interval of the emission data is calculated (confidence level ≥ 95%). The industry node calls the benchmark database to generate the industry benchmark analysis result. The main verification contract performs weighted fusion on the output results of the three types of nodes, and the weight coefficients are dynamically adjusted according to the historical accuracy and real-time status of the nodes.

[0013] Furthermore, the blockchain-based multi-node carbon emission intelligent contract verification method of the present invention further includes a policy update process: Parse the policy text through a natural language processing engine to generate verification rules in the form of an abstract syntax tree; Apply loop unrolling technology to optimize the contract code structure for reducing Gas consumption; Write the Merkle root of the effective policy into the main chain block header, and store the historical version and change impact analysis data in the side chain.

[0014] Furthermore, the blockchain-based multi-node carbon emission intelligent contract verification method of the present invention further includes: implementing attribute-based encryption on the standardized data packet and decrypting it when the access policy meets the preset standard; Adopt a digital signature algorithm resistant to quantum attacks, including CRYSTALS-Dilithium or ECDSA; Execute the threshold signature protocol every 24 hours to rotate the verification node keys stored distributively.

[0015] The beneficial effects of the present invention: By constructing a hierarchical blockchain architecture and deploying a dynamic rule engine, the present invention can monitor carbon emission data in real time and quickly trigger a multi-level verification process when anomalies or policy changes are detected. This improves the real-time update and accurate verification of carbon emission data, and helps the fair and efficient operation of the carbon emission rights trading market. By introducing a hierarchical sampling method with operating condition interval constraints, Monte Carlo simulation can effectively distinguish the fuel consumption characteristics under different production loads, making the generated simulated emission dataset closer to the real production scenario. Compared with the traditional random sampling method, the present invention reduces the confidence interval error rate of carbon emission calculation, significantly improving the credibility of the transaction settlement data.

[0016] The present invention introduces a verification network including regulatory nodes, audit nodes, and industry nodes, and realizes a consensus mechanism involving multiple parties. This mechanism enhances the trust among enterprises during collaboration, reduces the trust cost, and helps promote the healthy development of the carbon emission rights trading market.

[0017] The dynamic rule engine can flexibly respond to policy changes and distribute different verification tasks according to the types of trigger events. This enables the present invention to adapt to complex policy environments and meet the requirements of multi-dimensional verification, improving the flexibility and adaptability of carbon emission data management.

[0018] By adopting measures such as using edge computing nodes to collect data, a off-chain storage system to save raw data, and implementing security enhancement steps, the present invention improves the efficiency of data processing and enhances the security of data. This helps protect the confidentiality and integrity of carbon emission data and prevent data leakage and tampering.

[0019] In summary, these beneficial effects of the present invention provide strong support for the healthy development of the carbon emission rights trading market. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the drawings.

[0021] Figure 1 It is a timing diagram of the process of the multi-node carbon emission smart contract verification method based on blockchain provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments and corresponding drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention. The technical solutions provided by each embodiment of the present invention will be described in detail below with reference to the drawings.

[0023] To better understand the objectives of the present invention, the present invention will be further described in detail below.

[0024] As Figure 1 shown, the multi-node carbon emission smart contract verification method based on blockchain provided by the present invention includes: Step S1: Construct a hierarchical blockchain architecture, and the hierarchical blockchain architecture includes: Data layer, which uses a permissioned chain to store and verify metadata, and saves encrypted original carbon emission data through an off-chain storage system; Verification layer, which deploys a dynamic rule engine and a verification network including regulatory nodes, audit nodes, and industry nodes; Contract layer, which loads the main verification contract and the Merkle Patricia Tree policy library for storing policy files; In step S2, fuel consumption, laboratory data, and online carbon emission monitoring data of the enterprise side are collected through edge computing nodes, and the data is sequentially subjected to outlier filtering, format standardization, and time series alignment processing to generate a standardized data packet carrying a hash chain data fingerprint generated by a secure hash algorithm, and the secure hash algorithm includes SHA3-256 or SHA-256; In step S3, the standardized data packet generated in step S2 is input into the oracle node cluster. By implementing TLS mutual authentication and zero-knowledge proof verification, a cross-chain trusted channel with a third-party data source is established, associated data verified through reputation scoring is obtained, and the associated data is compared with the standardized data packet for consistency; In step S4, when any of the following conditions is detected, the main verification contract is triggered to start a multi-level verification process: the absolute value of the CO2 concentration mutation exceeds 15% in a non-start / stop state; the absolute value of the relative difference between the fuel-side calculated emissions and the online monitoring data at the emission side exceeds 10%; the policy file version is updated; the production volume and the emission change rate do not match. Trigger the main verification contract to start a multi-level verification process; In step S5, according to the trigger event type in step S4, the main verification contract distributes verification tasks to the verification layer nodes: Dispatch a compliance verification algorithm task to the regulatory node to verify whether the carbon emission data complies with the preset regulations; Dispatch a Monte Carlo simulation verification task to the audit node to generate a simulated emission data set and calculate the confidence interval; Dispatch a benchmark comparison task for peer enterprises to the industry node to generate an industry benchmark analysis result by calling the benchmark database; In step S6, each verification node submits the verification results in step S5 to the verification consensus module, and uses an improved Byzantine fault tolerance mechanism (PBFT improved algorithm, adjusting the fault tolerance threshold through dynamic weight) to perform weighted fusion on the verification results. After reaching a consensus, the blockchain data status is updated and a final data verification report is generated, and the weight coefficient of the weighted fusion is dynamically adjusted according to the historical accuracy and real-time status of the nodes; In step S7, when the policy library is updated, the new policy is simulated and run in a sandbox environment through the governance contract. After verification, the Merkle root of the policy library is updated, and the updated policy is synchronized to the dynamic rule engine through the contract layer, and at the same time, a double-chain graph recording the version change relationship is constructed.

[0025] Construct a hierarchical blockchain architecture: Data layer: Use a permissioned blockchain to store and verify metadata, enhancing data security and traceability. Meanwhile, save the encrypted original carbon emission data through an off-chain storage system, protecting data privacy while improving data access efficiency.

[0026] Verification layer: Deploy a dynamic rules engine that can parse and process carbon emission data in real time, improving data accuracy and compliance. Meanwhile, a verification network including regulatory nodes, audit nodes, and industry nodes forms a multi-party participation verification mechanism, enhancing the credibility of verification.

[0027] Contract layer: Load the main verification contract, which is the core of the verification process and is responsible for triggering and executing verification tasks. Meanwhile, store the Merkle Patricia Tree policy library of the policy file.

[0028] Data collection and processing: Collect enterprise-side fuel consumption, laboratory data, and online carbon emission monitoring data through edge computing nodes, and use the low latency and high efficiency characteristics of edge computing to achieve real-time data collection.

[0029] Filter out outliers, standardize the format, and align the time series for the collected data.

[0030] Generate a standardized data packet carrying the hash chain data fingerprint generated by a secure hash algorithm, where the secure hash algorithm includes SHA3-256 or SHA-256, providing a basis for the secure transmission and verification of data.

[0031] Data verification: Input the standardized data packet into the oracle node cluster, and establish a cross-chain trusted channel with a third-party data source through TLS mutual authentication and zero-knowledge proof verification.

[0032] The dynamic rules engine monitors the verified data in real time. When a specific event (such as a sudden change in CO2 concentration, an update of the policy file version, or a mismatch between the production volume and the emission change rate) is detected, it triggers the main verification contract to start a multi-level verification process.

[0033] The main verification contract distributes verification tasks to the verification layer nodes according to the type of trigger event, including compliance verification, Monte Carlo simulation verification, and benchmark comparison with peer enterprises, etc.

[0034] Consensus and report generation: Each verification node submits the verification result to the verification consensus module, and uses an improved Byzantine fault tolerance mechanism to reach a consensus.

[0035] Update the blockchain data status and generate a final data verification report to provide a scientific basis for carbon emission management.

[0036] Policy Update and Security Enhancement: When the policy library is updated, the new policy is simulated and run in a sandbox environment through a governance contract. After verification, the Merkle root of the policy library is updated.

[0037] Implement security enhancement steps, such as attribute-based encryption, digital signatures resistant to quantum attacks, and regular rotation of verification node keys, etc., to enhance the security and anti-attack capabilities of the data.

[0038] Advantages of the Method: Improve verification efficiency and accuracy: Through real-time monitoring and multi-level verification processes.

[0039] Reduce trust costs: Utilize blockchain technology to establish a multi-party participation verification mechanism and enhance the trust level during enterprise collaboration.

[0040] Adapt to policy changes and multi-dimensional verification requirements: The dynamic rule engine and policy update process can flexibly respond to policy changes and multi-dimensional verification requirements.

[0041] Enhance data security and anti-attack capabilities: Through security enhancement steps, enhance the security and anti-attack capabilities of the data, and protect the confidentiality and integrity of carbon emission data.

[0042] Specifically, in the multi-node carbon emission smart contract verification method based on blockchain of the present invention, in step S1, the off-chain storage system adopts horizontal sharding and vertical sharding strategies: Horizontal sharding distributes data in a distributed manner among the IPFS node cluster according to industry types; Vertical sharding separates basic monitoring data and derived verification metrics into different encrypted databases.

[0043] The off-chain storage system adopts horizontal sharding and vertical sharding strategies to optimize data storage and management. The specific implementation methods and advantages of these two sharding strategies are as follows: Horizontal Sharding Strategy: Implementation method: The horizontal sharding strategy mainly distributes data in a distributed manner among the IPFS (InterPlanetary File System) node cluster according to industry types. This means that data from different industries will be allocated to different IPFS nodes for storage.

[0044] Advantages: This sharding method helps to achieve classified storage and efficient access of data. By dispersing data storage across different IPFS nodes, data reliability and availability can be improved. Even if a certain node fails, the data on other nodes remains intact and available, thus reducing the impact of a single node failure on the entire system.

[0045] Vertical sharding strategy: Implementation method: The vertical sharding strategy separates the basic monitoring data from the derived verification metrics and stores them in different encrypted databases. The basic monitoring data is the original, unprocessed data, while the derived verification metrics are data calculated based on the basic data through certain algorithms and used for verification and validation.

[0046] Advantages: By storing these two types of data separately, the integrity and security of the data can be better protected. The basic monitoring data remains in its original state, facilitating subsequent processing and verification; while the derived verification metrics are used for dedicated verification and validation processes, improving the efficiency and accuracy of data processing. In addition, storing data in an encrypted database can enhance data security and prevent unauthorized access and tampering of the data.

[0047] In summary, the horizontal and vertical sharding strategies of the off-chain storage system together constitute an important data storage and management mechanism in this smart contract verification method, providing strong guarantees for data reliability, availability, and security.

[0048] Specifically, in the multi-node carbon emission smart contract verification method based on blockchain described in the present invention, the dynamic rule engine in step S4 includes: A rule parser that compiles a JSON / XML format policy file into executable verification logic; The rule parser uses the Antlr4 framework to parse the JSON / XML format policy file, generates an abstract syntax tree (AST), and generates executable verification logic through LLVM intermediate code; the differential hash comparator calculates the hash value of the policy file based on the SHA-256 algorithm and compares it with the historical hash value stored in the blockchain. If they are inconsistent, the version update process is triggered.

[0049] A differential hash comparator that identifies version changes by comparing the hash values of policy files; An event correlation analyzer that establishes a dynamic mapping relationship among production data, fuel data, and online carbon emission monitoring data.

[0050] The dynamic rule engine is a core component that is responsible for real-time monitoring and processing of multi-source carbon emission data and triggering corresponding verification processes according to preset rules and strategies. The dynamic rule engine includes three key components: a rule parser, a differential hash comparator, and an event correlation analyzer. Their respective functions and working methods are as follows: Rule Parser: Function: Compile policy files in formats such as JSON / XML into executable verification logic.

[0051] Working Method: The policy file includes various verification rules, threshold settings, triggering conditions, etc. The rule parser parses these policy files and transforms them into a form that the dynamic rule engine can understand and execute, thereby enhancing the verification of data in accordance with preset rules.

[0052] Differential Hash Comparator: Function: Identify version changes by comparing the hash values of policy files.

[0053] Working Method: The differential hash comparator calculates the hash value of the current policy file and compares it with the previously stored hash value. If the two are inconsistent, it indicates that the policy file has changed. At this time, the differential hash comparator will trigger an update process to make the verification logic in the dynamic rule engine consistent with the latest policy file.

[0054] Event Correlation Analyzer: Function: Establish a dynamic mapping relationship among production data, fuel data, and online carbon emission monitoring data.

[0055] Working Method: The event correlation analyzer can monitor carbon emission data and related production data in real time. Through data analysis techniques, it finds the correlation and change trends among these data. When specific triggering events (such as sudden changes in CO2 concentration, large differences between fuel-side and online carbon emission monitoring-side data, policy file version updates, or mismatches between production volume and emission change rates, etc.) are detected, the event correlation analyzer will match these events with preset verification rules and trigger corresponding verification processes, thereby enhancing the accuracy and compliance of carbon emission data.

[0056] These three components work together, enabling the dynamic rule engine to flexibly respond to complex policy changes and multi-dimensional verification requirements, providing strong support for the verification and certification of carbon emission data.

[0057] Specifically, for the blockchain-based multi-node carbon emission smart contract verification method described in the present invention, the distribution and verification task in step S5 includes phased processing: In the primary verification stage, the pre-compiled contract template is called to verify the data threshold with millisecond-level response; e2. In the intermediate verification stage, the secure multi-party computing protocol is activated to implement cross-verification of encrypted data among multiple nodes; In the intermediate verification stage, the secure multi-party computing protocol is activated to implement cross-verification of encrypted data among multiple nodes; In the in-depth verification stage, the LSTM prediction model is loaded. Using the enterprise's historical fuel consumption, production volume, and policy adjustment factors as input features, an emission trend prediction curve is generated through time window sliding training. When the relative deviation between the actual emission data and the prediction curve exceeds the threshold (±5%), an artificial review instruction is generated.

[0058] The distribution verification task in step S5 is carefully designed to be processed in stages to improve the accuracy and compliance of carbon emission data. These three stages are the primary verification stage, the intermediate verification stage, and the in-depth verification stage. They each undertake different verification responsibilities and work together to complete the distribution verification task.

[0059] Primary verification stage: In this stage, the system calls the pre-compiled contract template to verify the data threshold with millisecond-level response. This means that the system can quickly conduct a preliminary screening of the data. By checking whether the data exceeds the preset threshold range, it can quickly identify and exclude data that is significantly abnormal or does not conform to basic rules. The purpose of this step is to lighten the burden on subsequent verification stages and improve the overall verification efficiency.

[0060] Intermediate verification stage: The intermediate verification stage is more in-depth and complex. In this stage, the system activates the secure multi-party computing protocol to implement cross-verification of encrypted data among multiple nodes. This means that the regulatory node, the audit node, and the industry node will jointly participate in the verification process, each performing different verification tasks.

[0061] The regulatory node is responsible for running the compliance verification algorithm and outputting the regulatory compliance index; the audit node performs Monte Carlo simulation to calculate the confidence interval of the emission data and evaluate the reliability and accuracy of the data; the industry node calls the benchmark database to generate industry benchmark analysis results and compare the data with the industry average level.

[0062] The main verification contract performs weighted fusion on the output results of these three types of nodes. The weight coefficients are dynamically adjusted according to the historical accuracy and real-time status of the nodes to improve the comprehensiveness and accuracy of the verification results.

[0063] The calculation method of the weight coefficient is: W =α⋅Ai+β⋅Si; Among them, Ai is the historical accuracy rate of the node (based on the correct rate of verification results in the past 30 days), Si is the real-time status score (including indicators such as node online rate and response delay), and α and β are adjustment factors (the default values are 0.7 and 0.3 respectively).

[0064] Deep verification stage: The deep verification stage is a further verification and refinement of the intermediate verification results. In this stage, the system loads the LSTM prediction model for emission trend analysis to predict future emission trends and compare them with actual data.

[0065] The LSTM prediction model is constructed using the PyTorch framework. The input features include the historical fuel consumption (normalized) of the enterprise in the past 24 months, the production volume (smoothed quarterly), and the policy adjustment factor (represented by one-hot encoding). When training the model, the time window sliding method is used (the window size is 12 months, and the sliding step is 1 month). The loss function is the mean square error (MSE), and the optimizer is selected as Adam.

[0066] If there is a large deviation between the emission trend and the results of the prediction model, and the deviation exceeds the preset threshold, the system will generate an instruction for manual review. This means that for data that may have problems, the system will require manual final review and confirmation to improve the accuracy and compliance of the data.

[0067] Through the gradual deepening and refinement of these three stages, the distributed verification task can improve the accuracy and compliance of carbon emission data, providing strong support for the realization of the carbon neutrality goal.

[0068] Specifically, in the multi-node carbon emission intelligent contract verification method based on blockchain described in the present invention, the intermediate verification stage includes: The regulatory node runs a compliance verification algorithm and outputs a regulatory compliance index; The audit node assigns Monte Carlo simulation verification tasks. By randomly sampling fuel consumption parameters, carbon content parameters, and equipment operation status parameters, a simulation emission dataset of the order of 10⁵ is generated. The sampling of fuel consumption parameters, carbon content parameters, and equipment operation status parameters needs to meet the constraints of working condition interval division and parameter correlation. Calculate the confidence interval of the emission data (confidence level ≥ 95%). The industry node calls the benchmark database to generate an industry benchmark analysis result. The main verification contract performs weighted fusion on the output results of the three types of nodes, and the weight coefficients are dynamically adjusted according to the node historical accuracy rate and real-time status.

[0069] When performing Monte Carlo simulation verification, the audit node adopts a stratified random sampling strategy to improve the simulation accuracy. Specifically, it includes: Operating condition range division: Based on the historical operation data of enterprise equipment, calculate the distribution range of the load factor (current power / rated power), and divide the operating condition range at intervals of ±5%. For example, when the rated power of the equipment is 1000 kW, the actual power range corresponding to the load factor range of 85% - 90% is 850 - 900 kW.

[0070] Fuel consumption sampling: In the same operating condition range, extract the fuel consumption data set for a continuous time window (such as 1 hour), and select representative samples through the Latin Hypercube Sampling method to avoid sampling bias caused by operating condition fluctuations.

[0071] Parameter correlation constraint: Jointly sample the carbon content parameter and the equipment operating state parameters (such as combustion efficiency, flue gas temperature) to ensure that the physical correlation between the parameters meets the requirements of the actual production process.

[0072] In the intermediate verification stage, a variety of technical means are used to deeply verify the carbon emission data. The specific details of this stage are as follows: Regulatory node task: The regulatory node is mainly responsible for running the compliance verification algorithm. This algorithm will carefully check the carbon emission data to make it meet the relevant regulations and policy requirements.

[0073] After the algorithm execution is completed, the regulatory node will output a regulatory compliance index, which is an important reference indicator for evaluating data compliance.

[0074] Audit node task: The audit node performs Monte Carlo simulation, which is a statistical simulation method used to calculate the confidence interval of emission data.

[0075] By simulating a large number of possible situations, the audit node can evaluate the reliability and accuracy of the data, thereby enhancing the authenticity of the data.

[0076] Industry node task: The industry node will call the benchmark database to generate the industry benchmark analysis results.

[0077] By comparing the emission data of the current enterprise with the data of other enterprises in the same industry, the industry node can evaluate whether the emission level of the current enterprise is within a reasonable range.

[0078] In the intermediate verification stage, these three nodes each undertake different verification tasks, jointly constituting a comprehensive and in-depth verification system. In order to achieve the accuracy and fairness of the verification results, the main verification contract will perform weighted fusion processing on the output results of these three types of nodes.

[0079] Weighted fusion process of the main verification contract: The main verification contract first receives the output results of the regulatory nodes, audit nodes, and industry nodes.

[0080] Next, it dynamically adjusts the weight coefficients based on the historical accuracy rate and real-time status of the nodes. This means that the weights are updated in real time according to the actual performance of each node to achieve the fairness and accuracy of the verification results.

[0081] Finally, the main verification contract will perform a weighted sum of the results of each node according to their weight coefficients to obtain the final verification result. This process not only improves the accuracy and reliability of the verification but also enhances the flexibility and adaptability of the system.

[0082] Through the intermediate verification stage and the weighted fusion process of the main verification contract, this method can comprehensively and deeply verify carbon emission data, providing strong support for the healthy development of the carbon emission rights trading market.

[0083] Specifically, for the blockchain-based multi-node carbon emission intelligent contract verification method described in the present invention, the policy update process in step S7 includes: Parse the policy text through a natural language processing engine to generate verification rules in the form of an abstract syntax tree; Use the BERT pre-trained model to perform semantic parsing on the policy text, extract key constraint conditions (such as emission limits, monitoring frequencies), and generate verification rules in the form of an abstract syntax tree (AST) through rule template mapping. The template mapping table is stored in the policy library.

[0084] Apply loop unrolling technology to optimize the contract code structure to reduce Gas consumption; Write the Merkle root of the effective policy into the main chain block header, and store the historical versions and change impact analysis data in the side chain.

[0085] Step S7 specifically includes three steps: Policy rule generation: In this step, the system uses a natural language processing engine to parse the policy text. This technology can deeply understand the policy content and convert it into verification rules that can be understood by the computer.

[0086] The parsed verification rules are presented in the form of an abstract syntax tree, providing a clear and structured basis for subsequent policy execution.

[0087] Contract code optimization: To improve the execution efficiency of the intelligent contract and reduce the execution cost, the system applies loop unrolling technology to optimize the contract code structure.

[0088] By reducing unnecessary loops and repeated calculations, this step significantly reduces Gas consumption, making the execution of smart contracts on the blockchain more economical and efficient.

[0089] Policy Deployment and Version Management: After the policy update is completed, the system writes the Merkle root of the effective policy into the main chain block header.

[0090] At the same time, to support version management and change impact analysis, the system stores historical versions and change impact analysis data in the side chain. This not only retains the historical records of the policy but also facilitates subsequent version backtracking and change analysis.

[0091] Through the collaborative work of these three steps, the policy update process realizes the dynamic optimization and continuous improvement of the multi-node carbon emission smart contract verification method based on the blockchain. This not only improves the flexibility and adaptability of the system.

[0092] Specifically, for the multi-node carbon emission smart contract verification method based on the blockchain described in the present invention, the method further includes security enhancement steps: performing attribute-based encryption on the standardized data packet, and decrypting when the access policy satisfies (regulatory level ≥ 2) ∧ (jurisdiction matching); Adopting a digital signature algorithm resistant to quantum attacks, including CRYSTALS-Dilithium or ECDSA; Executing the threshold signature protocol every 24 hours to rotate the verification node keys stored distributively.

[0093] Performing attribute-based encryption (h1) on the standardized data packet: In this step, the system performs additional security protection on the processed and generated standardized data packet. Specifically, the attribute-based encryption technology is adopted, which means that the access rights of the data are set based on specific attributes or conditions.

[0094] In this example, only when the access policy satisfies (regulatory level ≥ 2) and (jurisdiction matching), in the attribute-based encryption policy, "regulatory level ≥ 2" is defined as passing the regulatory qualification of level 2 or above; "jurisdiction matching" is defined as the administrative region to which the IP address of the data access request belongs being the same as the registered place of the data source enterprise, will the data packet be decrypted and allowed to be accessed. This encryption method not only improves the security of the data but also realizes fine-grained access control, enabling only users who meet specific conditions to access the data.

[0095] Adopting a digital signature algorithm resistant to quantum attacks, including CRYSTALS-Dilithium or ECDSA (h2): To address potential future threats from quantum computers, the system introduces the CRYSTALS-Dilithium algorithm to generate digital signatures. This algorithm is specifically designed to resist attacks from quantum computers, ensuring that data signatures remain valid and secure even after the maturity of quantum computing technology.

[0096] Execute the threshold signature protocol every 24 hours to rotate the verification node keys (h3) stored distributively: To enhance the overall security and attack resistance of the system, the system implements a mechanism for regular key rotation. Specifically, the threshold signature protocol is executed every 24 hours to rotate the keys of the verification nodes in the distributed storage.

[0097] The threshold signature protocol is a distributed key generation and signature scheme that allows multiple nodes to jointly generate and verify keys, thereby enhancing the robustness and security of the system. By regularly rotating keys, the system can further reduce the risk of key leakage and ensure the continuous secure operation of the system.

[0098] These security enhancement steps together constitute the security guarantee system in the multi-node carbon emission smart contract verification method based on blockchain, aiming to enhance the confidentiality, integrity, and availability of data, while improving the overall security and attack resistance of the system.

[0099] Explanation of technical terms in this invention: Blockchain: A distributed ledger technology that enables multi-party trusted collaboration through a decentralized and immutable data storage method. In this invention, it is used for the distributed storage and verification of carbon emission data.

[0100] Hierarchical blockchain architecture: Includes a data layer, a verification layer, and a contract layer: Data layer: Uses a permissioned blockchain to store verification metadata and saves encrypted original carbon emission data through an off-chain storage system.

[0101] Verification layer: Deploys a dynamic rule engine and a verification network consisting of regulatory nodes, audit nodes, and industry nodes, responsible for multi-party collaborative data verification.

[0102] Contract layer: Loads the main verification contract and the Merkle Patricia Tree policy library for storing policy files, used to execute smart contract logic.

[0103] Permissioned Blockchain: A blockchain that only allows authorized nodes to participate.

[0104] Smart contract: An automatically executed programmed protocol used to trigger and execute verification tasks, such as the main verification contract and the governance contract.

[0105] Edge computing node: A computing device deployed at the enterprise side, used to collect real-time fuel consumption, laboratory data, and online monitoring data of carbon emissions, reducing data transmission latency.

[0106] Secure Hash Algorithm (SHA3-256 / SHA-256): An encryption algorithm that generates a hash fingerprint of data.

[0107] Standardized data packet: A data packet that has undergone outlier filtering, format standardization, and time series alignment, carrying a hash chain data fingerprint.

[0108] Off-chain storage system: Combining horizontal sharding (distributed storage by industry type in a cluster of IPFS nodes) and vertical sharding (separating basic data and derived metrics into different encrypted databases) to optimize data storage efficiency and security.

[0109] Oracle node cluster: A middleware connecting the blockchain and external data sources, establishing a cross-chain trusted channel through TLS mutual authentication and zero-knowledge proof verification to obtain third-party associated data.

[0110] Dynamic rule engine: Includes three core components: Rule parser: Compiles JSON / XML format policy files into executable verification logic.

[0111] Difference hash comparator: Identifies policy version changes by comparing SHA-256 hash values.

[0112] Event correlation analyzer: Establishes a dynamic mapping relationship between production data, fuel data, and carbon emission data, triggering a multi-level verification process.

[0113] Primary verification stage: Invokes a pre-compiled contract template for threshold verification (millisecond-level response).

[0114] Intermediate verification stage: Activates the Secure Multi-Party Computation protocol (SMPC) to achieve cross-verification of encrypted data among multiple nodes.

[0115] Deep verification stage: Loads the LSTM prediction model, generates an emission trend prediction curve through training with historical data, and triggers manual review when the deviation exceeds the threshold (±5%).

[0116] Improved Practical Byzantine Fault Tolerance mechanism (PBFT improved algorithm): By dynamically adjusting the node weight coefficient (based on historical accuracy and real-time status), it improves the fault tolerance threshold and achieves weighted consensus.

[0117] Monte Carlo simulation: Based on the historical operation data of the device, generates a simulation emission data set of the order of 10^5 through Latin hypercube sampling, and calculates the confidence interval (confidence level ≥ 95%).

[0118] LSTM Prediction Model: Long Short-Term Memory network. The input features include historical fuel consumption, production volume, and policy adjustment factors. Through time window sliding training, it predicts the emission trend.

[0119] Regulatory Compliance Index: The compliance score output by the regulatory node, reflecting the degree to which the data complies with the preset regulations.

[0120] Industry Benchmark Database: Stores the carbon emission data of enterprises in the same industry, used to generate industry benchmark analysis results.

[0121] Attribute-Based Encryption (ABE): An encryption method based on access policies (such as regulatory level ≥ 2 and jurisdiction matching), achieving fine-grained data access control.

[0122] Quantum-Resistant Digital Signature Algorithm: CRYSTALS-Dilithium: A post-quantum cryptography algorithm that resists quantum computer attacks.

[0123] ECDSA: Elliptic Curve Digital Signature Algorithm, used for conventional data signing.

[0124] Threshold Signature Protocol: Rotates the verification node key every 24 hours to prevent the risk of key leakage.

[0125] Merkle Patricia Tree Policy Library: A tree structure for storing policy files, supporting efficient hash verification.

[0126] Governance Contract: Simulates the operation of new policies in a sandbox environment, and updates the Merkle root of the policy library after successful verification.

[0127] Double-Chain Graph: A chain structure for recording the relationship of policy version changes. The main chain stores the effective policies, and the side chain stores historical versions and change impact analysis data.

[0128] TLS Mutual Authentication: A transport layer security protocol that ensures mutual authentication between the oracle node and the third-party data source.

[0129] IPFS (InterPlanetary File System): A distributed storage protocol for horizontal sharding storage of off-chain data.

[0130] Latin Hypercube Sampling: A stratified random sampling method that improves the accuracy of Monte Carlo simulation.

[0131] The present invention proposes a multi-node carbon emission smart contract verification method based on blockchain, effectively solving the limitations of smart contracts in carbon emission data verification, enabling it to adapt to complex policy changes and multi-dimensional verification requirements.

[0132] First, the present invention constructs a hierarchical blockchain architecture, including a data layer, a verification layer, and a contract layer. This architecture enables distributed storage and verification of carbon emission data among multiple nodes, improving the reliability and security of the data. The data layer uses a permissioned chain to store verification metadata and saves the encrypted original carbon emission data through an off-chain storage system.

[0133] Second, the present invention uses edge computing nodes to collect carbon emission monitoring data from the enterprise side, and performs outlier filtering, format standardization, and time series alignment processing on these data to generate a standardized data packet carrying a hash chain data fingerprint generated by a secure hash algorithm, where the secure hash algorithm includes SHA3-256 or SHA-256. This process improves the accuracy and consistency of the data, providing a reliable basis for subsequent data verification.

[0134] In terms of data verification, the present invention introduces a dynamic rule engine and a multi-level verification process. The dynamic rule engine can monitor carbon emission data in real time and trigger the main verification contract to start a multi-level verification process according to preset rules. These verification tasks include compliance verification, Monte Carlo simulation verification, and benchmark comparison with peer enterprises, etc., which can comprehensively cover multiple dimensions of carbon emission data.

[0135] In addition, the present invention also adopts an improved Byzantine fault tolerance mechanism to reach a consensus and update the blockchain data state. This mechanism can ensure that the verification results can still reach an agreement in the presence of malicious nodes, further improving the reliability and security of the system.

[0136] Finally, when the policy library is updated, the present invention simulates the operation of the new policy in a sandbox environment through a governance contract, updates the Merkle root of the policy library after verification, and synchronously constructs a double-chain graph recording the version change relationship.

[0137] In summary, the present invention effectively solves the deficiencies of smart contracts in carbon emission data verification by constructing a hierarchical blockchain architecture, using edge computing and oracle technologies to obtain and verify carbon emission data, deploying a dynamic rule engine and a multi-level verification process, and implementing security enhancement steps, enabling it to adapt to complex policy changes and multi-dimensional verification requirements.

Claims

1. A multi-node carbon emission smart contract verification method based on blockchain, characterized in that: include: Step S1, constructing a layered blockchain architecture, which includes a data layer, a verification layer, and a contract layer: Step S2, based on the data layer of the layered blockchain architecture, collect enterprise-side fuel consumption, laboratory data, and carbon emission online monitoring data through edge computing nodes, and pre-process the collected data to generate a standardized data packet carrying a hash chain data fingerprint generated by a secure hash algorithm; Step S3, input the standardized data packet generated in step S2 into the oracle node cluster, establish a cross-chain trusted channel with the third-party data source by implementing TLS two-way authentication and zero-knowledge proof verification, obtain the associated data verified by the reputation score, and compare the associated data with the standardized data packet for consistency; Step S4: when it is detected that the absolute value of the sudden change of CO2 concentration in the non-start-stop state exceeds 15%, the absolute value of the relative difference between the calculated emissions at the fuel end and the online monitoring data at the emission end exceeds 10%, the policy document version is updated; When the production volume and emission volume change rate do not match any of the conditions, the main verification contract is triggered to start the multi-level verification process. The online monitoring data at the emission end includes flue CO2 monitoring, multi-dimensional carbon data quality control comparison data inverted by drones or satellite images, and the fuel end calculates emissions as the basis for carbon emission rights trading settlement. Step S5: The main verification contract distributes verification tasks to the verification layer nodes according to the trigger event type of step S4 to obtain the verification results; In step S6, each verification node submits the verification result of step S5 to the verification consensus module, and uses the improved Byzantine fault tolerance mechanism to perform weighted fusion on the verification results. After reaching a consensus, the blockchain data status is updated and a final data verification report is generated. The weight coefficient of the weighted fusion is dynamically adjusted according to the historical accuracy and real-time status of the node.

2. The multi-node carbon emission smart contract verification method based on blockchain according to claim 1 is characterized in that: The layered blockchain architecture includes a data layer, a verification layer, and a contract layer, wherein: The data layer uses a permissioned chain to store verification metadata and saves encrypted raw carbon emission data through an off-chain storage system; Verification layer, deploying dynamic rule engines and verification networks including regulatory nodes, audit nodes, and industry nodes; Contract layer, loads the main verification contract and the Merkle Patricia Tree policy library that stores the policy file.

3. The multi-node carbon emission smart contract verification method based on blockchain according to claim 1 is characterized in that: Step S5: The main verification contract distributes verification tasks to the verification layer nodes according to the trigger event type of step S4 to obtain the verification results, where: Assign compliance verification algorithm tasks to regulatory nodes to verify whether carbon emission data complies with preset regulations; Assigning Monte Carlo simulation verification tasks to audit nodes to generate simulated emission data sets and calculate confidence intervals; Assign similar enterprise benchmark comparison tasks to industry nodes to call the benchmark database to generate industry benchmark analysis results.

4. The multi-node carbon emission smart contract verification method based on blockchain according to claim 1 is characterized in that: Also includes: Step S7: When the policy library is updated, the governance contract is used to simulate the operation of the new policy in the sandbox environment. After verification, the policy library Merkle root is updated, and the updated policy is synchronized to the dynamic rule engine through the contract layer. At the same time, a dual-chain graph that records the version change relationship is constructed.

5. The multi-node carbon emission smart contract verification method based on blockchain according to claim 2 is characterized in that: The off-chain storage system adopts horizontal sharding and vertical sharding strategies: Horizontal sharding distributes and stores data in IPFS node clusters by industry type; Vertical sharding separates basic monitoring data and derived verification indicators into different encrypted databases.

6. The multi-node carbon emission smart contract verification method based on blockchain according to claim 1 is characterized in that: The dynamic rule engine includes: Rule parser, compiles JSON or XML formatted policy files into executable validation logic; A differential hash comparer that identifies version changes by comparing policy file hash values; Event correlation analyzer establishes a dynamic mapping relationship between production data, fuel data and carbon emission online monitoring data.

7. The multi-node carbon emission smart contract verification method based on blockchain according to claim 1 is characterized in that: The distribution verification task of step S5 includes phased processing: In the primary verification stage, the precompiled contract template is called to perform millisecond-level response verification on the data threshold; In the intermediate verification stage, the secure multi-party computing protocol is activated to implement cross-verification of encrypted data among multiple nodes; During the deep verification stage, the LSTM prediction model of the long short-term memory network is loaded. The LSTM prediction model generates an emission trend prediction curve through time window sliding training. Its input features include the company's historical fuel consumption, production volume and policy adjustment factors. When the relative deviation between the actual emission data and the prediction curve exceeds the threshold of ±5%, a manual review instruction is generated.

8. The multi-node carbon emission smart contract verification method based on blockchain according to claim 7 is characterized in that: The intermediate verification stage includes: The regulatory node runs the compliance verification algorithm and outputs the regulatory compliance index; The audit node assigns the Monte Carlo simulation verification task and generates a simulated emission data set through the following steps: based on the historical operation data of the equipment, the operating condition interval is divided according to the load rate ±5%; within the same operating condition interval, the fuel consumption parameters of the continuous time window are randomly selected; Latin hypercube sampling is performed on the carbon content parameters and equipment operating status parameters to generate a simulated emission data set of the order of 10^5. The sampling of fuel consumption parameters, carbon content parameters and equipment operating status parameters must meet the operating condition interval division and parameter correlation constraints, and the confidence interval of the emission data is calculated. The industry node calls the benchmark database to generate industry benchmarking analysis results. The main verification contract performs weighted fusion on the output results of the three types of nodes, and the weight coefficient is dynamically adjusted according to the historical accuracy and real-time status of the node.

9. The multi-node carbon emission smart contract verification method based on blockchain according to claim 4 is characterized in that: It also includes the policy update process: Parse the policy text through the natural language processing engine and generate verification rules in the form of abstract syntax trees; Apply loop unrolling technology to optimize the contract code structure to reduce Gas consumption; The Merkle root of the effective policy is written into the main chain block header, and the historical version and change impact analysis data are stored in the side chain.

10. The multi-node carbon emission smart contract verification method based on blockchain according to claim 1 is characterized in that: It also includes: implementing attribute-based encryption on standardized data packets, and decrypting when the access policy meets the preset criteria; Use quantum-resistant digital signature algorithms, including CRYSTALS-Dilithium or ECDSA; The threshold signature protocol is executed every 24 hours to rotate the verification node keys in distributed storage.

Images