Green supply chain collaborative carbon accounting method and system

Abstract

The present application relates to the technical field of carbon management, and particularly relates to a green supply chain collaborative carbon accounting method and system. The method first constructs an environmental bill of materials structure; performs privacy-protected carbon data aggregation, each level of supplier locally trains a carbon accounting model using private data, uploads encrypted model gradient parameters to a central platform, the central platform updates a global carbon footprint prediction model and issues a notification; performs a bottom-up cascading transfer algorithm, uses the output carbon footprint of a child node as the input carbon footprint of a parent node to calculate step by step, and uses homomorphic encryption technology to complete addition operation in a ciphertext state; writes the carbon footprint hash value, accounting timestamp and digital signature submitted by each level of supplier into a blockchain, and marks abnormal data for rejection, so that the carbon footprint data is tamper-proof and traceable. The present application realizes trusted transmission and automatic calculation of supply chain carbon data by constructing an environmental BOM and a privacy computing framework.

Description

Technical Field

[0001] This invention relates to the field of carbon management technology, and in particular to a collaborative carbon accounting method and system for green supply chains. Background Technology

[0002] In the carbon footprint of a product's entire lifecycle, Scope III emissions (i.e., emissions from upstream and downstream supply chains) typically account for as much as 70%-90%. However, current supply chain carbon accounting faces significant technical hurdles:

[0003] First, there are data silos and difficulties in data penetration. The supply chain is complex (with first-tier, second-tier, and even N-tier suppliers), and upstream companies often regard energy consumption and process data as trade secrets and are unwilling to share detailed data. Downstream brand owners find it difficult to obtain accurate primary data and can only use rough estimates based on "spend-based" methods, which have huge errors and cannot reflect the true emission reduction effect.

[0004] Secondly, the BOM hierarchy is fragmented, and the transfer mechanism is lacking. Traditional bills of materials (BOMs) only include physical attributes (quantity, specifications) and lack a mechanism for carrying and transferring carbon footprints. The carbon footprint of raw materials cannot be automatically calculated down to semi-finished and finished products, resulting in a break in the accounting chain between suppliers at various levels. This requires repeated manual summarization, which is extremely inefficient and prone to errors.

[0005] Thirdly, there is a crisis of trust and the risk of falsification. In the absence of reliable technical means, carbon data reported by suppliers themselves is difficult to verify, posing a risk of "greenwashing" or data falsification. Brand owners lack effective technical means to verify the authenticity of data without disclosing trade secrets.

[0006] Fourth, collaborative computing is difficult. There is a lack of a unified platform to support secure multi-party collaboration, and existing point-to-point communication methods cannot support large-scale, high-frequency supply chain carbon data updates and accounting.

[0007] Therefore, there is an urgent need in this field for a green supply chain carbon accounting technology based on privacy computing and blockchain technology that can achieve data that is "usable but not visible", support automatic cascading transmission of carbon footprints, and have tamper-proof verification capabilities. Summary of the Invention

[0008] The present invention aims to at least partially solve one of the technical problems in the related art.

[0009] Therefore, the first objective of this invention is to propose a collaborative carbon accounting method for green supply chains, comprising the following steps: S1. Construct an environmental bill of materials structure, attach dynamically updated carbon factor objects to each material node of the traditional bill of materials, the carbon factor objects include a physical attribute layer and an environmental attribute layer, and establish carbon footprint dependency relationships between parent and child nodes. S2 aggregates carbon data with privacy protection through federated learning technology. Suppliers at each level train carbon accounting models locally using private data, upload encrypted model gradient parameters to the central platform, and the central platform aggregates, updates, and distributes the global carbon footprint prediction model. S3 executes a bottom-up cascading algorithm. Based on the carbon footprint data of suppliers at each level, it automatically uses the output carbon footprint of the child node as the input carbon footprint of the parent node for progressive convolution. During data transmission, homomorphic encryption technology is used to enable the platform to complete the addition operation in ciphertext. S4 writes the carbon footprint hash values, accounting timestamps, and digital signatures submitted by suppliers at all levels into the blockchain, and automatically executes logical verification rules through smart contracts to reject abnormal data, thereby achieving the immutability and reliable traceability of carbon footprint data.

[0010] In one embodiment of the present invention, S1 includes: The physical attribute layer includes material code, name, quantity, and unit; The environmental attribute layer includes unit carbon footprint value, data source type, accounting standard version, last update time, and digital signature; The carbon footprint dependency between parent and child nodes means that the output carbon footprint of the child node automatically becomes the input carbon footprint of the parent node.

[0011] In one embodiment of the present invention, S2 includes: Suppliers at each level use energy consumption and production data from their private databases to train lightweight carbon accounting models and obtain model gradient parameters. Suppliers at each level will encrypt the model gradient parameters using homomorphic encryption technology before uploading them to the central platform; The central platform aggregates gradients from all parties, updates the global carbon footprint prediction model, and then distributes the updated global model to each node.

[0012] In one embodiment of the present invention, S3 includes: The carbon footprint of product P at level k is calculated using the transfer formula, which is:

[0013] in, This represents the direct emissions during the assembly process of the k-th layer. Indicates the number of components i. This indicates the carbon footprint of components passed down from the previous level. This indicates the transport emission coefficient from this level to the next level.

[0014] In one embodiment of the present invention, step S3, which involves using homomorphic encryption technology during data transmission to enable the platform to perform addition operations in ciphertext state, includes: Using the homomorphic encryption function E, such that After decryption, the exact sum is obtained.

[0015] In one embodiment of the present invention, the method for marking abnormal data as rejected in step S4 is as follows: The smart contract automatically detects whether the energy consumption of a unit product is within the reasonable threshold of the industry and whether the upstream and downstream data are matched. If the verification fails, it will be automatically rejected and marked as abnormal.

[0016] In one embodiment of the present invention, it further includes: Generate carbon labels with blockchain hash values, allowing consumers or regulators to view a blockchain-certified carbon footprint path by scanning the product's QR code.

[0017] To achieve the above objectives, a second aspect of the present invention proposes a green supply chain collaborative carbon accounting system, comprising: The BOM carbon factor construction module is used to build an environmental bill of materials structure. It attaches dynamically updated carbon factor objects to each material node of the traditional bill of materials. The carbon factor objects contain physical attribute layers and environmental attribute layers, and establish carbon footprint dependencies between parent and child nodes. The federated learning privacy modeling module is used to aggregate carbon data with privacy protection through federated learning technology. Each level of supplier trains the carbon accounting model locally using private data and only uploads the encrypted model gradient parameters to the central platform. The central platform then aggregates and updates the global carbon footprint prediction model and distributes it. The encrypted hierarchical aggregation and accounting module is used to execute the bottom-up cascading algorithm. Based on the carbon footprint data of suppliers at each level, it automatically uses the output carbon footprint of the child node as the input carbon footprint of the parent node for hierarchical convolution and employs homomorphic encryption technology during data transmission to enable the platform to complete the addition operation in encrypted state. The blockchain trusted evidence storage and verification module is used to write the carbon footprint hash value, accounting timestamp and digital signature submitted by suppliers at all levels into the blockchain, and automatically executes the logical verification rules through smart contracts to reject abnormal data, so as to realize the immutability and trusted traceability of carbon footprint data.

[0018] In one embodiment of the present invention, the ciphertext hierarchical aggregation and accounting module includes: The carbon footprint of product P at level k is calculated using the transfer formula, which is:

[0019] in, This represents the direct emissions during the assembly process of the k-th layer. Indicates the number of components i. This indicates the carbon footprint of components passed down from the previous level. This indicates the transport emission coefficient from this level to the next level.

[0020] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0021] The methods, systems, and storage media of this invention solve the two major problems of "data silos" and "lack of trust" in supply chain carbon accounting through technological innovation, and realize the automated, reliable, and private transmission of carbon footprints, providing a solid technical foundation for building a green supply chain.

[0022] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a green supply chain collaborative carbon accounting method according to an embodiment of the present invention; Figure 2 This is a structural diagram of a green supply chain collaborative carbon accounting system according to an embodiment of the present invention. Detailed Implementation

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] The following description, with reference to the accompanying drawings, describes a green supply chain collaborative carbon accounting method and system proposed according to an embodiment of the present invention.

[0027] Example 1 like Figure 1As shown, this invention proposes a collaborative carbon accounting method for green supply chains, which includes four core steps: e-BOM structure construction, federated learning data aggregation, cascaded transfer algorithm execution, and blockchain notarization and verification.

[0028] Step 1: Building the e-BOM (Environmental Bill of Materials) structure.

[0029] Extend the traditional BOM to an e-BOM, and attach a dynamically updated "carbon factor object" to each material node: Physical attribute layer: Retain the original material code, name, quantity, and unit.

[0030] Environmental attribute layer: New fields include "unit carbon footprint value ( "Data Source Type (Measured / Estimated)", "Accounting Standard Version", "Last Update Time", "Digital Signature".

[0031] Hierarchical association: Establish carbon footprint dependency relationships between parent and child nodes, and the output carbon footprint of the child node automatically becomes the input carbon footprint of the parent node.

[0032] This step, by constructing the aforementioned environmental bill of materials structure, achieves deep integration and structured association between carbon footprint data and product material information, providing a unified framework and transmission path for carbon data across the entire supply chain. This effectively solves the technical problems of missing carbon footprint information and broken data transmission mechanisms in traditional bills of materials, and significantly improves the automation level and data consistency of carbon accounting.

[0033] Step 2: Federated learning data aggregation (privacy protection).

[0034] Federated learning technology can be used to resolve the conflict between data sharing and privacy protection. Specific steps include: Local training: Suppliers at each level deploy lightweight carbon accounting models locally and train model parameters using local private data (energy consumption, output), with the original data remaining within the domain.

[0035] Encrypted gradient upload: Suppliers only upload encrypted model gradient parameters (representing carbon emission patterns rather than specific values) to the central platform.

[0036] Global aggregation: The central platform aggregates gradients from all parties, updates the global carbon footprint prediction model, and then distributes it to each node.

[0037] This technical step effectively solves the problem of obtaining raw data in supply chain carbon accounting caused by data silos and the need to protect trade secrets by constructing a collaborative computing mechanism that makes data "usable but not visible". While ensuring the data sovereignty and privacy security of all participants, it significantly improves the accuracy and reliability of global carbon footprint prediction, and provides a feasible technical path for collaborative carbon accounting in large-scale, multi-level supply chains.

[0038] Step 3: Execute the cascading propagation algorithm.

[0039] This method employs a bottom-up cascading transfer algorithm, aiming to automatically and progressively accumulate and transfer carbon footprint values ​​across the material hierarchy based on carbon footprint data generated at each level of the supply chain. Specifically, it utilizes the parent-child node dependencies established in a pre-built environmental bill of materials, using the carbon footprint data output by downstream child nodes (such as component suppliers) as input parameters for their corresponding parent nodes (such as assembly manufacturers) when calculating their own product carbon footprints. This forms a continuous and traceable carbon footprint transfer chain from raw materials to the final product. By defining carbon footprint transfer rules between levels, the algorithm automatically integrates the contributions of all upstream child nodes in the carbon footprint calculation of each node, eliminating the need for manual intervention or data aggregation. To ensure that the original carbon footprint details of each supplier are not leaked during cross-enterprise and cross-level data flow, homomorphic encryption technology is introduced in the data transmission stage. Through this technology, data is uploaded to the computing platform in an encrypted state, allowing the platform to directly perform addition operations on the encrypted data without decryption, thereby completing the progressive carbon footprint calculation. After the calculation is completed, the result can be decrypted to obtain the accurate total carbon footprint. The entire process achieves "usable but invisible" data, effectively protecting the business secrets of all participating parties.

[0040] Specifically, this invention designs a bottom-up automatic carbon footprint accumulation algorithm that supports seamless consolidation of multi-level supplier data: Define the transfer formula: Let the carbon footprint of product P at the k-th level be... for:

[0041] in, This represents the direct emissions (range 1+2) during the assembly process of the k-th layer. Indicates the number of components i. This indicates the carbon footprint of a component passed down from the next higher level (supplier) (which already includes all emissions from its upstream sources). This indicates the transport emission coefficient from this level to the next level.

[0042] Meanwhile, homomorphic encryption technology is used during data transmission. This allows the platform to perform addition operations in encrypted form:

[0043] The exact sum was obtained after decryption, with no plaintext leaked throughout the process.

[0044] This technological approach significantly improves the automation and data security of supply chain carbon footprint accounting by building a bottom-up automated rollout mechanism and a homomorphic encryption privacy computing framework. It avoids the inefficiency and error-proneness of traditional manual aggregation, and fundamentally resolves the contradiction between cross-level data sharing and the protection of trade secrets, providing a solid technical foundation for the reliable transmission of carbon data in large-scale, multi-level supply chains.

[0045] Step 4: Blockchain notarization and smart contract verification.

[0046] After completing the privacy-preserving aggregation and cascaded calculation of carbon footprint data, further trusted storage and automated verification of the calculation results are required to ensure the authenticity, integrity, and non-repudiation of the data. To this end, this method writes key credential information of the carbon footprint data submitted by suppliers at all levels, including but not limited to carbon footprint hash values, calculation timestamps, and digital signatures, into the blockchain network. The distributed ledger nature of the blockchain ensures that this credential information is immutable once it is on the chain, and all operation records are traceable, thus building a transparent and legally binding trust foundation for supply chain carbon accounting. Based on this, smart contracts are deployed to execute automated logical verification rules. These rules can cover dimensions such as data reasonableness threshold judgment, upstream and downstream data consistency comparison, and accounting standard version matching. The smart contracts automatically review the submitted data according to preset verification logic. When data is detected to be outside the reasonable range or has logical contradictions, it is judged as abnormal data, and rejection operations are automatically executed and the abnormal status is marked, thereby achieving efficient data quality control without human intervention.

[0047] Specifically, step 4 includes: Data on-chain: The carbon footprint hash value, accounting timestamp, and digital signature submitted by suppliers at all levels are written into the consortium blockchain to ensure that the data is tamper-proof and traceable.

[0048] Smart contract verification: Deploy smart contracts to automatically execute logical verification rules (e.g., whether the energy consumption per unit of product is within the reasonable industry threshold? Whether the upstream and downstream data match?). If the verification fails, it is automatically rejected and marked as abnormal.

[0049] Trustworthy traceability: End consumers or regulatory agencies can scan the product's QR code to view a blockchain-certified carbon footprint roadmap and verify the authenticity of data at each level.

[0050] As a specific embodiment, the system for implementing the above method should include: Supply chain collaboration portal: Allows suppliers at all levels to log in, maintain e-BOM, and upload encrypted parameters.

[0051] Federated learning compute nodes: Lightweight compute containers distributed across various enterprises.

[0052] Cascaded computing engine: Performs multi-level carbon footprint convolution and homomorphic encryption operations.

[0053] Blockchain Evidence Preservation Module: Responsible for data hashing and on-chaining, smart contract execution, and traceability query interface.

[0054] Example 2 To provide a detailed explanation of the methods proposed in this invention, specific embodiments are used to further illustrate the methods below.

[0055] Taking a car manufacturer's calculation of the carbon footprint of its entire vehicle as an example, involving hundreds of suppliers in sectors such as batteries, steel, and electronics, the specific process includes: The lithium mine (Level N) calculates the carbon footprint of lithium concentrate locally and uploads the parameters in encrypted form.

[0056] The battery manufacturer (Level N-1) receives the encrypted parameters, calculates the battery pack's carbon footprint based on its own energy consumption, and then encrypts and uploads the data again.

[0057] Car manufacturers (brand owners) aggregate data from all components and use cascaded algorithms to calculate the carbon footprint of the entire vehicle.

[0058] Verification: The smart contract detected that the unit energy consumption uploaded by a steel mill was abnormally low (below the physical limit), and automatically triggered an alarm and requested a review.

[0059] Result: While protecting the trade secrets of each supplier, the automaker obtained vehicle carbon footprint data accurate to kg CO2e and generated carbon labels with blockchain hash values.

[0060] Example 3 like Figure 2 As shown, this invention proposes a green supply chain collaborative carbon accounting system 10, comprising: The BOM carbon factor construction module 100 is used to construct an environmental bill of materials structure. It attaches dynamically updated carbon factor objects to each material node of the traditional bill of materials. The carbon factor objects include a physical attribute layer and an environmental attribute layer, and establish carbon footprint dependency relationships between parent and child nodes. The Federated Learning Privacy Modeling Module 200 is used to aggregate carbon data with privacy protection through federated learning technology. Each level of supplier trains the carbon accounting model locally using private data and only uploads the encrypted model gradient parameters to the central platform. The central platform then aggregates and updates the global carbon footprint prediction model and distributes it. The ciphertext hierarchical aggregation and accounting module 300 is used to execute the bottom-up cascading algorithm. Based on the carbon footprint data of suppliers at each level, it automatically uses the output carbon footprint of the child node as the input carbon footprint of the parent node for hierarchical convolution and employs homomorphic encryption technology during data transmission to enable the platform to complete the addition operation in ciphertext state. The blockchain trusted evidence storage and verification module 400 is used to write the carbon footprint hash value, accounting timestamp and digital signature submitted by suppliers at all levels into the blockchain, and automatically executes the logical verification rules through smart contracts to reject abnormal data, so as to realize the immutability and trusted traceability of carbon footprint data.

[0061] Furthermore, the ciphertext hierarchical aggregation and accounting module includes: The carbon footprint of product P at level k is calculated using the transfer formula, which is:

[0062] in, This represents the direct emissions during the assembly process of the k-th layer. Indicates the number of components i. This indicates the carbon footprint of components passed down from the previous level. This indicates the transport emission coefficient from this level to the next level.

[0063] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described green supply chain collaborative carbon accounting method.

[0064] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0065] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A collaborative carbon accounting method for green supply chains, characterized in that, Includes the following steps: S1. Construct an environmental bill of materials structure, attach dynamically updated carbon factor objects to each material node of the traditional bill of materials, the carbon factor objects include a physical attribute layer and an environmental attribute layer, and establish carbon footprint dependency relationships between parent and child nodes. S2 aggregates carbon data with privacy protection through federated learning technology. Suppliers at each level train carbon accounting models locally using private data, upload encrypted model gradient parameters to the central platform, and the central platform aggregates, updates, and distributes the global carbon footprint prediction model. S3 executes a bottom-up cascading algorithm. Based on the carbon footprint data of suppliers at each level, it automatically uses the output carbon footprint of the child node as the input carbon footprint of the parent node for progressive convolution. During data transmission, homomorphic encryption technology is used to enable the platform to complete the addition operation in ciphertext. S4 writes the carbon footprint hash values, accounting timestamps, and digital signatures submitted by suppliers at all levels into the blockchain, and automatically executes logical verification rules through smart contracts to reject abnormal data, thereby achieving the immutability and reliable traceability of carbon footprint data.

2. The method as described in claim 1, characterized in that, S1 includes: The physical attribute layer includes material code, name, quantity, and unit; The environmental attribute layer includes unit carbon footprint value, data source type, accounting standard version, last update time, and digital signature; The carbon footprint dependency between parent and child nodes means that the output carbon footprint of the child node automatically becomes the input carbon footprint of the parent node.

3. The method as described in claim 1, characterized in that, S2 includes: Suppliers at each level use energy consumption and production data from their private databases to train lightweight carbon accounting models and obtain model gradient parameters. Suppliers at each level will encrypt the model gradient parameters using homomorphic encryption technology before uploading them to the central platform; The central platform aggregates gradients from all parties, updates the global carbon footprint prediction model, and then distributes the updated global model to each node.

4. The method as described in claim 1, characterized in that, S3 includes: The carbon footprint of product P at level k is calculated using the transfer formula, which is: in, This represents the direct emissions during the assembly process of the k-th layer. Indicates the number of components i. This indicates the carbon footprint of components passed down from the previous level. This indicates the transport emission coefficient from this level to the next level.

5. The method as described in claim 1, characterized in that, Step S3, which describes using homomorphic encryption during data transmission to enable the platform to perform addition operations in ciphertext, includes: Using the homomorphic encryption function E, such that After decryption, the exact sum is obtained.

6. The method as described in claim 1, characterized in that, The method for marking abnormal data as rejected in step S4 is as follows: The smart contract automatically detects whether the energy consumption of a unit product is within the reasonable threshold of the industry and whether the upstream and downstream data are matched. If the verification fails, it will be automatically rejected and marked as abnormal.

7. The method as described in claim 1, characterized in that, Also includes: Generate carbon labels with blockchain hash values, allowing consumers or regulators to view a blockchain-certified carbon footprint path by scanning the product's QR code.

8. A green supply chain collaborative carbon accounting system, characterized in that, include: The BOM carbon factor construction module is used to build an environmental bill of materials structure. It attaches dynamically updated carbon factor objects to each material node of the traditional bill of materials. The carbon factor objects contain physical attribute layers and environmental attribute layers, and establish carbon footprint dependencies between parent and child nodes. The federated learning privacy modeling module is used to aggregate carbon data with privacy protection through federated learning technology. Each level of supplier trains the carbon accounting model locally using private data and only uploads the encrypted model gradient parameters to the central platform. The central platform then aggregates and updates the global carbon footprint prediction model and distributes it. The encrypted hierarchical aggregation and accounting module is used to execute the bottom-up cascading algorithm. Based on the carbon footprint data of suppliers at each level, it automatically uses the output carbon footprint of the child node as the input carbon footprint of the parent node for hierarchical convolution and employs homomorphic encryption technology during data transmission to enable the platform to complete the addition operation in encrypted state. The blockchain trusted evidence storage and verification module is used to write the carbon footprint hash value, accounting timestamp and digital signature submitted by suppliers at all levels into the blockchain, and automatically executes the logical verification rules through smart contracts to reject abnormal data, so as to realize the immutability and trusted traceability of carbon footprint data.

9. The system as described in claim 8, characterized in that, The ciphertext hierarchical aggregation and accounting module includes: The carbon footprint of product P at level k is calculated using the transfer formula, which is: in, This represents the direct emissions during the assembly process of the k-th layer. Indicates the number of components i. This indicates the carbon footprint of components passed down from the previous level. This indicates the transport emission coefficient from this level to the next level.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as claimed in any one of claims 1-7.

Images