A carbon footprint accounting input data snapshot reference binding method
By generating snapshots of input data and performing binding verification and replay verification, the problem of inconsistent results in the carbon footprint accounting system is solved, ensuring the traceability and reproducibility of the data and improving the stability and credibility of the accounting results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SPECIAL EQUIP SAFETY SUPERVISION INSPECTION INST OF JIANGSU PROVINCE
- Filing Date
- 2026-05-12
- Publication Date
- 2026-07-31
AI Technical Summary
Existing carbon footprint accounting systems suffer from inconsistent results when the source data is updated, making them difficult to trace and reproduce. Furthermore, the lack of complete records of input data affects the reliability and verifiability of the results.
By generating snapshots of input data, source anchors, original layers, processing layers, and reference layers are established to record the source location, processing trajectory, and reference data of input items. Snapshot data is then called during accounting to perform binding verification and replay verification, ensuring the stability and traceability of the results.
It achieves stability and reproducibility of carbon footprint accounting results, improves data traceability and interpretability, and avoids historical results being overwritten or confused by new data.
Smart Images
Figure CN122489495A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of carbon footprint accounting technology, specifically a method for binding snapshot references to input data in carbon footprint accounting. Background Technology
[0002] With increasing requirements for product carbon footprint assessment and supply chain carbon data disclosure, companies need to obtain input data such as raw material inputs, energy consumption, transportation data, supplier declaration data, emission factors, and allocation rules from enterprise resource planning systems, energy management platforms, logistics systems, supplier platforms, manual reporting systems, and emission factor databases when conducting carbon footprint accounting. These data sources are scattered, updated frequently, and typically require processing such as unit standardization, time trimming, boundary filtering, field mapping, anomaly handling, missing data completion, and factor matching before being entered into the accounting model.
[0003] Existing carbon footprint accounting systems typically read real-time data directly from the source system, or only save the accounting results and some input values after the accounting is completed. When the material requisition records, electricity meter data, supplier documents, manually filled-in content, or emission factor versions in the source system are updated, different input data may be read when historical accounting results are recalculated, resulting in different results for the same accounting task at different times, affecting the reproducibility and reliability of carbon footprint results.
[0004] Meanwhile, existing systems often fail to fully record the source location, extraction conditions, original fields, context information, and processing steps from the original data to the final accounting reference data. Reviewers find it difficult to determine which original record, file segment, interface response, or emission factor version a particular accounting result references, making it difficult to trace back and replay the accounting results for verification.
[0005] Furthermore, when suppliers submit supplementary data, emission factors are updated, or accounting boundaries are adjusted, existing methods typically overwrite older data with the latest data, easily leading to confusion between historical and revised accounting versions. Therefore, it is necessary to provide a method for referencing and binding carbon footprint accounting input data snapshots. This method freezes the source data, origin information, extraction conditions, and processing trajectory of input items when they are actually referenced, and establishes a binding relationship with accounting nodes and rule versions, thereby improving the traceability, verifiability, and reproducibility of carbon footprint accounting results. Summary of the Invention
[0006] Based on the shortcomings of the prior art described above, the purpose of this invention is to provide a method for binding carbon footprint accounting input data snapshot references to solve the above-mentioned technical problems.
[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for binding snapshot references to carbon footprint accounting input data, comprising:
[0008] Receive carbon footprint accounting tasks, parse the accounting objects, accounting boundaries, accounting cycles and accounting rules in the carbon footprint accounting tasks, generate multiple nodes to be accounted for based on the accounting objects, accounting boundaries, accounting cycles and accounting rules, and generate an input reference list for each node to be accounted for. The input reference list includes multiple input items that the node to be accounted for needs to call for carbon footprint accounting.
[0009] For each input item, a source anchor point is established. The source anchor point is used to record the data source location and data extraction conditions of the source data corresponding to that input item.
[0010] When any input item meets the preset reference triggering condition, extract the source data, data source information, extraction condition information and context data corresponding to that input item, and generate an input data snapshot corresponding to that input item.
[0011] The input data snapshot is used to construct an original layer, a processing layer, and a reference layer. The original layer is used to save the original content of the source data at the time of reference. The processing layer is used to save the processing trajectory of the original content into computable data. The reference layer is used to save the final reference data that actually enters the corresponding computable node after being processed by the processing trajectory.
[0012] Generate reference binding records based on carbon footprint accounting task identifier, node identifier to be accounted for identifier, input item identifier, input data snapshot identifier, source anchor point, processing trajectory, and accounting rule version;
[0013] When performing carbon footprint accounting, the reference layer data in the input data snapshot corresponding to the reference binding record is called to perform the accounting and obtain the accounting result.
[0014] Before the accounting results are output, a binding closure check is performed on the complete correspondence between the accounting results and the referenced binding records. The carbon footprint accounting process is re-executed based on the input data snapshot for re-verification.
[0015] The present invention is further configured such that the nodes to be accounted for are generated according to one or more of the following: product life cycle stage, emission source category, process unit, material flow, energy consumption flow, transportation flow, and waste disposal flow;
[0016] Input items include one or more of the following: activity data, material data, energy consumption data, transportation data, emission factor data, supplier declaration data, allocation rule data, process monitoring data, waste disposal data, and supporting documentation data.
[0017] The present invention is further configured to establish a source anchor point for each input item, including:
[0018] When the input comes from the business database, record the system identifier, data table identifier, record primary key, field name, query conditions, and extraction time;
[0019] When the input comes from data returned by the interface, record the interface identifier, request parameters, response batch, response field path, and receiving time;
[0020] When the input comes from an uploaded file, record the file identifier, file version, page number range, field area, submission subject, and submission time.
[0021] When the input items are from manually entered data, record the subject of the entry, the template number, the round of entry, the confirmation process identifier, and the confirmation time;
[0022] When the input item comes from the emission factor library, record the factor library identifier, factor entry number, applicable region, applicable time period, applicable process, and factor version.
[0023] The present invention is further configured such that the preset reference triggering condition includes any one of the following conditions:
[0024] The corresponding input item enters the execution flow of the corresponding node to be calculated for the first time;
[0025] The corresponding input item completes the preset confirmation process and reaches a verifiable state;
[0026] The carbon footprint accounting task has entered the final stage of its cycle.
[0027] The reviewer initiates a review and lock operation on the corresponding input item;
[0028] The corresponding input item was detected as a dependent input item of the calculation result to be output, which is recorded in the result reference relationship table.
[0029] The present invention is further configured such that the context data includes the original record identifier, unit of measurement, collection time range, data source description, confirmation conclusion, related business document identifier, and supporting material identifier;
[0030] When the input item is interval-type input data, the context data also includes the interval start point, interval end point, sampling frequency, and aggregation method;
[0031] When the input item is a detailed set type of input data, the context data also includes the detailed item set, item sorting rules, and total number of items;
[0032] When the input is document-based proof data, the context data also includes document page numbers, document block locations, field labels, and fragment content summaries.
[0033] The present invention is further configured such that the original content stored in the original layer includes the original value, the original unit, the original field, the original source anchor point, and the original extraction time;
[0034] The processing traces saved in the processing layer include one or more of the following processing steps: unification of original units, filtering of boundary objects, pruning of time range, mapping of field calibers, handling of outliers, filling in missing values, matching of emission factors, and confirmation of final values.
[0035] The final reference data stored in the reference layer includes one or more of the following: numerical reference data, factor reference data, rule reference data, and supporting material reference identifiers, as well as the node identifier to be calculated and the version identifier of the calculation rule corresponding to the final reference data.
[0036] The present invention is further configured such that the processing layer records the preceding input, processing action, subsequent output, and processing execution entity for each processing step;
[0037] When the corresponding processing step is triggered by manual confirmation, the confirming entity, the reason for confirmation, and the confirmation time are also recorded.
[0038] When the corresponding processing steps are executed automatically by preset rules, the rule number, rule version, and triggering conditions are also recorded.
[0039] The present invention is further configured such that the reference binding record includes carbon footprint accounting task number, accounting object number, node number to be accounted for, input item number, input data snapshot number, source anchor point summary, processing trajectory summary, accounting rule version identifier, binding generation time, binding execution subject, and binding status;
[0040] After generating a reference binding record, the reference binding record is written to the sequential append binding ledger. The sequential append binding ledger appends and saves the newly added reference binding record, preserving the order relationship between the newly added reference binding record and the previous reference binding record.
[0041] The present invention is further configured such that when performing carbon footprint accounting, the accounting engine only calls the reference layer data in the input data snapshot corresponding to the reference binding record, and does not directly call the real-time data in the source system;
[0042] A result reference table is established for each intermediate result node and the final result node. The result reference table is used to record the set of input data snapshots corresponding to the current result node, the input category corresponding to each input data snapshot, the applicable accounting rule version, and the calculation path.
[0043] The present invention is further configured such that the binding closure verification includes: verifying whether all input items associated with the accounting result have corresponding reference binding records, verifying whether each input data snapshot has source anchor, processing trajectory and reference layer data at the same time, verifying whether the accounting rule version corresponding to the input data snapshot is consistent with the accounting rule version of the accounting result record, and verifying whether there are result nodes in the result reference relationship table that are missing corresponding input data snapshots or missing corresponding reference binding records;
[0044] The replay verification includes: extracting all input data snapshots from the accounting results in reverse; re-executing the carbon footprint accounting process based only on the extracted input data snapshots, processing trajectories, and accounting rule versions, without the real-time data of the source system; and determining that the accounting results pass the replay verification when the replay results meet the preset consistency conditions with the original accounting results.
[0045] When the source data corresponding to an input item for which a reference binding record has been generated is updated, the difference between the updated source data and the original input data snapshot is identified. When the difference occurs in the field involved in the formation of the reference layer data, the emission factor matching field, or the accounting boundary determination field, a new input data snapshot and a new reference binding record are generated, forming a new accounting version that is distinct from the historical accounting version. The input data snapshot and reference binding record corresponding to the historical accounting version remain unchanged.
[0046] This invention provides a method for binding references to carbon footprint accounting input data snapshots. It receives a carbon footprint accounting task, parses the accounting object, accounting boundary, accounting cycle, and accounting rules within the task, generates multiple nodes to be accounted for based on these criteria, and generates an input reference list for each node. This list includes multiple input items required for carbon footprint accounting by the node. A source anchor is established for each input item, recording the data source location and data extraction conditions of the corresponding source data. When any input item meets a preset reference trigger condition, the source data, data source information, extraction condition information, and context data corresponding to that input item are extracted to generate an input data snapshot. An original layer is constructed from the input data snapshot. The system consists of a processing layer and a referencing layer. The original layer stores the original content of the source data at the time it is referenced. The processing layer stores the processing trajectory from the original content to the computationally achievable data. The referencing layer stores the final referenced data that actually enters the corresponding computation node after processing by the trajectory. Reference binding records are generated based on the carbon footprint accounting task identifier, computation node identifier, input item identifier, input data snapshot identifier, source anchor point, processing trajectory, and accounting rule version. During carbon footprint accounting, the reference layer data in the input data snapshot corresponding to the reference binding record is called for accounting to obtain the accounting result. Before the accounting result is output, a binding closure check is performed on the complete correspondence between the accounting result and the reference binding record. The carbon footprint accounting process is re-executed based on the input data snapshot for re-verification. The beneficial effects include:
[0047] 1. By generating a snapshot of the input data when the input item meets the reference triggering conditions, and establishing a reference binding record between this snapshot and the carbon footprint accounting task, the node to be accounted for, the input item, and the version of the accounting rule, the accounting engine only calls the reference layer data in the bound snapshot when performing the accounting, and does not directly read the real-time data of the source system. Therefore, even if the source system data, supplier declaration content, or emission factor version changes later, it will not affect the historical accounting results, improving the stability and reproducibility of the carbon footprint accounting results;
[0048] 2. By establishing source anchors for each input item and dividing the input data snapshot into an original layer, a processing layer, and a referencing layer, the original layer stores the original data at the time of reference. The processing layer records the processing trajectory, including unit unification, boundary filtering, time clipping, field mapping, anomaly handling, missing data completion, and factor matching. The referencing layer stores the final referenced data that actually enters the accounting model. Reviewers can reverse-engineer the input source and verify the data processing process, improving the traceability and interpretability of the accounting basis.
[0049] 3. By performing binding closure verification and replay verification before outputting the accounting results, the system verifies the complete correspondence between input snapshots, reference binding records, rule versions, and result nodes. This allows for the re-execution of the accounting process even without real-time data from the source system. When the source data is subsequently updated, historical snapshots are not overwritten. Instead, new input data snapshots and reference binding records are generated when key fields change, forming a new accounting version and preventing historical results from being overwritten or confused by new data.
[0050] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description
[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:
[0052] Figure 1 The flowchart illustrates a carbon footprint accounting input data snapshot reference binding method as an exemplary embodiment of the present invention. Detailed Implementation
[0053] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.
[0054] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.
[0055] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.
[0056] A method for binding snapshot references to input data for carbon footprint accounting, such as Figure 1 As shown, it includes:
[0057] Receive carbon footprint accounting tasks, parse the accounting objects, accounting boundaries, accounting cycles and accounting rules in the carbon footprint accounting tasks, generate multiple nodes to be accounted for based on the accounting objects, accounting boundaries, accounting cycles and accounting rules, and generate an input reference list for each node to be accounted for. The input reference list includes multiple input items that the node to be accounted for needs to call for carbon footprint accounting.
[0058] For each input item, a source anchor point is established. The source anchor point is used to record the data source location and data extraction conditions of the source data corresponding to that input item.
[0059] When any input item meets the preset reference triggering condition, extract the source data, data source information, extraction condition information and context data corresponding to that input item, and generate an input data snapshot corresponding to that input item.
[0060] The input data snapshot is used to construct an original layer, a processing layer, and a reference layer. The original layer is used to save the original content of the source data at the time of reference. The processing layer is used to save the processing trajectory of the original content into computable data. The reference layer is used to save the final reference data that actually enters the corresponding computable node after being processed by the processing trajectory.
[0061] Generate reference binding records based on carbon footprint accounting task identifier, node identifier to be accounted for identifier, input item identifier, input data snapshot identifier, source anchor point, processing trajectory, and accounting rule version;
[0062] When performing carbon footprint accounting, the reference layer data in the input data snapshot corresponding to the reference binding record is called to perform the accounting and obtain the accounting result.
[0063] Before the accounting results are output, a binding closure check is performed on the complete correspondence between the accounting results and the referenced binding records. The carbon footprint accounting process is re-executed based on the input data snapshot for re-verification.
[0064] Specifically, the aforementioned carbon footprint accounting input data snapshot reference binding method addresses the problems of scattered input data sources, difficulty in reproducing historical accounting results due to subsequent changes in source data, and the lack of a stable binding relationship between accounting results and input basis during the carbon footprint accounting process. This method can be deployed in carbon footprint accounting systems, enterprise carbon management platforms, product lifecycle assessment systems, or supply chain carbon data management systems, and is executed jointly by servers, databases, accounting engines, and data acquisition interfaces.
[0065] Furthermore, a carbon footprint accounting task refers to the task of calculating the carbon footprint of an emission object within a specific product, batch of products, production cycle, or organizational boundary. A carbon footprint accounting task includes at least the accounting object, accounting boundary, accounting cycle, and accounting rules. The accounting object can be a specific product, a batch of products, a production line, or a supply chain delivery unit; the accounting boundary defines the scope of data included in the accounting; the accounting cycle defines the time frame of the data involved in the accounting; and the accounting rules define the calculation methods, factor matching rules, data pruning rules, and allocation rules used in the carbon footprint calculation.
[0066] A carbon footprint accounting node is a computational unit formed by breaking down a carbon footprint accounting task. Nodes can be divided according to product lifecycle stage, emission source category, process unit, material flow, energy consumption flow, transportation flow, or waste disposal flow. For example, for a carbon footprint accounting task of a specific injection-molded housing component, nodes for raw material input, electricity consumption, auxiliary material input, transportation, waste disposal, and emission factor matching can be generated. Each node has a corresponding input reference list, which records the input items required for carbon footprint accounting at that node.
[0067] Input items refer to the data objects that the node to be accounted for needs to reference when performing carbon footprint accounting. Input items can include activity data, material data, energy consumption data, transportation data, emission factor data, supplier declaration data, allocation rule data, process monitoring data, waste disposal data, and supporting documentation data. Activity data can include production volume, input volume, transportation volume, and processing volume; material data can include the input volume of raw materials, auxiliary materials, and packaging materials; energy consumption data can include the consumption of electricity, steam, natural gas, and compressed air; transportation data can include transportation distance, transportation mode, and load information; emission factor data can include electricity emission factors, material emission factors, transportation emission factors, or waste disposal factors; supplier declaration data can be carbon emission declarations of upstream products provided by suppliers; supporting documentation data can include invoices, transport documents, test reports, supplier declaration documents, electricity bills, or metering records.
[0068] Source anchors are used to characterize the data source location and data extraction conditions corresponding to the input item's source data. Source anchors are not simply data numbers, but rather a set of location information used by subsequent reviewers to trace the original source of the input item. When the input item originates from a business database, the source anchor includes the system identifier, data table identifier, record primary key, field name, query conditions, and extraction time; when the input item originates from interface returned data, the source anchor includes the interface identifier, request parameters, response batch, response field path, and receiving time; when the input item originates from an uploaded file, the source anchor includes the file identifier, file version, page number segment, field area, submitting entity, and submission time; when the input item originates from manually entered data, the source anchor includes the submitting entity, submission template number, submission round, confirmation process identifier, and confirmation time; when the input item originates from an emission factor library, the source anchor includes the factor library identifier, factor entry number, applicable region, applicable time period, applicable process, and factor version.
[0069] An input data snapshot is a data object created by freezing and saving the source data, data source information, extraction condition information, and context data corresponding to an input item when the reference trigger condition is met. Input data snapshots ensure that the input basis for an input item remains unchanged when it is referenced in a specific accounting task, a specific accounting node, or a specific accounting rule version. Even if the original data in the source system is subsequently modified, the input data snapshot retains its content at the time of generation.
[0070] Reference trigger conditions are used to determine when to generate input data snapshots. Reference trigger conditions can include: the corresponding input item first entering the execution flow of the corresponding accounting node; the corresponding input item completing the preset confirmation process and reaching an accountable state; the carbon footprint accounting task entering the periodic sealing stage; the reviewer initiating a review and locking operation on the corresponding input item; and the system detecting that the corresponding input item is recorded as a dependent input item for the accounting result to be output in the result reference relationship table. Through these trigger conditions, input data snapshots are not simply backed up at fixed intervals, but are frozen at the moment the input item is actually referenced by the accounting task, thus ensuring a clear correspondence between the snapshot and the accounting behavior.
[0071] The input data snapshot comprises three layers: the raw layer, the processing layer, and the referencing layer. The raw layer stores the original content of the source data at the time it is referenced, including original values, original units, original fields, original source anchors, and the original extraction time. The processing layer stores the processing trajectory from the raw content to the calculated data. This trajectory includes steps such as unit unification, boundary object filtering, time range trimming, field caliber mapping, outlier handling, missing value imputation, emission factor matching, and final value confirmation. The referencing layer stores the final referenced data that actually enters the corresponding calculation node after processing by the trajectory. The final referenced data can be numerical, factor-based, rule-based, or a supporting document reference identifier. For example, electricity consumption becomes numerical referenced data after time range trimming and unit unification; a factor entry in the emission factor library becomes factor-based referenced data after matching the applicable region and applicable time period; allocation rules become rule-based referenced data after version confirmation; and supplier reports become supporting document reference identifiers after document location and fragment confirmation.
[0072] Reference binding records are used to document the binding relationships between a specific input data snapshot and a specific carbon footprint accounting task, a node to be accounted for, an input item, and an accounting rule version. A reference binding record includes the carbon footprint accounting task number, accounting object number, node to be accounted for number, input item number, input data snapshot number, source anchor summary, processing trajectory summary, accounting rule version identifier, binding generation time, binding execution entity, and binding status. After generation, reference binding records are written to a sequential append-only binding ledger. The sequential append-only binding ledger saves newly added reference binding records in an append-only manner, preserving the order relationship between new reference binding records and previous reference binding records. Historical reference binding records already written are not overwritten by subsequent updates.
[0073] In this embodiment, the carbon footprint accounting system includes a task parsing unit, a source anchoring unit, a snapshot generation unit, a processing trajectory recording unit, a reference binding unit, an accounting execution unit, a closure verification unit, a replay verification unit, and a version fork unit. The task parsing unit receives and parses the carbon footprint accounting task; the source anchoring unit establishes source anchors for input items; the snapshot generation unit generates an input data snapshot when the reference triggering condition is met; the processing trajectory recording unit forms the original layer, processing layer, and reference layer; the reference binding unit generates reference binding records; the accounting execution unit performs carbon footprint accounting based on the reference layer data; the closure verification unit determines whether the accounting result corresponds completely with the reference binding record; the replay verification unit re-executes the accounting process under conditions detached from the source system's real-time data; and the version fork unit generates new input data snapshots and new reference binding records when the source data is updated.
[0074] In one specific embodiment, the company performs a product carbon footprint accounting for injection-molded housing components produced in January 2025. The accounting object is the injection-molded housing components, the accounting period is from January 1, 2025 to January 31, 2025, the accounting boundaries are raw material input, electricity consumption, packaging material input, and off-site transportation, and the accounting rule version is the company's Product Carbon Footprint Accounting Rule 2025 version. After receiving this carbon footprint accounting task, the raw material input node, electricity consumption node, packaging material node, transportation node, and emission factor matching node are parsed.
[0075] For raw material input nodes, the generated input reference list includes polycarbonate resin input quantity, recycled material input quantity, supplier declaration documents, and raw material emission factors. The polycarbonate resin input quantity is derived from the warehousing and material requisition records in the Enterprise Resource Planning (ERP) system. The source anchor record includes the system identifier, data table identifier, material requisition record primary key, material code field, material requisition quantity field, accounting cycle query conditions, and extraction time. The supplier declaration document is derived from the product carbon emission declaration document uploaded by the supplier. The source anchor record includes the file identifier, file version, page number section, field area, submitting entity, and submission time. The raw material emission factors are derived from the emission factor library. The source anchor record includes the factor library identifier, factor entry number, applicable region, applicable time period, applicable process, and factor version.
[0076] For the electricity consumption node, the generated input reference list includes metering data from the injection molding workshop, product output data, allocation rules, and electricity emission factors. The metering data from the injection molding workshop originates from the energy management platform interface, with the source anchor recording the interface identifier, request parameters, response batch, response field path, and receiving time. The product output data originates from the Manufacturing Execution System (MES), with the source anchor recording the system identifier, data table identifier, production batch primary key, output field, product code field, and extraction time. The allocation rules originate from the enterprise accounting rule base, with the source anchor recording the rule base identifier, rule number, rule version, applicable objects, and activation time. The electricity emission factors originate from the emission factor library, with the source anchor recording the factor entry number, applicable region, applicable year, and factor version.
[0077] For each transport node, the generated input reference list includes the transport order number, transport mode, transport distance, cargo weight, and transport emission factor. The transport order number, transport mode, and transport distance are derived from the logistics management system, with the source anchor recording the system identifier, transport order primary key, field name, query conditions, and retrieval time. The transport emission factor is derived from the emission factor database, with the source anchor recording the factor entry number, applicable transport mode, applicable region, and factor version.
[0078] When the raw material input node first enters the accounting execution process, it is determined that the polycarbonate resin input quantity meets the reference triggering condition. The snapshot generation unit extracts the source data, source anchor point, and context data corresponding to this input item to generate an input data snapshot. The original layer of this snapshot saves the original quantity, original unit, material code, record primary key, field name, query conditions, and extraction time of the warehousing record and material requisition record at the reference time. The processing layer of this snapshot records the processing of the original material requisition data, including removing material requisition records outside the accounting cycle, filtering material requisition records belonging to injection molded shell component product batches, unifying different units of measurement to kilograms, excluding trial production material records that have not entered the accounting boundary, and forming the material input quantity that can be used for the raw material input node. The reference layer of this snapshot saves the final material input quantity, unit of measurement, corresponding node identifier to be accounted for, and accounting rule version identifier that enter the raw material input node.
[0079] When a power consumption node enters the accounting execution process, it is determined that the meter readings meet the reference triggering conditions. The snapshot generation unit obtains the meter readings and interval power consumption within the accounting period from the energy management platform interface. The original layer of this snapshot saves the original power consumption, acquisition time, unit of measurement, interface identifier, request parameters, and response batch returned by the interface. The processing layer records time range clipping, abnormal reading removal, missing period completion, allocation processing between total workshop power consumption and product output, and unit unification processing. The reference layer saves the power consumption, unit of measurement, node identifier to be calculated, and accounting rule version identifier assigned to the injection molded housing component after allocation.
[0080] When the emission factor matching node is executed, it matches the corresponding factor entries in the emission factor library based on the type of raw materials, power region, transportation mode, and accounting cycle, and generates an input data snapshot for each matched factor. The original layer of the factor snapshot stores the original content of the factor entry, including factor name, factor value, applicable region, applicable time period, applicable process, data source description, and factor version. The processing layer stores the factor applicability matching process, including matching by material type, matching by region, matching by accounting cycle, and matching by process conditions. The reference layer stores the factor-type reference data, factor version identifier, node identifier to be calculated, and rule version identifier that finally enter the accounting model.
[0081] After each input data snapshot is generated, the reference binding unit generates a reference binding record. For example, for the polycarbonate resin input item, the reference binding record includes the carbon footprint accounting task number, the injection molding shell component accounting object number, the raw material input node number, the polycarbonate resin input item number, the corresponding input data snapshot number, the source anchor summary, the processing trajectory summary, the accounting rule version identifier, the binding generation time, the binding execution entity, and the binding status. The binding status can include generated, verified, locked, and invalidated. Once the reference binding record is written to the sequential append-only binding ledger, it will not be changed by subsequent modifications to the material requisition records in the enterprise resource planning system.
[0082] When performing carbon footprint accounting, the accounting execution unit no longer directly accesses real-time data from the enterprise resource planning system, energy management platform, logistics management system, or emission factor database. Instead, it only calls the reference layer data from the input data snapshot that has already generated reference binding records. For raw material input nodes, the accounting execution unit reads the material input quantity reference data and raw material emission factor reference data; for electricity consumption nodes, it reads the electricity consumption quantity reference data and electricity emission factor reference data; for transportation nodes, it reads the transportation quantity reference data and transportation emission factor reference data. The accounting execution unit performs calculations according to the corresponding accounting rule version, generates intermediate results for each node to be accounted for, and further summarizes them to form the final carbon footprint accounting result.
[0083] When generating intermediate and final results, the system establishes a result reference table. This table records the set of input data snapshots that each result node depends on, the input category corresponding to each snapshot, the applicable accounting rule version, and the calculation path. For example, the result reference table for the raw material input node records snapshots of polycarbonate resin input, recycled material input, supplier declaration documents, and raw material emission factors; the result reference table for the electricity consumption node records snapshots of electricity metering data, product output data, allocation rules, and electricity emission factors. The final carbon footprint result's result reference table further aggregates all input data snapshots corresponding to each intermediate result node, enabling the final result to be traced back to all input bases.
[0084] Before the calculation results are output, the closure verification unit performs binding closure verification. Binding closure verification includes: verifying whether all input items associated with the final calculation result have corresponding reference binding records; verifying whether each input data snapshot has source anchor, processing trajectory, and reference layer data simultaneously; verifying whether the calculation rule version corresponding to the input data snapshot is consistent with the calculation rule version of the calculation result record; and verifying whether there are any result nodes in the result reference relationship table that lack corresponding input data snapshots or corresponding reference binding records. When any input item lacks an input data snapshot, or any result node cannot be back-located to the corresponding reference binding record, the system marks the result node as unclosed and outputs the missing input item number, the node number to be calculated, and the reason for the missing information. Only when all input items involved in the formation of the calculation result have complete reference binding records, and all input data snapshots have source anchor, processing trajectory, and reference layer data, will the system mark the calculation result as bound closed.
[0085] After the binding closure verification passes, the replay verification unit performs replay verification. During replay verification, the system extracts all input data snapshots from the final calculation result and, under the state of real-time data from the isolated source system, re-executes the carbon footprint calculation process based only on the extracted input data snapshots, processing trajectories, and calculation rule versions. Real-time data from the isolated source system refers to the current real-time data in the Enterprise Resource Planning System, Energy Management Platform, Logistics Management System, Manual Reporting System, or Emission Factor Database that is no longer accessed during the replay verification process; instead, only the already bound input data snapshots are read. The replay results obtained from the replay verification are compared with the original calculation results for consistency judgment. When the numerical difference between the replay results and the original calculation results does not exceed the preset accuracy range, and the set of input data snapshots referenced during the replay process is consistent with the set of input data snapshots corresponding to the original calculation results, the calculation result is determined to have passed the replay verification. When the replay results are inconsistent with the original calculation results, the source of the difference is located according to the result reference relationship table, and the node to be calculated, input item, input data snapshot, and processing step to which the difference belongs are output.
[0086] After the accounting results are output, if the source data corresponding to the input items in the source system is updated, the version fork unit does not overwrite the historical input data snapshot. Instead, it identifies the differences between the updated source data and the original input data snapshot. Difference identification includes changes in data values, units of measurement, sources, confirmation status, factor versions, and applicable boundaries. When a difference occurs in a descriptive or memo field and does not affect the formation of the reference layer data, the system retains the historical accounting version and records the difference description. When a difference occurs in a field involved in the formation of the reference layer data, an emission factor matching field, or an accounting boundary determination field, the system generates a new input data snapshot and a new reference binding record, forming a new accounting version distinct from the historical accounting version. The input data snapshot, reference binding record, result reference relationship table, and accounting results corresponding to the historical accounting version remain unchanged.
[0087] For example, after the aforementioned calculation of the injection-molded housing component is completed, the supplier submits a new polycarbonate resin carbon emission declaration document, or the emission factor library updates the raw material emission factors to a new version. The system detects that this change involves emission factor matching fields or core numerical fields in the supplier's declaration data. Therefore, it does not overwrite historical snapshots but generates a new input data snapshot for the new supplier declaration document or the new emission factor entry. The new snapshot has a new snapshot number, a new source anchor, and a new processing trajectory, and establishes a reference binding record with the new calculation version. The original calculation version still corresponds to the original supplier declaration document snapshot and the original emission factor snapshot. Subsequent reviewers can either recreate the original calculation results based on the original snapshot or generate revised calculation results based on the new snapshot.
[0088] The above methods achieve input source location, reference time point freezing, three-layer snapshot construction, accounting node-level binding, snapshot-based accounting, result closure verification, replay verification detached from real-time source data, and version forking after source data updates. This method ensures that carbon footprint accounting results stably correspond to their input basis, avoids historical accounting result drift caused by subsequent changes in source system data, and improves the traceability, verifiability, and replayability of carbon footprint accounting results.
[0089] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A carbon footprint accounting input data snapshot reference binding method, characterized by, include: Receive carbon footprint accounting tasks, parse the accounting objects, accounting boundaries, accounting cycles and accounting rules in the carbon footprint accounting tasks, generate multiple nodes to be accounted for based on the accounting objects, accounting boundaries, accounting cycles and accounting rules, and generate an input reference list for each node to be accounted for. The input reference list includes multiple input items that the node to be accounted for needs to call for carbon footprint accounting. For each input item, a source anchor point is established. The source anchor point is used to record the data source location and data extraction conditions of the source data corresponding to that input item. When any input item meets the preset reference triggering condition, extract the source data, data source information, extraction condition information and context data corresponding to that input item, and generate an input data snapshot corresponding to that input item. The input data snapshot is used to construct an original layer, a processing layer, and a reference layer. The original layer is used to save the original content of the source data at the time of reference. The processing layer is used to save the processing trajectory of the original content into computable data. The reference layer is used to save the final reference data that actually enters the corresponding computable node after being processed by the processing trajectory. Generate reference binding records based on carbon footprint accounting task identifier, node identifier to be accounted for identifier, input item identifier, input data snapshot identifier, source anchor point, processing trajectory, and accounting rule version; When performing carbon footprint accounting, the reference layer data in the input data snapshot corresponding to the reference binding record is called to perform the accounting and obtain the accounting result. Before the accounting results are output, a binding closure check is performed on the complete correspondence between the accounting results and the referenced binding records. The carbon footprint accounting process is re-executed based on the input data snapshot for re-verification.
2. The carbon footprint accounting input data snapshot reference binding method of claim 1, wherein, The nodes to be accounted for are generated according to one or more of the following: product life cycle stage, emission source category, process unit, material flow, energy consumption flow, transportation flow, and waste disposal flow; Input items include one or more of the following: activity data, material data, energy consumption data, transportation data, emission factor data, supplier declaration data, allocation rule data, process monitoring data, waste disposal data, and supporting documentation data.
3. The carbon footprint accounting input data snapshot reference binding method of claim 1, wherein, For each input item, establish a source anchor point, including: When the input comes from the business database, record the system identifier, data table identifier, record primary key, field name, query conditions, and extraction time; When the input comes from data returned by the interface, record the interface identifier, request parameters, response batch, response field path, and receiving time; When the input comes from an uploaded file, record the file identifier, file version, page number range, field area, submission subject, and submission time. When the input items are from manually entered data, record the subject of the entry, the template number, the round of entry, the confirmation process identifier, and the confirmation time; When the input item comes from the emission factor library, record the factor library identifier, factor entry number, applicable region, applicable time period, applicable process, and factor version.
4. The carbon footprint accounting input data snapshot reference binding method according to claim 1, characterized in that, The default reference trigger conditions include any one of the following conditions: The corresponding input item enters the execution flow of the corresponding node to be calculated for the first time; The corresponding input item completes the preset confirmation process and reaches a verifiable state; The carbon footprint accounting task has entered the final stage of its cycle. The reviewer initiates a review and lock operation on the corresponding input item; The corresponding input item was detected as a dependent input item of the calculation result to be output, which is recorded in the result reference relationship table.
5. The carbon footprint accounting input data snapshot reference binding method according to claim 1, characterized in that, Contextual data includes original record identifier, unit of measurement, collection time range, data source description, confirmation conclusion, related business document identifier, and supporting document identifier; When the input item is interval-type input data, the context data also includes the interval start point, interval end point, sampling frequency, and aggregation method; When the input item is a detailed set type of input data, the context data also includes the detailed item set, item sorting rules, and total number of items; When the input is document-based proof data, the context data also includes document page numbers, document block locations, field labels, and fragment content summaries.
6. The carbon footprint accounting input data snapshot reference binding method according to claim 1, characterized in that, The original content stored in the original layer includes the original values, original units, original fields, original source anchors, and original extraction time. The processing traces saved in the processing layer include one or more of the following processing steps: unification of original units, filtering of boundary objects, pruning of time range, mapping of field calibers, handling of outliers, filling in missing values, matching of emission factors, and confirmation of final values. The final reference data stored in the reference layer includes one or more of the following: numerical reference data, factor reference data, rule reference data, and supporting material reference identifiers, as well as the node identifier to be calculated and the version identifier of the calculation rule corresponding to the final reference data.
7. The carbon footprint accounting input data snapshot reference binding method according to claim 6, characterized in that, The processing layer records the preceding inputs, processing actions, subsequent outputs, and the processing execution entity for each processing step; When the corresponding processing step is triggered by manual confirmation, the confirming entity, the reason for confirmation, and the confirmation time are also recorded. When the corresponding processing steps are executed automatically by preset rules, the rule number, rule version, and triggering conditions are also recorded.
8. The carbon footprint accounting input data snapshot reference binding method according to claim 1, characterized in that, The reference binding record includes the carbon footprint accounting task number, accounting object number, node number to be accounted for number, input item number, input data snapshot number, source anchor point summary, processing trajectory summary, accounting rule version identifier, binding generation time, binding execution subject, and binding status; After generating a reference binding record, the reference binding record is written to the sequential append binding ledger. The sequential append binding ledger appends and saves the newly added reference binding record, preserving the order relationship between the newly added reference binding record and the previous reference binding record.
9. The carbon footprint accounting input data snapshot reference binding method according to claim 1, characterized in that, When performing carbon footprint accounting, the accounting engine only calls the reference layer data in the input data snapshot corresponding to the reference binding record, and does not directly call the real-time data in the source system; A result reference table is established for each intermediate result node and the final result node. The result reference table is used to record the set of input data snapshots corresponding to the current result node, the input category corresponding to each input data snapshot, the applicable accounting rule version, and the calculation path.
10. A method for binding snapshot references to input data for carbon footprint accounting according to claim 1, characterized in that, The binding closure verification includes: verifying whether all input items associated with the accounting result have corresponding reference binding records; verifying whether each input data snapshot has source anchor, processing trajectory and reference layer data at the same time; verifying whether the accounting rule version corresponding to the input data snapshot is consistent with the accounting rule version of the accounting result record; and verifying whether there are result nodes in the result reference relationship table that are missing corresponding input data snapshots or missing corresponding reference binding records. The replay verification includes: extracting all input data snapshots from the accounting results in reverse; re-executing the carbon footprint accounting process based only on the extracted input data snapshots, processing trajectories, and accounting rule versions, without the real-time data of the source system; and determining that the accounting results pass the replay verification when the replay results meet the preset consistency conditions with the original accounting results. When the source data corresponding to an input item for which a reference binding record has been generated is updated, the difference between the updated source data and the original input data snapshot is identified. When the difference occurs in the field involved in the formation of the reference layer data, the emission factor matching field, or the accounting boundary determination field, a new input data snapshot and a new reference binding record are generated, forming a new accounting version that is distinct from the historical accounting version. The input data snapshot and reference binding record corresponding to the historical accounting version remain unchanged.