A data management method and system for carbon footprint of copper products

Abstract

The application discloses a copper product carbon footprint-oriented data management method and system, and relates to the technical field of carbon footprint data management. The method comprises the following steps: acquiring multi-source original data related to copper product carbon footprint accounting; performing source data quality verification and quality marking shunting based on data source types, data structure characteristics and preset carbon footprint data quality rules to obtain qualified source data and abnormal source data; storing the qualified source data as is to obtain source-attached data; performing data cleaning, field semantic alignment, missing data processing, unified coding mapping and metadata labeling on the source-attached data to obtain standardized detailed data; performing semantic classification, association and aggregation on the standardized detailed data to generate carbon footprint accounting parameter data and record data blood relationship information. Thus, the quality identification, cross-system association and traceability of copper product carbon footprint data are improved.

Description

Technical Field

[0001] This application relates to the field of carbon footprint data management technology, and in particular to a data management method and system for the carbon footprint of copper products. Background Technology

[0002] With increasing demands for green and low-carbon development, product carbon footprint has gradually become a crucial basis for the non-ferrous metals industry to conduct carbon emission management, supply chain collaboration, product evaluation, and external compliance reporting. The production process of copper products typically involves multiple stages, including raw material input, energy consumption, processing steps, equipment operation, product output, and transportation. The data types, collection frequencies, and management methods vary significantly across these stages. To obtain accurate carbon footprint results for copper products, it is generally necessary to collect, organize, and manage production process data, energy consumption data, material consumption data, equipment operation data, emission factor data, and external supply chain data.

[0003] In related technologies, enterprises typically store the aforementioned data separately through enterprise resource planning (ERP) systems, manufacturing execution systems (MES), industrial IoT platforms, energy management systems, spreadsheets, or shared files. When carbon footprint accounting is required, accounting personnel extract relevant data from different systems or files for organization and summarization. For enterprises with a high degree of informatization, some production and operation data can also be centrally stored and retrieved through general data warehouses or data platforms. However, these methods are more geared towards general production management, operation management, or equipment monitoring needs, and the data sources, data formats, statistical definitions, and coding rules are often not entirely consistent. At the same time, some data is prone to missing, duplicate, abnormal, inconsistent definitions, or unclear sources during collection, conversion, summarization, and use. When this data is further used for copper product carbon footprint accounting, it can easily lead to problems such as a large workload in preparing the basic accounting data, unstable cross-system data correlation, difficulty in timely identification of abnormal data, and difficulty in effectively tracing the accounting results back to the original data, thereby affecting the reliability of copper product carbon footprint data management and subsequent accounting work.

[0004] Therefore, in the management of carbon footprint data for copper products, the inconsistency of data from multiple sources, the difficulty in timely identification of the quality of basic accounting data, the insufficient accuracy of cross-system data correlation, and the difficulty in tracing carbon footprint accounting data have become urgent problems that need to be solved. Summary of the Invention

[0005] This application provides a data management method and system for the carbon footprint of copper products, aiming to solve the problems in the existing technology of copper product carbon footprint data management, such as inconsistent data from multiple sources, difficulty in timely identification of the quality of basic accounting data, insufficient accuracy of cross-system data correlation, and difficulty in tracing carbon footprint accounting data.

[0006] Firstly, a data management method for the carbon footprint of copper products, the method comprising: Obtain multi-source raw data related to the carbon footprint accounting of copper products. The multi-source raw data includes activity data, emission factor data, and supply chain related data. The activity data includes at least two of the following: production process data, energy consumption data, material consumption data, and equipment operation data. Based on the data source type, data structure characteristics, and preset carbon footprint data quality rules of the multi-source raw data, the multi-source raw data is subjected to source data quality verification, including format compliance verification, business rationality verification, and reference integrity verification. Based on the source data quality verification results, the multi-source raw data is quality-marked and diverted to obtain qualified source data and abnormal source data. The qualified source data is stored in its original form according to the source system and collection time to obtain the patch source data, and the abnormal source data and the quality verification record corresponding to the abnormal source data are stored. The source data is cleaned, field semantic alignment is performed, missing data is handled, unified encoding mapping is performed, and metadata is labeled to obtain standardized detailed data; During the data flow process of the multi-source raw data, the source-attached data, the standardized detailed data, the summary data, and the carbon footprint accounting parameter data, the data at each level is semantically categorized according to the copper product carbon footprint accounting related objects, and data association relationships between different accounting related objects are established based on the semantic classification results. The copper product carbon footprint accounting related objects include product objects, process objects, material objects, energy emission objects, equipment objects, and accounting result objects. Based on the aforementioned data relationships, the standardized detailed data is aggregated according to product, process, and time dimensions to obtain summary data for copper product carbon footprint accounting. According to the preset carbon footprint accounting rules, activity data values ​​and emission factor values ​​are extracted from the standardized detailed data and / or the summarized data to generate carbon footprint accounting parameter data. Record the data lineage information from the multi-source raw data to the carbon footprint accounting parameter data. The data lineage information includes data input / output identifiers and data processing rule information to establish a traceability association between the carbon footprint accounting parameter data and the corresponding multi-source raw data.

[0007] Optionally, in the above scheme, obtaining the multi-source raw data associated with the carbon footprint accounting of copper products includes: Obtain the source attributes, update frequency, and data structure information of candidate data sources to obtain data source feature information; Based on the data source feature information, the candidate data sources are divided into active data sources and emission factor data sources. The active data sources include at least two of the following: industrial IoT time-series data sources, production and operation structured data sources, and supply chain external exchange data sources. The emission factor data sources include at least one of the following: external databases, standard data sources, and literature reference data sources. Based on the activity data source and the emission factor data source, the corresponding data access strategies are determined to obtain the multi-source data access strategy. According to the multi-source data access strategy, data related to the carbon footprint accounting of copper products are collected from the candidate data sources to obtain initial collected data; The initial collected data is preprocessed to obtain the multi-source raw data.

[0008] Optionally, in the above scheme, the source data quality verification of the multi-source raw data is performed based on the data source type, data structure characteristics, and preset carbon footprint data quality rules. This verification includes format compliance verification, business rationality verification, and reference integrity verification. Based on the source data quality verification results, the multi-source raw data is then quality-marked and segmented to obtain qualified source data and abnormal source data. This includes: Based on the data structure characteristics, the multi-source raw data is subjected to field type validation, required field validation, timestamp format validation, and duplicate record validation to obtain the format compliance validation result. Based on the data source type, the data to be audited is determined from the data that has passed the format compliance check, and the data to be audited is obtained. Based on the preset carbon footprint data quality rules, the data set to be audited is subjected to material input-output consistency verification and / or process parameter range verification to obtain the business rationality verification result. Based on the preset coding dimension table, the material code, process code and equipment code in the data that has passed the format compliance check are checked for referential integrity, and the referential integrity check results are obtained. Based on the format compliance verification result, the business rationality verification result, and the reference integrity verification result, the source data quality verification result is determined; Based on the source data quality verification results, the multi-source raw data is quality-marked and split to obtain the qualified source data and the abnormal source data.

[0009] Optionally, in the above scheme, the step of performing quality marking and splitting on the multi-source raw data based on the source data quality verification result to obtain the qualified source data and the abnormal source data includes: When the format compliance verification result, the business rationality verification result, and the reference integrity verification result all meet the corresponding verification pass conditions, a qualified mark is added to the corresponding multi-source original data to obtain the qualified source data; When at least one of the format compliance verification result, the business rationality verification result, and the reference integrity verification result fails to meet the corresponding verification pass condition, an anomaly marker is added to the corresponding multi-source original data to obtain the abnormal source data; The quality verification record is generated based on the anomaly type, anomaly field, anomaly triggering rule, and quality verification time of the anomaly source data. The original data snapshot of the abnormal source data is associated with the quality verification record and stored to obtain the abnormal data processing record.

[0010] Optionally, in the above scheme, the step of performing data cleaning, field semantic alignment, missing data processing, unified encoding mapping, and metadata annotation on the source data to obtain standardized detailed data includes: The source data is cleaned by removing outliers and duplicate data to obtain cleaned data. Semantic alignment is performed on fields from different sources but with the same semantics in the cleaned data to obtain semantically aligned data. Missing fields in the semantic alignment data are processed for missing data, and missing processing identifiers are added to the fields that have undergone missing data processing to obtain completion marker data; Based on a preset standard coding system, the product code, process code, material code, energy code, and equipment code in the completed tag data are uniformly coded and mapped to obtain standardized coding data. Add information about the data source, update time, data quality score, and field meaning to the coded standardized data to obtain the standardized detailed data.

[0011] Optionally, in the above scheme, during the data flow process of the multi-source raw data, the source-attached data, the standardized detailed data, the summary data, and the carbon footprint accounting parameter data, the data at each level is semantically categorized according to the copper product carbon footprint accounting related objects, and data association relationships between different accounting related objects are established based on the semantic categorization results, including: Identify product identifiers, process identifiers, material identifiers, energy emission data identifiers, equipment identifiers, and accounting result data identifiers from data at each level to obtain a set of accounting object identifiers; Based on the set of accounting object identifiers, the data at each level are classified into the product object, the process object, the material object, the energy emission object, the equipment object, and the accounting result object, respectively, to obtain semantic classification results; Based on the semantic classification results, the product process association relationship between product objects and process objects, the process material association relationship between process objects and material objects, the process emission association relationship between process objects and energy emission objects, the process equipment association relationship between process objects and equipment objects, and the emission accounting association relationship between energy emission objects and accounting result objects are determined. Based on the product process association, the process material association, the process emission association, the process equipment association, and the emission accounting association, data association relationships are established between different accounting association objects.

[0012] Optionally, in the above scheme, the aggregation of the standardized detailed data based on the data association relationship, according to product dimension, process dimension, and time dimension, to obtain summary data for copper product carbon footprint accounting includes: Based on the data association, the target product object, target process object and target time range corresponding to the data to be aggregated are determined to obtain the aggregation dimension information; Based on the aggregated dimension information, corresponding material consumption details, energy consumption details, equipment operation details, and emission factor details are extracted from the standardized detailed data to obtain the detailed data to be aggregated. Based on preset aggregation rules that include aggregation keys and aggregation functions, the detailed data to be aggregated is aggregated to obtain summary indicator data under the product process time dimension. Record the correspondence between the detailed data to be aggregated, the aggregation key, the aggregation function, and the summary indicator data to obtain the aggregation path information; Based on the aggregated index data and the aggregated path information, the aggregated data for the carbon footprint accounting of copper products is obtained.

[0013] Optionally, in the above scheme, the step of extracting activity data values ​​and emission factor values ​​from the standardized detailed data and / or the summarized data according to preset carbon footprint accounting rules to generate carbon footprint accounting parameter data includes: Based on the preset carbon footprint accounting rules, the accounting boundary, accounting process and accounting parameter fields corresponding to the target copper product are determined to obtain the accounting parameter requirement information. Based on the accounting parameter requirement information, the corresponding activity data values ​​are extracted from the standardized detailed data and / or the summary data to obtain the activity data parameter set; Based on the accounting parameter requirement information and the result of the unified coding mapping, the corresponding emission factor values ​​are matched from the standardized detailed data and / or the summary data to obtain the emission factor parameter set. Based on the activity data parameter set, the emission factor parameter set, and the corresponding data quality score, carbon footprint accounting parameter data is generated, which includes activity data values, emission factor values, carbon emission calculation reference values, and data quality scores.

[0014] Optionally, in the above scheme, the data lineage information recording from the multi-source raw data to the carbon footprint accounting parameter data includes: During the process of generating the source data from the multi-source raw data, the original data identifier, source system identifier, collection time and quality verification record are recorded to obtain the source data lineage information; During the process of generating the standardized detailed data from the source data, the field semantic alignment rules, missing data handling rules, unified encoding mapping rules and metadata annotation information are recorded to obtain the detailed governance lineage information. During the process of generating the summary data from the standardized detailed data, the aggregation dimension information, preset aggregation rules and aggregation path information are recorded to obtain the summary processing lineage information; During the process of generating the carbon footprint accounting parameter data from the standardized detailed data and / or the summarized data, the source of activity data values, the source of emission factor values, the source of data quality scores, and the correspondence between accounting parameter fields are recorded to obtain parameter generation lineage information; Based on the source data lineage information, the detailed governance lineage information, the summary processing lineage information, and the parameter generation lineage information, the data lineage information is obtained by summarizing the data to establish a traceability association between the carbon footprint accounting parameter data and the corresponding multi-source original data.

[0015] Secondly, a data management system for the carbon footprint of copper products, the system comprising: A multi-source data acquisition unit is used to acquire multi-source raw data related to the carbon footprint accounting of copper products. The multi-source raw data includes activity data, emission factor data and supply chain related data. The activity data includes at least two of the following: production process data, energy consumption data, material consumption data and equipment operation data. The source data layer processing unit is used to perform source data quality checks on the multi-source raw data based on the data source type, data structure characteristics and preset carbon footprint data quality rules. These checks include format compliance checks, business rationality checks and reference integrity checks. The unit then performs quality marking and diversion on the multi-source raw data according to the source data quality check results to obtain qualified source data and abnormal source data. The source layer storage unit is used to store the qualified source data in its original form according to the source system and collection time to obtain the source data, and to store the abnormal source data and the quality verification record corresponding to the abnormal source data. The detail layer processing unit is used to perform data cleaning, field semantic alignment, missing data processing, unified encoding mapping, and metadata annotation on the source data to obtain standardized detail data. The subject domain management unit is used to semantically classify the data at each level according to the copper product carbon footprint accounting related objects during the data flow process of the multi-source raw data, the source data, the standardized detailed data, the summary data and the carbon footprint accounting parameter data, and establish the data association relationship between different accounting related objects based on the semantic classification results. The copper product carbon footprint accounting related objects include product objects, process objects, material objects, energy emission objects, equipment objects and accounting result objects. The aggregation layer processing unit is used to aggregate the standardized detailed data according to the product dimension, process dimension and time dimension based on the data association relationship to obtain the aggregated data for copper product carbon footprint accounting. The application layer output unit is used to extract activity data values ​​and emission factor values ​​from the standardized detailed data and / or the summarized data according to the preset carbon footprint accounting rules, and generate carbon footprint accounting parameter data. The lineage tracing unit is used to record data lineage information from the multi-source raw data to the carbon footprint accounting parameter data. The data lineage information includes data input / output identifiers and data processing rule information to establish a traceability association between the carbon footprint accounting parameter data and the corresponding multi-source raw data.

[0016] Compared with the prior art, this application has at least the following beneficial effects: This application, based on further analysis and research into existing technical problems, recognizes that existing technologies in copper product carbon footprint data management suffer from inconsistencies in the definitions of multi-source data, difficulty in timely identification of the quality of basic accounting data, insufficient accuracy of cross-system data correlation, and difficulties in tracing carbon footprint accounting data. By acquiring multi-source raw data related to copper product carbon footprint accounting, this application enables production process data, energy consumption data, material consumption data, equipment operation data, emission factor data, and supply chain-related data to enter a unified data management process. Furthermore, based on data source type, data structure characteristics, and preset carbon footprint data quality rules, source data quality verification and quality labeling are performed to ensure that data that does not meet requirements for format compliance, business rationality, or citation integrity can be identified and processed differently from qualified data, thereby reducing the risk of abnormal data directly entering the accounting chain. Finally, qualified source data is stored in its original form, and data cleaning and field semantic matching are performed on the source data. The system employs methods such as incomplete and missing data processing, unified coding mapping, and metadata annotation to organize data from different sources, with different definitions, and different codes into standardized, detailed data that is associative, identifiable, and evaluable. Furthermore, during data flow, semantic classification of data at each level is performed according to the copper product carbon footprint accounting related objects, establishing data associations between different accounting related objects. This creates a data association foundation for carbon footprint accounting among products, processes, materials, energy emissions, equipment, and accounting results. Based on this, summary data is generated according to product, process, and time dimensions. Activity data values ​​and emission factor values ​​are extracted according to preset carbon footprint accounting rules to generate carbon footprint accounting parameter data, transforming the basic data required for accounting from scattered data into directly usable parameterized data. Simultaneously, the system records the data lineage information from multi-source raw data to carbon footprint accounting parameter data, enabling the carbon footprint accounting parameter data to be traced back to the corresponding raw data and processing rules. Therefore, this application can improve the problems of inconsistent data from multiple sources, difficulty in timely identification of the quality of basic accounting data, insufficient accuracy of cross-system data correlation, and difficulty in tracing carbon footprint accounting data in the management of copper product carbon footprint data, thereby improving the accuracy, reliability and traceability of copper product carbon footprint data management. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a data management method for the carbon footprint of copper products provided in one embodiment of this application; Figure 2 This is a block diagram of the module architecture of a data management system for the carbon footprint of copper products, provided as an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] In one embodiment, such as Figure 1 As shown, a data management method for the carbon footprint of copper products is provided. This method can be executed by a server, a data management platform, a carbon footprint data management system, or computing devices deployed in an enterprise data center. The copper products may include cathode copper, copper wire, copper rods, copper foil, copper pipes, and other copper products, as well as symbiotic products related to the copper product production process, such as sulfuric acid, slag concentrate, gold, silver, or crude cobalt hydroxide. The multi-source raw data associated with the copper product carbon footprint accounting can come from enterprise resource planning systems, manufacturing execution systems, industrial IoT platforms, energy management systems, supply chain collaboration systems, third-party life cycle assessment databases, standard databases, literature repositories, or manually imported data files.

[0020] This embodiment first acquires multi-source raw data related to the carbon footprint accounting of copper products. This multi-source raw data includes activity data, emission factor data, and supply chain-related data. Activity data can include at least two of the following: production process data, energy consumption data, material consumption data, and equipment operation data. Production process data can include process number, production batch, product output, process start and end time, product flow, etc.; energy consumption data can include electricity consumption, natural gas consumption, heavy oil consumption, anthracite coal consumption, steam consumption, etc.; material consumption data can include raw material input, auxiliary material input, solder input, cathode copper input, scrap volume, and loss volume, etc.; equipment operation data can include current, voltage, temperature, flue gas flow rate, sulfur dioxide concentration, equipment operating time, and equipment load status, etc. Emission factor data can include electricity emission factors, fuel emission factors, purchased material emission factors, and default emission factors obtained from national standards, industry standards, third-party life cycle assessment databases, or literature. Supply chain-related data can include supplier carbon footprint statements, transportation and logistics data, purchased raw material source data, and product environmental data provided by suppliers, etc.

[0021] After acquiring multi-source raw data, source data quality verification is performed based on the data source type, data structure characteristics, and preset carbon footprint data quality rules. Data source type can be used to distinguish between industrial IoT time-series data, production and operation structured data, supply chain external exchange data, emission factor reference data, etc.; data structure characteristics can be used to distinguish between structured data, semi-structured data, unstructured data, and time-series data; preset carbon footprint data quality rules can include field format rules, required field rules, timestamp rules, coding rules, process threshold rules, material balance rules, and referential integrity rules. Source data quality verification includes format compliance verification, business rationality verification, and referential integrity verification. Format compliance verification determines whether field types, field lengths, timestamps, required fields, and duplicate records meet preset requirements; business rationality verification determines whether material input-output relationships, equipment operating parameters, and process operating parameters are within the reasonable range of the copper product production process; referential integrity verification determines whether material codes, process codes, equipment codes, and energy codes can be found in the corresponding standard coding tables or dimension tables.

[0022] Based on the source data quality verification results, the multi-source raw data is divided into qualified and abnormal source data by quality marking. Data that meets the verification pass conditions is marked as qualified and designated as qualified source data; data that does not meet at least one verification rule is marked as abnormal and designated as abnormal source data. For abnormal source data, information such as the abnormal type, abnormal fields, abnormal triggering rules, quality verification time, and original data snapshot can be further recorded to facilitate subsequent governance personnel in confirming, correcting, or reprocessing the causes of the abnormalities.

[0023] Subsequently, qualified source data is stored in its original form according to the source system and collection time, resulting in patched source data. This original form storage means that no transformations affecting business semantics are applied to the qualified source data during the patched source storage stage, preserving as much of the original fields, original codes, original timestamps, and original data formats as possible when the data entered the system. Specifically, partitions can be established according to the source system code and collection date. For example, patched source data partitions can be established according to source systems such as Enterprise Resource Planning (ERP), Manufacturing Execution System (MRP), Industrial Internet of Things (IIoT) platforms, and Energy Management Systems. Storage directories or data tables can then be further divided according to date, batch, or collection window. For abnormal source data, the abnormal source data and its corresponding quality verification records are stored, forming an abnormal data processing record.

[0024] After obtaining the source data, the data undergoes data cleaning, field semantic alignment, missing data handling, unified coding mapping, and metadata annotation to obtain standardized detailed data. Data cleaning may include removing duplicate records, eliminating obvious erroneous values, correcting format outliers, and processing invalid fields. Field semantic alignment may unify data fields with the same meaning but different field names in different systems to the same semantic granularity. For example, it may confirm the correspondence between fields representing power consumption, process power consumption, and equipment power consumption in different systems, and form fields that can be independently calculated according to the needs of carbon footprint accounting. Missing data handling can determine the processing method based on the missing data type, missing proportion, and data source. For example, it may use adjacent time window interpolation, batch mean interpolation, process statistical value interpolation, or manual confirmation, and add missing data handling labels to the processed fields. Unified coding mapping can map product codes, process codes, material codes, energy codes, and equipment codes from different systems to unified codes. Metadata annotation can add information such as data source, update time, update frequency, field meaning, quality score, and processing status to data tables and fields.

[0025] In this embodiment, during the data flow process of multi-source raw data, source-attached data, standardized detailed data, summary data, and carbon footprint accounting parameter data, the data at each level is semantically categorized according to the carbon footprint accounting related objects of copper products, and data association relationships are established between different accounting related objects based on the semantic classification results. The carbon footprint accounting related objects of copper products include product objects, process objects, material objects, energy emission objects, equipment objects, and accounting result objects. Product objects represent products such as cathode copper, copper wire, copper foil, copper rods, and copper pipes; process objects represent processes involved in copper product manufacturing such as smelting, casting, rolling, extrusion, heat treatment, surface treatment, testing, and packaging; material objects represent raw materials, auxiliary materials, solder, fuel, and purchased materials; energy emission objects represent electricity consumption, natural gas consumption, fuel consumption, carbon dioxide emissions, and emission factors; equipment objects represent key equipment such as flash furnaces, converters, electrolytic cells, induction furnaces, extruders, wire drawing machines, and rolling mills; and accounting result objects represent outcome data such as product carbon footprint factors, process carbon emission intensity, and data quality scores. Through semantic classification throughout the data flow process at each level, data originally scattered across different business systems and processing levels can be organized according to the object relationships required for copper product carbon footprint accounting.

[0026] After establishing data relationships between different accounting objects, standardized detailed data is aggregated based on these relationships, according to product, process, and time dimensions, to obtain summary data for copper product carbon footprint accounting. For example, summary data on process carbon emission intensity can be generated by product code, process code, and month; summary data on equipment energy efficiency evaluation can be generated by equipment code and week; and summary data required for product carbon footprint accounting can be generated by product code, process chain code, and accounting cycle. During the aggregation process, the correspondence between aggregation keys, aggregation functions, aggregation time ranges, input detailed data, and output summary data can be recorded for subsequent traceability.

[0027] Based on preset carbon footprint accounting rules, activity data values ​​and emission factor values ​​are extracted from standardized detailed data and / or aggregated data to generate carbon footprint accounting parameter data. The preset carbon footprint accounting rules may include product accounting boundaries, process boundaries, time boundaries, data priorities, emission factor matching rules, co-existing product allocation rules, and data quality evaluation rules. The carbon footprint accounting parameter data may include fields such as target product, target process, activity data values, emission factor values, carbon emission calculation reference values, data quality score, and data source identifier. This carbon footprint accounting parameter data can serve as the data foundation for product carbon footprint accounting, carbon emission report generation, third-party verification, and internal enterprise analysis.

[0028] Finally, the data lineage information from multi-source raw data to carbon footprint accounting parameter data is recorded. This lineage information includes data input / output identifiers and data processing rule information. Specifically, during the process of data moving from multi-source raw data to source data, from source data to standardized detailed data, from standardized detailed data to summary data, and from standardized detailed data and / or summary data to carbon footprint accounting parameter data, the corresponding input tables, output tables, field mappings, processing rules, aggregation rules, execution time, quality verification records, and data source identifiers are recorded. Therefore, when it is necessary to trace the source of a specific carbon footprint accounting parameter, one can trace back along the data lineage information from the carbon footprint accounting parameter data to the summary data, standardized detailed data, source data, and the corresponding multi-source raw data.

[0029] This embodiment enables the quality identification, original storage, standardization, semantic association, aggregation, and lineage tracing of data scattered across different systems, formats, and definitions during the copper product carbon footprint data management process. Because source data undergoes format compliance verification, business rationality verification, and reference integrity verification upon entry, the risk of abnormal data entering subsequent accounting processes is reduced. By establishing data associations between different accounting-related objects through unified coding mapping and semantic classification across all levels, the accuracy of cross-system data association is improved. Furthermore, recording the data lineage information from multi-source raw data to carbon footprint accounting parameter data enhances the traceability of carbon footprint accounting data. Therefore, this method can address the problems of scattered sources, inconsistent definitions, difficulty in timely data quality identification, unstable cross-system data association, and difficulties in tracing accounting data in the copper product carbon footprint data management process.

[0030] In some embodiments, when acquiring multi-source raw data associated with copper product carbon footprint accounting, the source attributes, update frequency, and data structure information of candidate data sources can be obtained first to acquire data source characteristic information. Source attributes may include the system to which the data source belongs, system interface type, data maintenance entity, data trust level, etc.; update frequency may include real-time updates, minute-level updates, hourly updates, daily updates, monthly updates, or irregular updates; data structure information may include structured tables, time-series data, semi-structured files, interface messages, spreadsheets, document files, etc.

[0031] Based on data source characteristics, candidate data sources are categorized into activity data sources and emission factor data sources. Activity data sources include at least two of the following: industrial IoT time-series data sources, production and operation structured data sources, and supply chain external exchange data sources. Industrial IoT time-series data sources can include sensor data sources distributed on key equipment such as flash furnaces, electrolytic cells, induction furnaces, extruders, and rolling mills, providing high-frequency equipment operating parameters such as current, voltage, temperature, flue gas flow rate, and gas concentration. Production and operation structured data sources can include enterprise resource planning systems, manufacturing execution systems, and energy management systems, providing structured data such as material consumption records, product output records, production batch records, and energy procurement records. Supply chain external exchange data sources can include supplier systems, logistics systems, external data interfaces, or data files submitted by suppliers, providing supplier carbon footprint statements, transportation distances, transportation methods, and environmental data of purchased products. Emission factor data sources include at least one of the following: external databases, standard data sources, and literature reference data sources. External databases can be life cycle assessment databases; standard data sources can be emission factor data from national standards, industry standards, or enterprise standards; and literature reference data sources can be process parameter reference values ​​from academic literature or technical documents.

[0032] Based on the determination of corresponding data access strategies for activity data sources and emission factor data sources, a multi-source data access strategy is obtained. For industrial IoT time-series data sources, real-time or near-real-time data acquisition can be achieved using open platform communication architecture, industrial Ethernet communication protocols, message queue telemetry transmission protocols, etc., with noise reduction filtering, outlier removal, and window aggregation performed at the edge gateway. For structured data sources from production operations, data access can be achieved using database change data capture, timed batch extraction, or database connection methods. For external supply chain exchange data sources, access can be achieved using interface calls, file transfer, timed polling, or event-driven methods. For emission factor data sources, access can be achieved using manual import, periodic synchronization, interface queries, or standard update monitoring methods.

[0033] The initial data is collected from candidate data sources based on a multi-source data access strategy to obtain data related to the carbon footprint accounting of copper products. For example, second- or millisecond-level industrial IoT data can be aggregated at the edge into minute-level statistical data before uploading; incremental capture can be performed on material consumption records from manufacturing execution systems; energy procurement data from enterprise resource planning systems can be extracted periodically; and emission factor data can be regularly maintained and version-recorded. Subsequently, the initial data is preprocessed to obtain multi-source raw data. Preprocessing may include interface message parsing, file format conversion, time-series data windowing, initial field extraction, source identifier addition, collection time recording, and basic anomaly filtering.

[0034] This embodiment employs differentiated data access strategies based on the source attributes, update frequency, and data structure information of different data sources, rather than using a single data import method to process all data sources. Since industrial IoT time-series data, production and operation structured data, supply chain external exchange data, and emission factor data differ in terms of timeliness, data format, and reliability, categorized access and preprocessing can improve the adaptability and completeness of multi-source raw data collection, providing a reliable data foundation for subsequent quality verification, standardized governance, and carbon footprint accounting parameter generation.

[0035] In some embodiments, when performing source data quality verification on multi-source raw data, the process first involves verifying field types, required fields, timestamp formats, and duplicate records based on data structure characteristics, yielding a format compliance verification result. Field type verification checks whether numeric fields are valid values, character fields conform to preset character sets and length requirements, and date fields are valid dates or timestamps. Required field verification checks whether key fields required for carbon footprint accounting are missing, such as whether fuel consumption records include process codes, fuel codes, consumption amounts, and timestamps, and whether material consumption records include material codes, batches, input amounts, and process codes. Timestamp format verification converts time fields from different sources into a unified time format and checks whether time fields are missing or exceed a reasonable range. Duplicate record verification determines the existence of duplicate records based on business primary keys, process codes, collection times, and source system identifiers.

[0036] Then, based on the data source type, the data to be audited is determined from the data that has passed the format compliance check, resulting in the data set to be audited. The data to be audited can include material consumption data, product output data, energy consumption data, and equipment operation data—data suitable for assessing business rationality. For emission factor data, supplier declaration data, or standard reference data, processing can be performed according to their data type using rules based on version validity, source credibility, or field integrity, without necessarily requiring mandatory material input-output consistency checks.

[0037] Based on preset carbon footprint data quality rules, the data set to be audited for business operations undergoes material input-output consistency verification and / or process parameter range verification to obtain business rationality verification results. Material input-output consistency verification can be based on the proportion of differences between the input, output, scrap, and process loss amounts in a specific process or batch. When the proportion of differences exceeds a preset threshold, the batch of data is deemed to have an abnormal material balance. Process parameter range verification can be based on the reasonable operating range of key parameters in copper smelting, copper processing, or electrolytic refining processes. For example, if current, voltage, temperature, flue gas flow rate, fuel consumption intensity, or equipment load exceed preset ranges, the process parameters are deemed abnormal.

[0038] Furthermore, based on the preset coding dimension table, referential integrity checks are performed on the material codes, process codes, and equipment codes in the data that have passed the format compliance check, yielding the referential integrity check results. The preset coding dimension table can include product dimension tables, process dimension tables, material dimension tables, energy dimension tables, and equipment dimension tables. Referential integrity checks are used to determine whether the codes appearing in the original data have been registered in the corresponding dimension tables. If a record references an unregistered temporary auxiliary material code, an undefined equipment code, or an invalid process code, the record can be marked as having incomplete references.

[0039] Based on the results of format compliance verification, business rationality verification, and reference integrity verification, the source data quality verification results are determined. Then, based on these results, the multi-source raw data is graded and categorized into qualified and abnormal source data. Specifically, for data that is format compliant, business rational, and has complete references, a source data quality verification result indicating that the verification has passed is generated; for data that fails any verification item, a source data quality verification result indicating that the verification has failed is generated, and the corresponding reason for the failure is recorded.

[0040] This embodiment of the solution goes beyond general data cleaning techniques such as null value detection, duplicate detection, and format checks. It further integrates the business characteristics of copper product carbon footprint data to identify material input-output relationships, process parameter ranges, and the integrity of coded references. Since different types of data can be determined beforehand to be subject to business auditing, and then the corresponding business rationality verification rules are applied, it avoids forcibly applying mismatched verification rules to inapplicable data sources. Therefore, this method improves the accuracy of source data quality identification and reduces the possibility of abnormal data and incorrectly referenced data entering the subsequent carbon footprint accounting data chain.

[0041] In some embodiments, when performing quality marking and distribution on multi-source raw data based on the source data quality verification results, when the format compliance verification result, business rationality verification result, and reference integrity verification result all meet the corresponding verification pass conditions, a qualified mark is added to the corresponding multi-source raw data to obtain qualified source data. The qualified mark may include information such as qualified status, verification pass time, verification rule version, and source system identifier, so that the corresponding quality status can be retained during subsequent source pasting storage.

[0042] When at least one of the format compliance check result, business rationality check result, and reference integrity check result fails to meet the corresponding pass condition, an anomaly flag is added to the corresponding multi-source raw data to obtain the anomaly source data. Anomaly flags can be used to distinguish between anomalies such as format anomalies, missing required fields, timestamp anomalies, duplicate records, material balance anomalies, process parameter anomalies, and incomplete references. For cases where the same data triggers multiple anomaly rules simultaneously, multiple anomaly flags can be recorded and sorted according to anomaly severity or processing priority.

[0043] Based on the anomaly type, anomaly fields, anomaly triggering rules, and quality verification time of the anomaly source data, a quality verification record is generated. The quality verification record can include information such as the source system of the anomaly data, business primary key, anomaly field name, anomaly field value, triggered verification rule, rule threshold, actual verification result, quality verification time, and processing status. For material input-output consistency anomalies, the record can record the input quantity, output quantity, scrap quantity, loss quantity, and difference ratio; for process parameter range anomalies, the record can record the corresponding parameter value, preset range, and out-of-limit type; for referential integrity anomalies, the record can record the coded value of the dimension table record that was not matched.

[0044] An anomaly data processing record is obtained by associating and storing the original data snapshot of the anomaly source data with the quality verification record. The original data snapshot is used to save the original state of the anomaly source data when it enters the system, while the quality verification record is used to explain the anomaly identification process and the cause of the anomaly. The anomaly data processing record can be stored in the anomaly area and anomaly alerts can be pushed to data governance personnel, carbon management personnel, or system administrators. Governance personnel can use the anomaly data processing record to perform data repair, dimension table supplementation, anomaly confirmation, or re-verification. For records that are confirmed to be due to actual operating conditions rather than data errors, a confirmation status can be added, and the data can proceed to the subsequent processing flow after meeting preset release conditions.

[0045] This embodiment enables the separate processing of qualified and abnormal data, preventing data that fails quality verification from directly entering the formal data processing chain. Simultaneously, abnormal source data is stored as a snapshot of the original data and quality verification records, preserving the basis for anomaly identification and subsequent processing. This method helps prevent the spread of low-quality data and facilitates the review, repair, and traceability of abnormal data, thereby improving the reliability of copper product carbon footprint data management.

[0046] In some embodiments, when performing data cleaning, field semantic alignment, missing data handling, unified encoding mapping, and metadata annotation on the source data, outlier cleanup and deduplication cleanup are first performed on the source data to obtain cleaned data. Outlier cleanup may include removing data that is clearly outside the physical range or equipment operating range, correcting outliers caused by unit conversion errors or format errors, or marking uncorrectable outliers as pending processing. Deduplication cleanup can identify duplicate records based on fields such as source system, business primary key, collection time, batch number, and process code, and determine the records to retain according to data quality score, update time, or source priority.

[0047] Subsequently, semantic alignment is performed on fields from different sources but with the same meaning in the cleaned data to obtain semantically aligned data. For example, if a field is named "Total Power Consumption" in one system, "Process Power Consumption" in another, and "Equipment Power Consumption" in yet another, their correspondence can be determined based on the field meaning, data acquisition equipment, process scope, and time granularity, and then aligned to a unified power consumption semantic field. In cases where there are differences in the definition of total power consumption and process power consumption, they can be decomposed into independently calculable sub-items based on equipment boundaries, process boundaries, and auxiliary system boundaries.

[0048] Furthermore, missing fields in the semantically aligned data undergo missing data processing, and missing data processing identifiers are added to the processed fields to obtain completed data. Missing data processing can be performed in different ways depending on the data type. For short-term missing industrial IoT time-series data, linear interpolation, averaging of adjacent windows, or historical data under the same operating conditions can be used for completion. For a small number of missing production and operation fields, data from the same batch, process, or product can be used for completion. For data with missing critical fields that cannot be automatically completed, the missing identifier can be retained and a manual confirmation process can be initiated. The missing data processing identifier can record the missing field, the missing time range, the processing method, and the version of the processing rule.

[0049] Based on a pre-defined standard coding system, product codes, process codes, material codes, energy codes, and equipment codes in the completed tag data are uniformly mapped to obtain standardized coding data. The pre-defined standard coding system can include product standard codes, process standard codes, material standard codes, energy standard codes, and equipment standard codes. For situations where the same fuel uses different names or language descriptions in different systems, they can be uniformly mapped to a standard energy code. For example, descriptions such as "anthracite" and "Anthracite" can be uniformly mapped to the same fuel code. Fuel categories can be represented using a combination of classification codes and serial numbers; for example, solid fuels, liquid fuels, and gaseous fuels can use different code segments. For process codes, processes with different names in different systems can be mapped to standard process nodes in a unified process chain.

[0050] Finally, data source, update time, data quality score, and field meaning information are added to the standardized data to obtain standardized detailed data. The data quality score can be determined based on factors such as data source reliability, quality verification results, missing data handling, imputation ratio, anomaly handling status, and field completeness. Field meaning information can be used to describe the business semantics, unit, data type, and applicable accounting scenarios of the fields. Standardized detailed data can serve as the foundation for subsequent summary processing and parameter generation, and as an important data basis for semantic classification and object association maintenance in subject domain management units.

[0051] This embodiment can further transform the original source data into standardized detailed data that is semantically clear, uniformly encoded, and of evaluable quality. Due to data cleaning, field semantic alignment, missing data handling, and unified encoding mapping, data association errors caused by inconsistent field definitions, missing data, and cross-system encoding inconsistencies can be reduced. Furthermore, the addition of metadata annotations and data quality scores ensures that subsequent summary processing and calculation parameters have interpretable data quality criteria.

[0052] In some embodiments, during the data flow process of multi-source raw data, source-attached data, standardized detailed data, summary data, and carbon footprint accounting parameter data, when semantically classifying the data at each level according to the copper product carbon footprint accounting related objects, product identifiers, process identifiers, material identifiers, energy emission data identifiers, equipment identifiers, and accounting result data identifiers can be identified from the data at each level to obtain a set of accounting object identifiers. Product identifiers can be used to indicate target products such as cathode copper, copper wire, copper rods, and copper foil; process identifiers can be used to indicate processes such as smelting, billet casting, rolling, extrusion, heat treatment, surface treatment, and testing and packaging; material identifiers can be used to indicate input materials such as raw materials, auxiliary materials, solder, and fuel; energy emission data identifiers can be used to indicate electricity consumption, natural gas consumption, fuel consumption, carbon dioxide emissions, or emission factors; equipment identifiers can be used to indicate equipment such as flash furnaces, converters, electrolytic cells, induction furnaces, extruders, wire drawing machines, and rolling mills; and accounting result data identifiers can be used to indicate result data such as product carbon footprint factors, process carbon emission intensity, or data quality scores.

[0053] Based on the set of accounting object identifiers, data at each level are categorized into product objects, process objects, material objects, energy emission objects, equipment objects, and accounting result objects, resulting in semantic classification. Product objects can be used to organize data such as product name, product code, specifications, product batch, and product output; process objects can be used to organize data such as process code, process name, process sequence, process input, and process output; material objects can be used to organize data such as material code, material name, input quantity, loss quantity, and material source; energy emission objects can be used to organize data such as energy type, consumption, emission factor, emission quantity, and emission range; equipment objects can be used to organize data such as equipment code, equipment name, equipment operating parameters, equipment process affiliation, and equipment energy consumption; and accounting result objects can be used to organize data such as product carbon footprint factor, process carbon emission intensity, co-existing product allocation results, and data quality score.

[0054] Based on semantic classification results, the following relationships are determined: product-process relationships between product objects and process objects; process-material relationships between process objects and material objects; process-emission relationships between process objects and energy emission objects; process-equipment relationships between process objects and equipment objects; and emission accounting relationships between energy emission objects and accounting result objects. Product-process relationships can represent the production process chain corresponding to a product; process-material relationships can represent the raw materials, auxiliary materials, or fuel consumed in a process; process-emission relationships can represent the energy consumption, emission factors, and emissions corresponding to a process; process-equipment relationships can represent which equipment performs a process and how equipment operation data is attributed to that process; and emission accounting relationships can represent how energy consumption data and emission factor data form accounting result data.

[0055] Based on product process relationships, process material relationships, process emission relationships, process equipment relationships, and emission accounting relationships, data relationships between different accounting-related objects are established. These data relationships can be implemented using relational tables, graph structures, association indexes, topic views, or data service interfaces. Taking copper rod products as an example, a network can be formed where the product object is copper rod, the process objects are smelting to billet casting to rolling and testing / packaging, the material objects are cathode copper input and auxiliary material input, the energy emission objects are power consumption of the induction furnace, power consumption of the rolling mill, natural gas consumption and corresponding emission factors, the equipment objects are operating parameters of the induction furnace and rolling mill, and the accounting result objects are the carbon footprint factor of the copper rod product and the carbon emission intensity distribution of each process.

[0056] This embodiment organizes standardized detailed data according to the object relationships required for copper product carbon footprint accounting, rather than storing it in isolation according to the data source system. By establishing data relationships between products, processes, materials, energy emissions, equipment, and accounting results, accounting personnel or subsequent procedures can obtain a complete data view along the product carbon footprint chain, thereby improving the accuracy of cross-system data association and the efficiency of organizing copper product carbon footprint accounting data.

[0057] In some embodiments, when performing aggregation processing based on data relationships, the target product object, target process object, and target time range corresponding to the data to be aggregated can be determined first based on the data relationships to obtain aggregation dimension information. The target product object can be cathode copper, copper wire, copper rod, etc.; the target process object can be electrolytic refining, smelting, rolling, extrusion, etc.; the target time range can be daily, weekly, monthly, batch cycle, accounting cycle, or user-specified time period. Aggregation dimension information can include product dimension, process dimension, time dimension, and necessary equipment dimension, energy type dimension, or material type dimension.

[0058] Based on aggregated dimensional information, corresponding detailed data on material consumption, energy consumption, equipment operation, and emission factors are extracted from standardized detailed data to obtain the detailed data to be aggregated. For example, when it is necessary to generate summary data for cathode copper in a certain month's electrolytic refining process, detailed data on electricity consumption, material input, equipment operation, and corresponding emission factors for that product, process, and month can be extracted. For equipment energy efficiency ranking, the runtime, output, electricity consumption, and related process data of the target equipment in the target week can be extracted.

[0059] Based on preset aggregation rules containing aggregation keys and aggregation functions, the detailed data to be aggregated is processed to obtain summary indicator data under the product process time dimension. Aggregation keys can include product code, process code, time period, equipment code, energy code, etc.; aggregation functions can include summation, averaging, maximum value, minimum value, weighted average, grouped statistics, proportional calculation, etc. For summarizing process carbon emission intensity, the process code and month can be used as aggregation keys to process energy consumption, material consumption, and corresponding emission factors; for equipment energy efficiency evaluation, the equipment code and week can be used as aggregation keys to aggregate equipment output, power consumption, and operating time.

[0060] The process records the correspondence between the detailed data to be aggregated, the aggregation key, the aggregation function, and the summarized indicator data to obtain the aggregation path information. The aggregation path information may include the input detail table, input fields, aggregation dimensions, aggregation functions, output summary table, output fields, and execution time. Based on the summarized indicator data and the aggregation path information, summarized data for copper product carbon footprint accounting is obtained. This summarized data includes both the summarized indicator results and the aggregation path information that can be used to trace the generation process of these results.

[0061] This embodiment can perform light aggregation of standardized detailed data for carbon footprint accounting according to product, process, and time dimensions, forming summary data that can be directly used for subsequent parameter extraction, analysis, display, or accounting services. Because the aggregation key, aggregation function, and aggregation path information are recorded during the aggregation process, not only is the efficiency of generating summary data improved, but the correspondence between the summary data and the underlying detailed data is also preserved, thus providing a basis for subsequent data traceability and audit verification.

[0062] In some embodiments, when generating carbon footprint accounting parameter data, the accounting boundaries, accounting processes, and accounting parameter fields corresponding to the target copper product can be determined first based on preset carbon footprint accounting rules to obtain the accounting parameter requirement information. The accounting boundaries can include the boundary ranges of raw material acquisition, production and processing, and product delivery, or can be set as specific process boundaries or specific product batch boundaries according to actual needs. Accounting processes can include smelting, billet casting, electrolytic refining, rolling, extrusion, heat treatment, surface treatment, inspection, and packaging. Accounting parameter fields can include product code, process code, activity data value, activity data unit, emission factor value, emission factor source, carbon emission calculation reference value, and data quality score.

[0063] Based on the accounting parameter requirements, corresponding activity data values ​​are extracted from standardized detailed data and / or aggregated data to obtain an activity data parameter set. Activity data values ​​may include product output, raw material input, auxiliary material consumption, electricity consumption, natural gas consumption, fuel consumption, equipment operating time, transportation distance, etc. For processes requiring detailed accounting, detailed-level activity data can be extracted from standardized detailed data; for scenarios requiring monthly, weekly, or batch-based accounting, aggregated activity data can be extracted from aggregated data.

[0064] Based on the accounting parameter requirements and the results of the unified coding mapping, corresponding emission factor values ​​are matched from standardized detailed data and / or aggregated data to obtain the emission factor parameter set. For example, emission factors for electricity, natural gas, anthracite, and heavy oil can be matched based on energy codes; emission factors for purchased materials can be matched based on material codes; and transportation emission factors can be matched based on transportation mode and distance. Since the unified coding mapping has been completed in the preceding steps, emission factor matching errors caused by inconsistent names of the same material or fuel in different systems can be reduced.

[0065] Based on the activity data parameter set, emission factor parameter set, and corresponding data quality score, carbon footprint accounting parameter data is generated, including activity data values, emission factor values, carbon emission calculation reference values, and data quality scores. The carbon emission calculation reference values ​​can be generated based on the activity data values ​​and emission factor values, providing a reference for subsequent formal accounting, report generation, or manual review. For production processes involving co-existing products, relevant activity data, emission factors, and calculation reference values ​​can be associated with the corresponding co-existing product allocation results according to preset allocation rules. The carbon footprint accounting parameter data can be provided to subsequent accounting tools, carbon management platforms, or auditing systems in the form of parameter wide tables, data interfaces, data marts, or file output.

[0066] This embodiment can further assemble standardized detailed data and summary data into parameter data that can directly support carbon footprint accounting based on preset carbon footprint accounting rules. Since both activity data values ​​and emission factor values ​​are extracted and matched based on unified coding mapping results, and the parameter data includes data quality scores, the accuracy of parameter matching and the reliability of the accounting basis data can be improved. Therefore, this method can reduce errors caused by manual aggregation, manual matching of emission factors, and manual judgment of data quality, and improve the standardization and efficiency of the copper product carbon footprint accounting preparation process.

[0067] In some embodiments, when recording data lineage information from multi-source raw data to carbon footprint accounting parameter data, the source data lineage information can be obtained by recording the original data identifier, source system identifier, collection time, and quality verification record during the process of generating source-attached data from multi-source raw data. The original data identifier can be the original record primary key, data file identifier, interface message number, or time-series data window number; the source system identifier can indicate that the data comes from an enterprise resource planning system, manufacturing execution system, industrial IoT platform, supply chain system, or external database; the quality verification record can include format compliance verification results, business rationality verification results, and reference integrity verification results.

[0068] In the process of generating standardized detailed data from source data, field semantic alignment rules, missing data handling rules, unified encoding mapping rules, and metadata annotation information are recorded to obtain detailed governance lineage information. Field semantic alignment rules are used to explain how fields from different sources are mapped to unified fields; missing data handling rules are used to explain the handling methods, processing parameters, and processing identifiers for missing fields; unified encoding mapping rules are used to explain the mapping relationship between the original encoding and the standard encoding; metadata annotation information is used to explain the data source, update time, quality score, and field meaning, etc.

[0069] In the process of generating summary data from standardized detailed data, aggregation dimension information, preset aggregation rules, and aggregation path information are recorded to obtain summary processing lineage information. Aggregation dimension information may include product dimensions, process dimensions, time dimensions, etc.; preset aggregation rules may include aggregation keys and aggregation functions; aggregation path information can explain which detailed data participated in the summary processing and how the summary indicator data was formed.

[0070] During the process of generating carbon footprint accounting parameter data from standardized detailed data and / or aggregated data, the sources of activity data values, emission factor values, data quality scores, and the correspondence between accounting parameter fields are recorded to obtain parameter generation lineage information. The source of activity data values ​​indicates which detailed or aggregated data the activity data in the parameter data originates from; the source of emission factor values ​​indicates whether the emission factors come from standard data sources, external databases, literature references, or manually maintained data; the source of data quality scores explains the basis for calculating the quality scores; and the correspondence between accounting parameter fields explains the relationship between the fields in the parameter wide table and the underlying data fields.

[0071] Based on source data lineage information, detailed governance lineage information, aggregated processing lineage information, and parameter-generated lineage information, a data lineage information is obtained to establish a traceability link between carbon footprint accounting parameter data and corresponding multi-source raw data. When it is necessary to trace the source of carbon emission data for a specific copper product, a specific process, and a specific time period, one can trace down from the carbon footprint accounting parameter data to the aggregated data, then down to the standardized detailed data, source-attached data, and multi-source raw data, while simultaneously viewing the corresponding quality verification records, field mapping rules, coding mapping rules, and aggregation rules. This traceability path can provide data support for third-party verification, internal audits, compliance reporting, and anomaly review.

[0072] This embodiment of the method can uniformly record key processing information in the data acquisition, quality verification, detailed governance, summary processing, and parameter generation processes as data lineage information. Since data lineage information records not only the data source but also processing rules, mapping rules, aggregation rules, and quality records, it can form a complete traceability chain from carbon footprint accounting parameter data to multi-source raw data. This method can improve the problems of unclear data sources, difficult-to-explain processing procedures, and insufficient audit and verification evidence in carbon footprint accounting.

[0073] In one embodiment, such as Figure 2 As shown, a data management system for the carbon footprint of copper products is provided. The system includes a multi-source data acquisition unit, a source data layer processing unit, a source-attached layer storage unit, a detail layer processing unit, a summary layer processing unit, an application layer output unit, a subject domain management unit, and a lineage tracing unit. The subject domain management unit and the lineage tracing unit are both located outside the main data processing chain and are integrated throughout the data flow process at each level. Each unit can be deployed on the same server or distributed across different servers, data nodes, or cloud platform instances. Data interaction between units can occur through databases, message queues, interface services, or data processing tasks.

[0074] The multi-source data acquisition unit is used to acquire multi-source raw data related to the carbon footprint accounting of copper products. This unit can connect to industrial IoT platforms, enterprise resource planning (ERP) systems, manufacturing execution systems (MES), energy management systems (EMS), supply chain systems, external databases, and manual import interfaces. For industrial IoT time-series data, the multi-source data acquisition unit can receive device sensor data via industrial communication protocols or message transmission protocols, and supports noise reduction filtering, outlier removal, and window aggregation at the edge gateway side. For structured production and operation data, the multi-source data acquisition unit can acquire material consumption, product output, and energy procurement data through database change data capture, periodic extraction, or interface calls. For emission factor data and supply chain-related data, the multi-source data acquisition unit can acquire them through file import, interface queries, periodic synchronization, or manual maintenance.

[0075] The source data layer processing unit performs source data quality verification on multi-source raw data based on the data source type, data structure characteristics, and preset carbon footprint data quality rules. Based on the source data quality verification results, it performs quality marking and streaming of the multi-source raw data to obtain qualified and abnormal source data. The source data layer processing unit can be configured with format compliance verification rules, business rationality verification rules, and reference integrity verification rules, and can generate quality verification records and abnormal data processing records.

[0076] The source-attachment layer storage unit is used to store qualified source data in its original form according to the source system and acquisition time, resulting in attached source data. It also stores abnormal source data and the corresponding quality verification records. The source-attachment layer storage unit can be partitioned according to source system code, acquisition date, data type, or batch, preserving the original fields, original codes, original formats, and original semantics of qualified source data when it entered the system. For abnormal source data, the source-attachment layer storage unit can use a method of associating original data snapshots with quality verification records to form an abnormal data area.

[0077] The detail layer processing unit performs data cleaning, field semantic alignment, missing data handling, unified encoding mapping, and metadata annotation on the source data to obtain standardized detail data. This unit can maintain a standard encoding system, field semantic mapping rules, missing data handling rules, and data quality scoring rules. Through this unit, heteronymous fields, heteronymous codes, and data of different granularities from different systems can be transformed into detailed data with unified semantics and unified encoding.

[0078] The subject domain management unit is not a sequential processing module within a single data processing layer. Instead, it is used to semantically categorize data at each level according to the copper product carbon footprint accounting related objects during the data flow process between the multi-source data acquisition unit, source data layer processing unit, source-attached layer storage unit, detail layer processing unit, summary layer processing unit, and application layer output unit. Based on the semantic classification results, it establishes data association relationships between different accounting related objects. The subject domain management unit can maintain product objects, process objects, material objects, energy emission objects, equipment objects, and accounting result objects, as well as the relationships between these objects. Through the subject domain management unit, an association network can be formed between product domains, process domains, material domains, energy emission domains, equipment domains, and result domains, enabling cross-object, cross-process, and cross-system association queries of data on the same product carbon footprint chain.

[0079] The aggregation layer processing unit is used to aggregate standardized detailed data based on data relationships, according to product, process, and time dimensions, to obtain aggregated data for copper product carbon footprint accounting. This unit can be configured with aggregation keys, aggregation functions, aggregation periods, and output indicator fields, and records aggregation path information. Through this unit, aggregated data on process carbon emission intensity, equipment energy efficiency, and product process cycle can be generated.

[0080] The application layer output unit extracts activity data values ​​and emission factor values ​​from standardized detailed data and / or summarized data according to preset carbon footprint accounting rules, generating carbon footprint accounting parameter data. The application layer output unit can generate a wide table of carbon footprint accounting parameters according to product, process, time range, and accounting boundary, and provide data services to carbon accounting tools, carbon management platforms, reporting systems, or auditing systems. The carbon footprint accounting parameter data may include activity data values, emission factor values, carbon emission calculation reference values, data quality scores, and data source identifiers, etc.

[0081] The lineage tracing unit records data lineage information from multi-source raw data to carbon footprint accounting parameter data. It records source data lineage information, detailed governance lineage information, summary processing lineage information, and parameter generation lineage information, establishing a traceability link between carbon footprint accounting parameter data and multi-source raw data. Furthermore, when a user queries the source of a specific carbon footprint accounting parameter, the lineage tracing unit can return the corresponding raw data source, quality verification records, field mapping rules, encoding mapping rules, aggregation rules, and parameter generation rules.

[0082] This system utilizes various functional units to collaboratively complete the collection, quality verification, original storage, detailed management, semantic association across all levels, summary processing, parameter output, and lineage tracing of copper product carbon footprint data. Because the system performs quality identification upon data entry, unified coding management at the detailed level, and establishes data associations between different accounting objects across all levels through the subject domain management unit, and records the complete data flow process through the lineage tracing unit, it can improve the standardization, accuracy, and traceability of copper product carbon footprint data management. This reduces the impact of inconsistent data definitions from multiple sources, difficulty in identifying abnormal data, inaccurate cross-system associations, and difficulty in tracing the source of accounting parameters on carbon footprint accounting.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A data management method for the carbon footprint of copper products, characterized in that, The method includes: Obtain multi-source raw data related to the carbon footprint accounting of copper products. The multi-source raw data includes activity data, emission factor data, and supply chain related data. The activity data includes at least two of the following: production process data, energy consumption data, material consumption data, and equipment operation data. Based on the data source type, data structure characteristics, and preset carbon footprint data quality rules of the multi-source raw data, the multi-source raw data is subjected to source data quality verification, including format compliance verification, business rationality verification, and reference integrity verification. Based on the source data quality verification results, the multi-source raw data is quality-marked and diverted to obtain qualified source data and abnormal source data. The qualified source data is stored in its original form according to the source system and collection time to obtain the patch source data, and the abnormal source data and the quality verification record corresponding to the abnormal source data are stored. The source data is cleaned, field semantic alignment is performed, missing data is handled, unified encoding mapping is performed, and metadata is labeled to obtain standardized detailed data; During the data flow process of the multi-source raw data, the source-attached data, the standardized detailed data, the summary data, and the carbon footprint accounting parameter data, the data at each level is semantically categorized according to the copper product carbon footprint accounting related objects, and data association relationships between different accounting related objects are established based on the semantic classification results. The copper product carbon footprint accounting related objects include product objects, process objects, material objects, energy emission objects, equipment objects, and accounting result objects. Based on the aforementioned data relationships, the standardized detailed data is aggregated according to product, process, and time dimensions to obtain summary data for copper product carbon footprint accounting. According to the preset carbon footprint accounting rules, activity data values ​​and emission factor values ​​are extracted from the standardized detailed data and / or the summarized data to generate carbon footprint accounting parameter data. Record the data lineage information from the multi-source raw data to the carbon footprint accounting parameter data. The data lineage information includes data input / output identifiers and data processing rule information to establish a traceability association between the carbon footprint accounting parameter data and the corresponding multi-source raw data.

2. The method according to claim 1, characterized in that, The acquisition of multi-source raw data associated with the carbon footprint accounting of copper products includes: Obtain the source attributes, update frequency, and data structure information of candidate data sources to obtain data source feature information; Based on the data source feature information, the candidate data sources are divided into active data sources and emission factor data sources. The active data sources include at least two of the following: industrial IoT time-series data sources, production and operation structured data sources, and supply chain external exchange data sources. The emission factor data sources include at least one of the following: external databases, standard data sources, and literature reference data sources. Based on the activity data source and the emission factor data source, the corresponding data access strategies are determined to obtain the multi-source data access strategy. According to the multi-source data access strategy, data related to the carbon footprint accounting of copper products are collected from the candidate data sources to obtain initial collected data; The initial collected data is preprocessed to obtain the multi-source raw data.

3. The method according to claim 2, characterized in that, Based on the data source types, data structure characteristics, and preset carbon footprint data quality rules of the multi-source raw data, source data quality checks are performed on the multi-source raw data, including format compliance checks, business rationality checks, and citation integrity checks. Based on the source data quality check results, the multi-source raw data is then quality-marked and segmented to obtain qualified source data and abnormal source data, including: Based on the data structure characteristics, the multi-source raw data is subjected to field type validation, required field validation, timestamp format validation, and duplicate record validation to obtain the format compliance validation result. Based on the data source type, the data to be audited is determined from the data that has passed the format compliance check, and the data to be audited is obtained. Based on the preset carbon footprint data quality rules, the data set to be audited is subjected to material input-output consistency verification and / or process parameter range verification to obtain the business rationality verification result. Based on the preset coding dimension table, the material code, process code and equipment code in the data that has passed the format compliance check are checked for referential integrity, and the referential integrity check results are obtained. Based on the format compliance verification result, the business rationality verification result, and the reference integrity verification result, the source data quality verification result is determined; Based on the source data quality verification results, the multi-source raw data is quality-marked and split to obtain the qualified source data and the abnormal source data.

4. The method according to claim 3, characterized in that, The step of performing quality marking and streaming on the multi-source raw data based on the source data quality verification results to obtain the qualified source data and the abnormal source data includes: When the format compliance verification result, the business rationality verification result, and the reference integrity verification result all meet the corresponding verification pass conditions, a qualified mark is added to the corresponding multi-source original data to obtain the qualified source data; When at least one of the format compliance verification result, the business rationality verification result, and the reference integrity verification result fails to meet the corresponding verification pass condition, an anomaly marker is added to the corresponding multi-source original data to obtain the abnormal source data; The quality verification record is generated based on the anomaly type, anomaly field, anomaly triggering rule, and quality verification time of the anomaly source data. The original data snapshot of the abnormal source data is associated with the quality verification record and stored to obtain the abnormal data processing record.

5. The method according to claim 1, characterized in that, The process of cleaning, semantic alignment, missing data processing, unified encoding mapping, and metadata annotation of the source data to obtain standardized detailed data includes: The source data is cleaned by removing outliers and duplicate data to obtain cleaned data. Semantic alignment is performed on fields from different sources but with the same semantics in the cleaned data to obtain semantically aligned data. Missing fields in the semantic alignment data are processed for missing data, and missing processing identifiers are added to the fields that have undergone missing data processing to obtain completion marker data; Based on a preset standard coding system, the product code, process code, material code, energy code, and equipment code in the completed tag data are uniformly coded and mapped to obtain standardized coding data. Add information about the data source, update time, data quality score, and field meaning to the coded standardized data to obtain the standardized detailed data.

6. The method according to claim 5, characterized in that, In the data flow process of the multi-source raw data, the source-attached data, the standardized detailed data, the summary data, and the carbon footprint accounting parameter data, the data at each level is semantically categorized according to the copper product carbon footprint accounting related objects, and data association relationships between different accounting related objects are established based on the semantic categorization results, including: Identify product identifiers, process identifiers, material identifiers, energy emission data identifiers, equipment identifiers, and accounting result data identifiers from data at each level to obtain a set of accounting object identifiers; Based on the set of accounting object identifiers, the data at each level are classified into the product object, the process object, the material object, the energy emission object, the equipment object, and the accounting result object, respectively, to obtain semantic classification results; Based on the semantic classification results, the product process association relationship between product objects and process objects, the process material association relationship between process objects and material objects, the process emission association relationship between process objects and energy emission objects, the process equipment association relationship between process objects and equipment objects, and the emission accounting association relationship between energy emission objects and accounting result objects are determined. Based on the product process association, the process material association, the process emission association, the process equipment association, and the emission accounting association, data association relationships are established between different accounting association objects.

7. The method according to claim 6, characterized in that, Based on the data relationships, the standardized detailed data is aggregated according to product, process, and time dimensions to obtain summary data for copper product carbon footprint accounting, including: Based on the data association, the target product object, target process object and target time range corresponding to the data to be aggregated are determined to obtain the aggregation dimension information; Based on the aggregated dimension information, corresponding material consumption details, energy consumption details, equipment operation details, and emission factor details are extracted from the standardized detailed data to obtain the detailed data to be aggregated. Based on preset aggregation rules that include aggregation keys and aggregation functions, the detailed data to be aggregated is aggregated to obtain summary indicator data under the product process time dimension. Record the correspondence between the detailed data to be aggregated, the aggregation key, the aggregation function, and the summary indicator data to obtain the aggregation path information; Based on the aggregated index data and the aggregated path information, the aggregated data for the carbon footprint accounting of copper products is obtained.

8. The method according to claim 7, characterized in that, The step of extracting activity data values ​​and emission factor values ​​from the standardized detailed data and / or the summarized data according to the preset carbon footprint accounting rules, and generating carbon footprint accounting parameter data, includes: Based on the preset carbon footprint accounting rules, the accounting boundary, accounting process and accounting parameter fields corresponding to the target copper product are determined to obtain the accounting parameter requirement information. Based on the accounting parameter requirement information, the corresponding activity data values ​​are extracted from the standardized detailed data and / or the summary data to obtain the activity data parameter set; Based on the accounting parameter requirement information and the result of the unified coding mapping, the corresponding emission factor values ​​are matched from the standardized detailed data and / or the summary data to obtain the emission factor parameter set. Based on the activity data parameter set, the emission factor parameter set, and the corresponding data quality score, carbon footprint accounting parameter data is generated, which includes activity data values, emission factor values, carbon emission calculation reference values, and data quality scores.

9. The method according to claim 8, characterized in that, The data lineage information recorded from the multi-source raw data to the carbon footprint accounting parameter data includes: During the process of generating the source data from the multi-source raw data, the original data identifier, source system identifier, collection time and quality verification record are recorded to obtain the source data lineage information; During the process of generating the standardized detailed data from the source data, the field semantic alignment rules, missing data handling rules, unified encoding mapping rules and metadata annotation information are recorded to obtain the detailed governance lineage information. During the process of generating the summary data from the standardized detailed data, the aggregation dimension information, preset aggregation rules and aggregation path information are recorded to obtain the summary processing lineage information; During the process of generating the carbon footprint accounting parameter data from the standardized detailed data and / or the summarized data, the source of activity data values, the source of emission factor values, the source of data quality scores, and the correspondence between accounting parameter fields are recorded to obtain parameter generation lineage information; Based on the source data lineage information, the detailed governance lineage information, the summary processing lineage information, and the parameter generation lineage information, the data lineage information is obtained by summarizing the data to establish a traceability association between the carbon footprint accounting parameter data and the corresponding multi-source original data.

10. A data management system for the carbon footprint of copper products, characterized in that, The system includes: A multi-source data acquisition unit is used to acquire multi-source raw data related to the carbon footprint accounting of copper products. The multi-source raw data includes activity data, emission factor data and supply chain related data. The activity data includes at least two of the following: production process data, energy consumption data, material consumption data and equipment operation data. The source data layer processing unit is used to perform source data quality checks on the multi-source raw data based on the data source type, data structure characteristics and preset carbon footprint data quality rules. These checks include format compliance checks, business rationality checks and reference integrity checks. The unit then performs quality marking and diversion on the multi-source raw data according to the source data quality check results to obtain qualified source data and abnormal source data. The source layer storage unit is used to store the qualified source data in its original form according to the source system and collection time to obtain the source data, and to store the abnormal source data and the quality verification record corresponding to the abnormal source data. The detail layer processing unit is used to perform data cleaning, field semantic alignment, missing data processing, unified encoding mapping, and metadata annotation on the source data to obtain standardized detail data. The subject domain management unit is used to semantically classify the data at each level according to the copper product carbon footprint accounting related objects during the data flow process of the multi-source raw data, the source data, the standardized detailed data, the summary data and the carbon footprint accounting parameter data, and establish the data association relationship between different accounting related objects based on the semantic classification results. The copper product carbon footprint accounting related objects include product objects, process objects, material objects, energy emission objects, equipment objects and accounting result objects. The aggregation layer processing unit is used to aggregate the standardized detailed data according to the product dimension, process dimension and time dimension based on the data association relationship to obtain the aggregated data for copper product carbon footprint accounting. The application layer output unit is used to extract activity data values ​​and emission factor values ​​from the standardized detailed data and / or the summarized data according to the preset carbon footprint accounting rules, and generate carbon footprint accounting parameter data. The lineage tracing unit is used to record data lineage information from the multi-source raw data to the carbon footprint accounting parameter data. The data lineage information includes data input / output identifiers and data processing rule information to establish a traceability association between the carbon footprint accounting parameter data and the corresponding multi-source raw data.

Images