Method for constructing multi-source carbon footprint database fused with regional industrial data and application

Abstract

The application discloses a method for constructing a multi-source carbon footprint database fused with regional industrial data and application thereof. The method collects regional statistical, industrial, enterprise, product and process carbon data, establishes standard carbon data records containing scale labels, accounting boundary fingerprints, sources, quality and uncertainty, constructs a regional-industry-enterprise-product-process cross-scale mapping network, forms mapping constraints through industrial input relations, material flow, energy flow and yield relations, calibrates missing data and corrects localization factors through two-way conservation of macro carbon total and micro accounting results, and forms a basic database and an industry special database. A carbon influence propagation model is constructed based on the database, carbon changes caused by emission reduction measures are iteratively propagated along the industrial correlation network, an energy structure, process parameter and production load adjustment scheme is generated under the constraint of regional total, and multi-scale carbon accounting, emission reduction decision and product carbon label verification are realized.

Description

Technical Field

[0001] This invention relates to the fields of data management technology and carbon emission management technology, and in particular to a method and application for constructing a multi-source carbon footprint database that integrates regional industrial data. Background Technology

[0002] With the increasing demand for green and low-carbon transformation in the manufacturing industry and the growing need for product carbon footprint management, carbon emission data has expanded from traditional annual energy consumption statistics to encompass multiple lifecycle stages, including raw material acquisition, production and processing, energy supply, logistics and transportation, product use, and waste disposal. To improve the efficiency of carbon emission accounting, existing technologies typically establish carbon emission factor databases or product carbon footprint databases. These databases collect data on enterprise energy consumption, raw material input, production operations, and environmental monitoring, and match this data with corresponding carbon emission factors to obtain carbon emission results for enterprises, products, or industrial parks. Some technologies further enhance the automation of carbon emission accounting through data cleaning, time-series alignment, quality scoring, dynamic emission factors, and intelligent analysis. Existing literature has indeed disclosed technologies such as multi-level carbon data organization, data quality control, and dynamic emission factors.

[0003] However, the data involved in regional industrial carbon emission management exhibits significant multi-scale characteristics. Regional statistical data is typically aggregated according to administrative regions, industry categories, and annual statistical periods; enterprise data is recorded according to organizational boundaries and production cycles; product carbon footprint data is calculated according to functional units and life cycle boundaries; and process data is further represented by high-frequency data such as equipment operation, material input and output, and energy consumption. Significant differences exist between data at different scales in terms of accounting objects, time granularity, functional units, and system boundaries. While existing technologies can organize carbon data at multiple levels such as groups, enterprises, processes, equipment, and products, and establish data relationships between data in the same time and space or for the same products, data at different scales are usually still calculated independently according to their respective accounting boundaries. There is a lack of cross-scale conversion mechanisms to ensure quantitative consistency between regional statistical carbon totals, industry carbon emissions, enterprise measured carbon emissions, and product and process carbon footprints.

[0004] On the other hand, the completeness of data in regional industrial databases is inconsistent. Key enterprises can provide high-precision production activity data through production management systems, energy management systems, and on-site sensing equipment, while some enterprises and links in the industrial chain can only obtain statistical data, industry averages, or general database factors. If general factors are directly used or the missing carbon emissions are allocated proportionally according to single indicators such as output or output value, it is easy to cause discrepancies between macro-statistical results and micro-accounting results, and may result in double counting and emission omissions. Some existing technologies propose to use material balance, energy balance, and data quality assessment to correct product carbon footprint data, and have also proposed to use real enterprise data to update regional or industry emission factors. However, these mainly correct for single products, enterprises, or factor data themselves, and it is still difficult to simultaneously use macro-carbon total constraints, micro-measured data quality, and industrial material and energy flow relationships to complete the unified calibration of multi-scale data.

[0005] Furthermore, existing carbon emission analysis results are mainly used to identify high-emission processes within enterprises, generate emission reduction recommendations, or plan multi-objective pathways for emission reduction technologies in industrial parks. For regional industrial systems, when an enterprise implements energy substitution, process optimization, or production adjustments, it not only changes its own carbon emissions but may also alter its raw material procurement, energy demand, and upstream and downstream product supply, thereby transmitting carbon emission changes along the supply and demand relationships between enterprises and industries.

[0006] Therefore, it is necessary to provide a method for constructing a multi-source carbon footprint database that can integrate regional industrial data. Summary of the Invention

[0007] In view of this, the purpose of this invention is to propose a method and application for constructing a multi-source carbon footprint database that integrates regional industrial data. This solution establishes a quantitative mapping relationship between regions, industries, enterprises, products, and processes based on the unified semantics of data accounting at different scales. It also performs bidirectional calibration of multi-scale carbon data by combining data quality, material balance, energy balance, and regional carbon total constraints. At the same time, it further utilizes the formed database to track the propagation of carbon emission impacts of emission reduction measures in industrial linkage networks, enabling industrial carbon emission decisions to simultaneously consider local production status changes and their linkage impact on regional total carbon emissions.

[0008] To achieve the above-mentioned technical objectives, the technical solution adopted by this invention is as follows: A method for constructing a multi-source carbon footprint database that integrates regional industrial data includes: S1. Acquire regional statistical data, industry production data, enterprise measured production data, product material data, and product process data. Perform standardization processing on the acquired data to generate standard carbon data records. Configure each standard carbon data record with scale label, accounting boundary fingerprint, time label, spatial location label, activity level data, emission factor, data quality weight, uncertainty, and version identifier. S2. Construct a basic carbon footprint database and an industry-specific database based on standard carbon data records. The basic carbon footprint database is used to store common carbon data across industries, and the industry-specific database is used to store carbon data of products, processes, materials, energy and equipment corresponding to industry production processes. S3. Based on the regional industrial input relationship, the product supply and demand relationship between enterprises, the product bill of materials, material flow, energy flow and process output relationship, construct a cross-scale carbon mapping network about region-industry-enterprise-product-process, determine the scale mapping coefficient between adjacent scale nodes and generate a scale aggregation matrix. S4. Calculate the upper-level carbon emission estimate from bottom to top based on the standard carbon data records corresponding to enterprises, products and processes. Compare the upper-level carbon emission estimate with the upper-level carbon constraints formed by regional statistical data and industry production data to determine the cross-scale carbon residual. S5. Using activity level data and corresponding corrections as variables to be calibrated, the data quality weight is used to determine the retention weight of the variables to be calibrated relative to the original values. A conservation calibration model is constructed that is simultaneously constrained by cross-scale carbon residuals, material balance, energy balance and production boundary. The calibrated standard carbon data records are obtained by solving the model. The carbon residuals of the upper scale are then applied in reverse to the corresponding lower scale nodes according to the cross-scale carbon mapping network. S6. Update the regional localized carbon data based on the calibrated standard carbon data records. Write the calibrated standard carbon data records, scale mapping relationship, data source relationship, calibration residual and version relationship into the carbon footprint basic database and industry-specific database to form a regional industrial multi-source carbon footprint database.

[0009] As one possible implementation, further, in step S1 of this solution, the standardization process includes: performing field mapping, unit unification, and time sequence alignment on the acquired data to generate standard carbon data records.

[0010] As one possible implementation, the scale label described in this solution is used to identify the regional scale, industry scale, enterprise scale, product scale, and / or process scale corresponding to the standard carbon data record; The accounting boundary fingerprint is formed by a combination of geographical boundary identifiers, organizational boundary identifiers, life cycle boundary identifiers, time granularity identifiers, functional unit identifiers, and / or greenhouse gas accounting scope identifiers; The data quality weight is calculated based on the reliability of the data source, time representativeness, regional representativeness, process representativeness, and / or data completeness, and is used to control the permissible correction degree of the standard carbon data record in the conservation calibration model.

[0011] As a possible implementation, further, in step S3 of this solution, the cross-scale carbon mapping network includes scale aggregation edges for connecting process nodes, product nodes, enterprise nodes, industry nodes and regional nodes, and industry supply and demand propagation edges for connecting enterprise nodes and industry nodes with product supply and demand relationships. Among them, for the upper-level node Corresponding lower-level nodes The scaling coefficients are: In the formula, Indicates the lower-level node For the parent node The scale mapping coefficients, It represents the physical contribution base determined based on actual output, material consumption, energy consumption, product bill of materials, and process output; Represents a set of nodes The Middle Each node relative to its parent node The physical contribution base, Indicates the relationship with the parent node Establish the set of subordinate nodes that form an aggregation relationship.

[0012] As one possible implementation, further, in step S5 of this scheme, the conservation calibration model is: in, This represents the optimal activity level data vector obtained by solving the conservation calibration model. This represents the calibrated activity level data vector. This represents the activity level data vector before calibration. This represents the data preservation weight matrix, determined by data quality weights and uncertainty. Represents the scale aggregation matrix, Represents the emission factor matrix, This indicates carbon constraints at a higher-level scale. This represents the reliable weight matrix for carbon constraints at higher scales. This represents the scale consistency constraint coefficient.

[0013] As a preferred implementation method, the conservation calibration model described in this scheme preferably also satisfies material balance constraints. Energy balance constraints and data allowable range constraints ; in, This represents the calibrated activity level data vector. This represents the material balance constraint matrix established based on the material input and output relationships of the processes. This represents the boundary vector corresponding to the material balance constraint. This represents the energy balance constraint matrix established based on the energy input and output relationships of equipment and processes; This represents the boundary vector corresponding to the energy balance constraint; , These represent the lower and upper limits of the activity level data, respectively, determined based on equipment capacity, process design values, and historical operating range.

[0014] As a preferred implementation method, step S6 of this solution further includes: The regional localization correction coefficient is determined based on the actual production data of enterprises that have completed the conservation calibration. The regional localization value of the emission factor is obtained by multiplying the general reference value of the emission factor by the regional localization correction coefficient. The industry-specific databases establish production process routes and product-process mapping relationships for the steel, chemical, cement, and battery industries, respectively. When standard carbon data records, emission factor versions, and process routes change, the set of affected nodes that have scale aggregation relationships and industry supply and demand propagation relationships with the changed nodes is determined based on the cross-scale carbon mapping network. Local scale aggregation and conservation calibration are performed on the set of affected nodes, and the data versions before and after the update are saved.

[0015] Based on the above, this solution also proposes an industrial carbon emission decision-making method based on multi-source carbon footprint data, which applies the aforementioned method for constructing a multi-source carbon footprint database that integrates regional industrial data, including: A01. Obtain carbon data for regions, industries, enterprises, products and processes from the multi-source carbon footprint database, as well as the carbon mapping relationships between different scales, and generate regional industrial carbon emission baselines based on data with the same accounting boundary fingerprints; A02. Construct an industrial carbon impact propagation network based on the supply and demand relationship of raw materials, energy and products among enterprises, and determine the carbon impact propagation matrix based on the physical input, supply ratio and node carbon intensity response relationship in the industrial carbon impact propagation network. A03. Obtain candidate adjustment parameters for energy substitution ratio, unit energy consumption of process, equipment operating load, raw material ratio and production volume, input the candidate adjustment parameters into the carbon emission response model of the corresponding process, and obtain the initial node carbon disturbance caused by the candidate adjustment parameters. A04. Based on the carbon impact propagation matrix, the initial node carbon disturbance is iteratively propagated along the industrial carbon impact propagation network to obtain the transmitted carbon disturbance of upstream enterprises, downstream enterprises and related industries. Based on the carbon mapping relationship between each scale, the initial node carbon disturbance and the transmitted carbon disturbance are aggregated to the regional scale to obtain the regional carbon emission change corresponding to the candidate adjustment parameter. A05. Verify the consistency between the change in regional carbon emissions and the regional carbon emission constraints, and solve the candidate adjustment parameters under the constraints of regional carbon emissions, industry output, enterprise production capacity, equipment operation, energy supply, and material balance. A06. Generate industrial carbon emission decision parameters based on the solution results. The industrial carbon emission decision parameters include the target energy substitution ratio, the target process unit energy consumption, the target equipment operating load, and the target production volume. Send the industrial carbon emission decision parameters to the corresponding enterprise's production management system and energy management system.

[0016] As a preferred implementation method, preferably, in step A04 of this scheme, when iteratively propagating the initial node carbon perturbation along the industry carbon impact propagation network based on the carbon impact propagation matrix, it includes: Let the initial nodal carbon perturbation be The carbon impact propagation matrix is , No. The carbon perturbation obtained from the second propagation is: when The propagation stops at a certain point, and the cumulative change in carbon emissions generated by the candidate adjustment parameter in the industrial linkage network is determined according to the following formula: in, , They are respectively , Carbon perturbations obtained from secondary propagation. Indicates the propagation convergence threshold. This represents the number of iterations required to reach the propagation convergence threshold. This represents the cumulative change in carbon emissions.

[0017] Based on the above, this solution also proposes a computer-readable storage medium, which stores at least one instruction, at least one program, code set, or instruction set. The at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the multi-source carbon footprint database construction method for integrating regional industrial data or the industrial carbon emission decision-making method based on multi-source carbon footprint data as described above.

[0018] By adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art: The present invention establishes unified scale labels, accounting boundary fingerprints, and cross-scale mapping relationships between the basic database and industry-specific databases, enabling regional statistics, industry data, enterprise measured data, product bills of materials, and process operation data to form a computable association between region, industry, enterprise, product, and process under the same accounting semantics. For data with different time granularities, functional units, accounting boundaries, and data quality, this solution utilizes quality weights and uncertainty classification for processing. Under the constraints of material balance, energy balance, and production boundaries, it performs bidirectional conservation calibration of macro-level carbon totals and micro-level accounting results, prioritizing the preservation of highly reliable measured data, while missing and low-reliability data are used for scale residual correction. This reduces accounting bias caused by direct application of general factors, simple proportional allocation, and repeated superposition of data at different scales, thereby improving the completeness, consistency, and traceability of regional local industrial carbon footprint data.

[0019] The present invention also converts energy substitution, process adjustment, equipment load and output changes into initial carbon disturbances at process nodes, and propagates them along the material supply and demand relationship between enterprises and the industrial linkage network to calculate the linkage impact of local emission reduction measures on the total carbon amount of upstream and downstream industries and the region; under the constraints of regional carbon constraints, industry output, equipment capacity, energy supply and material balance, emission reduction decision parameters are generated, thereby identifying the situation where local emission reduction is offset by upstream emission increases, and transforming the database from a static accounting data base into a basis for regional industrial carbon reduction decision-making. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a simplified implementation flowchart of the multi-source carbon footprint database construction method that integrates regional industrial data in this solution; Figure 2 This is a simplified implementation flowchart of the industrial carbon emission decision-making method based on multi-source carbon footprint data in this scheme. Detailed Implementation

[0022] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be particularly noted that the following embodiments are for illustrative purposes only and do not limit the scope of the invention. Similarly, the following embodiments are only some, not all, embodiments of the present invention, and all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Combination Figure 1 As shown in the figure, this embodiment proposes a method for constructing a multi-source carbon footprint database that integrates regional industrial data, which includes: S1. Acquire regional statistical data, industry production data, enterprise measured production data, product material data, and product process data. Perform standardization processing on the acquired data to generate standard carbon data records. Configure each standard carbon data record with scale label, accounting boundary fingerprint, time label, spatial location label, activity level data, emission factor, data quality weight, uncertainty, and version identifier. S2. Construct a basic carbon footprint database and an industry-specific database based on standard carbon data records. The basic carbon footprint database is used to store common carbon data across industries, and the industry-specific database is used to store carbon data of products, processes, materials, energy and equipment corresponding to industry production processes. S3. Based on the regional industrial input relationship, the product supply and demand relationship between enterprises, the product bill of materials, material flow, energy flow and process output relationship, construct a cross-scale carbon mapping network about region-industry-enterprise-product-process, determine the scale mapping coefficient between adjacent scale nodes and generate a scale aggregation matrix. S4. Calculate the upper-level carbon emission estimate from bottom to top based on the standard carbon data records corresponding to enterprises, products and processes. Compare the upper-level carbon emission estimate with the upper-level carbon constraints formed by regional statistical data and industry production data to determine the cross-scale carbon residual. S5. Using activity level data and corresponding corrections as variables to be calibrated, the data quality weight is used to determine the retention weight of the variables to be calibrated relative to the original values. A conservation calibration model is constructed that is simultaneously constrained by cross-scale carbon residuals, material balance, energy balance and production boundary. The calibrated standard carbon data records are obtained by solving the model. The carbon residuals of the upper scale are then applied in reverse to the corresponding lower scale nodes according to the cross-scale carbon mapping network. S6. Update the regional localized carbon data based on the calibrated standard carbon data records. Write the calibrated standard carbon data records, scale mapping relationship, data source relationship, calibration residual and version relationship into the carbon footprint basic database and industry-specific database to form a regional industrial multi-source carbon footprint database.

[0024] In this scheme, regional industrial carbon data exhibits significant heterogeneity. For example, regional statistical data typically uses administrative region-industry-year as the statistical unit, enterprise energy data uses enterprise-equipment-hour / day as the statistical unit, product data uses unit product-life cycle stage as the statistical unit, and process data uses production batch-equipment-material / energy input / output as the statistical unit. If only the field names and units of the above data are standardized without simultaneously recording their accounting objects, accounting boundaries, and data quality, it is easy to cause situations such as directly adding data from different statistical calibers, double counting, or incorrect matching of emission factors.

[0025] Therefore, to address the issue that carbon data from different sources, spatial scales, and accounting boundaries cannot directly participate in the same carbon footprint calculation, step S1 of this scheme achieves data cleaning through data standardization, transforming all subsequent data participating in cross-scale calculations into standard carbon data records with unified data semantics. Specifically, step S1 of this scheme includes: performing field mapping, unit unification, and time-series alignment on the acquired data to generate standard carbon data records.

[0026] In addition, in step S1 of this solution, the scale label is used to identify the regional scale, industry scale, enterprise scale, product scale and / or process scale corresponding to the standard carbon data record. The accounting boundary fingerprint is formed by a combination of geographical boundary identifiers, organizational boundary identifiers, life cycle boundary identifiers, time granularity identifiers, functional unit identifiers, and / or greenhouse gas accounting scope identifiers; The data quality weight is calculated based on the reliability of the data source, time representativeness, regional representativeness, process representativeness, and / or data completeness, and is used to control the permissible correction degree of the standard carbon data record in the conservation calibration model.

[0027] As an example, step S1 of this solution may include the following technical details: First, obtain datasets at five scales. Its definition is: in, This represents the original dataset at the regional scale. This represents the original dataset at the industry scale. This represents the raw data set at the enterprise scale; This represents the original data set for product dimensions; This represents the original data set for the process scale.

[0028] The data includes, at a regional scale, at least total regional energy consumption, industry output, regional electricity consumption structure, and regional greenhouse gas emissions statistics, i.e., regional statistical data; industry-scale data includes industry output, input of major raw materials, industry energy consumption, and typical industry process data, i.e., industry production data; enterprise-scale data comes from enterprise ERP, MES, EMS, and energy metering systems, i.e., actual production data; product-scale data includes product BOM, raw material sources, transportation distance, and life cycle process data, i.e., product material data; and process-scale data includes process material input and output, energy consumption, equipment operating time, production batches, and process output data, i.e., product process data.

[0029] After the above data is aggregated and mapped, it is then uniformized in units and aligned in time sequence to generate standard carbon data records.

[0030] Regarding the standardization of units and data formats, for the first The first of the original data Numeric fields First, standardize according to the pre-configured unit conversion factor: in, Indicates the first The first of the original data One original field value; This indicates the field value after unit standardization; Indicates the first The unit conversion factor for each field, from the original unit of measurement to the preset standard unit of measurement; the subscript of the numerical field. Indicates the original data record sequence number. Indicates the sequence number of the data field.

[0031] Regarding the standardization of units and data formats, for example, ton standard coal and kilogram standard coal are standardized to joule or ton standard coal, MWh and kWh are standardized to kWh, and kgCO2e and tCO2e are standardized to kgCO2e.

[0032] Regarding the unification of time scales, let the original time series be... The target unified time window is: For flow-type or consumption-type data, the activity level data in a unified window is obtained through time integration: For discrete sampled data, it can be further written as: in, Indicates the first Each standard carbon data record corresponds to the activity level data; Indicates the first Measurements of how activity data changes over time; Indicates the start time of the target accounting time window; Indicates the end time of the target accounting time window; Indicates the first Each sampling time; Indicates the first The sampling time interval corresponding to each sampling moment; Indicates the first time within the target time window The number of sampling points corresponding to the data type.

[0033] Therefore, regardless of whether the original data is in the minute, hour, day, or month range, the activity level corresponding to the target accounting cycle can be obtained.

[0034] After the above processing, this scheme obtains standard carbon data records. Based on these records, scale labels, boundary fingerprints, time labels, spatial location labels, activity level data, emission factors, data quality weights, uncertainties, and version identifiers are then configured on the data.

[0035] Regarding the configuration of the accounting boundary fingerprint, for data that has completed preliminary standardization, it is not immediately used for carbon accounting, but an accounting boundary fingerprint is generated for it, and its information is defined as follows: in, Indicates the first The accounting boundary fingerprint of a standard carbon data record; Indicates the first The geographical boundary identifier corresponding to each standard carbon data record; Indicates the first The organizational boundary identifier corresponding to each standard carbon data record; Indicates the first The life cycle boundary identifier corresponding to each standard carbon data record; Indicates the first The time granularity identifier corresponding to each standard carbon data record; Indicates the first The functional unit identifier corresponding to each standard carbon data record; Indicates the first The greenhouse gas accounting scope identifier corresponding to each standard carbon data record.

[0036] This calculated boundary fingerprint will be used to determine whether two records have direct aggregation conditions when establishing cross-scale mapping relationships.

[0037] For example, data belonging to the same year 2025 cannot necessarily be directly aggregated; if one is the Scope 1+Scope 2 emission at the enterprise organization level, and the other is the carbon footprint of a product from its initial production stage to becoming a product, then it must be mapped to the product-enterprise scale before it can participate in the calculation at the higher scale.

[0038] Regarding the determination of data quality weights, in order to enable subsequent steps to allocate cross-scale residuals according to the reliability of the data, this scheme uses data quality scores as data quality weights.

[0039] If data quality consists of five components: source reliability, temporal representativeness, regional representativeness, technological representativeness, and data completeness, then: and: in, Indicates the first Data quality score for each standard carbon data record; Indicates the first The data in the first Scores across each quality evaluation dimension; Indicates the first The weights corresponding to each data quality evaluation dimension; This indicates the dimension number for data quality evaluation.

[0040] In this scheme, the higher the data quality score, the closer the data is to the actual operating status of the regional industry.

[0041] For example, actual electricity consumption data obtained from online metering by enterprises can be assigned a higher source reliability score; while national average data from industry literature can serve as supplementary data, their regional and process representativeness is relatively low. The corresponding reduction.

[0042] To determine the data uncertainty, this scheme further synthesizes the instrument error, time aggregation error, and factor error into: in, Indicates the first The overall uncertainty of a standard carbon data record; Indicates the first Measurement uncertainty corresponding to each data point; Indicates the first Uncertainty arising from time aggregation or time interpolation in the data; Indicates the first The uncertainty arising from the emission factors corresponding to each data point.

[0043] In addition to the three data items mentioned above, this solution can extract, set, or map the scale labels, time labels, spatial location labels, activity level data, emission factors, and version identifiers configured for standard carbon data records, in conjunction with regional energy balance and industrial statistics, industry production statistics, domestic and international carbon emission factors, industry research data, enterprise ERP / MES / EMS / SCADA data, product BOM, production process routes, material input and output, energy consumption, and equipment operation data. This solution will not elaborate on these aspects further here.

[0044] Ultimately, a standard carbon data record is represented as: in, Indicates the first Standard carbon data record; Indicates the first A unique identifier for each standard carbon data record; Indicates the first Scale labels for standard carbon data records; Indicates the first The accounting boundary fingerprint of a standard carbon data record; Indicates the first The time stamp of each standard carbon data record; Indicates the first Spatial location tags for standard carbon data records; Indicates the first Each standard carbon data record corresponds to the region, industry, enterprise, product, or process object code; Indicates the first Each standard carbon data record corresponds to the activity level data; Indicates the relationship with the first Emission factors matched with activity level data; Indicates the first Data quality score for each standard carbon data record; Indicates the first The overall uncertainty of a standard carbon data record; Indicates the first A data version identifier for each standard carbon data record.

[0045] Therefore, after step S1, this scheme can obtain a standard carbon data record set. Its definition is: This standard carbon data record set This will subsequently be used as a database organization object, and its corresponding scale labels will be included. Calculate boundary fingerprints Activity level data Emission factors Data quality score Comprehensive uncertainty It continues to be passed to other steps as intermediate data reuse.

[0046] Because public carbon data is difficult to reuse when mixed with industry-specific industrial process data, and industry data cannot maintain a unified reference relationship with public factors, there are problems. For example, if completely independent databases are established for steel, chemicals, cement, and batteries, it will result in duplicate storage of public emission factors such as electricity, fuel, and transportation; conversely, if all industrial data is put into a general database, it is difficult to accurately represent industry-specific process relationships such as ironmaking, clinker calcination, chemical reactions, and battery material preparation. Therefore, step S2 of this solution will adopt a two-tier structure of basic database and industry-specific database for data management.

[0047] As an example, step S2 of this solution includes the following technical details: The standard carbon data record set obtained from S1 Based on object properties, they can be categorized as follows: in, This represents a set of common standard carbon data records applicable to multiple industries; This refers to a set of carbon data records based on specific industrial sectors and production processes.

[0048] The basic database of carbon footprint is denoted as: Industry-specific databases are denoted as: in, Represents the basic database of carbon footprint; This represents the set of common carbon data records in the basic database; Represents a set of general emission factors; Represents a set of codes for general raw materials, energy, transportation, and waste objects; This indicates an industry-specific database; This represents a set of carbon data records specific to industry standards. This represents a set of industry production process routes; It represents the set of intra-industry relationships between products, processes, equipment, materials, and energy.

[0049] Taking the steel industry as an example, the industry's production process routes set The process can be sequentially divided into sintering, ironmaking, steelmaking, continuous casting, and rolling; for the cement industry, it can be sequentially divided into raw material preparation, raw meal preparation, clinker calcination, and cement grinding.

[0050] In this scheme, each specific process node continues to reference the unified object code from step S1. and accounting boundary fingerprint Instead of establishing a new industry-specific coding system.

[0051] Furthermore, a reference relationship is established between basic database records and industry-specific database records, defined as follows: in, This represents a set of reference relationships between the basic database and industry-specific databases; This indicates the basic database standard carbon data record that is accessed by the industry-specific record; This indicates the use of industry-specific standard carbon data records that reference basic carbon data records; This indicates the reference type between basic data records and specific data records.

[0052] For example, when a steel company accesses regional power grid emission factors to record electricity consumption in its blast furnace process, these regional power grid factors are located in the carbon footprint database. Blast furnace process data is located in an industry-specific database. The two are linked by a set of references. Instead of repeatedly copying the same grid emission factor to the steel industry database, establish a reference.

[0053] After step S2, the final result of this solution is: in, This represents an initial two-layer carbon data structure that has not yet completed cross-scale conservation calibration. This initial data structure and its standard carbon data records are also shown. It continues to serve as a data source for subsequent steps in building cross-scale mapping networks.

[0054] In this scheme, although carbon data can be obtained at the five scales of region, industry, enterprise, product, and process, there is a lack of a quantitative mapping mechanism between the different scales. For example, the relationship between an enterprise belonging to a certain industry and a product produced by a certain enterprise in a general database is only a logical relationship. It cannot be used to determine how the carbon emissions of a product or process should be aggregated at the enterprise, industry, and regional scales, how shared energy among enterprises is allocated among different products, or how the macro-level carbon constraints of a certain industry are mapped back to specific enterprises and products.

[0055] Therefore, step S3 of this scheme establishes a cross-scale carbon mapping network with quantitative mapping coefficients.

[0056] As a possible implementation, further, in step S3 of this scheme, the cross-scale carbon mapping network includes scale aggregation edges for connecting process nodes, product nodes, enterprise nodes, industry nodes and regional nodes, and industry supply and demand propagation edges for connecting enterprise nodes and industry nodes with product supply and demand relationships.

[0057] As an example, step S3 of this solution may include the following technical details: Establish a cross-scale node set, and record the standard carbon data in S1 according to scale labels. Generate a set of nodes, defined as: in, This represents the complete set of nodes in a cross-scale carbon mapping network. Represents a set of nodes at the regional scale; Represents a set of industry-scale nodes; Represents the set of enterprise-scale nodes; Represents the set of product scale nodes; This represents the set of process scale nodes.

[0058] Based on this, a cross-scale carbon mapping network is established, defined as follows: in, This represents a cross-scale carbon mapping network representing regional industries. This represents the complete set of nodes in a cross-scale carbon mapping network. Represents the set of aggregated mapping edges between different scales; It represents the set of industry-related edges established between enterprises or industries based on the supply and demand relationship of materials, energy and products.

[0059] The current step mainly utilizes the aggregated mapping edge set between different scales. Complete vertical scaling; a set of industry linkage edges established between enterprises or industries based on the supply and demand relationships of materials, energy, and products. It is also stored to provide a basis for the subsequent propagation of carbon disturbances.

[0060] Determine the scale mapping coefficients for a given parent node. The set of its corresponding child nodes is denoted as: For each subordinate node Calculate the physical contribution base .

[0061] For example: When mapping processes to products, the actual material input or intermediate product output of the process can be used. When mapping products to enterprises, actual product output can be used; When mapping a company to an industry, it can use the physical output of the company's corresponding products; When mapping industries to regions, the physical output or energy activity of the corresponding industries within the region can be used.

[0062] Therefore, the scale mapping coefficients are defined for the parent node. Corresponding lower-level nodes The scaling coefficients are: thus: In the formula, Indicates the current subordinate node number; Indicates the current parent node number; Indicates the relationship with the parent node The associated child node traversal number, Indicates the lower-level node For the parent node The scale mapping coefficients, Indicates the lower-level node Relative to the parent node The physical contribution base is determined based on physical output, material consumption, energy consumption, product bill of materials, and process output. Represents a set of nodes The Middle Each node relative to its parent node The physical contribution base; Indicates the relationship with the parent node The set of all child nodes that establish an aggregation relationship.

[0063] The above approach uses physical quantities rather than simply enterprise output as the primary mapping basis, ensuring that the scaling transformation maintains a physical connection with the actual flow of materials, products, and energy.

[0064] Regarding the formation of the scale aggregation matrix, this scheme establishes the following aggregation matrices respectively: 1. , which represents the aggregation matrix that maps process scale to product scale; 2. , which represents the aggregation matrix that maps product scale to enterprise scale; 3. , which represents the aggregation matrix that maps the enterprise scale to the industry scale; 4. , which represents the aggregation matrix that maps industry scale to region scale.

[0065] Furthermore, a combination matrix spanning multiple scales is obtained. For example, when the process scale is directly aggregated to the region scale, it has the following definition: in, This represents a combined aggregation matrix that directly maps from the process scale to the region scale.

[0066] Therefore, through step S3, this scheme does not obtain a simple data hierarchy structure, but rather: a cross-scale carbon mapping network. and multiple aggregation matrices ( These matrices will be used subsequently for bottom-up aggregation of microscopic carbon data.

[0067] In this scheme, the carbon emissions obtained by aggregating measured data from enterprises, products, and processes are usually inconsistent with regional statistics or industry-verified data. Therefore, it is necessary to first quantify the degree of this inconsistency in order to conduct an authentic calibration.

[0068] The reasons for this inconsistency include: differences between regional statistical boundaries and enterprise accounting boundaries; lack of measured data for some small and medium-sized enterprises; use of industry averages for some activity level data; instrumentation errors in enterprise data; and different conversion methods used for macroeconomic statistical data.

[0069] Therefore, in this approach, one side's data cannot be forced to cover the other side's data. Instead, a quantitative relationship between micro-estimates, macro-constraints, and residuals needs to be established first.

[0070] Specifically, the carbon emissions at the micro-nodes are calculated first. For the first obtained in step S1 The standard carbon data record has the following carbon emissions before calibration: If the same node exists For this type of activity data: in, Indicates the first Carbon emissions before node calibration; Indicates the first Each standard carbon data record corresponds to the activity level data; Indicates the first The emission factor corresponding to each activity level data; Indicates the first The node Activity level data corresponding to each activity type; Indicates the first The node The emission factors corresponding to the activity type; Indicates the first The number of activity types that each node participates in for carbon accounting; Indicates the activity type number.

[0071] Arrange all activity level data to be calibrated as follows: Establish the emission factor matrix: The microscopic carbon emission vector is then: in, This represents a column vector consisting of the activity level data of all lower-level nodes before calibration. This represents a diagonal matrix of emission factors that corresponds one-to-one with the data for each activity level. Represents the carbon emission vector of the lower-level node; This indicates the number of active-level data points involved in the current cross-scale calibration.

[0072] In terms of upward aggregation to form the upper-level scale estimate, based on the scale aggregation matrix obtained in step S3... , where the scale aggregation matrix It can be derived from the corresponding matrix Or, if it is composed of its combination matrix, then: in, This represents the upper-scale carbon emission estimation vector obtained by bottom-up aggregation of micro data; This represents the scale aggregation matrix generated in step S3, which corresponds to the current calibration scale. Represents the emission factor matrix; This represents the active level data vector before calibration.

[0073] For example, if the consistency check is performed directly on enterprises and the industry, then: If process data is directly aggregated to the regional scale, then: In establishing macro-level carbon constraints and calculating residuals, higher-level carbon constraints are obtained from regional statistical data or industry-verified data. calculate: Substituting the aforementioned aggregation relation, we get: in, Represents a higher-scale carbon constraint vector formed from regional statistical data or industry-verified data; Define Indicates the first Carbon emission constraints for objects at higher-level scales; This indicates the number of parent-level scale objects participating in this consistency check; This represents the carbon emission estimation vector at the higher scale obtained by aggregating lower-level data upwards; This represents the carbon residual vector across scales.

[0074] Further calculation of relative residuals: in, Indicates the first Cross-scale relative carbon residuals of each superior node; Indicates the first The cross-scale carbon residuals corresponding to each higher-level node; Indicates the first The macroscopic carbon constraints corresponding to each higher-level node; This represents a default positive number used to avoid a denominator of zero.

[0075] when When the data is less than a preset consistency threshold, the current micro-level data can be considered to be basically consistent with the macro-level statistical constraints; when it exceeds the threshold, [the following will occur]. The relevant node set then proceeds to the next step to perform conservation calibration.

[0076] Through step S4, the core parameters that can be passed to step S5 include: a column vector composed of the activity level data of all lower-level nodes before calibration. Emission factor matrix Scale aggregation matrix A higher-scale carbon constraint vector formed from regional statistical data or industry-verified data. Cross-scale carbon residual vector , No. Data quality score of standard carbon data records , No. Comprehensive uncertainty of standard carbon data records .

[0077] In this scheme, when there are residuals between macro-statistical carbon constraints and micro-measured data, the difference cannot be simply allocated by the number of enterprises, output, or output value. Otherwise, the originally highly reliable enterprise measured data will be unreasonably modified.

[0078] For example: Company A has complete online metering data; Company B only has monthly energy bills; Company C lacks some production data and uses industry averages.

[0079] If there is a discrepancy between the total carbon emissions of the industry and the combined values ​​of the three companies, the more technically reasonable approach would be to: maintain company A's data as much as possible; allow company B to make moderate corrections; and have company C and any missing or incomplete data bear a larger share of the residual.

[0080] Therefore, step S5 of this scheme, based on the aforementioned processing, further... Data quality score of standard carbon data records , No. Comprehensive uncertainty of standard carbon data records Convert to the original value in a mathematical sense while maintaining the weight.

[0081] As one possible implementation, step S5 of this solution further includes the following technical details: To form a data-preserving weight matrix, define: Further normalization: form: in, Indicates the first The initial data for each activity level should retain their weights; Indicates the first obtained by S1 Data quality score for each standard carbon data record; Indicates the first obtained by S1 The overall uncertainty of a standard carbon data record; This represents a preset positive number to prevent numerical anomalies caused by zero uncertainty. Indicates the first The normalized data for each activity level retains its weight; This represents the data preservation weight matrix, which consists of the weights of all data. This indicates the number of activity level data points involved in the conservation calibration.

[0082] Therefore, high data quality score Low overall uncertainty The data has a large data preservation weight. This means that deviating from the original measurement will result in a higher objective function cost.

[0083] In deriving the conserved calibration objective function, let the calibrated activity level data be: in, Represents a calibrated activity level data vector; This represents the vector of activity level data before calibration; This represents the activity level data correction vector to be solved.

[0084] Firstly, the calibrated data must not deviate arbitrarily from the original measurement value; therefore, the first term is constructed as follows: At the same time, it is hoped that the upward aggregation result of the calibrated micro data will approximate the macro constraints obtained by S4. (A higher-scale carbon constraint vector formed from regional statistical data or industry-verified data), therefore, the second term is constructed: Therefore, the joint objective function is obtained: final: That is, the conservation calibration model is: in, This indicates the cost of deviation of the calibration result relative to the original data; This represents the cross-scale residual cost of the calibrated micro-aggregate value relative to the carbon constraint at the higher scale. This represents the cross-scale conservation calibration objective function; This represents the optimal activity level data vector obtained by solving the conservation calibration model. This represents the calibrated activity level data vector. This represents the activity level data vector before calibration. This represents the data preservation weight matrix, determined by data quality weights and uncertainty. Represents the scale aggregation matrix, Represents the emission factor matrix, This indicates carbon constraints at a higher-level scale. This represents the reliable weight matrix for carbon constraints at higher scales. This represents the scale consistency constraint coefficient, which is the adjustment coefficient between the goal of preserving the original data value and the goal of cross-scale consistency.

[0085] because: Substitute: We can obtain: Therefore, the objective function becomes: If we temporarily disregard other physical constraints, for Find the first derivative: make: get: and then: in, This represents the data correction vector generated by the cross-scale carbon residual; This indicates that the data retains the weight matrix; Represents the macroscopic carbon constraint confidence weight matrix; Represents the scale aggregation matrix; Represents the emission factor matrix; Represents the cross-scale carbon residual vector; This represents the scale consistency constraint coefficient.

[0086] As can be seen from the above expression, when a certain record When the corresponding weights are large, the corresponding activity level data is less likely to change significantly; while low-quality, high-uncertainty data are more likely to absorb cross-scale residuals.

[0087] Therefore, physical constraints must also be satisfied in the actual solution process, including: Material balance constraints:

[0088] Energy balance constraints:

[0089] Data allowable range constraints:

[0090] in, This represents the calibrated activity level data vector. This represents the material balance constraint matrix established based on the material input and output relationships of the processes. This represents the boundary vector corresponding to the material balance constraint. This represents the energy balance constraint matrix established based on the energy input and output relationships of equipment and processes; This represents the boundary vector corresponding to the energy balance constraint; , These represent the lower and upper limits of the activity level data, respectively, determined based on equipment capacity, process design values, and historical operating range.

[0091] Therefore, this solution does not ultimately result in arbitrary modification of enterprise data to meet macroeconomic aggregate requirements. Instead, it seeks the optimal activity level data under the combined influence of three constraints: data credibility, macroeconomic carbon, and industrial physical constraints. .

[0092] Further, the calibrated node carbon emissions were obtained: And calibration residuals: in, This represents the carbon emission vector corresponding to the next-level node after conservation calibration. This represents the cross-scale residual carbon after the conservation calibration is completed.

[0093] Ultimately, the activity level data in each standard carbon data record will be... The adjusted activity level data, after conservation calibration, is defined as follows: It also saves the activity level correction amount for standard carbon data records, which is defined as: in, Indicates the first The standard carbon data record records activity level data after conservation calibration; Indicates the first Activity level correction for each standard carbon data record.

[0094] Through the aforementioned steps, this scheme still faces the problem that the general carbon emission factor may not accurately reflect the energy structure, technological level and industrial technology conditions of the target region when generating regional localized carbon data and forming a multi-source carbon footprint database. Therefore, it is still necessary to use data that has been cross-scale conservation calibration to form regional localized carbon data.

[0095] Furthermore, if the database only stores the final carbon emission value without preserving the original data, calibration process, scale relationships, and version information, it will be impossible to explain why the carbon value changed later, nor will it be possible to perform traceable recalculation after emission factors or enterprise data are updated.

[0096] Therefore, step S6 of this scheme is responsible for both generating localized regional data and writing the complete computation chain formed by the aforementioned steps S1 to S5 into the final database.

[0097] As an example, step S6 of this solution includes the following technical details: For those belonging to the same region, industry, and product / process category A set of data calibrated by S5, whose sample set is defined as: Utilizing data quality and uncertainty The regional factor sample weights are redefined as follows: in, Indicates the use of generating the first A valid sample set of localized data for similar regions; Indicates sample The normalization weights corresponding to the generation of localized regional data; Indicates sample Data quality score; Indicates sample The overall uncertainty; This represents a default positive number used to avoid a denominator of zero.

[0098] Using the calibrated activity level and corresponding carbon emissions obtained from S5 above, the regional localized emission factor is calculated: in, Indicates the target region number Localized emission factors corresponding to a type of material, energy, or process; Indicates sample Carbon emissions after conservation calibration; Indicates sample Activity level data after conservation calibration; Indicates sample The corresponding regional localization factor is used to calculate the weight.

[0099] If the database already has a common benchmark factor It can also calculate regionalization correction factors: therefore: in, Indicates the first Regional localization correction coefficient for emission factors; Indicates the first Common benchmark values ​​for emission factors; Indicates the first Regional localization values ​​of emission factors.

[0100] The difference between the above correction logic and the existing scheme that directly corrects emission factors based on enterprise measured data lies in the fact that it uses data to generate regionally localized values. The data is not an unprocessed sample of enterprises, but rather carbon emissions that have been calibrated by simultaneously undergoing macro-level carbon constraints, data quality weights, material balance, and energy balance constraints in step S5. and activity level data .

[0101] Therefore, there is a verifiable consistency between the regionalization factor and the overall regional carbon constraint.

[0102] Regarding the establishment of final inventory records, this scheme expands the final carbon data record as follows: in, Indicates the first The final carbon data record for storage; Indicates a unique identifier for the data; Indicates scale label; Represents the boundary fingerprint of the accounting process; Indicates a time label; Indicates a spatial label; Represents object encoding; This represents the activity level data before calibration; This represents the activity level data after calibration; Indicates the original matching emission factor; Indicates the localized emission factor for the region; Indicates data quality score; Indicates the overall uncertainty of the data; This indicates the amount of correction for activity level data; This indicates the calibrated cross-scale residual information corresponding to this record; This represents data lineage information, which consists of data source, parent node, child node, and factor reference relationships. Indicates the data version identifier.

[0103] When this scheme forms the final database, it will combine the basic database, industry-specific databases, cross-scale carbon mapping networks, regional localization factors, calibration history, and data versions to form the final multi-source carbon footprint database, which is defined as follows: in, This represents the final regional industrial multi-source carbon footprint database; Represents the basic database of carbon footprint; This indicates an industry-specific database; This represents a cross-scale carbon mapping network relating to region, industry, enterprise, product, and process. This represents the set of regionally localized emission factors generated by S6; This represents the set of cross-scale residuals and the history of conserved calibration; This represents a collection of historical versions of all records, mapping relationships, and emission factors in the database.

[0104] Regarding database data updates, as a preferred implementation method, step S6 of this solution preferably further includes: The regional localization correction coefficient is determined based on the actual production data of enterprises that have completed the conservation calibration. The regional localization value of the emission factor is obtained by multiplying the general benchmark value of the emission factor by the regional localization correction coefficient. The industry-specific database establishes the production process routes and product-process mapping relationships for the steel industry, chemical industry, cement industry and battery industry respectively. When standard carbon data records, emission factor versions, and process routes change, the set of affected nodes that have scale aggregation relationships and industry supply and demand propagation relationships with the changed nodes is determined based on the cross-scale carbon mapping network. Local scale aggregation and conservation calibration are performed on the set of affected nodes, and the data versions before and after the update are saved.

[0105] Based on the established multi-source carbon footprint database, this solution also proposes an industrial carbon emission decision-making method based on multi-source carbon footprint data. This method utilizes the aforementioned method for constructing a multi-source carbon footprint database that integrates regional industrial data, and includes: A01. Obtain carbon data for regions, industries, enterprises, products and processes from the multi-source carbon footprint database, as well as the carbon mapping relationships between different scales, and generate regional industrial carbon emission baselines based on data with the same accounting boundary fingerprints; A02. Construct an industrial carbon impact propagation network based on the supply and demand relationship of raw materials, energy and products among enterprises, and determine the carbon impact propagation matrix based on the physical input, supply ratio and node carbon intensity response relationship in the industrial carbon impact propagation network. A03. Obtain candidate adjustment parameters for energy substitution ratio, unit energy consumption of process, equipment operating load, raw material ratio and production volume, input the candidate adjustment parameters into the carbon emission response model of the corresponding process, and obtain the initial node carbon disturbance caused by the candidate adjustment parameters. A04. Based on the carbon impact propagation matrix, the initial node carbon disturbance is iteratively propagated along the industrial carbon impact propagation network to obtain the transmitted carbon disturbance of upstream enterprises, downstream enterprises and related industries. Based on the carbon mapping relationship between each scale, the initial node carbon disturbance and the transmitted carbon disturbance are aggregated to the regional scale to obtain the regional carbon emission change corresponding to the candidate adjustment parameter. A05. Verify the consistency between the change in regional carbon emissions and the regional carbon emission constraints, and solve the candidate adjustment parameters under the constraints of regional carbon emissions, industry output, enterprise production capacity, equipment operation, energy supply, and material balance. A06. Generate industrial carbon emission decision parameters based on the solution results. The industrial carbon emission decision parameters include the target energy substitution ratio, the target process unit energy consumption, the target equipment operating load, and the target production volume. Send the industrial carbon emission decision parameters to the corresponding enterprise's production management system and energy management system.

[0106] Although carbon data from different enterprises, industries, products, and processes have been stored in the database, they cannot be directly used as the initial state of the same industrial carbon emission decision-making problem if the accounting cycle, life cycle boundary, or greenhouse gas accounting scope is inconsistent.

[0107] For example, an enterprise node stores Scope 1 and Scope 2 emissions for 2025, while a product node stores the carbon footprint of the entire production process for a certain quarter in 2026. If both are directly used as inputs to the same decision model, it will result in a mixture of time and boundary conditions in the baseline state.

[0108] Therefore, step A01 of this plan first uses the regional industrial multi-source carbon footprint database formed by the aforementioned plan. Data that meets the same decision-making and accounting conditions are extracted, and carbon emission baselines are established at five scales: region, industry, enterprise, product, and process.

[0109] As an example, step A01 of this solution includes the following technical details: Let the target accounting boundary corresponding to this industrial carbon emission decision be: in, This represents the target accounting boundary fingerprint corresponding to this industrial carbon emission decision; Indicates the target geographic accounting boundary; Indicates the target organization's accounting boundaries; Indicates the target lifecycle accounting boundary; Indicates the target accounting time range; Indicates the target functional unit; This indicates the target greenhouse gas accounting scope.

[0110] Each final database record formed in the aforementioned database construction scheme Accounting boundary fingerprint and Perform boundary matching.

[0111] Define the boundary matching function: in, Only when At that time, the first Each data record is added to the current decision dataset.

[0112] in, Indicates the first Does each carbon data record meet the boundaries of this decision-making accounting? Indicates the first Record number Matching status on each accounting boundary dimension; Indicates the first Record number Boundary information in each boundary dimension; It represents the numbering of six boundary dimensions: geography, organization, life cycle, time, functional unit, and accounting scope.

[0113] Therefore, we can conclude that: in, This represents the set of standard carbon data records that meet the criteria for this industrial carbon emission decision-making and accounting.

[0114] Establishing node carbon emission baselines is crucial for decision-making datasets. The first in Each node uses calibrated activity level data. and regional localized emission factors The node baseline carbon emissions are obtained, which are defined as follows: If the node involves For this type of activity data: in, Indicates the first The baseline carbon emissions of an enterprise, product, or process node at the initial decision-making stage; This represents activity level data after cross-scale conservation calibration. Indicates the localized emission factor for the region; Represents a node Corresponding to the Post-calibration activity level data for similar activities; Represents a node Corresponding to the Regional localized emission factors for similar activities; Represents a node The number of activity types included; Indicates the activity type number.

[0115] The resulting nodal carbon emission baseline vector is defined as follows: in, This represents the carbon emission baseline vector of all nodes involved in industrial carbon emission decision-making. This indicates the total number of nodes involved in this industrial carbon emission decision.

[0116] In terms of forming a unified baseline at the regional scale, the cross-scale aggregation matrix obtained by the aforementioned database construction method is continued to be used.

[0117] For example, the aggregation from process directly to the regional scale is defined as: but: in, This represents the combined-scale aggregation matrix used to aggregate carbon emissions at the current decision node to a regional scale; Represents the node carbon emission baseline vector; This indicates the regional industrial carbon emission baseline prior to the implementation of any emission reduction adjustments.

[0118] In this plan, if a certain enterprise or a certain process undergoes a change in emission reduction, its impact is not limited to that node, but may change the production activities of other nodes through raw material procurement, energy procurement and product supply relationships, thereby leading to new indirect carbon emission changes in the regional industrial system.

[0119] For example, a steel company increasing its use of scrap steel can reduce its use of iron ore and coke, but it will increase its demand for scrap steel; increasing the proportion of electric arc furnaces may increase regional electricity demand. Therefore, if only the direct carbon emission reductions of the steel company itself are calculated, the actual emission reductions in the region may be overestimated.

[0120] Based on this, in the decision-making method of this scheme, step A02 will consider the industry supply and demand sides. This is further converted into a carbon impact propagation matrix that can participate in numerical calculations, which includes: Establish industrial physical input coefficients for supply nodes. To demand nodes The raw materials, energy, or intermediate products provided are assumed to have a physical supply volume of [amount] during the base period. ,node Product output is .

[0121] The input coefficient corresponding to each unit of product is: in, Indicates the demand node The supply nodes required to produce each unit of product Product input volume; Indicates the supply node under the decision baseline state To demand nodes The quantity of goods provided; Indicates the demand node under the decision baseline state. Product output; Indicates the upstream supply node number; Indicates the downstream demand node number.

[0122] Therefore, when node When the production volume changes, the upstream node The resulting changes in demand are: in, Indicates the demand node Change in output relative to a baseline state; Indicates due to node Changes in output affecting supply nodes Changes in product demand.

[0123] In determining the marginal carbon intensity of nodes, this scheme uses the carbon emission baseline formed in step A01 above to calculate the initial value of the marginal carbon intensity per unit output of the nodes: in, Represents a node Carbon intensity per unit output under actual operating conditions in the current region; Represents a node Carbon emission baseline; Represents a node Production of physical products under baseline conditions.

[0124] Therefore, by node Upstream nodes caused by production changes The changes in carbon emissions are as follows: in, Represents a node Changes in production status are transmitted to nodes through supply and demand relationships. The resulting change in carbon emissions.

[0125] In this scheme, nodes The approximate self-carbon perturbation corresponding to the unit output change is: but: Substitute: get: in, This results in an industrial carbon impact propagation matrix, defined as: in, Represents a node When a unit of carbon emission disturbance is generated, it is transmitted to the nodes through the physical supply and demand relationship of the industry. The proportion of carbon disturbances formed; Indicates supply node Marginal carbon intensity per unit output; Indicates the demand node Marginal carbon intensity per unit output; Represents a node Unit product to node The physical input demand coefficient of the product; This represents the carbon impact propagation matrix of the entire regional industrial network, where the carbon impact propagation matrix... It is determined by the actual supply and demand of enterprises, the actual output of products, and the local carbon intensity of the region.

[0126] In this scheme, industrial control variables such as energy substitution ratio, unit energy consumption of process, equipment load, raw material ratio and production volume use different physical units. Therefore, these data cannot be directly entered into the industrial carbon transmission network and need to be converted into a unified nodal carbon emission change quantity first.

[0127] For example, the unit for natural gas substitution ratio is percentage; equipment load may be expressed in kW or percentage; raw material ratio is expressed in mass fraction; output is expressed in t / h; while industrial carbon propagation models use kgCO2e or tCO2e.

[0128] Therefore, it is necessary to establish the conversion relationship between changes in industrial control parameters, changes in process activity levels, and initial carbon disturbances at nodes; this includes: Establish an industrial regulation variable vector, assuming the current benchmark industrial operating parameters are: The candidate industrial control parameters are: but: in, This represents the current industrial operating baseline parameter vector; This represents the vector of candidate industrial regulation parameters to be evaluated. This represents a vector of changes in industrial control parameters relative to the current operating state. This indicates the number of adjustable industrial parameters involved in this carbon emission optimization. This represents the first element in the current industrial operating baseline parameter vector. One reference parameter, Indicates the first The candidate industrial regulation parameters to be evaluated may correspond to the following data information: energy substitution ratio; unit energy consumption setpoint for each process; equipment operating load; raw material ratio; and target production volume.

[0129] Establish the carbon emission response function for the process, for the first... For each process node, carbon emissions can be expressed as: in, Indicates process node Carbon emissions under given operating conditions; This represents the node carbon emission response function established based on the process material flow, energy flow, and equipment operation patterns; Indicates the action applied to the node. Industrial control parameters; Represents a node Currently unadjustable operating status parameters, including environmental status, order status, and equipment status; This indicates the localized emission factor for the region.

[0130] At the current benchmark operating point First-order expansion nearby: The Jacobian matrix is ​​as follows: Therefore, the initial carbon disturbance caused by the candidate industrial control parameters is: in, Here is the node carbon emission vector, which is related to the industrial adjustment parameters. The function, To be at the reference operating point The baseline carbon emission vector, This represents the initial node carbon perturbation vector formed after the industrial regulation parameters are directly applied to the target node. This represents the local response matrix of industrial regulation parameters to each carbon emission node; Indicates the first When an industrial regulation parameter changes by one unit, the node Local rate of change corresponding to carbon emissions; This represents the vector of changes in candidate industrial control parameters relative to the baseline operating parameters.

[0131] Regarding the source of the initial carbon perturbation, for example, the first Each process uses two types of energy simultaneously, with the baseline consumption being respectively... , The emission factors are respectively , Let the change in the proportion of energy 2 replacing energy 1 be... .

[0132] The changes in the consumption of the two types of energy are as follows: thereby: in, , The changes in energy consumption for energy sources 1 and 2 are shown below. Indicates process node The change in the proportion of energy substitution; This indicates the amount of energy consumed under baseline conditions; This indicates the regional localized emission factor corresponding to the energy source being replaced. This indicates the regional localized emission factor corresponding to alternative energy sources; Nodes representing changes directly caused by changes in the energy structure Initial carbon perturbation.

[0133] Therefore, this step unifies the industrial adjustment quantities of various physical units into nodes directly caused by changes in the energy structure. Initial carbon perturbation .

[0134] In the decision-making method of this scheme, the initial carbon disturbance is propagated through the industrial chain iterative propagation and aggregated to the regional scale through step A04. This step solves the problem that the reduction of carbon emissions in local processes does not equal the reduction of total carbon emissions in the region.

[0135] For example, the number of target companies decreases: However, this increases demand for upstream green power equipment materials, power supply, or alternative raw materials, potentially leading to an increase in related businesses. Therefore, the actual emission reduction in the region is only about: If higher-level supply chain relationships exist, changes in upstream companies may continue to affect their own suppliers, so multiple rounds of dissemination are necessary.

[0136] Based on this, step A04 of this solution, during multi-round propagation, includes the following technical details: In the first round of propagation, the propagation matrix obtained in step A02 is used. And the initial carbon perturbation obtained from A03, its function is defined as: in, This represents the initial node carbon disturbance formed by the direct effect of industrial regulation parameters; This indicates the first round of indirect carbon disturbance, which is formed through the transmission of the initial carbon disturbance via a layer of industrial supply and demand relationships. This represents the industrial carbon impact propagation matrix.

[0137] During multiple rounds of transmission, The second round is: We obtained the following in sequence: in, Indicates the process The first generation generated after the propagation of the industrial linkage relationship. Wheel carbon disturbance; Indicates the propagation rounds of carbon disturbances; The first element of the carbon influence propagation matrix represents the... Power of 1.

[0138] when: The transmission stopped at that time; among them, Indicates the first The L2 norm of the carbon perturbation vector; This represents the pre-set convergence threshold for carbon perturbation propagation.

[0139] Regarding the propagation convergence condition, to ensure convergence even when cyclical supply relationships exist in the industrial network, this scheme requires the propagation matrix to satisfy: ;in: Representing the propagation matrix Spectral radius.

[0140] If the original propagation matrix does not satisfy this condition, then a propagation damping factor is introduced: And satisfy: in, This represents the industry carbon impact propagation matrix after damping correction; Indicates the propagation damping coefficient; This represents the spectral radius of the original propagation matrix.

[0141] This plan will be followed up As the actual propagation matrix.

[0142] In calculating the cumulative industrial carbon disturbance, the total carbon disturbance is: Right now: in, This represents the cumulative nodal carbon emission change vector generated by the candidate industrial adjustment scheme throughout the entire industrial linkage network; This indicates the final propagation round when the propagation convergence condition is met.

[0143] In summary, in step A04 of this scheme, when iteratively propagating the initial node carbon perturbation along the industry carbon impact propagation network based on the carbon impact propagation matrix, let the initial node carbon perturbation be... The carbon impact propagation matrix is , No. The carbon perturbation obtained from the second propagation is: when The propagation stops at a certain point, and the cumulative change in carbon emissions generated by the candidate adjustment parameter in the industrial linkage network is determined according to the following formula: in, , They are respectively , Carbon perturbations obtained from secondary propagation. Indicates the propagation convergence threshold. This represents the number of iterations required to reach the propagation convergence threshold. This represents the cumulative change in carbon emissions.

[0144] Theoretically, when: ,and: At that time, according to the geometric series of the matrix: therefore: in, Representation and propagation matrix Identity matrices of the same order.

[0145] When this scheme is applied to a real system, the aforementioned iterative convergence method can be used, without the need to explicitly calculate the matrix inverse.

[0146] In terms of aggregation to the regional scale, this scheme calls the scale aggregation matrix that is continuously passed from A01. Its definition is: Therefore, the predicted regional carbon emissions after implementing the candidate industrial adjustment parameters are: in, This represents the regional-scale change in carbon emissions resulting from the propagation of candidate industrial adjustment schemes through all industry linkages. Represents a regional-scale aggregation matrix; Indicates the regional industrial carbon emission baseline; Indicates the implementation of candidate industrial adjustment parameters The total carbon emissions from regional industries are predicted in the future.

[0147] Step A05 of this decision-making method is to solve for the optimal adjustment parameters under regional carbon constraints and industrial physical constraints. Since this decision-making method cannot select the industrial adjustment scheme solely based on the minimum carbon emissions, excessive emission reduction may lead to insufficient output, equipment exceeding limits, insufficient energy supply capacity, or imbalance in raw material ratios. At the same time, it cannot only reduce the carbon emissions of the target enterprise itself while ignoring the carbon emissions transferred to other nodes in the industrial chain.

[0148] Therefore, the information that needs to be evaluated simultaneously includes: the region's actual total carbon emissions; carbon transfer in the industrial chain; carbon data uncertainty; and the degree of deviation from normal production status.

[0149] As an example, step A05 of this solution's decision-making method includes the following technical details: Establish the carbon transfer volume in the industrial chain, and define the target node set for this direct adjustment as follows: The set of non-target associated nodes is: The additional carbon emissions generated at non-target nodes during the localized emission reduction process are defined as follows: in, Indicates candidate industrial regulation parameters The resulting carbon transfer volume in the industrial chain; This represents the set of target nodes that directly implement industrial regulation. This represents a set of related nodes that are not directly regulated but are affected by industry propagation. Represents the nodes in the cumulative carbon perturbation vector The change in carbon emissions; This indicates that only the new carbon emissions generated by associated nodes are counted.

[0150] Therefore, if a company reduces its own CO2e by 1000 tCO2e, but upstream companies increase their CO2e by 300 tCO2e, then at least: This allows us to identify schemes that involve surface emission reduction and actual carbon transfer.

[0151] This plan also establishes a production operation deviation, defined as follows: To avoid the minimum carbon emission plan deviating significantly from normal production conditions, the deviation is: in, This indicates the overall operational deviation of the candidate industrial adjustment scheme from the normal production state; Indicates the first One candidate industrial regulation parameter; Indicates the first The baseline operating values ​​for each industrial control parameter; Indicates the first The permissible upper limit for each industrial control parameter; Indicates the first The lower limit of the allowable limit for each industrial regulation parameter; This indicates the number of adjustable industrial parameters.

[0152] Regarding the propagation of carbon emission calculation uncertainty, the database record called in step A01 itself contains the uncertainty formed during the database construction phase. .

[0153] This step expresses the uncertainty of the initial nodal carbon perturbation as a covariance matrix, defined as follows: Based on the definition determined in step A04: in: Then, according to the definition of linear propagation of covariance: Further aggregation to the regional scale: in, Denotes the covariance matrix corresponding to the initial carbon perturbation; Define Represents a node Uncertainty of the initial carbon perturbation; This represents the cumulative propagation matrix of industrial carbon disturbances; This represents the cumulative carbon perturbation covariance matrix after industrial propagation; This represents the transpose of the cumulative propagation matrix of industrial carbon disturbances. This represents the overall carbon emission calculation uncertainty of the candidate industrial adjustment scheme at the regional scale. This represents a regional-scale aggregation matrix.

[0154] To avoid different dimensions for carbon emissions, carbon transfer, uncertainty, and deviations from production status, benchmark values ​​are used for normalization. This scheme also establishes a comprehensive industrial carbon emission decision objective function, which includes: Establish a comprehensive decision objective function for candidate industrial adjustment schemes, defined as follows: And satisfy: ;in, This represents the comprehensive decision objective function value of the candidate industrial adjustment schemes; This indicates the regional carbon emissions after implementing the candidate scheme; Indicates the regional industrial carbon emission baseline; Indicates the amount of carbon transferred within the industrial chain; This represents a reference value used to normalize carbon transfer amounts across the supply chain. Indicates the uncertainty in calculating carbon emissions at the regional scale; This represents the normalized reference value for carbon emission uncertainty; Indicates the deviation of production operation status; Indicates the weight of the regional total carbon emission indicator; Indicates the weight of carbon transfer indicators in the industrial chain; Indicates the weight of the carbon emission uncertainty indicator; This indicates the deviation of the production operation status from the indicator weight.

[0155] To establish industrial constraints, regional carbon emission constraints must be met first: in, This indicates the upper limit of regional industrial carbon emissions allowed during the current decision-making cycle.

[0156] Meet industry minimum production constraints: in, Indicates the first candidate solution Projected output for each industry; Indicates the first The minimum output required for each industry to meet regional production needs.

[0157] Meets the equipment operating range: Meeting energy supply constraints: in, Indicates the candidate industrial solution for the first Total demand for energy sources; Indicates the current decision-making cycle number The available supply of this type of energy.

[0158] Satisfy material balance constraints: Where (H_M) represents the material balance constraint matrix; This represents a data vector of material and energy activity levels formed after the application of candidate industrial regulation parameters. This represents the material balance boundary vector.

[0159] Therefore, the final solution is: And simultaneously satisfy the above constraints.

[0160] in, This represents the optimal industrial carbon emission decision parameter vector obtained under the conditions of satisfying regional carbon constraints and industrial physical constraints.

[0161] Thus, step A05 completes a comprehensive assessment of local emission reductions, actual net emission reductions in the region, industrial carbon transfer, and industrial feasibility.

[0162] Because existing carbon emission analyses typically output abstract management information such as recommendations to reduce energy consumption and increase the proportion of renewable energy, they cannot be directly converted into industrial operation target parameters that can be used by enterprise MES, EMS, or equipment control systems.

[0163] Therefore, the decision-making method in this scheme does not take the emission reduction recommendation report as the endpoint, but rather uses the optimal industrial carbon emission decision parameter vector obtained in step A05. The mapping is to target operating parameters that can be identified by the enterprise's production and energy management systems.

[0164] Specifically, let: in, This represents the optimal industrial carbon emission decision parameter vector obtained from step A05; (Definition) Indicates the first The optimal energy substitution ratio for each target process; This indicates the number of processes involved in energy structure adjustment; Indicates the first The target unit product energy consumption value corresponding to each target process; Indicates the number of processes involved in adjusting the unit's energy consumption; Indicates the first The target operating load corresponding to each target device; Indicates the number of devices involved in the equipment load adjustment; Indicates the first The target output for a specific enterprise, product, or production line; This indicates the number of products or production lines involved in the production plan adjustment.

[0165] Converting decision parameters into system execution instructions, for example, the energy substitution ratio. This sets the energy supply ratio settings used by the Energy Management System (EMS): ; For the target unit energy consumption Energy consumption limits for the formation process: ; For equipment load This forms the load target corresponding to the equipment control system or MES: ; For target output This forms the production planning objectives in the MES: ; in, This indicates the number of documents issued to the energy management system. One energy structure adjustment target; Indicates the first Unit energy consumption control targets for each process; Indicates the first The target operating load of each target device; Indicates the first The target output plan for a product or production line.

[0166] This ultimately leads to: in, This represents the set of executable carbon emission decision parameters output to the enterprise's industrial information system.

[0167] As an example of a preferred implementation method, after the MES and EMS execute the above parameters, in one control cycle... At the end, actual operating data is collected again, and actual node carbon emissions are calculated according to the database structure established in the first independent claim. The predicted node carbon emissions are: Calculate model bias: in, This represents the node carbon emission vector obtained from actual monitoring after the implementation of industrial carbon emission decisions; This represents the post-execution node carbon emission vector predicted based on the industry propagation model; This represents the model prediction error between actual carbon emissions and predicted carbon emissions. This represents the decision-making, implementation, and feedback cycle for industrial carbon emissions.

[0168] when: At that time, the actual operating data will be rewritten into the regional industrial multi-source carbon footprint database formed by the first independent claim, and the marginal carbon intensity in step A02 will be updated. Input coefficient and propagation matrix .

[0169] in, This represents the prediction error threshold used to determine whether an industrial carbon propagation model needs to be recalibrated.

[0170] Through this step, this solution can further form a complete closed loop: carbon database, carbon emission decision-making, MES / EMS execution, actual industrial operation data, database updates, and decision-making for the next cycle.

[0171] The above description is only a part of the embodiments of the present invention and does not limit the scope of protection of the present invention. Any equivalent device or equivalent process transformation made based on the content of the present invention specification and drawings, or direct or indirect application in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for constructing a multi-source carbon footprint database that integrates regional industrial data, characterized in that, It includes: S1. Acquire regional statistical data, industry production data, enterprise measured production data, product material data, and product process data. Perform standardization processing on the acquired data to generate standard carbon data records. Configure each standard carbon data record with scale label, accounting boundary fingerprint, time label, spatial location label, activity level data, emission factor, data quality weight, uncertainty, and version identifier. S2. Construct a basic carbon footprint database and an industry-specific database based on standard carbon data records. The basic carbon footprint database is used to store common carbon data across industries, and the industry-specific database is used to store carbon data of products, processes, materials, energy and equipment corresponding to industry production processes. S3. Based on the regional industrial input relationship, the product supply and demand relationship between enterprises, the product bill of materials, material flow, energy flow and process output relationship, construct a cross-scale carbon mapping network about region-industry-enterprise-product-process, determine the scale mapping coefficient between adjacent scale nodes and generate a scale aggregation matrix. S4. Calculate the upper-level carbon emission estimate from bottom to top based on the standard carbon data records corresponding to enterprises, products and processes. Compare the upper-level carbon emission estimate with the upper-level carbon constraints formed by regional statistical data and industry production data to determine the cross-scale carbon residual. S5. Using activity level data and corresponding corrections as variables to be calibrated, the data quality weight is used to determine the retention weight of the variables to be calibrated relative to the original values. A conservation calibration model is constructed that is simultaneously constrained by cross-scale carbon residuals, material balance, energy balance and production boundary. The calibrated standard carbon data records are obtained by solving the model. The carbon residuals of the upper scale are then applied in reverse to the corresponding lower scale nodes according to the cross-scale carbon mapping network. S6. Update the regional localized carbon data based on the calibrated standard carbon data records. Write the calibrated standard carbon data records, scale mapping relationship, data source relationship, calibration residual and version relationship into the carbon footprint basic database and industry-specific database to form a regional industrial multi-source carbon footprint database.

2. The method for constructing a multi-source carbon footprint database that integrates regional industrial data as described in claim 1, characterized in that, In step S1, the standardization process includes: performing field mapping, unit unification, and time sequence alignment on the acquired data to generate standard carbon data records.

3. The method for constructing a multi-source carbon footprint database that integrates regional industrial data as described in claim 1, characterized in that, The scale label is used to identify the regional scale, industry scale, enterprise scale, product scale and / or process scale corresponding to the standard carbon data record; The accounting boundary fingerprint is formed by a combination of geographical boundary identifiers, organizational boundary identifiers, life cycle boundary identifiers, time granularity identifiers, functional unit identifiers, and / or greenhouse gas accounting scope identifiers; The data quality weight is calculated based on the reliability of the data source, time representativeness, regional representativeness, process representativeness, and / or data completeness, and is used to control the permissible correction degree of the standard carbon data record in the conservation calibration model.

4. The method for constructing a multi-source carbon footprint database that integrates regional industrial data as described in claim 1, characterized in that, In step S3, the cross-scale carbon mapping network includes scale aggregation edges for connecting process nodes, product nodes, enterprise nodes, industry nodes and regional nodes, and industry supply and demand propagation edges for connecting enterprise nodes and industry nodes with product supply and demand relationships. Among them, for the upper-level node Corresponding lower-level nodes The scaling coefficients are: In the formula, Indicates the lower-level node For the parent node The scale mapping coefficients, Indicates the lower-level node relative to the parent node The physical contribution base; Represents a set of nodes The Middle Each node relative to its parent node The physical contribution base, Indicates the relationship with the parent node Establish the set of subordinate nodes that form an aggregation relationship.

5. The method for constructing a multi-source carbon footprint database that integrates regional industrial data as described in claim 1, characterized in that, In step S5, the conservation calibration model is: in, This represents the optimal activity level data vector obtained by solving the conservation calibration model. This represents the calibrated activity level data vector. This represents the activity level data vector before calibration. This represents the data preservation weight matrix, determined by data quality weights and uncertainty. Represents the scale aggregation matrix, Represents the emission factor matrix, This indicates carbon constraints at a higher-level scale. This represents the reliable weight matrix for carbon constraints at higher scales. This represents the scale consistency constraint coefficient.

6. The method for constructing a multi-source carbon footprint database that integrates regional industrial data as described in claim 5, characterized in that, The conservation calibration model also satisfies material balance constraints. Energy balance constraints and data allowable range constraints ; in, This represents the calibrated activity level data vector. This represents the material balance constraint matrix established based on the material input and output relationships of the processes. This represents the boundary vector corresponding to the material balance constraint. This represents the energy balance constraint matrix established based on the energy input and output relationships of equipment and processes; This represents the boundary vector corresponding to the energy balance constraint; , These represent the lower and upper limits of the activity level data, respectively, determined based on equipment capacity, process design values, and historical operating range.

7. The method for constructing a multi-source carbon footprint database that integrates regional industrial data as described in claim 1, characterized in that, Step S6 also includes: The regional localization correction coefficient is determined based on the actual production data of enterprises that have completed the conservation calibration. The regional localization value of the emission factor is obtained by multiplying the general reference value of the emission factor by the regional localization correction coefficient. The industry-specific databases establish production process routes and product-process mapping relationships for the steel, chemical, cement, and battery industries, respectively. When standard carbon data records, emission factor versions, and process routes change, the set of affected nodes that have scale aggregation relationships and industry supply and demand propagation relationships with the changed nodes is determined based on the cross-scale carbon mapping network. Local scale aggregation and conservation calibration are performed on the set of affected nodes, and the data versions before and after the update are saved.

8. An industrial carbon emission decision-making method based on multi-source carbon footprint data, which applies the multi-source carbon footprint database construction method integrating regional industrial data as described in any one of claims 1 to 7, characterized in that, It includes: A01. Obtain carbon data for regions, industries, enterprises, products and processes from the multi-source carbon footprint database, as well as the carbon mapping relationships between different scales, and generate regional industrial carbon emission baselines based on data with the same accounting boundary fingerprints; A02. Construct an industrial carbon impact propagation network based on the supply and demand relationship of raw materials, energy and products among enterprises, and determine the carbon impact propagation matrix based on the physical input, supply ratio and node carbon intensity response relationship in the industrial carbon impact propagation network. A03. Obtain candidate adjustment parameters for energy substitution ratio, unit energy consumption of process, equipment operating load, raw material ratio and production volume, input the candidate adjustment parameters into the carbon emission response model of the corresponding process, and obtain the initial node carbon disturbance caused by the candidate adjustment parameters. A04. Based on the carbon impact propagation matrix, the initial node carbon disturbance is iteratively propagated along the industrial carbon impact propagation network to obtain the transmitted carbon disturbance of upstream enterprises, downstream enterprises and related industries. Based on the carbon mapping relationship between each scale, the initial node carbon disturbance and the transmitted carbon disturbance are aggregated to the regional scale to obtain the regional carbon emission change corresponding to the candidate adjustment parameter. A05. Verify the consistency between the change in regional carbon emissions and the regional carbon emission constraints, and solve the candidate adjustment parameters under the constraints of regional carbon emissions, industry output, enterprise production capacity, equipment operation, energy supply, and material balance. A06. Generate industrial carbon emission decision parameters based on the solution results. The industrial carbon emission decision parameters include the target energy substitution ratio, the target process unit energy consumption, the target equipment operating load, and the target production volume. Send the industrial carbon emission decision parameters to the corresponding enterprise's production management system and energy management system.

9. The industrial carbon emission decision-making method based on multi-source carbon footprint data as described in claim 7, characterized in that, In step A04, when the initial node carbon perturbation is iteratively propagated along the industry carbon impact propagation network according to the carbon impact propagation matrix, it includes: Let the initial nodal carbon perturbation be The carbon impact propagation matrix is , No. The carbon perturbation obtained from the second propagation is: when The propagation stops at a certain point, and the cumulative change in carbon emissions generated by the candidate adjustment parameter in the industrial linkage network is determined according to the following formula: in, , They are respectively , Carbon perturbations obtained from secondary propagation. Indicates the propagation convergence threshold. This represents the number of iterations required to reach the propagation convergence threshold. This represents the cumulative change in carbon emissions.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the method for constructing a multi-source carbon footprint database that integrates regional industrial data as described in any one of claims 1 to 7, or the industrial carbon emission decision-making method based on multi-source carbon footprint data as described in claim 8 or 9.

Images