Carbon footprint factor matching method and device and electronic equipment

Abstract

The invention provides a carbon footprint factor matching method and device and electronic equipment, and the method comprises the steps: obtaining a multi-source carbon footprint factor, carrying out the unified integration of first attribute feature word synonymous fields of the multi-source carbon footprint factor, aligning factor names and factor tags, and forming a carbon footprint factor library; extracting the structural composition of the target equipment and attribute feature words of production energy consumption of the target equipment, performing vectorization coding on the attribute feature words, and determining a coding vector of the target equipment; and performing multi-dimensional similarity matching on the coding vector and a carbon footprint factor library, and determining a carbon footprint factor source path of the target equipment according to a matching result. In carbon footprint factor accounting, automatic matching is realized, and the matching efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of carbon footprint accounting technology, and in particular to a carbon footprint factor matching method, apparatus and electronic equipment. Background Technology

[0002] Carbon footprint accounting is a key method for assessing greenhouse gas emissions throughout a product's life cycle. Effective carbon footprint accounting can effectively protect the environment, reduce pollution, and play a crucial role in controlling the product's life cycle. The completeness of the carbon footprint factor database directly affects the accuracy and comparability of the accounting results.

[0003] However, different industries have different products, and these products are rapidly evolving. Furthermore, due to the complexity of their composition, and the fact that carbon footprint factor databases are not entirely unified and comprehensive, carbon footprint accounting for products cannot be directly matched with the carbon footprint factor databases, thus making it impossible to effectively conduct carbon footprint accounting. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a carbon footprint factor matching method, apparatus and electronic device that solves or at least partially solves the above problems.

[0005] To achieve the above objectives, a first aspect of this application provides a carbon footprint factor matching method, comprising:

[0006] Obtain multi-source carbon footprint factors, and uniformly integrate and align factor names and factor tags by the synonym fields of the first attribute feature words of the multi-source carbon footprint factors to form a carbon footprint factor library;

[0007] Extract the structural composition and energy consumption attribute features of the target equipment, vectorize and encode the attribute features, and determine the encoding vector of the target equipment.

[0008] The encoded vector is matched with the carbon footprint factor library in multiple dimensions, and the source path of the carbon footprint factor of the target equipment is determined based on the matching results.

[0009] Optionally, multi-source carbon footprint factors are obtained, and the multi-source carbon footprint factors are uniformly integrated, aligned with factor names and factor tags, and a carbon footprint factor library is formed, including:

[0010] Establish a field mapping table, identify the semantics of the carbon footprint factors, and align the synonymous fields;

[0011] The attributes of the multi-source carbon footprint factors are determined based on the semantics, and the unified carbon footprint factors are aligned to determine a unified factor label.

[0012] The factor labels of the same carbon footprint factor are structured and the structured representation is assigned to the header of the field mapping table to determine the carbon footprint factor library. The structured representation includes at least the substance name, source, region, process, time, life cycle boundary and data quality level of the carbon footprint factor.

[0013] Optionally, extract the structural composition and energy consumption attribute features of the target equipment, perform vectorized encoding on the attribute features, and determine the encoding vector of the target equipment, including:

[0014] The standard structure of the target equipment and its production energy consumption list is determined in advance;

[0015] Determine the structural composition of the target equipment and its energy consumption list;

[0016] Identify the text fields of the list, accurately segment the text fields using the standard structure as a reference, and determine the structural composition and the attribute feature words of the energy used in production;

[0017] The attribute feature words are vectorized and encoded to determine feature vectors, and the weight of any attribute feature word is determined based on the feature vectors.

[0018] Optionally, the attribute feature words are vectorized and encoded to determine feature vectors, and the weight of any attribute feature word is determined based on the feature vectors, including:

[0019] Treat any of the aforementioned lists as an independent document, and determine the document set to form a corpus;

[0020] Determine the word frequency and inverse document frequency of any of the attribute feature words in the corpus, determine the feature vector, and determine the weight of any of the attribute feature words based on the word frequency and the inverse document frequency;

[0021] The attribute feature words are sorted according to their weights.

[0022] Using the independent documents as rows and the corresponding attribute feature words and their weights as columns, the feature matrix of the target equipment is determined.

[0023] Optionally, the standard structure includes at least the target equipment type, target equipment name, structural component name, structural component material type, structural component region, structural component process, and structural component year.

[0024] Optionally, the encoded vector is matched with the carbon footprint factor library using multi-dimensional semantic similarity, and the source path of the carbon footprint factor of the target equipment is determined based on the matching result, including:

[0025] Input the factor name and the corresponding factor label into the pre-trained semantic model to determine the semantic vector and feature label of the factor name and the corresponding factor label respectively;

[0026] Using the number of types of the feature tags as a dimension, the similarity between the semantic vector and the feature vector under any dimension is matched, and the source path of the carbon footprint factor of the target equipment is determined based on the matching result.

[0027] Optionally, using the number of types of the feature labels as a dimension, the similarity between the semantic vector and the feature vector is matched in any dimension, including:

[0028] In response to the feature label being a physical attribute label of the carbon footprint factor, the matching rule is determined to be a hard rule; the physical attribute label includes at least the material of the carbon footprint factor;

[0029] In response to the feature label being a non-physical attribute label of the carbon footprint factor, the matching rule is determined to be a soft rule; the non-physical attribute label includes at least the semantic relevance, geographical relevance, and temporal relevance of the carbon footprint factor;

[0030] According to the hard rule, if the semantic vector is the same as the feature vector, the match is successful; if the semantic vector is different from the feature vector, the match fails.

[0031] In response to a successful hard rule match, the similarity between the semantic vector and the feature vector is calculated based on the semantic relevance, geographical relevance, and temporal relevance of the carbon footprint factor, respectively, according to the soft rule, and the corresponding score is determined.

[0032] Optionally, based on the matching results, the source path of the carbon footprint factor of the target equipment is determined, including:

[0033] The proportions of the semantic relevance score, geographical relevance score, and temporal relevance score are determined respectively, and the total score of the carbon footprint factor of the target equipment is determined based on the proportions and the scores.

[0034] The carbon footprint factor and the source of the carbon footprint factor of the target equipment are determined based on the total score.

[0035] A second aspect of this application provides a carbon footprint factor matching device, comprising:

[0036] The acquisition module is used to acquire multi-source carbon footprint factors, and to uniformly integrate and align factor names and factor tags for the first attribute feature words of the multi-source carbon footprint factors, and form a carbon footprint factor library.

[0037] The encoding vector module is used to extract the structural composition of the target equipment and the attribute feature words of its production energy consumption, and to vectorize and encode the attribute feature words to determine the encoding vector of the target equipment.

[0038] The matching module is used to perform multi-dimensional similarity matching between the encoded vector and the carbon footprint factor library, and determine the source path of the carbon footprint factor of the target equipment based on the matching results.

[0039] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor, when executing the computer program, implements the method according to the first aspect.

[0040] As can be seen from the above, the carbon footprint factor matching method, apparatus, and electronic equipment provided in this application, by uniformly integrating and aligning factor names and tags of multi-source carbon footprint factors, facilitates direct use of factor names and tags from the carbon footprint factor library for carbon footprint accounting within the relevant field. Furthermore, by vectorizing the structural composition of the target equipment and its energy consumption attribute features, omissions of carbon footprint factors are avoided during the carbon footprint factor accounting process. Finally, through multi-dimensional similarity matching, the carbon footprint factor accounting of the target equipment becomes more granular and accurate. Simultaneously, manual accounting is avoided, improving the efficiency of carbon footprint factor accounting. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the carbon footprint factor matching method 100 according to an embodiment of this application;

[0043] Figure 2 This is a schematic diagram illustrating the formation of the carbon footprint factor library according to an embodiment of this application;

[0044] Figure 3 This is a schematic diagram of the encoding vector for determining the target equipment according to an embodiment of this application;

[0045] Figure 4 This is a schematic diagram illustrating the weighting of attribute feature words in an embodiment of this application.

[0046] Figure 5 This is a schematic diagram illustrating the source path of the carbon footprint factor of the target equipment in an embodiment of this application;

[0047] Figure 6 This is a schematic diagram of the carbon footprint factor matching device according to an embodiment of this application;

[0048] Figure 7 This is a schematic diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0050] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0051] As described in the background section, carbon footprint accounting is a key means of assessing greenhouse gas emissions throughout a product's life cycle. Effective carbon footprint accounting can effectively protect the environment, reduce pollution, and also play a crucial role in product life cycle control. The completeness of the carbon footprint factor database directly affects the accuracy and comparability of the accounting results.

[0052] Regarding carbon factor databases, firstly, they need further improvement. Secondly, different systems use different carbon factor databases; for example, Ecoinvent, GABI, and CLCD are mainstream carbon factor databases. These different databases use different standards, resulting in heterogeneity, such as inconsistent fields, units, and classification systems. Given the inadequacy of the carbon factor databases, this heterogeneity also makes it difficult for products to perform carbon footprint accounting.

[0053] In addition, different industries have different products, and these products are rapidly evolving and updated. Furthermore, due to the complexity of their composition, the matching work for carbon footprint factor accounting primarily relies on manual methods. This process involves item-by-item searching, judging, and selecting the product's composition. For example, if one carbon factor database lists polyisoprene and another lists rubber, while the product's structural composition list includes tires, the system cannot directly match tires with rubber and polyisoprene during the search. Manual verification based on experience is required to determine that the main component of tires is rubber, and the main chemical component of rubber is polyisoprene, significantly reducing the efficiency of product carbon factor accounting.

[0054] Furthermore, the inventors discovered that in the process of carbon footprint accounting, calculating the carbon footprint of only the materials in the product's structural components cannot fully obtain the product's carbon footprint factor. Moreover, on a macro level, the total carbon emissions are greater than the total carbon factor of all structural components of the product. The inventors further found that the carbon footprint accounting of any product lacks the energy consumption of its structural components during production; that is, the energy consumed in the technological processes and pathways of each component in the production of the product should also be included in the product's carbon footprint accounting.

[0055] In view of this, refer to Figure 1 The flowchart shown is a carbon footprint factor matching method 100 provided in this application embodiment. Method 100 begins with step S101, obtaining multi-source carbon footprint factors, and uniformly integrating and aligning factor names and factor tags to form a carbon footprint factor library.

[0056] In an exemplary embodiment, the multi-source carbon footprint factor indicates that the carbon footprint factor comes from different sources and has different standards.

[0057] In this embodiment, multi-source carbon footprint factors can be obtained by structurally analyzing mainstream carbon factor databases, such as the Ecoinvent, GABI, and CLCD carbon factor databases. Since different carbon factor databases use different standards, they need to be integrated into a unified standard to facilitate subsequent matching with the carbon footprint factors of the target equipment.

[0058] In an exemplary implementation, multi-source carbon footprint factors can be integrated based on the ILCD (International Reference Life Cycle Data System) to unify their fields, units, and classification system.

[0059] In an exemplary embodiment, since there are many carbon footprint factor databases in different standard systems, and each carbon footprint factor database also contains a lot of data, the unified integration of multi-source carbon footprint factors can be carried out in advance.

[0060] In some embodiments, reference Figure 2 As shown, step S101 includes:

[0061] S201. Establish a field mapping table, identify synonymous fields in multi-source carbon footprint factors, and align the synonymous fields.

[0062] Different carbon footprint factor databases may have different naming and organization methods for different fields in the carbon footprint factor, but they may be essentially the same. Therefore, aligning synonymous fields will result in a more standardized carbon footprint factor database and will also avoid omissions in subsequent matching processes.

[0063] For example, the fields in different carbon footprint factors generally include:

[0064] The main field types of the CPCD database are:

[0065] (1) Basic Information Category

[0066] Dataset ID (UUID)

[0067] Dataset Name (Chinese / English)

[0068] Data type (Unit process / System process / LCI result)

[0069] Industry / Product Classification (Based on CPC classification and Chinese industry standards)

[0070] Time period

[0071] Geographical scope (Region, e.g., China, provincial)

[0072] Technology level

[0073] Dataset Version

[0074] (2) Input / output list class

[0075] Input flows (raw materials, energy)

[0076] Output flows (products, waste, byproducts)

[0077] Units (e.g., kg, MJ, kWh)

[0078] Amount

[0079] (3) Elementary flows

[0080] Emissions to the environment (air, water, soil)

[0081] Resource use

[0082] Unit

[0083] Amount

[0084] (4) Supplementary Information

[0085] Data quality metrics (temporal, geographical, technological representativeness)

[0086] Source (reference)

[0087] Data processing methods (allocation, cut-off rules, modeling approach)

[0088] 2. Main field types of the Ecoinvent database

[0089] Ecoinvent's data is stored in ecoSpold2 format (XML), and the fields include:

[0090] (1) Data meta-information

[0091] activityName(process / activity name)

[0092] UUID (Unique Identifier)

[0093] geography (region)

[0094] timePeriod (time range)

[0095] technology (technical description)

[0096] datasetType(Unit process / System process / Result)

[0097] Classification (CPC, ISIC, internal classification)

[0098] (2) Input / output streams

[0099] Exchange (input / output records)

[0100] flowName(flow name)

[0101] flowType(product / waste / elementary)

[0102] unit(kg, MJ, m) 3 )

[0103] amount (numerical value)

[0104] isInput / isOutput (Direction)

[0105] uncertainty (distribution of uncertainty)

[0106] (3) Environmental flow

[0107] emission (to air, water, soil)

[0108] resourceUse(natural resources)

[0109] Category tags (elementary flow categories)

[0110] (4) Supplementary Information

[0111] Source (literature, data sources)

[0112] datasetOwner

[0113] allocationMethod

[0114] modelling approach.

[0115] It is clear that the two are not the same in terms of field description.

[0116] S202. Determine the attributes of multi-source carbon footprint factors based on semantics, align multi-source carbon footprint factors with the same attribute, and determine a unified factor label.

[0117] In this step, the field content of multi-source carbon footprint factors is not singular. It may include name, material, geographical information, purpose, etc. Each field represents a different attribute. However, the descriptions may differ in different carbon footprint factor databases. Therefore, it is necessary to unify, integrate and align the descriptions in different carbon footprint factor databases to form factor labels. For example, copper and metal in the label attributes can be aligned as the same attribute, and the unified factor label can be the region.

[0118] In an exemplary implementation, the field mapping table is shown in Table 1 below:

[0119] Table 1

[0120]

[0121] S203. Perform a structured representation of the factor labels for the same carbon footprint factor, and assign the structured representation to the header of the field mapping table to determine the carbon footprint factor library. The structured representation shall include at least the material name, source, region, process, time, life cycle boundary and data quality level of the carbon footprint factor.

[0122] In an exemplary implementation, the structured representation can be: Factor ID|Name|Source|Region|Year|Boundary|Data Quality Level.

[0123] Understandably, the resulting carbon footprint factor library contains a massive number of carbon footprint factors. Each carbon footprint factor forms a mapping table, which includes both different attributes of the carbon footprint factor and synonyms for any attribute. The resulting carbon footprint factor library is standardized and uniform in format. Furthermore, the aligned factor names and factor tags allow for more diverse matching during subsequent matching processes, avoiding omissions due to different names.

[0124] Meanwhile, the structured representation of the factor label for the carbon footprint factor in the header can be used for matching first. After the matching is completed, the contents of the field mapping table can be used for further matching to improve matching efficiency.

[0125] Subsequently, in step S102, the structural composition of the target equipment and the attribute feature words of its production energy are extracted, the attribute feature words are vectorized and encoded, and the encoding vector of the target equipment is determined.

[0126] In an exemplary embodiment, the target equipment can be electrical equipment, such as transformers and cables, or other products with complex carbon emission accounting and structural composition, without being specifically limited here.

[0127] In this embodiment, the structural composition and production energy consumption of the target equipment are extracted. If the structural composition and production energy consumption are directly matched, the matching relationship of the carbon footprint factor of the structural composition and production energy consumption cannot be directly obtained unless the descriptions are consistent.

[0128] For example, taking a transformer as the target equipment, its structural components include copper, silicon steel sheets, epoxy resin, etc., and the energy used in production includes electricity, natural gas, argon, etc. However, these data are represented differently in the carbon footprint factor library. The carbon footprint factor library contains carbon footprint factors with information such as natural gas, silicon steel, and copper, but the representation in the carbon footprint factor library is not necessarily natural gas, silicon steel, and copper. Therefore, there are significant differences in fields, information, and data.

[0129] By vectorizing attribute feature words, and then matching the vectorized codes with information in the carbon footprint factor library, the difference in expression is reduced.

[0130] In some embodiments, reference Figure 3 As shown, step S102 includes:

[0131] S301. Predetermine the standard structure of the target equipment's structural composition and its production energy consumption list.

[0132] In an exemplary embodiment, each structural component and its standard energy-consuming structure can be represented as follows:

[0133] Target equipment type | Target equipment name | Structural component name | Structural component materials and types | Structural component region | Structural component process | Structural component year.

[0134] S302. Determine the structural composition of the target equipment and its energy consumption list.

[0135] In this step, the structural composition information of the target equipment is first extracted, such as copper, silicon steel sheets, epoxy resin, life cycle, equipment type, etc. Combined with information on energy consumption such as supply location, production process, and technical route, it is convenient to extract attribute feature words from this information in the future.

[0136] In an exemplary embodiment, the structural composition and energy consumption information of the target equipment can be obtained through structural composition lists and energy consumption lists of structural compositions, such as Bill of Materials (BOM), Life Cycle Inventory (LCI), and procurement catalog.

[0137] S303. Identify the text fields of the list, accurately segment the text fields using the standard structure as a reference, and determine the structural composition and the attribute feature words of its production energy consumption.

[0138] In an exemplary implementation, the data in the list can first be clarified by removing special symbols, removing meaningless characters, and standardizing the format.

[0139] The text fields in the list can then be segmented using language models or word segmentation tools. Since the structural composition of the target equipment and the standard structure of its energy consumption list are already determined, segmenting the text fields using the standard structure as a reference improves the accuracy of word segmentation and avoids errors that could hinder effective carbon footprint factor matching and carbon footprint calculation. For example, segmenting "cold / rolled / silicon / steel / sheet" into "cold-rolled silicon steel sheet".

[0140] S304. Vectorize the attribute feature words to determine the feature vector, and determine the weight of any attribute feature word based on the feature vector.

[0141] In this step, during the subsequent matching of carbon footprint factors, directly matching attribute feature words as natural language with the carbon footprint database is impossible unless the expressions are completely identical. By vectorizing and encoding the attribute feature words into computer language, subsequent matching becomes easier. Simultaneously, the weights of the attribute feature words are determined; higher weights indicate greater importance and improve the discriminative power of the matching. For example, attribute feature words such as "copper," "silicon steel," and "epoxy resin" have higher feature vector weights.

[0142] In some embodiments, reference Figure 4 As shown, the attribute feature words are vectorized and encoded to determine feature vectors. The weight of any attribute feature word is then determined based on the feature vectors, including:

[0143] S401. Treat any list as an independent document and determine the document set to form a corpus.

[0144] S402. Determine the term frequency and inverse document frequency of any attribute feature word in the corpus, and determine the weight of any attribute feature word based on the term frequency and inverse document frequency.

[0145] In an exemplary implementation, the TF-IDF algorithm can be used to treat attribute feature words as feature terms, calculate their term frequency (TF) and inverse document frequency (IDF) values ​​in the corpus, and generate a weight for each attribute feature word. The higher the weight, the more important the attribute feature word is in an independent document.

[0146] S403. Sort the attribute feature words according to their weights.

[0147] S404. Using independent documents as rows, determine the feature matrix of the target equipment by identifying the corresponding attribute feature words and their weights in the columns.

[0148] The entire corpus is transformed into a feature matrix, and the values ​​in the matrix are the weights of the attribute feature words as they are transformed into feature vectors.

[0149] It should be noted that during the matching process between the encoding vector and the semantic vector, the feature matrix can be directly matched with the semantic vector.

[0150] In step S103, the encoding vector is matched with the carbon footprint factor library in multiple dimensions, and the source path of the carbon footprint factor of the target equipment is determined based on the matching results.

[0151] In the process of carbon footprint accounting for target equipment, different raw materials and their manufacturing processes for structural components may correspond to multiple sources of carbon footprint factors. Therefore, it is necessary to accurately match the structural composition of the target equipment with the carbon footprint factor library to determine the source paths of the target equipment's carbon footprint factors.

[0152] In some embodiments, reference Figure 5 As shown, step S103 includes:

[0153] S501. Input the factor name and its corresponding factor label into the pre-trained semantic model to determine the semantic vector and feature label of the factor name and its corresponding factor label, respectively.

[0154] For example, factor labels can be:

[0155] Material labels (e.g., steel, transformer oil);

[0156] Geographic tags (such as "CN", "Europe");

[0157] Time tags (e.g., 2018, 2020);

[0158] Process labels (such as "primary production" or "recycling");

[0159] Allocation type (e.g., "cut-off", "allocation").

[0160] In this step, it can also be determined that factor names and factor labels are natural language, while feature vectors are computer language. They cannot be directly matched. Therefore, the factor names and corresponding factor labels are converted into semantic vectors and feature labels for matching.

[0161] In an exemplary implementation, the pre-trained semantic model can be either a Word2Vec semantic model or a BERT semantic model, and there is no limitation on this.

[0162] S502. Using the number of feature label types as a dimension, match the similarity between the semantic vector and the feature vector under any dimension. Based on the matching results, determine the source path of the carbon footprint factor of the target equipment.

[0163] In this embodiment, similarity matching is performed in multiple dimensions, which can effectively determine the carbon footprint factor of the target equipment. Based on the determined carbon footprint factor, the source path of the carbon footprint factor can be determined, avoiding the inability to effectively identify the carbon footprint factor of the target equipment under a single dimension.

[0164] Furthermore, in order to better determine the carbon footprint factor of the target equipment, corresponding matching rules can be constructed during the similarity matching process between feature vectors and semantic vectors.

[0165] In some embodiments, the similarity between semantic vectors and feature vectors is matched in any dimension, using the number of feature label types as a dimension, including:

[0166] In response to the feature label being a physical attribute label of the carbon footprint factor, the matching rule is determined to be a hard rule; the physical attribute label includes at least the material of the carbon footprint factor.

[0167] It should be noted that the physical attribute label represents the physical properties of the carbon footprint factor itself, such as material.

[0168] For example, if the feature label is material, the material of the carbon footprint factor in the semantic vector is copper, while the material of the carbon footprint factor in the feature vector is aluminum, it is obviously impossible to achieve a match.

[0169] In response to a non-physical attribute label that is a carbon footprint factor, the matching rule is determined to be a soft rule; the non-physical attribute label includes at least the semantic relevance, geographical relevance and temporal relevance of the carbon footprint factor.

[0170] It should be noted that non-physical attribute tags represent the relevant attributes of the carbon footprint factor, such as region and process. These can be specific locations within a small area or vague locations within a large area.

[0171] For example, if the feature label is a region, the carbon footprint factor in the semantic vector is country X, and the carbon footprint factor in the feature vector is East Asia, since the regions are close, similarity matching can be performed based on regional relevance.

[0172] For example, the feature label is production process, the carbon footprint factor in the semantic vector is cold rolling, and the carbon footprint factor in the feature vector is low-alloyed. Obviously, they are just different in expression. Similarity is determined by semantic relevance for matching.

[0173] According to the hard rule, if the semantic vector and the feature vector are the same, the match is successful; if the semantic vector and the feature vector are different, the match fails.

[0174] In response to a successful hard rule match, the similarity between the semantic vector and the feature vector is calculated based on the semantic relevance, geographical relevance, and temporal relevance of the carbon footprint factor, and the corresponding score is determined according to the soft rules.

[0175] If a hard rule match fails, the physical attributes are different, so there is no need to perform further matching. However, if a hard rule match succeeds, soft rule matching can be used to achieve a more granular match.

[0176] In semantic relevance analysis, cosine similarity can be used to determine the similarity between semantic vectors and feature vectors. For example, cosine similarity can take values ​​from 0 to 1, with a larger value indicating a closer semantic relationship.

[0177] In an exemplary implementation, cosine similarity can be achieved by directly calling scikit-learn in Python. scikit-learn is a standard library in the field of machine learning.

[0178] For example, calculating the semantic relevance score of "cold-rolled silicon steel sheet":

[0179] The carbon footprint factor library contains tens of thousands of factor names (such as "low-alloyed steel" and "copper wire"), all of which are converted into strings of numbers (semantic vectors) through a model. The cosine similarity function is used to calculate the similarity score between the feature vector of "cold-rolled silicon steel sheet" and each semantic vector. Finally, the results are sorted by score, with the best match being "low-alloyed steel" (0.92 semantic similarity out of 1.0), while "copper wire" has a similarity of only 0.15 and is recommended for filtering.

[0180] In an exemplary implementation, for example, regional similarity scoring:

[0181] If the feature vector and the semantic vector represent the same region exactly, then it is given the first score. For example, the score is 1.

[0182] If the feature vector and the region represented by the semantic vector are adjacent, then it is the second score. For example, the score is 0.6-0.8.

[0183] If the geographical distance represented by the feature vector and the semantic vector differs by more than 1000km, it is a third score, for example, a score of 0, where the first score is greater than the second score, which is greater than the third score.

[0184] In an exemplary implementation, for example, time similarity scoring:

[0185] If the feature vector and the semantic vector represent the same time exactly, then it is rated as the fourth score. For example, the frequency score is 1.

[0186] If the time difference between the feature vector and the semantic vector is within the first threshold range, it is rated as the fifth score. For example, a time difference of 1-3 years results in a score of 0.8.

[0187] If the time difference between the feature vector and the semantic vector is greater than the first threshold range, it is rated as the sixth score. For example, a time difference of 5 years results in a score of 0.4.

[0188] It is understandable that for the same structural component, if the time difference is large, the process flow of that structural component will be different, and the energy consumption for production will also be different. Therefore, it is necessary to match the carbon footprint factor based on the time correlation.

[0189] In some embodiments, determining the carbon footprint factor source path of the target equipment based on the matching result includes:

[0190] The proportions of the semantic relevance score, geographical relevance score, and temporal relevance score are determined respectively, and the total score of the carbon footprint factor of the target equipment is determined based on the proportions and the scores.

[0191] In an exemplary embodiment, the total score can be expressed as: semantic relevance score * 0.5 + geographic relevance score * 0.3 + temporal relevance score * 0.2. Where 0.5, 0.3, and 0.2 represent the semantic relevance score, geographic relevance score, and temporal relevance score, respectively.

[0192] The carbon footprint factor and the source of the carbon footprint factor of the target equipment are determined based on the total score.

[0193] It is understandable that the carbon footprint factor database contains a massive amount of data, so there may be more than one matching result. The matching results can be sorted according to the score, and the closest result is determined first. Correspondingly, the carbon footprint factor database contains the source path and carbon emission standard value of the corresponding carbon footprint factor, which can be compared intuitively.

[0194] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0195] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0196] Based on the same inventive concept, corresponding to any of the above embodiments, this application also provides a carbon footprint factor matching device.

[0197] refer to Figure 6 The carbon footprint factor matching device includes:

[0198] The acquisition module 601 is used to acquire multi-source carbon footprint factors, and to uniformly integrate and align factor names and factor tags in the synonym fields of the first attribute feature words of the multi-source carbon footprint factors to form a carbon footprint factor library.

[0199] The encoding vector module 602 is used to extract the structural composition of the target equipment and the attribute feature words of its production energy consumption, and to vectorize and encode the attribute feature words to determine the encoding vector of the target equipment.

[0200] The matching module 603 is used to perform multi-dimensional similarity matching between the encoded vector and the carbon footprint factor library, and determine the source path of the carbon footprint factor of the target equipment based on the matching result.

[0201] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0202] The apparatus of the above embodiments is used to implement the corresponding carbon footprint factor matching method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0203] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the carbon footprint factor matching method described in any of the above embodiments.

[0204] Figure 7This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0205] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0206] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0207] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0208] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0209] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0210] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0211] The electronic devices described above are used to implement the corresponding carbon footprint factor matching methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0212] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the carbon footprint factor matching method as described in any of the above embodiments.

[0213] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0214] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the carbon footprint factor matching method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0215] Based on the same inventive concept, corresponding to the carbon footprint factor matching method described in any of the above embodiments, this disclosure also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer to cause the computer and / or the processor to perform the color correction method. Corresponding to the execution entity for each step in each embodiment of the color correction method, the processor executing the corresponding step can belong to the corresponding execution entity.

[0216] The computer program product of the above embodiments is used to cause the computer and / or the processor to execute the carbon footprint factor matching method as described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0217] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0218] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0219] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0220] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A carbon footprint factor matching method, characterized in that, include: Obtain multi-source carbon footprint factors, and uniformly integrate and align factor names and factor tags by the synonym fields of the first attribute feature words of the multi-source carbon footprint factors to form a carbon footprint factor library; Extract the structural composition and energy consumption attribute features of the target equipment, vectorize and encode the attribute features, and determine the encoding vector of the target equipment. The encoded vector is matched with the carbon footprint factor library in multiple dimensions, and the source path of the carbon footprint factor of the target equipment is determined based on the matching results.

2. The method according to claim 1, characterized in that, Obtain multi-source carbon footprint factors, and uniformly integrate and align factor names and factor tags to form a carbon footprint factor library, including: Establish a field mapping table, identify the semantics of the carbon footprint factors, and align the synonymous fields; The attributes of the multi-source carbon footprint factors are determined based on the semantics, and the unified carbon footprint factors are aligned to determine a unified factor label. The factor labels of the same carbon footprint factor are structured and the structured representation is assigned to the header of the field mapping table to determine the carbon footprint factor library. The structured representation includes at least the substance name, source, region, process, time, life cycle boundary and data quality level of the carbon footprint factor.

3. The method according to claim 1, characterized in that, Extracting the structural composition and energy consumption attribute features of the target equipment, vectorizing and encoding the attribute features, and determining the encoding vector of the target equipment, including: The standard structure of the target equipment and its production energy consumption list is determined in advance; Determine the structural composition of the target equipment and its energy consumption list; Identify the text fields of the list, accurately segment the text fields using the standard structure as a reference, and determine the structural composition and the attribute feature words of the energy used in production; The attribute feature words are vectorized and encoded to determine feature vectors, and the weight of any attribute feature word is determined based on the feature vectors.

4. The method according to claim 3, characterized in that, The process involves vectorizing the attribute feature words to determine feature vectors, and then determining the weight of any attribute feature word based on the feature vectors, including: Treat any of the aforementioned lists as an independent document, and determine the document set to form a corpus; Determine the word frequency and inverse document frequency of any of the attribute feature words in the corpus, determine the feature vector, and determine the weight of any of the attribute feature words based on the word frequency and the inverse document frequency; The attribute feature words are sorted according to their weights. Using the independent documents as rows and the corresponding attribute feature words and their weights as columns, the feature matrix of the target equipment is determined.

5. The method according to claim 3, characterized in that, The standard structure shall include at least the target equipment type, target equipment name, structural component name, structural component material type, structural component region, structural component process, and structural component year.

6. The method according to claim 5, characterized in that, The encoded vector is matched with the carbon footprint factor library using multi-dimensional semantic similarity analysis. Based on the matching results, the source path of the carbon footprint factor of the target equipment is determined, including: Input the factor name and the corresponding factor label into the pre-trained semantic model to determine the semantic vector and feature label of the factor name and the corresponding factor label respectively; Using the number of types of the feature tags as a dimension, the similarity between the semantic vector and the feature vector in any dimension is matched, and the source path of the carbon footprint factor of the target equipment is determined based on the matching results.

7. The method according to claim 6, characterized in that, Using the number of types of the feature labels as a dimension, the similarity between the semantic vector and the feature vector is matched in any dimension, including: In response to the feature label being a physical attribute label of the carbon footprint factor, the matching rule is determined to be a hard rule; the physical attribute label includes at least the material of the carbon footprint factor; In response to the feature label being a non-physical attribute label of the carbon footprint factor, the matching rule is determined to be a soft rule; the non-physical attribute label includes at least the semantic relevance, geographical relevance, and temporal relevance of the carbon footprint factor; According to the hard rule, if the semantic vector is the same as the feature vector, the match is successful; if the semantic vector is different from the feature vector, the match fails. In response to a successful hard rule match, the similarity between the semantic vector and the feature vector is calculated based on the semantic relevance, geographical relevance, and temporal relevance of the carbon footprint factor, and the corresponding score is determined, according to the soft rule.

8. The method according to claim 7, characterized in that, Based on the matching results, the source paths of the carbon footprint factors of the target equipment are determined, including: The proportions of the semantic relevance score, geographical relevance score, and temporal relevance score are determined respectively, and the total score of the carbon footprint factor of the target equipment is determined based on the proportions and the scores. The carbon footprint factor and the source of the carbon footprint factor of the target equipment are determined based on the total score.

9. A carbon footprint factor matching device, characterized in that, include: The acquisition module is used to acquire multi-source carbon footprint factors, and to uniformly integrate and align factor names and factor tags for the first attribute feature words of the multi-source carbon footprint factors, and form a carbon footprint factor library. The encoding vector module is used to extract the structural composition of the target equipment and the attribute feature words of its production energy consumption, and to vectorize and encode the attribute feature words to determine the encoding vector of the target equipment. The matching module is used to perform multi-dimensional similarity matching between the encoded vector and the carbon footprint factor library, and determine the source path of the carbon footprint factor of the target equipment based on the matching results.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor, when executing the computer program, implements the method according to any one of claims 1 to 8.

Images