Comparative analysis method and system of textile raw material carbon footprint quantification method

Abstract

The invention relates to the field of data processing, in particular to a comparative analysis method and system of a textile raw material carbon footprint quantification method. According to the scheme provided by the invention, a full-process solution for textile raw material carbon footprint quantification is established, material accurate classification is realized through fiber characteristic analysis, weights are intelligently configured by combining market policies and enterprise policies based on a dynamic method library matching differentiation accounting method, and finally economic benefits are improved through resource reuse optimization. According to the scheme provided by the invention, the end-to-end accurate accounting of the mixed fiber material is realized, and the dual goals of policy compliance and production economy are met at the same time.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a comparative analysis method and system for a textile raw material carbon footprint quantification method. BACKGROUND

[0002] At present, the global green supply chain construction is accelerating, and various international and domestic policies have included textiles in the regulatory scope. Precise carbon footprint quantification has become a core requirement for enterprises to respond to trade barriers and obtain green certification. In addition, the carbon footprint quantification of the textile industry has significant particularity. The raw material acquisition process involves the planting and breeding emissions of biological fibers such as cotton and wool, as well as the energy consumption of the extraction of petrochemical raw materials for chemical fibers such as polyester and nylon. The carbon source composition is complex and the accounting boundaries are different.

[0003] The existing carbon footprint quantification method for the textile industry has three key defects. First, it cannot effectively handle the differentiated accounting needs of mixed fiber materials, resulting in the mixing of carbon emission calculations for biological and chemical fibers. Second, the static accounting framework cannot adapt to the rapid changes in carbon regulatory policies in different markets. Finally, traditional methods ignore the reuse value of monitoring equipment and data resources in the production process, resulting in repeated investment and calculation bias. These problems seriously restrict the accuracy and economy of carbon footprint accounting. Therefore, a full-process solution for textile raw material carbon footprint quantification is needed, which takes into account the characteristics of fibers and market changes. SUMMARY

[0004] The purpose of the present application is to provide a full-process solution for textile raw material carbon footprint quantification.

[0005] According to a first aspect of the present application, a comparative analysis method for a textile raw material carbon footprint quantification method is proposed, comprising the following steps: S1 Fiber property analysis: receiving product material list data, analyzing the mass proportion and material properties of each fiber type, generating a fiber property vector, and the material properties at least including fiber source category, production process type and transportation radius; S2 Method library matching: according to the fiber property vector, matching a candidate method set from a pre-constructed carbon footprint quantification method library, each method in the method library being associated with four-dimensional evaluation indexes, including accuracy index A, cost index C, implementation complexity index D and data timeliness index T; S3 Multi-dimensional evaluation and weight configuration: based on target market compliance requirements and enterprise production strategies, dynamically configuring a weight matrix of the four-dimensional evaluation indexes, and calculating the comprehensive score of each candidate method; S4 Mixed material method combination: if the material list analyzed in step S1 contains multiple fiber types, selecting the optimal quantification method for each fiber type from the method library of S2 to generate a method combination scheme for mixed materials; S5 Resource Reuse Optimization: Analyze the reusability of data acquisition equipment and computing resources among different methods in the method combination scheme generated in step S4, and output the final carbon footprint accounting scheme.

[0006] According to some embodiments, in the method of the first aspect of the present invention, the construction of the method library in step S2 specifically includes: collecting historical carbon accounting data, including fiber raw material procurement records, production process parameters and original carbon accounting data; establishing a mapping relationship between fiber type and quantification method, and labeling the applicable fiber attribute range and four-dimensional evaluation index for each quantification method in the method library.

[0007] According to some embodiments, in the method of the first aspect of the present invention, updating the method library includes: based on the established mapping relationship between fiber types and quantification methods, when a new fiber type is input, calculating the similarity between the new fiber type and existing fiber types in the method library; selecting the existing fiber type with the highest similarity as the benchmark fiber type, and the corresponding method of the benchmark fiber type as the benchmark method; and correcting the four-dimensional evaluation index of the benchmark method according to the difference between the new fiber type and the benchmark fiber type.

[0008] According to some embodiments, in the method of the first aspect of the present invention, the four-dimensional evaluation index of the modified benchmark method includes: generating a first modified evaluation index based on the difference between the new fiber type and the benchmark fiber type through a linear regression model; obtaining a benchmark emission factor when the new fiber type has low-carbon certification attributes, and modifying the benchmark emission factor according to the certification level included in the low-carbon certification attributes to obtain a modified emission factor; calculating a second modified evaluation index based on the modified emission factor and the first modified evaluation index, and using the second modified evaluation index as the modified four-dimensional evaluation index.

[0009] According to some embodiments, in the method of the first aspect of the present invention, in step S3, the target market compliance requirements include carbon regulatory policy text, the enterprise production strategy includes product market positioning and corresponding preset weight templates, and the weight matrix of the four-dimensional evaluation indicators is dynamically configured based on the target market compliance requirements and the enterprise production strategy, including: Analyze the carbon regulatory policy text and output the policy intensity coefficient P∈[0,1]; The proportion of recycled materials R∈[0,1] in the fiber characteristic vector generated in step S1 is analyzed. Select a preset basic weight template based on product market positioning. ; The basic weight template is revised based on the policy intensity coefficient P and the proportion of recycled materials R. The indicator weights in the data are used to obtain the initial corrected weights; Based on the product market positioning, a relevant real-time transaction market is determined, and the initial correction weight is dynamically corrected according to the real-time transaction market.

[0010] According to some embodiments, in the method of the first aspect of the application, the parsing of the carbon regulatory policy text outputs a policy strength coefficient P ∈ [0, 1], which includes: Establish a policy keyword library, including encouragement and mandatory keywords; According to the policy keyword library, calculate the policy strength P based on the TF-IDF algorithm, following the formula: Where S is the time effectiveness of the policy.

[0011] According to some embodiments, in the method of the first aspect of the application, the regenerated material ratio R ∈ [0, 1] in the fiber characteristic vector generated in the parsing step S1 includes: Set different difference coefficients for different types of regenerated materials Where represents the conversion coefficient of the i-th type of regenerated material; Establish a regenerated material credibility verification mechanism, including: for regenerated materials with blockchain traceability, credibility t = 1; For traditionally declared regenerated materials, credibility t = 0.5; ; Calculate the regenerated material ratio according to the difference coefficient and the regenerated material credibility verification mechanism, with the formula: .

[0012] According to some embodiments, in the method of the first aspect of the application, the resource reuse optimization of step S5 specifically includes: According to the method combination scheme, calculate the reuse gain coefficient of each method: Where is the resource reuse degree of method i, is the cost correction coefficient, is the implementation complexity correction coefficient; According to the reuse gain coefficient, construct a decision matrix M with a modified dimension, where , , , are the original indicators of method i in the accuracy indicator A, cost indicator C, implementation complexity indicator D, and data timeliness indicator T, respectively, , are the cost gain coefficient and implementation complexity gain coefficient of method i, respectively, and the decision matrix M is: Based on the original indicators of each method in the method combination scheme and the decision matrix, calculate the comprehensive utility U, with the formula: The scheme with the highest value of the comprehensive utility U is taken as the final carbon footprint accounting scheme.

[0013] According to some embodiments, in the method of the first aspect of the application, the calculation of the multiplexing gain coefficient of each method according to the method combination scheme comprises: the multiplexable resource type includes device sharing and data multiplexing, and according to the type of the multiplexable resource, the multiplexing gain coefficient of each method is determined according to the following formula , wherein represents an edge weight coefficient, is an edge existence indicator function: the data transmission time consumption of the data multiplexing edge is obtained , wherein the data transmission time consumption is greater than the transmission time threshold, the multiplexing gain coefficient of each method is updated based on the formula . .

[0014] According to the second aspect of the application, a comparative analysis system of a textile raw material carbon footprint quantification method is provided, which is used to execute the method of the first aspect of the application, and the system comprises: a fiber characteristic analysis module, configured to receive bill of materials data of a product, analyze mass proportions and material attributes of each fiber type, generate a fiber characteristic vector, and the material attributes at least include a fiber source category, a production process type and a transportation radius; a method library matching module, configured to match a candidate method set from a pre-constructed carbon footprint quantification method library according to the fiber characteristic vector, each method in the method library is associated with four-dimensional evaluation indexes, including an accuracy index, a cost index, an implementation complexity index and a data timeliness index; a multi-dimensional evaluation and weight configuration module, configured to dynamically configure a weight matrix of the four-dimensional evaluation indexes based on target market compliance requirements and enterprise production strategies, and calculate a comprehensive score of each candidate method; a mixed material method combination module, configured to select an optimal quantification method for each fiber type from the method library of S2 if the bill of materials analyzed by the fiber characteristic analysis module contains multiple fiber types, and generate a method combination scheme for the mixed material; a resource multiplexing optimization module, configured to analyze the multiplexability of data acquisition devices and computing resources among different methods in the method combination scheme generated by the mixed material method combination module, and output a final carbon footprint accounting scheme.

[0015] The scheme provided by the application has the following beneficial effects: 1. Establish a comprehensive solution for quantifying the carbon footprint of textile raw materials: This solution achieves precise material classification through fiber characteristic analysis, matches differentiated accounting methods based on a dynamic method library, intelligently allocates weights in conjunction with market policies and corporate strategies, and ultimately improves economic efficiency through resource reuse optimization. This method is the first to achieve accurate end-to-end accounting for mixed fiber materials, simultaneously meeting the dual objectives of policy compliance and production economics.

[0016] 2. Construct a self-learning method library system: Through in-depth mining of historical data, an intelligent mapping between fiber characteristics and accounting methods is established. A similarity algorithm is used to automatically match new fiber types, and emission factors are dynamically adjusted based on low-carbon certification data. This enables the method library to continuously evolve, significantly improving the accounting efficiency for novel sustainable materials.

[0017] 3. Develop a policy-sensitive dynamic weight adjustment mechanism: By generating intensity indicators through real-time analysis of regulatory policy texts and combining them with actual enterprise production data (such as the proportion of recycled materials), the weight allocation is automatically optimized. This solution enables enterprises to meet the carbon management requirements of different markets with minimal compliance costs.

[0018] 4. Create a graph theory-based resource reuse optimization system: The accounting method is abstracted as nodes, and resource sharing relationships are abstracted as edges. Reuse benefits are quantified through topological analysis, and feedback is provided to adjust implementation complexity indicators. This method achieves global optimal allocation of carbon accounting resources in the textile production chain, significantly reducing monitoring costs. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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 exceeding the scope of protection claimed by the present invention.

[0020] Figure 1 This is a flowchart illustrating an embodiment 1000 of a comparative analysis method for quantifying the carbon footprint of textile raw materials according to the present invention. Figure 2 for Figure 1 A flowchart illustrating the refinement of step S2A in embodiment 1000; Figure 3 for Figure 1 A flowchart illustrating the refinement of step S2B in embodiment 1000; Figure 4 for Figure 3 A flowchart illustrating the refinement step S2Ba within step S2B; Figure 5 forFigure 1 Flowchart of step S3 in embodiment 1000; Figure 6 As Figure 5 Flowchart of refinement step S31 in step S3 in embodiment 1000; Figure 7 As Figure 5 Flowchart of refinement step S32 in step S3 in embodiment 1000; Figure 8 As Figure 1 Flowchart of step S5 in embodiment 1000; Figure 9 As Figure 8 Flowchart of refinement step S51 in step S5 in embodiment 1000; Figure 10 Structure diagram of embodiment 2000 of a comparative analysis system of a textile raw material carbon footprint quantification method of the present application. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0022] Referring to Figure 1 , Figure 1 Flowchart of embodiment 1000 of a comparative analysis method of a textile raw material carbon footprint quantification method of the present application. As Figure 1 shown, embodiment 1000 includes steps S1-S5.

[0023] In some specific embodiments, in step S1, a comparative analysis system of a textile raw material carbon footprint quantification method (such as a processor containing the system, hereinafter referred to as a processor) analyzes fiber characteristics, specifically including: the processor first receives product material list data, then analyzes the mass proportion and material attributes of each fiber type to generate a fiber characteristic vector. Among them, the material attributes at least include fiber source category, production process type and transportation radius.

[0024] Optionally, the fiber types include biological fibers and chemical fibers. Optionally, the biological fibers include plant fibers and animal fibers, and the plant fibers and animal fibers include various types of fibers. For example, the plant fibers include seed fibers represented by cotton and bast fibers represented by hemp fibers, and the animal fibers include hair fibers (such as wool, rabbit hair, camel hair) and secretion fibers (such as silk). For example, the chemical fibers include regenerated fibers (semi-synthetic fibers such as regenerated cellulose and regenerated protein) and synthetic fibers (such as polyester fibers).

[0025] In some embodiments, in step S2, the processor performs method library matching, specifically including: the processor matches a candidate method set from a pre-constructed carbon footprint quantification method library according to the fiber characteristic vector, and each method in the method library is associated with four-dimensional evaluation indexes including an accuracy index A, a cost index C, an implementation complexity index D, and a data timeliness index T.

[0026] In some embodiments, the construction process of the carbon footprint quantification method library in step S2 includes: collecting historical carbon accounting data, the historical data including fiber raw material procurement records, production process parameters, and carbon accounting raw data; establishing a mapping relationship between fiber types and quantification methods, and labeling the applicable fiber attribute range and four-dimensional evaluation indexes of each quantification method in the method library.

[0027] In some embodiments, in step S3, the processor performs multi-dimensional evaluation and weight configuration, dynamically configures a weight matrix of the four-dimensional evaluation indexes based on target market compliance requirements and enterprise production strategies, and calculates a comprehensive score of each candidate method.

[0028] In some embodiments, in step S3, the processor parses a carbon regulatory policy text to output a policy intensity coefficient P ∈ [0, 1]; parses a regenerated material proportion R ∈ [0, 1] in the fiber characteristic vector generated in step S1; selects a preset basic weight template according to the product market positioning ; corrects the index weight in the basic weight template based on the policy intensity coefficient P and the regenerated material proportion R to obtain an initial corrected weight; determines a related real-time transaction market based on the product market positioning, and dynamically corrects the initial corrected weight according to the real-time transaction market.

[0029] In some embodiments, in step S4, the processor derives a hybrid material method combination, specifically including: if the material list parsed in step S1 includes multiple fiber types, an optimal quantification method is selected from the method library in step S2 for each fiber type to generate a method combination scheme of the hybrid material. In some embodiments, in step S4, the processor selects a method combination corresponding to the hybrid material based on the corrected weight value and the comprehensive score derived in step S3, as the method combination scheme of the hybrid material.

[0030] In some embodiments, in step S5, the processor resource reuse optimization, specifically including: analyzing the reusability of data acquisition equipment and computing resources between different methods in the method combination scheme generated in step S4, outputting the final carbon footprint accounting scheme. Optionally, in step S5, the reusable resource types include device sharing and data reuse.

[0031] In some embodiments, in step S5, the processor derives the final carbon footprint accounting scheme, including: the processor calculates the reuse gain coefficient of each method according to the method combination scheme: Wherein, is the resource reuse degree of method i, is the cost correction coefficient, is the implementation complexity correction coefficient; According to the reuse gain coefficient, a decision matrix M with a correction dimension is constructed, wherein , , , are the original indicators of method i in the four dimensions of accuracy indicator A, cost indicator C, implementation complexity indicator D and data timeliness indicator T, respectively, , are the cost gain coefficient and the implementation complexity gain coefficient of method i, respectively, and the decision matrix M is: Based on the original indicators and the decision matrix of each method in the method combination scheme, the comprehensive utility U is calculated, and the formula is: The scheme with the highest value of comprehensive utility U is taken as the final carbon footprint accounting scheme.

[0032] According to the embodiment as shown in Figure 1 , the technical scheme of the present application establishes a full-process solution for quantifying the carbon footprint of textile raw materials: through fiber property analysis to realize accurate classification of materials, based on a dynamic method library to match differentiated accounting methods, combined with market policies and enterprise strategies to intelligently configure weights, and finally through resource reuse optimization to improve economic efficiency. This method first realizes the end-to-end accurate accounting of mixed fiber materials, while meeting the dual goals of policy compliance and production economy.

[0033] Figure 2 is Figure 1 a detailed step S2A of step S2 in embodiment 1000. As shown in Figure 2 , step S2A includes steps S21-S22.

[0034] In some embodiments, at step S21, the processor collects historical carbon accounting data, the historical data including fiber raw material purchase records, production process parameters and carbon accounting raw data. Optionally, at step S21, the processor obtains the historical carbon accounting data from an enterprise production database. Optionally, the fiber raw material purchase records include raw material origin and raw material transportation distance; the production process parameters include production energy consumption and auxiliary material consumption; and the carbon accounting raw data includes monitoring device type and sampling frequency.

[0035] In some embodiments, at step S22, the processor establishes a mapping relationship between fiber types and quantification methods, and labels the applicable fiber attribute range and four-dimensional evaluation indexes of each quantification method in the method library.

[0036] In some embodiments, at step S22, the processor extracts key attributes of each fiber type to constitute a fiber attribute range, which specifically includes: for biological fibers, fiber attributes include irrigation method and fertilizer use intensity; and for chemical fibers, fiber attributes include crude oil API degree and catalyst type. In some embodiments, at step S22, the processor calculates the four-dimensional evaluation index distribution of each quantification method through Monte Carlo simulation, which specifically includes: determining the precision index based on the error distribution of the measured value; determining the cost index based on the depreciation of the accounting equipment and the time consumption of manual work; determining the implementation complexity index based on a preset expert scoring level; and determining the data timeliness index based on the reciprocal of the data update cycle.

[0037] Figure 3 For Figure 1 the detailed step S2B of step S2 in embodiment 1000. As shown in FIG. 10B, step S2B includes steps S23-S25. Figure 3

[0038] In some embodiments, at step S23, the processor calculates the similarity between a new fiber type and an existing fiber type in the method library based on the established mapping relationship between fiber types and quantification methods. Optionally, the processor calculates the similarity between the new fiber type and the existing fiber type in the method library using the Euclidean distance. For example, in some embodiments, the attribute vector of the existing fiber type is , the attribute vector of the new fiber type is , and the similarity between the attribute vectors is: In some embodiments, the sum of the similarities between all attribute vectors corresponding to the new fiber type and the existing fiber type is taken as the similarity between the two.

[0039] ​In some embodiments, at step S24, the processor selects the existing fiber type with the highest similarity as the reference fiber type, and the corresponding method of the reference fiber type as the reference method. At step S25, the processor corrects the four-dimensional evaluation index of the reference method according to the difference between the new fiber type and the reference fiber type.

[0040] In some embodiments, at step S25, the process of correcting the four-dimensional evaluation index of the reference method by the processor includes: generating a first corrected evaluation index by a linear regression model based on the difference between the new fiber type and the reference fiber type; in the case that the new fiber type has a low-carbon certification attribute, obtaining a reference emission factor and correcting the reference emission factor according to a certification level included in the low-carbon certification attribute to obtain a corrected emission factor; calculating according to the corrected emission factor and the first corrected evaluation index to obtain a second corrected evaluation index, and taking the second corrected evaluation index as the corrected four-dimensional evaluation index.

[0041] According to the embodiments as shown in Figure 3 , the present application proposes a method library system with self-learning ability: an intelligent mapping of fiber characteristics and accounting methods is established through deep mining of historical data, a similarity algorithm is used to realize automatic matching of new fiber types, and an emission factor is dynamically corrected in combination with low-carbon certification data. This makes the method library have the ability of continuous evolution, and significantly improves the accounting efficiency of new sustainable materials.

[0042] Figure 4 For the flowchart of the refining step S2Ba in step S2B in Figure 3 , as shown in Figure 4 , step S2Ba includes steps S251-S253.

[0043] In some embodiments, at step S251, the processor generates a first corrected evaluation index by a linear regression model based on the difference between the new fiber type and the reference fiber type.

[0044] In some embodiments, at step S251, the processor defines a difference vector . Wherein, for continuous attributes (for example, transportation distance), is an absolute value difference; for categorical attributes (for example, catalyst type), is a Hamming distance. For example, in some embodiments, at step S251, the linear regression model is defined as: , . Wherein is a transportation distance difference vector, is an energy consumption difference vector, is a catalyst type difference vector, The efficiency difference vector, and a, b, g, d are difference coefficients corresponding to the attribute vector, respectively.

[0045] In some embodiments, at step S252, the processor obtains a baseline emission factor in the case that the new fiber type has a low-carbon certification attribute, and corrects the baseline emission factor according to a certification level included in the low-carbon certification attribute to obtain a corrected emission factor.

[0046] In some embodiments, at step S252, the processor obtains a baseline emission factor from the LCA database , and multiplies the baseline emission factor by a discount coefficient k according to the certification level, so that the corrected emission factor is .

[0047] In some embodiments, at step S253, the processor calculates a second corrected evaluation index according to the corrected emission factor and the first corrected evaluation index, and takes the second corrected evaluation index as the corrected four-dimensional evaluation index. In some embodiments, the corrected emission factor directly affects the baseline value of the precision index, for example, the carbon emission data calculated according to the emission factor and the monitoring data obtained based on the actual monitoring equipment are compared to obtain the accounting precision index A. In some embodiments, the first corrected evaluation index affects the values of the precision index A and the cost index C based on the linear regression model corresponding to the first corrected evaluation index.

[0048] Figure 5 For Figure 1 the flowchart of step S3 in embodiment 1000. As Figure 5 shown, step S3 includes steps S31-S35.

[0049] In some embodiments, at step S3, the target market compliance requirements include a carbon regulatory policy text, and the enterprise production strategy includes product market positioning and a corresponding preset weight template.

[0050] In some embodiments, at step S31, the processor parses the carbon regulatory policy text and outputs a policy strength coefficient P ∈ [0, 1]. In some embodiments, at step S31, the processor first establishes a policy keyword library including encouragement type keywords and mandatory type keywords, and then calculates the policy strength P based on the TF-IDF algorithm according to the policy keyword library.

[0051] In some embodiments, at step S32, the processor parses the recycled material proportion R ∈ [0, 1] in the fiber characteristic vector generated at step S1, specifically including: the processor sets a difference coefficient for different types of recycled materials, where ​Let represent the conversion factor for the i-th type of recycled material; the processor establishes a credibility verification mechanism for recycled materials, including: for recycled materials with blockchain traceability, the credibility t = For recycled materials declared in the traditional way, the credibility t= According to the coefficient of difference The formula for calculating the proportion of recycled materials using the recycled materials credibility verification mechanism is as follows: .

[0052] In some specific embodiments, in step S33, the processor selects a preset basic weight template based on the product market positioning. Optionally, product market positioning includes the company's place of registration and product distribution. For example, in some specific embodiments, the product is mainly sold in the EU / North America region, and the initial basic weight template... =[0.5,0.2,0.1,0.2]. For example, in some specific embodiments, the product is mainly sold in Asia. =[0.3,0.4,0.2,0.1].

[0053] In some specific embodiments, in step S34, the processor modifies the basic weight template based on the policy intensity coefficient P and the proportion of recycled materials R. The index weights are used to obtain the initial correction weights. In some specific embodiments, the calculation of the initial correction weights specifically includes: when the proportion of recycled materials is lower than a preset recycled material threshold, If the proportion of recycled materials exceeds the threshold for recycled materials, the weighting will be adjusted accordingly. Optionally, the threshold for recycled materials is 0.3.

[0054] In some specific embodiments, in step S35, the processor determines the relevant real-time trading market based on the product's market positioning and dynamically adjusts the initial adjustment weight according to the real-time trading market. In some specific embodiments, in step S35, the processor accesses the carbon trading market API corresponding to the market where the product is mainly sold to obtain the current carbon price. And monitor the carbon footprint disclosure values ​​of competing products of the same type. Then calculate the market competition factor α, following the formula: In step S35, the processor generates the final weight matrix. Follow the formula below: In some specific embodiments, in step S35, the processor determines the weight matrix based on the final weight matrix. Calculate the overall score for each method.

[0055] According to such Figure 5In the illustrated embodiment, the present application proposes a policy-sensitive weight dynamic adjustment mechanism: by real-time analysis of regulatory policy texts to generate intensity indicators, combined with actual production data of enterprises (such as the proportion of recycled materials), the weight distribution is automatically optimized. This scheme enables enterprises to meet the carbon management requirements of different markets at the minimum compliance cost.

[0056] Figure 6 For Figure 5 the refinement step S31 in step S3. As shown in Figure 6 step S31 includes steps S311-S312.

[0057] In some embodiments, in step S311, the processor establishes a policy keyword library, including encouragement type vocabulary and mandatory type vocabulary. Among them, the encouragement type vocabulary includes vocabulary that represents the recommended but not mandatory requirement in the carbon emission policy of the target market, such as encouragement, suggestion, hope, etc.; the mandatory type vocabulary includes vocabulary that represents the mandatory requirement in the carbon emission policy of the target market, such as requirement, must, mandatory, etc.

[0058] In some embodiments, in step S312, the processor calculates the policy intensity P based on the TF-IDF algorithm according to the policy keyword library, following the formula: where S_time represents the policy effective time urgency. Optionally, the value range of S_time is [0, 1]. In some embodiments, in step S312, the processor performs keyword labeling and timeliness analysis based on the established policy keyword library using the TF-IDF algorithm. Among them, the mandatory type vocabulary quantization coefficient and the encouragement type vocabulary quantization coefficient , and the timeliness coefficient are obtained according to the keyword labeling results in the policy.

[0059] Figure 7 For Figure 5 the refinement step S32 in step S3. As shown in Figure 7 step S32 includes steps S321-S323.

[0060] In some embodiments, in step S321, the processor sets a difference coefficient for different types of recycled materials, where represents the conversion coefficient of the i-th type of recycled material. In some embodiments, in step S321, the carbon footprint quantification process of recycled materials has particularity, mainly reflected in the accuracy difference of carbon emissions affected by the source and traceability mechanism of recycled materials. In some embodiments, recycled materials affect the final result of carbon footprint quantification of products, so their proportion needs to be considered.

[0061] In some embodiments, at step S322, the processor establishes a recycled material credibility verification mechanism, including: for recycled materials with blockchain traceability, credibility t= ; for traditionally declared recycled materials, credibility t= In some embodiments, the credibility of recycled materials with blockchain traceability is greater than that of traditionally declared recycled materials. Optionally, t1=1.0, t2=0.7.

[0062] In some embodiments, at step S323, the processor calculates the recycled material proportion according to the difference coefficient and the recycled material credibility verification mechanism, and the formula is: In step S323, the logic of calculating the recycled material proportion is: the recycled material proportion is determined by the product of the credibility verification mechanism and the difference coefficient, and the minimum limit is used to limit the recycled material proportion within 1, so as to avoid errors in accounting.

[0063] In the embodiment shown in Figure 7 , the comparative analysis scheme of the textile raw material carbon footprint quantification method of the present application fully considers the influence of recycled materials on method selection, including: 1. Special methods are needed for carbon footprint accounting of recycled material fibers (such as recycled polyester), such as tracing the source of waste materials (bottle chips / waste yarn), affecting the timeliness weight of data; the energy consumption of depolymerization process fluctuates greatly, which requires higher precision index weight; 2. Market demand for green premium: products containing recycled materials usually have an environmental premium, which requires higher precision carbon data to support green certification (such as GRS certification requires an error of <10%).

[0064] Figure 8 For Figure 1 the flowchart of step S5 in embodiment 1000. As Figure 8 shown, step S5 includes steps S51-S53.

[0065] At step S51, the processor calculates the reuse gain coefficient of each method according to the method combination scheme: wherein, is the resource reuse degree of method i, is the cost correction coefficient, is the implementation complexity correction coefficient.

[0066] In some embodiments, at step S51, the processor constructs a method combination graph, taking each quantification method in the method combination scheme generated in step S4 as a node, and establishing a directed edge according to the type of reusable resources, indicating the resource reuse degree of each node method. In some embodiments, at step S51, the cost correction coefficient The calculation basis includes a 10% cost reduction per shared resource and a complexity correction coefficient The calculation basis includes a 5% complexity increase per shared resource. In step S51, the processor quantifies the impact of resource multiplexing on cost and implementation complexity, with a negative sign indicating cost reduction and a positive sign indicating complexity increase.

[0067] In some embodiments, in step S52, the processor constructs a decision matrix M with modified dimensions according to the multiplexing gain coefficient, where 、 、 、 are the original indicators of method i in the four dimensions of accuracy indicator A, cost indicator C, implementation complexity indicator D, and data timeliness indicator T, respectively, 、 are the cost gain coefficient and the implementation complexity gain coefficient of method i, respectively, and the decision matrix M is: In some embodiments, in step S52, the first four columns of the decision matrix M are the original indicators, and the last two columns are the dynamically calculated correction coefficients, separated by a separator to facilitate the calculation of the overall resource multiplexing situation in subsequent step S53.

[0068] In some embodiments, in step S53, the processor calculates the comprehensive utility U based on the original indicators of each method in the method combination scheme and the decision matrix, with the formula: The scheme with the highest value of comprehensive utility U is selected as the final carbon footprint accounting scheme.

[0069] In step S53, 、 、 、 are the weight coefficients in the weight matrix determined in step S3 as shown in Figure 5 , corresponding to the accuracy indicator A, the cost indicator C, the implementation complexity indicator D, and the data timeliness indicator T, respectively. For example, in some embodiments, the weight matrix determined based on the target market compliance requirements and the enterprise production strategy is [0.5, 0.2, 0.1, 0.2]. 、 represent the original accuracy indicator A and the data timeliness indicator T, 、 represent the modified cost indicator C and the implementation complexity indicator D, respectively.

[0070] In some embodiments, in step S53, the processor derives the optimized final carbon footprint accounting scheme according to the resource reuse condition. The calculation logic is: keeping the precision index A and the data timeliness T unchanged, dynamically adjusting the cost index C and the implementation complexity index D, and then obtaining the final evaluation quantitative index, i.e., the comprehensive utility U, by weighted summation, so as to obtain the optimized carbon footprint accounting scheme.

[0071] In some embodiments as shown in FIG. 5, Figure 8 In some embodiments as shown in FIG. 5, The method combination includes: method 1: satellite remote sensing (A=0.8, C=0.4, D=0.6, T=0.7); method 2: RFID tracking (A=0.7, C=0.5, D=0.8, T=0.6); the sharing relationship includes: method 1 and method 2 share the RFID reader (k=1); after step S52 is corrected, the cost of method 1 is C'=0.4×(1-0.1×1)=0.36; the efficiency of method 2 is D'=0.8×(1+0.05×1)=0.84; the comprehensive utilities of method 1 and method 2 are respectively: ; and the finally obtained optimized carbon footprint accounting scheme includes: the efficiency optimization scheme method 1, the balanced scheme combination, and the cost sensitive scheme method 2.

[0072] According to the embodiment as shown in FIG. 5, Figure 8 The technical scheme provided by the present application creates a resource reuse optimization system based on graph theory: the accounting method is abstracted as a node, and the resource sharing relationship is abstracted as an edge. The reuse benefit is quantified by topological analysis, and the cost and implementation complexity indexes are corrected. This method realizes the global optimization configuration of carbon accounting resources on the textile production chain, and greatly reduces the monitoring cost.

[0073] Figure 9 For the flowchart of the refinement step S51 in step S5 in FIG. 5, Figure 8 As shown in FIG. 5, Figure 9 Step S51 includes step S511-step S512.

[0074] Step S511, the reusable resource type includes device sharing and data reuse. According to the type of reusable resources, the following formula is used to determine , wherein represents the edge weight coefficient, is an edge existence indicator function: In some embodiments, device sharing refers to the sharing of physical monitoring devices, such as GPS locators and spectrometers, by two nodes. Alternatively, data reuse refers to the output data of a predecessor method that can be directly used as the input of a successor method. Alternatively, the value of the edge existence indicator function may be 0 or 1, wherein​ = 1 indicates that the edge exists, = 0 indicates that the edge does not exist. Optionally, in some embodiments, the degree of influence of device sharing is greater than that of data reuse, and the weight of the device sharing edge is 0.9, and the weight of the data reuse edge is 0.4.

[0075] In step S512, the data transmission time consumption of the data reuse edge is obtained , and the data transmission time consumption is greater than a transmission time threshold, the implementation complexity gain coefficient is updated based on the formula .

[0076] In some embodiments, the data transmission time consumption has an impact on the benefit of data reuse, and therefore a transmission time threshold is set to quantify the impact. For example, the transmission time threshold is 0.005s. In some embodiments, if the data transmission time consumption is greater than the transmission time threshold, the implementation complexity gain coefficient is updated according to the calculation logic that the implementation complexity gain coefficient decreases as the data transmission time consumption increases.

[0077] Figure 10 FIG. 2 is a structural schematic diagram of an embodiment 2000 of a comparative analysis system of a textile raw material carbon footprint quantification method according to the present application. As shown in the figure, the embodiment 2000 comprises a fiber property analysis module 201, a method library matching module 202, a multi-dimensional evaluation and weight configuration module 203, a hybrid material method combination module 204, and a resource reuse optimization module 205. Figure 10

[0078] In some embodiments, the fiber property analysis module 201 performs fiber property analysis, specifically including: the fiber property analysis module 201 first receives the product bill of materials data, then analyzes the mass proportion and material attributes of each fiber type, and generates a fiber property vector. The material attributes at least include fiber source category, production process type, and transportation radius.

[0079] Optionally, the fiber types include biological fibers and chemical fibers. Optionally, the biological fibers include plant fibers and animal fibers, and the plant fibers and animal fibers include various types of fibers. For example, the plant fibers include seed fibers represented by cotton and bast fibers represented by hemp fibers, and the animal fibers include hair fibers (such as wool, rabbit hair, camel hair) and excretion fibers (such as silk). For example, the chemical fibers include regenerated fibers (semi-synthetic fibers such as regenerated cellulose and regenerated protein) and synthetic fibers (such as polyester fibers).​

[0080] In some embodiments, the method library matching module 202 performs method library matching, specifically including: the method library matching module 202 matches a candidate method set from a pre-constructed carbon footprint quantification method library according to the fiber characteristic vector, each method in the method library being associated with four-dimensional evaluation indexes, including an accuracy index A, a cost index C, an implementation complexity index D, and a data timeliness index T.

[0081] In some embodiments, the construction process of the carbon footprint quantification method library includes: collecting historical carbon accounting data, the historical data including fiber raw material procurement records, production process parameters, and carbon accounting raw data; establishing a mapping relationship between fiber types and quantification methods, and labeling the applicable fiber attribute range and four-dimensional evaluation indexes for each quantification method in the method library.

[0082] In some embodiments, the multi-dimensional evaluation and weight configuration module 203 performs multi-dimensional evaluation and weight configuration, dynamically configures a weight matrix of the four-dimensional evaluation indexes based on target market compliance requirements and enterprise production strategies, and calculates the comprehensive score of each candidate method.

[0083] In some embodiments, the multi-dimensional evaluation and weight configuration module 203 analyzes the carbon regulatory policy text and outputs a policy intensity coefficient P ∈ [0, 1]; analyzes the recycled material proportion R ∈ [0, 1] in the fiber characteristic vector generated by the fiber characteristic analysis module 201; selects a preset basic weight template according to the product market positioning ; based on the policy intensity coefficient P and the recycled material proportion R, corrects the index weight in the basic weight template to obtain an initial corrected weight; determines the relevant real-time transaction market based on the product market positioning, and dynamically corrects the initial corrected weight according to the real-time transaction market.

[0084] In some embodiments, the mixed material method combination module 204 derives a mixed material method combination, specifically including: if the material list analyzed by the fiber characteristic analysis module 201 contains multiple fiber types, selecting the optimal quantification method for each fiber type from the method library constructed by the method library matching module 202 to generate a method combination scheme for the mixed material. In some embodiments, the mixed material method combination module 204 selects the method combination corresponding to the mixed material based on the corrected weight value and the comprehensive score derived by the multi-dimensional evaluation and weight configuration module 203, as the method combination scheme for the mixed material.

[0085] In some embodiments, the resource reuse optimization module 205 optimizes resource reuse, specifically including: analyzing the reusability of data acquisition equipment and computing resources among different methods in the method combination scheme generated by the mixed material method combination module 204, and outputting a final carbon footprint accounting scheme. Optionally, the reusable resource types include equipment sharing and data reuse.

[0086] In some embodiments, the process of the resource reuse optimization module 205 to derive the final carbon footprint accounting scheme includes: According to the method combination scheme, calculate the reuse gain coefficient of each method: Wherein, is the resource reuse degree of method i, is the cost correction coefficient, is the implementation complexity correction coefficient; According to the reuse gain coefficient, construct the decision matrix M with the corrected dimensions, wherein , , , are the original indicators of method i in the four dimensions of accuracy indicator A, cost indicator C, implementation complexity indicator D and data timeliness indicator T, respectively, , are the cost gain coefficient and the implementation complexity gain coefficient of method i, respectively, and the decision matrix M is: Calculate the comprehensive utility U based on the original indicators and the decision matrix of each method in the method combination scheme, and the formula is: Take the scheme with the highest value of the comprehensive utility U as the final carbon footprint accounting scheme.

[0087] The above has introduced the embodiments of the present application in detail, and the principle and implementation mode of the present application have been described by applying specific examples. The above embodiment descriptions are only used to help understand the method of the present application and its core idea. Meanwhile, the changes or deformations made by the skilled in the art according to the idea of the present application, based on the specific implementation mode and application range of the present application, all belong to the protection range of the present application. In summary, the content of the present specification should not be understood as a limitation of the present application.

Claims

1. A comparative analysis method for quantifying the carbon footprint of textile raw materials, characterized in that, Includes the following steps: S1 Fiber Characteristic Analysis: Receives the product's bill of materials data, analyzes the mass percentage and material properties of each fiber type, and generates a fiber characteristic vector. The material properties include at least the fiber source category, production process type, and transportation radius. S2 Method Library Matching: Based on the fiber characteristic vector, a set of candidate methods is matched from a pre-built carbon footprint quantification method library. Each method in the method library is associated with a four-dimensional evaluation index, including accuracy index A, cost index C, implementation complexity index D, and data timeliness index T. S3 Multidimensional Assessment and Weight Configuration: Based on the target market compliance requirements and the company's production strategy, the weight matrix of the four-dimensional assessment indicators is dynamically configured, and the comprehensive score of each candidate method is calculated. S4 Hybrid Material Method Combination: If the material list parsed in step S1 contains multiple fiber types, select the optimal quantification method from the method library in S2 for each fiber type to generate a hybrid material method combination scheme; S5 Resource Reuse Optimization: Analyze the reusability of data acquisition equipment and computing resources among different methods in the method combination scheme generated in step S4, and output the final carbon footprint accounting scheme.

2. The method according to claim 1, characterized in that, The construction of the method library in step S2 specifically includes: Collect historical carbon accounting data, including fiber raw material purchase records, production process parameters, and raw carbon accounting data; Establish a mapping relationship between fiber type and quantification method, and label the applicable fiber attribute range and the four-dimensional evaluation index for each quantification method in the method library.

3. The method according to claim 2, characterized in that, The updates to the method library include: Based on the mapping relationship between fiber type and quantification method established in claim 2, when a new fiber type is input, the similarity between the new fiber type and the fiber types already existing in the method library is calculated; The existing fiber type with the highest similarity is selected as the benchmark fiber type, and the corresponding method of the benchmark fiber type is used as the benchmark method. Based on the difference between the new fiber type and the benchmark fiber type, the four-dimensional evaluation index of the benchmark method is modified.

4. The method according to claim 3, characterized in that, The four-dimensional evaluation metrics of the modified benchmark method include: Based on the difference between the new fiber type and the benchmark fiber type, a first corrected evaluation index is generated through a linear regression model; If the new fiber type has low-carbon certification attributes, a baseline emission factor is obtained, and the baseline emission factor is corrected according to the certification level included in the low-carbon certification attributes to obtain a corrected emission factor. The second revised assessment index is calculated based on the revised emission factor and the first revised assessment index, and the second revised assessment index is used as the revised four-dimensional assessment index.

5. The method according to claim 1, characterized in that, In step S3, the target market compliance requirements include carbon regulatory policy text, and the enterprise production strategy includes product market positioning and corresponding preset weight template. Based on the target market compliance requirements and the enterprise production strategy, the weight matrix of the four-dimensional evaluation indicators is dynamically configured by parsing the carbon regulatory policy text and outputting the policy intensity coefficient P∈[0,1]. The proportion of recycled materials R∈[0,1] in the fiber characteristic vector generated in step S1 is analyzed; Select a preset basic weight template based on the product market positioning. ; Based on the policy intensity coefficient P and the proportion of recycled materials R, the basic weight template is revised. The indicator weights in the data are used to obtain the initial corrected weights; Based on the product market positioning, a relevant real-time trading market is determined, and the initial adjustment weight is dynamically adjusted according to the real-time trading market.

6. The method according to claim 5, characterized in that, The process of parsing the carbon regulatory policy text and outputting the policy intensity coefficient P∈[0,1] includes: Establish a policy keyword database, including both encouragement-related and mandatory terms; Based on the policy keyword database, the policy intensity P is calculated using the TF-IDF algorithm, following the formula: Among them, S timeliness indicates the urgency of the policy taking effect.

7. The method according to claim 6, characterized in that, The step of parsing the proportion of recycled material R∈[0,1] in the fiber characteristic vector generated in step S1 includes: Differential coefficients are set for different types of recycled materials. ,in Represents the conversion factor for the i-th type of recycled material; Establish a credibility verification mechanism for recycled materials, including: for recycled materials with blockchain traceability, a credibility t = For recycled materials declared in the traditional way, the credibility t= ; According to the difference coefficient The proportion of recycled materials is calculated using the aforementioned recycled material credibility verification mechanism, and the formula is as follows: .

8. The method according to claim 1, characterized in that, The resource reuse optimization step S5 specifically includes: Based on the method combination scheme, calculate the multiplexing gain coefficient of each method: in, Let i be the resource reuse degree. This is the cost adjustment factor. To implement a complexity correction factor; Construct a decision matrix M with modified dimension based on the multiplexing gain coefficients, where , , , These are the original metrics for method i across four dimensions: accuracy (A), cost (C), implementation complexity (D), and data timeliness (T). , The decision matrix M represents the cost gain coefficient and implementation complexity gain coefficient for method i, respectively. The comprehensive utility U is calculated based on the original indicators of each method in the method combination scheme and the decision matrix, using the following formula: The scheme with the highest comprehensive utility U value will be used as the final carbon footprint accounting scheme.

9. The method according to claim 8, characterized in that, The calculation of the multiplexing gain coefficient of each method according to the method combination scheme includes: Reusable resource types include device sharing and data reuse. The type of reusable resource is determined according to the following formula. ,in Represents the edge weight coefficient. An indicator function for the existence of edges: Obtain the data transmission time of the data reuse edge. During the data transmission time When the transmission time exceeds the threshold, based on the formula Update implementation complexity gain coefficient .

10. A comparative analysis system for quantifying the carbon footprint of textile raw materials, used to perform the method described in any one of claims 1-9, characterized in that, include: The fiber characteristic analysis module is used to receive the product's bill of materials data, analyze the mass percentage and material properties of each fiber type, and generate a fiber characteristic vector. The material properties include at least the fiber source category, production process type, and transportation radius. The method library matching module is used to match a set of candidate methods from a pre-built carbon footprint quantification method library based on the fiber characteristic vector. Each method in the method library is associated with a four-dimensional evaluation index, including accuracy index, cost index, implementation complexity index and data timeliness index. The multidimensional evaluation and weight configuration module is used to dynamically configure the weight matrix of the four-dimensional evaluation indicators based on the target market compliance requirements and the company's production strategy, and to calculate the comprehensive score of each candidate method. The hybrid material method combination module is used to select the optimal quantification method from the method library of S2 for each fiber type if the material list analyzed by the fiber property analysis module contains multiple fiber types, and generate a hybrid material method combination scheme. The resource reuse optimization module is used to analyze the reusability of data acquisition equipment and computing resources among different methods in the method combination scheme generated by the hybrid material method combination module, and output the final carbon footprint accounting scheme.

Images