Method and System for Analyzing Uncertainty of Product Carbon Footprint Based on Manufacturing Scenario Constraints

Abstract

The present invention relates to the technical field of carbon footprint uncertainty analysis, and discloses a method and system for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints. The method includes: removing noise from the initial sample set to obtain a to-be-predicted sample set after noise removal; inputting the to-be-predicted sample set after noise removal into a carbon footprint regression prediction model to obtain the carbon footprint prediction value corresponding to each to-be-predicted sample in the to-be-predicted sample set after noise removal; calculating the coefficient of dispersion according to the carbon footprint prediction value; if the coefficient of dispersion is higher than the set threshold, it indicates that the parameters of the current to-be-predicted sample need to be modified; after obtaining all the to-be-predicted samples whose parameters need to be modified, screening out the key parameters for each to-be-predicted sample, modifying the key parameters to obtain the modified to-be-predicted samples, and further obtaining the modified to-be-predicted sample set, and carrying out uncertainty analysis based on the manufacturing scenario information and constraints to ensure the rationality and accuracy of the results.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon footprint uncertainty analysis in the processing and manufacturing stage of electromechanical products, and particularly to a method and system for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints. Background Art

[0002] The carbon emissions of the manufacturing industry have an important impact on issues such as global warming and climate change. The accuracy of carbon emissions results in the manufacturing stage plays a decisive role in subsequent optimization decisions. However, the wide range of data sources and complex information associations in the manufacturing stage make the credibility of carbon emissions results questionable, which may lead to decision-making errors. Therefore, it is urgent to carry out the uncertainty analysis of carbon footprint in the manufacturing stage to improve the credibility of the evaluation results. The uncertainty analysis of carbon footprint in the manufacturing stage includes two key contents: (1) comprehensive identification and expression of information related to carbon footprint in the manufacturing stage; (2) reasonable uncertainty analysis methods. However, existing research mostly focuses on the influencing factors directly related to carbon emissions in the manufacturing stage, and some factors with a greater impact on carbon emissions have not been considered (such as energy type, geographical location, etc.), and the complex association constraints between information lack analysis and expression, seriously affecting the accuracy of carbon emissions evaluation results. In addition, the uncertainty analysis method focuses on analyzing the trends and laws of output results, ignoring the accuracy of input information. Summary of the Invention

[0003] To solve the deficiencies of the prior art, the present invention provides a method and system for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints; identifying and expressing the information and association constraints in the manufacturing stage, and carrying out uncertainty analysis based on the manufacturing scenario information and constraints to ensure the rationality and accuracy of the results.

[0004] On the one hand, a method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints is provided, including:

[0005] (1): Obtain a set of samples to be predicted, where the set of samples to be predicted includes: a number of samples to be predicted, and each sample to be predicted includes: a number of attribute parameters to be predicted under a manufacturing scenario; randomly sample the set of samples to be predicted to obtain an initial set of samples; remove noise from the initial set of samples to obtain a set of samples to be predicted after noise removal;

[0006] (2): Input the set of samples to be predicted after noise removal into a carbon footprint regression prediction model to obtain the carbon footprint prediction value corresponding to each sample to be predicted in the set of samples to be predicted after noise removal;

[0007] (3): Calculate the coefficient of variation based on the predicted carbon footprint value. If the coefficient of variation is higher than the set threshold, it indicates that the parameters of the current sample to be predicted need to be modified. After obtaining all the samples to be predicted whose parameters need to be modified, proceed to (4). If the coefficient of variation is lower than the set threshold, it indicates that the parameters of the current sample to be predicted do not need to be modified.

[0008] (4): Screen out the key parameters for each sample to be predicted, modify the key parameters, obtain the modified samples to be predicted, and then obtain the set of modified samples to be predicted, and return to (1).

[0009] On the other hand, a product carbon footprint uncertainty analysis system based on manufacturing scenario constraints is provided, including:

[0010] An acquisition module, which is configured to: acquire a set of samples to be predicted, the set of samples to be predicted including: several samples to be predicted, and each sample to be predicted including: several attribute parameters to be predicted under a manufacturing scenario; perform random sampling on the set of samples to be predicted to obtain an initial set of samples; perform noise elimination on the initial set of samples to obtain a set of samples to be predicted after noise elimination.

[0011] A prediction module, which is configured to: input the set of samples to be predicted after noise elimination into a carbon footprint regression prediction model to obtain the predicted carbon footprint value corresponding to each sample to be predicted in the set of samples to be predicted after noise elimination.

[0012] A calculation module, which is configured to: calculate the coefficient of variation based on the predicted carbon footprint value. If the coefficient of variation is higher than the set threshold, it indicates that the parameters of the current sample to be predicted need to be modified. After obtaining all the samples to be predicted whose parameters need to be modified, enter the modification module. If the coefficient of variation is lower than the set threshold, it indicates that the parameters of the current sample to be predicted do not need to be modified.

[0013] A modification module, which is configured to: screen out the key parameters for each sample to be predicted, modify the key parameters, obtain the modified samples to be predicted, and then obtain the set of modified samples to be predicted, and return to the acquisition module.

[0014] The above technical solution has the following advantages or beneficial effects:

[0015] Comprehensively analyze the activity characteristics in the manufacturing stage, propose the concept of manufacturing scenarios, identify and analyze the attribute information of each manufacturing scenario and the associated constraints between the information, and achieve the standardized expression of the carbon footprint-related information and constraints in the manufacturing stage. Use the random forest algorithm to quickly generate a prediction model for the carbon footprint. Based on the distribution function and associated constraints of the manufacturing scenario attribute information, generate sample data through Monte Carlo analysis, and perform noise screening on the sample data based on the comprehensive sorting method of the importance of manufacturing scenario attribute parameters. Considering the completion time, cost, environment, and constraint indicators of the manufacturing process, judge the solution priority. Solve the constraint satisfaction problem according to the priority order, efficiently screen and order the initial sample set that meets the constraints, and ensure the rationality and accuracy of the input sample data. Further, conduct uncertainty analysis on the carbon footprint results through the coefficient of variation, identify the key parameters that have a greater impact on carbon emissions in the manufacturing stage, and reduce the uncertainty of carbon emissions and improve the credibility of the results by improving and enhancing the key parameters.

[0016] The present invention proposes the concept of manufacturing scenarios to comprehensively express the carbon footprint-related information, identifies and defines the attribute information of manufacturing scenarios and their associated constraints, and realizes the comprehensive and accurate description of the information in the manufacturing stage. Predict the carbon footprint results through the prediction model of the random forest algorithm. Further, based on the distribution function of the manufacturing scenario attribute information and its associated constraints, define the noise screening priority of the sample data according to the comprehensive sorting of the importance of manufacturing scenario attribute parameters, solve the constraint satisfaction problem according to the priority, screen and eliminate the samples that do not meet the requirements, and ensure the rationality of the input variables for uncertainty analysis. Analyze the uncertainty of the carbon footprint results through the coefficient of variation, and improve the credibility of the results by improving and enhancing the key parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings forming a part of this invention are used to provide a further understanding of the invention. The illustrative embodiments and descriptions thereof of the invention are used to explain the invention and do not constitute an improper limitation of the invention.

[0018] Figure 1 It is a flowchart of the method for the first embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0020] Manufacturing is the main method for the shaping of products or parts, and the results of its carbon emission assessment are of great significance for subsequent optimization decisions and carbon reduction. However, the extensive data sources, uncontrollable acquisition errors, and simplified assessment models in the manufacturing stage lead to doubts about the credibility of carbon emission results in this stage, which may result in decision-making mistakes. Therefore, it is urgent to develop an uncertainty analysis method for the manufacturing stage to identify the key parameters that have a greater impact on carbon emissions in the manufacturing stage. By improving and enhancing the key parameters, the credibility of the results can be improved, and the low-carbon development of the manufacturing industry can be promoted.

[0021] Uncertainty analysis can identify the impact of information data on product carbon emissions, effectively estimate the ways and trends of changes in these data, and then trace back and guide the design. It is a hot issue in carbon emission assessment research. Through the uncertainty analysis of information, the credibility and robustness of carbon emission results can be effectively evaluated. The uncertain analysis of the carbon footprint in the manufacturing stage includes two key aspects: (1) the comprehensive identification and expression of information related to the carbon footprint in the manufacturing stage; (2) a reasonable uncertainty analysis method. In terms of information identification and expression, existing research mostly focuses on the influencing factors directly related to carbon emissions in the manufacturing stage, such as processing power, processing time, etc. Some factors that have a greater impact on carbon emissions have not been considered (for example, energy type, geographical location, etc.). In addition, there are complex correlation relationships and constraints among various pieces of information, and research on the expression of information correlation constraints has not been carried out, restricting the information traceability and seriously affecting the accuracy of carbon emission assessment. In the design of information uncertainty analysis and processing, probability theory, fuzzy theory, and grey system theory have been widely applied, mainly through various methods such as Monte Carlo analysis and multiple regression analysis. Uncertainty results are mainly measured by the changes in input information and output results, and different carbon emission results correspond to different manufacturing scenarios. The rationality and accuracy of input variables have a decisive impact on the results of uncertainty analysis. However, existing uncertainty analysis mostly focuses on the analysis and improvement of output results, lacks research on the rationality and accuracy of input information, and ignores the complex correlation constraints among the manufacturing scenario attribute information in the manufacturing process, making it difficult to ensure the rationality and accuracy of uncertainty analysis results.

[0022] Example 1, as Figure 1 shown, this example provides a method for uncertainty analysis of product carbon footprint based on manufacturing scenario constraints, including:

[0023] S101: Obtain a set of samples to be predicted, where the set of samples to be predicted includes: several samples to be predicted, and each sample to be predicted includes: several attribute parameters to be predicted under a manufacturing scenario; randomly sample the set of samples to be predicted to obtain an initial set of samples; remove noise from the initial set of samples to obtain a set of samples to be predicted after noise removal;

[0024] S102: Input the set of samples to be predicted after noise removal into the carbon footprint regression prediction model to obtain the carbon footprint prediction values corresponding to each sample to be predicted in the set of samples to be predicted after noise removal;

[0025] S103: Calculate the coefficient of dispersion based on the carbon footprint prediction values; if the coefficient of dispersion is higher than the set threshold, it indicates that the parameters of the current sample to be predicted need to be modified; after obtaining all the samples to be predicted whose parameters need to be modified, proceed to S104; if the coefficient of dispersion is lower than the set threshold, it indicates that the parameters of the current sample to be predicted do not need to be modified;

[0026] S104: Screen out the key parameters for each sample to be predicted, modify the key parameters to obtain the modified samples to be predicted, and further obtain the set of modified samples to be predicted, then return to S101.

[0027] Furthermore, in S101: Randomly sample the set of samples to be predicted to obtain an initial sample set, including:

[0028] (1-1): Confirm the probability distribution function of the manufacturing scenario attribute parameters in the set of samples to be predicted;

[0029] (1-2): According to the probability distribution function of the manufacturing scenario attribute parameters, use Monte Carlo analysis to randomly sample the manufacturing scenario attribute parameters to obtain an initial sample data set.

[0030] Furthermore, in (1-1): Confirm the probability distribution function of the manufacturing scenario attribute parameters, and the probability distribution function includes: the uniform distribution function of discrete variables and the normal distribution function of continuous variable pairs.

[0031] Furthermore, in (1-2): According to the probability distribution function of the manufacturing scenario attribute parameters, use Monte Carlo analysis to randomly sample the manufacturing scenario attribute parameters to obtain an initial sample data set, including:

[0032] If the attribute parameter conforms to the normal distribution, randomly obtain values according to the normal distribution;

[0033] If the attribute parameter conforms to the uniform distribution, randomly obtain values according to the uniform distribution.

[0034] It should be understood that based on the theoretical basis of probability theory, deeply analyze the characteristics of various attribute information in the manufacturing scenario, and deduce the probability distribution functions of these attributes through mathematical statistical fitting of the data.

[0035] These manufacturing scenario attributes can be divided into discrete type and continuous type, and the distribution patterns include uniform distribution, normal distribution, etc.

[0036] Among them, information such as process type, geographical location, energy type, and equipment type in different manufacturing scenario attributes belongs to discrete variables, and their distribution function conforms to the discrete uniform distribution:

[0037] , ;

[0038] Among them, is the discrete variable attribute parameter of the input that conforms to the uniform distribution, is the minimum value of the parameter, is the maximum value of the parameter.

[0039] The quantitative data of each manufacturing scenario, including processing power, processing time, etc., its uncertainty mainly comes from the error between the design information data and the actual processing data, and the distribution of these continuous variables conforms to the normal distribution:

[0040]

[0041] Among them is the continuous variable attribute parameter of the input that conforms to the normal distribution.

[0042] Using Monte Carlo analysis, according to the probability distribution function of the established manufacturing scenario attribute information, random sampling is performed on the input parameters to determine the number of iterations N of the Monte Carlo simulation, and a large number of input parameters that meet the specified requirements are generated .

[0043] Furthermore, the S101: perform noise elimination on the initial sample set to obtain the sample set to be predicted after noise elimination, including:

[0044] (2-1): For each sample to be predicted, calculate the comprehensive weight of each attribute parameter to be predicted;

[0045] (2-2): Sort all the attribute parameters to be predicted in descending order according to the comprehensive weight, and judge whether each attribute parameter to be predicted meets the association constraint in turn according to the order of the weight from large to small;

[0046] (2-3): If it meets the association constraint, continue to judge the next parameter;

[0047] (2-4): If it does not meet the association constraint, delete the current sample to be predicted from the sample set to be predicted to obtain the sample set to be predicted after noise elimination.

[0048] Furthermore, the (2-1): For each sample to be predicted, calculate the comprehensive weight of each attribute parameter to be predicted, including:

[0049] The comprehensive weight of the manufacturing scenario attribute parameter is:

[0050] .

[0051] Among them, represents the importance of the th attribute parameter pair for the completion time, represents the importance of the th attribute parameter pair for the cost, represents the importance of the th attribute parameter pair for the environment, represents the constraint importance of the th attribute parameter pair.

[0052] Among them, the importance of the th attribute parameter pair for the completion time , is obtained by inputting the scoring matrix of the th attribute parameter pair for the completion time into the analytic hierarchy process; for the time objective, the 1-9 scale method is used to compare the scenario attribute parameters pairwise. Among them, 1 means that the two parameters are equally important, 3 means that one parameter is slightly more important than the other, 5 means that one parameter is significantly more important than the other, 7 means that one parameter is strongly more important than the other, 9 means that one parameter is extremely more important than the other; 2, 4, 6, 8 are the intermediate values of the above adjacent judgments. The judgment matrix , represents the degree of importance of the attribute parameter relative to the attribute parameter . For example, the following judgment matrix P can be obtained:

[0053]

[0054] The importance of the th attribute parameter pair for the cost , is obtained by inputting the scoring matrix of the th attribute parameter pair for the cost into the analytic hierarchy process; the matrix acquisition method is the same as that for the time objective.

[0055] The importance of the th attribute parameter pair for the environment is obtained by inputting the scoring matrix of the th attribute parameter pair for the environment into the analytic hierarchy process; the matrix acquisition method is the same as that for the time objective.

[0056] The constraint importance of the th attribute parameter. By analyzing the number of constraints of each manufacturing scenario attribute parameter, the number of constraints imposed on the th manufacturing scenario attribute parameter is , then the The constraint importance of the manufacturing scenario attribute parameters is .

[0057] Furthermore, the association constraint includes: range constraint , sequence constraint and combination constraint .

[0058] Range constraint refers to the range limitation of each attribute parameter affected by the external environment, etc. in order to complete the specified processing task, which is generally expressed in the form of an inequality, as shown in the following formula. For example, in order to meet the delivery date requirement, the processing time range is restricted.

[0059] Range constraint , and its expression is:

[0060] ;

[0061] In the formula, is the information of the i th scenario attribute parameter; refers to the minimum threshold of the attribute information; refers to the maximum threshold of the attribute information.

[0062] Sequence constraint , which means that there is a certain sequence relationship between the scenario attribute parameters. When a certain parameter is determined, the range of another parameter is also determined immediately. For example, when the equipment type of the equipment attribute is determined, the processing power range of the parameter attribute is also determined immediately.

[0063] Sequence constraint , and its expression is:

[0064]

[0065] In the formula, , respectively represent the attribute parameters of the manufacturing scenario, and [a, b] is the parameter range corresponding to the sequence constraint.

[0066] Combination constraint , which means that there are two or more manufacturing scenario attribute parameters jointly determining the value of another attribute parameter. For example, the energy type, emission coefficient of the energy attribute and the geographical location of the description attribute. When the energy type and geographical location are both determined, the corresponding emission coefficient value can be obtained.

[0067] Combination constraint , and its expression is:

[0068]

[0069] In the formula, , and respectively represent the attribute parameters of the manufacturing scenario.

[0070] The present invention converts the association constraints between the manufacturing scenario attributes into a constraint satisfaction problem, and screens the sample data by solving the constraint satisfaction problem.

[0071] The constraint satisfaction problem mainly consists of a set of variables, the optional value range of each variable, and a finite constraint set of variables, and is mainly represented by a triple model, that is , where: , represents the variables of the problem, is the number of variables. , where represents the variable 's optional value range. , represents the constraints between variables.

[0072] For each sample, the variable is the manufacturing scenario attribute parameter ; the value range D is the value of each manufacturing scenario attribute parameter; the constraint C is the constructed constraint model , , .

[0073] It should be understood that in order to quickly and efficiently screen out each sample that does not meet the constraints, a comprehensive sorting method for the importance of manufacturing scenario attribute parameters is proposed to judge and solve the priority of constraint solving. This method faces each attribute parameter of the manufacturing scenario, comprehensively analyzes the constraints between parameters and the impact of each parameter on production at present, and conducts research based on the importance of attribute parameters.

[0074] During the product manufacturing process, the enterprise comprehensively considers the delivery time, benefits, and environmental issues, and sets three main production goals, including the completion time Time, cost Cost, and environment Envi. Among them, the completion time refers to the total time required for a product or component to complete all processes from the start of manufacturing to meet the predetermined quality standards. The cost refers to the total cost required to manufacture one or more components, including various expenditures such as raw material procurement, labor input, equipment depreciation, energy consumption, management fees, and quality control during the production process. The environment mainly refers to the environmental impact composed of the total amount of greenhouse gases such as carbon dioxide directly or indirectly released into the atmosphere during the process of manufacturing components due to activities such as energy consumption, raw material use, and waste treatment, which reflects the contribution degree of the production process to climate change. During the manufacturing process, the manufacturing scenario attribute parameters have different degrees of influence on the production goals. Facing the manufacturing scenario attribute parameters and production goals, three judgment matrices for the completion time, cost, and environment are respectively constructed through the analytic hierarchy process. By calculating the importance of each attribute parameter to each goal, that is, the weight belonging to each goal: , , . Among them ; . represents the th manufacturing scenario attribute parameter; is the number of manufacturing attribute parameters; , , respectively represent the weights of the th manufacturing scenario attribute parameter for the completion time, cost, and environment.

[0075] In addition, the constraints among the manufacturing scenario attribute parameters also have an important impact on the production goals. Based on the constructed constraint model, analyze the constraints of each manufacturing scenario attribute parameter. If the number of constraints imposed on the th manufacturing scenario attribute parameter is Ni, then the constraint weight of the th manufacturing scenario attribute parameter is .

[0076] Calculate and sort the weights of all manufacturing scenario attribute parameters through the above method . The higher the ranking, the greater the weight, indicating that the parameter The more important it is for production, the higher its solution priority. Nodes with larger comprehensive weights in the input parameters are processed first, and the constraint problems of this parameter in each sample are solved. Check whether the constraints meet the defined constraint conditions to reduce the complexity of the solution. Specific process: Sort the variables in each sample data by priority, expand the constraint check for variables with larger comprehensive weights, trace back to the associated constraints through forward verification, and determine whether the constraint requirements are met. If a sample does not meet a certain constraint condition, it is deleted from the sample set. It can be implemented using the deletion operation of the data structure, such as using the remove method in a list or by reconstructing a new list that only contains samples that meet the constraints.

[0077] After screening, a sample set that meets the manufacturing scenario constraints is obtained , According to the prediction model obtained by random forest training, calculate its corresponding carbon footprint value . Combine the sample and its carbon footprint value into a new data pair .

[0078] Furthermore, S102: Input the set of samples to be predicted after removing noise into the carbon footprint regression prediction model to obtain the carbon footprint prediction values corresponding to each sample to be predicted in the set of samples to be predicted after removing noise; wherein, the process of obtaining the carbon footprint regression prediction model includes:

[0079] Take all historical attribute parameters under the manufacturing scenario corresponding to each sample in the training set as the input of the random forest model, and take the historical carbon footprint corresponding to the current sample as the output of the random forest model, and train the random forest model to obtain the carbon footprint regression prediction model.

[0080] It should be understood that according to the prediction model obtained by random forest training, calculate its corresponding carbon footprint value:

[0081] .

[0082] Combine the sample and its carbon footprint value into a new data pair .

[0083] It should be understood that the sample set and its carbon footprint data pairs are divided into a training set and a test set. Usually, they can be divided according to the ratio of 70% for the training set and 30% for the test set. Standardize the data of the training set and the test set so that the numerical ranges of different variables have similar scales, which is convenient for the random forest algorithm to learn and predict. Determine the parameters of the random forest model, such as the number of decision trees being 100 and the maximum depth of each decision tree being 10. Use the training set data to train the random forest model to let the model learn the relationship between the sample features (manufacturing scenario variables) and the carbon footprint values. Use the trained random forest model to predict the carbon footprint of the test set to obtain the predicted carbon footprint value Carbon. Finally, evaluate the performance of the prediction model through the mean absolute error A regression prediction model dedicated to estimating carbon emissions in different processing and manufacturing stages is constructed. This method can achieve efficient and accurate prediction of carbon emissions in the manufacturing process.

[0084]

[0085] In the formula, is the number of input samples; is the actual carbon footprint result; The predicted carbon footprint result.

[0086] Random forest is a powerful machine learning algorithm that uses a set of different decision trees for classification or regression prediction. By aggregating and combining the prediction results of these trees, the random forest can significantly improve the accuracy and robustness of the overall model and reduce the risk of overfitting. Its core principle mainly revolves around a set of decision trees constructed from samples randomly sampled with replacement and randomly selected features.

[0087] However, in view of the diversity and complex correlation of information in the manufacturing scenario, although the random sampling mechanism in the random forest is simple and effective, when the number of samples is large, the required calculation and processing time are long, and the solution efficiency is low. In addition, there are also constraint problems among the manufacturing scenario information. For example, there is a constraint relationship between the equipment type and the processing power of the parameter attributes There are differences in the processing power ranges corresponding to equipment with different advanced levels. The sampling method of the random forest has not considered the constraints among the information, and it is difficult to ensure the rationality of the training samples, which affects the accuracy of the prediction results. Therefore, the present invention adopts the Monte Carlo analysis method, combines the constraint satisfaction problem with Monte Carlo sampling to ensure the accuracy of the sample data, and effectively supports the prediction and uncertainty analysis of the carbon footprint.

[0088] Furthermore, the training set includes: a number of samples, and each sample includes: a number of historical attribute parameters under a manufacturing scenario and the historical carbon footprint under this manufacturing scenario.

[0089] It should be understood that the carbon footprint refers to the total amount of greenhouse gas emissions directly and indirectly generated by an activity, product or individual throughout its life cycle, mainly carbon dioxide, but also including other greenhouse gases such as methane and nitrous oxide, which are usually converted to carbon dioxide equivalents according to their global warming potential values).

[0090] Combined with diverse data sources integrating on-site collection and enterprise ERP / CAM / CAPP systems, collect the attribute data of the manufacturing scenario, including descriptive attributes, processing attributes, energy attributes, parameter attributes and equipment attributes, ensure the accuracy and integrity of the data, and use it as the input variable for uncertainty analysis The output variable is the carbon emissions corresponding to each manufacturing scenario .

[0091] Furthermore, several historical attribute parameters under a manufacturing scenario include:

[0092] Descriptive attributes, processing attributes, energy attributes, parameter attributes and equipment attributes;

[0093] The descriptive attributes include: geographical location and time;

[0094] The processing attributes include: processing types such as forging, casting and turning;

[0095] The energy attributes include: emission factors and energy types such as solar power generation, thermal power generation and hydropower generation;

[0096] The parameter attributes include: processing time and processing power;

[0097] The equipment attributes include: equipment type and equipment life.

[0098] It should be understood that product processing and manufacturing is a process of transforming raw materials or components into finished or semi-finished products with specific functions, shapes, and qualities through a series of technological operations and processing. This process generally includes multiple stages such as the selection and preparation of raw materials, processing and shaping, assembly and debugging, etc. Carbon emissions are generated in each stage of the processing and manufacturing process, and their magnitudes are affected by various factors, including differences in process input and output inventories, specifically involving the consumption of resources and energy as well as gas emissions. In different scenarios, there are significant differences in the state of the product and its external environment, which have a profound impact on the collected carbon emission data. These data are closely related to specific scenarios, and each scenario generates different data sets. For example, even under the same heat treatment process, using different processing equipment or operating powers may result in different carbon emission results. Therefore, in order to systematically and accurately depict the green attributes in the manufacturing process, the present invention proposes the concept of "manufacturing scenario". This concept can uniformly express and describe all relevant information that affects carbon emissions in the manufacturing process, thereby contributing to the accurate assessment of carbon emissions.

[0099] A manufacturing scenario refers to the characterization of activities occurring in various processing and manufacturing processes, including time, location, environment, activities, etc. The process can be the entire technical process or consist of several process segments. The time, location, and environmental background data of the processing and manufacturing stage are closely related to the resource attributes, energy attributes, and environmental emission attributes of the product. In order to improve the accuracy and reliability of these data, the attributes of the manufacturing scenario are divided into descriptive attributes , processing attributes , energy attributes , parameter attributes and equipment attributes .

[0100] Descriptive attributes , including the time and geographical location of the manufacturing stage, etc., are expressed as: . The time specifies the exact moment when the activity occurs, including year, month, day, or hour and minute, which has an important impact on the storage of the process inventory data and the selection of the reference year for subsequent environmental impact assessment. The geographical location includes the place of origin, destination, or specific operating location.

[0101] Processing attributes , which describe the characteristics of the processing and manufacturing activities during execution, are mainly used to express the processing type of this process, . The corresponding processing types of this attribute include casting, forging, welding, heat treatment, machining, painting, assembly, etc.

[0102] Energy attributes , used to describe different types and forms of energy consumed in various stages such as processing and manufacturing, including energy types , emission factor , . These energy resources can be obtained directly, such as natural energy like solar energy, wind energy, geothermal energy, etc., or can be converted into useful energy through means such as processing and conversion, such as coal, oil, natural gas, electricity, etc. Among them, electricity can be further divided into thermal power, hydropower, and hybrid power, etc.

[0103] Parameter attributes , is a set of data describing various operating conditions and process parameters in the processing and manufacturing process, including processing power , processing time , expressed as . These parameters have a crucial impact on the quality and performance of the final product.

[0104] Equipment attributes , refers to the equipment types of the relevant equipment used to complete the processing task (such as machining centers, CNC machine tools, ordinary machine tools, etc.), equipment life , expressed as: .

[0105] Based on the above description, the manufacturing scenario is defined as follows:

[0106]

[0107] Among them, , , , , .

[0108] Furthermore, the S103: According to the carbon footprint prediction value, calculate the dispersion coefficient, including:

[0109] Uncertainty analysis evaluates the carbon emission results under various manufacturing scenarios, using the dispersion coefficient as the main metric to quantify the uncertainty in these results.

[0110]

[0111] In the formula, is the number of samples; is the carbon footprint result, is the mean of the carbon footprint results.

[0112] If the coefficient of variation is lower than the set threshold, it indicates that the result has high credibility and robustness. Otherwise, if the coefficient of variation is higher than the set threshold, it means that the input parameters need to be improved to enhance the robustness of the result and ensure that the coefficient of variation meets the established criteria, thereby improving the reliability of carbon emissions.

[0113] It should be understood that the above technical solution is for a more in-depth understanding of how changes in the input manufacturing scenario affect the carbon emissions of the product manufacturing process, and an uncertainty analysis of the manufacturing scenario is carried out. This uncertainty analysis includes two key aspects: (1) how to quickly and accurately obtain the output result based on the input information; (2) how to comprehensively evaluate the impact of the input on the output. The CV result determines whether the coefficient of variation meets the threshold, and this coefficient can generally be set by enterprise designers based on experience or requirements. Generally, the threshold of the coefficient of variation is set to 0.15.

[0114] Furthermore, the S104: screening out key parameters for each sample to be predicted and modifying the key parameters to obtain the modified sample to be predicted, includes:

[0115] According to the comprehensive weight of the manufacturing scenario attribute parameters , in accordance with the comprehensive weight Sort the attribute parameters in descending order, and select the top 50% of the attribute parameters as key parameters to obtain the key parameter group . These parameters have a more important potential impact on the carbon footprint compared to other parameters.

[0116] For the key parameter group, construct a multiple linear regression model:

[0117]

[0118] Among them, are the coefficients to be determined, is the error term.

[0119] Train the multiple linear regression model with the input sample data, determine the coefficients to be determined by the least squares method, analyze the linear relationship between the key parameters and the carbon footprint, and preliminarily determine the sensitivity of the key parameters.

[0120] For the multiple linear regression model, directly use the regression coefficient as the sensitivity index, the larger the absolute value of , the more sensitive the parameter. Take the attribute parameters with the absolute value of

[0121] greater than the set threshold as key parameters.

[0122] Increase or decrease the identified key parameters by 5% to 10%, recalculate the carbon footprint and evaluate the optimization effect, calculate the CV index until a better optimization effect is achieved to meet the enterprise requirements.

[0123] Example Two

[0124] This embodiment provides a product carbon footprint uncertainty analysis system based on manufacturing scenario constraints, including:

[0125] An acquisition module, which is configured to: acquire a set of samples to be predicted, the set of samples to be predicted including: a number of samples to be predicted, and each sample to be predicted including: a number of attribute parameters to be predicted under a manufacturing scenario; perform random sampling on the set of samples to be predicted to obtain an initial set of samples; perform noise elimination on the initial set of samples to obtain a set of samples to be predicted after noise elimination;

[0126] A prediction module, which is configured to: input the set of samples to be predicted after noise elimination into a carbon footprint regression prediction model to obtain the carbon footprint prediction value corresponding to each sample to be predicted in the set of samples to be predicted after noise elimination;

[0127] A calculation module, which is configured to: calculate the coefficient of variation according to the carbon footprint prediction value; if the coefficient of variation is higher than a set threshold, it indicates that the parameters of the current sample to be predicted need to be modified; after obtaining all the samples to be predicted whose parameters need to be modified, enter the modification module; if the coefficient of variation is lower than the set threshold, it indicates that the parameters of the current sample to be predicted do not need to be modified;

[0128] A modification module, which is configured to: screen out the key parameters for each sample to be predicted, modify the key parameters to obtain the modified samples to be predicted, and further obtain a set of modified samples to be predicted, and return to the acquisition module.

[0129] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints, characterized in that Including: Obtain a set of samples to be predicted, where the set of samples to be predicted includes: a number of samples to be predicted, and each sample to be predicted includes: a number of attribute parameters to be predicted in a manufacturing scenario; randomly sample the set of samples to be predicted to obtain an initial set of samples; remove noise from the initial set of samples to obtain a set of samples to be predicted after noise removal. Input the set of samples to be predicted after noise removal into the carbon footprint regression prediction model to obtain the carbon footprint prediction value corresponding to each sample to be predicted in the set of samples to be predicted after noise removal. Calculate the coefficient of dispersion based on the carbon footprint prediction value; if the coefficient of dispersion is higher than the set threshold, it indicates that the parameters of the current sample to be predicted need to be modified; after obtaining all the samples to be predicted whose parameters need to be modified, proceed to the next step; if the coefficient of dispersion is lower than the set threshold, it indicates that the parameters of the current sample to be predicted do not need to be modified. Screen out the key parameters for each sample to be predicted, modify the key parameters to obtain the modified samples to be predicted, and then obtain the modified set of samples to be predicted, and return to the step of obtaining the set of samples to be predicted. For each sample to be predicted, calculate the comprehensive weight of each attribute parameter to be predicted. Sort all the attribute parameters to be predicted in descending order according to the comprehensive weight, and sequentially determine whether each attribute parameter to be predicted meets the association constraint in descending order of the weight. The associated constraints include: range constraints , sequential constraints and combinatorial constraints ; Range constraint refers to the range limitation of each attribute parameter affected by the external environment to complete the specified processing task; range constraint and its expression is: ； In the formula, is the th scene attribute parameter information; refers to the minimum threshold of the attribute information; refers to the maximum threshold of the attribute information; Sequential constraint , which means there is a certain sequential relationship between scene attribute parameters; sequential constraint , and its expression is: ； In the formula, , respectively represent the attribute parameters of the manufacturing scenario, is the parameter range corresponding to the sequential constraint; Combined constraint refers to the situation where there are two or more manufacturing scenario attribute parameters jointly determining the value of another attribute parameter; combined constraint and its expression is: ； In the formula, , and respectively represent the attribute parameters of the manufacturing scenario.

2. The method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints according to claim 1, wherein Randomly sample the set of samples to be predicted to obtain an initial set of samples, including: (1-1): Confirm the probability distribution function of the manufacturing scenario attribute parameters in the set of samples to be predicted; the probability distribution function includes: the uniform distribution function of discrete variables and the normal distribution function of continuous variable pairs. (1-2): According to the probability distribution function of the manufacturing scenario attribute parameters, use Monte Carlo analysis to randomly sample the manufacturing scenario attribute parameters to obtain an initial sample data set; if the attribute parameter conforms to the normal distribution, randomly take values according to the normal distribution; if the attribute parameter conforms to the uniform distribution, randomly take values according to the uniform distribution.

3. The method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints according to claim 1, characterized in that, Remove noise from the initial set of samples to obtain a set of samples to be predicted after noise removal, including: If it meets the association constraint, continue to judge the next parameter. If it does not meet the association constraint, delete the current sample to be predicted from the set of samples to be predicted to obtain a set of samples to be predicted after noise removal.

4. The method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints according to claim 1, characterized in that For each sample to be predicted, calculate the comprehensive weight of each attribute parameter to be predicted, including: Comprehensive weight of manufacturing scenario attribute parameters is as follows: ； Among them, represents the importance of the th attribute parameter pair to the completion time, represents the importance of the th attribute parameter pair to the cost, represents the importance of the th attribute parameter pair to the environment, represents the constraint importance of the th attribute parameter pair.

5. The method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints according to claim 1, wherein, The several attribute parameters to be predicted in a manufacturing scenario include: description attributes, processing attributes, energy attributes, parameter attributes, and equipment attributes; the description attributes include: geographical location and time; the processing attributes include: forging, casting, and turning; the energy attributes include: solar power generation, thermal power generation, and hydropower generation; the parameter attributes include: processing time and processing power; the equipment attributes include: equipment type and equipment life.

6. The method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints according to claim 1, wherein, Calculate the coefficient of dispersion based on the carbon footprint prediction value, including: ； where M is the number of samples, is the carbon footprint result, is the mean value of the carbon footprint results.

7. The method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints according to claim 1, characterized in that, Screen out the key parameters for each sample to be predicted, modify the key parameters, and obtain the modified sample to be predicted, including: According to the comprehensive weight of manufacturing scenario attribute parameters , sort the attribute parameters in descending order according to the comprehensive weight , select the top 50% of the attribute parameters as key parameters to obtain the key parameter group ; Construct a multiple linear regression model for the key parameter group: ； Among them, is the coefficient to be determined, takes values in the range of 0, 1, 2…m, is the error term; Train the multiple linear regression model with the input sample data, determine the coefficients to be determined by the least squares method, analyze the linear relationship between the key parameters and the carbon footprint, and preliminarily determine the sensitivity of the key parameters; For the multiple linear regression model, directly use the regression coefficient as the sensitivity index. The larger the absolute value of , the more sensitive the parameter. Take the attribute parameter whose absolute value of is greater than the set threshold as the key parameter.

8. The method for analyzing the uncertainty of product carbon footprint based on manufacturing scenario constraints as described in claim 1, characterized in that, Screen out the key parameters for each sample to be predicted, and modify the key parameters, including: increasing or decreasing the identified key parameters by 5% to 10%.

9. A product carbon footprint uncertainty analysis system based on manufacturing scenario constraints, characterized in that Including: An acquisition module, which is configured to: acquire a set of samples to be predicted, the set of samples to be predicted includes: a number of samples to be predicted, and each sample to be predicted includes: a number of attribute parameters to be predicted in a manufacturing scenario; randomly sample the set of samples to be predicted to obtain an initial set of samples; remove noise from the initial set of samples to obtain a set of samples to be predicted after noise removal; A prediction module, which is configured to: input the set of samples to be predicted after noise removal into the carbon footprint regression prediction model to obtain the carbon footprint prediction value corresponding to each sample to be predicted in the set of samples to be predicted after noise removal; A calculation module, which is configured to: calculate the coefficient of dispersion according to the carbon footprint prediction value; if the coefficient of dispersion is higher than the set threshold, it means that the parameters of the current sample to be predicted need to be modified; after obtaining all the samples to be predicted whose parameters need to be modified, enter the modification module; if the coefficient of dispersion is lower than the set threshold, it means that the parameters of the current sample to be predicted do not need to be modified; A modification module, which is configured to: screen out the key parameters for each sample to be predicted, modify the key parameters, obtain the modified sample to be predicted, and further obtain the modified set of samples to be predicted, and return to the acquisition module; For each sample to be predicted, calculate the comprehensive weight of each attribute parameter to be predicted; Sort all the attribute parameters to be predicted in descending order according to the comprehensive weight, and sequentially determine whether each attribute parameter to be predicted meets the association constraint in descending order of the weight; The associated constraints include: range constraints , sequential constraints and combination constraints ; Range constraint refers to the range limitations of various attribute parameters affected by the external environment to complete the specified processing tasks; range constraint whose expression is: ； In the formula, is the th scene attribute parameter information; refers to the minimum threshold of the attribute information; refers to the maximum threshold of the attribute information; Sequential Constraint , which means there is a certain sequential relationship between scene attribute parameters; Sequential Constraint , and its expression is: ； In the formula, , respectively represent the attribute parameters of the manufacturing scenario, is the parameter range corresponding to the sequential constraint; Combined constraint refers to the situation where the values of at least two manufacturing scenario attribute parameters jointly determine the value of another attribute parameter; combined constraint and its expression is: ； In the formula, , and respectively represent the attribute parameters of the manufacturing scenario.

Images