System and method for collecting and analyzing full-life-cycle carbon footprint data of product

Abstract

The invention provides a product full-life-cycle carbon footprint data acquisition and analysis system and method, relates to the field of data acquisition and analysis, and is used for performing data analysis and semantic understanding on product full-life-cycle data by introducing an artificial intelligence data processing technology, and calculating an initial total value of carbon emission of each stage of a product. Then, a carbon emission correction factor is obtained based on the semantic information of each stage and the relation between the semantic information and the relation between the semantic information and the relation between the semantic information and the relation between the semantic information and the relation between the semantic information and the relation, so that the difference in each processing process is reflected more accurately, and dynamic changes of different product processes and external conditions are better adapted; and the total value of the carbon emission in the whole life cycle of the product, which is closer to reality, is obtained. Therefore, the real environmental influence of the product can be reflected more accurately, more reliable product full-life-cycle carbon footprint data support is provided for enterprises, and effective emission reduction measures are formulated.

Description

Technical Field

[0001] The present application relates to the field of data collection and analysis, and more specifically, to a system and method for collecting and analyzing carbon footprint data of a product throughout its life cycle. Background Art

[0002] In the context of increasingly serious global climate change, it is becoming increasingly important for companies to evaluate and manage the carbon footprint of their products throughout their life cycle. Products will generate carbon emissions at different stages of their life cycle, such as from the acquisition of raw materials, manufacturing, transportation, use to final recycling or disposal. The accumulation of these emissions forms the carbon footprint of the product, which is crucial for evaluating the environmental performance of the product. The evaluation of the carbon footprint of the product throughout its life cycle not only helps to identify and reduce carbon emissions, but also enhances the company's image of sustainable development and meets consumers' expectations for environmental responsibility.

[0003] However, the traditional product carbon footprint data collection and calculation method usually obtains the total carbon emission value by directly accumulating the carbon emissions of each stage of the product. This carbon emission accumulation method is usually based on static data, does not take into account the temporal characteristics of the product throughout its entire life cycle, and ignores the interactions and influences between different stages. For example, waste generated during the mining of raw materials may affect carbon emissions during the transportation stage, but traditional calculation methods may not fully consider these interdependent relationships. In addition, the production batches, usage conditions, recycling methods, etc. of each product may be different, which will result in different carbon emissions for each treatment even in the same type of production process. Therefore, the simple accumulation method is difficult to adapt to this variability.

[0004] Therefore, an optimized product life cycle carbon footprint data collection and analysis solution is desired. Summary of the invention

[0005] In order to solve the above-mentioned technical problems, the present application is proposed. The embodiment of the present application provides a product life cycle carbon footprint data collection and analysis system and method, which introduces more sophisticated artificial intelligence data processing technology to perform data analysis and semantic understanding of product life cycle data, and calculates the initial total carbon emissions of each stage of the product. Then, based on the semantic information of each stage and the connection between each other, the carbon emission correction factor is obtained to make corresponding corrections to the initial total carbon emissions, so as to more accurately reflect the differences in each processing process, better adapt to the dynamic changes of different product processes and external conditions, and obtain a total carbon emissions value of the product life cycle that is closer to the actual value. In this way, the real environmental impact of the product can be reflected more accurately, providing enterprises with more reliable product life cycle carbon footprint data support, and formulating effective emission reduction measures.

[0006] According to one aspect of the present application, a method for collecting and analyzing carbon footprint data of a product throughout its life cycle is provided, which comprises:

[0007] Obtain product life cycle data, including raw material acquisition data, manufacturing data, transportation data, usage data, and recycling and settlement data;

[0008] Calculate the initial total value of carbon emissions based on the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data;

[0009] After semantic coding of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data, the semantic coding features of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data are subjected to time series information propagation processing based on the decay of the principal component of the node feature to obtain the principal component time series propagation aggregation feature of the product life cycle data, including: principal component extraction of the semantic coding features of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data to obtain a sequence of principal component representation of the semantic coding features of the current phase data and a sequence of principal component representation of the semantic coding features of the historical phase data; time series decay dynamic aggregation of the sequence of principal component representation of the semantic coding features of the current phase data and a sequence of principal component representation of the semantic coding features of the historical phase data to obtain the principal component time series propagation aggregation feature of the product life cycle data;

[0010] Based on the principal component time series propagation aggregation characteristics of the product's full life cycle data, a carbon emission correction factor is generated, and the total value of the product's full life cycle carbon emissions is calculated.

[0011] According to another aspect of the present application, a product life cycle carbon footprint data collection and analysis system is provided, which includes:

[0012] A product life cycle data acquisition module, which is used to acquire product life cycle data, including raw material acquisition stage data, manufacturing stage data, transportation stage data, use stage data and recycling and settlement stage data;

[0013] A carbon emission initial total value calculation module, used to calculate the carbon emission initial total value based on the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data;

[0014] The node feature principal component decay time series information propagation processing module is used for semantically encoding the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data, and then performing time series information propagation processing based on the node feature principal component decay on the semantic coding features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data to obtain the principal component time series propagation aggregation features of the product life cycle data, including: extracting the principal components of the semantic coding features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data to obtain the sequence of the principal component representation of the semantic coding features of the current stage data and the principal component representation of the semantic coding features of the historical stage data; performing time series decay dynamic aggregation on the sequence of the principal component representation of the semantic coding features of the current stage data and the principal component representation of the semantic coding features of the historical stage data to obtain the principal component time series propagation aggregation features of the product life cycle data;

[0015] The module for calculating the total carbon emissions of a product over its entire life cycle is used to generate a carbon emission correction factor based on the time series propagation aggregation characteristics of the principal components of the product's entire life cycle data, and calculate the total carbon emissions of the product over its entire life cycle.

[0016] Compared with the prior art, the present application provides a product life cycle carbon footprint data collection and analysis system and method, which introduces more sophisticated artificial intelligence data processing technology to perform data analysis and semantic understanding of product life cycle data, and calculate the initial total carbon emissions of each stage of the product. Then, based on the semantic information of each stage and the connection between each other, the carbon emission correction factor is obtained to make corresponding corrections to the initial total carbon emissions, so as to more accurately reflect the differences in each processing process, better adapt to the dynamic changes of different product processes and external conditions, and obtain a total carbon emissions value of the product life cycle that is closer to the actual value. In this way, the real environmental impact of the product can be reflected more accurately, providing enterprises with more reliable product life cycle carbon footprint data support, and formulating effective emission reduction measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 The present invention is a flowchart of a method for collecting and analyzing carbon footprint data of a product throughout its life cycle according to an embodiment of the present application.

[0019] Figure 2 In order to semantically encode the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the usage stage data and the recycling and settlement stage data in the product life cycle carbon footprint data collection and analysis method according to an embodiment of the present application, the semantic coding features of the raw material acquisition stage data, the semantic coding features of the manufacturing stage data, the semantic coding features of the transportation stage data, the semantic coding features of the usage stage data and the semantic coding features of the recycling and settlement stage data are subjected to time series information propagation processing based on the attenuation of the principal components of node features to obtain a flowchart of the time series propagation aggregation features of the principal components of the product life cycle data.

[0020] Figure 3 A flowchart of the method for collecting and analyzing carbon footprint data for the entire life cycle of a product according to an embodiment of the present application is provided for performing temporal information propagation processing based on the attenuation of principal components of node features on the semantic coding features of data in the raw material acquisition stage, the manufacturing stage, the transportation stage, the usage stage, and the recycling and settlement stage to obtain the principal component temporal propagation aggregation features of data for the entire life cycle of the product.

[0021] Figure 4 In the product life cycle carbon footprint data collection and analysis method according to an embodiment of the present application, the semantic coding features of the data in the raw material acquisition stage, the semantic coding features of the data in the manufacturing stage, the semantic coding features of the data in the transportation stage, the semantic coding features of the data in the use stage, and the semantic coding features of the data in the recycling and settlement stage are processed by time series information propagation based on the attenuation of the principal components of node features to obtain a data flow diagram of the time series propagation aggregation features of the principal components of the product life cycle data.

[0022] Figure 5 This is a system block diagram of a product life cycle carbon footprint data collection and analysis system according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0024] The life cycle carbon footprint refers to the total carbon emissions generated by a product throughout its life cycle, from the acquisition of raw materials, manufacturing, transportation, use to final recycling or disposal. This concept covers the environmental impact of each stage of the product, emphasizing that carbon emissions at all stages must be considered when evaluating and managing the sustainability of products.

[0025] Traditional carbon footprint calculation methods usually use the direct accumulation of carbon emissions at each stage of the product to obtain the total emission value. This accumulation method based on static data fails to fully consider the temporal characteristics of the product life cycle and ignores the interaction between different stages.

[0026] In response to the above technical problems, this application proposes a method for collecting and analyzing carbon footprint data for the entire life cycle of a product. Figure 1 Flow chart of the method for collecting and analyzing product life cycle carbon footprint data according to an embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the method for collecting and analyzing carbon footprint data of the product life cycle includes: S110, acquiring product life cycle data, wherein the product life cycle data includes raw material acquisition stage data, manufacturing stage data, transportation stage data, use stage data and recycling and settlement stage data; S120, calculating the initial total carbon emission value based on the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data; S130, after semantically encoding the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data, the semantic encoding features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data are subjected to time series information propagation processing based on the attenuation of the principal component of the node feature to obtain the principal component time series propagation aggregation feature of the product life cycle data; S140, generating a carbon emission correction factor based on the principal component time series propagation aggregation feature of the product life cycle data, and calculating the total carbon emission value of the product life cycle.

[0027] In the above-mentioned product life cycle carbon footprint data collection and analysis method, the step S110 obtains product life cycle data, and the product life cycle data includes raw material acquisition stage data, manufacturing stage data, transportation stage data, use stage data and recycling and settlement stage data. It should be understood that obtaining product life cycle data can comprehensively evaluate the impact of the product on the environment throughout its life cycle, including raw material acquisition, manufacturing, transportation, use and recycling stages. These data not only help companies identify and optimize resource use and emissions in their production processes, but also ensure compliance, meet consumer expectations for sustainability, and provide a scientific basis for strategic decision-making.

[0028] In the above-mentioned product life cycle carbon footprint data collection and analysis method, the step S120 calculates the initial total value of carbon emissions based on the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data. It should be understood that by calculating and summarizing the carbon emissions of each stage, it is possible to ensure a comprehensive understanding of the carbon emissions of all important stages in the product life cycle, thereby calculating the initial total value of carbon emissions of each stage of the product, thereby providing a comprehensive environmental impact assessment for the enterprise.

[0029] Specifically, in step S120, based on the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data, the initial total carbon emission value is calculated, including: based on the raw material acquisition stage data, determining the carbon emission value of the raw material acquisition stage; based on the manufacturing stage data, determining the carbon emission value of the manufacturing stage; based on the transportation stage data, determining the carbon emission value of the transportation stage; based on the use stage data, determining the carbon emission value of the use stage; based on the recycling and settlement stage data, determining the carbon emission value of the recycling and settlement stage; calculating the sum of the carbon emission value of the raw material acquisition stage, the carbon emission value of the manufacturing stage, the carbon emission value of the transportation stage, the carbon emission value of the use stage and the carbon emission value of the recycling and settlement stage as the initial total carbon emission value. It is worth mentioning that in the raw material acquisition stage, the life cycle assessment (LCA) method is required to collect the type, quantity, acquisition method and environmental impact of the supply chain of the raw materials used, and then determine the carbon emission value of the raw material acquisition stage based on energy consumption and related carbon emission factors. Next, in the manufacturing stage, it is necessary to track information such as the types of energy used in the production process, production efficiency, waste emissions and their treatment methods to determine the carbon emission value of the manufacturing stage. Among them, in the transportation stage, data such as transportation distance, transportation method, load and fuel usage are collected to calculate the carbon emissions of the product during transportation. Furthermore, in the use stage, data such as user usage frequency, time and conditions are collected to determine the carbon emission value of the use stage. Finally, in the recycling and settlement stage, it is necessary to record the product's recycling method, treatment process and possible reuse or resource recovery to determine the carbon emission value of the recycling and processing stage.

[0030] In the above-mentioned product life cycle carbon footprint data collection and analysis method, in step S130, after semantic coding of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data, the semantic coding features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data are processed based on the time series information propagation of the node feature principal component attenuation to obtain the product life cycle data principal component time series propagation aggregation features. It should be understood that the process of semantic coding involves labeling and classifying the data so that data from different sources and types can be uniformly understood and processed. This process not only improves the readability of the data, but also promotes the correlation between data at different stages. After the data semantic coding is completed, the data at each stage can extract its semantic coding features, which include the key indicators and attributes of each stage. The extraction of these features not only covers the original information of the data, but also contains the potential relationship and mutual influence between the data, forming the node features of each stage. Next, the semantic coding features of these stages are processed by principal component attenuation based on node features to reduce the dimension and complexity of the data. Principal component analysis (PCA) is an effective data dimensionality reduction technology that makes subsequent analysis more efficient by retaining the most important feature information and removing redundant and noisy data. This process can highlight the contribution of each stage to the full life cycle data while eliminating unnecessary details to make the data more focused and intuitive. Time series information propagation processing is to dynamically analyze these reduced features to obtain trends and patterns of data changes over time. By building a time series model, enterprises can identify the changes in different life cycle stages in the time series and observe the causal relationship and influence between the stages. This information propagation processing can also capture the dynamic characteristics of data changes over time, allowing enterprises to better predict future trends. Finally, by aggregating the features of all stages, the principal component time series propagation aggregation feature of the product's full life cycle data is formed. This feature integrates the key information of each stage and provides enterprises with a comprehensive perspective to understand the full life cycle performance of the product.

[0031] Figure 2After semantic coding of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data in the product life cycle carbon footprint data collection and analysis method according to the embodiment of the present application, the semantic coding features of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data are processed based on the time series information propagation of the node feature principal component attenuation to obtain the flow chart of the time series propagation aggregation features of the principal component of the product life cycle data. Figure 2 As shown, the step S130 includes: S131, semantically encoding the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data to obtain the semantic coding features of the raw material acquisition stage data, the semantic coding features of the manufacturing stage data, the semantic coding features of the transportation stage data, the semantic coding features of the use stage data and the semantic coding features of the recycling and settlement stage data; S132, performing time series information propagation processing based on node feature principal component attenuation on the semantic coding features of the raw material acquisition stage data, the semantic coding features of the manufacturing stage data, the semantic coding features of the transportation stage data, the semantic coding features of the use stage data and the semantic coding features of the recycling and settlement stage data to obtain the principal component time series propagation aggregation features of the product's entire life cycle data.

[0032] Specifically, in step S131, the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data are semantically encoded to obtain the semantic encoding features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data. It should be understood that by semantically encoding the data at each stage in the product life cycle, it is possible to understand the semantic information from the data at different stages of the product and extract the important semantic features in these stages, which helps to identify and emphasize the important factors that affect the subsequent carbon footprint calculation, such as different production batches, different use conditions, and different recycling methods in each stage of the product life cycle, and provide a basis for the subsequent carbon emission correction and calculation tasks.

[0033] More specifically, in an embodiment of the present application, the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the usage phase data and the recycling and settlement phase data are semantically encoded, including: semantically encoding the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the usage phase data and the recycling and settlement phase data to obtain a semantic encoding vector of the raw material acquisition phase data as the semantic encoding feature of the raw material acquisition phase data, a semantic encoding vector of the manufacturing phase data as the semantic encoding feature of the manufacturing phase data, a semantic encoding vector of the transportation phase data as the semantic encoding feature of the transportation phase data, a semantic encoding vector of the usage phase data as the semantic encoding feature of the usage phase data and a semantic encoding vector of the recycling and settlement phase data as the semantic encoding feature of the recycling and settlement phase data.

[0034] Specifically, in step S132, the semantic coding features of the data in the raw material acquisition stage, the semantic coding features of the data in the manufacturing stage, the semantic coding features of the data in the transportation stage, the semantic coding features of the data in the use stage, and the semantic coding features of the data in the recycling and settlement stage are processed by time series information propagation based on the decay of the principal component of the node feature to obtain the time series propagation aggregation features of the principal component of the product life cycle data. It should be understood that the product will undergo various changes at different stages of its life cycle (for example, from raw material acquisition to final recycling and processing), and these changes have time series characteristics and will affect the subsequent calculation of carbon emissions. Therefore, the time series information propagation processing based on the decay of the principal component of the node feature can more comprehensively and accurately understand the interaction and time series correlation relationship between the data semantics of each stage in the product life cycle, so as to more accurately correct the carbon emissions. It is worth mentioning that the time series information propagation processing method based on the decay of the principal component of the node feature is to capture the trend of the time series node features of the semantics of each stage changing over time by combining the principal component analysis and the gradient decay mechanism, and use these trends for effective information propagation and aggregation. Specifically, the process of time series information propagation processing based on the attenuation of node feature principal components is to perform principal component analysis on the node feature vectors at each time node (the data semantic coding vector of the raw material acquisition stage, the data semantic coding vector of the manufacturing stage, the data semantic coding vector of the transportation stage, the data semantic coding vector of the use stage, and the data semantic coding vector of the recycling processing and settlement stage) to extract the most representative stage data semantic principal components, and then calculate the principal component change rate and its corresponding attenuation entropy factor between different stage time nodes, and further introduce time span modulation and gating function to filter important information, and finally generate a product full life cycle data principal component time series propagation aggregation vector that can reflect the dynamic changes of semantic node characteristics of each stage in the time series as the product full life cycle data principal component time series propagation aggregation feature.

[0035] Figure 3 A flowchart of the method for collecting and analyzing carbon footprint data for the entire life cycle of a product according to an embodiment of the present application is provided for performing temporal information propagation processing based on the attenuation of principal components of node features on the semantic coding features of data in the raw material acquisition stage, the manufacturing stage, the transportation stage, the usage stage, and the recycling and settlement stage to obtain the principal component temporal propagation aggregation features of data for the entire life cycle of the product. Figure 4 In the product life cycle carbon footprint data collection and analysis method according to the embodiment of the present application, the semantic coding features of the raw material acquisition stage data, the semantic coding features of the manufacturing stage data, the semantic coding features of the transportation stage data, the semantic coding features of the use stage data, and the semantic coding features of the recycling and settlement stage data are processed based on the time series information propagation of the node feature principal component attenuation to obtain the data flow diagram of the time series propagation aggregation features of the product life cycle data principal component. Figure 3 and Figure 4 As shown, the step S132 includes: S1321, performing principal component extraction on the semantic coding features of the data in the raw material acquisition stage, the semantic coding features of the data in the manufacturing stage, the semantic coding features of the data in the transportation stage, the semantic coding features of the data in the use stage, and the semantic coding features of the data in the recycling and settlement stage to obtain a sequence of principal component representations of the data semantic coding features in the current stage and principal component representations of the data semantic coding features in the historical stage; S1322, performing time-series attenuation dynamic aggregation on the sequence of principal component representations of the data semantic coding features in the current stage and principal component representations of the data semantic coding features in the historical stage to obtain the principal component time-series propagation aggregation features of the data throughout the product life cycle.

[0036] Specifically, in step S1321, principal component extraction is performed on the semantic coding features of the data in the raw material acquisition stage, the semantic coding features of the data in the manufacturing stage, the semantic coding features of the data in the transportation stage, the semantic coding features of the data in the use stage, and the semantic coding features of the data in the recycling and settlement stage to obtain a sequence of principal component representations of the semantic coding features of the current stage data and principal component representations of the semantic coding features of the historical stage data. It should be understood that by using principal component analysis (PCA), the most representative features can be extracted from high-dimensional data and the impact of interference factors can be reduced. This helps to focus on the semantic key factors of each stage that truly affect carbon emissions, while reducing the complexity of the data processing process.

[0037] Specifically, in the step S1321, it includes: performing eigenvalue-based principal component extraction on the raw material acquisition stage data semantic coding vector, the manufacturing stage data semantic coding vector, the transportation stage data semantic coding vector and the use stage data semantic coding vector to obtain the raw material acquisition stage data semantic coding principal component representation vector, the manufacturing stage data semantic coding principal component representation vector, the transportation stage data semantic coding principal component representation vector and the use stage data semantic coding principal component representation vector as the sequence of the historical stage data semantic coding principal component representation, and performing eigenvalue-based principal component extraction on the recycling and settlement stage data semantic coding vector to obtain the recycling and settlement stage data semantic coding principal component representation vector as the current stage data semantic coding feature principal component representation.

[0038] Specifically, in the step S1322, the sequence of principal component representations of the semantic coding features of the current stage data and the principal component representations of the semantic coding features of the historical stage data are dynamically aggregated with time series attenuation to obtain the principal component time series propagation aggregation features of the product life cycle data, including: S1322-1, respectively calculating the absolute factors of the time series propagation attenuation entropy of the semantic coding principal component representation vector of the raw material acquisition stage data, the semantic coding principal component representation vector of the manufacturing stage data, the semantic coding principal component representation vector of the transportation stage data and the semantic coding principal component representation vector of the use stage data relative to the semantic coding principal component representation vector of the recycling and settlement stage data to obtain a sequence of absolute factors of the time series propagation attenuation entropy of the stage data; S1322-2, calculating the absolute factors of the time series propagation attenuation entropy of the stage data based on the time span between the semantic coding principal component representation vector of the raw material acquisition stage data, the semantic coding principal component representation vector of the manufacturing stage data, the semantic coding principal component representation vector of the transportation stage data and the semantic coding principal component representation vector of the use stage data and the semantic coding principal component representation vector of the recycling and settlement stage data. The absolute factors of the propagation attenuation entropy of the time series of each stage data in the sequence of factors are modulated in the time dimension to obtain a sequence of the propagation attenuation entropy factors of the time series span modulation of the stage data; S1322-3, the sequence of the propagation attenuation entropy factors of the time series span modulation of the stage data is input into the information transmission screening module based on the gating function to obtain a sequence of the propagation attenuation weights of the time series span modulation of the stage data; S1322-4, based on the sequence of the propagation attenuation weights of the time series span modulation of the stage data, the weighted sum of the semantic encoding principal component representation vector of the raw material acquisition stage data, the semantic encoding principal component representation vector of the manufacturing stage data, the semantic encoding principal component representation vector of the transportation stage data and the semantic encoding principal component representation vector of the use stage data is calculated to obtain the significant transmission aggregation representation vector of the principal component of the historical stage data semantic encoding; S1322-5, the positional sum of the significant transmission aggregation representation vector of the principal component of the historical stage data semantic encoding and the principal component representation vector of the semantic encoding of the recycling, processing and settlement stage data is calculated to obtain the principal component time series propagation aggregation vector of the product life cycle data as the principal component time series propagation aggregation feature of the product life cycle data.

[0039] Specifically, in step S1322-1, the absolute factors of the time series propagation decay entropy of the semantic encoding principal component representation vector of the data in the raw material acquisition stage, the semantic encoding principal component representation vector of the data in the manufacturing stage, the semantic encoding principal component representation vector of the data in the transportation stage, and the semantic encoding principal component representation vector of the data in the use stage relative to the semantic encoding principal component representation vector of the data in the recycling processing and settlement stage are calculated to obtain a sequence of absolute factors of the time series propagation decay entropy of the stage data. It should be understood that in order to further evaluate the degree of change in the time node feature distribution between the semantics of different stages, that is, the semantic association and interaction of each stage in the whole life cycle of the product. To this end, the absolute factors of the time series propagation decay entropy between the semantic encoding principal component representation vector of each stage data relative to the semantic encoding principal component representation vector of the last stage are calculated. This is a measurement method for quantifying the uncertainty of random variables, which is used to capture the trend of the evolution of the semantic features of the product in each stage over time and the information loss.

[0040] Specifically, the step S1322-1 includes: calculating the positional division of the corresponding positions of the semantic coding principal component representation vector of the raw material acquisition stage data and the semantic coding principal component representation vector of the recycling and settlement stage data to obtain the stage data semantic coding feature principal component attenuation vector; calculating the logarithmic function value with base two of the absolute value of each eigenvalue of the stage data semantic coding feature principal component attenuation vector to obtain the stage data semantic coding feature principal component attenuation logarithmic vector; calculating the positional dot product of the semantic coding principal component representation vector of the raw material acquisition stage data and the stage data semantic coding feature principal component attenuation logarithmic vector, and then performing positional multiplication on each positional eigenvalue in the obtained dot product vector to obtain the stage data time series propagation attenuation value; calculating the exponential function of the stage data time series propagation attenuation value with the natural constant e as the base to obtain the stage data time series propagation attenuation entropy absolute factor.

[0041] Specifically, in step S1322-2, the absolute factors of the time series propagation attenuation entropy of each stage data in the sequence of absolute factors of the time series propagation attenuation entropy of the stage data are modulated in the time dimension based on the time spans between the principal component representation vector of the semantic encoding of the raw material acquisition stage data, the principal component representation vector of the semantic encoding of the manufacturing stage data, the principal component representation vector of the semantic encoding of the transportation stage data, and the principal component representation vector of the semantic encoding of the use stage data and the principal component representation vector of the semantic encoding of the recycling processing settlement stage data, so as to obtain a sequence of the propagation attenuation entropy factors modulated by the time series span of the stage data. It should be understood that in order to further enhance the model's ability to understand information transmission under long time intervals, the time span factor is introduced to adjust the above-mentioned attenuation entropy factors to form a sequence of the propagation attenuation entropy factors modulated by the time series span of the stage data. This adjustment enables a more comprehensive and reasonable processing of the long-distance time series dependency relationship between the various stages in the entire life cycle of the product, ensuring that even large differences between the semantic time series nodes of the stage data that are far apart are considered to be changes within the normal range.

[0042] Specifically, the step S1322-2 includes: subtracting the timestamp of the principal component representation vector of the semantic encoding of the recycling processing settlement stage data from the timestamp of the principal component representation vector of the semantic encoding of the raw material acquisition stage data, and rounding down to obtain the stage data time span value; calculating an exponential function with the natural constant e as the base and the stage data time span value as the exponent to obtain the stage data time span modulation value; dividing the stage data timing propagation attenuation entropy absolute factor corresponding to the principal component representation vector of the semantic encoding of the raw material acquisition stage data by the stage data time span modulation value to obtain the stage data timing span modulation propagation attenuation entropy factor.

[0043] Specifically, in step S1322-3, the sequence of the phase data timing span modulated propagation attenuation entropy factors is input into the information transmission screening module based on the gating function to obtain the sequence of the phase data timing span modulated propagation attenuation weights. It should be understood that the information transmission screening module of the gating mechanism is similar to the design in the RNN architecture such as LSTM, which adaptively determines which timing nodes of the phase data semantics should be focused on and expressed according to the context, so that important information can be screened out and given an adaptive weight, thereby ensuring that the factors that have the greatest impact on carbon emissions receive due attention.

[0044] Specifically, the step S1322-3 includes: comparing each stage data timing span modulation propagation attenuation entropy factor in the sequence of the stage data timing span modulation propagation attenuation entropy factors with a predetermined threshold to obtain a sequence of the stage data timing span modulation propagation attenuation weights; wherein, in response to the stage data timing span modulation propagation attenuation entropy factor being greater than the predetermined threshold, the stage data timing span modulation propagation attenuation entropy factor greater than the predetermined threshold is input into a sigmoid function; in response to the stage data timing span modulation propagation attenuation entropy factor being less than or equal to the predetermined threshold, the stage data timing span modulation propagation attenuation entropy factor less than or equal to the predetermined threshold is set to zero.

[0045] Specifically, in step S1322-4, based on the sequence of propagation attenuation weights modulated by the time span of the stage data, the weighted sum of the semantically encoded principal component representation vector of the raw material acquisition stage data, the semantically encoded principal component representation vector of the manufacturing stage data, the semantically encoded principal component representation vector of the transportation stage data and the semantically encoded principal component representation vector of the use stage data is further calculated to obtain the aggregation characteristics of the significant information transmission of the semantic principal components of the data of other stages except the recycling, processing and settlement stage during the entire life cycle of the product.

[0046] Specifically, in step S1322-5, the positional sum of the main component significant transfer aggregation representation vector of the historical stage data semantic encoding and the main component representation vector of the recycling and settlement stage data semantic encoding is calculated to obtain the main component time series propagation aggregation vector of the product life cycle data as the main component time series propagation aggregation feature of the product life cycle data. It should be understood that the main component significant transfer aggregation representation vector of the historical stage data semantic encoding and the main component representation vector of the recycling and settlement stage data semantic encoding are added by position to obtain the final output, that is, the main component time series propagation aggregation vector of the product life cycle data as the main component time series propagation aggregation feature of the product life cycle data. This feature vector integrates the key semantic information of the main component data of each stage in the product life cycle and the time series interaction relationship, which reflects the state transition of the product in each stage, helps to deeply understand the time series semantic features of each stage of the product, so as to adapt to the dynamic changes of different product processes and external conditions, provide a basis for comprehensive evaluation of the environmental impact of the product, and provide a solid foundation for comprehensive understanding of the carbon emissions in the product life cycle.

[0047] Specifically, the semantic coding features of the raw material acquisition stage data, the semantic coding features of the manufacturing stage data, the semantic coding features of the transportation stage data, the semantic coding features of the use stage data, and the semantic coding features of the recycling and settlement stage data are processed by time series information propagation based on the decay of the principal component of node features to obtain the principal component time series propagation aggregation features of the product life cycle data;

[0048] Among them, the stage semantic temporal information propagation formula is:

[0049] X={x 1 ,x 2 ,x 3 ,x 4 ,x 5}

[0050] v i = PCA(x i )

[0051]

[0052] V={v 1 ,v 2 ,v 3 ,v 4 ,v 5}

[0053]

[0054]

[0055] Among them, x 1 ,x 2 ,x 3 ,x 4 ,x 5 are the semantic coding vectors of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data, and the recycling processing settlement phase data, respectively. X is the sequence of the phase data semantic coding vectors, x i is the i-th stage data semantic encoding vector in the sequence of the stage data semantic encoding vector, PCA(x i ) is x i Perform principal component extraction based on eigenvalues, v i For x i The corresponding stage data semantic encoding principal component representation vector, For x i The transposed vector of n is the number of eigenvalues ​​in the semantic encoding vector of the data in the stage, C i is x iThe corresponding stage data covariance matrix, U i is x i The corresponding stage data principal component orthogonal matrix, For U i The transposed matrix, v i1 ,v i2 ,…,v im is the principal component vector of each stage data in the sequence of the principal component vector of the stage data, Λ i For x i The corresponding stage data diagonal matrix, diag(λ i1 ,λ i2 ,…,λ im ) is the matrix whose diagonal elements are λ i1 ,λ i2 ,…,λ im The diagonal matrix of the stage data, λ i1 ,λ i2 ,…,λ im are the weight values ​​of the principal component vectors of the data at each stage, is the k value corresponding to the maximum value returned, j is the maximum approximate matching value, V is the sequence of the principal component representation vector of the semantic encoding of the stage data, and v 1 ,v 2 ,v 3 ,v 4 ,v 5 are respectively the first, second, third, fourth and fifth stage data semantic encoding principal component representation vectors in the sequence of the stage data semantic encoding principal component representation vectors, v i The eigenvalues ​​at each position in , v 5 The eigenvalues ​​of each position in, L is v i The number of eigenvalues ​​in the equation, log represents the logarithmic function value with base 2, exp(·) represents the exponential function with the natural constant e as the base, d i v i and v 5 The absolute factor of the decay entropy of the phase data time series propagation between the two, t 5 and t i Respectively represent the timestamps of the principal component representation vectors of the 5th and ith phase data semantic encodings, For floor rounding operation, e (i→5) Yes i and v 5 The stage data timing span modulation propagation attenuation entropy factor between them, mask(·) is the masking process, sigmoid(·) is the sigmoid function, θ is the predetermined threshold, w iis the timing span modulation propagation attenuation weight of each stage data in the sequence of the stage data timing span modulation propagation attenuation weight, and s is the principal component propagation aggregation representation vector of the semantic timing of the device stage data.

[0056] In the above-mentioned product life cycle carbon footprint data collection and analysis method, the step S140 generates a carbon emission correction factor based on the main component time series propagation aggregation characteristics of the product life cycle data, and calculates the total carbon emissions of the product life cycle. It should be understood that the main component time series propagation aggregation characteristics of the product life cycle data provide key indicators and trends of carbon emissions at each stage. These characteristics integrate the data of each stage of raw material acquisition, manufacturing, transportation, use and recycling, forming a comprehensive view to help enterprises identify the emission characteristics and relative importance of each stage. The establishment of this aggregation feature makes the assessment of carbon emissions not limited to a single time point, but takes into account the changes in the time series and the interactive relationship between different stages. Next, by analyzing these aggregation features, a carbon emission correction factor can be generated. This factor is an adjustment coefficient used to correct the emission data deviation caused by external factors (such as climate change, etc.). Finally, by applying the calculated correction factor to the carbon emission data of each stage, the enterprise can obtain a more accurate total carbon emission value of the whole life cycle.

[0057] Specifically, in the step S140, a carbon emission correction factor is generated based on the time series propagation aggregation characteristics of the main components of the product's life cycle data, and the total value of the product's carbon emissions over its life cycle is calculated, including: inputting the time series propagation aggregation vector of the main components of the product's life cycle data into a decoder-based carbon emission correction factor generator to obtain the carbon emission correction factor; multiplying the carbon emission correction factor by the initial total value of carbon emissions to obtain the total value of the product's carbon emissions over its life cycle.

[0058] It should be understood that the principal component time series propagation aggregation vector of the product life cycle data is input into the decoder-based carbon emission correction factor generator to obtain the carbon emission correction factor. In other words, the semantic principal component time series propagation aggregation representation information of each stage of the product life cycle data is used to perform decoding regression to obtain the carbon emission correction factor. Then, the carbon emission correction factor is multiplied by the initial total value of carbon emissions to obtain the total carbon emissions of the product throughout its life cycle. In this way, the differences in each processing process can be more accurately reflected, so as to better adapt to the dynamic changes of different product processes and external conditions, and take into account the mutual influence between each stage and the importance of timing to obtain a total carbon emission value of the product throughout its life cycle that is closer to the actual value. This method can not only more accurately reflect the real environmental impact of the product, but also provide enterprises with more reliable product life cycle carbon footprint data support, helping them to formulate effective emission reduction measures.

[0059] Preferably, when the semantic coding vector of the raw material acquisition stage data, the semantic coding vector of the manufacturing stage data, the semantic coding vector of the transportation stage data, the semantic coding vector of the use stage data and the semantic coding vector of the recycling and settlement stage data respectively represent the encoded semantic features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data, when performing feature time series dynamic propagation aggregation based on the attenuation of the node feature principal component gradient, the feature principal component gradient attenuation under the differences in the encoding semantic patterns of each data will have different propagation directionality, causing the product life cycle data principal component time series propagation aggregation vector obtained by dynamic propagation aggregation to have a complex interactive aggregation spatial structure. Therefore, when the product life cycle data principal component time series propagation aggregation vector is input into the decoder-based carbon emission correction factor generator, it is expected to improve the decoding regression convergence and generalization effect of the product life cycle data principal component time series propagation aggregation vector under the complex aggregation spatial structure.

[0060] Therefore, the applicant of the present application considers optimizing the principal component time series propagation aggregation vector of the product life cycle data when inputting the principal component time series propagation aggregation vector of the product life cycle data into the decoder-based carbon emission correction factor generator. The optimization process includes: calculating the absolute value sum of each eigenvalue of the principal component time series propagation aggregation vector of the product life cycle data to obtain the first principal component time series propagation aggregation spatial structure value of the product life cycle data, and calculating the square root of the sum of the squares of each eigenvalue of the principal component time series propagation aggregation vector of the product life cycle data to obtain the second principal component time series propagation aggregation spatial structure value of the product life cycle data:

[0061]

[0062] Among them, v i Represents the eigenvalues ​​of the principal component time series propagation aggregation vector of the product life cycle data, ∑ represents the sum function, |·| represents the absolute value function, represents a real number set, α represents the spatial structure value of the principal component of the first product's full life cycle data in time series, β represents the spatial structure value of the principal component of the second product's full life cycle data in time series, represents the set of real numbers, L represents the length of the vector;

[0063] Multiply each eigenvalue of the principal component time series propagation aggregation vector of the product life cycle data by the first principal component time series propagation aggregation spatial structure value of the product life cycle data and the second principal component time series propagation aggregation spatial structure value of the product life cycle data to obtain the first principal component time series propagation aggregation structure reference value of the product life cycle data and the second principal component time series propagation aggregation structure reference value of the product life cycle data corresponding to each eigenvalue; multiply each eigenvalue of the principal component time series propagation aggregation vector of the product life cycle data by the length of the principal component time series propagation aggregation vector of the product life cycle data and the square root of the length to obtain the first principal component time series propagation aggregation scale transformation value of the product life cycle data and the second principal component time series propagation aggregation scale transformation value of the product life cycle data corresponding to each eigenvalue; multiply the first principal component time series propagation aggregation vector of the product life cycle data by the length of the principal component time series propagation aggregation vector of the product life cycle data and the square root of the length The first product life cycle data principal component time series propagation aggregation transformation adjustment value is obtained by dividing the structure reference value by the difference between the first product life cycle data principal component time series propagation aggregation spatial structure value and the first product life cycle data principal component time series propagation aggregation scale transformation value to obtain the first product life cycle data principal component time series propagation aggregation transformation adjustment value; the second product life cycle data principal component time series propagation aggregation structure reference value is divided by the difference between the second product life cycle data principal component time series propagation aggregation spatial structure value and the second product life cycle data principal component time series propagation aggregation scale transformation value to obtain the second product life cycle data principal component time series propagation aggregation transformation adjustment value; and, the weighted sum of the first product life cycle data principal component time series propagation aggregation transformation adjustment value and the second product life cycle data principal component time series propagation aggregation transformation adjustment value is calculated to obtain each eigenvalue of the optimized product life cycle data principal component time series propagation aggregation vector.

[0064] Here, the principal component time series propagation aggregation vector of the product life cycle data, denoted as V, is optimized as follows:

[0065] V'=V 1 ⊕ ( ω⊙V 2 )

[0066] exist and In the case of:

[0067]

[0068] v 1i =(α×v i ) / (α-L×v i )

[0069] Among them, v irepresents the eigenvalues ​​of the principal component time series propagation aggregation vector of the product life cycle data, α represents the spatial structure value of the principal component time series propagation aggregation of the first product life cycle data, β represents the spatial structure value of the principal component time series propagation aggregation of the second product life cycle data, v 1i represents the principal component time series propagation aggregation transformation adjustment value of the first product life cycle data, v 2i V represents the principal component time series propagation aggregation transformation adjustment value of the second product life cycle data, 1 represents the first product life cycle data principal component time series propagation aggregation transformation adjustment feature vector obtained by quantizing the plurality of first product life cycle data principal component time series propagation aggregation transformation adjustment values, V 2 represents the principal component time series propagation aggregation transformation adjustment feature vector of the second product life cycle data obtained by quantizing the principal component time series propagation aggregation transformation adjustment value vectors of the second product life cycle data, V' represents the optimized principal component time series propagation aggregation vector of the product life cycle data, ⊕ represents addition by position, ω represents the principal component time series propagation aggregation adjustment weighted value, ⊙ represents multiplication by position, represents the set of real numbers, and L represents the length of the vector.

[0070] That is, for the spatial structure information of the feature set of the principal component time series propagation aggregation vector of the product life cycle data in the high-dimensional space, the scale-based frame transformation of each eigenvalue of the principal component time series propagation aggregation vector of the product life cycle data is performed by taking the class norm space structured representation of the principal component time series propagation aggregation vector of the product life cycle data as the reference window, and realizing the spatial structure-based frame attention weight adjustment of each eigenvalue of the principal component time series propagation aggregation vector of the product life cycle data, so as to ensure the spatial transformation invariance of the principal component time series propagation aggregation vector of the product life cycle data under the interaction of feature space, thereby improving the convergence and generalization effect of the decoding regression of the feature set of the principal component time series propagation aggregation vector of the product life cycle data under the complex spatial structure representation, and improving the accuracy of the carbon emission correction factor obtained by inputting it into the decoder-based carbon emission correction factor generator. In this way, the initial total value of carbon emissions can be corrected more accurately, and the dynamic changes of different product processes and external conditions can be better adapted to obtain a total value of carbon emissions of the product life cycle that is closer to the actual total value, providing enterprises with more reliable product life cycle carbon footprint data support and helping them to formulate effective emission reduction measures.

[0071] In summary, a method for collecting and analyzing product life cycle carbon footprint data based on an embodiment of the present application is explained, which introduces more sophisticated artificial intelligence data processing technology to perform data analysis and semantic understanding of product life cycle data, and calculates the initial total carbon emissions of each stage of the product. Then, the carbon emission correction factor is obtained based on the semantic information of each stage and the connection between each other, so as to make corresponding corrections to the initial total carbon emissions, so as to more accurately reflect the differences in each processing process, better adapt to the dynamic changes of different product processes and external conditions, and obtain a total carbon emissions value of the product life cycle that is closer to the actual value. In this way, the real environmental impact of the product can be reflected more accurately, providing enterprises with more reliable product life cycle carbon footprint data support, and formulating effective emission reduction measures.

[0072] Figure 5 : is a system block diagram of a product life cycle carbon footprint data collection and analysis system according to an embodiment of the present application. Figure 5 As shown, according to the product life cycle carbon footprint data collection and analysis system 100 of the embodiment of the present application, it includes: a product life cycle data acquisition module 110, which is used to acquire product life cycle data, and the product life cycle data includes raw material acquisition stage data, manufacturing stage data, transportation stage data, use stage data and recycling and settlement stage data; a carbon emission initial total value calculation module 120, which is used to calculate the initial total value of carbon emissions based on the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data; a node feature principal component attenuation time series information propagation processing module 130, which is used to process the raw material acquisition stage data After semantic encoding of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data, the semantic encoding features of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data are processed with time series information propagation based on the attenuation of the principal component of the node characteristics to obtain the principal component time series propagation aggregation characteristics of the product's entire life cycle data; a module 140 for calculating the total value of carbon emissions over the entire life cycle of the product is used to generate a carbon emission correction factor based on the principal component time series propagation aggregation characteristics of the product's entire life cycle data, and calculate the total value of carbon emissions over the product's entire life cycle.

[0073] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned product life cycle carbon footprint data collection and analysis system have been referenced above. Figures 1 to 4 The product life cycle carbon footprint data collection and analysis method has been introduced in detail in the description of the product life cycle carbon footprint data collection and analysis method, and therefore, its repeated description will be omitted.

[0074] As described above, the product life cycle carbon footprint data collection and analysis system 100 according to the embodiment of the present application can be implemented in various terminal devices. In one example, the product life cycle carbon footprint data collection and analysis system 100 can be integrated into the terminal device as a software module and / or hardware module. For example, the product life cycle carbon footprint data collection and analysis system 100 can be a software module in the operating system of the terminal device, or it can be an application developed for the terminal device; of course, the product life cycle carbon footprint data collection and analysis system 100 can also be one of the many hardware modules of the terminal device.

[0075] Alternatively, in another example, the product life cycle carbon footprint data collection and analysis system 100 and the terminal device may also be separate devices, and the product life cycle carbon footprint data collection and analysis system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0076] In summary, a product life cycle carbon footprint data collection and analysis system based on the embodiment of the present application is explained, which introduces more sophisticated artificial intelligence data processing technology to perform data analysis and semantic understanding of product life cycle data, and calculates the initial total carbon emissions of each stage of the product. Then, the carbon emission correction factor is obtained based on the semantic information of each stage and the connection between each other, so as to make corresponding corrections to the initial total carbon emissions, so as to more accurately reflect the differences in each processing process, better adapt to the dynamic changes of different product processes and external conditions, and obtain a total carbon emissions value of the product life cycle that is closer to the actual value. In this way, the real environmental impact of the product can be reflected more accurately, providing enterprises with more reliable product life cycle carbon footprint data support, and formulating effective emission reduction measures.

Claims

1. A method for collecting and analyzing carbon footprint data of a product throughout its life cycle, characterized in that: include: Obtain product life cycle data, including raw material acquisition data, manufacturing data, transportation data, usage data, and recycling and settlement data; Calculate the initial total value of carbon emissions based on the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data; After semantic coding of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data, the semantic coding features of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data are subjected to time series information propagation processing based on the decay of the principal component of the node feature to obtain the principal component time series propagation aggregation feature of the product life cycle data, including: principal component extraction of the semantic coding features of the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data to obtain a sequence of principal component representation of the semantic coding features of the current phase data and a sequence of principal component representation of the semantic coding features of the historical phase data; time series decay dynamic aggregation of the sequence of principal component representation of the semantic coding features of the current phase data and a sequence of principal component representation of the semantic coding features of the historical phase data to obtain the principal component time series propagation aggregation feature of the product life cycle data; Based on the principal component time series propagation aggregation characteristics of the product's full life cycle data, a carbon emission correction factor is generated, and the total value of the product's full life cycle carbon emissions is calculated.

2. The method for collecting and analyzing product life cycle carbon footprint data according to claim 1, characterized in that: Based on the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the use phase data and the recycling and settlement phase data, the initial total value of carbon emissions is calculated, including: Based on the raw material acquisition stage data, determine the carbon emission value of the raw material acquisition stage; Based on the manufacturing stage data, determining the carbon emission value of the manufacturing stage; Based on the transportation stage data, determining the carbon emission value of the transportation stage; Based on the usage phase data, determining a carbon emission value of the usage phase; Based on the data of the recycling and processing settlement stage, determine the carbon emission value of the recycling and processing settlement stage; The sum of the carbon emission values ​​of the raw material acquisition stage, the manufacturing stage, the transportation stage, the use stage and the recycling and settlement stage is calculated as the initial total carbon emission value.

3. The method for collecting and analyzing product life cycle carbon footprint data according to claim 2, characterized in that: The raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the usage phase data and the recycling and settlement phase data are semantically encoded, including: semantically encoding the raw material acquisition phase data, the manufacturing phase data, the transportation phase data, the usage phase data and the recycling and settlement phase data to obtain a semantic encoding vector of the raw material acquisition phase data as the semantic encoding feature of the raw material acquisition phase data, a semantic encoding vector of the manufacturing phase data as the semantic encoding feature of the manufacturing phase data, a semantic encoding vector of the transportation phase data as the semantic encoding feature of the transportation phase data, a semantic encoding vector of the usage phase data as the semantic encoding feature of the usage phase data and a semantic encoding vector of the recycling and settlement phase data as the semantic encoding feature of the recycling and settlement phase data.

4. The method for collecting and analyzing product life cycle carbon footprint data according to claim 3 is characterized in that: The principal component extraction is performed on the semantic coding features of the raw material acquisition stage data, the semantic coding features of the manufacturing stage data, the semantic coding features of the transportation stage data, the semantic coding features of the use stage data, and the semantic coding features of the recycling and settlement stage data to obtain a sequence of principal component representations of the current stage data semantic coding features and principal component representations of the historical stage data semantic coding features, including: Performing principal component extraction based on eigenvalues ​​on the semantic coding vector of raw material acquisition phase data, the semantic coding vector of manufacturing phase data, the semantic coding vector of transportation phase data and the semantic coding vector of use phase data respectively to obtain the principal component representation vector of raw material acquisition phase data semantic coding, the principal component representation vector of manufacturing phase data semantic coding, the principal component representation vector of transportation phase data semantic coding and the principal component representation vector of use phase data semantic coding as the sequence of principal component representation of historical phase data semantic coding; The principal component extraction based on the eigenvalue is performed on the semantic coding vector of the recycling settlement stage data to obtain the principal component representation vector of the semantic coding of the recycling settlement stage data as the principal component representation of the semantic coding features of the current stage data.

5. The method for collecting and analyzing product life cycle carbon footprint data according to claim 4 is characterized in that: The sequence of the principal component representation of the semantic coding feature of the current stage data and the principal component representation of the semantic coding feature of the historical stage data is dynamically aggregated to obtain the principal component time series propagation aggregation feature of the product life cycle data, including: Respectively calculate the time series propagation attenuation entropy absolute factors of the semantic encoding principal component representation vector of the raw material acquisition stage data, the semantic encoding principal component representation vector of the manufacturing stage data, the semantic encoding principal component representation vector of the transportation stage data, and the semantic encoding principal component representation vector of the use stage data relative to the semantic encoding principal component representation vector of the recycling processing settlement stage data to obtain a sequence of the time series propagation attenuation entropy absolute factors of the stage data; Based on the time spans between the semantic encoding principal component representation vector of the raw material acquisition stage data, the semantic encoding principal component representation vector of the manufacturing stage data, the semantic encoding principal component representation vector of the transportation stage data and the semantic encoding principal component representation vector of the use stage data and the semantic encoding principal component representation vector of the recycling processing settlement stage data, respectively, each stage data time series propagation attenuation entropy absolute factor in the sequence of the stage data time series propagation attenuation entropy absolute factors is modulated in the time dimension to obtain a sequence of stage data time series span modulated propagation attenuation entropy factors; Inputting the sequence of the phase data timing span modulation propagation attenuation entropy factors into the information transfer screening module based on the gating function to obtain the sequence of the phase data timing span modulation propagation attenuation weights; Based on the sequence of the phase data time series span modulation propagation attenuation weights, the weighted sum of the semantic encoding principal component representation vector of the raw material acquisition phase data, the semantic encoding principal component representation vector of the manufacturing phase data, the semantic encoding principal component representation vector of the transportation phase data and the semantic encoding principal component representation vector of the use phase data is calculated to obtain the significant transfer aggregation representation vector of the historical phase data semantic encoding principal component; Calculate the positional sum of the semantically encoded principal component significant transfer aggregation representation vector of the historical stage data and the semantically encoded principal component representation vector of the recycling processing and settlement stage data to obtain the principal component time series propagation aggregation vector of the product's full life cycle data as the principal component time series propagation aggregation feature of the product's full life cycle data.

6. The method for collecting and analyzing product life cycle carbon footprint data according to claim 5, characterized in that: Respectively calculating the time series propagation attenuation entropy absolute factors of the semantic encoding principal component representation vector of the raw material acquisition stage data, the semantic encoding principal component representation vector of the manufacturing stage data, the semantic encoding principal component representation vector of the transportation stage data, and the semantic encoding principal component representation vector of the use stage data relative to the semantic encoding principal component representation vector of the recycling processing settlement stage data to obtain a sequence of the time series propagation attenuation entropy absolute factors of the stage data, including: Calculate the positional division of the principal component representation vector of the semantic encoding of the raw material acquisition stage data and the principal component representation vector of the semantic encoding of the recycling processing settlement stage data to obtain the principal component attenuation vector of the semantic encoding feature of the stage data; Calculate the base-two logarithmic function value of the absolute value of each eigenvalue of the principal component attenuation vector of the semantic encoding feature of the stage data to obtain the principal component attenuation logarithmic vector of the semantic encoding feature of the stage data; After calculating the positional dot product of the semantic coding principal component representation vector of the raw material acquisition stage data and the attenuation logarithm vector of the semantic coding feature principal component of the stage data, each positional feature value in the obtained dot product vector is multiplied by position point to obtain the stage data time series propagation attenuation value; An exponential function of the phase data timing propagation attenuation value with the natural constant e as the base is calculated to obtain the phase data timing propagation attenuation entropy absolute factor.

7. The method for collecting and analyzing product life cycle carbon footprint data according to claim 6, characterized in that: Based on the time spans between the semantic encoding principal component representation vector of the raw material acquisition stage data, the semantic encoding principal component representation vector of the manufacturing stage data, the semantic encoding principal component representation vector of the transportation stage data, and the semantic encoding principal component representation vector of the use stage data and the semantic encoding principal component representation vector of the recycling processing settlement stage data, the absolute factors of the time series propagation attenuation entropy of each stage data in the sequence of the absolute factors of the time series propagation attenuation entropy of the stage data are modulated in the time dimension to obtain a sequence of the propagation attenuation entropy factors modulated by the time series span of the stage data, including: Subtract the timestamp of the semantic encoding principal component representation vector of the recycling settlement stage data from the timestamp of the semantic encoding principal component representation vector of the raw material acquisition stage data, and round them down to obtain a stage data time span value; Calculate an exponential function with the natural constant e as the base and the time span value of the stage data as the exponent to obtain a modulation value of the time span of the stage data; The stage data timing propagation attenuation entropy absolute factor corresponding to the principal component representation vector of the raw material acquisition stage data semantic encoding is divided by the stage data time span modulation value to obtain the stage data timing span modulation propagation attenuation entropy factor.

8. The method for collecting and analyzing product life cycle carbon footprint data according to claim 7, characterized in that: Inputting the sequence of the phase data timing span modulation propagation attenuation entropy factors into the information transfer screening module based on the gating function to obtain the sequence of the phase data timing span modulation propagation attenuation weights, including: Comparing each stage data timing span modulation propagation attenuation entropy factor in the sequence of the stage data timing span modulation propagation attenuation entropy factors with a predetermined threshold value to obtain a sequence of the stage data timing span modulation propagation attenuation weights; Among them, in response to the stage data timing span modulation propagation attenuation entropy factor being greater than the predetermined threshold, the stage data timing span modulation propagation attenuation entropy factor greater than the predetermined threshold is input into the sigmoid function; in response to the stage data timing span modulation propagation attenuation entropy factor being less than or equal to the predetermined threshold, the stage data timing span modulation propagation attenuation entropy factor less than or equal to the predetermined threshold is set to zero.

9. The method for collecting and analyzing product life cycle carbon footprint data according to claim 8, characterized in that: Based on the principal component time series propagation aggregation characteristics of the product life cycle data, a carbon emission correction factor is generated, and the total carbon emission value of the product life cycle is calculated, including: Inputting the principal component time series propagation aggregation vector of the product life cycle data into a decoder-based carbon emission correction factor generator to obtain the carbon emission correction factor; The carbon emission correction factor is multiplied by the initial total carbon emission value to obtain the total carbon emission value of the product over its entire life cycle.

10. A product life cycle carbon footprint data collection and analysis system, characterized in that: include: A product life cycle data acquisition module, which is used to acquire product life cycle data, including raw material acquisition stage data, manufacturing stage data, transportation stage data, use stage data and recycling and settlement stage data; A carbon emission initial total value calculation module, used to calculate the carbon emission initial total value based on the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data; The node feature principal component decay time series information propagation processing module is used for semantically encoding the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data, and then performing time series information propagation processing based on the node feature principal component decay on the semantic coding features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data to obtain the principal component time series propagation aggregation features of the product life cycle data, including: extracting the principal components of the semantic coding features of the raw material acquisition stage data, the manufacturing stage data, the transportation stage data, the use stage data and the recycling and settlement stage data to obtain the sequence of the principal component representation of the semantic coding features of the current stage data and the principal component representation of the semantic coding features of the historical stage data; performing time series decay dynamic aggregation on the sequence of the principal component representation of the semantic coding features of the current stage data and the principal component representation of the semantic coding features of the historical stage data to obtain the principal component time series propagation aggregation features of the product life cycle data; The module for calculating the total carbon emissions of a product over its entire life cycle is used to generate a carbon emission correction factor based on the time series propagation aggregation characteristics of the principal components of the product's entire life cycle data, and calculate the total carbon emissions of the product over its entire life cycle.

Images