A Monitoring Method and System for Carbon Emissions in the Physicalization Stage of Railway Infrastructure

Abstract

This application relates to the field of carbon emission monitoring technology. Specifically, it discloses a carbon emission monitoring method and system for the materialization stage of railway infrastructure. After receiving the material list for the materialization stage uploaded by the user, it uses data processing technology based on deep learning to analyze the material data in the material list for the materialization stage. By perceiving the context semantic association of each item of material data, it realizes the abnormal diagnosis of the material list. Then, after confirming that the material data is correct, it calculates the total carbon emission value for the materialization stage based on the carbon emission per unit mass corresponding to each material, and gives corresponding carbon emission early warning prompts according to the offset of the total carbon emission value for the materialization stage relative to the expected value. This application can achieve accurate accounting of carbon emissions in the materialization stage of railway infrastructure, effectively avoid carbon emission calculation deviations caused by inaccurate data, and provide strong support for the low-carbon and sustainable development of the railway industry.

Description

Technical Field

[0001] This application relates to the technical field of carbon emission monitoring, and more specifically, to a method and system for monitoring carbon emissions during the materialization stage of railway infrastructure. Background Art

[0002] During the construction of railway infrastructure, the materialization stage is the period when resources are most concentratedly invested, involving the use of a large number of different types of materials. Accurately monitoring carbon emissions during this stage is crucial for achieving low-carbon and sustainable development in the railway industry. As a key document recording the material usage during the materialization stage, the accuracy of the data in the material list directly affects the reliability of carbon emission calculations.

[0003] Traditionally, the monitoring of carbon emissions during the materialization stage of railway infrastructure relied on manual statistics and calculations. This method is not only time-consuming and laborious but also prone to large deviations in assessment results due to inaccurate or missing data. In addition, with the continuous expansion of the scale of railway construction projects and the increasing richness of material types, the amount of data in the material list has increased exponentially, which further exacerbates the difficulty of verifying the data in the material list. If incorrect material data is not discovered and corrected, it will lead to deviations in the total carbon emission value calculated based on it, which not only cannot provide an accurate basis for project decision-making but may also mislead the formulation of subsequent carbon emission management strategies.

[0004] Therefore, an optimized method and system for monitoring carbon emissions during the materialization stage of railway infrastructure are needed to solve the above technical problems. Summary of the Invention

[0005] To solve the above technical problems, this application is proposed.

[0006] According to one aspect of this application, a system for monitoring carbon emissions during the materialization stage of railway infrastructure is provided, which includes:

[0007] A materialization stage list upload module for receiving the material list during the materialization stage uploaded by a user. The material list during the materialization stage includes the material name and the usage amount of each material. Among them, the materialization stage list upload module is also used to perform material analysis on the material list during the materialization stage to achieve list anomaly diagnosis;

[0008] A regional emission factor collection module for obtaining the regional emission factors corresponding to each material from a public database. The regional emission factor is the carbon emission per unit mass;

[0009] A total carbon emission value calculation module for calculating the total carbon emission value during the materialization stage based on the regional emission factors corresponding to each material and the material list during the materialization stage;

[0010] The total carbon emission target setting module is used to extract the expected value of the total carbon emission in the materialization stage from the background database;

[0011] The carbon emission monitoring module is used to calculate the difference between the total carbon emission value in the materialization stage and the expected value of the total carbon emission in the materialization stage to obtain the total carbon emission target offset value, and then determine whether to generate a carbon emission warning prompt based on the comparison between the total carbon emission target offset value and a preset threshold;

[0012] Specifically, the materialization stage list uploading module includes:

[0013] The material list structured coding unit is used to perform structured coding on the material list in the materialization stage to obtain a set of material property-usage semantic description coding vectors;

[0014] The inter-material context dynamic walking unit is used to perform causal-triggered inter-material dynamic walking semantic association coding on the set of material property-usage semantic description coding vectors to obtain inter-material context dynamic walking semantic coding vectors;

[0015] The anomaly diagnosis unit is used to determine a diagnosis result based on the inter-material context dynamic walking semantic coding vectors, and the diagnosis result is used to indicate whether there is an anomaly in the material list in the materialization stage.

[0016] Specifically, the material list structured coding unit is used for:

[0017] Performing semantic embedding coding on each material in the material list in the materialization stage to obtain a set of material semantic description coding vectors;

[0018] Adding the usage amount of the corresponding material to the tail of each material semantic description coding vector in the set of material semantic description coding vectors to obtain the set of material property-usage semantic description coding vectors.

[0019] Specifically, the inter-material context dynamic walking unit includes:

[0020] The implicit feature mining subunit is used to perform implicit feature mining on each material property-usage semantic description coding vector in the set of material property-usage semantic description coding vectors to obtain a set of material property-usage deep implicit semantic feature coding vectors;

[0021] The causal association topological feature extraction subunit is used to extract the semantic causal association topological features between the set of material property-usage deep implicit semantic feature coding vectors to obtain a semantic causal association topological feature matrix between material data;

[0022] A context semantic association encoding subunit, configured to perform multi-level dynamic walk semantic association encoding on a set of material property-usage semantic description encoding vectors based on the semantic causal association topological feature matrix between the material data, so as to obtain a context dynamic walk semantic encoding vector between the materials.

[0023] Specifically, the causal association topological feature extraction subunit includes:

[0024] A causal association factor calculation secondary subunit, configured to calculate a semantic causal association factor between any two material property-usage depth implicit semantic feature encoding vectors in the set of material property-usage depth implicit semantic feature encoding vectors, so as to obtain a semantic causal association topological matrix between material data composed of multiple semantic causal association factors between material data;

[0025] A causal trigger secondary subunit, configured to input the semantic causal association topological matrix between the material data into a causal trigger network based on a gated activation function to obtain the semantic causal association topological feature matrix between the material data.

[0026] Specifically, the causal association factor calculation secondary subunit is configured to:

[0027] Perform semantic association encoding on any two material property-usage depth implicit semantic feature encoding vectors in the set of material property-usage depth implicit semantic feature encoding vectors to obtain a set of semantic association encoding matrices between material data;

[0028] Based on the statistical features of each semantic association encoding matrix between the material data in the set of semantic association encoding matrices between the material data, calculate the corresponding semantic causal association factor between the material data to obtain the multiple semantic causal association factors between the material data, where the statistical features include the feature variance, maximum eigenvalue, feature mean, and causal association energy bias term of the semantic association encoding matrix between the material data.

[0029] Specifically, if the feature variance of the semantic association encoding matrix between the material data is greater than or equal to a preset threshold, calculate the average weighted value of the Euclidean distances between every two material property-usage depth implicit semantic feature encoding vectors in the set of material property-usage depth implicit semantic feature encoding vectors as the causal association energy bias term;

[0030] If the feature variance of the semantic association encoding matrix between the material data is less than the preset threshold, calculate the weighted value of the feature mean of the semantic association encoding matrix between the material data as the causal association energy bias term.

[0031] Specifically, the context semantic association encoding subunit is configured to:

[0032] Input the semantic causal association topological feature matrix between the material data and the set of material property-usage semantic description encoding vectors into a dynamic random walk encoder based on a graph convolutional neural network model to obtain a surface context dynamic random walk semantic encoding vector between the material data;

[0033] Input the semantic causal association topological feature matrix between the material data and the set of material property-usage deep implicit semantic feature encoding vectors into the dynamic random walk encoder based on the graph convolutional neural network model to obtain a hidden layer context dynamic random walk semantic encoding vector between the material data;

[0034] Fuse the hidden layer context dynamic random walk semantic encoding vector between the material data and the surface context dynamic random walk semantic encoding vector between the material data to obtain a context dynamic random walk semantic encoding vector between the materials.

[0035] Specifically, the anomaly diagnosis unit is used for:

[0036] Input the context dynamic random walk semantic encoding vector between the materials into an inventory diagnoser based on a classifier to obtain the diagnosis result.

[0037] According to another aspect of the present application, a method for monitoring carbon emissions in the physicalization stage of railway infrastructure is provided, which includes:

[0038] Receive a physicalization stage material list uploaded by a user, where the physicalization stage material list includes material names and the usage amounts of each material. Among them, the physicalization stage list uploading module is also used to perform material analysis on the physicalization stage material list to achieve list anomaly diagnosis;

[0039] Obtain the regional emission factors corresponding to each material from a public database, where the regional emission factor is the carbon emission per unit mass;

[0040] Based on the regional emission factors corresponding to each material and the physicalization stage material list, calculate the total carbon emission value in the physicalization stage;

[0041] Extract the expected total carbon emission value in the physicalization stage from the background database;

[0042] Calculate the difference between the total carbon emission value in the physicalization stage and the expected total carbon emission value in the physicalization stage to obtain a total carbon emission target offset value, and then based on the comparison between the total carbon emission target offset value and a preset threshold, determine whether to generate a carbon emission warning prompt.

[0043] Specifically, performing material analysis on the physicalization stage material list to achieve list anomaly diagnosis includes:

[0044] Structurally encode the material list in the materialization stage to obtain a set of material property - dosage semantic description encoding vectors;

[0045] Perform causal - trigger - based dynamic random - walk semantic association encoding on the set of material property - dosage semantic description encoding vectors to obtain inter - material context - aware dynamic random - walk semantic encoding vectors;

[0046] Based on the inter - material context - aware dynamic random - walk semantic encoding vectors, determine a diagnosis result, where the diagnosis result is used to indicate whether there is an abnormality in the material list in the materialization stage.

[0047] The present application has at least the following technical effects:

[0048] Compared with the prior art, for the railway infrastructure materialization - stage carbon - emission monitoring method and system provided by the present application, after receiving the material list in the materialization stage uploaded by the user, it uses deep - learning - based data - processing technology to analyze the material data in the material list in the materialization stage. By performing context - semantic association perception on each item of material data, it realizes the abnormal diagnosis of the material list. Then, after confirming that the material data is correct, it calculates the total carbon - emission value in the materialization stage based on the carbon emission per unit mass corresponding to each material, and gives corresponding carbon - emission early - warning prompts according to the offset of the total carbon - emission value in the materialization stage relative to the expected value. It can achieve accurate accounting of the carbon emissions in the railway infrastructure materialization stage, effectively avoid carbon - emission calculation deviations caused by inaccurate data, and provide strong support for the low - carbon and sustainable development of the railway industry. Description of the Drawings

[0049] By describing the embodiments of the present application in more detail in conjunction with the drawings, the above - mentioned and other objects, features, and advantages of the present application will become more obvious. The drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification. They are used together with the embodiments of the present application to explain the present application, and do not constitute a limitation to the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0050] Figure 1 It is a block diagram of a railway infrastructure materialization - stage carbon - emission monitoring system according to an embodiment of the present application.

[0051] Figure 2 It is a block diagram of a materialization - stage list uploading module in a railway infrastructure materialization - stage carbon - emission monitoring system according to an embodiment of the present application.

[0052] Figure 3 It is a schematic diagram of data flow of the materialization - stage list in a railway infrastructure materialization - stage carbon - emission monitoring system according to an embodiment of the present application.

[0053] Figure 4Block diagram of the material context dynamic walking unit in the carbon emission monitoring system for the railway infrastructure in the materialization stage according to an embodiment of the present application.

[0054] Figure 5 Block diagram of the causal association topological feature extraction subunit in the carbon emission monitoring system for the railway infrastructure in the materialization stage according to an embodiment of the present application.

[0055] Figure 6 Flowchart of the carbon emission monitoring method for the railway infrastructure in the materialization stage according to an embodiment of the present application. Detailed implementation manners

[0056] As shown in the present application and the claims, unless the context clearly indicates otherwise, words such as "a", "an", "one", and / or "the" are not specifically singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of the clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0057] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server. The modules are merely illustrative, and different aspects of the system and method can use different modules.

[0058] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations before or below are not necessarily executed precisely in order. On the contrary, as needed, various steps can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0059] Next, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0060] It should be noted that all actions of obtaining data in the present application are carried out on the premise of complying with the corresponding data protection regulations and policies of the country where the data is located and obtaining the authorization given by the owner of the corresponding device.

[0061] Figure 1 Block diagram of the carbon emission monitoring system for the railway infrastructure in the materialization stage according to an embodiment of the present application. As Figure 1As shown, the carbon emission monitoring system 100 for the materialization stage of railway infrastructure includes: a materialization stage list uploading module 110, configured to receive a materialization stage material list uploaded by a user, where the materialization stage material list includes material names and the usage amounts of each material. Among them, the materialization stage list uploading module is further configured to perform material analysis on the materialization stage material list to achieve list anomaly diagnosis; a regional emission factor collection module 120, configured to obtain the regional emission factors corresponding to each material from a public database, where the regional emission factor is the carbon emission amount per unit mass; a total carbon emission value calculation module 130, configured to calculate the total carbon emission value for the materialization stage based on the regional emission factors corresponding to each material and the materialization stage material list; a total carbon emission target setting module 140, configured to extract the expected total carbon emission value for the materialization stage from a background database; and a carbon emission monitoring module 150, configured to calculate the difference between the total carbon emission value for the materialization stage and the expected total carbon emission value for the materialization stage to obtain a total carbon emission target deviation value, and then determine whether to generate a carbon emission warning prompt based on the comparison between the total carbon emission target deviation value and a preset threshold.

[0062] In the above carbon emission monitoring system for the materialization stage of railway infrastructure, the materialization stage list uploading module 110 is configured to receive a materialization stage material list uploaded by a user, where the materialization stage material list includes material names and the usage amounts of each material. Among them, the materialization stage list uploading module is further configured to perform material analysis on the materialization stage material list to achieve list anomaly diagnosis. It should be understood that the material list contains detailed information about various materials used in the materialization stage of railway infrastructure, such as material types, specifications, quantities, etc. As the basic data for calculating carbon emissions in the materialization stage, the accuracy of its data has a decisive effect on the accuracy of subsequent total carbon emission calculations. Any incorrect or unreasonable data entry, such as miswriting of material usage amounts or confusion of material types, may lead to serious deviations in the final carbon emission calculation results. Therefore, after receiving the materialization stage material list uploaded by the user, the materialization stage list uploading module further uses deep learning technology to perform data analysis on the material data in the materialization stage material list to identify whether there are anomalies in the material list, thereby avoiding waste of subsequent computing resources, reducing the repeated verification and correction work caused by data errors, and improving the efficiency and credibility of the entire carbon emission monitoring process. Among them, Figure 2 is a block diagram of the materialization stage list uploading module in the carbon emission monitoring system for the materialization stage of railway infrastructure according to an embodiment of the present application. Figure 3 is a schematic diagram of the data flow of the materialization stage list in the carbon emission monitoring system for the materialization stage of railway infrastructure according to an embodiment of the present application. As Figure 2 and Figure 3As shown, the materialization stage list upload module 110 includes: a material list structured encoding unit 111 for performing structured encoding on the materialization stage material list to obtain a set of material property-usage semantic description encoding vectors; an inter-material context dynamic random walk unit 112 for performing causal-triggered inter-material dynamic random walk semantic association encoding on the set of material property-usage semantic description encoding vectors to obtain inter-material context dynamic random walk semantic encoding vectors; and an anomaly diagnosis unit 113 for determining a diagnosis result based on the inter-material context dynamic random walk semantic encoding vectors, where the diagnosis result is used to indicate whether there is an anomaly in the materialization stage material list.

[0063] Specifically, in a specific example of the present application, the material list structured encoding unit 111 is configured to: perform semantic embedding encoding on each material in the materialization stage material list to obtain a set of material semantic description encoding vectors. Specifically, in order to perform material data analysis and anomaly recognition on the materialization stage material list using deep learning algorithms, it is necessary to convert each material data into a numerical vector form that can be understood and processed by a computer. Based on this, in the present application, for the text information such as the name and specification of each material in the materialization stage material list, a pre-trained semantic embedding model is used to perform semantic embedding encoding on it to convert the text information into a numerical representation in a high-dimensional vector space, obtaining a set of material semantic description encoding vectors. In an embodiment of the present application, the Word2Vec model is used to perform semantic embedding encoding on each material information in the materialization stage material list. The Word2Vec model has been trained on a large-scale corpus and has strong semantic representation capabilities. It can accurately convert material information into vector form and map similar words to close positions in the vector space, thereby retaining the semantic association between materials. In this way, even in the face of material information with different expressions but the same essence, it can be accurately identified and characterized accordingly, thereby improving the analysis accuracy of material data in the material list and the accuracy of anomaly recognition.

[0064] Specifically, in a specific example of the present application, the material list structured coding unit 111 is further configured to: add the dosage of the corresponding material to the tail of each material semantic description coding vector in the set of material semantic description coding vectors to obtain the set of material property-dosage semantic description coding vectors. It should be understood that since the material dosage is an important dimension for measuring the contribution of materials to carbon emissions, the accuracy of its data is the key to ensuring the reliability of the carbon emission calculation results. Therefore, in the present application, the dosage information of each material in the materialized stage material list is further appended to the tail of the corresponding material semantic description coding vector in numerical form to form a composite vector representation that combines material properties and dosages, and obtain the set of material property-dosage semantic description coding vectors, so as to ensure that subsequent data analysis can consider both the type and dosage of materials at the same time, and more accurately judge the data rationality of the materialized stage material list.

[0065] Specifically, the inter-material context dynamic walking unit 112 is configured to perform causal-triggered inter-material dynamic walking semantic association coding on the set of material property-dosage semantic description coding vectors to obtain inter-material context dynamic walking semantic coding vectors. Specifically, in the actual application scenario, there is a certain causal association relationship between the material data items in the materialized stage material list. For example, in the construction of railway bridges, the dosages of materials such as steel, concrete, and admixtures will have a certain range and association relationship due to factors such as design standards and construction techniques; in the process of laying railway tracks, the dosage of steel rails will have a causal association with the dosages of materials such as fasteners and sleepers. Based on this, the present application proposes a causal-triggered inter-material dynamic walking semantic association coding method. By analyzing the causal association strength between the material data items, it simulates the dynamic walking process of the material data on the causal chain, and performs context dynamic walking coding on the set of material property-dosage semantic description coding vectors, so as to more deeply understand the internal logic between the data items in the material list, such as the reasonable proportion range of the dosages of steel and concrete under a certain construction technique, and the corresponding relationship between the dosage of admixtures and the concrete strength grade, etc., thereby obtaining inter-material context dynamic walking semantic coding vectors that integrate inter-material dynamic association information. In this way, the potential relationships and abnormal patterns between the data items in the material list can be captured more accurately, and the unreasonable points in the data can be identified, such as materials with abnormal dosages, materials with inconsistent specifications, etc., thereby providing a more reliable data basis for the subsequent abnormal diagnosis of the material list.

[0066] Figure 4 It is a block diagram of the inter-material context dynamic walking unit in the carbon emission monitoring system for the materialized stage of railway infrastructure according to an embodiment of the present application. As Figure 4As shown, the material - context dynamic walking unit 112 includes: an implicit feature mining subunit 1121, configured to perform implicit feature mining on each material property - dosage semantic description encoding vector in the set of material property - dosage semantic description encoding vectors to obtain a set of material property - dosage deep implicit semantic feature encoding vectors; a causal association topological feature extraction subunit 1122, configured to extract semantic causal association topological features between the set of material property - dosage deep implicit semantic feature encoding vectors to obtain a semantic causal association topological feature matrix between material data; and a context semantic association encoding subunit 1123, configured to perform multi - level dynamic walking semantic association encoding on the set of material property - dosage semantic description encoding vectors based on the semantic causal association topological feature matrix between material data to obtain the material - context dynamic walking semantic encoding vectors.

[0067] More specifically, the implicit feature mining subunit 1121 is represented by the formula:

[0068] ;

[0069] ;

[0070] ;

[0071] Wherein, represents the set of material property - dosage semantic description encoding vectors, , , and respectively represent the 1st, 2nd, th, and th material property - dosage semantic description encoding vectors in the set of material property - dosage semantic description encoding vectors, represents the number of vectors in the set of material property - dosage semantic description encoding vectors, represents a 1×1 convolution operation, represents an activation function, represents the set of material property - dosage deep implicit semantic feature encoding vectors, , , and respectively represent the material property - dosage deep implicit semantic feature encoding vectors corresponding to , , and .

[0072] Specifically, due to the complex relationships among railway construction materials, it is difficult to discover potential patterns from surface data alone. Therefore, this application extracts deep features from the material property - usage semantic description encoding vectors to generate a more extensive semantic representation, namely the material property - usage deep implicit semantic feature encoding vector, thereby providing richer data information for accurately analyzing the causal relationships among materials in subsequent steps and making the subsequent analysis more accurate and comprehensive.

[0073] Figure 5 The block diagram of the causal association topological feature extraction subunit in the carbon emission monitoring system for the physicalization stage of railway infrastructure according to an embodiment of this application. As Figure 5 shown, the causal association topological feature extraction subunit 1122 includes: a causal association factor calculation secondary subunit 11221, configured to calculate the semantic causal association factors between any two material property - usage deep implicit semantic feature encoding vectors in the set of material property - usage deep implicit semantic feature encoding vectors to obtain a material data semantic causal association topological matrix composed of multiple material data semantic causal association factors; a causal trigger secondary subunit 11222, configured to input the material data semantic causal association topological matrix into a causal trigger network based on a gated activation function to obtain the material data semantic causal association topological feature matrix.

[0074] In a specific example of this application, the causal association factor calculation secondary subunit 11221 is configured to: perform semantic association encoding on any two material property - usage deep implicit semantic feature encoding vectors in the set of material property - usage deep implicit semantic feature encoding vectors to obtain a set of material data semantic association encoding matrices; calculate the corresponding material data semantic causal association factors based on the statistical features of each material data semantic association encoding matrix in the set of material data semantic association encoding matrices to obtain the multiple material data semantic causal association factors, where the statistical features include the feature variance, maximum eigenvalue, feature mean, and causal association energy bias term of the material data semantic association encoding matrix. The above - mentioned causal association factor calculation secondary subunit 11221 is represented by the formula:

[0075] ;

[0076] ;

[0077] where represents the th material property - usage deep implicit semantic feature encoding vector in the set of material property - usage deep implicit semantic feature encoding vectors, represents the transpose of the vector, represents and The semantic association coding matrix between corresponding material data Indicates the characteristic variance of the semantic association coding matrix between material data Indicates the characteristic mean of the matrix Indicates the causal association energy bias term Indicates taking the maximum value Indicates and The semantic causal association factor between corresponding material data Indicates the semantic causal association topology matrix between material data 、 、 and respectively represent the semantic causal association factors between the corresponding two material property - dosage depth implicit semantic feature coding vectors in the set of the material property - dosage depth implicit semantic feature coding vectors

[0078] Specifically, for any two material property - dosage depth implicit semantic feature coding vectors, perform semantic association coding to construct the semantic association coding matrix between material data, and use the causal association energy metric function to analyze the feature distribution of the semantic association coding matrix between material data, so as to realize the explicit quantitative coding expression of the causal association between any two material property - dosage depth implicit semantic feature coding vectors in the set of the material property - dosage depth implicit semantic feature coding vectors, and obtain the corresponding semantic causal association factors between material data

[0079] Particularly, in a preferred example of the present application, if the characteristic variance of the semantic association coding matrix between material data is greater than or equal to a preset threshold, calculate the average weighted value of the Euclidean distances between each two material property - dosage depth implicit semantic feature coding vectors in the set of the material property - dosage depth implicit semantic feature coding vectors as the causal association energy bias term; if the characteristic variance of the semantic association coding matrix between material data is less than the preset threshold, calculate the weighted value of the characteristic mean of the semantic association coding matrix between material data as the causal association energy bias term, which is expressed by the formula as

[0080] ;

[0081] Wherein Indicates the predetermined threshold Indicates the Euclidean distance metric function and respectively represent different weight parameters is The length of, that is, the number of vectors in the set of the material property - dosage depth implicit semantic feature coding vectors

[0082] Specifically, by regarding the low-level causal associations in the semantic association coding matrix of the material data as molecular-level relationships inferred based on statistical correlations, it is possible to further perform intervention prediction of causal association energy on the basis of the global fine-grained statistical association representation, so as to study the causal association fine-grained structure and its dynamic regulation of the semantic association coding matrix of the material data in terms of the high-dimensional and heterogeneous representation of causal genomics. Among them, when the aggregative distribution representation of the causal graph (i.e., the semantic association coding matrix of the material data) is greater than a predetermined threshold, a bias occurs during the integration of source data based on the matrix graph node effect representation of the semantic causal association factor between the material data, while when the aggregative distribution representation of the causal graph is less than the predetermined threshold, the condensed structure modeling can be directly performed through feature pattern integration compression. In this way, not only can the causal association energy in the semantic association coding matrix of the material data be encoded and described, but also the implicit causal intervention prediction results can be condensed, thereby obtaining a more efficient revelation of key causal associations.

[0083] In a specific example of the present application, the causal trigger secondary subunit 11222 is expressed by the formula:

[0084] ;

[0085] wherein, represents the normalization exponential function, represents the causal trigger network, represents the gating threshold, represents the topological feature matrix of the semantic causal association between the material data.

[0086] Specifically, the topological matrix of the semantic causal association between the material data is input into the causal trigger network based on the gating activation function, so as to utilize the gating mechanism to highlight important causal association information, weaken or remove irrelevant or interfering information, and generate the topological feature matrix of the semantic causal association between the material data, making the causal association features more prominent, more accurately reflecting the causal relationship between the materials, and providing more effective data for subsequent analysis.

[0087] More specifically, the context semantic association coding subunit 1123 is used to: input the topological feature matrix of the semantic causal association between the material data and the set of semantic description coding vectors of the material property-usage into the dynamic random walk encoder based on the graph convolutional neural network model to obtain the surface context dynamic random walk semantic coding vector between the material data, which is expressed by the formula:

[0088] ;

[0089] wherein, represents the graph convolutional neural network, Represents the surface context dynamic walking semantic encoding vector between material data.

[0090] Specifically, after deeply optimizing the association topology structure between material data using the causal trigger network, the context semantic association between the material property-usage semantic description encoding vectors is further extracted through a graph convolutional neural network. Through the propagation of each material data semantic description feature in the causal association topology structure, from local to global, the explicit semantics between nodes are gradually recursively aggregated to generate the surface context dynamic walking semantic encoding vector between material data.

[0091] More specifically, the context semantic association encoding subunit 1123 is further used to: input the semantic causal association topology feature matrix between the material data and the set of material property-usage deep implicit semantic feature encoding vectors into the dynamic walking encoder based on the graph convolutional neural network model to obtain the hidden layer context dynamic walking semantic encoding vector between material data, which is expressed by the formula:

[0092] ;

[0093] Where, Represents the hidden layer context dynamic walking semantic encoding vector between material data.

[0094] Among them, the material property-usage deep implicit semantic feature encoding vector contains deeper implicit semantic information of material data compared to the original material property-usage semantic description encoding vector. By inputting the semantic causal association topology feature matrix between the material data and the set of material property-usage deep implicit semantic feature encoding vectors into the graph convolutional neural network model, the implicit context semantic association between material data can be further mined, revealing more subtle and deep semantic connections between material data. Similarly, the graph convolutional neural network gradually fuses local features into global features by layer-by-layer passing and aggregating node information, thereby generating the hidden layer context dynamic walking semantic encoding vector between material data.

[0095] More specifically, the context semantic association encoding subunit 1123 is also used to: fuse the hidden layer context dynamic walking semantic encoding vector between the material data and the surface context dynamic walking semantic encoding vector between the material data to obtain the context dynamic walking semantic encoding vector between materials, which is expressed by the formula:

[0096] ;

[0097] Where, Represents the fusion weight parameter, Represents the context dynamic walking semantic encoding vector between materials.

[0098] Specifically, considering that the feature dimensions of the surface context dynamic random walk semantic encoding vector between material data and the hidden context dynamic random walk semantic encoding vector between material data are complementary, therefore, in this application, through a weighted summation fusion strategy, the importance distribution of the surface context dynamic random walk semantic encoding vector between material data and the hidden context dynamic random walk semantic encoding vector between material data during the fusion process is adjusted to ensure that the generated context dynamic random walk semantic encoding vector between materials has stronger discrimination ability and representation integrity.

[0099] Specifically, in a specific example of this application, the anomaly diagnosis unit 113 is configured to: input the context dynamic random walk semantic encoding vector between materials into a manifest diagnosis device based on a classifier to obtain the diagnosis result, and the diagnosis result is used to indicate whether there is an anomaly in the material manifest during the materialization phase. Specifically, the classifier is trained on a large number of known normal and abnormal material manifest data to learn the classification boundary between the distribution patterns of normal material manifest data and abnormal material manifest distribution data. Thus, after receiving the context dynamic random walk semantic encoding vector between materials, it can learn and identify its feature patterns to accurately determine whether it deviates from the normal data distribution range and output the corresponding diagnosis result. If the diagnosis result shows that there is an anomaly in the material manifest during the materialization phase, the system will immediately issue a corresponding anomaly prompt or suggestion to the user, reminding the user to check and correct the incorrect data in the material manifest during the materialization phase.

[0100] In the above carbon emission monitoring system for railway infrastructure during the materialization phase, the regional emission factor acquisition module 120 is configured to obtain the regional emission factors corresponding to each material from a public database, and the regional emission factor is the carbon emission per unit mass. Specifically, after confirming that the material manifest data during the materialization phase is correct, the carbon emissions can be calculated using the material manifest data during the materialization phase. Here, considering that there are significant differences in energy structure, production process, environmental protection standards, etc. in different regions, the carbon emissions of the same material during the production process in different regions will be different. For example, in regions where coal is the main energy source, the carbon emissions during cement production are relatively high; while in regions rich in hydropower resources, the carbon emissions during cement production are relatively low due to the use of clean electricity. Therefore, this application further obtains the regional emission factors corresponding to each material from an authoritative public database, that is, the carbon emission per unit mass of the material during the production process in the corresponding region, to ensure that the carbon emission calculation result can truly reflect the carbon emissions during the material production process.

[0101] Specifically, the bill of materials needs to record in detail all the materials used in the project and their corresponding quantities. At the same time, for each material, obtain its unit mass carbon emissions in a specific region from a reliable source. These emission factors reflect the greenhouse gas emissions generated during the entire physical and chemical processes such as production and transportation of different materials. Since the production processes and energy consumption patterns of the same material may vary in different regions, it is particularly crucial to select the emission factors that best match the location of the project. This step involves analyzing the differences between different regions. For example, the carbon emissions of cement production may vary significantly between the eastern coastal regions and the western inland regions of China because of the differences in fuel use efficiency, power supply structure, etc. Therefore, when collecting emission factor data, regional factors must be considered and the data source closest to the project location should be selected as much as possible.

[0102] After determining the specific emission factors of the required materials, the next task is to integrate this data into the system. In this process, attention needs to be paid to the consistency and compatibility of data formats. Since the data formats from different sources may vary, it is necessary to preprocess the original data to ensure that all data can be successfully imported into the system's database. This includes, but is not limited to, unifying measurement units (such as converting all carbon emissions into the form of tons of CO2 / ton of material), standardizing data field names, eliminating duplicate records, etc. Only in this way can it be ensured that the subsequent calculation module can correctly read and use this data.

[0103] Given the progress of technology, the production technology and energy consumption patterns of materials may change over time, and the corresponding emission factors need to be updated regularly. To this end, automated retrieval tools can be set up to regularly scan selected public databases to find the latest emission factor data and compare it with the data in the existing database. If new and more accurate data is found, the old data should be replaced in a timely manner to ensure that the system always operates based on the most accurate information.

[0104] In addition, to improve the quality and usability of the data, an expert review mechanism can also be introduced. Invite experts in fields such as materials science and environmental engineering to review the collected emission factor data, especially those data points that may have greater uncertainty or controversy. Experts can verify the rationality and accuracy of the data and give improvement suggestions by referring to the latest research results, experimental data, etc. This approach can not only enhance the credibility of the data but also promote interdisciplinary cooperation and communication, and drive the development of related fields.

[0105] Since the emission factor is directly related to the environmental performance evaluation results of enterprises, strict security measures must be taken to prevent data leakage. Using encryption technology to store and transmit data is an effective method; at the same time, formulating detailed data access permission rules to only allow authorized personnel to access corresponding levels of data according to their work requirements. In addition, regular data backup is also essential to prevent the accidental loss of important information.

[0106] In the above carbon emission monitoring system for the physicalization stage of railway infrastructure, the total carbon emission value calculation module 130 is used to calculate the total carbon emission value for the physicalization stage based on the regional emission factors corresponding to each material and the material list for the physicalization stage. Specifically, according to the law of conservation of matter, the carbon emissions of each material during physicalization processes such as production and transportation are proportional to the amount of the material used. Multiplying the amount of each material by its corresponding unit mass carbon emission (i.e., the regional emission factor) can obtain the carbon emissions of each material. Then, summing up the carbon emissions of all materials can obtain the total carbon emission value for the physicalization stage.

[0107] To ensure the efficiency and accuracy of the calculation process, automated tools can be used to assist. By writing specialized algorithm scripts or using existing software platforms, a large amount of data can be processed quickly. This approach is particularly applicable to large projects, where the material list may contain thousands of different materials and their usage amounts. Automated tools can automatically extract data from the material list according to preset rules and perform paired calculations with the corresponding emission factors. At the same time, these tools can also detect and mark outliers or inconsistencies in real time for manual review and correction.

[0108] After calculating the carbon emissions of each material, the next step is to aggregate all the separately calculated results, that is, to sum up the carbon emission values of all materials to obtain the total carbon emission value for the entire physicalization stage. During this process, special attention must be paid to the consistency of data formats. For example, the carbon emissions of all materials should be uniformly converted to the same measurement unit for accurate summation. In addition, consideration should also be given to how to handle the values after the decimal point to ensure that the final result is neither distorted nor can reflect the actual situation. Usually, retaining an appropriate number of decimal places helps to improve the accuracy of the result.

[0109] In the above carbon emission monitoring system for the materialization stage of railway infrastructure, the total carbon emission target setting module 140 is used to extract the expected value of the total carbon emission in the materialization stage from the background database. Here, the expected value of the total carbon emission in the materialization stage is the carbon emission target set by the project in the planning stage based on various factors such as its own sustainable development goals, industry standards, and empirical data of similar projects. It provides an accurate and reliable benchmark for the comparative analysis of the total carbon emission, helps the project team better understand the carbon emission situation of the project, timely discover the potential risks of carbon emission exceeding the standard, and take corresponding measures for adjustment and optimization to ensure that the project can successfully achieve the established carbon emission target.

[0110] Specifically, in the initial stage of project planning, the project team sets the carbon emission target based on various factors such as its own sustainable development goals, industry standards, and empirical data of similar projects. For example, when formulating the carbon emission target for a railway infrastructure construction project, factors such as the environmental policy requirements of the project location, construction technical conditions, material selection, and their corresponding carbon emission situations need to be comprehensively considered. In addition, the actual carbon emission data of domestic and foreign similar projects also need to be referred to. By analyzing these data, a challenging yet feasible expected value of the total carbon emission is determined.

[0111] Next, the expected values of the total carbon emission determined after careful consideration are entered into the background database. To ensure the accuracy and integrity of the data, the entry work must be completed by personnel with a professional knowledge background and requires a strict review process. During this process, not only the specific values need to be recorded, but also information such as the source, calculation method, and relevant assumptions of each data point needs to be detailed. This not only improves the transparency of the data but also facilitates the traceability and verification of the data in the future.

[0112] In the background database, a data structure is constructed specifically for storing and managing these expected values of the total carbon emission. This structure should have good scalability and flexibility to adapt to the changing needs of different types of projects. Considering the differences in material types, construction processes, etc. that may be involved in different projects, a data model that can be dynamically adjusted needs to be designed. For example, for some projects using new environmentally friendly materials or innovative construction technologies, corresponding fields can be reserved in the data model to facilitate the flexible addition of new types of carbon emission data. At the same time, to facilitate management and query, a classification system also needs to be established to classify and organize the data according to dimensions such as project type, geographical location, and implementation time.

[0113] When it is necessary to extract the expected value of the total carbon emissions during the materialization phase from the back-end database, the system will automatically locate the correct data record based on the relevant parameters input by the user. The parameters mentioned here include, but are not limited to, basic information such as the unique identifier of the project, the region it belongs to, and the expected start time. By precisely matching these parameters, the expected value of the total carbon emissions of the corresponding project can be found quickly and accurately. It is worth noting that during this process, the system also needs to have a certain degree of fault tolerance, that is, when encountering the situation of partial parameter missing or incomplete matching, it can provide reasonable alternative solutions or prompt the user to supplement the necessary information.

[0114] To ensure that the extracted expected value of the total carbon emissions always remains up-to-date, it is essential to update the data in the back-end database regularly. With the changes in the external environment, such as the upgrade of industry standards, the originally set carbon emission targets may no longer be applicable. At this time, it is necessary to adjust the relevant records in the database in a timely manner. For this purpose, a special working group can be established to track the latest policy trends and technological progress, and revise the data in the database accordingly. In addition, automated tools can be introduced to regularly scan relevant policy documents and industry reports, and once new information that affects the carbon emission target is found, immediately trigger the data update process.

[0115] On the premise of ensuring data security, authorized users are allowed to access the expected value of the total carbon emissions in the back-end database. Given that these data are crucial for the successful implementation of the project, strict security measures must be taken to prevent unauthorized access or data leakage. Specific practices include, but are not limited to: using encryption technology to protect the security of data transmission; setting detailed permission control rules to ensure that only authenticated users can view or modify specific data records; regularly backing up the database to prevent the accidental loss of important information. At the same time, a complete audit mechanism should be established to record every data access and modification behavior for future tracking and review.

[0116] In the above carbon emission monitoring system for the materialization stage of railway infrastructure, the carbon emission monitoring module 150 is used to calculate the difference between the total carbon emission value in the materialization stage and the expected total carbon emission value in the materialization stage to obtain the total carbon emission target deviation value, and then based on the comparison between the total carbon emission target deviation value and a preset threshold, determine whether to generate a carbon emission warning prompt. Specifically, by calculating the difference between the actual total carbon emission value in the materialization stage and the expected value, the deviation degree between the actual carbon emission and the expected target is quantified to obtain the total carbon emission target deviation value. At the same time, the deviation value is further compared with the preset threshold to determine whether the total carbon emission exceeds a reasonable range. The preset threshold is comprehensively determined according to factors such as the risk tolerance of the project, the strictness of carbon emission control, and industry standards. If the deviation value is greater than the preset threshold, it indicates that the actual carbon emission deviates too much from the expected target, and a carbon emission warning prompt will be automatically generated to attract the attention of the project team, timely analyze the reasons and take corresponding emission reduction measures, including but not limited to optimizing material selection, reducing the usage of high-carbon emission materials, improving the production process, enhancing energy utilization efficiency, and exploring and using low-carbon technologies and clean energy. Conversely, if the deviation value is within the preset threshold range, it indicates that the actual carbon emission is basically consistent with the expected target, and the project team can continue to carry out carbon emission management and control according to the established plan.

[0117] In summary, the carbon emission monitoring system for the materialization stage of railway infrastructure based on the embodiments of the present application is clarified. After receiving the material list in the materialization stage uploaded by the user, it uses data processing technology based on deep learning to analyze the material data in the material list in the materialization stage. By perceiving the context semantic association of each material data, it realizes the abnormal diagnosis of the material list. Then, after confirming that the material data is correct, it calculates the total carbon emission value in the materialization stage based on the unit mass carbon emission of each material, and makes corresponding carbon emission warning prompts according to the deviation of the total carbon emission value in the materialization stage relative to the expected value. In this way, accurate accounting of carbon emissions in the materialization stage of railway infrastructure can be achieved, effectively avoiding carbon emission calculation deviations caused by inaccurate data, and providing strong support for the low-carbon sustainable development of the railway industry.

[0118] Furthermore, a carbon emission monitoring method for the materialization stage of railway infrastructure is also provided.

[0119] Figure 6 is a flowchart of the carbon emission monitoring method for the materialization stage of railway infrastructure according to the embodiments of the present application. As Figure 6As shown, the carbon emission monitoring method for the materialization stage of the railway infrastructure includes the following steps: S1, receiving the material list for the materialization stage uploaded by the user. The material list for the materialization stage includes the material names and the usage amounts of each material. Among them, the material list upload module for the materialization stage is also used to perform material analysis on the material list for the materialization stage to achieve list anomaly diagnosis; S2, obtaining the regional emission factors corresponding to each material from the public database. The regional emission factor is the carbon emission per unit mass; S3, calculating the total carbon emission value for the materialization stage based on the regional emission factors corresponding to each material and the material list for the materialization stage; S4, extracting the expected total carbon emission value for the materialization stage from the background database; S5, calculating the difference between the total carbon emission value for the materialization stage and the expected total carbon emission value for the materialization stage to obtain the target deviation value of the total carbon emission, and then based on the comparison between the target deviation value of the total carbon emission and the preset threshold, determining whether to generate a carbon emission warning prompt.

[0120] The basic principles of the present invention have been described above in combination with specific embodiments. However, it should be noted that the advantages, advantages, effects, etc. mentioned in the present invention are only examples and not limitations. It cannot be considered that these advantages, advantages, effects, etc. are essential for each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purposes of illustration and easy understanding, rather than limitations. The above details do not limit the present invention to necessarily adopt the above specific details to be implemented.

[0121] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative. For example, the unit division is only a logical function division, and there can be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

[0123] In addition, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. The multiple elements stated in the system claims can also be implemented by one element through software or hardware.

[0124] Finally, it should be noted that the above description has been given for purposes of illustration and description. In addition, the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A carbon emission monitoring system for railway infrastructure in the materialization stage, characterized in that: include: A materialization stage list uploading module is used to receive a material list of the materialization stage uploaded by a user, wherein the material list of the materialization stage includes material names and usage amounts of each material, wherein the materialization stage list uploading module is also used to perform material analysis on the material list of the materialization stage to realize abnormality diagnosis of the list; A regional emission factor acquisition module is used to obtain the regional emission factors corresponding to each material from a public database, where the regional emission factors are carbon emissions per unit mass; A carbon emission total value calculation module, used to calculate the total carbon emission value of the physical and chemical stage based on the regional emission factors corresponding to the various materials and the physical and chemical stage material list; The carbon emission total target setting module is used to extract the expected value of carbon emission total in the chemical stage from the background database; A carbon emission monitoring module is used to calculate the difference between the total carbon emission value of the materialization stage and the expected value of the total carbon emission of the materialization stage to obtain a total carbon emission target offset value, and then determine whether to generate a carbon emission warning prompt based on the comparison between the total carbon emission target offset value and a preset threshold value; The materialization stage list upload module includes: A material list structured coding unit, used for performing structured coding on the material list of the physicalization stage to obtain a set of material property-amount semantic description coding vectors; A material context dynamic walk unit, used for performing causally triggered material context dynamic walk semantic association coding on the set of material property-usage semantic description coding vectors to obtain a material context dynamic walk semantic coding vector; An abnormality diagnosis unit, used to determine a diagnosis result based on the contextual dynamic walk semantic coding vector between the materials, wherein the diagnosis result is used to indicate whether there is an abnormality in the material list at the materialization stage; The inter-material context dynamic walking unit includes: An implicit feature mining subunit is used to perform implicit feature mining on each material property-usage semantic description coding vector in the set of material property-usage semantic description coding vectors to obtain a set of material property-usage deep implicit semantic feature coding vectors; A causal association topological feature extraction subunit is used to extract the semantic causal association topological features between the sets of material property-usage deep implicit semantic feature encoding vectors to obtain a semantic causal association topological feature matrix between material data; A context semantic association coding subunit, for performing multi-level dynamic walking semantic association coding on the set of material property-amount semantic description coding vectors based on the semantic causal association topological feature matrix between the material data to obtain the context dynamic walking semantic coding vector between the materials; The context semantic association encoding subunit is used to: Inputting the semantic causal association topological feature matrix between the material data and the set of the material property-usage semantic description encoding vectors into a dynamic walking encoder based on a graph convolutional neural network model to obtain a surface context dynamic walking semantic encoding vector between the material data; Inputting the semantic causal association topological feature matrix between the material data and the set of the material property-usage deep implicit semantic feature encoding vectors into the dynamic walking encoder based on the graph convolutional neural network model to obtain the hidden layer context dynamic walking semantic encoding vector between the material data; The hidden layer context dynamic wandering semantic coding vector between the material data and the surface layer context dynamic wandering semantic coding vector between the material data are fused to obtain the context dynamic wandering semantic coding vector between the materials.

2. The railway infrastructure materialization stage carbon emission monitoring system according to claim 1 is characterized in that: The bill of materials structured coding unit is used to: Performing semantic embedding coding on each material in the material list of the materialization stage to obtain a set of material semantic description coding vectors; The amount of the corresponding material is added to the end of each material semantic description coding vector in the set of material semantic description coding vectors to obtain the set of material property-amount semantic description coding vectors.

3. The railway infrastructure materialization stage carbon emission monitoring system according to claim 2 is characterized in that: The causal association topological feature extraction subunit includes: A causal association factor calculation secondary subunit is used to calculate the material data semantic causal association factor between any two material property-usage depth implicit semantic feature coding vectors in the set of material property-usage depth implicit semantic feature coding vectors to obtain a material data semantic causal association topological matrix composed of multiple material data semantic causal association factors; The causal triggering secondary subunit is used to input the semantic causal association topological matrix between the material data into the causal triggering network based on the gated activation function to obtain the semantic causal association topological feature matrix between the material data.

4. The railway infrastructure materialization stage carbon emission monitoring system according to claim 3 is characterized in that: The causal correlation factor calculation secondary subunit is used to: Performing semantic association coding on any two material property-usage depth implicit semantic feature coding vectors in the set of material property-usage depth implicit semantic feature coding vectors to obtain a set of semantic association coding matrices between material data; Based on the statistical characteristics of each semantic association coding matrix between material data in the set of semantic association coding matrices between material data, the corresponding semantic causal association factors between material data are calculated to obtain the multiple semantic causal association factors between material data, wherein the statistical characteristics include the characteristic variance, maximum eigenvalue, characteristic mean and causal association energy bias term of the semantic association coding matrix between material data.

5. The railway infrastructure materialization stage carbon emission monitoring system according to claim 4 is characterized in that: If the feature variance of the semantic association coding matrix between the material data is greater than or equal to a preset threshold, the average weighted value of the Euclidean distance between each two material property-usage depth implicit semantic feature coding vectors in the set of the material property-usage depth implicit semantic feature coding vectors is calculated as the causal association energy bias term; If the feature variance of the semantic association coding matrix between the material data is less than a preset threshold, a weighted value of the feature mean of the semantic association coding matrix between the material data is calculated as the causal association energy bias item.

6. The railway infrastructure materialization stage carbon emission monitoring system according to claim 5 is characterized in that: The abnormality diagnosis unit is used for: The context dynamic walk semantic encoding vector between materials is input into a classifier-based inventory diagnostician to obtain the diagnostic result.

7. A method for monitoring carbon emissions during the materialization phase of railway infrastructure, applied to the system for monitoring carbon emissions during the materialization phase of railway infrastructure as claimed in claim 6, characterized in that: include: receiving a material list of the physical and chemical stage uploaded by a user, wherein the material list of the physical and chemical stage includes material names and usage amounts of each material, and performing material analysis on the material list of the physical and chemical stage to realize abnormality diagnosis of the material list; Obtain the regional emission factor corresponding to each material from a public database, where the regional emission factor is the carbon emission per unit mass; Calculate the total carbon emission value of the physical and chemical stage based on the regional emission factors corresponding to each material and the list of materials in the physical and chemical stage; Extract the expected value of total carbon emissions in the chemical process from the backend database; The difference between the total carbon emission value in the materialization stage and the expected value of the total carbon emission in the materialization stage is calculated to obtain the total carbon emission target offset value, and then based on the comparison between the total carbon emission target offset value and the preset threshold, it is determined whether to generate a carbon emission warning prompt.

8. The method for monitoring carbon emissions during the materialization phase of railway infrastructure according to claim 7, characterized in that: Perform material analysis on the material list of the physical and chemical stage to achieve abnormality diagnosis of the material list, including: Performing structured coding on the material list of the physicalization stage to obtain a set of material property-amount semantic description coding vectors; Performing causally triggered dynamic walking semantic association coding between materials on the set of material property-usage semantic description coding vectors to obtain contextual dynamic walking semantic coding vectors between materials; Based on the contextual dynamic walk semantic coding vector between the materials, a diagnosis result is determined, and the diagnosis result is used to indicate whether there is an abnormality in the material list in the materialization stage.

Images