A method for predicting carbon emissions from engineering projects based on knowledge graph

Abstract

The present invention discloses a method for predicting carbon emissions of engineering projects based on a knowledge graph, which belongs to the field of carbon emission processing technology and includes the following steps: S1. According to the carbon emissions of transportation equipment and construction equipment in the engineering project, a number of entity nodes and a number of edges are determined to construct a knowledge graph for the engineering project; S2. According to the characteristic parameters of the knowledge graph corresponding to the engineering project, the knowledge graph factors of the engineering project are determined; S3. According to the knowledge graph factors, a carbon emission prediction result is generated for the engineering project. The present invention uses knowledge graph factors for prediction. Compared with some complex multi-factor comprehensive prediction models, it reduces unnecessary calculation links and improves prediction efficiency. It can provide carbon emission prediction results for engineering projects in a relatively short time, meeting the demand for rapid decision support in actual engineering.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon emission processing, and specifically relates to a method for predicting carbon emissions from engineering projects based on a knowledge graph. Background Art

[0002] As a significant area of ​​energy consumption and carbon emissions, accurate carbon emission forecasting for engineering projects is crucial for developing scientific and rational energy conservation and emission reduction strategies and achieving sustainable development. Accurate carbon emission forecasting can help project managers plan resource allocation, optimize construction processes, and select low-carbon equipment and technologies, effectively reducing the carbon intensity of engineering projects and minimizing their negative impact on the environment.

[0003] Currently, carbon emission forecasting for engineering projects primarily relies on traditional statistical models and empirical formulas. These methods, typically based on historical data and simple linear regression analysis, can provide estimates of carbon emissions to a certain extent, but they have numerous limitations.

[0004] Traditional methods rely on limited historical data and fail to fully account for factors such as equipment type, resulting in significant discrepancies between forecasts and actual conditions. Furthermore, during project implementation, carbon emissions can change dynamically with factors such as construction progress and equipment usage. Traditional methods typically employ static forecasting models that fail to reflect these dynamic changes in real time, making them inadequate for precise carbon emissions management required by engineering projects.

[0005] As an emerging knowledge representation and management technology, knowledge graph can effectively represent and process complex knowledge and relationships by organizing and storing entities and relationships in the form of a graph structure. In recent years, knowledge graph technology has been widely used in many fields, such as intelligent search, recommendation systems, and medical diagnosis, and has achieved remarkable results. In the field of engineering projects, knowledge graphs can integrate various information of engineering projects, including equipment information, to form a comprehensive and systematic knowledge system. Through knowledge graphs, the correlation between various elements in engineering projects can be clearly displayed, providing strong support for in-depth analysis and understanding of the carbon emission mechanism of engineering projects. Therefore, the present invention proposes a method for predicting carbon emissions of engineering projects based on knowledge graphs. Summary of the Invention

[0006] In order to solve the above problems, the present invention proposes a carbon emission prediction method for engineering projects based on knowledge graph.

[0007] The technical solution of the present invention is: a method for predicting carbon emissions from engineering projects based on knowledge graphs, comprising the following steps:

[0008] S1. Based on the carbon emissions of transportation equipment and construction equipment in the project, determine several entity nodes and several edges to build a knowledge graph for the project;

[0009] S2. Determine the knowledge graph factor of the engineering project based on the characteristic parameters of the knowledge graph corresponding to the engineering project;

[0010] S3. Generate carbon emission prediction results for engineering projects based on knowledge graph factors.

[0011] Furthermore, S1 includes the following sub-steps:

[0012] S11. The transportation equipment that generates carbon emissions in the project is considered as a transportation equipment ontology set, and the construction equipment that generates carbon emissions in the project is considered as a construction equipment ontology set;

[0013] S12. Constructing attribute relationships for each ontology in the transportation equipment ontology set;

[0014] S13, constructing attribute relationships for each ontology in the construction equipment ontology set;

[0015] S14. Take all ontologies of the transportation equipment ontology set and the construction equipment ontology set as entity nodes of the knowledge graph, and take attribute relationships as edges of the knowledge graph to construct the knowledge graph of the engineering project.

[0016] The beneficial effect of this further solution is that, in this invention, the construction equipment entity set includes all construction equipment that has generated carbon emissions in the project to date (such as excavators, cranes, concrete mixing plants, and welders). Transportation equipment includes trucks, cranes, and pump trucks. By combining the equipment entities as entity nodes and attribute relationships as edges, the equipment in actual projects is integrated with the knowledge graph structure. This intuitively presents the relationships between equipment and their attributes, facilitating the integration and analysis of information related to project carbon emissions.

[0017] Furthermore, in S12, the attribute values ​​of the ontology in the transportation equipment ontology set are used as attribute relations, and the attribute values The calculation formula is:

[0018] ;

[0019] Where, Indicates that the ontology in the transportation equipment ontology set is The actual carbon emissions generated at any moment, Indicates the total time, Represents the average of the real carbon emissions generated by the total time of the transport equipment ontology set, Represents the exponential function.

[0020] Furthermore, S13 includes the following sub-steps:

[0021] S131. Obtain the real carbon emission sequence of the entity in the construction equipment collection and predicting carbon emission sequences ;in, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at time 1, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at time 2, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at 3 moments, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at 4 moments, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, Indicates the total time;

[0022] S132. Calculate the root mean square error between the actual carbon emission series and the predicted carbon emission series;

[0023] S133, calculating the absolute value of the difference between the actual carbon emission value and the root mean square error of the construction equipment body set at each moment, and determining the moment corresponding to the minimum absolute value;

[0024] S134. Calculate the attribute value of the entity according to the time corresponding to the minimum absolute value;

[0025] S135. Use the attribute values ​​of the ontology in the transportation equipment ontology set as attribute relationships.

[0026] The beneficial effect of the above further solution is that in the present invention, modern engineering equipment undergoes life testing and carbon emissions testing before being put into operation, so each device has its own estimated carbon emissions at each moment. The actual carbon emission sequence and predicted carbon emission sequence at different moments cover data from multiple time points from the initial moment to the total moment K. By calculating the absolute value of the difference between the actual carbon emission value and the root mean square error at each moment, determining the moment corresponding to the minimum absolute value, it is possible to accurately find the moment closest to the root mean square error, thereby mapping the attribute values ​​of the construction equipment body to the transportation equipment body set, forming an attribute relationship that reflects the attribute characteristics of the equipment under the optimal predicted performance.

[0027] Furthermore, in S134, the attribute value of the ontology in the construction equipment ontology set The calculation formula is:

[0028]

[0029] Where, Represents the carbon emission change curve of the entities in the construction equipment entity set, represents the time corresponding to the minimum absolute value, Indicates the total time, Indicates the actual carbon emission value of the entity at time 1, Indicates that the entity is The actual carbon emission value at the moment, Indicates the average calculation of the actual carbon emissions over a period of time.

[0030] Constructing data into a curve helps you intuitively understand the data trend and assists in selecting the appropriate integration method. The upper and lower limits of the integration can be adjusted appropriately.

[0031] Furthermore, S2 includes the following sub-steps:

[0032] S21. Determine the entity granularity of the knowledge graph corresponding to the engineering project;

[0033] S22, extracting the node degree of each node in the knowledge graph corresponding to the engineering project;

[0034] S23. The ratio of the node degree to the entity granularity of each node is used as the weighted weight of the node, and the weighted sum of all node betweenness centralities is performed to obtain the knowledge graph factor of the engineering project.

[0035] The beneficial effect of the above further scheme is that: in the present invention, the frequency of a node as a shortest path bridge measures its control power in the propagation of carbon emissions. The ratio of node degree to entity granularity is used as the weighted weight, taking into account the relative importance of the node in the knowledge graph, and combining the overall scale of the knowledge graph (entity granularity). This weighting method can, to a certain extent, balance the importance differences of nodes in knowledge graphs of different sizes. Node betweenness centrality reflects the ability of a node to connect other node pairs as a "bridge" in the knowledge graph network. The knowledge graph factor is obtained by weighted summation of all node betweenness centralities, attempting to comprehensively consider the importance of each node in the knowledge graph in the overall structure, thereby obtaining a factor that can represent the overall characteristics of the knowledge graph, providing a quantitative indicator based on the knowledge graph structure for subsequent carbon emission prediction. Setting the optimal time period and calculating the average of the total real carbon emissions within the time period takes into account the temporal dynamic changes of carbon emissions of engineering projects. By selecting representative time periods for average calculation, data fluctuations can be smoothed to a certain extent, making the prediction results more stable.

[0036] Furthermore, in S21, the adjacency matrix of the knowledge graph of the engineering project is extracted, and the row and column values ​​of the adjacency matrix are used as the entity granularity of the knowledge graph.

[0037] A two-dimensional matrix is ​​used to represent the relationship between vertices in a knowledge graph. The adjacency matrix is ​​a data structure used to represent a graph structure. The rows and columns correspond to entities in the knowledge graph, and the elements in the matrix represent the connection relationship between entities.

[0038] Furthermore, in S3, an optimal period is set, and the product of the average total real carbon emission value of the engineering project within the optimal period and the knowledge graph factor is used as the carbon emission prediction result of the engineering project.

[0039] The optimal time period can be set manually, for example, the last three moments can be used as the optimal time period, or the last five moments can be used as the optimal time period. Try to choose the last stable moment as the optimal time period.

[0040] The beneficial effects of the present invention are:

[0041] (1) This invention constructs a knowledge graph based on the entity nodes and edges of the carbon emissions of transportation equipment and construction equipment, which can effectively integrate the carbon emission-related information of these two types of key equipment in engineering projects. The carbon emission data of different equipment have different characteristics and sources, which can be organized uniformly through the knowledge graph to achieve data association and fusion;

[0042] (2) The present invention determines the knowledge graph factor based on the knowledge graph feature parameters, deeply mines the information contained in the knowledge graph, extracts the relationship data containing a large number of nodes and edges, and can convert complex graph structure information into quantitative indicators that can be used for prediction, giving full play to the role of knowledge graph in analysis and prediction;

[0043] (3) The present invention uses knowledge graph factors for prediction. Compared with some complex multi-factor comprehensive prediction models, it reduces unnecessary calculation links and improves prediction efficiency. It can provide carbon emission prediction results for engineering projects in a relatively short time, meeting the demand for rapid decision-making support in actual engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flowchart of the carbon emission prediction method for engineering projects based on knowledge graph. DETAILED DESCRIPTION

[0045] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0046] like Figure 1 As shown, the present invention provides a method for predicting carbon emissions of engineering projects based on a knowledge graph, comprising the following steps:

[0047] S1. Based on the carbon emissions of transportation equipment and construction equipment in the project, determine several entity nodes and several edges to build a knowledge graph for the project;

[0048] S2. Determine the knowledge graph factor of the engineering project based on the characteristic parameters of the knowledge graph corresponding to the engineering project;

[0049] S3. Generate carbon emission prediction results for engineering projects based on knowledge graph factors.

[0050] In this embodiment of the present invention, S1 includes the following sub-steps:

[0051] S11. The transportation equipment that generates carbon emissions in the project is considered as a transportation equipment ontology set, and the construction equipment that generates carbon emissions in the project is considered as a construction equipment ontology set;

[0052] S12. Constructing attribute relationships for each ontology in the transportation equipment ontology set;

[0053] S13, constructing attribute relationships for each ontology in the construction equipment ontology set;

[0054] S14. Take all ontologies of the transportation equipment ontology set and the construction equipment ontology set as entity nodes of the knowledge graph, and take attribute relationships as edges of the knowledge graph to construct the knowledge graph of the engineering project.

[0055] In this paper, the construction equipment entity set includes all construction equipment that has generated carbon emissions from the project to date (such as excavators, cranes, concrete mixing plants, and welders). Transportation equipment includes trucks, cranes, and pump trucks. By combining the equipment entities as entity nodes and their attribute relationships as edges, this approach integrates the equipment in actual projects with the knowledge graph structure. This intuitively presents the relationships between equipment and their attributes, facilitating the integration and analysis of information related to project carbon emissions.

[0056] In the embodiment of the present invention, in S12, the attribute value of the ontology in the transportation equipment ontology set is used as the attribute relationship, and its attribute value The calculation formula is:

[0057] ;

[0058] Where, Indicates that the ontology in the transportation equipment ontology set is The actual carbon emissions generated at any moment, Indicates the total time, Represents the average of the real carbon emissions generated by the total time of the transport equipment ontology set, Represents the exponential function.

[0059] In this embodiment of the present invention, S13 includes the following sub-steps:

[0060] S131. Obtain the real carbon emission sequence of the entity in the construction equipment collection and predicting carbon emission sequences ;in, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at time 1, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at time 2, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at 3 moments, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at 4 moments, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, Indicates the total time;

[0061] S132. Calculate the root mean square error between the actual carbon emission series and the predicted carbon emission series;

[0062] S133, calculating the absolute value of the difference between the actual carbon emission value and the root mean square error of the construction equipment body set at each moment, and determining the moment corresponding to the minimum absolute value;

[0063] S134. Calculate the attribute value of the entity according to the time corresponding to the minimum absolute value;

[0064] S135. Use the attribute values ​​of the ontology in the transportation equipment ontology set as attribute relationships.

[0065] In this invention, modern engineering equipment undergoes lifespan testing and carbon emissions testing before being put into operation. Therefore, each device has its own estimated carbon emissions at each moment. The actual carbon emission sequence and the predicted carbon emission sequence at different moments cover data from multiple time points from the initial moment to the total moment K. By calculating the absolute value of the difference between the actual carbon emission value and the root mean square error at each moment, and determining the moment corresponding to the minimum absolute value, it is possible to accurately find the moment closest to the root mean square error, thereby mapping the attribute values ​​of the construction equipment body to the set of transportation equipment bodies, forming an attribute relationship that reflects the attribute characteristics of the equipment under optimal prediction performance.

[0066] In the embodiment of the present invention, in S134, the attribute value of the ontology in the construction equipment ontology set The calculation formula is:

[0067]

[0068] Where, Represents the carbon emission change curve of the entities in the construction equipment entity set, represents the time corresponding to the minimum absolute value, Indicates the total time, Indicates the actual carbon emission value of the entity at time 1, Indicates that the entity is The actual carbon emission value at the moment, Indicates the average calculation of the actual carbon emissions over a period of time.

[0069] Constructing data into a curve helps you intuitively understand the data trend and assists in selecting the appropriate integration method. The upper and lower limits of the integration can be adjusted appropriately.

[0070] In this embodiment of the present invention, S2 includes the following sub-steps:

[0071] S21. Determine the entity granularity of the knowledge graph corresponding to the engineering project;

[0072] S22, extracting the node degree of each node in the knowledge graph corresponding to the engineering project;

[0073] S23. The ratio of the node degree to the entity granularity of each node is used as the weighted weight of the node, and the weighted sum of all node betweenness centralities is performed to obtain the knowledge graph factor of the engineering project.

[0074] In the present invention, the frequency of a node as a shortest path bridge measures its control over the propagation of carbon emissions. The ratio of node degree to entity granularity is used as a weighted weight, taking into account the relative importance of the node in the knowledge graph, and combining the overall scale of the knowledge graph (entity granularity). This weighting method can, to a certain extent, balance the importance differences of nodes in knowledge graphs of different sizes. Node betweenness centrality reflects the ability of a node to connect other node pairs as a "bridge" in the knowledge graph network. The knowledge graph factor is obtained by weighted summation of all node betweenness centralities, attempting to comprehensively consider the importance of each node in the knowledge graph in the overall structure, thereby obtaining a factor that can represent the overall characteristics of the knowledge graph, providing a quantitative indicator based on the knowledge graph structure for subsequent carbon emission predictions. The optimal time period is set and the average of the total real carbon emissions within the time period is calculated. The temporal dynamic changes of carbon emissions of engineering projects are taken into account. By selecting representative time periods for average calculation, data fluctuations can be smoothed to a certain extent, making the prediction results more stable.

[0075] In an embodiment of the present invention, in S21, the adjacency matrix of the knowledge graph of the engineering project is extracted, and the row and column values ​​of the adjacency matrix are used as the entity granularity of the knowledge graph.

[0076] A two-dimensional matrix is ​​used to represent the relationship between vertices in a knowledge graph. The adjacency matrix is ​​a data structure used to represent a graph structure. The rows and columns correspond to entities in the knowledge graph, and the elements in the matrix represent the connection relationship between entities.

[0077] In an embodiment of the present invention, in S3, an optimal time period is set, and the product of the average total real carbon emission value of the engineering project within the optimal time period and the knowledge graph factor is used as the carbon emission prediction result of the engineering project.

[0078] The optimal time period can be set manually, for example, the last three moments can be used as the optimal time period, or the last five moments can be used as the optimal time period. Try to choose the last stable moment as the optimal time period.

[0079] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A method for predicting carbon emissions from engineering projects based on knowledge graphs, characterized in that: The following steps are involved: S1. Based on the carbon emissions of transportation equipment and construction equipment in the project, determine several entity nodes and several edges to build a knowledge graph for the project; S2. Determine the knowledge graph factor of the engineering project based on the characteristic parameters of the knowledge graph corresponding to the engineering project; S3. Generate carbon emission prediction results for engineering projects based on knowledge graph factors; The S1 comprises the following sub-steps: S11. The transportation equipment that generates carbon emissions in the project is considered as a transportation equipment ontology set, and the construction equipment that generates carbon emissions in the project is considered as a construction equipment ontology set; S12. Constructing attribute relationships for each ontology in the transportation equipment ontology set; S13, constructing attribute relationships for each ontology in the construction equipment ontology set; S14. Build a knowledge graph for the engineering project by taking all ontologies of the transportation equipment ontology set and the construction equipment ontology set as entity nodes of the knowledge graph and attribute relationships as edges of the knowledge graph; In the above S12, the attribute value of the ontology in the transportation equipment ontology set is used as the attribute relationship, and its attribute value The calculation formula is: ； Where, Indicates that the ontology in the transportation equipment ontology set is The actual carbon emissions generated at any moment, Indicates the total time, Represents the average of the real carbon emissions generated by the total time of the transport equipment ontology set, represents the exponential function; The S13 includes the following sub-steps: S131. Obtain the real carbon emission sequence of the entity in the construction equipment collection and predicting carbon emission sequences ;in, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at time 1, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at time 2, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at 3 moments, and They represent the actual carbon emission value and the estimated carbon emission value of the entity in the construction equipment set at 4 moments, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, and Respectively represent the entities in the construction equipment set The actual carbon emission value and estimated carbon emission value at the moment, Indicates the total time; S132. Calculate the root mean square error between the actual carbon emission series and the predicted carbon emission series; S133, calculating the absolute value of the difference between the actual carbon emission value and the root mean square error of the construction equipment body set at each moment, and determining the moment corresponding to the minimum absolute value; S134. Calculate the attribute value of the entity according to the time corresponding to the minimum absolute value; S135, taking the attribute values ​​of the ontology in the construction equipment ontology set as attribute relations; In the above S134, the attribute value of the ontology in the construction equipment ontology set The calculation formula is: ； Where, Represents the carbon emission change curve of the entities in the construction equipment entity set, represents the time corresponding to the minimum absolute value, Indicates the total time, Indicates the actual carbon emission value of the entity at time 1, Indicates that the entity is The actual carbon emission value at the moment, Indicates the average calculation of the actual carbon emissions over a period of time.

2. The method for predicting carbon emissions from engineering projects based on knowledge graph according to claim 1 is characterized in that: The S2 includes the following sub-steps: S21. Determine the entity granularity of the knowledge graph corresponding to the engineering project; S22, extracting the node degree of each node in the knowledge graph corresponding to the engineering project; S23. The ratio of the node degree to the entity granularity of each node is used as the weighted weight of the node, and the weighted sum of all node betweenness centralities is performed to obtain the knowledge graph factor of the engineering project.

3. The method for predicting carbon emissions from engineering projects based on knowledge graph according to claim 2 is characterized in that: In S21, the adjacency matrix of the knowledge graph of the engineering project is extracted, and the row and column values ​​of the adjacency matrix are used as the entity granularity of the knowledge graph.

4. The method for predicting carbon emissions from engineering projects based on knowledge graph according to claim 1 is characterized in that: In S3, an optimal period is set, and the product of the average of the total real carbon emission values ​​of the engineering project within the optimal period and the knowledge graph factor is used as the carbon emission prediction result of the engineering project.

Images