Urban carbon metering method and prediction method and device based on knowledge graph

Abstract

The present disclosure provides a knowledge graph-based urban carbon measurement method and device, which can be applied to the field of low-carbon management. The method comprises: constructing an initial knowledge graph of carbon activity content in a city for a city area, obtaining carbon activity attribute data and time attribute data of the initial subject attribute node according to the initial knowledge graph, and the time attribute data corresponding to the time node; performing a carbon emission calculation task according to the carbon activity attribute data and the time attribute data, and the carbon emission calculation parameters related to the carbon activity node to obtain subject carbon emission information; updating the initial carbon emission information node having an attribute relationship with the carbon activity node according to the subject carbon emission information to obtain a target knowledge graph, and obtaining an urban carbon measurement result according to the target knowledge graph. The present application can be used for recording and storing multi-dimensional network information of urban carbon activities and relationship analysis, and can realize the carbon emission prediction function of urban areas, and realize efficient low-carbon management.

Description

Technical Field

[0001] This disclosure relates to the field of low-carbon governance, and more specifically to a knowledge graph-based urban carbon measurement method, prediction method, and device. Background Technology

[0002] Urban carbon emissions are a major source of global carbon emissions. Studying the characteristics of communities and transportation during urban operations and their corresponding carbon emissions is crucial for revealing the mechanisms of urban carbon emissions and achieving urban carbon reduction and control. The inventors have found that current low-carbon city governance is mainly limited to total carbon emission accounting and is highly lagging. It fails to integrate and correlate urban carbon emission activities from a bottom-up perspective, considering all factors such as the subject, form, time, and spatial location, and the types and relationships of urban built environment and urban operational activities. This makes it difficult to finely characterize and trace the carbon activities of specific subjects, and prevents a multi-dimensional expression of the complex mapping relationship between carbon activities and urban spatial elements. Therefore, it is necessary to conduct refined and intelligent modeling and analysis based on urban carbon emission data to reveal the mapping relationship and impact mechanisms between the urban built environment and carbon emissions. Summary of the Invention

[0003] In view of the above problems, this disclosure provides a knowledge graph-based urban carbon measurement method, prediction method, and apparatus.

[0004] According to the first aspect of this disclosure, a knowledge graph-based urban carbon measurement method is provided, comprising: constructing an initial knowledge graph of carbon activities within an urban area, wherein the initial knowledge graph includes an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph; the initial carbon emission subject subgraph includes a subject root node representing four types of carbon emission subjects, and initial subject attribute nodes having attribute relationships with the subject root node; the four types of carbon emission subjects include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects; the initial activity information subgraph includes carbon activity nodes related to carbon emission subjects; the time information subgraph includes time nodes representing the time or period of carbon activity occurrence; and the initial carbon emission subgraph includes carbon... For activity-related carbon emission nodes, the root nodes of each subgraph in the initial knowledge graph have attribute relationships, including at least one of the following: containment, generation, location, connection, time, and attribute. Based on the initial knowledge graph, carbon activity attribute data and time attribute data of the initial subject attribute nodes are obtained, with the time attribute data corresponding to the time nodes. Based on the carbon activity attribute data and time attribute data of the initial subject attribute nodes, as well as the carbon emission calculation parameters related to the carbon activity nodes, a carbon emission calculation task is performed to obtain the subject carbon emission information related to the initial subject attribute nodes. Based on the subject carbon emission information, the initial carbon emission information nodes with attribute relationships to the carbon activity nodes are updated to obtain the target knowledge graph. Based on the target knowledge graph, the city carbon measurement results are obtained.

[0005] According to a second aspect of this disclosure, a carbon emission prediction method based on a target knowledge graph is provided, comprising: determining a target knowledge graph related to carbon activities within a city, the target knowledge graph including a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph, wherein the nodes of the time information subgraph represent time attribute data related to urban carbon emission activities, the nodes of the carbon emission subject subgraph include subject root nodes related to land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects, as well as subject attribute nodes having attribute relationships with the subject root nodes, and the carbon emission subgraph... The nodes in the activity information subgraph represent the carbon emission information of the carbon emission subject, and the nodes in the activity information subgraph represent the carbon activity attribute data of the carbon emission subject. The carbon emission subject is related to the time attribute data. The triples in the target knowledge graph are determined. The triples include target carbon emission subject nodes, target carbon activity nodes, and target carbon emission information nodes with attribute relationships. The target carbon activity nodes represent the carbon activity attribute data of the carbon emission subject, and the target carbon emission information nodes represent the carbon emission information of the carbon emission subject. The graph structure corresponding to the triples is input into the prediction model, and the carbon emission prediction results of the city are output.

[0006] A third aspect of this disclosure provides a knowledge graph-based urban carbon metering device, comprising: a construction module for constructing an initial knowledge graph of carbon activities within an urban area, wherein the initial knowledge graph includes an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph. The initial carbon emission subject subgraph includes a subject root node representing four types of carbon emission subjects, and initial subject attribute nodes having attribute relationships with the subject root node. The four types of carbon emission subjects include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects. The initial activity information subgraph includes carbon activity nodes related to carbon emission subjects. The time information subgraph includes time nodes representing the time or period of carbon activity occurrence. The initial carbon emission subgraph includes carbon emission amount nodes related to carbon activities. Each of the initial knowledge graph subgraphs has a subject root node. The root nodes have attribute relationships, which include at least one of the following: contain, generate, located, connected, time, and attribute; an acquisition module is used to acquire carbon activity attribute data and time attribute data of the initial subject attribute nodes based on the initial knowledge graph, with the time attribute data corresponding to the time nodes; a subject carbon emission information acquisition module is used to perform carbon emission calculation tasks based on the carbon activity attribute data and time attribute data of the initial subject attribute nodes, as well as the carbon emission calculation parameters related to the carbon activity nodes, to obtain subject carbon emission information related to the initial subject attribute nodes; a target knowledge graph acquisition module is used to update the initial carbon emission information nodes with attribute relationships to the carbon activity nodes based on the subject carbon emission information to obtain the target knowledge graph; and a city carbon measurement result acquisition module is used to obtain the city carbon measurement result based on the target knowledge graph.

[0007] The fourth aspect of this disclosure provides a carbon emission prediction device based on a target knowledge graph, comprising: a processing module for determining a target knowledge graph related to carbon activities within a city, the target knowledge graph including a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph, wherein the nodes of the time information subgraph represent time attribute data related to urban carbon emission activities, the nodes of the carbon emission subject subgraph include subject root nodes related to land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects, and subject attribute nodes having attribute relationships with the subject root nodes, and the nodes of the carbon emission subgraph represent carbon emission... The system comprises three modules: a target carbon emission subject information module and a target carbon emission information module. The target carbon emission subject's nodes represent carbon activity attribute data related to the carbon emission subject, and the carbon emission subject is associated with time attribute data. A triplet determination module is used to determine triples in the target knowledge graph. Triples include target carbon emission subject nodes, target carbon activity nodes, and target carbon emission information nodes with attribute relationships. Target carbon activity nodes represent the carbon activity attribute data of the carbon emission subject, and target carbon emission information nodes represent the subject's carbon emission information. A prediction result output module is used to input the graph structure corresponding to the triples into the prediction model and output the city's carbon emission prediction results.

[0008] According to embodiments of this disclosure, for urban areas, an initial knowledge graph of carbon activities within the city can be constructed. The initial knowledge graph may include an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph. The initial carbon emission subject subgraph may include subject root nodes representing four types of carbon emission subjects, and initial subject attribute nodes that have attribute relationships with the subject root nodes. The four types of carbon emission subjects may include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects. The initial activity information subgraph may include carbon activity nodes related to carbon emission subjects. The time information subgraph may include time nodes representing the time or period of carbon activity occurrence. The initial carbon emission subgraph may include carbon emission amount nodes related to carbon activities. Each subject root node in the initial knowledge graph subgraph has attribute relationships, which may include at least one of the following: containing, generating, located, connected, time, and attribute. Based on the initial knowledge graph, initial subject attributes can be obtained. The system generates carbon activity attribute data and time attribute data for initial subject attribute nodes, with the time attribute data corresponding to the time nodes. Based on the initial subject attribute node's carbon activity attribute data and time attribute data, as well as the carbon emission calculation parameters related to the carbon activity node, a carbon emission calculation task is performed to obtain the subject carbon emission information related to the initial subject attribute node. The initial carbon emission information nodes with attribute relationships to the carbon activity node are updated based on the subject carbon emission information, resulting in a target knowledge graph. Based on the target knowledge graph, the city's carbon measurement results are obtained. The constructed target knowledge graph efficiently integrates temporal, spatial, and subject-related process information related to carbon activities within the city. Through the interconnection of relationships, a multi-dimensional network knowledge structure is formed, representing and storing various elements related to carbon activities in the form of a directed graph. This improves the accuracy and efficiency of automated carbon emission measurement, realizes a comprehensive expression of the mapping relationship between carbon emissions and planning, facilitates accurate traceability of carbon measurement results, and provides efficient support for carbon reduction and control. Attached Figure Description

[0009] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0010] Figure 1 A flowchart of a knowledge graph-based urban carbon metering method according to an embodiment of the present disclosure is shown;

[0011] Figure 2 A schematic diagram of a target knowledge graph according to an embodiment of the present disclosure is shown;

[0012] Figure 3 A flowchart of a carbon emission prediction method based on a target knowledge graph according to an embodiment of the present disclosure is shown;

[0013] Figure 4A structural block diagram of a knowledge graph-based urban carbon metering device according to an embodiment of the present disclosure is shown;

[0014] Figure 5 A structural block diagram of a carbon emission prediction device based on a target knowledge graph according to an embodiment of the present disclosure is shown; and

[0015] Figure 6 A block diagram of an electronic device suitable for implementing a knowledge graph-based urban carbon measurement and prediction method is shown according to embodiments of the present disclosure. Detailed Implementation

[0016] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0017] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0018] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0019] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0020] Urban carbon emissions are a major source of global carbon emissions. Studying the characteristics of communities and transportation during urban operations and their corresponding carbon emissions is crucial for revealing the mechanisms of urban carbon emissions and achieving urban carbon reduction and control. The inventors have found that current low-carbon city governance is mainly limited to total carbon emission accounting and is highly lagging. It fails to integrate and correlate urban carbon emission activities from a bottom-up perspective, considering all factors such as the subject, form, time, and spatial location, and the types and relationships of urban built environment and urban operational activities. This makes it difficult to finely characterize and trace the carbon activities of specific subjects, and prevents a multi-dimensional expression of the complex mapping relationship between carbon activities and urban spatial elements. Therefore, it is necessary to conduct refined and intelligent modeling and analysis based on urban carbon emission data to reveal the mapping relationship and impact mechanisms between the urban built environment and carbon emissions.

[0021] In view of this, this disclosure provides a knowledge graph-based urban carbon measurement method, including: constructing an initial knowledge graph of carbon activities within an urban area, the initial knowledge graph including an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph; the initial carbon emission subject subgraph including subject root nodes representing four types of carbon emission subjects, and initial subject attribute nodes having attribute relationships with the subject root nodes; the four types of carbon emission subjects including land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects; the initial activity information subgraph including carbon activity nodes related to the carbon emission subjects; the time information subgraph including time nodes representing the time or period of carbon activity occurrence; and the initial carbon emission subgraph including nodes related to carbon activities. The initial knowledge graph subgraphs have attribute relationships between their respective root nodes, which include at least one of the following: containment, generation, location, connection, time, and attribute. Based on the initial knowledge graph, the carbon activity attribute data and time attribute data of the initial subject attribute nodes are obtained, with the time attribute data corresponding to the time nodes. A carbon emission calculation task is performed based on the initial subject attribute node's carbon activity attribute data and time attribute data, as well as the carbon emission calculation parameters related to the carbon activity nodes, to obtain the subject carbon emission information related to the initial subject attribute nodes. The initial carbon emission information nodes with attribute relationships to the carbon activity nodes are updated based on the subject carbon emission information to obtain the target knowledge graph. Based on the target knowledge graph, the city carbon measurement results are obtained.

[0022] In the technical solution disclosed herein, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse.

[0023] Figure 1 A flowchart of a knowledge graph-based urban carbon metering method according to an embodiment of this disclosure is shown.

[0024] like Figure 1 As shown, the knowledge graph-based urban carbon measurement method of this embodiment includes operations S110 to S150.

[0025] In operation S110, an initial knowledge graph of carbon activities within the city is constructed for urban areas.

[0026] During operation S120, based on the initial knowledge graph, carbon activity attribute data and time attribute data of the initial subject attribute nodes are obtained, with the time attribute data corresponding to the time nodes.

[0027] In operation S130, a carbon emission calculation task is performed based on the carbon activity attribute data and time attribute data of the initial subject attribute node, as well as the carbon emission calculation parameters related to the carbon activity node, to obtain the subject carbon emission information related to the initial subject attribute node.

[0028] In operation S140, the initial carbon emission information nodes that have attribute relationships with carbon activity nodes are updated based on the main carbon emission information to obtain the target knowledge graph.

[0029] In operation S150, the city carbon measurement results are obtained based on the target knowledge graph.

[0030] According to embodiments of this disclosure, the initial knowledge graph may include an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph. The initial carbon emission subject subgraph may include a subject root node representing four types of carbon emission subjects, and initial subject attribute nodes that have attribute relationships with the subject root node. For example, the subject root node may be a land parcel node representing land parcel information subjects, and the initial subject attribute nodes may include plot ratio nodes, population nodes, etc. The initial subject attribute nodes may be leaf nodes in the initial carbon emission subject subgraph other than the subject root node.

[0031] According to embodiments of this disclosure, the four types of carbon emission subjects may include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects, but are not limited thereto, and this disclosure does not limit them. The initial activity information subgraph may include carbon activity nodes related to the carbon emission subjects. Carbon activity nodes may include waste treatment nodes, industrial process nodes, land use nodes, energy activity nodes, etc., but are not limited thereto, and embodiments of this disclosure do not limit them. The time information subgraph may include time nodes representing the time or period in which carbon activity occurs. Time nodes may include year nodes, month nodes, and day nodes, etc., but are not limited thereto, and embodiments of this disclosure do not limit them. The initial carbon emission subgraph may include carbon emission amount nodes related to carbon activities.

[0032] Optionally, the nodes in the initial knowledge graph are not filled with specific data. The nodes in the initial knowledge graph include data parameters that need to be obtained, such as "area parameters" and "window-to-wall ratio parameters". The main root nodes of each subgraph of the initial knowledge graph have attribute relationships. The attribute relationships include at least one of the following: contain, generate, located, connected, time, attribute, but are not limited to these. The embodiments of this disclosure do not specifically limit the attribute relationships.

[0033] Optionally, based on the initial knowledge graph, carbon activity attribute data and time attribute data of the initial subject attribute nodes can be obtained, with the time attribute data corresponding to time nodes. Specifically, carbon activity attribute data can include specific parameters, such as room parameters, building parameters, vehicle parameters, travel parameters, road parameters, etc. Travel parameters can include travel time, driving speed, etc., and road parameters can include road length, road grade, etc. These carbon activity attribute data are not listed exhaustively here.

[0034] Optionally, a carbon emission calculation task can be performed based on the carbon activity attribute data and time attribute data of the initial subject attribute node, as well as the carbon emission calculation parameters related to the carbon activity node, to obtain the subject carbon emission information related to the initial subject attribute node. Subject carbon emission information could be, for example, carbon emission information for each vehicle, carbon emission information for each room, etc.

[0035] According to embodiments of this disclosure, obtaining carbon activity attribute data requires cleaning, removing noise and outliers from the initially acquired raw data. This disclosure employs a bottom-up carbon emission accounting method proposed in authoritative reports, namely the carbon emission factor method (carbon emission calculation parameters), as shown in Table 1. Taking travel as an example, the carbon emission information of the travel subject related to the travel carbon emission node can be equal to the vehicle's travel distance multiplied by the carbon emission factor per unit distance.

[0036] Table 1 Carbon emission measurement formula

[0037]

[0038] According to embodiments of this disclosure, a target knowledge graph can be obtained by updating the initial carbon emission information nodes that have attribute relationships with the carbon activity nodes based on the main carbon emission information update. The urban carbon measurement results can be calculated based on the obtained target knowledge graph, and the urban carbon measurement results can characterize the results after measuring and statistically analyzing the urban carbon emissions.

[0039] It should be noted that the carbon metering involved in the embodiments of this disclosure can be understood as carbon emission metering. For example, carbon metering may include carbon emission amount and other carbon emission-related information in a preset time period. The embodiments of this disclosure will not be described in detail here.

[0040] According to embodiments of this disclosure, for urban areas, an initial knowledge graph of carbon activities within the city can be constructed. The initial knowledge graph may include an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph. The initial carbon emission subject subgraph may include subject root nodes representing four types of carbon emission subjects, and initial subject attribute nodes that have attribute relationships with the subject root nodes. The four types of carbon emission subjects may include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects. The initial activity information subgraph may include carbon activity nodes related to carbon emission subjects. The time information subgraph may include time nodes representing the time or period of carbon activity occurrence. The initial carbon emission subgraph may include carbon emission amount nodes related to carbon activities. Each subject root node in the initial knowledge graph subgraph has attribute relationships, which may include at least one of the following: containing, generating, located, connected, time, and attribute. Based on the initial knowledge graph, initial subject attributes can be obtained. The system generates carbon activity attribute data and time attribute data for initial subject attribute nodes, with the time attribute data corresponding to the time nodes. Based on the initial subject attribute node's carbon activity attribute data and time attribute data, as well as the carbon emission calculation parameters related to the carbon activity node, a carbon emission calculation task is performed to obtain the subject carbon emission information related to the initial subject attribute node. The initial carbon emission information nodes with attribute relationships to the carbon activity node are updated based on the subject carbon emission information, resulting in a target knowledge graph. Based on the target knowledge graph, the city's carbon measurement results are obtained. The constructed target knowledge graph efficiently integrates temporal, spatial, and subject-related process information related to carbon activities within the city. Through the interconnection of relationships, a multi-dimensional network knowledge structure is formed, representing and storing various elements related to carbon activities in the form of a directed graph. This improves the accuracy and efficiency of automated carbon emission measurement, realizes a comprehensive expression of the mapping relationship between carbon emissions and planning, facilitates accurate traceability of carbon measurement results, and provides efficient support for carbon reduction and control.

[0041] According to embodiments of this disclosure, the carbon emission main subgraph of the target knowledge graph includes a land parcel information subgraph, a functional information subgraph, a location information subgraph, and a movement information subgraph; the main root node of the land parcel information subgraph is a land parcel node that represents the main land parcel information, and the main attribute nodes of the land parcel information subgraph include land parcel attribute nodes that have attribute relationships with the land parcel nodes. The land parcel attribute nodes include at least one of the following: area node, population node, plot ratio node, building height node, and building density node, wherein the land parcel nodes have a generation relationship with the carbon activity nodes in the activity information subgraph.

[0042] For the sub-map of land parcel information, different land parcel characteristics will also affect the city's carbon emissions. Different land parcel characteristics will directly affect the city's carbon emissions. Therefore, land parcel factors can include area, population, plot ratio, building height, building density, etc., but are not limited to these. This disclosure does not limit the land parcel factors. The calculation methods for relevant attributes such as land parcel factors and building factors are shown in Table 2.

[0043] Table 2 Parameter Calculation Method

[0044]

[0045] According to embodiments of this disclosure, the mobile information subgraph may include travel nodes representing the subjects of mobile information, and the subject attribute nodes of the mobile information subgraph may include vehicle nodes that have a generation relationship with the travel nodes. The vehicle attribute nodes that have an attribute relationship with the vehicle nodes may include at least one of the following: fuel type node, type node, and displacement node; the travel node and the land parcel node may have an inclusion relationship, the travel node and the time node in the time subgraph may have a time relationship, and the vehicle node and the energy activity node in the activity information subgraph may have a generation relationship.

[0046] According to embodiments of this disclosure, the mobility information subgraph may include travel nodes and vehicle nodes that have a relationship with the travel nodes. Differences in the technical parameters of the vehicles themselves are closely related to the total amount and intensity of traffic carbon emissions. Therefore, to explore the correlation between different technical parameters of vehicles and their carbon activities, and also to study the connection between these technical data and vehicle operation and carbon emissions more deeply, vehicle attribute nodes may include at least one of the following: fuel type node, emission standard node, vehicle age node, vehicle gross vehicle weight node, etc., but are not limited to these, and embodiments of this disclosure do not limit this. Secondly, vehicle travel activities are also a fundamental cause of carbon emissions. In this disclosure, travel attribute nodes that have an attribute relationship with travel nodes may include at least one of the following: driving speed node, traversed road segment node, and travel time node, but are not limited to these, and embodiments of this disclosure do not limit this.

[0047] According to embodiments of this disclosure, the functional information subgraph includes land use type nodes that represent the main body of functional information. The type attribute nodes that have attribute relationships with the land use type nodes include at least one of the following: industrial land use nodes, warehousing land use nodes, green space nodes, transportation land use nodes, commercial service land use nodes, public service land use nodes, and residential land use nodes; the land use type nodes have attribute relationships with the plot nodes.

[0048] Regarding the functional information subgraph, the land use type nodes in the functional information subgraph and the land parcel nodes in the land parcel information subgraph have attribute relationships. The land use type attribute also affects the city's carbon emissions. Type attribute nodes may also include industrial land use nodes, warehousing land use nodes, green space nodes, transportation land use nodes, commercial service land use nodes, public service land use nodes, and residential land use nodes, etc., but are not limited to these. This disclosure does not specifically limit the attributes of land use types.

[0049] According to embodiments of this disclosure, the location information subgraph may include road segment nodes representing the main body of location information. Nodes with a location relationship with road segment nodes may include at least one of the following: administrative region nodes and road nodes. Road segment attribute nodes with an attribute relationship with road segment nodes may include at least one of the following: road segment length nodes, road segment level nodes, and road segment width nodes. Administrative region attribute nodes with an inclusion relationship with administrative region nodes may include at least one of the following: district nodes and street nodes. Road attribute nodes with an attribute relationship with road nodes may include road name nodes. Nodes with a connection relationship with road segment nodes may include intersection nodes. Intersection attribute nodes with an attribute relationship with intersection nodes may include at least one of the following: longitude nodes and latitude nodes. Road segment nodes and travel nodes may have a location relationship, and road segment nodes and land parcel nodes may have an adjacent relationship.

[0050] According to embodiments of this disclosure, the target knowledge graph may further include a location information subgraph, the main root node of which is a road segment node, and road segment attribute nodes that have attribute relationships with the road segment nodes. The road segment attribute nodes include at least one of the following: road segment length node, road segment level node, and road segment width node.

[0051] According to embodiments of this disclosure, administrative region attribute nodes that have an inclusion relationship with administrative region nodes may include at least one of the following: district nodes and street nodes, but are not limited thereto, and embodiments of this disclosure do not limit this. Road attribute nodes that have an attribute relationship with road nodes may include road name nodes, nodes that have a connection relationship with road segment nodes may include intersection nodes, and intersection attribute nodes that have an attribute relationship with intersection nodes may include at least one of the following: longitude nodes and latitude nodes. Road segment nodes and travel nodes may have a location relationship, and road segment nodes and land parcel nodes may have an adjacent relationship.

[0052] According to embodiments of this disclosure, the carbon activity nodes of the activity information subgraph may include at least one of the following: waste treatment nodes, land use nodes, industrial process nodes, and energy activity nodes. The energy activity nodes may include at least one of the following: gas consumption nodes, water consumption nodes, fuel oil consumption nodes, and electricity consumption nodes. The carbon emission subgraph may include carbon emission amount nodes related to carbon activities. Carbon emission amount nodes and carbon activity nodes may have an inclusion relationship, and carbon emission amount nodes and time nodes may have a temporal relationship. The time subgraph may include multiple time nodes representing different carbon activity attributes, and the time attribute data related to carbon emission activities is stored in the data layer of the time nodes.

[0053] For the activity information subgraph, the plot nodes in the plot information subgraph are related to the carbon activity nodes in the activity information subgraph. Operations on plots generate related carbon activities, such as waste disposal, land use, and industrial processes. Simultaneously, these operations also generate related energy activities, which lead to corresponding carbon emissions, such as gas consumption, fuel oil consumption, electricity consumption, and water consumption. These energy consumptions directly or indirectly affect the city's carbon emissions.

[0054] For the time information subgraph, the relevant carbon emissions will be emitted during a certain period of time, and the amount of carbon emissions can be determined based on the specific time of the specific event. The time nodes in the time information subgraph can include time attribute nodes, such as year nodes, month nodes, day nodes, time nodes, and event nodes, etc., but are not limited to these, and the embodiments of this disclosure do not limit them.

[0055] Figure 2 A schematic diagram of a target knowledge graph according to an embodiment of the present disclosure is shown.

[0056] like Figure 2 As shown, the target knowledge graph of this embodiment includes a plot information subgraph 210, a functional information subgraph 220, an activity information subgraph 230, a time information subgraph 240, a movement information subgraph 250, a location information subgraph 260, and a carbon emission subgraph 270.

[0057] According to embodiments of this disclosure, the land parcel information sub-graph 210 may include land parcel nodes. Land parcel attribute nodes that have attribute relationships with the land parcel nodes can represent land parcel-related data in the urban area, such as area nodes, population nodes, plot ratio nodes, building height nodes, building density nodes, etc. The land parcel attribute nodes also have corresponding attribute values.

[0058] According to embodiments of this disclosure, the functional information sub-graph 220 may include land use type nodes, wherein the type attribute nodes related to the land use type nodes include industrial land nodes, warehousing land nodes, green space nodes, transportation land nodes, commercial service land nodes, public service land nodes, and residential land nodes. The type attribute nodes related to the land use type nodes also have corresponding attribute values. Furthermore, land use type nodes and plot nodes have attribute relationships.

[0059] According to embodiments of this disclosure, the activity information sub-graph 230 may include carbon activity nodes. Carbon activity attribute nodes that are inclusive of carbon activity nodes may include waste treatment nodes, land use nodes, industrial process nodes, and energy activity nodes. Energy consumption nodes that are inclusive of energy activity nodes may include gas consumption nodes, fuel oil consumption nodes, electricity consumption nodes, and water consumption nodes. Each energy consumption node and carbon activity attribute node has a corresponding attribute value and is related to both carbon activity nodes and land parcel nodes.

[0060] According to embodiments of this disclosure, the time information subgraph 240 may include time nodes. Time attribute nodes connected to time nodes can characterize the time data of relevant carbon emissions in the urban area, such as year nodes, month nodes, day nodes, time nodes, and event nodes.

[0061] According to embodiments of this disclosure, the mobility information subgraph 250 may include travel nodes, vehicle nodes that have a generation relationship with travel nodes, and vehicle attribute nodes that have an attribute relationship with vehicle nodes. Vehicle attribute nodes connected to vehicle nodes can represent relevant attribute information related to vehicles. Vehicle attribute nodes may include fuel type nodes, type nodes, and displacement nodes, and each vehicle attribute node has a corresponding attribute value. Travel nodes and land parcel nodes have an inclusion relationship, and travel nodes and road segment nodes have a location relationship.

[0062] According to embodiments of this disclosure, the location information sub-graph 260 may include road segment nodes representing the main body of location information. Nodes with a location relationship with the road segment nodes include at least one of the following: administrative region nodes and road nodes. Road segment attribute nodes with an attribute relationship with the road segment nodes include at least one of the following: road segment length nodes, road segment level nodes, and road segment width nodes. Administrative region attribute nodes with an inclusion relationship with the administrative region nodes may include at least one of the following: district nodes and street nodes. Road attribute nodes with an attribute relationship with the road nodes may include road name nodes. Nodes with a connection relationship with the road segment nodes may include intersection nodes. Intersection attribute nodes with an attribute relationship with the intersection nodes may include at least one of the following: longitude nodes and latitude nodes. Road segment nodes and travel nodes may have a location relationship, and road segment nodes and land parcel nodes may have an adjacent relationship.

[0063] According to embodiments of this disclosure, the carbon emission submap 270 includes carbon emission nodes, which have a temporal relationship with the time nodes in the time information submap 240. The carbon emission nodes in the carbon emission submap can characterize a series of carbon emissions generated by related activities in an urban area.

[0064] Figure 3 A flowchart of a carbon emission prediction method based on a target knowledge graph according to an embodiment of the present disclosure is shown.

[0065] like Figure 3 As shown, the carbon emission prediction method based on the target knowledge graph in this embodiment includes operations S310 to S330.

[0066] In operation S310, a target knowledge graph related to carbon activities within the city is identified.

[0067] In operation S320, the triples in the target knowledge graph are identified.

[0068] When operating S330, the graph structure corresponding to the triple is input into the prediction model, and the carbon emission prediction results for the city are output.

[0069] According to embodiments of this disclosure, the target knowledge graph may include a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph. The nodes of the time information subgraph can represent time attribute data related to urban carbon emission activities. The nodes of the carbon emission subject subgraph may include subject root nodes related to land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects, as well as subject attribute nodes that have attribute relationships with the subject root nodes. The nodes of the carbon emission subgraph can represent subject carbon emission information related to the carbon emission subject. The nodes of the activity information subgraph can represent carbon activity attribute data related to the carbon emission subject. The carbon emission subject is related to the time attribute data.

[0070] According to an embodiment of this disclosure, taking a certain urban area as an example, the relevant data for this urban area can be collected from authoritative statistical departments, energy companies, utility companies, research institutions, etc. The collected data mainly includes various types of data related to urban communities and vehicle travel trajectories within the city's jurisdiction, including but not limited to urban community resource consumption data and data reflecting the community's physical environment and socio-economic characteristics. This data can include building outline data, land use data, point of interest (POI) data, vehicle travel trajectory data, socio-economic data, climate data, etc., as shown in Table 3.

[0071] Table 3 Data List

[0072]

[0073] According to embodiments of this disclosure, the target knowledge graph focuses on representing urban communities and the mechanisms of carbon emission generation from travel. The construction of the target knowledge graph can include the construction of a schema layer and a data layer. Regarding the schema layer, this disclosure constructs the urban target knowledge graph schema layer from top to bottom based on expert experience. That is, it starts with high-level abstract concepts or domain knowledge, gradually refining and expanding the knowledge graph schema layer, and determining reasonable abstract levels and relational structures for entities based on domain experts. The constructed schema layer can include six categories: land parcel information subgraphs, functional information subgraphs, mobility information subgraphs, activity information subgraphs, time information subgraphs, and location information subgraphs. These six categories of subgraphs also include two structures: "entity-relationship-entity" and "entity-attribute-attribute value." Ontologies and their attribute features, as well as attribute relationships between entities, are defined in these six categories of subgraphs. Attribute relationships can include, but are not limited to, terms such as inclusion, generation, location, connection, time, and attribute. Embodiments of this disclosure do not specifically limit these relationships. Among them, attribute features of different types of ontologies were defined for six types of subgraphs, namely, land parcel information subgraph, functional information subgraph, movement information subgraph, activity information subgraph, time information subgraph, and location information subgraph, as shown in Table 4.

[0074] Table 4. Target Knowledge Graph Ontology and its Description

[0075]

[0076] According to embodiments of this disclosure, the data layer of the city target knowledge graph is constructed under the guidance of the schema layer. The data layer uses fact triples as units to store specific data information, where each entity in the triple has its own attributes and attribute values. Table 5 shows the structure of the target knowledge graph.

[0077] Table 5 Knowledge Graph Structure

[0078]

[0079] According to embodiments of this disclosure, a graph database reflecting the ontological attribute feature index system can be constructed to represent urban community and travel carbon emissions. This graph database includes graph data of communities and their attribute information, graph data of community building information, image data of community resident information, graph data of urban climate information, graph data of vehicle and travel trajectory attribute information, and graph data of urban carbon emissions. After collecting relevant urban area data, the data needs to be cleaned, noise and outliers removed, and standardized to ensure comparability of data from different sources. Then, through knowledge extraction and knowledge fusion, the target knowledge graph is constructed. This disclosure can perform knowledge extraction based on structured data, i.e., structured tabular data has clearly defined fields. Relationship extraction is achieved through field-entity mapping and inter-table connections, thereby obtaining relevant entity names, attribute information, and attribute relationships (i.e., edge relationships between nodes) from the database. Attribute relationship extraction is key to connecting various entities and forming rich knowledge. Secondly, knowledge fusion involves integrating different types of heterogeneous data after knowledge extraction within a unified schema framework, as shown in Table 6. The extracted data units are manually proofread to merge similar items, eliminate ambiguity, and unify and integrate them, including entity alignment, relationship alignment, and attribute alignment. A graph database can then be used to store the target knowledge graph.

[0080] Table 6 Comparison before and after knowledge integration

[0081]

[0082] According to embodiments of this disclosure, the target knowledge graph may include a land parcel information subgraph, a functional information subgraph, an activity information subgraph, and a time information subgraph. Land parcel nodes in the land parcel information subgraph and land use type nodes in the functional information subgraph have attribute relationships. Land parcel nodes in the land parcel information subgraph and carbon activity nodes in the activity information subgraph have generation relationships. Nodes in the time information subgraph represent time attribute data related to urban carbon emission activities. Nodes in the land parcel information subgraph represent urban land parcel attribute data. Nodes in the functional information subgraph represent urban land use type attribute data. Nodes in the activity information subgraph represent carbon activity attribute data related to carbon emission entities. Carbon emission entities are related to time attribute data.

[0083] According to embodiments of this disclosure, triples in a target knowledge graph can be determined. A triple may include a target carbon emission subject node, a target carbon activity node, and a target carbon emission information node with attribute relationships. The target carbon activity node can represent the carbon activity attribute data of the carbon emission subject, and the target carbon emission information node can represent the subject's carbon emission information.

[0084] According to embodiments of this disclosure, entity relation extraction can be used to extract nodes from a target knowledge graph. A triple is an ordered set of three elements; its structure is clear and concise, fully representing entities, relations, and the semantic associations between them. Examples include: community - located on - street; vehicle - generates - travel trajectory; community - feature parameter - floor area ratio value, etc.

[0085] According to embodiments of this disclosure, a graph structure corresponding to a triple is input into a prediction model, and the city carbon emission prediction result is output. Each triple can be converted into a graph structure, where nodes are entities and edges are relations. Node data includes entity sequence number, entity type, etc., and edge data includes entity relation type, etc.

[0086] According to embodiments of this disclosure, multiple relationships, such as the time and spatial form of carbon emission activities, are linked together in the form of triples. Embedding multiple triples into a knowledge graph forms a complete database of carbon emission measurement records. Within the constructed target knowledge graph, spatial, temporal, and subject-specific tracing can be achieved through a bottom-up approach. The triples obtained through tracing can effectively analyze the causes of carbon emissions and provide new ideas for carbon reduction and control.

[0087] According to embodiments of this disclosure, a target knowledge graph related to urban carbon activities can be determined. This target knowledge graph may include a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph. Nodes in the time information subgraph can represent time attribute data related to urban carbon emission activities. Nodes in the carbon emission subject subgraph may include root nodes related to land parcel information, functional information, location information, and mobility information, as well as subject attribute nodes with attribute relationships to the root nodes. Nodes in the carbon emission subgraph can represent subject carbon emission information related to the carbon emission subject. Nodes in the activity information subgraph can represent carbon activity attribute data related to the carbon emission subject. The carbon emission subject is related to the time attribute data. Triples in the target knowledge graph can be determined. Each triple may include a target carbon emission subject node, a target carbon activity node, and a target carbon emission information node with attribute relationships. The target carbon activity node can represent carbon... The carbon activity attribute data of the emitting entities, and the target carbon emission information nodes can represent the main carbon emission information of the emitting entities. The graph structure corresponding to the triples is input into the prediction model to output the carbon emission prediction results of the city. The entity-relationship-entity triples are defined at the micro level, which improves the accuracy of automated measurement and assessment of urban carbon emissions and provides decision support for the management of urban carbon activities. By combining the target knowledge graph with the carbon emission-related data of the urban area, the inherent correlation of multi-dimensional data is shown, which improves the accuracy of urban carbon emission prediction. Data is collected from the aspects of carbon emission time, space and causes, which improves the accuracy and professionalism of the matching between the target knowledge graph and the urban carbon emission prediction model. It solves the problems of high data cost, complex knowledge construction and difficulty in guaranteeing prediction effect in urban carbon control planning, improves the degree of matching between output content and real results, and effectively captures useful information in the target knowledge graph for carbon emission prediction.

[0088] According to embodiments of this disclosure, the time attribute data includes multiple data sets arranged chronologically, each corresponding one-to-one with a target knowledge graph. The prediction model includes a graph neural network and a time series network. Inputting the graph structure corresponding to the triples into the prediction model and outputting the city's carbon emission prediction result includes: inputting the graph structure corresponding to the time attribute data into the graph neural network and outputting a graph embedding feature matrix, wherein multiple graph embedding feature matrices have a temporal relationship, determined based on the time attribute data corresponding to each of the multiple graph structures; and inputting the multiple graph embedding feature matrices with a temporal relationship into the time series network to output the city's carbon emission prediction result.

[0089] According to embodiments of this disclosure, the time attribute data may include multiple data sets arranged chronologically. These multiple time attribute data sets may correspond one-to-one with multiple target knowledge graphs, and the prediction model may include graph neural networks and time series networks.

[0090] According to embodiments of this disclosure, in a static knowledge graph stored corresponding to a certain time attribute data, the definitions of nodes and relationships are fixed. The target knowledge graph adds a time series to the static knowledge graph, expanding the static triples into a set of quadruples that include a time dimension. Each of these contains a corresponding static knowledge graph. The target knowledge graph (UCEKG) is a multi-relation directed label graph, which can be represented as UCEKG= ,in, E represents the set of vertices (entities) in the knowledge graph; R represents the set of edges (relationships) in the knowledge graph. Indicates on date The set of facts that exist, that is, in There are occasional nodes , ∈E, ∈R exists .

[0091] According to embodiments of this disclosure, graph structures corresponding to time attribute data can be input into a graph neural network to output graph embedding feature matrices. Multiple graph embedding feature matrices have temporal relationships. These temporal relationships can be determined based on the time attribute data corresponding to each of the multiple graph structures. Multiple graph embedding feature matrices with temporal relationships can be input into a time series network to output predicted carbon emissions for a city.

[0092] According to embodiments of this disclosure, time series networks may include, but are not limited to, long short-term memory networks, and embodiments of this disclosure do not limit time series networks.

[0093] According to embodiments of this disclosure, by associating multiple time attribute data with multiple target knowledge graphs, the carbon emissions of a city can be viewed from a timeline perspective. A fact set is set within the knowledge graph; for example, if no transportation is used on a certain day, there is no fact set between the transportation node and the vehicle node. The next stage can be predicted based on the facts existing within the time attribute data, thereby improving prediction accuracy. Using the method provided by this disclosure, carbon emission predictions can be made more accurate. For example, if a school day is the first day of the semester, more transportation is used, resulting in relatively higher carbon emissions. If the carbon emissions of that day are used to predict the carbon emissions of the next day, the predicted result may differ significantly from the actual result. However, if multiple target knowledge graphs of a certain time attribute data are used to predict the carbon emissions of a certain day, the matching result between the output content and the actual result can be greatly improved, effectively capturing useful information from the target knowledge graph for carbon emission prediction.

[0094] According to embodiments of this disclosure, a graph neural network includes an input layer, a graph convolutional layer, and an output layer. A graph structure corresponding to temporal attribute data is input into the graph convolutional neural network, and a graph embedding feature matrix is ​​output. Specifically, the graph structure corresponding to the temporal attribute data can be processed based on the input layer to obtain a node feature matrix related to the nodes and an adjacency matrix representing the connection relationships between nodes in the graph structure. The node feature matrix and the adjacency matrix are then input into the graph convolutional layer for feature aggregation to obtain a spatial feature matrix. Finally, the spatial feature matrix is ​​processed based on the output layer to obtain the graph embedding feature matrix.

[0095] According to embodiments of this disclosure, a graph neural network may include an input layer, a graph convolutional layer, and an output layer. The graph structure corresponding to the time attribute data can be processed based on the input layer to obtain a node feature matrix associated with the nodes, and an adjacency matrix representing the connection relationships between nodes in the graph structure.

[0096] According to embodiments of this disclosure, the node feature matrix and adjacency matrix can be input into a graph convolutional layer for feature aggregation to obtain a spatial feature matrix. The spatial feature matrix can then be processed based on the output layer to obtain a graph embedding feature matrix.

[0097] According to embodiments of this disclosure, feature aggregation, i.e., multiplication of the node feature matrix and the adjacency matrix, can be used to aggregate the features of each node's neighboring nodes, thereby converting the input features into 16-dimensional features. Then, feature transformation is used, i.e., processing the aggregated spatial feature matrix through an activation function to generate a new node representation, i.e., a graph embedding feature matrix, converting the 16-dimensional features into a feature count equal to the number of categories in the dataset.

[0098] According to embodiments of this disclosure, the aggregated feature matrix is ​​processed through a linear layer and an activation function. The linear layer is typically a weight matrix used to map features from the input dimension to the hidden layer dimension. The activation function introduces non-linearity, enabling the network to learn more complex features. The output of the previous layer is used as the input to the next layer, and the steps of neighbor aggregation and feature transformation are repeated.

[0099] According to embodiments of this disclosure, the training process of a prediction model may include defining an optimizer and a loss function, clearing previous gradients, forward propagation, calculating the loss on the training set, backpropagation, and optimization. The evaluation process of the prediction model may include setting the model to evaluation mode, disabling gradient computation, forward propagation, and calculating performance metrics on the test set. The acquired dataset can be divided into training and test sets, then the optimizer can be used to tune the model parameters to minimize the loss. A context manager can be used to disable gradient computation during model evaluation to save memory and computational resources. Finally, a loss function is used at the output layer to measure the difference between the predicted and actual values.

[0100] According to embodiments of this disclosure, model evaluation and optimization can be performed using methods such as cross-validation to ensure the model's generalization ability. Based on the evaluation results, the model is optimized and adjusted. The mean squared error (MSE) loss function is used to calculate the average of the squared errors between the predicted and actual values ​​for each sample, thereby measuring the model's predictive performance and improving prediction accuracy and stability. The MSE loss function is shown in formula (1).

[0101] （1）

[0102] According to embodiments of this disclosure, the constructed target knowledge graph for urban carbon emission measurement and control provides a more detailed representation of the intrinsic relationships between multidimensional data. It defines entity-relationship-entity triples at the micro-level, improving the accuracy of automated carbon emission measurement and assessment, and providing decision support for urban carbon activity management. The constructed structured triples achieve a unified expression and visual representation of multimodal data knowledge, aggregating data from the perspectives of carbon emission time, space, and causes. This improves the accuracy and professionalism of the matching between the target knowledge graph and the carbon emission prediction model, solving problems such as high data costs, complex knowledge construction, and difficulty in guaranteeing prediction results in carbon control planning. The provided carbon emission prediction method is easily expandable and optimized, improving the matching degree between output content and actual results, and effectively capturing useful information from the target knowledge graph for carbon emission prediction.

[0103] Based on the aforementioned knowledge graph-based urban carbon measurement method, this disclosure also provides a knowledge graph-based urban carbon measurement device. The following will combine... Figure 4 The device is described in detail.

[0104] Figure 4 A structural block diagram of a knowledge graph-based urban carbon metering device according to an embodiment of the present disclosure is shown.

[0105] like Figure 4 As shown, the knowledge graph-based urban carbon metering device 400 of this embodiment includes a construction module 410, an acquisition module 420, a main carbon emission information acquisition module 430, a target knowledge graph acquisition module 440, and an urban carbon metering result acquisition module 450.

[0106] Module 410 is used to construct an initial knowledge graph of carbon activities within an urban area. The initial knowledge graph includes an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph. The initial carbon emission subject subgraph includes subject root nodes representing four types of carbon emission subjects, as well as initial subject attribute nodes that have attribute relationships with the subject root nodes. The four types of carbon emission subjects include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects. The initial activity information subgraph includes carbon activity nodes related to carbon emission subjects. The time information subgraph includes time nodes representing the time or period when carbon activities occur. The initial carbon emission subgraph includes carbon emission amount nodes related to carbon activities. The subject root nodes of each subgraph in the initial knowledge graph have attribute relationships, which include at least one of the following: contain, generate, located, connected, time, and attribute.

[0107] The acquisition module 420 is used to acquire carbon activity attribute data and time attribute data of the initial subject attribute nodes based on the initial knowledge graph. The time attribute data corresponds to the time nodes.

[0108] The main carbon emission information acquisition module 430 is used to perform carbon emission calculation tasks based on the carbon activity attribute data and time attribute data of the initial main attribute node, as well as the carbon emission calculation parameters related to the carbon activity node, to obtain the main carbon emission information related to the initial main attribute node.

[0109] The target knowledge graph is obtained by module 440, which is used to update the initial carbon emission information nodes that have attribute relationships with carbon activity nodes based on the subject's carbon emission information, thereby obtaining the target knowledge graph.

[0110] The Urban Carbon Measurement Results Module 450 is used to obtain urban carbon measurement results based on the target knowledge graph.

[0111] According to embodiments of this disclosure, for urban areas, an initial knowledge graph of carbon activities within the city can be constructed. The initial knowledge graph may include an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph. The initial carbon emission subject subgraph may include subject root nodes representing four types of carbon emission subjects, and initial subject attribute nodes that have attribute relationships with the subject root nodes. The four types of carbon emission subjects may include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects. The initial activity information subgraph may include carbon activity nodes related to carbon emission subjects. The time information subgraph may include time nodes representing the time or period of carbon activity occurrence. The initial carbon emission subgraph may include carbon emission amount nodes related to carbon activities. Each subject root node in the initial knowledge graph subgraph has attribute relationships, which may include at least one of the following: containing, generating, located, connected, time, and attribute. Based on the initial knowledge graph, initial subject attributes can be obtained. The system generates carbon activity attribute data and time attribute data for initial subject attribute nodes, with the time attribute data corresponding to the time nodes. Based on the initial subject attribute node's carbon activity attribute data and time attribute data, as well as the carbon emission calculation parameters related to the carbon activity node, a carbon emission calculation task is performed to obtain the subject carbon emission information related to the initial subject attribute node. The initial carbon emission information nodes with attribute relationships to the carbon activity node are updated based on the subject carbon emission information, resulting in a target knowledge graph. Based on the target knowledge graph, the city's carbon measurement results are obtained. The constructed target knowledge graph efficiently integrates temporal, spatial, and subject-related process information related to carbon activities within the city. Through the interconnection of relationships, a multi-dimensional network knowledge structure is formed, representing and storing various elements related to carbon activities in the form of a directed graph. This improves the accuracy and efficiency of automated carbon emission measurement, realizes a comprehensive expression of the mapping relationship between carbon emissions and planning, facilitates accurate traceability of carbon measurement results, and provides efficient support for carbon reduction and control.

[0112] Based on the aforementioned carbon emission prediction method based on target knowledge graphs, this disclosure also provides a carbon emission prediction device based on target knowledge graphs. The following will combine... Figure 5 The device is described in detail.

[0113] Figure 5 A structural block diagram of a carbon emission prediction device based on a target knowledge graph according to an embodiment of the present disclosure is shown.

[0114] like Figure 5 As shown, the carbon emission prediction device 500 based on target knowledge graph in this embodiment includes a processing module 510, a triplet determination module 520, and a prediction result output module 530.

[0115] Processing module 510 is used to determine the target knowledge graph related to carbon activities within the city. The target knowledge graph includes a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph. The nodes of the time information subgraph represent time attribute data related to urban carbon emission activities. The nodes of the carbon emission subject subgraph include subject root nodes related to land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects, as well as subject attribute nodes with attribute relationships to subject root nodes. The nodes of the carbon emission subgraph represent subject carbon emission information related to the carbon emission subject. The nodes of the activity information subgraph represent carbon activity attribute data related to the carbon emission subject. The carbon emission subject is related to the time attribute data.

[0116] The triplet determination module 520 is used to determine triples in the target knowledge graph. The triples include target carbon emission subject nodes, target carbon activity nodes, and target carbon emission information nodes with attribute relationships. The target carbon activity nodes represent the carbon activity attribute data of the carbon emission subject, and the target carbon emission information nodes represent the subject carbon emission information of the carbon emission subject.

[0117] The prediction result output module 530 is used to input the graph structure corresponding to the triple into the prediction model and output the carbon emission prediction results of the city.

[0118] According to embodiments of this disclosure, a target knowledge graph related to urban carbon activities can be determined. This target knowledge graph may include a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph. Nodes in the time information subgraph can represent time attribute data related to urban carbon emission activities. Nodes in the carbon emission subject subgraph may include root nodes related to land parcel information, functional information, location information, and mobility information, as well as subject attribute nodes with attribute relationships to the root nodes. Nodes in the carbon emission subgraph can represent subject carbon emission information related to the carbon emission subject. Nodes in the activity information subgraph can represent carbon activity attribute data related to the carbon emission subject. The carbon emission subject is related to the time attribute data. Triples in the target knowledge graph can be determined. Each triple may include a target carbon emission subject node, a target carbon activity node, and a target carbon emission information node with attribute relationships. The target carbon activity node can represent carbon... The carbon activity attribute data of the emitting entities, and the target carbon emission information nodes can represent the main carbon emission information of the emitting entities. The graph structure corresponding to the triples is input into the prediction model to output the carbon emission prediction results of the city. The entity-relationship-entity triples are defined at the micro level, which improves the accuracy of automated measurement and assessment of urban carbon emissions and provides decision support for the management of urban carbon activities. By combining the target knowledge graph with the carbon emission-related data of the urban area, the inherent correlation of multi-dimensional data is shown, which improves the accuracy of urban carbon emission prediction. Data is collected from the aspects of carbon emission time, space and causes, which improves the accuracy and professionalism of the matching between the target knowledge graph and the urban carbon emission prediction model. It solves the problems of high data cost, complex knowledge construction and difficulty in guaranteeing prediction effect in urban carbon control planning, improves the degree of matching between output content and real results, and effectively captures useful information in the target knowledge graph for carbon emission prediction.

[0119] According to embodiments of this disclosure, any plurality of modules, including modules, submodules, units, and subunits, can be combined into one module for implementation, or any one of these modules can be split into multiple modules. Alternatively, at least a portion of the functionality of one or more of these modules can be combined with at least a portion of the functionality of other modules and implemented in one module. According to embodiments of this disclosure, at least one of modules, submodules, units, and subunits can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these. Alternatively, at least one of modules, submodules, units, and subunits can be at least partially implemented as a computer program module, which, when run, can perform corresponding functions.

[0120] Figure 6 A block diagram of an electronic device suitable for implementing a knowledge graph-based urban carbon measurement and prediction method is shown according to embodiments of the present disclosure.

[0121] like Figure 6 As shown, an electronic device 600 according to an embodiment of this disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage portion 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this disclosure.

[0122] RAM 603 stores various programs and data required for the operation of electronic device 600. Processor 601, ROM 602, and RAM 603 are interconnected via bus 604. Processor 601 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 602 and / or RAM 603. It should be noted that the programs may also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0123] According to embodiments of this disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to a bus 604. The electronic device 600 may also include one or more of the following components connected to the I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 610 as needed so that computer programs read from it can be installed into the storage section 608 as needed.

[0124] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0125] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 602 and / or RAM 603 and / or one or more memories other than ROM 602 and RAM 603 described above.

[0126] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code enables the computer system to implement the knowledge graph-based urban carbon measurement and prediction methods provided in embodiments of this disclosure.

[0127] When the computer program is executed by the processor 601, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0128] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 609, and / or installed from the removable medium 611. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0129] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0130] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on a user's computing device, partially on a user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0131] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0132] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0133] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A knowledge graph-based urban carbon metering method, characterized in that, Comprise: For urban areas, an initial knowledge graph of carbon activity content within the city is constructed, wherein the initial knowledge graph includes an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph, the initial carbon emission subject subgraph includes a subject root node representing four types of carbon emission subjects, and an initial subject attribute node having an edge relationship with the subject root node, the four types of carbon emission subjects include plot information subjects, function information subjects, location information subjects, and movement information subjects, the plot information subjects represent plot information related to land use types, carbon activities, road segments, and trips, the function information subjects represent land use types related to plots, the location information subjects represent road segment information related to plots and trips, and the movement information subjects represent movement information related to plots, road segments, and carbon activities, the initial activity information subgraph includes carbon activity nodes related to the carbon emission subjects, the time information subgraph includes time nodes representing the time or period of carbon activity, and the initial carbon emission subgraph includes carbon emission amount nodes related to carbon activities, the subject root nodes of the initial knowledge graph subgraphs have an edge relationship between each other, and the edge relationship includes at least one of the following: contains, generates, is located, is connected, time, and attribute; According to the initial knowledge graph, carbon activity attribute data and time attribute data of the initial subject attribute node are obtained, and the time attribute data corresponds to the time node; According to the carbon activity attribute data and the time attribute data of the initial subject attribute node, and the carbon emission calculation parameters related to the carbon activity nodes, a carbon emission calculation task is performed to obtain subject carbon emission information related to the initial subject attribute node; and According to the subject carbon emission information, the initial carbon emission information node having an edge relationship with the carbon activity node is updated to obtain a target knowledge graph; According to the target knowledge graph, a city carbon measurement result is obtained, The carbon emission subject subgraph of the target knowledge graph includes a plot information subgraph, a function information subgraph, a location information subgraph, and a movement information subgraph; the subject root node of the plot information subgraph is a plot node representing the plot information subject, and the subject attribute nodes of the plot information subgraph include plot attribute nodes having attribute relationships with the plot node, and the plot attribute nodes include at least one of the following: area nodes, number of people nodes, volume rate nodes, building height nodes, and building density nodes; Wherein, the plot node has a generation relationship with the carbon activity nodes in the activity information subgraph, the movement information subgraph includes a trip node representing the movement information subject, and the subject attribute nodes of the movement information subgraph include vehicle nodes having a generation relationship with the trip node, and vehicle attribute nodes having an attribute relationship with the vehicle nodes include at least one of the following: fuel type nodes, type nodes, and displacement nodes. The travel node and the land parcel node have an inclusion relationship; the travel node and the time node in the time subgraph have a time relationship; the vehicle node and the energy activity node in the activity information subgraph have a generation relationship; the functional information subgraph includes land use type nodes that represent the main body of functional information; and the type attribute nodes that have an edge relationship with the land use type nodes include at least one of the following: industrial land use node, warehousing land use node, green space node, transportation land use node, commercial service land use node, public service land use node, and residential land use node. The land use type node and the plot node have attribute relationships. The location information subgraph includes road segment nodes that represent the main body of location information. Nodes that have a location relationship with the road segment node include at least one of the following: administrative region nodes and road nodes. Road segment attribute nodes that have an attribute relationship with the road segment node include at least one of the following: road segment length nodes, road segment level nodes, and road segment width nodes. Administrative region attribute nodes that have an inclusion relationship with the administrative region node include at least one of the following: district nodes and street nodes. Road attribute nodes that have an attribute relationship with the road node include road name nodes. Nodes that have a connection relationship with the road segment node include intersection nodes. Intersection attribute nodes that have an attribute relationship with the intersection nodes include at least one of the following: longitude nodes and latitude nodes. The road segment node and the travel node are located in the same position. The road segment node and the land parcel node are adjacent to each other. The carbon activity nodes of the activity information subgraph include at least one of the following: waste treatment node, land use node, industrial process node and energy activity node. The energy activity node includes at least one of the following: gas consumption node, water consumption node, fuel oil consumption node and electricity consumption node. The carbon emission submap includes carbon emission nodes related to carbon activities, the carbon emission nodes are inclusive of the carbon activity nodes, and the carbon emission nodes are temporally related to the time nodes; The time subgraph includes multiple time nodes that characterize different carbon activity attributes, and the time attribute data related to the carbon emission activities is stored in the data layer of the time nodes.

2. A carbon emission prediction method based on a target knowledge graph, characterized in that, The method includes: The method according to claim 1 determines a target knowledge graph related to carbon activities within a city, the target knowledge graph including a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph, wherein the nodes of the time information subgraph represent time attribute data related to urban carbon emission activities, the nodes of the carbon emission subject subgraph include subject root nodes related to land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects, and subject attribute nodes with edge relationships to the subject root nodes, the nodes of the carbon emission subgraph represent subject carbon emission information related to the carbon emission subjects, the nodes of the activity information subgraph represent carbon activity attribute data related to the carbon emission subjects, and the carbon emission subjects are related to the time attribute data; The triples in the target knowledge graph are determined. The triples include a target carbon emission subject node, a target carbon activity node, and a target carbon emission information node with edge relationships. The target carbon activity node represents the carbon activity attribute data of the carbon emission subject, and the target carbon emission information node represents the subject carbon emission information of the carbon emission subject. The graph structure corresponding to the triple is input into the prediction model, and the carbon emission prediction results of the city are output.

3. The method of claim 2, wherein, The time attribute data is multiple, and the multiple time attribute data are arranged in chronological order. The multiple time attribute data correspond one-to-one with the multiple target knowledge graphs. The prediction model includes graph neural networks and time series networks. The step of inputting the graph structure corresponding to the triplet into the prediction model and outputting the carbon emission prediction results for the city includes: The graph structure corresponding to the time attribute data is input into the graph neural network, and a graph embedding feature matrix is ​​output. The graph embedding feature matrices have a temporal relationship, which is determined based on the time attribute data corresponding to each of the graph structures. The graph embedding feature matrices with temporal relationships are input into the time series network to output the carbon emission prediction results for the city.

4. A knowledge graph-based urban carbon metering device, characterized in that, include: A construction module is used to build an initial knowledge graph of carbon activities within an urban area. This initial knowledge graph includes an initial carbon emission subject subgraph, an initial activity information subgraph, an initial carbon emission subgraph, and a time information subgraph. The initial carbon emission subject subgraph includes a root node representing four types of carbon emission subjects, and initial subject attribute nodes with edge relationships to the root node. The four types of carbon emission subjects include land parcel information subjects, functional information subjects, location information subjects, and mobility information subjects. The land parcel information subjects represent land parcel information related to land use type, carbon activities, road segments, and travel. The information subject represents the land use type related to the land parcel; the location information subject represents road segment information related to the land parcel and travel; the mobility information subject represents mobility information related to the land parcel, road segment, and carbon activity; the initial activity information subgraph includes carbon activity nodes related to the carbon emission subject; the time information subgraph includes time nodes representing the moment or period of carbon activity occurrence; the initial carbon emission subgraph includes carbon emission amount nodes related to carbon activity; and the root nodes of each subject in the initial knowledge graph subgraph have edge relationships, which include at least one of the following: contain, generate, located, connected, time, and attribute. The acquisition module is used to acquire carbon activity attribute data and time attribute data of the initial subject attribute nodes based on the initial knowledge graph, wherein the time attribute data corresponds to the time nodes; The main carbon emission information acquisition module is used to perform a carbon emission calculation task based on the carbon activity attribute data and time attribute data of the initial main attribute node, and the carbon emission calculation parameters related to the carbon activity node, to obtain the main carbon emission information related to the initial main attribute node; and The target knowledge graph acquisition module is used to update the initial carbon emission information nodes that have edge relationships with the carbon activity nodes based on the main carbon emission information, thereby obtaining the target knowledge graph; The urban carbon measurement result acquisition module is used to obtain urban carbon measurement results based on the target knowledge graph, and the knowledge graph-based urban carbon measurement device is used to execute the method according to claim 1. 5.A device for predicting carbon emission based on a target knowledge graph, characterized in that, The device includes: A processing module is configured to determine a target knowledge graph related to urban carbon activities according to the method described in claim 1. The target knowledge graph includes a carbon emission subject subgraph, an activity information subgraph, a carbon emission subgraph, and a time information subgraph. The nodes of the time information subgraph represent time attribute data related to urban carbon emission activities. The nodes of the carbon emission subject subgraph include a subject root node related to land parcel information, functional information, location information, and mobility information, as well as subject attribute nodes with edge relationships to the subject root node. The nodes of the carbon emission subgraph represent subject carbon emission information related to the carbon emission subject. The nodes of the activity information subgraph represent carbon activity attribute data related to the carbon emission subject. The carbon emission subject is related to the time attribute data. The triplet determination module is used to determine triples in the target knowledge graph. The triples include a target carbon emission subject node, a target carbon activity node, and a target carbon emission information node with edge relationships. The target carbon activity node represents the carbon activity attribute data of the carbon emission subject, and the target carbon emission information node represents the subject carbon emission information of the carbon emission subject. The prediction result output module is used to input the graph structure corresponding to the triple into the prediction model and output the carbon emission prediction results of the city.

Images