BIM-based component-level knowledge graph construction method and system

Abstract

The invention provides a BIM-based component-level knowledge graph construction method and system, and the method comprises the steps: loading a preset export plug-in, and exporting the component data of each component in a building information model in the form of a semi-structured data format file; analyzing the exported file by adopting an analysis tool to extract component data, and storing the component data into a component data structure; for hierarchical path parameters in each component data structure, generating a knowledge graph entity and mapping a front node name and component data into entity attributes; searching a post node of each node in a recursive traversal mode to obtain a hierarchical relationship between the components; and based on the hierarchical relationship, the knowledge graph entity and the corresponding entity attribute, generating a knowledge graph corresponding to the building information model. According to the method, the problems of low construction efficiency of the domain knowledge graph of the railway building information model and insufficient semantic relationship mining can be solved; and the construction efficiency and the information accuracy of the knowledge graph are improved.

Description

Technical Field

[0001] This invention relates to the field of railway building information modeling technology, and in particular to a method and system for constructing a component-level knowledge graph based on BIM. Background Technology

[0002] In fields such as railways, engineering, and construction management, Building Information Modeling (BIM) technology, through the integration and visualization of geometric and non-geometric information, can digitally represent the physical and functional characteristics of project facilities, covering information throughout the entire lifecycle from design, construction, and operation and maintenance. Currently, most BIM model-related technologies focus on model visualization, construction progress management, and safety risk management, failing to fully explore its deeper semantic relationships and knowledge value. Therefore, knowledge graphs, as a tool for structured representation of knowledge and its relationships, are widely used. By constructing component-level knowledge graphs based on BIM models, the geometric and non-geometric information of the BIM model can be effectively integrated, while establishing relationships between components. This provides support for further mining, analysis, and utilization of semantic information and knowledge within the BIM model, thereby achieving more efficient and intelligent management and decision-making.

[0003] Currently, traditional knowledge graph construction methods include: manually collecting data from data sources, obtaining structured data from relational databases, establishing a mapping relationship between structured data and ontology models, extracting entities and entity relationships that meet the requirements of ontology models from structured data, obtaining knowledge graph triples based on entities and entity relationships, and then constructing the knowledge graph; or, obtaining unstructured text, using a trained entity relationship acquisition model and a trained relationship classification model to obtain entity relationship results from the unstructured text based on the knowledge graph, obtaining knowledge graph triples based on the entity relationship results, and updating the knowledge graph.

[0004] However, while the above methods can integrate structured and unstructured data to construct knowledge graphs, knowledge acquisition still heavily relies on the performance of pre-defined ontology models and machine learning models. Their adaptability to complex and ever-changing business scenarios needs further improvement, and the accuracy and completeness of knowledge extraction are also limited by the models' limitations. For example, in the field of railway building information modeling, large and complex engineering projects often involve massive amounts of components and their complex attributes and relationships. Manually constructing component-level knowledge graphs requires significant manpower, resources, and time, and is prone to errors and omissions. Relying on simple data extraction tools to extract information from building information models makes it difficult to accurately identify and extract rule-based data that meets the requirements for knowledge graph construction. This results in low efficiency in knowledge graph construction and insufficient semantic relationship mining in the field of railway building information modeling. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a method and system for constructing a component-level knowledge graph based on BIM, to eliminate or improve one or more defects existing in the prior art. It can solve the problems of low efficiency in knowledge graph construction and insufficient semantic relationship mining in the field of railway building information modeling.

[0006] One aspect of the present invention provides a method for constructing a component-level knowledge graph based on BIM, the method comprising the following steps: Load a preset export plugin into the building information modeling software to export the component data of each component in the building information model in the form of a semi-structured data format file; the component data includes the identity description parameters and hierarchical path parameters of each component in the building information model; The semi-structured data format file is parsed using a preset parsing tool to extract component data and store it in a predefined component data structure; For the hierarchical path parameters in each component data structure, the node name of the last node is extracted as the entity name to generate a knowledge graph entity, the preceding node of the last node is extracted as the preceding node name, and the corresponding component data and preceding node name are mapped to the entity attributes of the generated knowledge graph entity. Starting from the top-level node in the hierarchical path parameters, the successor node of each node is found through recursive traversal to obtain the hierarchical relationship between components; the hierarchical relationship is used to indicate the predecessor and successor nodes of each component. Based on hierarchical relationships, knowledge graph entities, and corresponding entity attributes, a knowledge graph corresponding to the building information model is generated.

[0007] In some embodiments of the present invention, after extracting the node name of the last node as the entity name to generate the knowledge graph entity for the hierarchical path parameters in each component data structure, the method further includes: Compare the entity name of the current knowledge graph entity with the entity names of the already generated knowledge graph entities to determine if there is a name duplication. In the case of duplicate names, check whether the preceding node name of the entity with the duplicate name is consistent with the preceding node name of the current knowledge graph entity; If the names of the preceding nodes are the same, delete the current knowledge graph entity.

[0008] In some embodiments of the present invention, the top-level node in the hierarchical path parameters is used as the starting node, and the successor node of each node is found by recursive traversal to obtain the hierarchical relationship between components, including: For each found node, determine whether the name of the preceding node in the corresponding knowledge graph entity attribute is consistent with the name of the preceding node in the hierarchical path parameter; If the names are identical, retain the current node.

[0009] In some embodiments of the present invention, the component identity parameter includes a unique component identifier; the unique component identifier is used to associate entities in the knowledge graph with the corresponding component geometric models. After generating the knowledge graph corresponding to the building information model based on hierarchical relationships, knowledge graph entities, and corresponding entity attributes, the process also includes: In response to the received selection operation applied to entities in the knowledge graph, the unique identifier of the component corresponding to the selected entity is determined as the model location identifier; Based on the pre-defined mapping relationship between the unique identifier of a component and its geometric model, the location model locates the position of the geometric model of the component corresponding to the identifier in the building information model.

[0010] In some embodiments of the present invention, after generating the knowledge graph corresponding to the building information model based on hierarchical relationships, knowledge graph entities, and corresponding entity attributes, the method further includes: In response to the received selection operation applied to the geometric model of a component in the building information model, the unique identifier of the component corresponding to the selected model is determined as the entity positioning identifier. Based on a preset mapping relationship, the location entity is identified as the position of the knowledge graph entity corresponding to the entity's location identifier in the knowledge graph.

[0011] In some embodiments of the present invention, before loading a preset export plugin into the building information modeling software and exporting the component data of each component in the building information model in the form of a semi-structured data format file, the method further includes: Integrity checks are performed on the building information model, and the results are obtained. If the detection results indicate that there are missing component geometric models in the building information model, the building information model is repaired based on the component data corresponding to the missing models.

[0012] In some embodiments of the present invention, the semi-structured data format file includes an Extensible Markup Language (XML) file; the parsing tool includes an XML parser.

[0013] In some embodiments of the present invention, the building information modeling software includes at least two types of software; different building information modeling software has different preset export plugins.

[0014] Another aspect of the present invention provides a BIM-based component-level knowledge graph construction system, including a processor, a memory, and a computer program / instructions stored in the memory. The processor is used to execute the computer program / instructions. When the computer program / instructions are executed, the system implements the steps of the BIM-based component-level knowledge graph construction method described above.

[0015] Another aspect of the present invention provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps of a method for constructing a BIM-based component-level knowledge graph as described above.

[0016] The method and system for constructing a component-level knowledge graph based on BIM of the present invention can solve the problems of low efficiency in knowledge graph construction and insufficient semantic relationship mining in the field of railway building information modeling. Through a preset export plugin, a semi-structured data format file containing rule definitions of component data is exported from the building information model, which can accurately identify and extract rule-based data that meets the requirements for knowledge graph construction. Then, with the help of corresponding file parsing technology, components, component data, and relationships between components are quickly extracted in batches without relying on complex and cumbersome procedures or manual data processing and conversion. This significantly reduces manual intervention and data processing time, thereby significantly improving the efficiency of knowledge graph construction and effectively meeting the timeliness requirements of knowledge graph updates for component updates during railway engineering construction. At the same time, avoiding manual construction of the knowledge graph also reduces errors or omissions in knowledge graph information caused by human error, further improving the accuracy of the knowledge graph.

[0017] Furthermore, by using the unique component code corresponding to each component in the Building Information Model (BIM), a bidirectional positioning mechanism is achieved between component entities in the knowledge graph and components in the BIM. When querying a component entity in the knowledge graph, the bidirectional positioning mechanism can quickly locate the corresponding component and related information in the original BIM, allowing users to intuitively view the specific situation of the component in the BIM. When viewing a component in the BIM, the bidirectional positioning mechanism can quickly link it to the component entities and relationships associated with that component in the knowledge graph, breaking down information silos between the knowledge graph and the BIM, improving the efficiency of information interconnection and collaborative application, and assisting users in making rapid decisions. This comprehensively enhances the intelligent level of railway engineering lifecycle management.

[0018] Furthermore, for different Building Information Modeling (BIM) software and BIM models with different data formats, corresponding preset export plugins are used to export XML files containing component data using a universal XML standard. These XML files are then parsed using an XML parsing tool, effectively resolving compatibility issues in cross-platform and cross-stage integration of multi-source heterogeneous BIM data. Through a standardized data export and parsing process, BIM models from different modeling software are processed uniformly, enabling lossless extraction and structured conversion of key data such as component hierarchical paths and component identity parameters. Moreover, the universal XML standard defines strict tag structures and data type constraints, ensuring the standardization and verifiability of exported files, improving the accuracy and automation of data parsing, reducing dependence on specific BIM platforms, and providing a high-quality, standardized input source for subsequent knowledge graph construction, thus enhancing the overall scalability and engineering applicability of the solution.

[0019] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the description and drawings.

[0020] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0021] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. In the drawings: Figure 1 This is a flowchart illustrating a method for constructing a component-level knowledge graph based on BIM, as provided in an embodiment of the present invention.

[0022] Figure 2 Another flowchart of a method for constructing a component-level knowledge graph based on BIM, as provided in an embodiment of the present invention.

[0023] Figure 3 This is an example diagram of an XML file provided in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0025] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0026] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0027] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0028] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0029] The following section provides a detailed description of a BIM-based component-level knowledge graph construction method provided in this application.

[0030] like Figure 1 As shown, an embodiment of this application provides a method for constructing a component-level knowledge graph based on BIM. The implementation of this method can rely on a computer program, which can run on computer devices such as smartphones, tablets, and personal computers, or run on a server. This embodiment does not limit the operating entity of the method. The method includes at least steps S101 to S105: Step S101: Load the preset export plugin in the building information modeling software and export the component data of each component in the building information model in the form of a semi-structured data format file.

[0031] Building Information Modeling (BIM) software refers to digital tools used to create, manage, and analyze Building Information Models (BIM). Different BIM software programs may use the same or different data formats for their BIM models. Examples include Industry Foundation Classes (IFC) for cross-platform data sharing in the architecture, engineering, and construction industries; the Revit file format (RVT) for architectural design, structural analysis, and construction management in Revit software; and the Design file format (DGN) suitable for infrastructure and civil engineering in BIM software such as MicroStation.

[0032] In this embodiment, the Building Information Model (BIM) can cover various engineering scenarios. For example, the BIM can be a BIM model for railway engineering projects, including BIM models for tunnel engineering, bridge engineering, or roadbed engineering. This embodiment does not limit the type of BIM software or the source of the BIM model.

[0033] Preset export plugins refer to code plugins developed and integrated based on the specific format of the building information model (such as the .rvt format of Revit, the .dgn format of MicroStation, etc.) and calling the corresponding software development kit (SDK, such as the DGN software development kit) or application programming interface (API, such as the Autodesk application programming interface).

[0034] Specifically, the preset export plugin is used to separate the geometric data and non-geometric data corresponding to each component in the Building Information Model (BIM), and to export the component data in a semi-structured data format file according to preset rules. Geometric data refers to the component geometric model corresponding to each component in the BIM; non-geometric data includes the component data corresponding to each component, including but not limited to component identity parameters, hierarchical path parameters, etc.

[0035] Meanwhile, in the Building Information Modeling (BIM), each component's geometric model is pre-assigned a unique identifier. This unique identifier is used to uniquely identify the component, and different components have different unique identifiers. The unique identifier can be numbers, letters, or a combination of numbers and letters; this embodiment does not limit the implementation method of the unique identifier.

[0036] In this embodiment, the unique component identifier is used to dynamically associate the geometric and non-geometric data corresponding to the same component. For example, in building information modeling software, when a click operation is received on a component in the building information model, the corresponding component data is found and displayed through the pre-associated unique component identifier.

[0037] Preset rules refer to pre-defined data definition rules. Component data includes the component identity parameters and hierarchical path parameters of each component in the building information model. Correspondingly, data definition rules include data definition rules corresponding to component identity parameters and data definition rules corresponding to hierarchical path parameters.

[0038] The component identity parameters include component name (Name), component unique identifier (UUID), structure code (EBS), classification code (IFD), and other identity description parameters. The hierarchical path parameters are used to indicate the hierarchical path of the component in the building information model. For example, taking the building information model as a bridge project model and the component as pile 1 under pier 2 of the bridge, the corresponding hierarchical path parameters can be represented as "bridge-substructure-pier 2-pile 1".

[0039] In some embodiments of the present invention, the semi-structured data format file includes an Extensible Markup Language (XML) file. In this case, the preset rules are based on the "Guidelines for the Compilation of Railway Engineering Information Exchange Templates" developed by the Railway BIM Alliance, which predefines standardized rules for XML tag structures, element attributes, and hierarchical relationships.

[0040] For example, taking the Building Information Model (BIM) of a bridge project as an example, the BIM includes the geometric model of the components corresponding to the "piers"; (Refer to...) Figure 2 The pre-defined export plugin separates the building information model module corresponding to the bridge into geometric data and non-geometric data. The geometric and non-geometric data are dynamically linked through unique component identifiers. The non-geometric data is exported as an XML file; the XML file corresponding to the "pier" is as follows: Figure 3 As shown, all component parameters of the "pier body" component are contained in the label.<CRBIM Bridges And Culverts Component Properties> Below, the component name "pier body" is in the "Name" tag under the "ID" tag; the "unique identifier" corresponding to the component's geometric model is in the "UUID" tag under the "ID" tag; the structural code is in the "EBS" tag under the "ID" tag; and the classification code is in the "IFD53" tag under the "ID" tag. The hierarchical path serves as the location information in... Under the field <nodepath>In the field.

[0041] In other embodiments of the present invention, the semi-structured data format file also includes a JSON file. In this case, the preset rules are based on the "Guidelines for the Compilation of Railway Engineering Information Exchange Templates" formulated by the Railway BIM Alliance, which pre-defines standardized rules for JSON tag structures, element attributes, and hierarchical relationships. This embodiment does not limit the file type of the semi-structured data format file or the implementation method of the preset rules.

[0042] In addition, in actual implementation, component data may also include other parameters, such as Project Information (PI) parameters, Location Information (LI) parameters, Organization Role (RL) parameters, Asset Information (AM) parameters, Manufacturing Information (MF) parameters, Asset & Maintenance Information (AM) parameters, Geometric Information (GI) parameters, Technical Information (TI) parameters, etc. This embodiment does not limit the implementation method of component parameters.

[0043] The location information includes, but is not limited to, starting mileage, ending mileage, center mileage, elevation, left and right sides, left and right distance, and node path.

[0044] Organizational roles include, but are not limited to, ConstructionUnitName and DesignUnitName.

[0045] Project information includes, but is not limited to, project name (EntryName), line name (NameOfLineBelonging), and design unit name (DesignUnitName).

[0046] The generated information includes, but is not limited to, price, term of validity, manufacturer, supplier, actual size, product certification, manufacturing standards, and installation method.

[0047] Asset information includes, but is not limited to, cost, tax, procurement information, quantity, and ownership.

[0048] Maintenance information includes, but is not limited to, supplier, repair time, maintenance content, and status of spare parts.

[0049] Geometric information includes, but is not limited to, shape description, top shape, top surface longitudinal dimension, top surface horizontal dimension, height, slope rate, and volume.

[0050] Technical information includes, but is not limited to, type name, concrete strength grade, reinforcement strength grade, environmental category, construction technology, pier and abutment number, and pier or column body or abutment body number.

[0051] In addition, since a building information model may contain models corresponding to at least two different projects (such as tunnel projects, bridge projects, etc.), and these models may use different data formats (such as RVT, DGN, etc.); or the building information model of the same project may contain component geometric models in multiple formats, it is necessary to select the appropriate building information modeling software according to the specific data format of the model, and develop a dedicated preset export plugin for each software to ensure that the geometric and non-geometric data of all components are exported completely and accurately.

[0052] Specifically, in some embodiments of the present invention, the building information modeling software includes at least two types of software; different building information modeling software has different preset export plugins.

[0053] In addition, to further ensure the integrity of the exported data, an integrity check must be performed on the Building Information Model (BIM) before exporting component data through the preset export plugin to ensure the model's completeness and accuracy. This integrity check examines whether there are any damaged or missing component geometric models in the BIM. If damaged or missing component geometric models are found, they must be repaired or supplemented using the original component design information and construction records.

[0054] Specifically, before loading a preset export plugin into the building information modeling software and exporting the component data of each component in the building information model in the form of a semi-structured data format file, the process also includes: performing an integrity check on the building information model and obtaining the check results; if the check results indicate that there are missing component geometric models in the building information model, repairing the building information model based on the component data corresponding to the missing models.

[0055] Step S102: Use a preset parsing tool to perform parsing processing on the semi-structured data format file, extract component data and store it in a predefined component data structure.

[0056] Specifically, after the preset export plugin exports a semi-structured data format file, a suitable parsing tool is selected to parse the tags and elements in the semi-structured data format file in order to extract the component identity parameters and hierarchical path parameters corresponding to the component.

[0057] In some embodiments of the present invention, the semi-structured data format file includes an Extensible Markup Language (EXPLAIN) file.

[0058] like Figure 2 As shown, when the semi-structured data format file includes an Extensible Markup Language (XML) file, after exporting the XML file, the corresponding XML parsing tool is used to parse the XML file. Accordingly, the parsing tool includes an XML parser, such as the xml.etree.ElementTree module in the Python programming language, or other parsers conforming to the XML 1.0 specification (such as lxml, BeautifulSoup, or the DOM / SAX parser in Java).

[0059] In practice, the semi-structured data format file can also be other types of semi-structured data format files, such as JSON files. The corresponding parsing tool can also be a parsing tool for JSON files. This embodiment does not limit the types of semi-structured data format files and parsing tools.

[0060] After extracting component data from the semi-structured data format file using a parsing tool, the data is stored in a suitable component data structure for subsequent processing. The component data structure includes, but is not limited to, data structures such as lists, dictionaries, and tree structures; this embodiment does not limit the implementation method of the component data structure.

[0061] Step S103: For the hierarchical path parameters in each component data structure, extract the node name of the last node as the entity name to generate a knowledge graph entity, extract the predecessor node of the last node as the predecessor node name, and map the predecessor node name and the corresponding component parameter to entity attributes.

[0062] Each component data structure includes all the tags for the corresponding component, including tags corresponding to component identity parameters (such as "Name" tag, "UUID" tag, etc.) and hierarchical path parameter tags (such as "NodePath" tag).

[0063] For each component's corresponding component data structure, extract the hierarchical path parameters from the component data structure according to the hierarchical path parameter labels, and use the node name of the last node in the hierarchical path parameters as the knowledge graph entity name, and create the knowledge graph entity corresponding to the current component. For example... Figure 2 As shown, the component data corresponding to the current component is written into the attributes of the created knowledge graph entity, including but not limited to identity description parameters, location information parameters, project information parameters, geometric information parameters, etc.

[0064] Additionally, the attributes of a knowledge graph entity include a preceding node attribute, which indicates the preceding node or entity of the current component. Therefore, after creating the knowledge graph entity corresponding to the current component, it is necessary to extract the node name of the second-to-last node in the hierarchical path parameters and write it into the preceding node attribute. If there is only one node in the hierarchical path parameters, then that node is considered the top-level node, and the preceding node attribute in the corresponding knowledge graph entity is empty.

[0065] For example, taking the hierarchical path parameter "Bridge-Substructure-Pier 2-Pile 1" as an example, the knowledge graph entities that need to be created include the "Bridge" entity, the "Substructure" entity, the "Pier 2" entity, and the "Pile 1" entity; among them, the preceding node attribute of the "Bridge" entity is empty; the preceding node attribute of the "Substructure" entity is "Bridge"; the preceding node attribute of the "Pier 2" entity is "Substructure"; and the preceding attribute of the "Pile 1" entity is "Pier 2".

[0066] In some embodiments of the present invention, the component identity parameter includes the component name. Based on this, to ensure the correctness of the entity name of the created knowledge graph entity, it is necessary to compare the entity name with the corresponding component name. If the comparison result indicates that the names are the same, the currently created knowledge graph entity is retained; otherwise, the currently created knowledge graph entity is deleted, and a prompt message is generated indicating that the name in the hierarchical path parameter is different from the name in the component identity parameter.

[0067] In addition, during the creation of knowledge graph entities, to avoid creating multiple knowledge graph entities from the same component, it is necessary to compare the entity name of the currently created knowledge graph entity with the entity names of already created knowledge graph entities to remove duplicates. If an entity with the same name as the currently created knowledge graph entity already exists, the currently created knowledge graph entity will be retained.

[0068] For example, taking a bridge engineering model as an example from a Building Information Modeling (BIM) model; the bridge structure includes: Level 1: Bridge, Level 2: Superstructure, Substructure, Level 3: Pier 1, Pier 2, and Level 4: Pile 1, Pile 2, Pile 3, Pile 4; where the preceding entity of the "Substructure" entity is "Bridge". If the "Substructure" entity is created repeatedly, it will cause duplicate entities and relationships in the knowledge graph, and the duplicate "Substructure" entities need to be deleted; if there are two existing entities in the knowledge graph named "Pile 1", and if the preceding node of both is "Pier 1", it means that both entities represent Pile 1 of Pier 1, and the "Pile 1" entity created after deletion is sufficient.

[0069] Specifically, for each component data structure, after extracting the node name of the last node as the entity name to generate a knowledge graph entity, the process further includes: comparing the entity name of the current knowledge graph entity with the entity names of the already generated knowledge graph entities to determine if there is a name duplication; if there is a name duplication, checking if the preceding node name of the entity with the duplicate name is consistent with the preceding node name of the current knowledge graph entity; if the preceding node names are consistent, deleting the current knowledge graph entity.

[0070] If the names of the preceding nodes are inconsistent, the currently created knowledge graph will be retained.

[0071] For example, taking a bridge engineering model as an example of a building information model; the structure of a bridge includes: Level 1: bridge, Level 2: superstructure, substructure, Level 3: pier 1, pier 2, and Level 4: pile 1, pile 2, pile 3, pile 4; if there are two existing knowledge graph entities with the entity name "pile 1", and the preceding node of the first "pile 1" entity is "pier 1", and the preceding node of the second "pile 1" entity is "pier 2", then it means that although these two entities have the same name, they represent different components, so both are retained.

[0072] Step S104: Using the top-level node in the hierarchical path as the starting node, the successor node of each node is found through recursive traversal to obtain the hierarchical relationship between components. The hierarchical relationship is used to indicate the predecessor and successor nodes of each component.

[0073] Specifically, starting from the top-level node of the hierarchical path parameters (such as a bridge), a recursive traversal method is used to search for the successor nodes of each layer, starting from the top-level node, based on the hierarchical path parameters in the structural data of each component. The hierarchical relationships between components are then found through incremental updates based on the hierarchical path parameters corresponding to each component.

[0074] for example: Figure 3 For the information of the component "pier body", a hierarchical relationship such as "xxx Railway Grand Bridge - Substructure - Pier No. 1 - Pier Body" is created based on the component. Then, by traversing the hierarchical path parameters of other components, incremental updates are made on the basis of "xxx Railway Grand Bridge - Substructure - Pier No. 1 - Pier Body" to finally obtain the complete hierarchical relationship.

[0075] Meanwhile, in the process of finding subsequent nodes based on the hierarchical path parameters of each component, it is necessary to verify whether the attributes of the preceding nodes corresponding to the found subsequent nodes in the knowledge graph are consistent with the names of the preceding nodes that should exist in the hierarchical path, so as to ensure that the constructed hierarchical relationship is accurate.

[0076] Specifically, taking the top-level node in the hierarchical path as the starting node, the successor node of each node is found through recursive traversal to obtain the hierarchical relationship between components, including: for each found node, it is determined whether the name of the predecessor node in the corresponding knowledge graph entity attribute is consistent with the name of the predecessor node in the hierarchical path; if the names are consistent, the current node is retained.

[0077] Step S105: Generate the knowledge graph corresponding to the building information model based on the hierarchical relationship, knowledge graph entities and corresponding entity attributes.

[0078] In some embodiments of the present invention, the component identity parameter includes a unique component identifier, which serves as a key link for associating knowledge graph entities with their corresponding component geometric models after the knowledge graph is generated.

[0079] Specifically, to achieve deep integration of knowledge graphs and building information models (BIMs), a bidirectional positioning interface is developed in the BIM software, establishing a bidirectional mapping mechanism between knowledge graph entities and BIM components based on the model's unique identifier. This mechanism supports information linkage and rapid navigation between the two systems, improving data interaction efficiency and user experience.

[0080] Specifically, after generating the knowledge graph corresponding to the building information model based on hierarchical relationships, knowledge graph entities, and their attributes, the system further supports the following two types of interactive operations: The first category involves locating building information model components from knowledge graph entities.

[0081] Specifically, in response to the received selection operation applied to entities in the knowledge graph, the unique identifier of the component corresponding to the selected entity is determined as the model positioning identifier; based on the preset mapping relationship between the component unique identifier and the component geometric model, the position of the component geometric model corresponding to the model positioning identifier in the building information model is located.

[0082] The second category involves locating knowledge graph entities from building information model components.

[0083] Specifically, in response to the received selection operation applied to the geometric model of a component in the building information model, the unique identifier of the component corresponding to the selected model is determined as the entity location identifier; based on the preset mapping relationship, the position of the knowledge graph entity corresponding to the entity location identifier in the knowledge graph is located.

[0084] In summary, this embodiment provides a method for constructing a component-level knowledge graph based on BIM. By loading a preset export plugin into the Building Information Modeling (BIM) software, component data of each component in the BIM is exported in a semi-structured data format file. The component data includes identity description parameters and hierarchical path parameters for each component in the BIM. A preset parsing tool is used to parse the semi-structured data format file, extracting the component data and storing it in a predefined component data structure. For the hierarchical path parameters in each component data structure, the node name of the last node is extracted as the entity name to generate a knowledge graph entity, and the predecessor node of the last node is extracted as the predecessor node name. The corresponding component data and predecessor node names are mapped to the entity attributes of the generated knowledge graph entity. Using the top-level node in the hierarchical path parameters as the starting node, the successor nodes of each node are found through recursive traversal to obtain the hierarchical relationship between components. The hierarchical relationship is used to indicate the predecessor and successor nodes of each component. Node components; based on hierarchical relationships, knowledge graph entities, and corresponding entity attributes, a knowledge graph corresponding to the Building Information Model (BIM) is generated; this addresses the problems of low efficiency in knowledge graph construction and insufficient semantic relationship mining in the railway BIM field; through a preset export plugin, semi-structured data format files containing rule definitions of component data are exported from the BIM, accurately identifying and extracting rule-based data that meets the requirements for knowledge graph construction; then, with the help of corresponding file parsing technology, components, component data, and relationships between components are quickly extracted in batches, without relying on complex and cumbersome procedures or manual data processing and conversion, significantly reducing manual intervention and data processing time, thereby significantly improving the efficiency of knowledge graph construction and effectively meeting the timeliness requirements of knowledge graph updates for component updates during railway engineering construction; at the same time, avoiding manual knowledge graph construction also reduces errors or omissions in knowledge graph information caused by human error, further improving the accuracy of the knowledge graph.

[0085] Furthermore, by using the unique component code corresponding to each component in the Building Information Model (BIM), a bidirectional positioning mechanism is achieved between component entities in the knowledge graph and components in the BIM. When querying a component entity in the knowledge graph, the bidirectional positioning mechanism can quickly locate the corresponding component and related information in the original BIM, allowing users to intuitively view the specific situation of the component in the BIM. When viewing a component in the BIM, the bidirectional positioning mechanism can quickly link it to the component entities and relationships associated with that component in the knowledge graph, breaking down information silos between the knowledge graph and the BIM, improving the efficiency of information interconnection and collaborative application, and assisting users in making rapid decisions. This comprehensively enhances the intelligent level of railway engineering lifecycle management.

[0086] Furthermore, for different Building Information Modeling (BIM) software and BIM models with different data formats, corresponding preset export plugins are used to export XML files containing component data using a universal XML standard. These XML files are then parsed using an XML parsing tool, effectively solving compatibility issues in cross-platform and cross-stage integration of multi-source heterogeneous BIM data. Through a standardized data export and parsing process, BIM models from different modeling software can be processed uniformly, achieving lossless extraction and structured conversion of key data such as component hierarchical paths, attribute information, and geometric identifiers. Moreover, the universal XML standard defines strict tag structure and data type constraints, ensuring the standardization and verifiability of exported files, improving the accuracy and automation of data parsing. This not only reduces dependence on specific BIM platforms but also provides a high-quality, standardized input source for subsequent knowledge graph construction, enhancing the overall scalability and engineering applicability of the solution.

[0087] Corresponding to the above method, the present invention also provides a BIM-based component-level knowledge graph construction system. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the BIM-based component-level knowledge graph construction method described above.

[0088] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned method for constructing a BIM-based component-level knowledge graph. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0089] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0090] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0091] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0092] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.< / nodepath>

Claims

1. A method for constructing a component-level knowledge graph based on BIM, characterized in that, The method includes: A preset export plugin is loaded into the building information modeling software to export the component data of each component in the building information model in the form of a semi-structured data format file; the component data includes the identity description parameters and hierarchical path parameters of each component in the building information model; The semi-structured data format file is parsed using a preset parsing tool to extract the component data and store it in a predefined component data structure; For each component data structure, the node name of the last node is extracted as the entity name to generate a knowledge graph entity. The preceding node of the last node is extracted as the preceding node name, and the corresponding component data and the preceding node name are mapped to the entity attributes of the generated knowledge graph entity. Using the top-level node in the hierarchical path parameters as the starting node, the successor node of each node is found through recursive traversal to obtain the hierarchical relationship between components; the hierarchical relationship is used to indicate the predecessor and successor node components of each component. Based on the hierarchical relationship, the knowledge graph entities, and the corresponding entity attributes, a knowledge graph corresponding to the building information model is generated.

2. The method according to claim 1, characterized in that, After extracting the node name of the last node as the entity name to generate the knowledge graph entity for the hierarchical path parameters in each component data structure, the method further includes: Compare the entity name of the current knowledge graph entity with the entity names of the already generated knowledge graph entities to determine if there is a name duplication. In the case of duplicate names, check whether the preceding node name of the entity with the duplicate name is consistent with the preceding node name of the current knowledge graph entity; If the names of the preceding nodes are the same, the current knowledge graph entity will be deleted.

3. The method according to claim 1, characterized in that, The process of finding the hierarchical relationship between components by using the top-level node in the hierarchical path parameters as the starting node and recursively traversing to find the successor node of each node includes: For each found node, determine whether the name of the preceding node in the corresponding knowledge graph entity attribute is consistent with the name of the preceding node in the hierarchical path parameter; If the names are identical, retain the current node.

4. The method according to claim 1, characterized in that, The component identity parameter includes a unique component identifier; the unique component identifier is used to associate entities in the knowledge graph with the corresponding component geometric models. After generating the knowledge graph corresponding to the building information model based on the hierarchical relationship, the knowledge graph entities, and the corresponding entity attributes, the process further includes: In response to the received selection operation applied to entities in the knowledge graph, the unique identifier of the component corresponding to the selected entity is determined as the model location identifier; Based on the preset mapping relationship between the unique component identifier and the component geometric model, the position of the component geometric model corresponding to the model positioning identifier in the building information model is located.

5. The method according to claim 4, characterized in that, After generating the knowledge graph corresponding to the building information model based on the hierarchical relationship, the knowledge graph entities, and the corresponding entity attributes, the process further includes: In response to the received selection operation applied to the geometric model of a component in the building information model, the unique identifier of the component corresponding to the selected model is determined as the entity positioning identifier; Based on the preset mapping relationship, the position of the knowledge graph entity corresponding to the entity location identifier in the knowledge graph is located.

6. The method according to claim 1, characterized in that, Before loading a preset export plugin into the building information modeling software and exporting the component data of each component in the building information model in the form of a semi-structured data format file, the process also includes: The integrity of the building information model is checked, and the results are obtained. If the detection result indicates that there is a missing component geometric model in the building information model, the building information model is repaired based on the component data corresponding to the missing model.

7. The method according to claim 1, characterized in that, The semi-structured data format file includes an Extensible Markup Language (XML) file; the parsing tool includes an XML parser.

8. The method according to claim 1, characterized in that, The building information modeling software includes at least two types of software; different building information modeling software has different preset export plugins.

9. A knowledge graph construction system based on Building Information Modeling, comprising a processor, a memory, and computer programs / instructions stored in the memory, characterized in that, The processor is configured to execute the computer program / instructions, and when the computer program / instructions are executed, the system implements the steps of the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method as described in any one of claims 1 to 8.

Images