Knowledge-guided railway bridge construction process visual simulation method

Abstract

The application discloses a knowledge-guided railway bridge construction process visual simulation method and belongs to the field of railway bridge construction, and solves the problems that the existing railway bridge construction process involves many elements, it is difficult to capture the relationship, the behavior state description is not clear, with the advancement of the construction process, the bridge structure is iteratively updated, the environment changes are complex, and it is difficult to establish the correlation between the bridge main body model and the visual simulation environment. The railway bridge construction data is obtained, including GIS data, field data and construction data; the railway bridge construction scene knowledge graph of the three-domain correlation of entities-process-behavior is constructed based on the obtained railway bridge construction data; and the knowledge-guided railway bridge construction process dynamic visual simulation is carried out based on the railway bridge construction scene knowledge graph. The application is used for railway bridge construction process visual simulation.

Description

TECHNICAL FIELD

[0001] The application relates to a knowledge-guided railway bridge construction process visualization simulation method for railway bridge construction process visualization simulation and belongs to the field of railway bridge construction. BACKGROUND

[0002] A railway bridge is a large, complex and delicate dynamic system engineering, which has the characteristics of long span time, many risk factors and rapid construction process changes. The working condition of the railway bridge construction process is complex and has great fluidity, and the relationship between the bridge main body and the surrounding environment is dynamically variable, thus bringing great safety risks. Therefore, accurate control of the construction process is crucial, and through virtual simulation technology, risks can be predicted and controlled in advance by pre-playing the operation process in a virtual environment to ensure the safety of the construction process and improve engineering management efficiency and quality.

[0003] At present, the simulation method of the railway bridge construction process usually relies on parameterized modeling software to design bridge component models according to construction needs, and the models are moved and assembled according to the actual operation situation to simulate the construction process. Since different construction processes require different component models and different operation operations have no unified standard, the model organization and scheduling are difficult, and the simulation process has poor adaptability. Moreover, this method usually considers the fluidity of the model in isolation and ignores the spatio-temporal correlation between the construction process and the surrounding geographical environment. However, in the actual construction process, the scene objects and the environment often change dynamically with time, so it is necessary to improve and optimize this simulation method to accurately describe the behavior state of the scene entity and establish an efficient mapping relationship between the engineering main body model and the geographical environment to support the rapid visualization of various types of railway bridge construction processes.

[0004] A knowledge graph is an advanced knowledge management technology, and its core component unit is a triple of "entity-relation-entity" or "entity-relation-attribute". These triples are connected to each other through the attributes of the entities and the relevant relationships between the entities to form a comprehensive and interconnected knowledge network, and provide a basis for bridge construction simulation processes. However, the existing knowledge graph technology has the problems of weak field object perception and difficult behavior state description when facing complex bridge construction processes, which makes it difficult to fully cope with the changing conditions in the construction process in actual application. Therefore, it is urgent to establish a model that can accurately describe the behavior state of the scene entity and improve the ability of the knowledge graph to describe dynamic spatio-temporal relationships in order to better support the scene modeling of railway engineering and the efficient reuse of field knowledge.

[0005] In summary, when visualizing the railway bridge construction process, the following technical problems exist:

[0006] 1. In the process of railway bridge construction, there are many involved elements, it is difficult to capture the relationship, and the behavior state is not clearly described. As the construction process advances, the bridge structure is constantly updated and the environment changes complexly, making it difficult to establish the correlation between the bridge main model and the visual simulation environment.

[0007] 2. The simulation process of the railway bridge construction process has a large amount of scene information to build, and it is difficult to organize and schedule. Moreover, the construction process of the bridge is complex and diverse in technology and method, and there is a lack of flexible simulation means. SUMMARY

[0008] The purpose of the present application is to provide a knowledge-guided railway bridge construction process visualization simulation method, which solves the problem of the existing technology railway bridge construction process involving many elements, difficult to capture the relationship, and the behavior state is not clearly described. As the construction process advances, the bridge structure is constantly updated and the environment changes complexly, making it difficult to establish the correlation between the bridge main model and the visual simulation environment.

[0009] In order to achieve the above purpose, the technical scheme adopted by the present application is:

[0010] A knowledge-guided railway bridge construction process visualization simulation method, comprising the following steps:

[0011] Step 1, obtaining railway bridge construction data, including GIS data, field data and construction data;

[0012] Step 2, constructing an entity-process-behavior three-domain associated railway bridge construction scene knowledge graph based on the obtained railway bridge construction data;

[0013] Step 3, based on the railway bridge construction scene knowledge graph, the knowledge-guided railway bridge construction process dynamic visualization simulation is carried out.

[0014] Further, the specific steps of step 2 are:

[0015] Step 2.1, based on the construction data, the behavior state of each construction model in the railway bridge construction scene under different construction stages is integrated, and an entity-process-behavior three-domain associated expression model is given, wherein the construction data includes bridge structure and construction process;

[0016] Step 2.2, based on the obtained data and the given entity-process-behavior three-domain associated expression model, the railway bridge construction scene knowledge graph is constructed, and the specific steps are:

[0017] Step 2.21: Based on the given entity-process-behavior three-domain association expression model, extract the three-domain concepts from the domain data in the railway bridge construction scenario, and extract the domain ontology of each concept in the three-domain concepts, thus completing the construction of the knowledge graph pattern layer. The domain data includes design data, model data, and monitoring data. The three-domain concepts include entity domain, process domain, and behavior domain. The domain ontology refers to the pattern layer, which includes the three concept nodes of entity domain, process domain, and behavior domain and their association relationships. The association relationships include hierarchical relationships, semantic relationships, and spatiotemporal coupling relationships.

[0018] The steps for extracting entity domains are as follows:

[0019] Create a bounding box for the entire railway bridge, and select the bottommost component model of the railway bridge within the bounding box as the starting point.

[0020] Create a bounding box for the component model based on the starting point, and emit rays from the six faces of the bounding box of the component model to obtain the relationship and category between the component models that intersect with the rays. The relationship is set as an adjacency relationship and the same category is set as a group. The column and group are used as the storage method for the railway bridge entity domain.

[0021] Starting from the component model where the rays intersect, rays are emitted to continue building the railway bridge entity domain until the construction is complete;

[0022] The steps for extracting process domains and behavior domains are as follows:

[0023] First, at a macro scale, based on the acquired domain data, construction data, and extracted entity domains, the construction plans, design data, and construction schedules of railway bridges are analyzed manually to divide the construction into different stages.

[0024] Secondly, at the meso-level, each construction stage is broken down into specific construction tasks, and these tasks are associated and bound to entity objects in the entity domain to ensure that each task can be further mapped to specific operational behaviors.

[0025] Subsequently, at the micro scale, each task is broken down into specific operational steps. The construction process and specific operational behaviors of each entity object in the entity domain are extracted. For all entity objects in the entity domain, the construction process and specific operational behaviors are organized and combined to form process sequences and operation sequences. Based on the process sequences, the time arrangement and correlation of the construction tasks corresponding to all entity objects in the entity domain are sorted out to establish a complete construction process architecture that reflects the execution order of construction tasks, thus constructing the process domain.

[0026] Based on the operation sequence, the specific behaviors and interaction processes of each entity object in the entity domain are recorded to form a behavior network that reflects the actions and interactions of each entity object, thus constructing the behavior domain.

[0027] Step 2.22: After the knowledge graph pattern layer is constructed, the concept nodes and relationships in the domain ontology are mapped to the entity nodes and relationships in the data layer to realize the organization and storage of data, that is, the construction of the data layer is realized.

[0028] Step 2.23: Then, the open-source graph database Neo4j is used to store and visualize the triple information in the data layer, thus obtaining an instantiated knowledge graph of the railway bridge construction scenario.

[0029] Furthermore, the specific steps of step 3 are as follows:

[0030] Step 3.1: Construct event-driven railway bridge construction scenarios on demand based on the railway bridge construction scenario knowledge graph. The specific steps are as follows:

[0031] Step 3.11: Using railway bridge construction events as the driving force, based on construction data and design data, perform artificial analysis on the component models, geographical environment and construction process associated with railway bridge construction events, identify the characteristic entities, process flow and behavioral elements required in each railway bridge construction event, and map them into the railway bridge construction scenario knowledge graph to identify the entity nodes in the railway bridge construction events and their interrelationships.

[0032] Step 3.12: Based on the GIS data and the entity nodes and their relationships obtained in Step 3.11, model the entity objects represented by the entity nodes respectively. At the same time, integrate multi-source data to establish a geographic environment model. Based on the geographic environment model and the requirements of the railway bridge construction event, construct a virtual railway bridge construction scenario. Here, the entity objects are the categories of component models in the entity domain, including the main railway bridge project and the geographic environment. For the main railway bridge project, modeling is carried out through a hierarchical assembly method. At the same time, the relationships of the main project are encapsulated. The GIS data includes terrain data and image data. The multi-source data includes elevation data (DEM), document data (DOM), vector data, and point cloud data.

[0033] Step 3.2: Visualize and simulate the railway bridge construction process based on the constructed railway bridge construction scenario, specifically as follows:

[0034] A railway bridge construction event model is constructed to capture and represent the dynamic changes in spatiotemporal changes. This model introduces a time axis to correlate the changes of objects in three-dimensional space with time during the railway bridge construction process, forming a four-dimensional spatiotemporal framework to describe the state changes and task progress of various entities during the construction. The railway bridge construction event model is defined as follows:

[0035] M = [T, t, P, E]

[0036] in,

[0037] Where T represents the occurrence time of a railway bridge construction event, used to express the time node when a railway bridge construction event occurs; t represents the duration of a railway bridge construction event, used to represent the duration of each railway bridge construction event; P represents the pose of the component model, used to determine the position and orientation information of each component model involved in the railway bridge construction event; E represents the type of railway bridge construction event, used to describe the type of operation event in the railway bridge construction process; R represents the rotation matrix of the component model; and p represents the three-dimensional coordinate information of the component model.

[0038] In the osgEarth 3D engine, the parameters in the railway bridge construction event model in the event sequence are semantically mapped to animation parameters in keyframes to obtain animation parameter files. Then, the tinyxml2 library is used to parse the animation parameter files. After parsing, the osg::Animationpath interface is used to obtain the frame animation of the component model changing over time. The frame animation is then loaded into the constructed railway bridge construction scene to achieve a visual representation of the construction process.

[0039] Compared with the prior art, the advantages of the present invention are as follows:

[0040] I. This invention designs a three-domain association expression model of entity-process-behavior and proposes a method for automatically extracting bridge entities based on bounding boxes. From this, entities and relationships are extracted to establish an ontology of the bridge construction scenario domain, and a knowledge graph is built accordingly. This enables flexible storage and dynamic expression of knowledge. Specifically, by using the method of bounding boxes and ray intersection, component models in railway bridges can be automatically identified and extracted, and the spatial adjacency relationships between components can be accurately captured. This method not only improves the efficiency and accuracy of railway bridge entity domain extraction but also provides reliable basic data support for subsequent knowledge graph construction. It solves the problems of difficult entity domain extraction, inaccurate classification, and ambiguous hierarchical relationships in existing technologies, achieving the technical effect of automated, accurate, and structured extraction of railway bridge entity domains. At the same time, by extracting process and behavior domains, a clear expression of the construction process architecture and dynamic tracking of entity object behavior can be achieved. This effectively solves the problem that the complex association between construction process and object behavior is difficult to express clearly in existing technologies, improving the accuracy of construction process simulation and ensuring the overall controllability of the construction process.

[0041] Second, this invention proposes a knowledge-guided dynamic visualization simulation method for railway bridge construction. It uses events as the driving force and constructs a virtual construction scenario through three-domain analysis, clarifies the characteristics of scenario elements, defines an event model that takes into account spatiotemporal changes, and realizes the visualization simulation of the railway bridge construction process.

[0042] Third, this invention uses a four-dimensional spatiotemporal railway bridge event model to treat the entire railway bridge construction process as an event tree. By setting different parameters in the event model, the construction events of each tree node are simulated, and the parameters are instantiated to obtain the key frame parameters of each railway bridge construction event to achieve simulation.

[0043] Fourth, this invention uses animation to describe the spatiotemporal narrative of railway bridge construction in the osgEarth 3D engine, and intuitively displays each step of construction in chronological order in the 3D scene, clearly expressing the temporal sequence and dynamic changes of the construction process, enhancing the realism and immersion of the simulation, and improving the flexibility and practicality of the simulation. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a general framework diagram of the present invention;

[0046] Figure 2 This is a schematic diagram of the entity-process-behavior three-domain association expression model in this invention;

[0047] Figure 3 This is a schematic diagram of the entity domain extraction process in the bridge construction scenario of this invention;

[0048] Figure 4 This is the process for constructing a knowledge graph of railway bridge construction scenarios in this invention;

[0049] Figure 5 This is a schematic diagram of the three-domain analysis of the event-driven bridge scene in this invention;

[0050] Figure 6 This is a schematic diagram of the scene modeling process in this invention;

[0051] Figure 7 This is a schematic diagram of a virtual scene for railway bridge construction in this invention;

[0052] Figure 8 This is a schematic diagram of the simulation process of the railway bridge construction event model in this invention;

[0053] Figure 9 This is a schematic diagram of the instantiation process of the event model in the bridge construction process of this invention;

[0054] Figure 10 This is a schematic diagram of the process domain and behavior domain extraction process in this invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] like Figure 1 As shown, it mainly includes acquiring railway bridge construction data, constructing a knowledge graph of railway bridge construction scenarios with three domains of entity-process-behavior association, and knowledge-guided dynamic visualization simulation of the railway bridge construction process.

[0057] First, acquire railway bridge construction data, including GIS data, domain data, and construction data;

[0058] Based on the acquired railway bridge construction data, a knowledge graph of the railway bridge construction scenario is constructed, which is a three-domain association of entity, process, and behavior. The railway bridge construction scenario knowledge graph includes a schema layer and a data layer. For the schema layer, an ontology library is typically used for management. The ontology determines the concept nodes and relationships in the knowledge graph and is a crucial basis for its construction. Based on the detailed classification of concepts such as physical entities, construction processes, and behavioral states in the bridge scenario within the three-domain association expression model, a domain ontology containing hierarchical relationships, semantic relationships, and spatiotemporal coupling relationships between concepts can be extracted, thus completing the construction of the knowledge graph schema layer. The data layer mainly contains data and relationships refined from the schema layer. By mapping the concept node-relationship in the ontology structure to the entity node-relationship in the data layer, data organization and storage are achieved, thus completing the construction of the data layer. Subsequently, this invention uses the open-source graph database Neo4j to store and visualize the triple information in the data layer. This knowledge graph, instantiated using a railway bridge construction scenario domain ontology, can realize a dynamic construction knowledge management process, thereby ensuring the accuracy of railway bridge construction simulation. The specific steps are as follows:

[0059] Traditional knowledge graphs often suffer from shortcomings when dealing with complex engineering projects, such as unclear information organization, difficulty in capturing dynamic changes, and unclear behavioral descriptions. To address these issues, this invention fully considers bridge structure and construction procedures, integrates the behavioral states of various objects (component models) at different construction stages, and designs an entity-process-behavior three-domain association expression model by extracting entities from the railway bridge scene and analyzing the construction process. This achieves efficient management of information across the entire railway bridge construction scenario. Specifically, it integrates the behavioral states of various building models in the railway bridge construction scenario at different construction stages based on construction data, and provides an entity-process-behavior three-domain association expression model, where the construction data includes bridge structure and construction procedures.

[0060] Based on the acquired data and the given entity-process-behavior three-domain association representation model, a knowledge graph of the railway bridge construction scenario is constructed. The specific steps are as follows:

[0061] Based on the given entity-process-behavior three-domain association expression model, the domain data in the railway bridge construction scenario is extracted into three-domain concepts, and the domain ontology of each concept in the three-domain concepts is extracted, thus completing the construction of the knowledge graph pattern layer. The domain data includes design data, model data, and monitoring data. The three-domain concepts include entity domain, process domain, and behavior domain. The domain ontology refers to the pattern layer, which includes the three concept nodes of entity domain, process domain, and behavior domain and their association relationships. The association relationships include hierarchical relationships, semantic relationships, and spatiotemporal coupling relationships.

[0062] In this three-domain relational representation model, the entity domain encompasses all physical and abstract objects in the bridge construction process, including various components of the railway bridge, construction equipment, environmental information, etc. Each entity object has specific attributes and states. Therefore, to ensure the complete representation of entity objects in the model, a bounding box traversal of the bridge scene model is used for extraction. The specific process is as follows: Figure 3 After extraction, a list and groups of entity fields are obtained. The list represents the connection relationships between bridge entities, and the groups represent entities of the same type. The steps for extracting entity fields are as follows:

[0063] Create a bounding box for the entire railway bridge, and select the bottommost component model of the railway bridge within the bounding box as the starting point.

[0064] Create a bounding box for the component model based on the starting point, and emit rays from the six faces of the bounding box of the component model to obtain the relationship and category between the component models that intersect with the rays. The relationship is set as an adjacency relationship and the same category is set as a group. The column and group are used as the storage method for the railway bridge entity domain.

[0065] Starting from the component model where the rays intersect, rays are emitted to continue building the railway bridge entity domain until the construction is complete;

[0066] The process domain describes the various stages and steps of bridge construction, as well as the logical relationships between these stages and steps. It reflects the spatiotemporal changes of entities and relationships during construction and is the core of bridge construction progress. The behavior domain describes the specific operations and behaviors performed by various entities during construction. It is the key execution layer in the bridge construction process, ensuring the smooth completion of various construction tasks. Information from the process and behavior domains is mainly obtained through prior knowledge, design data, and research reports.

[0067] In the integrated expression model of the above entity-process-behavior three-domain association, the behavior domain drives the progress of the construction process through specific operations, the process domain steps are reflected in the state changes of entity objects, and the changes in the entity domain in turn affect the subsequent construction behavior.

[0068] The steps for extracting process domains and behavior domains are as follows:

[0069] First, at a macro scale, based on the acquired domain data, construction data, and extracted entity domains, the construction plans, design data, and construction schedules of railway bridges are analyzed manually to divide the construction into different stages.

[0070] Secondly, at the meso-level, each construction stage is broken down into specific construction tasks, and these tasks are associated and bound to entity objects in the entity domain to ensure that each task can be further mapped to specific operational behaviors.

[0071] Subsequently, at the micro scale, each task is broken down into specific operational steps. The construction process and specific operational behaviors of each entity object in the entity domain are extracted. For all entity objects in the entity domain, the construction process and specific operational behaviors are organized and combined to form process sequences and operation sequences. Based on the process sequences, the time arrangement and correlation of the construction tasks corresponding to all entity objects in the entity domain are sorted out to establish a complete construction process architecture that reflects the execution order of construction tasks, thus constructing the process domain.

[0072] Based on the operation sequence, the specific behaviors and interaction processes of each entity object in the entity domain are recorded to form a behavior network that reflects the actions and interactions of each entity object, thus constructing the behavior domain.

[0073] After the knowledge graph pattern layer is constructed, the concept nodes and relationships in the domain ontology are mapped to the entity nodes and relationships in the data layer, thereby realizing the organization and storage of data, which means the construction of the data layer is realized.

[0074] Then, the open-source graph database Neo4j is used to store and visualize the triple information in the data layer, thus obtaining an instantiated knowledge graph of the railway bridge construction scenario.

[0075] A dynamic visualization simulation of the railway bridge construction process based on a knowledge graph of railway bridge construction scenarios.

[0076] To simulate the construction process of railway bridges, it is first necessary to establish models of the main bridge structure and related geographical environment to generate a virtual scene. However, the construction machinery models and geographical environment involved in railway construction have complex semantic information and relationships. How to integrate, manage, and apply models in long-distance railway projects becomes a key issue. To this end, this case uses construction geographical events as the driving force, analyzes the characteristics of bridge construction scenarios from three dimensions: entities, processes, and behaviors, and identifies key objects in the corresponding scenarios and the relationships between them by mapping event features to a knowledge graph.

[0077] After obtaining the scene entity-relationships through the above process, these entities are modeled separately, and the pose of the models is constrained by the analyzed multi-level semantic relationships to construct a virtual scene of the construction process. The modeling process is as follows: Figure 6 As shown. The smallest unit for modeling the main bridge structure is the primitive model. A primitive model library with adaptive scheduling requirements is established, while semantic information is encapsulated to support the spatial analysis of model objects. The main structure model is formed through hierarchical assembly. The combination and assembly of primitive models forms component models, and the assembly of component models constitutes the main structure model. The geographic environment model is generated by integrating multi-source spatiotemporal data, such as DEM, DOM, and vector data.

[0078] Considering the significant management challenges due to the numerous elements in the construction scene model, this project utilizes the osgEarth 3D earth simulation engine, employing technologies such as the PagedLOD algorithm, multithreading, and multi-layered model-terrain-image fusion to manage the model and generate the scene. In osgEarth, all layer objects are organized within a scene tree structure, allowing for model movement, scaling, and rotation through manipulation of scene nodes. The osg::Group container node manages all bridge main engineering models within the scene, enabling model organization and scheduling through node manipulation. The osgEarth::MapNode node manages geographic environmental information, including imagery, elevation, vector data, and plotting data. Subsequently, the PagedLOD algorithm is used to divide the entire scene into blocks, creating a hierarchical model from low to high detail for each block. Finally, only the blocks within the view frustum are rendered, resulting in a 3D railway information scene, as shown below. Figure 7 As shown.

[0079] This involves constructing event-driven railway bridge construction scenarios on demand based on a knowledge graph of railway bridge construction scenarios. The specific steps are as follows:

[0080] Driven by railway bridge construction events, this study analyzes the component models, geographical environment, and construction processes associated with railway bridge construction events based on construction data and design data. It identifies the characteristic entities, technological processes, and behavioral elements required in each railway bridge construction event and maps them to a railway bridge construction scenario knowledge graph, thereby identifying the entity nodes in the railway bridge construction events and their interrelationships.

[0081] Based on GIS data and the obtained entity nodes and their relationships, the entity objects represented by the entity nodes are modeled separately. At the same time, a geographic environment model is established by integrating multi-source data. Based on the geographic environment model and the requirements of the railway bridge construction event, a virtual railway bridge construction scenario is constructed. Here, the entity objects are the categories of component models in the entity domain, including the main railway bridge project and the geographic environment. For the main railway bridge project, it is modeled by a hierarchical assembly method, while encapsulating the relationships of the main project. The GIS data includes terrain data and image data, and the multi-source data includes elevation data (DEM), document data (DOM), vector data, and point cloud data.

[0082] Railway bridge construction is not only a demonstration of engineering technology, but also a complex process involving dynamic changes in spatiotemporal information. During railway bridge construction, the positions and states of various objects and elements constantly change; these changes are continuous and interconnected in both time and space. To accurately manage and simulate each stage of railway bridge construction, this project establishes an event model capable of capturing and representing these dynamic spatiotemporal changes. This model introduces a timeline, linking the changes of objects in three-dimensional space with time during construction, forming a four-dimensional spatiotemporal framework to describe the state changes and task progress of various objects during construction. Through semantic mapping to animation parameters in keyframes and parsing to obtain frame animations, the construction process is visualized. The specific process is as follows: Figure 8 As shown, this is a visual simulation of the railway bridge construction process based on a constructed railway bridge construction scenario, specifically:

[0083] Based on the constructed railway bridge construction scenario, a railway bridge construction event model is built to capture and represent the dynamic changes in spatiotemporal changes. This model introduces a time axis to correlate the changes of objects in three-dimensional space during the railway bridge construction process with time, forming a four-dimensional spatiotemporal framework to describe the state changes and task progress of various entities during the railway bridge construction process. The railway bridge construction event model is defined as follows:

[0084] M = [T, t, P, E]

[0085] in,

[0086] Where T represents the occurrence time of a railway bridge construction event, used to express the time node when a railway bridge construction event occurs; t represents the duration of a railway bridge construction event, used to represent the duration of each railway bridge construction event; P represents the pose of the component model, used to determine the position and orientation information of each component model involved in the railway bridge construction event; R represents the type of railway bridge construction event, used to describe the type of operation event in the railway bridge construction process; R represents the rotation matrix of the component model; and p represents the three-dimensional coordinate information of the component model.

[0087] In the aforementioned railway bridge construction event model, the spatial displacement or transformation process of the railway bridge construction event model can be regarded as one or more events. When a certain time node is reached, the event is triggered, and the events involved by the entity within a certain period of time are completely constrained by time and space. Extending the railway bridge construction event model to the entire railway bridge construction process enables accurate simulation and management of the entire bridge construction process. Specifically, the entire railway bridge construction event can first be regarded as an event directory tree, with each event sequence process operation considered as a child node in the directory tree. The operation parameters covered by each child node element are obtained from the domain and construction knowledge graph, and encapsulated and stored in an XML file. This operation yields the parameter file after the event model is instantiated. Subsequently, the parameter files of each child node are mapped to keyframes arranged in time sequence. By integrating these keyframes, the animation parameter file of the bridge construction process can be obtained. The overall process is as follows: Figure 9 As shown, the animation parameter file is parsed using the animation interface provided by osgEarth and applied to the relevant models involved in the railway bridge construction event to obtain frame animations. Loading these frame animations into the constructed railway bridge construction scene allows for the correlation between the construction process and the surrounding geographical environment, thus achieving dynamic simulation of the railway bridge construction process. Specifically, in the osgEarth 3D engine, the parameters in each railway bridge construction event model in the event sequence are semantically mapped to animation parameters in keyframes to obtain the animation parameter file. This file is then parsed using the tinyxml2 library. After parsing, the frame animations of the component models changing over time are obtained using the osg::AnimationPath interface. These frame animations are then loaded into the constructed railway bridge construction scene to achieve a visual representation of the construction process.

[0088] Therefore, this invention enables on-demand construction of railway construction scenarios while accurately representing the behavioral states of scene entities. Compared with existing technologies, the advantages and features of this invention are as follows: It designs a three-domain association expression model of entity-process-behavior, extracts entities and relationships from it to establish a bridge construction scenario domain ontology, and builds a knowledge graph accordingly, achieving flexible knowledge storage and dynamic expression; it proposes a knowledge-guided dynamic visualization simulation method for railway bridge construction processes, using events as the driving force to clarify the characteristics of scene elements, define an event model that considers spatiotemporal changes, and create frame animations of the railway bridge construction process in the osgEarth 3D earth simulation engine, achieving visualized simulation of the construction process. Because this method integrates the knowledge graph of railway bridge construction scenarios and considers the four-dimensional spatiotemporal state of specific scene objects during the construction process, it can flexibly adapt to various stages of bridge construction projects, thus possessing strong universality and being suitable for simulating various railway bridge construction processes.

[0089] To address the aforementioned technical challenges, this project analyzes the characteristics of bridge construction scenarios, designs a three-domain association expression model of entity-process-behavior, extracts scenario entities and relationships, constructs a knowledge graph of railway construction scenarios, and clarifies the spatiotemporal behavioral states of objects involved in railway bridge construction events. Simultaneously, using events as the driving force, it clarifies the entities, processes, and behaviors of the construction process, establishes a virtual simulation platform based on the osgEarth 3D engine, and describes the dynamic evolution of railway bridge construction events in detail in a 3D virtual scene through the event model, mapping it to animation keyframes. By parsing the keyframes, it completes the virtual simulation of the railway bridge construction process, solving problems such as unclear organization of element information, difficulty in capturing dynamic environmental changes, and unclear description of object behavior in existing railway bridge construction process simulation methods.

Claims

1. A knowledge-guided visualization simulation method for railway bridge construction process, characterized in that, Includes the following steps: Step 1: Obtain railway bridge construction data, including GIS data, domain data, and construction data; Step 2: Construct a knowledge graph of railway bridge construction scenarios based on the acquired railway bridge construction data, which is related to the three domains of entity, process, and behavior. The specific steps are as follows: Step 2.1: Integrate the behavioral states of various construction models in the railway bridge construction scenario under different construction stages based on construction data, and give a three-domain association expression model of entity-process-behavior, where construction data includes bridge structure and construction procedures; Step 2.2: Based on the acquired data and the given entity-process-behavior three-domain association expression model, construct a knowledge graph for the railway bridge construction scenario. The specific steps are as follows: Step 2.21: Based on the given entity-process-behavior three-domain association expression model, extract the three-domain concepts from the domain data in the railway bridge construction scenario, and extract the domain ontology of each concept in the three-domain concepts, thus completing the construction of the knowledge graph pattern layer. The domain data includes design data, model data, and monitoring data. The three-domain concepts include entity domain, process domain, and behavior domain. The domain ontology refers to the pattern layer, which includes the three concept nodes of entity domain, process domain, and behavior domain and their association relationships. The association relationships include hierarchical relationships, semantic relationships, and spatiotemporal coupling relationships. The steps for extracting entity domains are as follows: Create a bounding box for the entire railway bridge, and select the bottommost component model of the railway bridge within the bounding box as the starting point. Create a bounding box for the component model based on the starting point, and emit rays from the six faces of the bounding box of the component model to obtain the relationship and category between the component models that intersect with the rays. The relationship is set as an adjacency relationship and the same category is set as a group. The column and group are used as the storage method for the railway bridge entity domain. Starting from the component model where the rays intersect, rays are emitted to continue building the railway bridge entity domain until the construction is complete; The steps for extracting process domains and behavior domains are as follows: First, at a macro scale, based on the acquired domain data, construction data, and extracted entity domains, the construction plans, design data, and construction schedules of railway bridges are analyzed manually to divide the construction into different stages. Secondly, at the meso-level, each construction stage is broken down into specific construction tasks, and these tasks are associated and bound to entity objects in the entity domain to ensure that each task can be further mapped to specific operational behaviors. Subsequently, at the micro scale, each task is broken down into specific operational steps. The construction process and specific operational behaviors of each entity object in the entity domain are extracted. For all entity objects in the entity domain, the construction process and specific operational behaviors are organized and combined to form process sequences and operation sequences. Based on the process sequences, the time arrangement and correlation of the construction tasks corresponding to all entity objects in the entity domain are sorted out to establish a complete construction process architecture that reflects the execution order of construction tasks, thus constructing the process domain. Based on the operation sequence, the specific behaviors and interaction processes of each entity object in the entity domain are recorded to form a behavior network that reflects the actions and interactions of each entity object, thus constructing the behavior domain. Step 2.22: After the knowledge graph pattern layer is constructed, the concept nodes and relationships in the domain ontology are mapped to the entity nodes and relationships in the data layer to realize the organization and storage of data, that is, the construction of the data layer is realized. Step 2.23: Then, the open-source graph database Neo4j is used to store and visualize the triple information in the data layer, thus obtaining an instantiated knowledge graph of the railway bridge construction scenario. Step 3: Dynamic visualization simulation of the railway bridge construction process based on knowledge graph of railway bridge construction scenario.

2. The knowledge-guided visualization simulation method for railway bridge construction process according to claim 1, characterized in that, The specific steps of step 3 are as follows: Step 3.1: Construct event-driven railway bridge construction scenarios on demand based on the railway bridge construction scenario knowledge graph. The specific steps are as follows: Step 3.11: Using railway bridge construction events as the driving force, based on construction data and design data, perform artificial analysis on the component models, geographical environment and construction process associated with railway bridge construction events, identify the characteristic entities, process flow and behavioral elements required in each railway bridge construction event, and map them into the railway bridge construction scenario knowledge graph to identify the entity nodes in the railway bridge construction events and their interrelationships. Step 3.12: Based on the GIS data and the entity nodes and their relationships obtained in Step 3.11, model the entity objects represented by the entity nodes respectively. At the same time, integrate multi-source data to establish a geographic environment model. Based on the geographic environment model and the requirements of the railway bridge construction event, construct a virtual railway bridge construction scenario. Here, the entity objects are the categories of component models in the entity domain, including the main railway bridge project and the geographic environment. For the main railway bridge project, modeling is carried out through a hierarchical assembly method. At the same time, the relationships of the main project are encapsulated. The GIS data includes terrain data and image data. The multi-source data includes elevation data (DEM), document data (DOM), vector data, and point cloud data. Step 3.2: Visualize and simulate the railway bridge construction process based on the constructed railway bridge construction scenario, specifically as follows: A railway bridge construction event model is constructed to capture and represent the dynamic changes in spatiotemporal changes. This model introduces a time axis to correlate the changes of objects in three-dimensional space with time during the railway bridge construction process, forming a four-dimensional spatiotemporal framework to describe the state changes and task progress of various entities during construction. The railway bridge construction event model is defined as follows: in, This indicates the time when a railway bridge construction event occurred, used to express the specific time point in the construction of a railway bridge. This indicates the duration of a railway bridge construction event, used to represent the duration of each railway bridge construction event. This represents the pose of the component model, used to determine the position and orientation information of each component model involved in railway bridge construction events. This indicates the type of railway bridge construction event, used to describe the types of operational events during the railway bridge construction process. The rotation matrix represents the component model. Represents the three-dimensional coordinate information of the component model; In the osgEarth 3D engine, the parameters in the railway bridge construction event model in the event sequence are semantically mapped to the animation parameters in the keyframes to obtain the animation parameter file. Then, the animation parameter file is parsed by the tinyxml2 library. After parsing, the frame animation of the component model changing over time is obtained by using the osg::Animationpath interface. The frame animation is then loaded into the constructed railway bridge construction scene to realize the visualization of the construction process.

Images