A pipe network digital twin scene generation method

Abstract

This invention discloses a method for generating digital twin scenes of pipeline networks. First, the pipeline network is classified and spatially calculated. Second, a model connection rule base is established based on pruning rules. Then, a 3D instantiated model of the pipeline network facilities is built, and geometric and texture information is extracted and encapsulated, while retaining the orientation information index. Next, 3D tile organization and distributed storage of the pipeline network data are performed. A loose octree-based 3D tile construction technique is used to generate tile data, ensuring the correctness of the 3D tile data. Distributed storage is used for tile data storage, breaking the constraints of resource usage and data capacity, improving the 3D tile retrieval and retrieval speed, and reducing the data loading pressure on the digital twin scene. Finally, the data is rendered to generate the digital twin scene of the pipeline network. This method significantly reduces data storage, reduces data redundancy, and greatly improves the efficiency of large-scale 3D pipeline network modeling and rendering.

Description

Technical Field

[0001] This invention relates to the field of digital twin scenarios, specifically to a method for generating a pipeline digital twin scenario. Background Technology

[0002] Digital twin scenarios often contain identical models. Traditional methods require modeling each entity separately. For urban underground pipelines (water supply, drainage, gas, etc.), which contain numerous similar facilities (such as pipe sections, wells, valves, fire hydrants, meters, etc.), this not only severely impacts modeling efficiency but also creates enormous storage pressure. As the data volume of various component models increases dramatically, it consumes and occupies a large amount of computer storage resources, putting immense pressure on the rendering engine and causing slow scene loading and rendering stutters.

[0003] Furthermore, to ensure efficient rendering of massive amounts of 3D data in digital twin scenarios, large-scale 3D scene data needs to be divided into several tile units according to certain rules, and organized into 3D tiles using an efficient spatial index structure. Traditional tile construction methods result in an exponential increase in the number of tiles as the tile hierarchy increases, leading to low production efficiency and putting pressure on tile storage and retrieval, which is detrimental to network transmission and parsing rendering. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a method for generating digital twin scenarios for pipeline networks, comprising the following steps:

[0005] S1. Classification and spatial calculation of the pipeline network: The pipeline network includes urban underground pipeline networks such as water supply networks, drainage networks, gas networks, power networks, and communication networks; the classification is based on the cross-sectional attribute information and topological relationships of the pipe segments; the spatial calculation includes pipeline elevation calculation and pipeline orientation calculation.

[0006] S2. Establish a model connection algorithm rule base: In the actual modeling process, pipe points are the endpoints after being trimmed and supplemented according to pipe point characteristics, pipe segment length, and pipe diameter. The model connection algorithm rule base includes corresponding connection algorithms matched by different connection objects; according to the model connection rules, square pipes, round pipes, tees, multi-way valves, elbows, and valve facilities are quickly and seamlessly connected to achieve accurate matching between pipe point objects and pipe segment objects;

[0007] S3. Establish a 3D instantiated model of the pipeline network: Organize pipe segments and pipe points of the same type as unit models, containing only the geometric and texture information of the model; for pipe points and pipelines in different spatial locations, only their spatial position and orientation information are retained;

[0008] S4. Establish a three-dimensional tile organization and distributed storage for pipeline network data, storing the tiles in a distributed manner in the form of tile data and tile indexes using the 3D Tiles format;

[0009] S5. Render the data to generate a digital twin scene of the pipeline network. Use Cesium to parse the 3D tile data of the pipeline network and directly locate the corresponding tile through the tile index; quickly access the data and load geometric information according to the tile range and field of view, and use a facility global sharing mechanism for rendering; and integrate data such as terrain elevation, terrain imagery, roads and bridges, and building white models to complete the rendering of the 3D pipeline network scene through the WebGL engine; perform virtual pipeline network mapping according to the actual pipeline network status, and access monitoring, operation and maintenance, and analysis data in a visual way to generate a digital twin scene of the pipeline network.

[0010] Furthermore, S1 specifically refers to:

[0011] S11. Based on the cross-sectional properties of the pipe sections, the pipe sections are classified into circular pipe sections and square pipe sections; the pipe points are classified into auxiliary pipe points and non-auxiliary pipe points. Auxiliary pipe points include valves, fire hydrants, wells, storm drains, streetlights, gate valves, air vents, condensate tanks, and meters. Non-auxiliary pipe points include tees, crosses, elbows, and reducers. Among them, auxiliary pipe points are further divided into ground-mounted auxiliary points and non-ground-mounted auxiliary points based on whether they are attached to the ground. Ground-mounted auxiliary points include wells, fire hydrants, storm drains, and streetlights. Non-ground-mounted auxiliary points include valves, gate valves, air vents, condensate tanks, and meters.

[0012] S12. Calculate the pipeline point elevation based on ground elevation, burial depth, burial method, and pipe diameter: Burial methods include top burial depth, center burial depth, and bottom burial depth; when the burial method is top burial depth, the pipeline point elevation is subtracted from the pipe segment radius; when the burial method is bottom burial depth, the pipeline point elevation is added to the pipe segment radius; when the burial method is center burial depth, the pipeline point elevation remains unchanged.

[0013] S13. Perform attitude calculations on the corresponding pipelines and pipe points based on the pipeline and pipe point attribute data: If the pipe point is a ground-mounted attachment type, scale the attachment according to the pipe diameter of the connected pipe section to ensure that the pipeline and pipe point match; if the pipe point is a non-ground-mounted attachment type, scale the attachment according to the burial depth and the pipe diameter of the connected pipe section to ensure that the pipeline, pipe point and ground match.

[0014] S14. Determine the actual direction based on the spatial information of pipelines and pipe points, match the directions of the appurtenances with the starting and ending points of the pipe segments, and determine the connection relationship between models through topology analysis and pose reduction.

[0015] Furthermore, the model connection algorithm rule base in S2 includes:

[0016] Dual-segment connection: Pipe point modeling is performed based on the angle between the pipe segment cross-section and the pipe segment;

[0017] Multi-segment connection: Model pipe points or construct wells to connect them using the largest pipe segment cross-section;

[0018] Ground-hugging model: Automatically matches the model to the scene while ensuring that the model is ground-hugging and connected to underground pipe sections;

[0019] Pipeline valve connection: The size and orientation of valves are automatically calculated and adjusted based on pipeline parameters;

[0020] There is a height difference between the pipe segments: the vertical pipe segment model is automatically constructed based on the height difference to complete the connection;

[0021] Cut-off processing: Based on the pipe segment type and parameters, the cut-off is automatically matched and adjusted to complete the connection.

[0022] Furthermore, S3 specifically refers to: for facilities of the same type 1, 2, 3, ..., n, each facility contains geometric, texture, size, and orientation information; wherein, facilities of the same type include pipe sections of the same type and auxiliary facilities of the same type; geometric and texture information occupies a large amount of memory, so the facilities are encapsulated into unit models through 3D modeling tools to uniformly represent the facilities; size and orientation information occupies a small amount of memory, so they are uniformly stored through an indexing mechanism to form an instantiated model of this type of facility.

[0023] Furthermore, S4 specifically includes:

[0024] S41. A 3D tile construction technique based on a loose octree is used to generate tile data. According to the slicing level, each level of 3D tile is divided into eight 3D sub-tiles. Each tile is uniformly numbered according to the level. The loose octree dynamically calculates the range of 3D tiles according to the actual size of the model. If the model size exceeds the range of 3D tiles, the tiles are expanded to include the model data, and the size of its parent 3D tile is recalculated. If there is no model data within the range of 3D tiles, the 3D tile is removed from the queue to avoid invalid searches.

[0025] S42. In 3D Tiles format, tiles are stored in the form of tile data and tile indexes respectively. Tile data includes texture, attributes, and geometry. After instantiation, it is organized and saved in a distributed storage manner in layers and blocks. The tile index establishes a tile description file that matches the root tile. This file records the tile range, visibility distance parameters, tile data references, and child tile information.

[0026] Furthermore, the facility global sharing mechanism in S5 renders the same type of facility across the entire scene precisely to a specified location based on the instantiated model and spatial location information through different pose transformations during the rendering process, and the pose data of the facility is obtained through the pose index.

[0027] This invention provides a method for generating digital twin scenarios of pipeline networks, which has the following beneficial effects:

[0028] By instantiating facilities, an instance model is created for each type of facility. Geometric and texture information is extracted and encapsulated, while the orientation information index is preserved. This significantly reduces data storage, minimizes data redundancy, and greatly improves the efficiency of large-scale 3D pipeline network modeling and rendering. A loose octree-based 3D tile construction technique is used to generate tile data, ensuring the accuracy of 3D tile sizes. Distributed storage is employed for tile data storage, breaking constraints on resource usage and data capacity. This improves the speed of 3D tile retrieval and retrieval, reduces the data loading pressure on digital twin scenes, accelerates the browsing speed of pipeline network digital twin scenes, and enhances the user experience. Attached Figure Description

[0029] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0030] Figure 1 This is a flowchart of a method for generating a digital twin scenario for a pipeline network according to the present invention;

[0031] Figure 2 This is a model connection rule diagram for establishing a model connection rule base in an embodiment of the present invention;

[0032] Figure 3 This is a structural diagram of the instantiation organization method for establishing a three-dimensional instantiation model of pipeline facilities according to an embodiment of the present invention;

[0033] Figure 4 These are partial 3D model diagrams from the 3D instantiated model of pipeline facilities established in this embodiment of the invention;

[0034] Figure 5 This is a diagram of a loose octree structure based on a three-dimensional tile construction technique using a loose octree, according to an embodiment of the present invention. Detailed Implementation

[0035] like Figure 1 As shown: This invention provides a technical solution: a method for generating a digital twin scenario for a pipeline network, comprising the following steps:

[0036] S1. Classification and spatial calculation of the pipeline network: The pipeline network includes urban underground pipeline networks such as water supply pipelines, drainage pipelines, gas pipelines, power pipelines, and communication pipelines; the classification is based on the cross-sectional attribute information and topological relationship of the pipe segments; the spatial calculation includes pipeline elevation calculation and pipeline orientation calculation.

[0037] S11. Based on the cross-sectional properties of the pipe sections, the pipe sections are classified into circular pipe sections and square pipe sections; the pipe points are classified into auxiliary pipe points and non-auxiliary pipe points. Auxiliary pipe points include valves, fire hydrants, wells, storm drains, streetlights, gate valves, air vents, condensate tanks, and meters. Non-auxiliary pipe points include tees, crosses, elbows, and reducers. Among them, auxiliary pipe points are further divided into ground-mounted auxiliary points and non-ground-mounted auxiliary points based on whether they are attached to the ground. Ground-mounted auxiliary points include wells, fire hydrants, storm drains, and streetlights. Non-ground-mounted auxiliary points include valves, gate valves, air vents, condensate tanks, and meters.

[0038] S12. Calculate the pipeline point elevation based on ground elevation, burial depth, burial method, and pipe diameter: Burial methods include top burial depth, center burial depth, and bottom burial depth; when the burial method is top burial depth, the pipeline point elevation is subtracted from the pipe segment radius; when the burial method is bottom burial depth, the pipeline point elevation is added to the pipe segment radius; when the burial method is center burial depth, the pipeline point elevation remains unchanged.

[0039] S13. Perform attitude calculations on the corresponding pipelines and pipe points based on the pipeline and pipe point attribute data: If the pipe point is a ground-mounted attachment type, scale the attachment according to the pipe diameter of the connected pipe section to ensure that the pipeline and pipe point match; if the pipe point is a non-ground-mounted attachment type, scale the attachment according to the burial depth and the pipe diameter of the connected pipe section to ensure that the pipeline, pipe point and ground match.

[0040] S14. Determine the actual direction based on the spatial information of pipelines and pipe points, match the directions of the appurtenances with the starting and ending points of the pipe segments, and determine the connection relationship between models through topology analysis and pose reduction.

[0041] S2. Establish a model connection algorithm rule base: In the actual modeling process, pipe points are the endpoints after being trimmed and supplemented based on pipe point characteristics, pipe segment length, and pipe diameter. The model connection algorithm rule base includes corresponding connection algorithms matched by different connection objects. Based on the model connection rules, square pipes, round pipes, tees, multi-way valves, elbows, and valves are quickly and seamlessly connected, achieving accurate matching between pipe point objects and pipe segment objects. Some model connection rules are as follows: Figure 2 As shown.

[0042] The model connection algorithm rule base includes:

[0043] (1) Two-segment connection: Pipe point modeling is performed based on the angle between the pipe segment cross section and the pipe segment;

[0044] (2) Multi-segment connection: Model pipe points or construct wells to connect them using the largest pipe segment cross-section;

[0045] (3) Ground-hugging model: Automatically matches the model scene while ensuring that the model is ground-hugging and connected to the underground pipe section;

[0046] (4) Pipeline valve connection: The size and posture of the valves are automatically calculated and changed based on the pipeline parameters;

[0047] (5) There is a height difference between the pipe segments: The vertical pipe segment model is automatically constructed based on the height difference to complete the connection;

[0048] (6) Cut-off processing: Based on the pipe section type and parameters, the cut-off is automatically matched and adjusted to complete the connection.

[0049] S3. Establish a 3D instantiated model of the pipeline network: Organize pipe segments (e.g., circular pipe segments, square pipe segments) and pipe points (e.g., tees, tees, and accessories) of the same type as unit models, containing only the geometric and texture information of the model; for pipe points and pipelines in different spatial locations, only their spatial position and orientation information are retained. The organization method of facility instantiation is as follows: Figure 3 As shown, for facilities of the same type 1, 2, 3, ..., n, each facility contains geometric, texture, size, and orientation information; facilities of the same type include pipe sections and auxiliary facilities of the same type. For facilities of the same type, geometric and texture information consumes a lot of memory. Therefore, the facilities are encapsulated into unit models using 3D modeling tools for a unified representation. The size and orientation information of each facility consumes less memory and is stored uniformly through an indexing mechanism, forming an instantiated model of that type of facility.

[0050] S4. Tile data is generated using a 3D tile construction technique based on a loose octree. For example... Figure 4 As shown, based on the slicing level, each level of 3D tile is divided into eight 3D sub-tiles, and each tile is uniformly numbered according to its level (e.g., 0, 01, 001, etc.). The loose octree serves two purposes: firstly, it dynamically calculates the 3D tile range based on the actual model size, expanding the tile to include model data when the model size exceeds the range, and recalculating the size of its parent 3D tile; secondly, it removes 3D tiles from the queue when no model data exists within their range, avoiding invalid searches.

[0051] Tiles are stored in 3D Tiles format, consisting of tile data and tile indexes. Tile data includes textures, attributes, and geometry, and after instantiation, it is organized and saved in a distributed storage manner, using layers and blocks. The tile index establishes a tile description file matching the root tile, which records the tile extent, visibility distance parameters, references to tile data, and child tile information.

[0052] S51. By parsing 3D tile data of the pipeline network using Cesium, model loading can be achieved without any plugins. The corresponding tile can be directly located via tile index, and data and geometric information can be quickly accessed and loaded based on the tile range and view area. A global facility sharing mechanism is used for rendering. This mechanism accurately renders facilities of the same type across the entire scene to a specified location based on instantiated models and spatial location information, through different pose transformations (scaling, translation, rotation). The facility's pose data is obtained through pose indexes.

[0053] S52. It integrates terrain elevation, terrain imagery, road and bridge data, and building white models, and uses the WebGL engine to render the 3D pipeline scene.

[0054] S53. Based on the actual pipeline network status, perform virtual pipeline network mapping, and access various monitoring, operation and maintenance, and analysis data in a visualized manner to generate a digital twin scenario of the pipeline network.

Claims

1. A method for generating a digital twin scenario for a pipeline network, characterized in that, Includes the following steps: S1. Classification and spatial calculation of the pipeline network: The pipeline network referred to is the urban underground pipeline network; The classification is based on the pipe section cross-sectional attribute information and topological relationships; The spatial calculation includes pipeline and pipe point elevation calculation and pipeline and pipe point attitude calculation; S2. Establish a model connection algorithm rule base: Define model connection rules, trim and supplement pipe segments according to pipe point characteristics, pipe segment length, and pipe diameter. Each endpoint of the trimmed and supplemented pipe segment is used as a pipe point in the actual modeling process. The model connection algorithm rule base includes corresponding connection algorithms matched by different connection objects. According to the model connection rules, facilities in the pipeline network are quickly and seamlessly connected to achieve accurate matching between pipe point objects and pipe segment objects. S3. Establish a 3D instantiated model of the pipeline network: Organize pipe segments and pipe points of the same type as unit models, and retain only the geometric and texture information of the model; for pipe points and pipelines in different spatial locations, only their spatial position and orientation information are retained; S4. Establish a three-dimensional tile organization and distributed storage for pipeline network data, storing the tiles in a distributed manner in the form of tile data and tile indexes using the 3D Tiles format; S5. Render the data to generate a digital twin scene of the pipeline network. Use Cesium to parse the 3D tile data of the pipeline network and directly locate the corresponding tile through the tile index. Quickly access the data and load geometric information according to the tile range and field of view, and use the facility global sharing mechanism for rendering. It integrates prior geographical information and uses the WebGL engine to render the 3D pipeline network scene; it performs virtual pipeline network mapping based on the actual pipeline network status, and accesses monitoring, operation and maintenance, and analysis data in a visual way to generate a digital twin scene of the pipeline network. Specifically, S1 is: S11. Based on the cross-sectional properties of the pipe segments, the pipe segments are divided into circular pipe segments and square pipe segments; pipe points are divided into accessory pipe points and non-accessory pipe points; and accessory pipe points are divided into ground-mounted accessories and non-ground-mounted accessories based on whether they are attached to the ground. S12. Calculate the pipeline point elevation based on ground elevation, burial depth, burial method, and pipe diameter: Burial methods include top burial depth, center burial depth, and bottom burial depth; when the burial method is top burial depth, the pipeline point elevation is subtracted from the pipe segment radius; when the burial method is bottom burial depth, the pipeline point elevation is added to the pipe segment radius; when the burial method is center burial depth, the pipeline point elevation remains unchanged. S13. Calculate the attitude of the corresponding pipelines and pipe points based on the pipeline and pipe point attribute data: If the pipe point is a ground-mounted attachment type, scale the attachment according to the pipe diameter of the connected pipe section to ensure that the pipeline and pipe point match; if the pipe point is a non-ground-mounted attachment type, scale the attachment according to the burial depth and the pipe diameter of the connected pipe section to ensure that the pipeline, pipe point and ground match. S14. Determine the actual direction based on the spatial information of pipelines and pipe points, match the directions of the appurtenances with the starting and ending points of the pipe segments, and determine the connection relationship between models through topology analysis and pose reduction.

2. The method for generating a digital twin scenario for a pipeline network according to claim 1, characterized in that, The model connection algorithm rule base in S2 includes: Dual-segment connection: Pipe point modeling is performed based on the angle between the pipe segment cross-section and the pipe segment; Multi-segment connection: Model pipe points or construct wells to connect them using the largest pipe segment cross-section; Ground-hugging model: Automatically matches the model to the scene while ensuring that the model is ground-hugging and connected to underground pipe sections; Pipeline valve connection: The size and orientation of valves are automatically calculated and adjusted based on pipeline parameters; There is a height difference between the pipe segments: the vertical pipe segment model is automatically constructed based on the height difference to complete the connection; Cut-off processing: Based on the pipe segment type and parameters, the cut-off is automatically matched and adjusted to complete the connection.

3. The method for generating a digital twin scenario for a pipeline network according to claim 1, characterized in that, Specifically, S3 is as follows: For facilities of the same type 1, 2, 3, ..., n, each facility contains geometric, texture, size, and orientation information; wherein, facilities of the same type include pipe sections of the same type and auxiliary facilities of the same type; geometric and texture information are encapsulated into unit models by 3D modeling tools to uniformly represent the facilities; size and orientation information are uniformly stored by establishing an indexing mechanism to form an instantiated model of the facility of this type.

4. The method for generating a digital twin scenario for a pipeline network according to claim 1, characterized in that, Specifically, S4 is: S41. A 3D tile construction technique based on a loose octree is used to generate tile data. According to the slicing level, each level of 3D tile is divided into eight 3D sub-tiles. Each tile is uniformly numbered according to the level. The loose octree dynamically calculates the range of 3D tiles according to the actual size of the model. If the model size exceeds the range of 3D tiles, the tiles are expanded to include the model data, and the size of its parent 3D tile is recalculated. If there is no model data within the range of 3D tiles, the 3D tile is removed from the queue. S42. In 3D Tiles format, tiles are stored separately in the form of tile data and tile index. Tile data includes texture, attributes, and geometry. After instantiation, it is organized and saved in a distributed storage manner in layers and blocks. The tile index establishes a tile description file that matches the root tile. This file records the tile range, visibility distance parameters, tile data references, and child tile information.

5. The method for generating a digital twin scenario for a pipeline network according to claim 1, characterized in that, The facility global sharing mechanism described in S5 renders facilities of the same type across the entire scene precisely to a specified location based on instantiated models and spatial location information, through different pose transformations, during the rendering process. The pose data of the facilities is obtained through the pose index.

Images