Hydropower project hub scene scanning model automatic identification and replacement system and method

Abstract

The application discloses a system and method for automatically identifying and replacing a hydropower engineering hub scene scanning model, and belongs to the technical field of engineering three-dimensional visualization. The technical problem to be solved is that the scanning model precision of a hydropower engineering and the platform bearing property are difficult to be considered together, and there is a lack of effective means for automatically identifying and replacing a hydropower engineering hub scene scanning model. Technical solution points are as follows: based on a three-dimensional visualization engine architecture, the steps of sequentially performing the steps of constructing a classified parameterized hydropower engineering model database, importing and processing an engineering site scanning model and a terrain model, dividing the model into blocks and identifying the categories of the blocks, generating a lightweight standard model to replace the original model, displaying the replacement results and providing an external data calling interface are performed, so that the automatic identification and standardized replacement of the scanning model are realized.

Description

Technical Field

[0001] This invention belongs to the field of engineering 3D visualization technology, specifically relating to an automatic identification and replacement system and method for scanning models of hydropower engineering hub scenes. Background Technology

[0002] Large-scale scene models of hydropower projects, encompassing all disciplines, serve as the foundational data for 3D visualization management and intelligent management of the projects. They are a core element of the digital and intelligent construction of hydropower projects. However, hydropower projects inherently possess characteristics such as large land areas, dispersed key structures, numerous professional disciplines involved, and a wide variety of supporting engineering equipment and construction machinery. This leads to irreconcilable technical contradictions in setting the accuracy of the on-site scanning model. If the accuracy of the scanned model is set too low, the spatial relationships between the key buildings, equipment, and construction machinery in the project cannot be clearly presented, making it difficult to meet the needs of detailed viewing and data application of the 3D model at each stage of the hydropower project. If the accuracy of the scanned model is set too high, the number of facets in the model will increase significantly, placing extremely high demands on the digital platform and hardware equipment that support model display and data processing, making it impossible to achieve a complete and smooth browsing of the entire project's key scene model. Therefore, in the actual application of 3D visualization platforms for hydropower projects, it is often necessary to adapt to the platform and hardware capabilities by reducing the model accuracy or shrinking the model display range, which seriously affects the actual application effect of 3D visualization of hydropower project key scenes.

[0003] Currently, there is no effective automated processing method for on-site scanning models of hydropower engineering hubs. It is impossible to achieve lightweight and standardized processing of scanning models while preserving engineering details. It is also difficult to balance the needs for detailed model display with the needs for smooth operation of digital platforms. This problem has become a key pain point restricting the application of 3D visualization technology in the digital construction of hydropower projects.

[0004] In view of this, the present invention is hereby proposed. Summary of the Invention

[0005] To address the aforementioned technical problems in the existing technology, this invention provides an automatic identification and replacement system and method for scanning models of hydropower engineering hub scenes, which solves the problem that it is difficult to balance the accuracy of hydropower engineering scanning models with the platform's load-bearing capacity, and that there is a lack of effective means for automated identification and lightweight replacement of hydropower engineering hub scene scanning models.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: The first aspect is the automatic identification and replacement system for scanning models of hydropower engineering hub scenes, including: Scanned Model Import Module: This module imports the scanned hydropower engineering hub scene model and engineering area terrain model from the engineering site into the 3D visualization engine architecture. It reads, marks, organizes, and visualizes the model data, and outputs the processed model data to the automatic model recognition module. Model database construction module: used to collect BIM models of hydropower projects, extract model feature parameters, convert BIM models into standard driveable parametric models, build and store a classified hydropower project model database, and provide model data and feature parameter query and call for the model automatic recognition module and the engineering model replacement module; Automatic Model Recognition Module: This module receives model data from the scanned model import module, performs model segmentation and recognition, segments the hydropower engineering hub scene model, identifies the category and sub-category of each model segment by combining the model data and feature parameters from the model database construction module, adds the category information to the attribute information of the corresponding model segment, and outputs the marked model segment data to the engineering model replacement module. Engineering Model Replacement Module: This module receives the tagged model block data output by the automatic model recognition module, performs model generation and replacement, combines the model data from the model database construction module, generates matching lightweight blocks based on the category information of the model blocks, replaces the original model blocks in the hydropower engineering hub scene model with the lightweight blocks, performs stylistic adjustments on the lightweight blocks, and outputs the replaced model results to the model display module. Model Display Module: This module receives the replaced model output from the engineering model replacement module, carries and displays the hydropower engineering hub scene model, generates a model structure tree, and provides a data interface for external engineering system platforms to call model-related results and data.

[0007] Furthermore, the scanning model import module includes: Scanning Model Import Unit: Equipped with a data conversion interface, it is used to read and import engineering 3D models acquired by UAVs or 3D scanning equipment, and supports reading and conversion of mainstream model file formats; Model data reading unit: used to read, mark and integrate the basic data of the imported model, realize model naming and numbering, integrate the scanned model according to geographic coordinate information and calculate the geometric information such as the model's volume, number of faces, and basic dimensions; Scanning Results Carrying Unit: Used to call the 3D visualization engine architecture to visualize and carry the engineering site scanning model and related data.

[0008] Furthermore, the model database construction module includes: Engineering Model Input Unit: Used to read and input BIM models of hydropower projects, parse the model files and import them into the 3D visualization engine architecture, and parse and associate information such as model category, geometric data and engineering parameters; Model Feature Extraction Unit: Used to call the model feature parameter extraction algorithm to extract feature data such as volume, control dimensions, geometric structure, component surfaces, and color of the BIM model; Model parametric unit: Used to parametrically modify the main structural dimensions of the BIM model of hydropower project and convert it into a standard driveable parametric model; Model data construction unit: used to organize model and feature data, build the model database structure tree system and model data storage mapping relationship, and integrate and sort various types of model and parameter data.

[0009] Furthermore, the execution flow of the model feature parameter extraction algorithm is as follows: Enter the system models of each sub-specialty of hydropower engineering with known categories and attribute information parameters; Screen and classify similar models, measure the control dimensions and volume of the models, calculate the mean of the model control dimensions samples under the corresponding category and model as the basic size feature parameter, and calculate the mean volume as the basic volume feature parameter; Read the structural feature information of the model and calculate the mean under the corresponding category to form a structural feature parameter group for the main category and sub-category of the model. The structural feature information includes the number of sub-categories, the number of model faces, and the key dimensions of the structure. Read the RGB color parameter values ​​of the model, eliminate samples with large differences, and calculate the mean, which is used as the color feature parameter of the corresponding category model; Organize the basic size feature parameters, basic volume feature parameters, structural feature parameter groups, and color feature parameters of each category and subcategory model, and store them in the model database.

[0010] Furthermore, the automatic model recognition module includes: Model Block Processing Unit: Used to call the model block algorithm, calculate the differences between the model cross section and the terrain of the engineering area and the planar outline of the scene, divide the surface range of the model block from the hydropower engineering hub scene model, and pack and reassemble the corresponding model surfaces into model blocks; Model block recognition unit: It is used to call the model recognition algorithm and the data in the model database to analyze and recognize the model blocks one by one, automatically identify and mark the category information of the hydropower project hub buildings, equipment, construction machinery, etc. corresponding to the model blocks, and add the category and component name to the attribute information of the model blocks.

[0011] Furthermore, the execution flow of the model block division algorithm is as follows: The full-hub scene scanning model of the hydropower project and the terrain model of the project area are imported into the 3D visualization engine architecture, and the 3D spatial superposition and overlap of the two models are completed through geographic information coordinates; Adjust the model coordinate system, using the XY plane with Z=0 as the ground plane, and perform multiple sections on the two models parallel to the Z-axis according to the X and Y-axis coordinates to obtain the YZ and XZ section planes and mark the corresponding section outlines of the two models. The contour line difference judgment formula is used to determine the difference between the contour lines of the two models in each section plane, and the contour lines of the scene model that are significantly different from the terrain contour lines are marked. Organize the marked contour lines in three-dimensional space, connect the contour lines at intervals to generate curved surfaces, wrap the corresponding terrain curved surfaces with the curved surfaces to form three-dimensional entities, and package them into model blocks. Package all parts in the three-dimensional space that differ from the terrain model into preliminary model blocks; The entire hub scene model is judged and filtered for scene planes. The model is divided into grids according to the accuracy. The coordinates of the center point of the grid are read. The minimum detection range of the model's surface area is set. The detection range is determined by the relationship between the four coordinate line segments. The detection results are marked. The judgment of all minimum surface areas is completed in a loop. For the smallest patch region marked as being on the same plane, check whether its neighboring patch regions are on the same scene plane. Add a unified scene plane number to the patches on the same plane. Repeat this process to determine and mark all the smallest patch regions and their adjacent regions. Organize the judgment results, organize the model facets on the same plane according to the scene plane number, and calculate the expression formula of each scene plane within the allowable error accuracy range; Read the scene plane expression formula and construct the corresponding expression for the current scene plane. The coordinate axes are based on the direction of the scene plane normal. The axis is the scene plane. Value 0 flat; along The shaft is divided into multiple segments. , Planar sectioning: Mark the model outline and plane axis in each sectioning plane; determine the difference between the model outline and plane axis using a difference judgment formula; repeat the process to determine the difference between the outlines of all sectioning planes. Organize the marked contour lines in three-dimensional space, connect the contour lines at intervals to generate curved surfaces, wrap the corresponding scene planes with the curved surfaces to form three-dimensional entities, package them into model blocks, and repeat the block operation of all model facets that are different from the plane in the scene plane. All model blocks are organized hierarchically and structurally into 3D entities, and then output to the engineering model replacement module.

[0012] Furthermore, the execution flow of the model recognition algorithm is as follows: Read the model blocks and decompose and isolate them according to their internal hierarchical relationships to prepare for the recognition operation; Measure the volume and control dimensions of the isolated model blocks, call the basic size feature parameters and basic volume feature parameters of the model database, perform preliminary category screening of the model blocks through the size and volume screening formula, and record the results; Retrieve the structural feature parameter group and color feature parameter corresponding to the preliminary screening category, measure various parameters of the model block and its internal sub-model block, accurately identify the category of the model block through the multi-feature weighted judgment formula, and label the identification results into the attribute information of the model block; The loop operation completes the identification and attribute labeling of all model blocks in the hydropower project hub scene model, which can then be used by subsequent modules.

[0013] Furthermore, the engineering model replacement module includes: Model generation unit: Used to call the model generation algorithm, measure the key dimensions of the model blocks of the identified categories and models, search for the corresponding standardized model from the model database and fill in the measurement parameters, and drive the generation of lightweight models with matching dimensions; Model replacement unit: used to call the model replacement algorithm to replace the original model blocks in the hydropower project hub scene model with the lightweight version; Model Adjustment Unit: Used to replace the non-parametric driven parts generated in lightweighting with a model structure that conforms to the style of engineering architecture and equipment.

[0014] Furthermore, the execution flow of the model generation algorithm is as follows: Read each model block sequentially, and read the category information of the model block; Based on the model block categories, retrieve the corresponding main body and substructure categories, control size ranges, key size ranges, and feature parameter group data ranges from the model database; Based on the model recognition results, measure and read the feature dimensions or parameter group data in the corresponding structural parts of the model blocks; Retrieve the standard, driverable parametric model of the corresponding category and model from the model database, fill in the feature data read from the measurement, and generate a lightweight model through parameter-driven generation; The loop operation generates lightweight versions of all model blocks for use by subsequent modules and algorithms.

[0015] Furthermore, the execution flow of the model replacement algorithm is as follows: Based on the model recognition results and the lightweight main structure positioning standard, the main structure of the original model blocks is identified, and its spatial positioning coordinates are measured and read. In the scene model of the hydropower project hub, delete the model facets of the original model blocks, read the terrain surface or scene plane where the original model blocks are located, fill in the missing model facets along the extension trend of the surface / plane, and assign model materials to the missing parts of the model according to the adjacent model facets. Place the corresponding lightweight component at the spatial positioning coordinates of the main structure of the original model block; The loop operation completes the replacement of all model blocks, organizes and updates the replaced hydropower project hub scene model in three-dimensional space, and provides it for display by the model display module and for external system calls.

[0016] Secondly, the method for automatic identification and replacement of scanning models of hydropower project hub scenes, applied to the aforementioned automatic identification and replacement system for scanning models of hydropower project hub scenes, includes: S1. Construct a hydropower engineering model database: collect hydropower engineering BIM models, extract the feature parameters of the BIM models, convert the BIM models into standard driveable parametric models, construct and store a classified hydropower engineering model database, and provide model data and feature parameter query and call for subsequent steps; S2. Scanned Model Import and Processing: Import the scanned hydropower project hub scene model and the terrain model of the project area into the 3D visualization engine architecture. Read, mark, organize and visualize the imported model data, and output the processed model data. S3. Model Segmentation and Category Recognition: Receive the processed model data output from step S2, segment the hydropower engineering hub scene model, combine the model data and feature parameters in the model database constructed in step S1, identify the category and sub-category of each model segment, add the category information to the attribute information of the corresponding model segment, and output the marked model segment data. S4. Model generation and replacement processing: Receive the marked model block data output from step S3, combine it with the model data in the model database constructed in step S1, generate a matching lightweight standard model based on the category information of the model block, replace the original model block in the hydropower engineering hub scene model with the lightweight standard model, perform stylistic adjustments on the lightweight standard model, and output the replaced model result. S5. Model Display and Data Output: Receive the replaced model output from step S4, carry and display the hydropower engineering hub scene model, generate a model structure tree, and provide a data interface for external engineering system platforms to call the model-related results and data.

[0017] The beneficial effects of this invention are as follows: This method effectively addresses the challenge of balancing the accuracy of scanning models for hydropower projects with the platform's capacity. While preserving the detailed features of the hydropower project hub scene model, it replaces the original scanning model with a lightweight standard model, reducing the overall polygon count and adapting to the capacity of digital platforms and hardware devices. This enables complete and smooth browsing and display of the entire hydropower project hub scene model. Simultaneously, it solves the problem of the lack of automated identification and effective lightweight replacement methods for hydropower project hub scene scanning models. Through a standardized process, it automates the entire process from model import and segmented identification to replacement and display, eliminating the need for manual modification of each model. This improves the efficiency and standardization of hydropower project scanning model processing, providing lightweight and standardized 3D model foundational data support for the digital and intelligent construction of hydropower projects. Attached Figure Description

[0018] Figure 1 Architecture diagram of the automatic identification and replacement system for hydropower engineering hub scene scanning model provided in this embodiment of the invention; Figure 2 The execution flowchart of the model block division algorithm provided in the embodiment of the present invention is shown. Detailed Implementation

[0019] The technical solution of the present invention will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are not all embodiments of the present invention. All other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0020] It should be noted that, unless otherwise specifically stated, the relative arrangement and numerical expressions of the components and steps described in these embodiments should not be construed as limiting the scope of the invention.

[0021] The following description of exemplary embodiments is merely illustrative and is not intended to limit the invention or its application or use in any way. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail herein, but where applicable, such techniques, methods, and apparatus should be considered part of this specification.

[0022] Example 1 See Figure 1 , Figure 1 This is an architecture diagram of the automatic identification and replacement system for hydropower engineering hub scene scanning models proposed in this invention, which may specifically include: M1, Scanned Model Import Module: Used to import the scanned hydropower engineering hub scene model and engineering area terrain model from the engineering site into the 3D visualization engine architecture. It reads, marks, organizes, and visualizes the model data, and outputs the processed model data to the automatic model recognition module; specifically including: M11, Scanning Model Import Unit: Equipped with a data conversion interface, it is used to read and import engineering 3D models collected by UAVs or 3D scanning equipment into various units of the scanning model import module. It supports reading and conversion of mainstream model file formats, and the imported data serves as the basis for visual data collected on the engineering site. M12, Model Data Reading Unit: Used to read, mark and integrate the basic data of the imported model. It can perform preliminary naming and numbering of the scanned model, and integrate and average the scanned models of each segment according to geographic coordinate information. At the same time, it reads and calculates various geometric information such as model volume, number of faces, and dimensions of each basic part during the process. M13, Scanning Results Carrying Unit: Used to call the 3D visualization engine architecture to visualize the engineering site scanning model and related data, providing a space for displaying and viewing a visualized 3D model of the entire engineering hub and its corresponding data.

[0023] M2, Model Database Construction Module: Used to collect BIM models of hydropower projects, call the model algorithm module to extract model feature parameters, convert BIM models into standard, driveable parametric models, build and store a categorized hydropower project model database, and provide query and retrieval of model data and feature parameters for the model automatic recognition module and the engineering model replacement module; specifically including: M21, Engineering Model Input Unit: Used to read and input a large number of hydropower engineering BIM models. After parsing the model files, it is imported into the 3D visualization engine architecture. During the input process, the category, geometric data, engineering parameters, etc. of each BIM model file are parsed and input. Various basic data information in the model file attributes are extracted and associated with the corresponding model category and sub-category. M22, Model Feature Extraction Unit: Used to call the model feature parameter extraction algorithm to extract feature data such as volume, control dimensions, geometric structure, component surfaces, and color of the BIM model; provides a program interface to extract the corresponding model feature data under each category by calling the model feature parameter extraction algorithm. The extracted feature data includes: The model includes corresponding volume feature data, control dimension feature data, geometric structure feature data, surface feature data, and color feature data, and a sub-module database system is constructed through the model database to associate each feature data with the corresponding BIM model file. Specifically, the model feature parameter extraction algorithm supports the application functions of the model feature extraction submodule. It extracts feature data of various models from a large number of engineering BIM models based on model category and subcategory comparison, which is then analyzed and judged by the model automatic recognition module. The specific algorithm flow is as follows: M221. Input a large number of known categories and attribute information parameters for various sub-professional system models of hydropower projects; M222. Screen and classify similar models, measure the control dimensions and volume of the models, calculate the mean of the model control dimensions samples under the corresponding category and model as the basic size feature parameter, and calculate the mean volume as the basic volume feature parameter. Specifically, the model category and sub-category information is read, and models of the same type are filtered out for classification and organization. The control dimensions of all collected models in each category and sub-category are measured. The control dimensions are the minimum rectangular dimensions of the entire model. The control dimensions of models in each category and sub-category are collected and organized. The models are divided according to the corresponding model parameters in the model information of each category, and the mean of the maximum control dimensions of the models in the corresponding category and model is calculated. This mean is used as the basic size characteristic parameters of each category and model. The mean volume of each category and model is calculated as the basic volume characteristic parameters of each category and model. M223. Read the structural feature information of the model and calculate the mean under the corresponding category to form a structural feature parameter group for the main category and sub-category of the model. The structural feature information includes the number of sub-categories, the number of model faces, and the key dimensions of the structure. Specifically, based on the model category classification, a large amount of structural feature information of each model category and its subcategories is read. The structural feature information includes the number of subcategories in the main category model to which the model belongs, the number of faces in the main category and its internal subcategory models, the key dimensions of the main category structure and the subcategory structure, and the geometric dimensions of the main functional structural parts in the main category and subcategory models determined by classification and function.

[0024] For example, in the dam hub structure category, the main dam structure dimensions will be used as key dimensions. Various equipment and supporting structures on the dam crest will affect the control dimension data of the dam model, but will not affect the key structural dimension data of the dam model. The average values ​​of the structural feature information data for each of the above structures under the corresponding category are extracted from a large number of models to form the structural feature parameter groups for the main category and subcategories of the model. M224. Read the RGB parameter values ​​of the model, eliminate samples with large differences, and calculate the mean as the color feature parameter of the corresponding category model. Specifically, based on the model sub-category division, a large amount of color feature information of each model category and sub-category is read, the color RGB parameter values ​​of the corresponding model category and sub-category are collected and organized, the model color parameter samples with large differences in each corresponding category are eliminated, and only the sample color data of the color distribution interval of each category model is selected and the average value is taken as the color feature parameter of the model of that category. M225. Organize the basic dimensional feature parameters, basic volume feature parameters, structural feature parameter groups, and color feature parameters of each category and subcategory model, and store them in the model database.

[0025] Specifically, it collects and organizes the basic size characteristic parameters, basic volume characteristic parameters, structural characteristic parameter groups, and color characteristic parameter data of each category and subcategory model obtained through analysis and calculation of a large amount of model data, for subsequent use by various functional modules and algorithms.

[0026] M23, Model Parametric Unit: Used to parametrically modify the main structural dimensions of the BIM model of hydropower projects and convert it into a standard driveable parametric model; First, based on the model's attribute information and parameter specifications, the corresponding model category is called, or corresponding model sub-category modules are added or deleted. Then, utilizing the attributes and functions established during BIM model construction, the control dimensions of each structural element in the model are modified according to the input main structural dimensions. This adjusts the dimensions, volume, structural spacing, and other data of each structural component, enabling automatic modification and generation of the model's structure and geometry through parametric adjustments. This data is then used by the subsequent engineering model replacement module's related function algorithms.

[0027] M24, Model Data Construction Unit: Used to collect and organize various types of model data. Based on the BIM model categories for each project, it constructs a model database structure tree and model data storage mapping relationships, integrating and sorting various models and their corresponding data parameters. The collected data includes the name of each BIM model, its category and subcategories, model geometric dimensions, driveable geometric parameters, project attribute data, and various model feature data, which can be accessed and queried by various functional modules and corresponding algorithms.

[0028] M3, Automatic Model Recognition Module: This module receives model data from the scanned model import module, calls the model algorithm module to perform model segmentation and recognition, performs segmentation processing on the hydropower engineering hub scene model, and identifies the category and subcategory of each model segment by combining model data and feature parameters from the model database construction module. It adds the category information to the attribute information of the corresponding model segment and outputs the tagged model segment data to the engineering model replacement module. Specifically, it includes: M31, Model Block Processing Unit: Provides a program operation and input interface for calling the model block algorithm. By calculating the differences between the model cross section and the terrain of the engineering area and the planar outline of the scene, it can divide the potential system models such as buildings, equipment, and construction machinery related to engineering construction from the complete engineering hub scene model into surface ranges in the large scene model, and package and reorganize the model surfaces within the corresponding range into model blocks for subsequent functional modules to perform further identification and replacement operations. Specifically, the model segmentation algorithm supports the application functions of the model segmentation processing submodule. Through the algorithm flow, it segments the potential buildings, structures, equipment, construction machinery, and other models within the engineering hub scene model into blocks, which are then used by subsequent algorithm flows to analyze and identify each module. (Refer to...) Figure 2 The specific algorithm execution flow is as follows: M311. Import the full-hub scene scanning model of the hydropower project and the terrain model of the project area into the 3D visualization engine architecture, and complete the 3D spatial superposition and overlap of the two models through geographic information coordinates; Specifically, the application interface provided by the model automatic recognition module imports the full-hub scene model of the hydropower project and the terrain model of the project area scanned on site into the three-dimensional visualization engine architecture of the system, and uses geographic information coordinates to make the two models superimpose and overlap in three-dimensional space.

[0029] M312, Adjust the model coordinate system, The XY plane is set as the ground plane. The scanned hydropower project's entire hub scene model and the terrain model of the project area are then compared in three-dimensional space, parallel to... The shaft is designed according to precision requirements. The axis coordinates are cut multiple times; Get multiple of the current scene The cutting planes are marked with the outlines of the entire hub scene model and the terrain model of the engineering area.

[0030] M313. Use the contour line difference judgment formula to judge the difference between the two model contour lines in each section plane, and mark the contour line of the hydropower project hub scene model that has obvious differences from the terrain contour line. Specifically, for each The outlines of the whole hub scene model and the engineering area terrain model in the cross section are marked, and the formula is used to judge whether there are obvious differences between the outlines of the two models in each cross section. Let X take a fixed value Taking the cutting surface as an example, the specific formula for judgment is as follows:

[0031] Among them, the function for The corresponding y-values ​​on the contour line obtained by cutting the full-hub scene model in the section plane Value; Function for The corresponding y-values ​​on the contour line obtained by cutting the terrain model of the engineering area in the section plane Value; n is the value set according to the precision requirements for each Along the minimum line segment detection range of the cutting surface The number of values ​​that are evenly divided across the axis; Within the minimum line segment detection range, each Axis values, For the first value, This is the last value taken. The accuracy coefficient for the difference between the scene and terrain outlines; The judgment formula shows that within the minimum line segment detection range, if... Sum of differences and If the difference exceeds a certain percentage required for discrimination accuracy, it is considered that the outline of the entire hub scene model and the terrain model of the engineering area differs significantly. Similarly, various values ​​of Y can be obtained. Formula for judging the difference of model contour lines in the cross section.

[0032] When the above judgment formula requirements are met, the line segment is marked as a full-hub scene model outline that differs significantly from the terrain outline. When the above judgment formula requirements are not met, the next outline segment is judged, and the judgment operation in step M313 is repeated for all... The differences in the contour lines in the cross-section are used to make judgments.

[0033] M314. Organize the marked contour lines in three-dimensional space, connect the contour lines at intervals to generate curved surfaces, wrap the corresponding terrain curved surfaces with the curved surfaces to form three-dimensional entities, and package them into model blocks. Specifically, each The outlines of each hub scene model that are identified as having significant differences in the cross-section are organized and displayed in 3D space. The outlines are connected to each other at intervals to generate surfaces. The generated surfaces are used to wrap the terrain surfaces corresponding to the terrain models of the engineering area within the range, forming 3D solids and packaging them into model blocks. Steps M311-M313 are repeated to package all parts in 3D space that differ from the terrain model into preliminary model blocks. This method performs preliminary identification, classification, and packaging of scanned model patches of key buildings, structures, construction machinery, electromechanical equipment, etc., that are different from the terrain surface.

[0034] M315. Repeat steps M311 to M314 to package all parts in the three-dimensional space that differ from the terrain model into preliminary model blocks. M316. The scene plane is judged and filtered for the entire hub scene model. The model is divided into grids according to the accuracy. The coordinates of the center point of the grid are read. The minimum detection range of the model surface area is set. The detection range is determined by the relationship between the four coordinate line segments. The detection results are marked. The judgment of all minimum surface areas is completed in a loop. Specifically, scene planes are identified and screened within the full-hub scene model. These scene planes are portions of the hydropower project's full-hub scene model that can be approximated as planes. Due to the mechanisms of various model scanning equipment and the actual conditions at the engineering site, the actual scanned models cannot contain the absolute or theoretical planes required for the engineering design. Therefore, it is necessary to determine whether the scanned model is a scene plane. When model faces within a certain spatial range exhibit characteristics of approximating a unified plane in geometry, these model faces are considered to be within the same scene plane. The scene plane identification method is as follows: Based on the accuracy, the model's faces are meshed, and the coordinates of the center points of each mesh are read. According to the input discrimination accuracy requirements, the minimum detection range of the model's face area is set. Within this range, the coordinates of three mesh center points are randomly selected, and each mesh center point is then obtained as the fourth point. Line segments are formed between each pair of points based on their coordinates. It is determined whether the line segments formed by the four points are parallel or intersecting within the set error accuracy range in three-dimensional space. If they are, the detection range of the model's face area is considered to be in the same plane; if not, it is considered that the detection range of the model's face area is not in the same plane. After marking the detection result of the model's face area, the detection and judgment of whether the next adjacent minimum face area is in the same scene plane is performed repeatedly until all minimum face areas are judged, and then step M317 is performed.

[0035] M317. For the smallest face regions marked as being on the same plane, check whether their adjacent face regions are on the same scene plane. Add a unified scene plane number to the face regions on the same plane. Repeat this process to determine and mark all the smallest face regions and their adjacent regions. Specifically, for the minimum patch area detection range marked as the monitoring area in the same scene plane in step M317, the minimum detection range in the adjacent model patches is detected again. Four random points distributed in two different model patch minimum detection ranges are randomly selected. According to the method in step M317, it is determined whether these four points are in the same scene plane. If they are in the same scene plane, the corresponding scene plane number is added to the two adjacent minimum patches. The detection range of the next adjacent minimum patch area (and not the minimum patch detection area that was judged in step M316 to be no longer in the same plane) is judged in a loop. If they are not in the same scene, the judgment of the scene plane of that area ends. In step M317, the next smallest patch detection area that has been marked as being in the same scene plane in step M316 is judged to see if it is in the same scene plane as its neighboring detection areas. This process continues until all the smallest independent detection areas and their adjacent areas are judged and marked accordingly. Then, step M318 is performed.

[0036] M318. Organize the judgment results, organize the model faces of the same plane according to the scene plane number, and calculate the expression formula of each scene plane within the allowable error accuracy range. Specifically, the judgment results of steps M316 and M317 are sorted out, and the model patches in the same scene plane are sorted out according to the scene plane number to which each minimum detection range belongs. After sorting out, the expression formula of each scene plane that is approximately regarded as a plane within the allowable range of error accuracy is calculated, and step M319 is performed.

[0037] The scene plane classification obtained by this method will mark all adjacent regions belonging to the same scene plane, but it will bypass abnormally convex or concave model planes in adjacent planes that cannot be classified into the current model plane. Subsequent algorithm processes will identify the differences in the model in the scene plane.

[0038] M319. Read the calculation plane expression formulas corresponding to each scene plane, and construct the corresponding [formula] for the current scene plane. The coordinate axes are based on the direction of the scene plane normal. The axis is the scene plane. Value 0 flat; M3110, Based on minimum coordinate accuracy, along... The shaft is divided into multiple segments. , Planar sectioning: Mark the model outline and plane axis in each sectioning plane; determine the difference between the model outline and plane axis using a difference judgment formula; repeat the process to determine the difference between the outlines of all sectioning planes. by Take a fixed value Taking the cross-section as an example, the formula for judging differences is as follows:

[0039] Among them, the function For construction based on the current scene plane , , In coordinate system Take a fixed value In the cross-section, the contour lines obtained by cutting the entire hub scene model are different. The corresponding value Value; n is set according to the precision requirements for each Along the minimum line segment detection range of the cutting surface The number of values ​​that are evenly divided across the axis; The values ​​for each Y-axis within the minimum line segment detection range are: For the first value, This is the last value taken. This is the accuracy coefficient for the difference in the outline of the corresponding model under the current scene plane; The difference judgment formula shows that within the minimum line segment detection range, if each... The sum of the absolute values ​​of the values If the difference exceeds a certain percentage required for discrimination accuracy, it is considered that the model's contour line differs significantly from the scene plane's section line within the entire hub scene model. Similarly, the difference can be obtained in the constructed coordinate system. Each of the fixed values Formula for judging the difference between the model outline and the scene plane in the section plane.

[0040] When the above judgment formula requirements are met, the line segment is marked as a full-hub scene model outline that has a significant difference from the corresponding scene plane section line. When the above judgment formula requirements are not met, the next outline line segment is judged, and the judgment operation in step M3110 is repeated in the coordinate system constructed for all scene planes. , The differences in the contour lines in the cross-section are used to make judgments.

[0041] M3111: Organize the marked contour lines in three-dimensional space, connect the contour lines at intervals to generate curved surfaces, wrap the corresponding scene planes with the curved surfaces to form three-dimensional entities, package them into model blocks, and repeat the block operation of all model facets that are different from the plane in the scene plane. Specifically, each , The outlines of each hub scene model that are identified as having significant differences in the cross-section are organized and displayed in 3D space. The outlines are then connected to each other at intervals to generate surfaces. The corresponding scene planes within the generated surfaces are then wrapped to form 3D solids. These are packaged into model blocks. Steps M319 to M3111 are repeated until the model block operation is completed for all model faces that are different from the planes in all scene planes. Then, step M3112 is performed (including model faces that have already undergone preliminary model block operation in steps M314 and M315).

[0042] This method performs preliminary identification, classification, and packaging of scanned model patches of various buildings, structures, construction machinery, electromechanical equipment, etc., placed on the plane within the scene of the engineering hub scene model; For example, buildings and facilities on the ground of the owner's and construction party's camp, buildings and equipment on the dam top, various construction machinery on the construction leveling site, equipment inside the factory and buildings and on each floor, and various equipment installed on the structural walls or roof can be identified. Any model body that can be regarded as an extension of the model plane part in the engineering hub scene model can be identified, and it can be used to further separate the model structure and internal substructure to be identified from the scene model and perform block processing.

[0043] M3112 organizes all model blocks into 3D entities by level and structure, and outputs them to the engineering model replacement module.

[0044] M32, Model Block Recognition Unit: Provides a program operation and input interface for calling model recognition algorithms and data in the model database. It analyzes and recognizes each model block generated by the previous functional modules one by one. Through the algorithm, it evaluates and judges the feature indicators of each model block, automatically identifies and marks the category information of hydropower engineering hub buildings, equipment, construction machinery, construction equipment and vehicles corresponding to each model block in the large scene model, obtains the category and component name of each model block and its internal sub-part module blocks, and adds them to the attribute information of each model block.

[0045] Specifically, the execution flow of the model recognition algorithm is as follows: M321: Read the model blocks and disassemble and isolate them according to their internal hierarchical relationships to prepare for the recognition operation; Specifically, each model block of each preceding process is read, and each block is decomposed and isolated according to the result relationship within the block, in preparation for recognition in subsequent algorithm processes; M322. Measure the volume and control dimensions of the isolated model blocks, call the basic dimension feature parameters and basic volume feature parameters of the model database, perform preliminary category screening of the model blocks through the dimension and volume screening formula, and record the results. Specifically, the isolated model blocks and their sub-model blocks are read, and the volume and control dimensions of the model blocks are measured in three-dimensional space. The basic dimensional and volume characteristic parameters of each category are read from the model database, and a preliminary screening is performed using a size and volume screening formula to initially identify the possible corresponding hydropower engineering model categories. The size and volume screening formula is as follows:

[0046] in, , , The maximum control dimensions for the length, width, and height of the current model block are measured. For the corresponding model category to be judged, the maximum control size data of length, width, and height corresponding to the basic size feature parameters in the model database; This refers to the volume data obtained by measuring the current model in three-dimensional space. These are the basic volume feature parameters of the corresponding model to be discriminated; Adjustment factor for size determination; Volume judgment adjustment coefficient; To initially determine the accuracy requirement coefficient for the current model category; , , and The values ​​are matched and the differences are calculated in a way that minimizes the differences between each size. The accuracy requirement for the values ​​is related to the size, importance, and complexity of the model category.

[0047] After measuring the current block model size and volume parameters, read the basic size and volume feature parameters of each type of model from the model database, and calculate and judge each type by substituting them into the size and volume screening formula. If the size and volume screening formula is satisfied, it is initially screened as the model type that the current model block may correspond to. Record the possible corresponding types of the current model block. After repeating step M322 to perform preliminary screening, identification and recording of all model blocks, proceed to step M323.

[0048] M323: Retrieve the structural feature parameter group and color feature parameter corresponding to the preliminary screening category, measure various parameters of the model block and its internal sub-model block, accurately identify the category of the model block through the multi-feature weighted judgment formula, and label the identification results into the attribute information of the model block; Specifically, after obtaining the initial model category range, the structural feature parameter groups and color feature parameter data of possible categories are read from the model database. For each judgment category, various parameters of the model block and its internal sub-model blocks are read or measured. The weighted formula for the corresponding category is used for calculation and judgment. During this process, the potential feature parameter data of each model block are combined and measured from multiple angles. Priority is given to selecting category feature data from the model database that are similar to the measured data combination relationship of the current model block for formula judgment. The multi-feature weighted judgment formula is as follows:

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055] in, The difference in the number of identifiable model substructure blocks in the current model block; The number of sub-model blocks that can be divided into internally by the model partitioning algorithm for the current model block; This represents the number of internal substructure features corresponding to the current model category to be identified, extracted from the model database. The difference in the number of faces of the main structural model in the current model block; The number of faces of the main structural part of the model obtained by measuring the current model in blocks; This refers to the number of surface features of the main structural part of the model corresponding to the current model category to be identified, extracted from the model database. The squared difference of the main structural dimensions of the current model block; The dimensions of the main structural parts are generally the length, width, and height of the main body of the model. For special model structures, the corresponding main structural dimensions such as diameter, difference between long and short sides, and distance between parallel surfaces may also be indicated. This refers to the main feature structure size feature parameter data corresponding to the current model category to be identified, extracted from the model database; Scan the textures and material color differences for the main body of the model; Scan the RGB parameter values ​​of the main colors of the texture map for the main body of the current model block; The RGB parameter values ​​of the main structural features corresponding to the current model category to be identified are extracted from the model database; This is the sum of the differences in the number of faces of each substructure of the model; This represents the number of faces of a certain substructure model within the current model block during the calculation process. To extract the corresponding substructure model surface feature parameters from the model database for the current model substructure to be identified; This is the sum of the squared differences of the main dimensions of each substructure block in the model; Control the dimensions of each substructure part of the current model; The current model category to be identified is extracted from the model database, and the corresponding substructure control size feature parameter data is calculated by matching it with the measurement data according to the minimum difference. The total number of sub-model structures contained in the corresponding feature parameter category to be judged in the current model block; The weighted judgment weights for each comparison difference are set from 1 to 6, decreasing sequentially, to express the importance of the weight ratio in the model block recognition judgment; The coefficients are used to determine the required accuracy for classifying blocks in the current model. The accuracy requirement is related to the difficulty of judging the model category, the complexity of the model, and the number of substructures.

[0056] If the weighted score is greater than the accuracy requirement, the model block is considered not to belong to the category corresponding to the currently selected feature parameter group data, and the calculation continues to select other possible categories of feature parameter groups from the model database. If the weighted score is less than the accuracy requirement, the model block is considered to belong to the category corresponding to the currently selected feature parameter group data, and the annotation is filled in in the model block attribute information before continuing to step M324. If the accuracy requirement cannot be met after calculating the feature parameters of all possible categories, the operator shall perform operations such as accuracy adjustment, supplementing or modifying new category feature parameters, and modifying the model block area. After completion, the weighted calculation and identification judgment in step M323 shall be repeated until the formula calculation and accuracy value requirements are met.

[0057] M324, the loop operation completes the identification and attribute labeling of all model blocks in the hydropower project hub scene model, which can be called by subsequent modules; Specifically, the current model block and its corresponding category or subcategory are recorded. Steps M321 to M324 are repeated to perform weighted calculations and identification on all model blocks in the entire engineering hub model scene and mark the corresponding information in the attribute information of each model block for subsequent application functions and algorithm processes of the model replacement module.

[0058] M4, Engineering Model Replacement Module: This module receives the tagged model block data output by the automatic model recognition module, calls the model algorithm module to perform model generation and replacement, combines the model data from the model database construction module, generates matching lightweight blocks based on the category information of the model blocks, replaces the original model blocks in the hydropower engineering hub scene model with the lightweight blocks, performs stylistic adjustments on the lightweight blocks, and outputs the replaced model results to the model display module; specifically including: M41, Model Generation Unit: Provides a program interface for calling the model generation algorithm, measuring the key dimensions of each part of the model blocks whose category and model have been determined, searching the corresponding standardized model in the model database, filling in the measurement parameters, driving the model structure to generate the corresponding size model, and finally confirming it with the operator, so that subsequent modules can replace it in the whole engineering hub scene model. Specifically, the execution flow of the model generation algorithm is as follows: M411: Read each model block sequentially, and read the category information of the model block; M412. Based on the model block category, retrieve the corresponding main body and substructure categories, control size range, key size range, and feature parameter group data range from the model database; M413. Based on the model recognition results, measure and read the feature dimensions or parameter group data in the corresponding structural parts of the model blocks. M414. Retrieve the standard driveable parameterized model of the corresponding category from the model database. Take the data obtained by measuring and reading the feature parameters of each structural part in the current model block and fill it into the standard driveable parameterized model of the corresponding category by calling the model feature parameterization submodule. Generate the corresponding lightweight standard model through parameter driving. M415, loop through steps M411 to M414 to complete the measurement and reading of feature dimensions or parameter group data for all model blocks, and generate the corresponding lightweight standard model for subsequent model replacement submodules and algorithm flow calls.

[0059] M42, Model Replacement Unit: Used to call the model replacement algorithm to replace the original model blocks in the hydropower engineering hub scene model with the lightweight version; Specifically, the program's interactive interface calls a model replacement algorithm to replace the identified model blocks in the entire engineering hub scene model with standardized, driveable parameter BIM models generated by the model generation submodule. This reduces the number of faces in the entire engineering hub scene model and improves the overall model's flexibility and utilization efficiency. The execution flow of the model replacement algorithm is as follows: M421. Based on the model recognition sub-process and model recognition algorithm flow, filter and analyze the data of each feature parameter group of the current model block and calculate the results. At the same time, refer to the main structure positioning standard when the corresponding lightweight standard model is created, identify the corresponding main structure in the model block, and measure and read the spatial positioning coordinates of the main structure of the current model block. M422. In the hydropower project's entire scene model, delete the model facets contained in the current model block. When reading the model block algorithm, read the terrain surface or scene plane where the current model block is located. Extend along the original terrain surface or scene plane to fill in the missing model facets after the model block is deleted, and assign model materials to the missing parts based on the adjacent model facets. M423. Place the corresponding lightweight standard model generated through parameter-driven operation at the positioning coordinates of the main structure of the original model block. M424, repeat steps M421 to M423 to complete the coordinate reading of all model blocks, model deletion, terrain or scene plane processing and model placement and replacement, and organize, update and replace the completed hydropower project hub scene model in three-dimensional space for the model display module to summarize and display and other systems to call.

[0060] M43, Model Adjustment Unit: Used to replace the non-parametrically driven parts generated in lightweighting with a model structure that conforms to the style of engineering architecture and equipment.

[0061] For example, this submodule can be used to replace the architectural style and design of the model structure of the rooftop, structural columns, and other parts of the project owner's office building; it can also be used to replace the colors of the project equipment, some additional auxiliary functional structures, and special appearance structures. This submodule function can make the models generated by the model replacement module conform to the unified design style of the project, avoiding the need to modify each model one by one.

[0062] M5, Model Display Module: Used to receive the replaced model results output by the Engineering Model Replacement Module, carry and display the hydropower engineering hub scene model, generate the model structure tree, and provide data interfaces for external engineering system platforms to call model-related results and data; Specifically, the system utilizes a 3D visualization engine architecture to host and display the identified, replaced, and lightweighted model files of the entire hydropower project. The complete project model can be fully displayed in the display module, and a model structure tree can be generated based on the information obtained by the automatic model identification module, according to the category and subcategory of each model block.

[0063] The system allows users to view standardized models obtained by replacing corresponding model blocks using a structure tree. It also provides an interface for querying engineering and geometric information of the models and for parametrically modifying them. Furthermore, it provides a data interface for other engineering system platforms to call and display the results and data of automatic identification and replacement of hydropower engineering hub scene scanning models.

[0064] In addition, the 3D visualization engine architecture is the underlying program architecture for various basic 3D visualization functions such as graphic 3D space display, model visualization, attribute information carrying and association. It is used to create and support the 3D space required by the above-mentioned functional modules and algorithm processes, and provides various basic analysis and calculation functions related to image measurement, data analysis, spatial calculation, etc.

[0065] M6, Model Algorithm Module: Provides algorithmic support for the model database construction module, model automatic recognition module, and engineering model replacement module, enabling algorithmic functions such as model feature parameter extraction, model segmentation, model recognition, model generation, and model replacement.

[0066] Example 2 The automatic identification and replacement method for hydropower engineering hub scene scanning models proposed in this invention employs the automatic identification and replacement system for hydropower engineering hub scene scanning models proposed in this invention, and may specifically include: S1. Construct a hydropower engineering model database: collect hydropower engineering BIM models, extract the feature parameters of the BIM models, convert the BIM models into standard driveable parametric models, construct and store a classified hydropower engineering model database, and provide model data and feature parameter query and call for subsequent steps; S2. Scanned Model Import and Processing: Import the scanned hydropower project hub scene model and the terrain model of the project area into the 3D visualization engine architecture. Read, mark, organize and visualize the imported model data, and output the processed model data. S3. Model Segmentation and Category Recognition: Receive the processed model data output from step S2, segment the hydropower engineering hub scene model, combine the model data and feature parameters in the model database constructed in step S1, identify the category and sub-category of each model segment, add the category information to the attribute information of the corresponding model segment, and output the marked model segment data. S4. Model generation and replacement processing: Receive the marked model block data output in step S3, combine it with the model data in the model database constructed in step S1, generate a matching lightweight standard model based on the category information of the model blocks, replace the original model blocks in the hydropower engineering hub scene model with the lightweight standard model, perform stylistic adjustments on the lightweight standard model, and output the replaced model results. S5. Model Display and Data Output: Receive the replaced model output from step S4, carry and display the hydropower engineering hub scene model, generate a model structure tree, and provide a data interface for external engineering system platforms to call the model-related results and data.

[0067] The above specific embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A system for automatic identification and replacement of scene scanning models in a hydropower project hub, characterized in that, include: Scanned Model Import Module: This module imports the scanned hydropower engineering hub scene model and engineering area terrain model from the engineering site into the 3D visualization engine architecture. It reads, marks, organizes, and visualizes the model data, and outputs the processed model data to the automatic model recognition module. Model database construction module: used to collect BIM models of hydropower projects, extract model feature parameters, convert BIM models into standard driveable parametric models, build and store a classified hydropower project model database, and provide model data and feature parameter query and call for the model automatic recognition module and the engineering model replacement module; Automatic Model Recognition Module: This module receives model data from the scanned model import module, performs model segmentation and recognition, segments the hydropower engineering hub scene model, identifies the category and sub-category of each model segment by combining the model data and feature parameters from the model database construction module, adds the category information to the attribute information of the corresponding model segment, and outputs the marked model segment data to the engineering model replacement module. Engineering Model Replacement Module: This module receives the tagged model block data output by the automatic model recognition module, performs model generation and replacement, combines the model data from the model database construction module, generates matching lightweight blocks based on the category information of the model blocks, replaces the original model blocks in the hydropower engineering hub scene model with the lightweight blocks, performs stylistic adjustments on the lightweight blocks, and outputs the replaced model results to the model display module. Model Display Module: This module receives the replaced model output from the engineering model replacement module, carries and displays the hydropower engineering hub scene model, generates a model structure tree, and provides a data interface for external engineering system platforms to call model-related results and data.

2. The system according to claim 1, wherein, The scanning model import module includes: Scanning Model Import Unit: Equipped with a data conversion interface, it is used to read and import engineering 3D models acquired by UAVs or 3D scanning equipment, and supports reading and conversion of mainstream model file formats; Model data reading unit: used to read, mark and integrate the basic data of the imported model, realize model naming and numbering, integrate the scanned model according to geographic coordinate information and calculate the model's volume, number of faces, and basic geometric information. Scanning Results Carrying Unit: Used to call the 3D visualization engine architecture to visualize and carry the engineering site scanning model and related data.

3. The system according to claim 1, wherein, The model database construction module includes: Engineering Model Input Unit: Used to read and input BIM models of hydropower projects, parse the model files and import them into the 3D visualization engine architecture, and parse and associate the model's category, geometric data and engineering parameter information; Model Feature Extraction Unit: Used to call the model feature parameter extraction algorithm to extract the volume, control dimensions, geometric structure, component surfaces, and color feature data of the BIM model; Model parametric unit: Used to parametrically modify the main structural dimensions of the BIM model of hydropower project and convert it into a standard driveable parametric model; Model data construction unit: used to organize model and feature data, build the model database structure tree system and model data storage mapping relationship, and integrate and sort various types of model and parameter data.

4. The system according to claim 3, wherein, The execution flow of the model feature parameter extraction algorithm is as follows: Enter the system models of each sub-specialty of hydropower engineering with known categories and attribute information parameters; Screen and classify similar models, measure the control dimensions and volume of the models, calculate the mean of the model control dimensions samples under the corresponding category and model as the basic size feature parameter, and calculate the mean volume as the basic volume feature parameter; Read the structural feature information of the model and calculate the mean under the corresponding category to form a structural feature parameter group for the main category and sub-category of the model. The structural feature information includes the number of sub-categories, the number of model faces, and the key dimensions of the structure. Read the RGB color parameter values ​​of the model, eliminate samples with large differences, and calculate the mean, which is used as the color feature parameter of the corresponding category model; Organize the basic size feature parameters, basic volume feature parameters, structural feature parameter groups, and color feature parameters of each category and subcategory model, and store them in the model database.

5. The hydroelectric project hub scene scanning model automatic identification and replacement system of claim 1, wherein, The automatic model recognition module includes: Model Block Processing Unit: Used to call the model block algorithm, calculate the differences between the model cross section and the terrain of the engineering area and the planar outline of the scene, divide the surface range of the model block from the hydropower engineering hub scene model, and pack and reassemble the corresponding model surfaces into model blocks; Model block recognition unit: It is used to call the model recognition algorithm and the data in the model database to analyze and recognize the model blocks one by one, automatically identify and mark the category information of the hydropower project hub buildings, equipment and construction machinery corresponding to the model blocks, and add the category and component name to the attribute information of the model blocks.

6. The hydroelectric project hub scene scanning model automatic identification and replacement system of claim 5, wherein, The execution flow of the model block division algorithm is as follows: The full-hub scene scanning model of the hydropower project and the terrain model of the project area are imported into the 3D visualization engine architecture, and the 3D spatial superposition and overlap of the two models are completed through geographic information coordinates; Adjust the model coordinate system, using the XY plane with Z=0 as the ground plane, and perform multiple sections on the two models parallel to the Z-axis according to the X and Y-axis coordinates to obtain the YZ and XZ section planes and mark the corresponding section outlines of the two models. The contour line difference judgment formula is used to determine the difference between the contour lines of the two models in each section plane, and the contour lines of the scene model that are significantly different from the terrain contour lines are marked. Organize the marked contour lines in three-dimensional space, connect the contour lines at intervals to generate curved surfaces, wrap the corresponding terrain curved surfaces with the curved surfaces to form three-dimensional entities, and package them into model blocks. Package all parts in the three-dimensional space that differ from the terrain model into preliminary model blocks; The entire hub scene model is judged and filtered for scene planes. The model is divided into grids according to the accuracy. The coordinates of the center point of the grid are read. The minimum detection range of the model's surface area is set. The detection range is determined by the relationship between the four coordinate line segments. The detection results are marked. The judgment of all minimum surface areas is completed in a loop. For the smallest patch region marked as being on the same plane, check whether its neighboring patch regions are on the same scene plane. Add a unified scene plane number to the patches on the same plane. Repeat this process to determine and mark all the smallest patch regions and their adjacent regions. Organize the judgment results, organize the model facets on the same plane according to the scene plane number, and calculate the expression formula of each scene plane within the allowable error accuracy range; Read the scene plane expression formula, for the current scene plane to build the corresponding Coordinate axis, with the scene plane normal direction as Axis, with the scene plane as The value of 0 Plane; along axis to perform multi-section , plane section, in each section, mark the model contour line and the plane axis, determine the difference between the model contour line and the plane axis through the difference judgment formula, and complete the contour line difference judgment of all sections in a loop. Organize the marked contour lines in three-dimensional space, connect the contour lines at intervals to generate curved surfaces, wrap the corresponding scene planes with the curved surfaces to form three-dimensional entities, package them into model blocks, and repeat the block operation of all model facets that are different from the plane in the scene plane. All model blocks are organized hierarchically and structurally into 3D entities, and then output to the engineering model replacement module.

7. The system according to claim 5, wherein, The execution flow of the model recognition algorithm is as follows: Read the model blocks and decompose and isolate them according to their internal hierarchical relationships to prepare for the recognition operation; Measure the volume and control dimensions of the isolated model blocks, call the basic size feature parameters and basic volume feature parameters of the model database, perform preliminary category screening of the model blocks through the size and volume screening formula, and record the results; Retrieve the structural feature parameter group and color feature parameter corresponding to the preliminary screening category, measure various parameters of the model block and its internal sub-model block, accurately identify the category of the model block through the multi-feature weighted judgment formula, and label the identification results into the attribute information of the model block; The loop operation completes the identification and attribute labeling of all model blocks in the hydropower project hub scene model, which can then be used by subsequent modules.

8. The hydroelectric project hub scene scanning model automatic identification and replacement system of claim 1, wherein, The engineering model replacement module includes: Model generation unit: Used to call the model generation algorithm, measure the key dimensions of the model blocks of the identified categories and models, search for the corresponding standardized model from the model database and fill in the measurement parameters, and drive the generation of lightweight models with matching dimensions; Model replacement unit: Used to call the model replacement algorithm to replace the original model blocks in the hydropower project hub scene model with lightweight ones; Model Adjustment Unit: Used to replace the non-parametric driven parts generated in lightweighting with a model structure that conforms to the style of engineering architecture and equipment.

9. The automatic identification and replacement system for the scanning model of a hydropower engineering hub scene according to claim 8, characterized in that, The execution flow of the model generation algorithm is as follows: Read each model block sequentially, and read the category information of the model block; Based on the model block categories, retrieve the corresponding main body and substructure categories, control size ranges, key size ranges, and feature parameter group data ranges from the model database; Based on the model recognition results, measure and read the feature dimensions or parameter group data in the corresponding structural parts of the model blocks; Retrieve the standard, driverable parametric model of the corresponding category and model from the model database, fill in the feature data read from the measurement, and generate a lightweight model through parameter-driven generation; The loop operation generates lightweight versions of all model blocks for use by subsequent modules and algorithms.

10. The automatic identification and replacement system for the scanning model of a hydropower engineering hub scene according to claim 8, characterized in that, The execution flow of the model replacement algorithm is as follows: Based on the model recognition results and the lightweight main structure positioning standard, the main structure of the original model blocks is identified, and its spatial positioning coordinates are measured and read. In the scene model of the hydropower project hub, delete the model facets of the original model blocks, read the terrain surface or scene plane where the original model blocks are located, fill in the missing model facets along the extension trend of the surface / plane, and assign model materials to the missing parts of the model according to the adjacent model facets. Place the corresponding lightweight component at the spatial positioning coordinates of the main structure of the original model block; The loop operation completes the replacement of all model blocks, organizes and updates the replaced hydropower project hub scene model in three-dimensional space, and provides it for display by the model display module and for external system calls.

11. A method for automatic identification and replacement of scanning models of hydropower engineering hub scenes, applied to the automatic identification and replacement system for scanning models of hydropower engineering hub scenes as described in any one of claims 1-10, characterized in that it includes: S1. Construct a hydropower engineering model database: collect hydropower engineering BIM models, extract the feature parameters of the BIM models, convert the BIM models into standard driveable parametric models, construct and store a classified hydropower engineering model database, and provide model data and feature parameter query and call for subsequent steps; S2. Scanned Model Import and Processing: Import the scanned hydropower project hub scene model and the terrain model of the project area into the 3D visualization engine architecture. Read, mark, organize and visualize the imported model data, and output the processed model data. S3. Model Segmentation and Category Recognition: Receive the processed model data output from step S2, segment the hydropower engineering hub scene model, combine the model data and feature parameters in the model database constructed in step S1, identify the category and sub-category of each model segment, add the category information to the attribute information of the corresponding model segment, and output the marked model segment data. S4. Model generation and replacement processing: Receive the marked model block data output in step S3, combine it with the model data in the model database constructed in step S1, generate a matching lightweight standard model based on the category information of the model blocks, replace the original model blocks in the hydropower engineering hub scene model with the lightweight standard model, perform stylistic adjustments on the lightweight standard model, and output the replaced model results. S5. Model Display and Data Output: Receive the replaced model output from step S4, carry and display the hydropower engineering hub scene model, generate a model structure tree, and provide a data interface for external engineering system platforms to call the model-related results and data.

Images