Three-dimensional modeling method and system for an elevator shaft

Abstract

The application provides a three-dimensional modeling method and system for an elevator shaft, relates to the technical field of three-dimensional modeling, and comprises the following steps: after stripping the building information of the shaft, a shaft basic model is constructed; after collecting real-time shaft point clouds, K-layer shaft segmentation point clouds are obtained by cutting; P modeling attention point identifiers of P synchronous segmentation point clouds in the first layer shaft segmentation point clouds are extracted, and spatial modeling is performed in parallel to obtain a first layer sub-component model; until the K-layer shaft segmentation point clouds are iteratively modeled and assembled, a shaft sub-component assembly model is obtained; after the shaft basic model is projected to the shaft sub-component assembly model, model fusion filling is performed through difference area replacement, and a shaft three-dimensional model is output. The application solves the technical problem in the prior art that three-dimensional modeling is often based on design drawings or engineering drawings, accurate capture of actual space data of the shaft is lacked, the modeling precision of the shaft is low, and in turn, docking errors occur during actual construction, thereby causing safety hazards.

Description

Technical Field

[0001] This invention relates to the field of 3D modeling technology, and more specifically to a 3D modeling method and system for elevator shafts. Background Technology

[0002] As a special part of a building structure, the 3D modeling of elevator shafts not only needs to meet the accuracy of spatial structure, but also needs to handle complex structural components and intricate assembly relationships. Traditional methods often rely on design drawings or engineering drawings for 3D modeling, which lacks accurate capture of the actual spatial data of the elevator shaft. Although BIM models can provide relatively accurate design information, the lack of real-time shaft data results in low modeling accuracy. This leads to misalignment errors during the installation of various shaft components during construction and assembly, affecting project quality and construction progress. It can even cause the shaft structure to fail to properly connect with the elevator equipment after construction, creating serious safety hazards. Summary of the Invention

[0003] This application provides a three-dimensional modeling method and system for elevator shafts, aiming to solve the technical problem that existing technologies often rely on design drawings or engineering drawings for three-dimensional modeling, which lacks accurate capture of the actual spatial data of the shaft, resulting in low shaft modeling accuracy and consequently causing docking errors during actual construction, thus creating safety hazards.

[0004] The first aspect disclosed in this application provides a three-dimensional modeling method for elevator shafts. The method includes: after extracting shaft building information from local architectural design information, constructing a basic shaft model of the elevator shaft to be modeled based on the shaft building information; after collecting real-time shaft point clouds of the elevator shaft to be modeled, obtaining K-layer shaft segmentation point clouds by cropping from the real-time shaft point clouds based on the spatial dependency features of the shaft building; extracting P modeling interest markers from P synchronous segmentation point clouds in the first-layer shaft segmentation point cloud, and performing spatial modeling on the P synchronous segmentation point clouds in parallel based on the P modeling interest markers to obtain a first-layer component model; using the first-layer component model as an assembly reference, performing component modeling of the second-layer shaft segmentation point cloud based on spatial assembly relationships, until iteratively modeling and assembling the K-layer shaft segmentation point clouds to obtain a shaft component assembly model; projecting the shaft basic model onto the shaft component assembly model, performing model fusion and filling through difference region replacement, and outputting a three-dimensional shaft model.

[0005] The second aspect of this application discloses a 3D modeling system for elevator shafts. This system is used in the aforementioned 3D modeling method for elevator shafts. The system includes: a basic model construction module, used to extract shaft building information from local architectural design information and then construct a basic model of the elevator shaft to be modeled based on the shaft building information; a segmented point cloud acquisition module, used to collect real-time shaft point clouds of the elevator shaft to be modeled and then obtain K-layer segmented point clouds from the real-time shaft point clouds based on the shaft building spatial dependency features; and a spatial modeling module, used to extract P synchronous points from the first-layer segmented point cloud of the shaft. The system identifies P modeling interest points in the segmented point cloud and performs spatial modeling on the P synchronously segmented point clouds in parallel based on the P modeling interest points to obtain a first-layer component model. The component modeling module is used to perform component modeling on the second-layer shaft segmented point cloud based on the spatial assembly relationship, using the first-layer component model as the assembly reference, until the K-layer shaft segmented point cloud is iteratively modeled and assembled to obtain a shaft component assembly model. The model fusion and filling module is used to project the shaft base model onto the shaft component assembly model, and then perform model fusion and filling by replacing the difference regions to output a 3D shaft model.

[0006] One or more technical solutions provided in this application have at least the following beneficial effects:

[0007] By extracting shaft architectural information from local building design information, the construction of the shaft model accurately reflects the actual building information. This process eliminates redundant and irrelevant information, retaining only design data related to the shaft structure. Based on the extracted shaft architectural information, a basic shaft model is constructed, providing a starting point for subsequent modeling work and ensuring that the model meets engineering requirements. Real-time shaft point cloud acquisition ensures that the modeling process reflects the actual spatial situation of the shaft, avoiding the limitations of relying solely on design drawings or static models. K-layer shaft segmentation point clouds are obtained by cropping from the real-time shaft point cloud based on the shaft architectural spatial dependency characteristics. This processing method can improve model accuracy. Parallel processing of P synchronous segmentation point clouds significantly accelerates the modeling process. Each synchronous segmentation point cloud is modeled independently, and multiple modeling containers are used for parallel processing, reducing the latency caused by serial processing in traditional modeling methods. By extracting modeling interest markers, it is ensured that key parts of the model can be highlighted during the modeling process. By focusing on and precisely processing the data, the accuracy and operability of the model are further improved. Utilizing spatial assembly relationships, the first-layer component model is used as a benchmark for modeling the second-layer shaft segmentation point cloud. Modeling and assembly are performed layer by layer, ensuring that each layer of the model can be correctly connected and matched, avoiding assembly problems caused by hierarchical misalignment or improper connection. Through an iterative process, the shaft components are processed layer by layer, ensuring that the connections between each component and the overall spatial relationship are accurately modeled. This layered modeling and assembly method can effectively avoid errors or omissions during the overall model assembly. By projecting the shaft base model onto the shaft component assembly model and replacing and filling the difference areas, the differences between the initially constructed shaft base model and the shaft component assembly model obtained through point cloud data and spatial assembly are resolved. After difference filling and model fusion, the final output shaft 3D model can accurately reflect the actual structure and functional requirements of the shaft, improving the accuracy of 3D modeling.

[0008] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of a three-dimensional modeling method for elevator shafts provided in an embodiment of this application.

[0010] Figure 2 This is a schematic diagram of a three-dimensional modeling system for elevator shafts provided in an embodiment of this application.

[0011] Figure labeling: Basic model construction module 10, segmentation point cloud acquisition module 20, spatial modeling module 30, component modeling module 40, model fusion and filling module 50. Detailed Implementation

[0012] This application provides a three-dimensional modeling method and system for elevator shafts, which solves the technical problem that existing technologies often rely on design drawings or engineering drawings for three-dimensional modeling, lacking accurate capture of the actual spatial data of the shaft, resulting in low shaft modeling accuracy, which in turn leads to docking errors during actual construction and causes safety hazards.

[0013] After introducing the basic principles of this application, various non-limiting embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0014] Example 1, as Figure 1 As shown in the figure, this application provides a three-dimensional modeling method for elevator shafts, the method comprising:

[0015] After extracting the shaft building information from the local building design information, a basic model of the elevator shaft to be modeled is constructed based on the shaft building information.

[0016] Local architectural design information refers to architectural design documents, drawings, or CAD files. Shaft architectural information is extracted from this information, including the dimensions, shape, structural hierarchy, supporting structure, and entrance and exit locations of the elevator shaft. Based on this information, modeling software, such as BIM modeling tools, is used to model the basic geometry of the shaft. This preliminary depiction of the shaft's spatial structure, hierarchy, and the location and dimensions of its various components forms the starting point for subsequent modeling.

[0017] After collecting the real-time point cloud of the elevator shaft to be modeled, the K-level shaft segmentation point cloud is obtained by cropping from the real-time point cloud based on the architectural spatial dependence features of the shaft.

[0018] Laser scanning technology was used to acquire real-time point clouds of the elevator shaft to be modeled. The point cloud data contains three-dimensional information of the shaft space, with each point having spatial coordinates (X, Y, Z) and reflecting the actual geometric features of the shaft. Within the shaft, the spatial positions of different areas and structures are interdependent, and this dependency determines the shaft's segmentation method. For example, some structural units may be spatially connected or superimposed, or certain parts may have a specific order, such as the upper and lower layers of the shaft. These factors need to be considered during the trimming process. Based on the spatial dependency characteristics of the shaft's architecture, the real-time shaft point cloud is spatially segmented into multiple levels, i.e., K-level shaft segmentation point clouds. Each level represents a specific part or area of ​​the shaft, such as the guide rails, supports, or walls of each level. The trimming process is based on the spatial characteristics of the shaft structure, such as building floor height and the distribution of supporting structures, segmenting the point cloud data hierarchically to ensure that each segmented level of the shaft's point cloud accurately represents each part of the shaft.

[0019] Extract P modeling interest identifiers from P synchronous segmented point clouds in the first layer shaft segmentation point cloud, and perform spatial modeling on the P synchronous segmented point clouds in parallel based on the P modeling interest identifiers to obtain the first layer component model.

[0020] P synchronized segmented point clouds are extracted from the first-level shaft segmented point cloud. Synchronous segmented point clouds refer to point cloud data that are spatially related or require joint modeling, such as point cloud data belonging to the same building component. These synchronized point clouds represent specific parts of the shaft, such as guide rails, support structures, or local areas of the walls, and they are geometrically connected or aligned. Each synchronized segmented point cloud is associated with a modeling concern identifier, which is a marker used to guide the modeling process. It can be an indication of geometric features, structural features, or modeling constraints of a specific area. For example, modeling concern identifiers for the main structure include scan shaft wall flatness and verticality deviation; modeling concern identifiers for the guide rail system include guide rail spacing and verticality; modeling concern identifiers for supports and corbels include spatial coordinate positioning and levelness calibration; and modeling concern identifiers for floor door openings include doorway dimensions and threshold height uniformity. These ensure consistent modeling methods for each synchronized segmented point cloud and also help identify key geometric features of each synchronized segmented point cloud, such as details of nodes, edges, or faces.

[0021] Based on P modeling interest markers, P synchronous segmented point clouds are loaded into a parallel modeling container for modeling. The parallel modeling container is a tool for efficiently processing different modeling tasks in parallel; each container can handle the modeling task of one synchronous segmented point cloud simultaneously. During modeling, a corresponding 3D model is constructed based on the geometric features of the synchronous segmented point clouds. Through parallel modeling, the first-layer component model is obtained. Each synchronous segmented point cloud generates a corresponding component model, and the P component models together constitute the first-layer model of the shaft.

[0022] Using the first layer of component model as the assembly benchmark, the second layer of shaft segmentation point cloud is modeled based on spatial assembly relationships, until the K-layer shaft segmentation point cloud is iteratively modeled and assembled to obtain the shaft component assembly model.

[0023] The first-layer component model serves as the assembly benchmark, providing a spatial framework to guide subsequent modeling and assembly work. During the modeling of the second-layer shaft segmentation point cloud, the spatial relationships between each component are identified. These spatial assembly relationships describe the relative positions, dimensional fits, and connection methods between structural units, ensuring that the second-layer component model can be correctly assembled into the first-layer component model. Based on the second-layer shaft segmentation point cloud data and the spatial assembly relationships, the second-layer components are modeled. After the second-layer component modeling is completed, further adjustments are made based on the spatial assembly relationships, ensuring correct alignment between newly added components and existing component models. During assembly, assembly deviations may occur due to modeling errors or point cloud data accuracy issues. In such cases, dynamic compensation is required to correct these deviations and ensure precise alignment between the second-layer and first-layer component models.

[0024] Following the steps above, continue modeling and assembling the point cloud of the K-2 layer shaft segmentation after the second layer. The modeling process of each layer is based on the assembly model of the previous layer as a reference, and modeling is carried out according to the corresponding spatial assembly relationship. After modeling and assembling each layer, the assembly model of the entire shaft sub-components is finally completed. This model contains all the sub-components of the K layer shaft and accurately reflects their spatial assembly relationship.

[0025] After projecting the basic model of the shaft onto the assembly model of the shaft components, the model is fused and filled by replacing the differences in the region, and the three-dimensional model of the shaft is output.

[0026] Projecting the shaft foundation model onto the shaft component assembly model means aligning and combining the geometric information of the foundation model with the spatial structure of the component assembly model. There will be some differences between the projected shaft foundation model and the shaft component assembly model. These differences are usually caused by changes in certain details, shapes, or structures during the generation process of the foundation model and the component model. For example, some components may exist in the foundation model but be missing or have different dimensions in the assembly model, or the positions of some components may be slightly different, possibly due to actual modeling or measurement errors.

[0027] For identified discrepancies, corrections are made through replacement or adjustment. During this process, parts of the shaft foundation model that do not conform to the actual structure are replaced with corresponding areas from the shaft component assembly model. This ensures that the final 3D shaft model accurately reflects the actual condition of the building shaft. After replacing the discrepancies, model fusion and filling are performed. This involves merging the data from the shaft foundation model and the shaft component assembly model to ensure seamless integration. This includes smoothing seams and boundaries to remove visual or structural discontinuities; and filling in the details of the model to ensure all areas are fully modeled without omission. After model fusion and filling, a complete 3D shaft model is finally generated. This model not only includes all the components of the shaft but also reflects their spatial assembly relationships, dimensions, and structural features.

[0028] Furthermore, the method for obtaining a K-layer segmented point cloud of the shaft from the real-time shaft point cloud based on the spatial dependency characteristics of the shaft architecture includes:

[0029] The basic model of the elevator shaft is segmented based on functional attributes to obtain multiple shaft component segmentation surfaces, each of which has a modeling concern identifier. The multiple shaft component segmentation surfaces are hierarchically associated according to the spatial dependency relationship of the shaft construction sequence to obtain K-layer shaft component segmentation surfaces. After collecting the real-time shaft point cloud of the elevator shaft to be modeled, the real-time shaft point cloud is mapped and clipped using the K-layer shaft component segmentation surfaces to generate the K-layer shaft segmentation point cloud.

[0030] Functional attributes include different functional units of the shaft, such as support structures, guide rails, shaft walls, and floor slabs. Each part has different geometric shapes and structural features, and their positions and functions in space are different. Based on these functional attributes, the basic shaft model is divided into multiple shaft component segments. Each shaft component segment represents a functional or structural unit within the shaft. For example, the support frame of one shaft floor corresponds to one segment, and the guide rails of another floor constitute another segment. On each shaft component segment, modeling concerns are identified. These modeling concerns refer to specific points, areas, or geometric features that require special attention during the modeling process. They help to accurately model and optimize each segment in subsequent modeling stages. These identifications include important geometric features such as edges, connection nodes, or areas requiring special treatment.

[0031] Based on the construction sequence of the shaft, the spatial dependencies between different shaft component sections are analyzed. These dependencies refer to the spatial order and interrelationships of different structural units. For example, some sections must be located at certain positions on other sections; some sections must be constructed before other sections; and some sections need to be connected or docked with other sections. Based on the construction sequence and spatial dependencies, multiple shaft component sections are hierarchically associated to obtain K layers of shaft component sections. The K-layer hierarchical structure represents different construction stages or areas of the shaft, including different structural layers such as guide rail layers, support layers, and shaft wall layers. Each layer contains multiple related sections, representing the functional or structural units of that layer. Each layer of shaft component sections describes different areas and structural parts of the shaft, possessing a clear spatial structure and sequence.

[0032] Real-time point cloud data of the elevator shaft to be modeled was acquired using laser scanning technology. This point cloud data contains three-dimensional information about the shaft space, with each point having spatial coordinates (X, Y, Z) and reflecting the actual geometric features of the shaft. Based on the K-level shaft segmentation surfaces, the real-time point cloud was mapped. Mapping aligns the point cloud data with these segmentation surfaces, ensuring that the point cloud region corresponding to each segmentation surface is accurately mapped to the actual structure of the shaft. During mapping, the point cloud was cropped according to the spatial regions defined by the K-level shaft segmentation surfaces. The cropped point cloud only contains the regions related to each segmentation surface, removing unnecessary parts. This ensures that each segmentation surface corresponds to a specific region or functional unit within the shaft. After mapping and cropping, K-level shaft segmentation point clouds are generated. Each layer of the shaft segmentation point cloud represents a portion of the shaft, corresponding to different building floors or functional areas, providing accurate three-dimensional data for subsequent modeling.

[0033] Furthermore, the method also includes:

[0034] The structural units corresponding to the segmented surfaces of the multiple shafts are aggregated to obtain Q standard structural units. A structural unit matrix is ​​constructed based on the Q standard structural units. Q sets of interference risk feature sets of the Q standard structural units are accessed via a network, wherein each set of interference risk feature sets consists of an interference structural unit ID and an interference risk geometric parameter threshold. Guided by the interference structural unit IDs of the Q sets of interference risk feature sets, the interference risk geometric parameter thresholds of the Q sets of interference risk feature sets are mapped and filled into the structural unit matrix to obtain a structural risk feature matrix.

[0035] Multiple shaft segments are divided based on functional attributes. Each segment represents a functional or structural unit within the shaft, such as guide rails, supports, shaft walls, or floor slabs. A shaft structural unit refers to a component with an independent function in architectural design. The structural units corresponding to the multiple shaft segments are aggregated according to their functional attributes, geometric features, and spatial location. The purpose of aggregation is to combine similar or functionally related shaft structural units into standard structural units. For example, all shaft structural units related to guide rails are aggregated into one standard structural unit, and all shaft structural units related to supports are aggregated into another standard structural unit. The aggregated structural units form Q standard structural units according to their function and spatial layout.

[0036] The structural unit matrix represents the relative relationships, spatial positions, and geometric features of Q standard structural units. This matrix contains all the information of the Q standard structural units and organizes them into a matrix form. The elements of the matrix include the position coordinates, dimensions, shape, material properties, and other information of each standard structural unit. It can also include the connection relationships, mating relationships, or relative positions between them.

[0037] Different structural units within a shaft may experience collisions or interference during assembly or operation. For example, guide rails may collide with supports, leading to an unreasonable spatial layout and affecting the normal use of the shaft. The interference risk feature set is a collection of data analyzing these potential collisions and interferences. Network access refers to dynamically acquiring the interference risk feature set associated with each standard structural unit by connecting to an external database or system. These interference risk feature sets are provided through past experience, databases, or design specifications.

[0038] Each interference risk feature set consists of an interference structural unit ID and an interference risk geometric parameter threshold. The interference structural unit ID is the ID that identifies the structural unit involved in the interference, such as the ID of a guide rail or a support. The interference risk geometric parameter threshold represents the minimum safe distance or geometric dimension between two structural units. Interference will occur if this threshold is exceeded. For example, if the distance between a guide rail and a support is less than a certain threshold, it indicates that there is a risk of collision between them.

[0039] Based on the Q-group interference risk structural feature set, the interference risk geometric parameter threshold of each interference structural unit ID is mapped to the structural unit matrix. Mapping means filling the interference risk geometric parameter threshold into the corresponding position of the structural unit matrix according to the interference structural unit ID. Through the mapping operation, the resulting structural risk feature matrix contains the interference risk information between each standard structural unit. This matrix not only shows the position and size information of the structural units, but also shows the interference risk, and clearly identifies which areas have problems through the risk threshold, providing a basis for subsequent risk assessment and optimization.

[0040] Furthermore, the method also includes:

[0041] Extract the real-time building geometric parameters of the K-1 layer associated interference shaft node group from the 3D model of the shaft; use the first associated interference shaft node group as a dual retrieval condition to locate the first association matrix node in the structural risk feature matrix coordinates; extract the first interference risk geometric parameter threshold of the first association matrix node; compare whether the first real-time building geometric parameters fall within the first interference risk geometric parameter threshold, and output the first real-time interference risk feature of the first associated interference shaft node group; and so on, perform interference risk traversal identification in the 3D model of the shaft to locate and identify multiple real-time interference risk features.

[0042] The shaft is a multi-layered structure. Each layer is not only physically interconnected but may also have functional dependencies and interferences. This step focuses on the interference analysis of layer K-1, meaning it analyzes the interference between all layers below layer K and the layer above it. Layer K itself does not interfere with other layers; therefore, the interference analysis begins from layer K-1. Related interference shaft node groups refer to key nodes in the shaft structure that may interfere with other parts. These nodes can be connection points, interface points, or any potential intersections. Interference nodes are the key objects for spatial analysis and calculation, helping to understand the spatial relationships between different components and layers. The real-time architectural geometric parameters of layer K-1 related interference shaft node groups are extracted, including the position, shape, size, and angles of the components. Extracting these real-time architectural geometric parameters provides accurate spatial data, laying the foundation for interference risk analysis.

[0043] The first correlated interference shaft node group is a subset of the K-1 layer correlated interference shaft node group. The dual search conditions refer to locating and matching relevant nodes in the structural risk feature matrix using two search criteria. Specifically, the first search condition is the K-1 layer correlated interference shaft node group, and the second search condition is the first correlated interference shaft node group. This dual search helps to accurately find relevant nodes in the structural risk feature matrix and further assess the risk. Through dual searching, the nodes of the first correlation matrix are located.

[0044] Extract the first interference risk geometric parameter threshold of the first correlation matrix node. The interference risk geometric parameter threshold represents the minimum safe distance or geometric size between two structural units. Interference will occur if this threshold is exceeded.

[0045] The first real-time building geometric parameters are compared with the first interference risk geometric parameter threshold. Specifically, if the first real-time building geometric parameters, such as location and size, fall within the range of the first interference risk geometric parameter threshold, it indicates that these structural units have an interference risk. For example, if the minimum distance between two components is less than a set safety threshold, then these two structural units are considered to have an interference risk. After comparison, the first real-time interference risk characteristics of the first associated interference shaft node group are output, including the type of interference, such as collision, size mismatch, insufficient space, etc., as well as the specific location of the interference and the severity level of the interference.

[0046] In the 3D model of the shaft, all nodes are traversed sequentially, and interference risk identification is performed in a manner similar to the steps described above. For each node, relevant real-time building geometric parameters are extracted and compared with the corresponding interference risk geometric parameter threshold. Through the traversal process, multiple real-time interference risk features are identified, and these areas are marked. Each marked area includes the structural unit of the interference, the location of the interference, the type of interference, and its severity.

[0047] Furthermore, the method involves segmenting the basic model of the shaft based on functional attributes to obtain multiple shaft segmentation surfaces, the method of which includes:

[0048] Engineering characteristics are extracted from the shaft building information to obtain multiple functional building structures; geometric boundaries of the multiple functional building structures are identified in the shaft foundation model to obtain multiple sets of structural segmentation surfaces; based on multiple standard engineering attributes of the multiple functional building structures, and using the multiple sets of structural segmentation surfaces as data extraction constraints, functional attributes of the multiple sets of segmentation surfaces are extracted from the shaft foundation model; the functional attributes of the multiple sets of segmentation surfaces are used as modeling focus identifiers and mapped and bound to the multiple sets of structural segmentation surfaces to output multiple sets of shaft component segmentation surfaces; the shaft component segmentation surfaces are grouped and split to output the multiple shaft component segmentation surfaces.

[0049] The building information of the shaft includes the dimensions, shape, structural hierarchy, supporting structure, and the location of the shaft entrance and exit. The engineering characteristics of the shaft building information are extracted. The engineering characteristics refer to the specific functions, layout and construction features of the shaft structure. For example, the shaft includes various functional units such as supporting structure, guide rails, shaft walls, floor slabs and pipes. Through the extraction of engineering characteristics, various functional building structures are obtained. Each functional building structure corresponds to a specific part or functional unit of the shaft, such as supporting structure, guide rail system, shaft walls and floor slabs.

[0050] Geometric boundaries refer to the external shape, dimensions, outline, or limits of a structural unit. For example, the boundary of a support structure is the dimensions of its bottom and top, while the geometric boundary of a guide rail is its installation position and dimensions. For multi-functional building structures, geometric boundary identification is performed in the shaft foundation model. This process involves analyzing and extracting spatial information from the building data to determine the specific location and boundaries of each functional building structure and generate corresponding structural segmentation surfaces. Structural segmentation surfaces are two-dimensional or three-dimensional surfaces that represent the geometric characteristics of a structural unit. For example, a shaft wall corresponds to one segmentation surface, while a support structure corresponds to multiple segmentation surfaces.

[0051] Standard engineering attributes refer to the engineering characteristics and parameters associated with each type of functional building structure, such as dimensions, materials, load-bearing capacity, and durability. These attributes are obtained through engineering design standards, building codes, or industry standards. For example, guide rail systems have specific dimensional and material requirements, while support structures have specific load-bearing and strength requirements. Based on existing structural partitions, standard engineering attributes are used to further extract functional attributes associated with each partition. This means that it is necessary not only to determine the location and dimensions of each partition but also to identify its structural characteristics. Based on the constraints of standard engineering attributes and structural partitions, multiple sets of functional attributes for each partition are extracted. For example, the partitions of guide rails indicate their material type and installation requirements, while the partitions of support structures identify their load-bearing capacity and installation method.

[0052] During the modeling process, the functional attributes of the segmented surfaces require special attention. These are designated as modeling focus points. For example, a particular segmented surface may require special modeling treatment, such as specific materials, dimensions, or connection methods. This information, as modeling focus points, guides the modeling process. Multiple sets of segmented surface functional attributes are used as modeling focus points and bound to the corresponding structural segmented surfaces through mapping operations. This ensures that each set of structural segmented surfaces possesses not only geometric information but also related engineering characteristics. Finally, after mapping and binding operations, multiple sets of shaft segmented surfaces are output, reflecting the spatial layout and functional characteristics of various components within the shaft structure.

[0053] The shaft's component segments are grouped and decomposed, meaning a large structural unit of the shaft is broken down into multiple smaller components for more refined modeling and analysis. The decomposition can be based on factors such as structural function, geometry, connection method, and construction sequence. For example, a section of support structure in the shaft might be divided into support columns and support beams based on function; a portion of the shaft wall might be divided into multiple regions due to different materials. After decomposition, multiple shaft component segments are output, each representing a smaller building unit within the shaft with independent geometric characteristics and functional attributes.

[0054] Furthermore, the method involves extracting P modeling interest markers from P synchronously segmented point clouds in the first-layer shaft segmentation point cloud, and performing spatial modeling on the P synchronously segmented point clouds in parallel based on the P modeling interest markers to obtain the first-layer component model.

[0055] A pre-constructed parallel modeling container cluster is provided, comprising M parallel modeling containers, the output of which is connected to an identifier keyword matching unit. P modeling interest identifiers are loaded into the M identifier keyword matching units of the M modeling containers for data input matching, locating P dynamically allocated containers, where P ≤ 2M. Based on the matching association between the P synchronous segmented point clouds and the P dynamically allocated containers, the P synchronous segmented point clouds are mapped and loaded into the P dynamically allocated containers. The parallel modeling container cluster is then started to perform spatial modeling, outputting a first-layer component model including P component models.

[0056] In complex modeling tasks, parallel processing is employed to accelerate computation and modeling speed. A parallel modeling container cluster consists of M modeling containers connected in parallel. Each modeling container is responsible for handling an independent task within the wellbore model. These containers can handle different modeling tasks simultaneously, significantly improving overall processing efficiency. Each modeling container performs its computation independently, and its output is connected to an identifier keyword matching unit. The identifier keyword matching unit's task is to match and identify the modeling results based on modeling interest identifiers. It can identify key features in the output results of different modeling containers and match them with known identifiers.

[0057] P modeling concern identifiers are loaded into each identifier keyword matching unit in the parallel modeling container cluster. These identifier keyword matching units identify and match relevant modeling tasks based on the input modeling concern identifiers. For example, if a modeling container is responsible for processing structural segmentation surfaces of a specific region, the identifier keyword matching unit finds the corresponding task based on the loaded modeling concern identifiers, ensuring that the correct region or component is assigned to the appropriate modeling container for processing. Assigning tasks to suitable modeling containers based on the matching results is a dynamic process. The P modeling concern identifiers are dynamically assigned to P modeling containers based on the matching results. Here, the P dynamically assigned containers refer to containers assigned to specific tasks, and their number is less than or equal to 2M, meaning tasks can be allocated to multiple containers simultaneously, ensuring maximum resource utilization. Dynamic allocation ensures that each modeling container can handle tasks it is good at, avoiding resource waste or uneven load distribution. By flexibly allocating resources among multiple modeling containers, the efficient operation of the system can be ensured.

[0058] Based on the matching association between the synchronous segmented point clouds and the dynamically allocated containers, P synchronous segmented point clouds are loaded into the corresponding P dynamically allocated containers, with each dynamically allocated container receiving the corresponding synchronous segmented point cloud. Then, a parallel modeling container cluster is started, meaning that the P dynamically allocated containers perform modeling tasks simultaneously, with each container handling an independent modeling task. Multiple tasks are processed in parallel, accelerating the entire modeling process. During spatial modeling, each dynamically allocated container generates corresponding component models based on the loaded synchronous segmented point clouds. These component models represent the specific functional units of the shaft. Based on all P component models, the first layer of component models is formed. This is the first model layer in the entire shaft modeling process, containing the basic components of the shaft structure and serving as the foundation for subsequent hierarchical modeling.

[0059] Furthermore, using the first-layer component model as the assembly benchmark, the second-layer shaft segmentation point cloud is modeled based on spatial assembly relationships, and this process continues iteratively until the K-layer shaft segmentation point cloud is assembled, resulting in a shaft component assembly model. The method includes:

[0060] After extracting P groups of assembly nodes from the P synchronous segmented point clouds, assemble the P sub-component models in the first-layer sub-component model according to the P groups of assembly nodes to obtain the first-layer assembly model; perform sub-component modeling based on the second-layer shaft segmented point cloud to obtain the second-layer sub-component model; according to the spatial assembly constraint rules, assemble the second-layer sub-component model into the first-layer assembly model to perform dynamic compensation for assembly deviation to obtain the second-layer assembly model; and so on, until iteratively modeling and assembling the remaining K-2 layers of shaft segmented point clouds, and outputting the shaft sub-component assembly model.

[0061] Assembly nodes refer to the connection points or joints between different components in the shaft. They are key nodes that constitute the 3D model of the shaft. They are usually the interfaces, connection parts or intersections of components. P sets of assembly nodes are extracted from P synchronous segmented point clouds. These nodes represent the connection points of various components in the shaft. Each synchronous segmented point cloud corresponds to a set of assembly nodes to ensure that the various components can be correctly connected during the assembly process.

[0062] In the previous steps, each synchronous segmented point cloud has been modeled as a sub-component model. Based on the extracted P groups of assembly nodes, P sub-component models are assembled into the first-layer sub-component model. The assembly process is based on the node positions and geometric relationships to ensure that each sub-component model is correctly connected with other components in space. Through assembly, the first-layer assembly model is obtained, which is the first-level component assembly result in the overall structure of the shaft.

[0063] In the process of modeling the shaft, the shaft is divided into multiple layers. The second layer of shaft segmentation point cloud represents the second layer in the shaft. Using the second layer of shaft segmentation point cloud, component modeling is performed. Similar to the first layer of component model, the second layer of modeling also extracts geometric features from the point cloud data to construct a three-dimensional model representing each component. Through the modeling process, the second layer of component model is generated, which describes the second layer structure of the shaft.

[0064] Spatial assembly constraint rules refer to the constraints defined during the modeling process for the relative positions, dimensions, angles, and other relationships between different structural units. These constraints ensure that sub-components at different levels can be precisely aligned in three-dimensional space. Constraints include: positional constraints, such as the need for two sub-components to be aligned; dimensional constraints, such as the need for certain components to maintain a specific distance; and angular constraints, such as the need for components to meet design requirements at their angles.

[0065] The second-layer component model is assembled onto the first-layer assembly model. Based on the spatial assembly constraint rules, the position, size, and angle of the second-layer component model are adjusted to match the first-layer assembly model. This step ensures the structural connection and spatial docking between the two layers.

[0066] Due to errors in point cloud data acquisition, precision limitations in the modeling process, or other factors, there may be slight deviations between the first-layer assembly model and the second-layer component model. For example, components in the second layer may deviate slightly from their design positions, leading to misalignment during assembly. To ensure assembly accuracy, dynamic compensation technology is applied to automatically adjust the position of the second-layer component model, ensuring precise alignment with the first-layer assembly model. Dynamic compensation includes adjusting minute positions and rotation angles of components to ensure the assembly result meets design requirements. Through this process, the second-layer component model and the first-layer assembly model are successfully aligned, generating the second-layer assembly model. This model demonstrates the structural relationship between the first and second layers and adjusts for any assembly deviations, ensuring a precise connection between the two layers.

[0067] After obtaining the second-layer assembly model, the remaining layers are processed. Each layer's component model is sequentially docked with the previous layer's assembly model according to spatial assembly constraints. This process is recursive; each layer must be assembled with the previous layer's assembly model. The remaining K-2 layers of shaft segmentation point clouds are processed sequentially, gradually completing the modeling and assembly of each layer. During the assembly process of each layer, assembly deviations are dynamically compensated to ensure that each component is precisely docked with the previous layer's model. After the assembly of all K layers of shaft segmentation point clouds is completed, the shaft component assembly model is output. This model displays the complete structure of the shaft, including all layers of components and their interrelationships, ensuring the spatial and structural consistency of the shaft.

[0068] Furthermore, by hierarchically associating the multiple shaft component segmentation surfaces according to the spatial dependency relationship of the shaft construction sequence, a K-layer shaft component segmentation surface is obtained. The method includes:

[0069] Based on the construction sequence of the shaft, the spatial dependencies of the multiple shaft component segmentation surfaces are parsed and output; multiple geometric spatial anchor points of the multiple shaft component segmentation surfaces are extracted; based on the spatial projection relationship of the multiple geometric spatial anchor points and the multiple spatial dependencies, the spatially confined hierarchical association of the multiple shaft component segmentation surfaces is performed, and the K-layer shaft component segmentation surfaces are output.

[0070] The construction sequence of a hoistway refers to the specific order in which the various components of the hoistway are installed during actual construction. For example, the support structure needs to be installed first, followed by the guide rails, hoistway walls, and other components. The construction sequence determines the spatial relationships and installation dependencies between hoistway components. Some components must be built or installed first to provide support or space for the installation of subsequent components. The hoistway construction sequence analyzes and outputs multiple sets of spatial dependencies between the segmented surfaces of the hoistway. Spatial dependencies refer to the mutual dependence between different components in terms of spatial position. For example, there is a positional dependency between the support structure and the guide rails; the guide rails can only be correctly installed if the support frame is installed.

[0071] Multiple geometric space anchor points are extracted from the segmented surfaces of multiple shafts. Geometric space anchor points refer to the key geometric feature points in the shaft structure. They are reference points for subsequent modeling and spatial assembly. These geometric space anchor points are usually the connection points, intersection points or feature points of key parts of the shaft components. They are used to guide the spatial positioning and assembly in the modeling process. For example, the corners, nodes or intersections between components of the shaft can be used as geometric space anchor points. They can ensure that different segmented surfaces or components are accurately connected in space.

[0072] Spatial projection relationship refers to the projection relationship or relative position between geometric spatial anchor points. In short, it is to project a certain geometric spatial anchor point onto another reference point or structural unit in space to determine their relative positions. Spatial projection relationship can ensure the accurate docking of shaft components at different levels in space, so that components at each level can be correctly connected and matched.

[0073] By combining spatial projection relationships and multiple sets of spatial dependencies, a spatially confined hierarchical association is performed on the segmented surfaces of multiple shafts. This means that, based on the spatial projection relationships and multiple sets of spatial dependencies, the segmented surfaces of multiple shafts are divided into multiple levels (such as level K), ensuring that the components of each level can correctly connect to the level above. The spatially confined hierarchical association helps to clarify how shaft components at different levels are relatively positioned in space, ensuring that each component can correctly connect to other components according to its spatial dependencies and projection relationships. Through the spatially confined hierarchical association, the final output is the K-level shaft segmented surfaces. These segmented surfaces represent different hierarchical structures of the shaft, accurately representing the relationships between different components or regions in space.

[0074] Furthermore, if the first associated interference wellhead node group, as a dual retrieval condition, results in an empty set in the structural risk feature matrix, then the first real-time interference risk feature is set to 0.

[0075] Using the first associated interference shaft node group as a dual search condition, the system searches the structural risk feature matrix for interference risk features related to these nodes. If no match is found, it indicates that there is no interference problem between these nodes. In this case, the search result is an empty set, meaning that no interference risk features related to the first associated interference shaft node group were found. This means that under the given conditions, there will be no interference between these nodes. At this point, the first real-time interference risk feature is set to 0, which means that there is no real-time interference risk between the first associated interference shaft node group, and no further processing or adjustment is required.

[0076] In summary, the three-dimensional modeling method for elevator shafts provided in this application has the following technical effects:

[0077] By extracting shaft architectural information from local building design information, the construction of the shaft model accurately reflects the actual building information. This process eliminates redundant and irrelevant information, retaining only design data related to the shaft structure. Based on the extracted shaft architectural information, a basic shaft model is constructed, providing a starting point for subsequent modeling work and ensuring that the model meets engineering requirements. Real-time shaft point cloud acquisition ensures that the modeling process reflects the actual spatial situation of the shaft, avoiding the limitations of relying solely on design drawings or static models. K-layer shaft segmentation point clouds are obtained by cropping from the real-time shaft point cloud based on the shaft architectural spatial dependency characteristics. This processing method can improve model accuracy. Parallel processing of P synchronous segmentation point clouds significantly accelerates the modeling process. Each synchronous segmentation point cloud is modeled independently, and multiple modeling containers are used for parallel processing, reducing the latency caused by serial processing in traditional modeling methods. By extracting modeling interest markers, it is ensured that key parts of the model can be highlighted during the modeling process. By focusing on and precisely processing the data, the accuracy and operability of the model are further improved. Utilizing spatial assembly relationships, the first-layer component model is used as a benchmark for modeling the second-layer shaft segmentation point cloud. Modeling and assembly are performed layer by layer, ensuring that each layer of the model can be correctly connected and matched, avoiding assembly problems caused by hierarchical misalignment or improper connection. Through an iterative process, the shaft components are processed layer by layer, ensuring that the connections between each component and the overall spatial relationship are accurately modeled. This layered modeling and assembly method can effectively avoid errors or omissions during the overall model assembly. By projecting the shaft base model onto the shaft component assembly model and replacing and filling the difference areas, the differences between the initially constructed shaft base model and the shaft component assembly model obtained through point cloud data and spatial assembly are resolved. After difference filling and model fusion, the final output shaft 3D model can accurately reflect the actual structure and functional requirements of the shaft, improving the accuracy of 3D modeling.

[0078] Example 2, based on the same inventive concept as the three-dimensional modeling method for elevator shafts in the foregoing examples, such as... Figure 2 As shown in the figure, this application embodiment provides a three-dimensional modeling system for elevator shafts, the system comprising:

[0079] The basic model construction module 10 is used to extract shaft building information from local building design information and then construct a basic model of the elevator shaft to be modeled based on the shaft building information.

[0080] The segmentation point cloud acquisition module 20 is used to collect the real-time point cloud of the elevator shaft to be modeled, and then obtain the K-layer shaft segmentation point cloud by cropping from the real-time point cloud based on the architectural space dependency features of the shaft.

[0081] The spatial modeling module 30 is used to extract P modeling interest markers from P synchronous segmented point clouds in the first layer shaft segmentation point cloud, and to perform spatial modeling on the P synchronous segmented point clouds in parallel based on the P modeling interest markers to obtain the first layer component model.

[0082] The component modeling module 40 is used to model the second layer of shaft segmentation point cloud based on the spatial assembly relationship, using the first layer of component model as the assembly reference, until the K layer of shaft segmentation point cloud is iteratively modeled and assembled to obtain the shaft component assembly model.

[0083] The model fusion and filling module 50 is used to project the shaft foundation model onto the shaft component assembly model, and then perform model fusion and filling by replacing the difference areas to output the shaft three-dimensional model.

[0084] Furthermore, the segmented point cloud acquisition module 20 is used to perform the following operation steps:

[0085] The basic model of the elevator shaft is segmented based on functional attributes to obtain multiple shaft component segmentation surfaces, each of which has a modeling concern identifier. The multiple shaft component segmentation surfaces are hierarchically associated according to the spatial dependency relationship of the shaft construction sequence to obtain K-layer shaft component segmentation surfaces. After collecting the real-time shaft point cloud of the elevator shaft to be modeled, the real-time shaft point cloud is mapped and clipped using the K-layer shaft component segmentation surfaces to generate the K-layer shaft segmentation point cloud.

[0086] Furthermore, the segmented point cloud acquisition module 20 is used to perform the following operation steps:

[0087] The structural units corresponding to the segmented surfaces of the multiple shafts are aggregated to obtain Q standard structural units. A structural unit matrix is ​​constructed based on the Q standard structural units. Q sets of interference risk feature sets of the Q standard structural units are accessed via a network, wherein each set of interference risk feature sets consists of an interference structural unit ID and an interference risk geometric parameter threshold. Guided by the interference structural unit IDs of the Q sets of interference risk feature sets, the interference risk geometric parameter thresholds of the Q sets of interference risk feature sets are mapped and filled into the structural unit matrix to obtain a structural risk feature matrix.

[0088] Furthermore, the segmented point cloud acquisition module 20 is used to perform the following operation steps:

[0089] Extract the real-time building geometric parameters of the K-1 layer associated interference shaft node group from the 3D model of the shaft; use the first associated interference shaft node group as a dual retrieval condition to locate the first association matrix node in the structural risk feature matrix coordinates; extract the first interference risk geometric parameter threshold of the first association matrix node; compare whether the first real-time building geometric parameters fall within the first interference risk geometric parameter threshold, and output the first real-time interference risk feature of the first associated interference shaft node group; and so on, perform interference risk traversal identification in the 3D model of the shaft to locate and identify multiple real-time interference risk features.

[0090] Furthermore, the segmented point cloud acquisition module 20 is used to perform the following operation steps:

[0091] Engineering characteristics are extracted from the shaft building information to obtain multiple functional building structures; geometric boundaries of the multiple functional building structures are identified in the shaft foundation model to obtain multiple sets of structural segmentation surfaces; based on multiple standard engineering attributes of the multiple functional building structures, and using the multiple sets of structural segmentation surfaces as data extraction constraints, functional attributes of the multiple sets of segmentation surfaces are extracted from the shaft foundation model; the functional attributes of the multiple sets of segmentation surfaces are used as modeling focus identifiers and mapped and bound to the multiple sets of structural segmentation surfaces to output multiple sets of shaft component segmentation surfaces; the shaft component segmentation surfaces are grouped and split to output the multiple shaft component segmentation surfaces.

[0092] Furthermore, the spatial modeling module 30 is used to perform the following operation steps:

[0093] A pre-constructed parallel modeling container cluster is provided, comprising M parallel modeling containers, the output of which is connected to an identifier keyword matching unit. P modeling interest identifiers are loaded into the M identifier keyword matching units of the M modeling containers for data input matching, locating P dynamically allocated containers, where P ≤ 2M. Based on the matching association between the P synchronous segmented point clouds and the P dynamically allocated containers, the P synchronous segmented point clouds are mapped and loaded into the P dynamically allocated containers. The parallel modeling container cluster is then started to perform spatial modeling, outputting a first-layer component model including P component models.

[0094] Furthermore, the component modeling module 40 is used to perform the following operation steps:

[0095] After extracting P groups of assembly nodes from the P synchronous segmented point clouds, assemble the P sub-component models in the first-layer sub-component model according to the P groups of assembly nodes to obtain the first-layer assembly model; perform sub-component modeling based on the second-layer shaft segmented point cloud to obtain the second-layer sub-component model; according to the spatial assembly constraint rules, assemble the second-layer sub-component model into the first-layer assembly model to perform dynamic compensation for assembly deviation to obtain the second-layer assembly model; and so on, until iteratively modeling and assembling the remaining K-2 layers of shaft segmented point clouds, and outputting the shaft sub-component assembly model.

[0096] Furthermore, the segmented point cloud acquisition module 20 is used to perform the following operation steps:

[0097] Based on the construction sequence of the shaft, the spatial dependencies of the multiple shaft component segmentation surfaces are parsed and output; multiple geometric spatial anchor points of the multiple shaft component segmentation surfaces are extracted; based on the spatial projection relationship of the multiple geometric spatial anchor points and the multiple spatial dependencies, the spatially confined hierarchical association of the multiple shaft component segmentation surfaces is performed, and the K-layer shaft component segmentation surfaces are output.

[0098] Furthermore, if the first associated interference wellhead node group, as a dual retrieval condition, results in an empty set in the structural risk feature matrix, then the first real-time interference risk feature is set to 0.

[0099] Through the foregoing detailed description of the three-dimensional modeling method for elevator shafts, those skilled in the art can clearly understand the three-dimensional modeling system for elevator shafts in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the method section.

[0100] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for three-dimensional modeling of an elevator shaft, characterized in that, The method comprises: After stripping the shaft building information from the local building design information, a shaft base model of the to-be-modeled elevator shaft is constructed based on the shaft building information; After collecting real-time shaft point clouds of the to-be-modeled elevator shaft, K-layer shaft segmentation point clouds are obtained by cutting the real-time shaft point clouds based on shaft building space dependency characteristics; P modeling attention point identifiers of P synchronous segmentation point clouds in the first-layer shaft segmentation point cloud are extracted, and spatial modeling is performed on the P synchronous segmentation point clouds in parallel according to the P modeling attention point identifiers, to obtain a first-layer sub-component model; With the first-layer sub-component model as an assembly reference, sub-component modeling of the second-layer shaft segmentation point cloud is performed based on spatial assembly relationship, and the K-layer shaft segmentation point clouds are iteratively modeled and assembled, to obtain a shaft sub-component assembly model; After projecting the shaft base model to the shaft sub-component assembly model, model fusion filling is performed through difference area replacement, to output a shaft three-dimensional model; Wherein, the K-layer shaft segmentation point clouds are obtained by cutting the real-time shaft point clouds based on shaft building space dependency characteristics, and the method comprises: The shaft base model is segmented based on functional attributes to obtain a plurality of shaft component segmentation surfaces, wherein each shaft component segmentation surface has a modeling attention point identifier, and the modeling attention point identifier is a label for guiding the modeling process, including the indication of the geometric characteristics, structural characteristics or modeling constraint conditions of a certain area; The plurality of shaft component segmentation surfaces are hierarchically associated according to the spatial dependency relationship of the shaft building construction sequence, to obtain K-layer shaft component segmentation surfaces; After collecting real-time shaft point clouds of the to-be-modeled elevator shaft, the K-layer shaft component segmentation surfaces are used for mapping and cutting the real-time shaft point clouds, to generate the K-layer shaft segmentation point clouds; The plurality of shaft component segmentation surfaces are hierarchically associated according to the spatial dependency relationship of the shaft building construction sequence, to obtain K-layer shaft component segmentation surfaces, and the method comprises: According to the shaft building construction sequence, a plurality of groups of spatial dependency relationships of the plurality of shaft component segmentation surfaces are analyzed and output; A plurality of geometric spatial anchor points of the plurality of shaft component segmentation surfaces are extracted; According to the spatial projection relationship of the plurality of geometric spatial anchor points and the plurality of groups of spatial dependency relationships, the plurality of shaft component segmentation surfaces are hierarchically associated by spatial limiting, to output the K-layer shaft component segmentation surfaces.

2. The method for three-dimensional modeling of an elevator shaft according to claim 1, characterized in that, The method further comprises: A plurality of shaft structure units corresponding to the plurality of shaft component segmentation surfaces are aggregated in building structure, to obtain Q standard structure units; A structure unit matrix is constructed based on the Q standard structure units; Q groups of interference risk feature sets of the Q standard structure units are called in network, wherein each interference risk feature set is composed of an interference structure unit ID and an interference risk geometric parameter threshold; With the interference structure unit ID of the Q groups of interference risk feature sets as a guide, the interference risk geometric parameter threshold in the Q groups of interference risk feature sets is mapped and filled to the structure unit matrix, to obtain a structure risk feature matrix.

3. The method for three-dimensional modeling of an elevator shaft according to claim 2, characterized in that, The method further comprises: K-1 real-time building geometric parameters of a K-1-layer associated interference shaft node group are extracted from the shaft three-dimensional model; Locate a first associated interference shaft node group as a double search condition in the structure risk feature matrix coordinate positioning first associated matrix node; Extract the first interference risk geometric parameter threshold of the first associated matrix node; Compare whether the first real-time building geometric parameter falls within the first interference risk geometric parameter threshold, and output the first real-time interference risk feature of the first associated interference shaft node group; By analogy, the shaft three-dimensional model is iteratively modeled and identified for interference risk, and multiple real-time interference risk features are located and identified.

4. The method for three-dimensional modeling of an elevator shaft according to claim 1, characterized in that, The shaft base model is segmented based on functional attributes to obtain multiple shaft component segmentation surfaces, and the method comprises: Engineering properties of the shaft building information are extracted to obtain multiple functional building structures; Geometric boundary identification of the multiple functional building structures is performed on the shaft base model to obtain multiple sets of structure segmentation surfaces; According to multiple standard engineering properties of the multiple functional building structures, multiple sets of segmentation surface functional attributes are extracted from the shaft base model as data extraction constraints; The multiple sets of segmentation surface functional attributes are mapped and bound to the multiple sets of structure segmentation surfaces as modeling focus point identification content, and multiple sets of shaft component segmentation surfaces are output; The shaft component segmentation surfaces are split into groups, and the multiple shaft component segmentation surfaces are output.

5. The method for three-dimensional modeling of an elevator shaft according to claim 1, characterized in that, P modeling focus point identifications of P synchronous segmentation point clouds in the first layer shaft segmentation point cloud are extracted, and spatial modeling is performed on the P synchronous segmentation point clouds in parallel according to the P modeling focus point identifications to obtain a first layer sub-component model, and the method comprises: A parallel modeling container cluster is pre-constructed, wherein the parallel modeling container cluster comprises M modeling containers connected in parallel, and an identification keyword matching unit is connected to an output end of each modeling container; The P modeling focus point identifications are loaded into M identification keyword matching units of the M modeling containers for data input matching, and P dynamic allocation containers are located, wherein P≤2M; After the P synchronous segmentation point clouds are mapped and loaded into the P dynamic allocation containers according to the matching association between the P synchronous segmentation point clouds and the P dynamic allocation containers, the parallel modeling container cluster is started to perform spatial modeling, and the first layer sub-component model comprising P sub-component models is output.

6. The method for three-dimensional modeling of an elevator shaft according to claim 5, characterized in that, The first layer sub-component model is taken as an assembly reference, and sub-component modeling of the second layer shaft segmentation point cloud is performed based on spatial assembly relationship, and the K layer shaft segmentation point cloud is iteratively modeled and assembled to obtain a shaft sub-component assembly model, and the method comprises: After P assembly nodes are extracted from the P synchronous segmentation point clouds, the P sub-component models in the first layer sub-component model are assembled according to the P assembly nodes to obtain a first layer assembly model; Sub-component modeling is performed based on the second layer shaft segmentation point cloud to obtain a second layer sub-component model; According to the spatial assembly constraint rule, the second layer sub-component model is assembled to the first layer assembly model to perform dynamic compensation of assembly deviation, and a second layer assembly model is obtained; By analogy, the remaining K-2 layer shaft segmentation point clouds are iteratively modeled and assembled, and the shaft sub-component assembly model is output.

7. The method for three-dimensional modeling of an elevator shaft according to claim 3, characterized in that, If the first associated interference shaft node group as a double search condition in the search result of the structure risk feature matrix is an empty set, the first real-time interference risk feature is set to 0.

8. A three-dimensional modeling system for an elevator hoistway, characterized by The system comprises: a basic model construction module, configured to construct a shaft basic model of an elevator shaft to be modeled based on shaft construction information after stripping the shaft construction information from local building design information; a segmented point cloud acquisition module, configured to obtain K-layer shaft segmented point clouds by cropping from real-time shaft point clouds of the elevator shaft to be modeled based on shaft construction space dependency features; a space modeling module, configured to extract P modeling attention point identifiers of P synchronous segmented point clouds in a first-layer shaft segmented point cloud, and perform space modeling on the P synchronous segmented point clouds in parallel according to the P modeling attention point identifiers, to obtain a first-layer sub-component model; a sub-component modeling module, configured to perform sub-component modeling on a second-layer shaft segmented point cloud based on a space assembly relationship, taking the first-layer sub-component model as an assembly reference, until the K-layer shaft segmented point clouds are iteratively modeled and assembled, to obtain a shaft sub-component assembly model; a model fusion filling module, configured to perform model fusion filling by difference area replacement after projecting the shaft basic model to the shaft sub-component assembly model, and output a shaft three-dimensional model.

Images