A three-dimensional collaborative design method for national space planning based on BIM-GIS fusion

Abstract

The application discloses a kind of based on BIM-GIS fusion's land space planning three-dimensional collaborative design method, it is related to spatial planning technical field, including based on IFC standard and CityGML standard, constructs the BIM-GIS semantic mapping ontology model including BIM semantic layer, GIS semantic layer and mapping bridge layer.The application realizes the controllable conversion of BIM building component semantics to GIS geographic feature semantics by constructing semantic mapping ontology model and cross-scale geometric mapping algorithm, the complete transmission of key planning attribute is guaranteed in conversion process by attribute inheritance chain, by hierarchical tree index structure and the LOD scheduling strategy based on sight-planning level dual constraint, the adaptive switching of data details is realized, by constructing three-dimensional parameterized management and control model and real-time compliance check, improve the planning scheme design efficiency, multi-scheme three-dimensional comparison and difference highlight function make decision maker can intuitively, quickly grasp scheme difference, decision efficiency is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of spatial planning technology, specifically to a three-dimensional collaborative design method for land spatial planning based on BIM-GIS integration. Background Technology

[0002] The Ministry of Housing and Urban-Rural Development continues to promote the in-depth application of BIM technology in the field of engineering construction. However, currently, BIM technology mainly serves the micro-scale of building engineering, while GIS technology mainly serves the macro-scale of cities and regions. The two lack effective means of integration at the intermediate scale of land spatial planning, which combines macro-control and micro-implementation.

[0003] Currently, the application of BIM-GIS fusion technology in the field of land spatial planning mainly faces the following problems: (1) The semantic gap between multi-scale data has not yet been bridged. BIM models use building components (walls, slabs, columns, beams, pipes, etc.) as basic semantic units, while GIS data uses geographic features (points, lines, surfaces, volumes) as basic semantic units. The two semantic systems have fundamental differences in terms of conceptual hierarchy, attribute granularity, and coordinate accuracy. Existing solutions usually use simple geometric format conversion, which only achieves visual "stitching" and does not establish semantic mapping relationships.

[0004] (2) Lack of a multi-scale LOD adaptive conversion mechanism for territorial spatial planning. Territorial spatial planning covers multiple spatial levels with varying scales. When viewing at different levels, the Level of Detail (LOD) of the BIM model needs to be adaptively adjusted. Existing BIM-GIS integration solutions typically provide static models with fixed LOD levels, lacking an adaptive mechanism to dynamically adjust the LOD based on planning level and viewing distance. This results in rendering stutters when loading too much detail at the macro level, while insufficient detail at the micro level fails to support refined design.

[0005] (3) The coordination and integration between the 3D modeling of the "three lines" control lines and BIM design is lacking. The current land spatial planning mainly expresses the "three lines" control in a two-dimensional plane manner—the ecological protection red line is a closed polygon in a plane, the permanent basic farmland is a plot in a plane, and the urban development boundary is a boundary line in a plane. At the planning implementation level, the BIM design of building projects needs to determine whether the scheme meets the 3D control requirements of the "three lines". Existing technology lacks a method to extend the "three lines" from a two-dimensional plane to a three-dimensional control volume and to perform real-time compliance verification with the BIM design model. As a result, conflicts between the BIM design scheme and the "three lines" control requirements are often only discovered in the approval process, causing repeated modifications.

[0006] (4) The multi-role collaborative design process lacks a unified three-dimensional data foundation. Territorial spatial planning involves multiple participants, each using different professional software. Existing collaboration methods mainly rely on file transfer and meeting communication, lacking a real-time collaborative design platform based on a unified three-dimensional data foundation. Design changes by each party cannot be synchronized in real time, and conflicts cannot be automatically detected, which seriously restricts the efficiency and quality of planning.

[0007] (5) Lack of three-dimensional comparison and quantitative evaluation capabilities for planning schemes. Existing planning scheme comparison usually relies on rendering and manual reporting. It lacks the ability to automatically calculate quantitative indicators based on BIM-GIS fusion data, making it difficult to quickly compare the differences in key indicators of multiple planning schemes in a three-dimensional scene, which is not conducive to scientific decision-making. Summary of the Invention

[0008] To address the aforementioned technical issues, this paper presents a three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion. This technical solution resolves the problems mentioned in the background, such as the lack of a multi-scale data semantic gap, the absence of a multi-scale LOD adaptive conversion mechanism for land spatial planning, the lack of collaborative connection between the three-dimensional modeling of the "three lines" control lines and BIM design, the lack of a unified three-dimensional data foundation for the multi-role collaborative design process, and the lack of three-dimensional comparison and quantitative evaluation capabilities for planning schemes.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion includes: Based on the IFC standard and the CityGML standard, a BIM-GIS semantic mapping ontology model is constructed, which includes a BIM semantic layer, a GIS semantic layer and a mapping bridging layer. Obtain land and space planning and design schemes; Based on the mapping bridging layer, the planning hierarchy is divided to obtain the corresponding detailed level data; Based on the three-dimensional control volume of the control lines in the mapping bridging layer, the compliance verification of the land space planning and design scheme is carried out. Obtain information on changes to the scheme design, including information on participating parties and corresponding change operation information; Based on the scheme design change information, perform operational conflict detection and obtain the scheme change results; Based on the land and space planning design scheme, the system automatically extracts quantitative indicators for land parcel and area planning and compares the schemes.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a three-dimensional collaborative design method for land and space planning based on BIM-GIS integration. By constructing a semantic mapping ontology model and a cross-scale geometric mapping algorithm, it achieves a controllable conversion of BIM building component semantics to GIS geographic element semantics. During the conversion process, the complete transfer of key planning attributes is ensured through attribute inheritance chains. Through a hierarchical tree index structure and a LOD scheduling strategy based on dual constraints of sight distance and planning level, adaptive switching of data details is achieved. By constructing a three-dimensional parametric control model and real-time compliance verification, the efficiency of planning scheme design is improved. The three-dimensional comparison of multiple schemes and the difference highlighting function enable decision-makers to intuitively and quickly grasp the differences between schemes, significantly improving decision-making efficiency. Attached Figure Description

[0011] Figure 1 This is a flowchart of a three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion proposed in this invention; Figure 2 This is a schematic diagram of the three-layer architecture of the BIM-GIS semantic mapping ontology model (TP-SMO) in this invention; Figure 3 This is a flowchart of the cross-scale geometric mapping algorithm (TP-CSGM) in this invention; Figure 4 This is a schematic diagram of the attribute inheritance chain structure in this invention; Figure 5 This is a flowchart of the adaptive LOD scheduling process in this invention; Figure 6 This is a schematic diagram of the three-dimensional control body model of the control line in this invention; Figure 7 This is a flowchart illustrating the real-time compliance verification process for BIM design schemes and control models in this invention. Figure 8 This is a diagram of the event-driven multi-role collaborative design framework architecture in this invention; Figure 9 This is a flowchart of the three-level conflict pre-detection process in this invention; Figure 10 This is a schematic diagram of the three-dimensional quantitative comparison interface for planning schemes in this invention; Figure 11 This is a schematic diagram of the collaborative design conflict detection results in Embodiment 1 of the present invention; Figure 12 This is a schematic diagram of the cross-sectional analysis of the underground space and control structure in Embodiment 2 of the present invention. Figure 13 This is a structural block diagram of a three-dimensional collaborative design system for land spatial planning based on BIM-GIS fusion proposed in this invention. Detailed Implementation

[0012] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0013] Reference Figure 1 - Figure 9 As shown in the figure, a three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion in an embodiment of the present invention includes: Based on the IFC standard and the CityGML standard, a BIM-GIS semantic mapping ontology model is constructed, which includes a BIM semantic layer, a GIS semantic layer and a mapping bridging layer. Obtain land and space planning and design schemes; Specifically, based on the IFC and CityGML standards, a BIM-GIS semantic mapping ontology model is constructed, comprising a BIM semantic layer, a GIS semantic layer, and a mapping bridging layer, including: Based on the attribute set extension mechanism of the IFC4 standard, a new planning-related attribute extension set is added to the IFC standard semantic framework. The planning-related attribute extension set is encapsulated in the attribute set definition format of the IFC standard and includes land use code field, plot ratio constraint field, building height limit field, setback distance field and the code field of the planning management unit to which it belongs. Bind the planning-related attribute extension set to the entity type corresponding to the IFC4 standard to generate BIM semantic layer feature definitions; Based on the national land spatial planning standards, business specification information is obtained, which includes the core control information dimensions of three types of control objects: ecological protection red line, permanent basic farmland, and urban development boundary. Among them, the core control information dimensions include four dimensions: control space scope association, control rule constraints, compliance judgment conditions and vertical elevation benchmark; Based on business specification information, obtain the control attributes corresponding to each core control information dimension. The control attributes include the unique identifier of control line elements, the set of control rule entries, the Boolean expression for compliance judgment, and the national elevation datum code. Based on the extracted control attributes, field-based abstract modeling is performed, and the unique identifier of the control line element is abstracted into the control line three-dimensional control volume associated field. The field data type is string, which stores the unique identifier of the associated three-dimensional control volume. The set of control rule entries is abstracted into a control rule set field. The field data type is a structured array, and each array element contains three subfields: rule type, constraint parameters, and applicable conditions. The compliance judgment Boolean expression is abstracted into a compliance judgment logical expression field. The field data type is string, and it stores the compliance judgment condition expression connected by Boolean logical operators. The national elevation datum code is abstracted into an elevation datum field, and the field data type is an enumeration type; Based on the application domain extension mechanism of CityGML3.0, the above fields are encapsulated into a set of management semantic extensions; Bind the control semantic extension set to CityGML3.0 to obtain the GIS semantic layer feature definition; Based on the feature definitions of the BIM semantic layer and the GIS semantic layer, a mapping bridging layer is constructed to define the geometric transformation rules and attribute transformation rules between BIM semantics and GIS semantics; Based on the BIM semantic layer, GIS semantic layer, and mapping bridging layer, a BIM-GIS semantic mapping ontology model is constructed.

[0014] In this solution, the planning-related attribute extension set is bound to the IfcBuilding and IfcSpace entity types in the IFC4 standard. When obtaining the control attributes corresponding to each core control information dimension based on business specification information, specific control attributes corresponding to the above four dimensions are extracted from the control business process based on current normative documents such as the "Guidelines for Land and Sea Use Classification in Territorial Spatial Survey, Planning, and Land Use Control" and the "Measures for the Management of Ecological Protection Red Lines." When abstracting the national elevation datum code into an elevation datum field, selectable values ​​include the 1985 National Elevation Datum, the 1956 Yellow Sea Elevation Datum, and local elevation datums. Based on the application domain extension mechanism of CityGML 3.0, the XML of the extension module... The schema definition file declares the name, data type, value range, and bound object type of each field. These fields are encapsulated into a control semantic extension set, which is then bound to the Building and LandUse objects of CityGML. This allows the extended GIS semantic elements to carry complete 3D control attribute information, completing the definition and injection of the control semantic module. The control line 3D control volume association field in the control semantic extension set stores the unique identifier of the subsequently constructed 3D parametric control volume, establishing a reference association between GIS semantic elements and 3D control entities. Once the 3D parametric control volume is constructed, its unique identifier is backfilled into the corresponding GIS element's association field, achieving the binding of semantic attributes with geometric control entities.

[0015] Specifically, based on the feature definitions of the BIM semantic layer and the GIS semantic layer, a mapping bridging layer is constructed, which includes: Obtain the BIM models of buildings in the land spatial planning and design scheme; Based on the geometric vertex coordinates of all building components in the building BIM model, obtain the axis-aligned bounding box corresponding to each building component. The axis-aligned bounding box is determined by the extreme coordinates of the building component. Based on the axis-aligned bounding boxes corresponding to each building component, an axis-aligned bounding box hierarchy tree is constructed by aggregating from bottom to top according to spatial containment relationships. Based on the principle of minimum description length, the component set is optimally binary partitioned to obtain the building component group; The building component groups near the building's perimeter are labeled as the shell component set, and the building component groups located inside the building are labeled as the interior component set; Based on the shell component set, the corresponding continuous contour triangular mesh is obtained using the Alpha Shape algorithm; In this embodiment, based on the surface triangular patches of all components in the shell component set, the alpha radius parameter of the Alpha Shape algorithm is set to twice the average wall thickness of the building. Triangles with side lengths exceeding twice the alpha radius are filtered out through the alpha composite, and the continuous contour triangular mesh of the building shell is extracted. Based on continuous contour triangular mesh, the building space is voxelized, and the occupying voxels are marked as solid voxels. The moving cube algorithm is used to extract isosurfaces from the boundaries of solid voxels and reconstruct a simplified boundary representation shell. Understandably, based on a continuous contour triangular mesh, the architectural space is voxelized at a resolution of 1cm, dividing the architectural space into a regular cubic mesh. Each voxel has integer coordinates.

[0016] For each voxel, determine whether its center point is located inside the building entity: Using a continuous contour triangular mesh as the boundary of the entity, the voxel center point is determined by ray casting or signed distance field (SDF) to determine whether it is inside or outside the entity boundary. If it is inside, it occupies the voxel. The occupying voxels are marked as solid voxels. The moving cube algorithm is used to extract isosurfaces from the boundaries of solid voxels and reconstruct them into simplified boundary representation shells, thus completing the geometric simplification from the component level to the building shell level. Based on the set of internal components, the components are grouped according to the floors to which they belong, and the weighted average net height of the horizontal projected area of ​​all internal spaces on the corresponding floors of each group is taken as the net height of that group. In the formula, Indicates net height. Indicates the first The horizontal projected area of ​​each building component Indicates the first The clear height of a building component; the clear height represents the effective height of the building component itself. This indicates the total number of building components in the group; Based on the projected contours of each group of internal components on the horizontal plane, the Douglas-Peucker algorithm is used to simplify the contours and obtain simplified internal polygons. Using the simplified polygons corresponding to each group of internal components as the base and the net height of the group as the top height, a prism is generated, retaining the floor number, area and net height, to obtain the simplified internal space. Obtain roof type information, which indicates the roof type corresponding to the building; Based on the boundary representation of the outer shell, the roof surface is simplified based on the roof type information, and the wall surfaces are uniformly extracted as vertical facades to obtain the building block model. Based on the projected outline of the building block model on the horizontal plane, the Douglas-Peucker algorithm is used to simplify the outline and obtain simplified polygons. In this embodiment, when using the Douglas-Peucker algorithm to simplify the outline, the simplification distance threshold is set to 2% of the average side length of the building, compressing the number of outline vertices to less than 30% of the original number, thus obtaining a simplified polygon. Based on the boundary representation of the outer shell, simplified height information is obtained by using the area-weighted average of each floor elevation. In the formula, Indicates elevation. Indicates the building number The horizontal projected area of ​​the layer. Indicates the building number The top slab elevation of a floor indicates the height of the building from a reference surface (such as the ground). This represents the total number of floors in the building. A simplified building model is obtained by generating a prism with a simplified polygon as the base and a simplified height as the top height. Based on the boundary representation of the outer shell, the simplified internal space, the building block model, and the simplified building model, obtain cross-scale geometric mapping information.

[0017] In this scheme, based on the axis-aligned bounding boxes corresponding to each building component, an axis-aligned bounding box hierarchy tree is constructed by aggregating from bottom to top according to spatial containment relationships. The specific steps are as follows: S100: Based on the geometric vertex coordinates of each building component in the BIM model, calculate the axis-aligned bounding box for each component. The bounding box is determined by six extreme coordinates, and the calculation formula is as follows: In the formula, Indicates the first The axis-aligned bounding box corresponding to each building component For the first The minimum X-coordinate among all vertices of a building component; the same applies to other extreme values. S200: Align each building component and its axis with the bounding box as the initial leaf node of the hierarchy tree; S300: Determine the spatial inclusion relationship between any two initial leaf nodes based on their bounding box extreme coordinates; For the initial leaf nodes A and B, if the bounding box of node A satisfies: Then it is determined that the bounding box of node A completely contains the bounding box of node B; Based on the inclusion relationship determination result, the number of child nodes that are completely included by each node is counted and recorded as the inclusion degree; S400: Mark nodes with a containment degree of 0 as bottom-level leaf nodes, and mark nodes with a containment degree greater than 0 as candidate parent nodes; S500: Starting from the bottom leaf nodes, attach each bottom leaf node to the candidate parent node that completely contains it and has the smallest containment degree, and make it the direct child node of that candidate parent node. If multiple candidate parent nodes completely contain the bottom leaf node and have the same containment degree, the candidate parent node with the smallest bounding box volume is selected as the direct parent node. S600: Based on the parent node with attached child nodes, calculate the bounding box of its parent node. The parent node's bounding box is the spatial union of the bounding boxes of all its direct child nodes, i.e.: in, This represents the minimum X-coordinate of the parent node's bounding box. This represents the maximum X-coordinate of the parent node's bounding box. This represents the minimum X-coordinate of the bounding box of the direct child node corresponding to the parent node. This represents the maximum X-coordinate of the bounding box of the direct child node corresponding to the parent node. Show the set of direct child nodes corresponding to the parent node The extreme values ​​of the parent node's bounding box in the Y and Z directions are calculated similarly; S700: For a parent node that has completed the mounting of its child nodes, recount the number of nodes that contain it. If the parent node is completely contained by another node, then mount it to the corresponding parent node to form the next level of aggregation. S800: For nodes that do not have an inclusion relationship, calculate the Euclidean distance between the centers of the bounding boxes of any two nodes. The calculation formula is as follows: In the formula, For the first The center coordinates of the bounding box of each node; Select the two closest nodes and merge them into a new parent node. The bounding box of the new parent node is the union of the bounding boxes of the two child nodes. S900: Repeat steps S600-S800, aggregating upwards layer by layer until all nodes are aggregated into a unique root node. The bounding box of the root node is the global spatial union of the bounding boxes of all components, thus completing the construction of the axis-aligned bounding box hierarchy tree.

[0018] In this embodiment, when performing optimal binary partitioning on the component set, the partitioning objective is to minimize the sum of the total bounding box volume of the two groups of components after partitioning and the partitioning code length. The partitioning code length includes the subtree / component belonging code and the node structure code. When performing optimal binary partitioning on the component set, the corresponding axis-aligned bounding box hierarchy tree is also divided into two subtrees, including the left subtree and the right subtree. Subtree / component attribution codes are used to describe which components are assigned to the left subtree and which are assigned to the right subtree; For a set of n components, the optimal binary partition is equivalent to choosing a subset partition. Therefore, if 1 bit is used to represent "whether component i belongs to the left subtree or the right subtree", then n components require n bits, and the subtree / component attribution encoding length is n. The node structure encoding is used to describe the type information of nodes in each subtree, i.e., whether they are internal nodes or leaf nodes. Each internal node needs 1 bit to identify itself as an internal node. An internal node is a node that has both a parent node and a child node. The subtree / component attribution encoding length is the sum of the lengths of the internal nodes of the two subtrees.

[0019] In this embodiment, the roof type can be obtained by reading the building BIM model, or by clustering the top triangular facets into types such as flat roof, double-slope roof, and four-slope roof through normal vector clustering to achieve roof type identification. Based on the identified roof type, select the corresponding geometric template from the predefined parametric model primitive library; for example: Flat roof: a simple flat box.

[0020] A gable roof is a triangular prism consisting of two sloping rectangular faces and a rectangular base.

[0021] Four-sloped roof: A pyramidal platform consisting of four trapezoidal or triangular faces and a rectangular base; The parametric model primitive library represents the geometric models corresponding to different roofs, including: Shape parameters: These are internal properties that describe the "look" of a primitive, such as the width, length, wall height, and roof height of a house. Different types of roofs have different combinations of shape parameters.

[0022] Taking a hip roof as an example, its shape parameters typically include: width, bottom length, ridge length, and roof height.

[0023] Different primitives have different shape parameters, but their position parameters are universal.

[0024] Position parameters: These define the "where" and "facing" of the primitive, typically including the rotation angle along the Z-axis and three translations. The parametric model primitive library determines the minimum set of parameters required to describe the geometry of each primitive type by identifying the types of roofs and building structures that need to be included. These parameters include shape parameters (length, width, height, etc.) and position parameters (rotation, translation).

[0025] Define a local coordinate system for each primitive and fit a mathematical formula to calculate the coordinates of all vertices using their shape and position parameters.

[0026] Define faces and topological relationships, determine which vertices constitute a face (wall face, roof face), and the connections and adjacency relationships between these faces; This is how the definition of primitive types is achieved; Based on the boundary representation of the outer shell, obtain the roof parameters, including the roof profile, ridge line, and eaves height / roof height; The roof outline is a polygon projected onto a horizontal plane. The outline can be extracted and simplified from point clouds or mesh boundaries using the Alpha Shape algorithm.

[0027] The ridge line is the intersection of the slopes of a pitched roof, including the location and length of the intersection. Eaves height / roof height refers to the height of the roof surface; Substitute roof parameters (such as roof profile, ridge line, height, etc.) into the parametric template to generate a simplified geometric model of the roof.

[0028] Understandably, BIM and GIS differ by orders of magnitude in geometric representation precision (BIM at the millimeter level, GIS at the centimeter to meter level). Direct conversion would lead to an explosion in data volume or loss of semantics. Therefore, cross-scale geometric mapping is necessary. In this solution, BIM is simplified as follows: (a) LOD6→LOD3 (Component Level→Building Shell + Interior Space): Building components (walls, slabs, columns, etc.) in the BIM are aggregated according to their spatial enclosure relationships. First, an AABB bounding box hierarchy tree of all components within the building is constructed. Then, an optimal partitioning algorithm based on the Minimum Description Length (MDL) principle is used to divide the component set into "shell component set" and "internal component set". The shell component set extracts the building shell contour surface (triangular mesh) using the Alpha Shape algorithm, and then generates a simplified B-rep shell through voxelization-surface reconstruction (1cm resolution). The internal component set is grouped by floor, and each group is extracted as a simplified space, retaining key attributes such as net height and area.

[0029] (b) LOD3 → LOD2 (Building Shell → Block Model with Roof Structure): Automatic roof type identification is performed on the LOD3 shell (based on normal vector clustering, clustering the top triangular facets into flat roofs, gable roofs, four-sloped roofs, etc.), and then the roof surface is parametrically reconstructed and simplified according to the roof type. Walls are uniformly made vertical, and concave and convex details are removed. Finally, a CityGML LOD2 level Block model (including accurate roof structure) is generated.

[0030] (c) LOD2→LOD1 (Block Model→Simplified Cuboid): Calculate the projected outline of the LOD2 model on the XY plane, and simplify it into convex polygons or L-shaped polygons using the Douglas-Peucker algorithm. The height is taken as the weighted average of the floor elevations. Finally, a simplified cuboid or prism at the CityGML LOD1 level is generated.

[0031] During each simplification process, a Property Inheritance Chain (PIC) is maintained to record the traceability relationship between the simplified geometry and the original BIM components.

[0032] Specifically, based on the feature definitions of the BIM semantic layer and the GIS semantic layer, a mapping bridging layer is constructed, which also includes: The BIM model is simplified into a boundary representation shell and a simplified internal space as a first-level geometric simplification; the boundary representation shell is simplified into a building block model as a second-level geometric simplification; the building block model is simplified into a simplified building model as a third-level geometric simplification. Based on each level of geometric simplification process, an attribute inheritance chain node is established for each simplified geometry, recording the simplified geometry identifier, the source geometry identifier list, the inherited attribute list, the lost attribute list, and the reverse reconstruction prompts; Obtain attribute inheritance information based on the attribute type of building components; Among them, structural parameter attributes are directly mapped to the corresponding attributes of GIS building objects; For functional type attributes, the BIM functional type of building components is mapped to the land use and sea use classification code in GIS; All attribute inheritance chain nodes are linked together according to a simplified hierarchy to form a complete attribute tracing chain.

[0033] In this scheme, structural parameters in BIM (story height, total height, number of stories, foundation depth, etc.) are directly mapped to corresponding attributes of GIS building objects through cross-geometric transformation. IfcSpace function types in BIM (such as "office," "residential," "shopping mall," etc.) are mapped to land use and sea use classification codes in GIS (such as "0701 urban residential land," "0901 commercial service land," etc.), and the mapping relationship is stored in a pre-defined lookup table. When a building contains multiple functional spaces, the dominant function is determined according to the area proportion of each functional space. If the maximum area proportion does not exceed 60%, it is marked as "mixed-use land." A two-way binding is established between the planning extension attributes (PlanExtProps) on the BIM side and the control semantics (CtrlSemantics) on the GIS side. When the control rules on the GIS side are updated (such as adjusting building height limits), a compliance check on the BIM side is automatically triggered through an event notification mechanism; when design changes on the BIM side lead to changes in building height, an update request is automatically initiated to the GIS side.

[0034] Based on the mapping bridging layer, the planning hierarchy is divided to obtain the corresponding detailed level data; Specifically, based on the mapping bridging layer, the planning hierarchy is divided to obtain the corresponding detailed level data, including: Based on the axis-aligned bounding box hierarchy tree, the planning level is allocated according to the tree depth to obtain the planning level corresponding to each node. The planning level includes the city area, central urban area, management unit, land parcel and building. Among them, if the tree depth corresponding to the node in the axis-aligned bounding box hierarchy tree is [0, 2], it is classified as a city area; if the tree depth is [3, 5], it is classified as a central urban area; if the tree depth is [6, 8], it is classified as a management unit; if the tree depth is [9, 11], it is classified as a plot of land; if the tree depth is greater than or equal to 12, it is classified as a building. Based on the simplified model in the cross-scale geometric mapping information, the details of the geometric data are divided to obtain the geometric data detail level information. Each level of the simplified model in the cross-scale geometric mapping information corresponds to a geometric data detail level. Based on the current camera position, obtain the distance from the center of the bounding box of each node to the camera, and determine the target detail level of each node according to the distance threshold; Based on the target level of detail for each node, load the geometric data corresponding to the level of detail. If a user-specified planning level exists, the corresponding node of that level will be forcibly loaded to the specified detail level, and the data detail level of other level nodes will be reduced.

[0035] In this scheme, within each leaf node, based on the BIM-GIS elements falling within that spatial range, multiple LOD-level geometric data are generated using a cross-geometric transformation algorithm, mapping the data detail levels to the model simplification levels. The BIM and GIS data corresponding to the LOD1 model (simplified cuboid) in the cross-geometric transformation algorithm are used as LOD1 detail level data; the BIM and GIS data corresponding to LOD2 (mass model) are used as LOD2 detail level data; the BIM and GIS data corresponding to LOD3 (building shell + interior space) are used as LOD3 detail level data; and the BIM data corresponding to LOD6 (component level) is used as LOD4 detail level data. After serialization, the data at each level is stored in the local file system or object storage, and data references are recorded in the leaf nodes.

[0036] Preferably, in this embodiment, the component-level BIM data can be further divided: I. BIM Data LOD Level Classification BIM data is divided into 6 levels based on geometric accuracy and semantic integrity, corresponding to the viewing needs of different stages of architectural design and different planning levels: LOD1 A block model (cubic prism or prism) only represents the overall enclosing space of the building. Building name, land use, total height, total number of floors Conceptual planning stage City / Central Urban Area Urban overall landscape assessment and skyline analysis LOD2 A block model with a roof structure, distinguishing between roof types (flat roof / pitched roof). Add building function type, building age, and structural type. Solution Phase Central urban area / management unit Preliminary estimates of functional layout assessment and sunlight analysis for the area LOD3 A precise model including main components such as walls, floors, and columns, with component dimensional errors ≤10cm. Increase component material, fire resistance rating, and load-bearing properties. Preliminary design phase Management Unit / Plot Land development intensity assessment and precise calculation of plot ratio LOD4 Includes detailed components such as doors and windows, stairs, and embedded parts for main pipelines. Add equipment parameters, pipeline routing, and interface locations. Construction drawing design stage land parcel Building and municipal pipeline collision detection, planning indicator verification LOD5 Includes all building components (including decorative components, insulation layer, waterproof layer, etc.), with a dimensional error ≤1cm. Add construction techniques, material specifications, and cost information. Construction drawing refinement stage land parcel Detailed design review and construction plan simulation LOD6 The as-built model is consistent with the actual built environment, with a dimensional error of ≤5mm. Add completion date, actual material batch, and quality acceptance information. Completion and Acceptance Stage Plot / Building Completion planning verification and urban renewal assessment Basis for division: Refer to the definition of LOD in the "Unified Standard for Building Information Modeling Applications" GB / T 51212-2016; Referencing the LOD Specification grading standards of the US BIMForum; Determined based on the viewing accuracy requirements of different levels of territorial spatial planning; II. LOD Level Classification of GIS Data GIS data is classified into four levels based on geometric representation complexity and semantic completeness, corresponding to the CityGML 3.0 standard's grading specification: LOD0 2.5D terrain surface (TIN triangulation) + building outline Elevation value, building footprint area, number of building stories 1:10000 topographic map, DEM data City Topographical analysis and macro-level assessment of land use layout LOD1 Simplified block model (cubic prism or prism, height is the number of layers × estimated layer height) Building function, building height, and building age Planning permit data, remote sensing interpretation City / Central Urban Area Urban overall morphology analysis and plot ratio estimation LOD2 A block model with precise roof structure (distinguishing between flat roofs, pitched roofs, and curved roofs). Add building structure type, building material, and roof type Oblique photogrammetry, laser scanning Central urban area / management unit Urban landscape assessment, skyline analysis, and sunshine calculation LOD3 A precise architectural model with internal structure (including components such as walls, floors, doors, and windows). Increase the division of the building's interior space, the function of each floor, and the evacuation routes. As-built surveying and lightweight conversion of BIM models Management Unit / Plot Detailed planning of the site, management of underground space, and collaborative design. Basis for division: Refer to the definitions of LOD0-LOD3 in the OGC CityGML 3.0 Encoding Standard; Refer to the data accuracy requirements in the "Technical Specifications for Urban Information Modeling (CIM) Basic Platform"; The accuracy requirements of different levels in the "five-level, three-category" system of national land spatial planning are determined.

[0037] In this embodiment, when determining the target detail level of each node based on the distance thresholds, the determination of the view distance thresholds D2, D3, and D4 follows the following principles: (1) City-level (coverage area 50km×50km): D4=8000m: Beyond this distance, LOD1 level data will be displayed; D3=4000m: Displays LOD2 level data between D3 and D4; D2=1000m: Displays LOD3 level data between D2 and D3; Data smaller than D2 is displayed as LOD4 / LOD5 / LOD6 level data; (2) Central urban area level (covering an area of ​​10km × 10km): D4=2000m: Beyond this distance, LOD1 level data will be displayed; D3=1000m: Displays LOD2 level data between D3 and D4; D2=300m: Displays LOD3 level data between D2 and D3; Data smaller than D2 is displayed as LOD4 / LOD5 / LOD6 level data; (3) Management unit level (coverage area 5km×5km): D4=1000m: Beyond this distance, LOD1 level data will be displayed; D3=500m: Displays LOD2 level data between D3 and D4; D2=100m: Displays LOD3 level data between D2 and D3; Displaying LOD4 / LOD5 / LOD6 level data when D2 is less than D2 (4) Plot level (covering an area of ​​500m × 500m): D4=200m: Beyond this distance, LOD1 level data will be displayed; D3=100m: Displays LOD2 level data between D3 and D4; D2=30m: Displays LOD3 level data between D2 and D3; Data smaller than D2 is displayed as LOD4 / LOD5 / LOD6 level data; (5) Building level (covering an area of ​​50m×50m), displaying LOD4 / LOD5 / LOD6 level data; If a user-specified planning level exists (e.g., switching to the land parcel level), the corresponding node of that level will be forcibly loaded to the specified level of detail (LOD3), and the level of detail of other level nodes will be reduced to LOD1.

[0038] In this solution, the threshold can be set according to actual needs. The methods for determining the threshold include: Calculated based on human visual resolution and screen pixel density; Ensure that the transition between adjacent LOD levels is not obvious at the current viewing distance (visually continuous transition). To optimize rendering performance, we aim to minimize data loading while maintaining visual quality.

[0039] When the total amount of loaded data exceeds the video memory limit When the default is 80% of GPU memory, low-priority LOD data is evicted in ascending order of loading priority until the total usage drops to a minimum. The following eviction strategy adopts LRU-K (Least Recently Used with K references) to avoid frequent load-eviction oscillations and prioritize the retention of high-priority nodes with high-LOD level data.

[0040] The formula for calculating the load priority of each LOD entry for each node is as follows: in, For loading priority, These are the weighting coefficients (default values ​​are 0.4, 0.4, and 0.2 respectively). The visibility score for this node from the current viewpoint is the ratio of the visible volume of the node's axis-aligned bounding box to the total volume of the axis-aligned bounding box. The data freshness score represents the ratio of the amount of data updated in the current update cycle to the total amount of data in the node. In this embodiment, the update cycle is 24 hours.

[0041] LOD data loading employs an asynchronous incremental approach. Lower LOD level data is loaded first (for quickly rendering a coarse model), followed by progressive refinement of higher LOD level data in the background. Data loading utilizes HTTP / 2 multiplexing or WebSocket streaming, and CDN caching is supported for acceleration.

[0042] Based on the three-dimensional control volume of the control lines in the mapping bridging layer, the compliance verification of the land space planning and design scheme is carried out. Specifically, based on the three-dimensional control volume of the control lines in the mapping bridging layer, compliance verification of the land spatial planning and design scheme is carried out, including: Based on the needs of national land spatial planning, a three-dimensional control system for control lines is constructed, which includes a three-dimensional control system for ecological protection red lines, a three-dimensional control system for permanent basic farmland, and a three-dimensional control system for urban development boundaries. Based on the spatial coordinates of the BIM model, and using the axis-aligned bounding box hierarchy tree, obtain the corresponding management unit level nodes and the three-dimensional control volume of the control lines. The elevation of the highest point in the BIM model is compared point by point with the corresponding control line 3D control body height limit surface. If the elevation of the highest point exceeds the height limit surface, a height limit violation is generated, and the violation amount is the difference between the elevation of the highest point and the elevation of the height limit surface. Based on the minimum distance between the BIM model axis-aligned bounding box and the boundary of the control line 3D control body on the horizontal plane, if the minimum distance is less than the setback distance of the control line 3D control body, a setback violation will be generated. In this embodiment, the minimum distance between the BIM model and the boundary of the control body in the XY plane is used for approximate calculation. If the minimum distance is less than the setback distance of the three-dimensional control body of the control line, the distance is further determined based on the GJK (Gilbert-Johnson-Keerthi) algorithm distance solution to determine whether the distance is less than the setback distance of the three-dimensional control body of the control line. Based on the 3D collision detection between the underground part of the BIM model and the permanent basic farmland control body, if the underground part of the BIM model intrudes into the protection depth range of the cultivated layer, an underground space intrusion violation will be generated. The system compares the functional usage codes and control rule entries based on the BIM model to check whether the building functions meet the access requirements of the area, generates land use compatibility test results, and obtains compliance verification information for land space planning and design schemes.

[0043] In this plan, based on the needs of land and space planning, a three-dimensional control system for the control line is constructed, specifically including: (a) Three-dimensional control system for ecological protection red lines The three-dimensional management of ecological protection red lines involves not only the planar scope but also vertical management (such as the clearance height requirements for ecological corridors and the protection of groundwater levels in water conservation areas). The three-dimensional management volume of ecological protection red lines is parameterized as follows:

[0044] in, It is a planar base polygon (CGCS2000 coordinates). The surface elevation is generated by interpolating the DEM into a TIN triangulation network. The height limit for ecological corridors is determined by referring to a table based on the type of ecological corridor, such as a height limit of 200m for bird migration corridors and 50m for general ecological corridors. The outer buffer distance is determined according to the protection level, such as 500m for a first-level protection zone and 200m for a second-level protection zone. It is a set of ecological constraint rules, including a list of prohibited construction types and expressions for access conditions.

[0045] The geometric upper boundary of the three-dimensional control volume is Curved surface, lower boundary is Curved surface, side surface is A polygon is a surface body stretched along the surface normal to the upper and lower boundaries. The buffer region is modeled separately as an outer wrapper, and restrictive controls are implemented within the buffer.

[0046] (b) Three-dimensional control system for permanent basic farmland The three-dimensional management of permanent basic farmland focuses on the protection of topsoil and the restriction of underground space utilization. The three-dimensional management volume of permanent basic farmland is parameterized as follows:

[0047] in, The protected area of ​​farmland is a polygon. This refers to the depth range of the topsoil (typically from the surface to -1.5m, determined by soil profile data). Height limits are imposed on above-ground structures (generally not exceeding 2m, but facilities such as agricultural greenhouses are permitted). The set of control rules includes the prohibited depth of topsoil excavation, the minimum soil cover depth for laying underground pipelines, and the types of agricultural supporting facilities that are allowed to be constructed.

[0048] The upper boundary of the three-dimensional control volume is above the Earth's surface. The horizontal plane of the height, with its lower boundary being the bottom boundary of the cultivated layer. Below a certain depth (default -3m, to ensure safe isolation below the cultivated layer).

[0049] (c) Three-dimensional control system for urban development boundaries The three-dimensional control of urban development boundaries involves both above-ground development intensity (floor area ratio, building height limits) and underground space development scope. The three-dimensional control volume of urban development boundaries is parameterized as follows:

[0050] in, The area to be developed is a polygon. The maximum floor area ratio for the zone. The height restriction surface for buildings in the zone (which can be determined by a unified height restriction or aviation height restriction surface). The depth of underground space development is limited (usually -15m to -30m). The vertical stratification scheme (dividing the development space vertically into underground space layer, surface space layer, low-level space layer, middle-level space layer, high-level space layer, etc.) For the set of control rules.

[0051] The three-dimensional control body is pressed in the vertical direction The system is divided into layers, with each layer assigned different control rules (e.g., the underground space layer allows the construction of municipal utility tunnels and parking facilities, while the lower-level space layer has a height limit of 24m), forming a "vertical zoning control" model.

[0052] Obtain information on changes to the scheme design, including information on participating parties and corresponding change operation information; Based on the scheme design change information, perform operational conflict detection and obtain the scheme change results; Specifically, based on the scheme design change information, operational conflict detection is performed to obtain the scheme change results, including: Obtain the design change operations of each participant, and formalize each change operation into a triplet structure of operation type, operation target element reference path and operation parameters. The operation type includes three categories: creation, modification and deletion. Based on the distributed version management of the central collaborative model repository, operation transformation is performed on two concurrent operations to ensure that the final state is consistent after different clients execute in different orders; For two creation operations, if different features are created in the same location, they are both retained. If the same feature is created and the globally unique identifier is the same, the later one is marked as a conflict and the user is notified. For two modification operations, if different attributes of the same element are modified, the modifications are merged; if different values ​​of the same attribute are modified, the winning strategy is written last based on the priority of the participants. In the event of a conflict between deletion and modification operations, the deletion operation takes precedence, and the modification operation is marked as invalid and the modifier is notified.

[0053] In this solution, spatial partitioning locks are used for operations involving changes in geometric location (such as moving building locations). The 3D space is divided into several locked partitions according to the OLT leaf nodes. Before modifying the geometry, the operator needs to acquire the write lock of the corresponding partition. The granularity of the lock is configurable (default is OLT leaf node level) and supports lock escalation (when the operation involves multiple adjacent nodes, it is escalated to the parent node level lock).

[0054] Perform the following three-level pre-checks before each change submission: Level 1: Immediate pre-screening (<1 second) Check for collisions between the modified feature and the AABB bounding boxes of surrounding confirmed features. The scope is limited to a 50m space outside the modified feature's bounding box. If a collision is detected, immediately highlight the collision area in red on the client and prevent submission.

[0055] Level 2: Quick Pre-screen (1-5 seconds) For changes that pass the real-time pre-inspection, perform precise geometric collision detection (GJK+EPA algorithm). Simultaneously, check if the changed planning indicators exceed limits (such as calculating the plot ratio, building density, green space ratio, etc.). If problems are found, display an orange warning on the client side, allowing forced submission but requiring a reason.

[0056] Level 3: Global Pre-Check (<30 seconds) For changes that pass the rapid pre-inspection, a global impact analysis is performed on the server side. This checks whether the change triggers a cascading effect (e.g., changes in sunshine hours due to building relocation affecting sunshine compliance of adjacent plots, changes in traffic flow, etc.). The global pre-inspection is executed asynchronously, without blocking current operations, and the results are pushed to relevant participants upon completion.

[0057] Based on the land and space planning design scheme, the system automatically extracts quantitative indicators for land parcel and area planning and compares the schemes.

[0058] This scheme automatically extracts the quantitative indicators for land parcel-level and area-level planning in the territorial spatial planning and design scheme, including: (a) Land parcel level indicators: Floor area ratio (FAR) ): In the formula, For the first plot of land The sum of the floor areas of all floors of the building. The area of ​​the land parcel. The total number of buildings within the plot, building area, and floor information are directly extracted from the BIM model.

[0059] Building density ( ): In the formula Let be the base projection area of ​​the j-th building (the projection area of ​​the BIM model on the XY plane).

[0060] Green space ratio represents the ratio of green space area to land parcel area in a GIS land use classification layer.

[0061] Building height restriction compliance rate refers to the percentage of buildings within a plot whose highest point is below the height restriction value.

[0062] (b) District-level indicators: Skyline outline: Extracts the projection envelope of the top outlines of all buildings within the area on the specified viewing direction.

[0063] Sunlight Analysis: Based on the BIM building model and terrain data, the cumulative sunshine hours algorithm (calculated on three typical days: spring equinox, autumn equinox, and winter solstice) is used to calculate the sunshine hours of each window sill of the residential buildings within the plot.

[0064] Understandably, the information such as building outline, floor area, and height in the indicator calculation formula are all directly and accurately extracted from the BIM model (rather than estimated values ​​from the simplified GIS model), which is the accuracy advantage brought by the integration of BIM and GIS.

[0065] Reference Figure 10 As shown, this scheme allows for the comparison and selection of land spatial planning and design schemes, and can simultaneously load multiple planning schemes in the same 3D scene (supporting up to 5 schemes for side-by-side comparison), providing the following comparison functions: (a) Indicator dashboard comparison: The planning indicators of each scheme at the plot level and area level are displayed side by side in the form of radar chart / bar chart, and the indicators can be filtered and sorted.

[0066] (b) 3D scene sliding comparison: The 3D scene is divided into left and right (or multiple) views, and each view loads a different scheme. It supports two modes: synchronous linkage (rotation and scaling operations are synchronized to all views) and independent operation.

[0067] (c) Highlight differences: When comparing two schemes, the difference areas between the two schemes are marked with colors (green for new buildings, red for demolished buildings, and yellow for modified buildings), and a list of differences is provided (including changes in area, height, and function).

[0068] (d) Compliance Overview: Automatically perform "three-line" compliance checks on each plan and summarize the compliance status of each plan in tabular form (number of violations, distribution of violation types, area involved, etc.) to help decision-makers quickly screen compliant plans.

[0069] Reference Figure 11 As shown in Example 1: Application of BIM-GIS collaborative design in the preparation of a prefecture-level city's territorial spatial plan. Scenario Description: A prefecture-level city (with a central urban area of ​​approximately 120 km²) is conducting a territorial spatial planning process. The planning and design schemes for approximately 3,200 existing buildings and 5 key areas within the central urban area need to be incorporated into a unified BIM-GIS 3D collaborative platform for compilation and review. Participating units include a planning institute (responsible for the overall scheme and delineation of the "three lines" on the GIS side), an architectural design institute (responsible for the BIM design of architectural schemes in the key areas), and a municipal design institute (responsible for the BIM design of municipal pipelines), totaling three units.

[0070] Implementation steps: Step 1: Basic Data Preparation and TP-SMO Modeling Collect existing building BIM models (partially obtained through reverse modeling via laser scanning, LOD3 level; partly from design BIM models, LOD4 level) and GIS basic data (DEM, DOM, current land use, "three lines" delineation results, etc.). Construct a TP-SMO semantic mapping ontology model to establish a mapping relationship between each BIM building and GIS semantics.

[0071] TP-CSGM (Trans-Scale Geometric Transformation) was performed on 3200 buildings: LOD4→LOD3 took an average of 0.8 seconds per building, LOD3→LOD2 took an average of 0.3 seconds per building, and LOD2→LOD1 took an average of 0.1 seconds per building. After the transformation, the average length of the attribute inheritance chain was 3.2 levels, and the attribute retention rate was 94.3%.

[0072] Step 2: Build the OLT index An OLT index was constructed for all data within a 120km² area. The root node covers the entire central urban area (bounding box approximately 12km × 10km × 500m), with the hierarchical tree extending to Depth 13. The final OLT node count is approximately 2.4 million, including approximately 180,000 leaf nodes containing BIM data. Each leaf node carries data from 1 to 6 LOD levels, with a total data volume of approximately 860GB (all levels combined).

[0073] Step 3: Modeling the "Three-Line" 3D Control System The city's ecological protection red line, permanent basic farmland, and urban development boundary were each constructed as a three-dimensional parameterized control system. The ecological protection red line control system involves 12 protection zones, covering a total area of ​​approximately 35 km², with height restrictions set at 50m and 200m respectively, based on the type of ecological corridor. The permanent basic farmland control system involves 8 protected areas, covering a total area of ​​approximately 28 km², with the topsoil depth determined based on soil survey data (ranging from 0.8m to 1.8m). The urban development boundary control system is divided into 3 vertical zones (underground -15m, surface -24m low layer, and 24m-100m mid-to-high layer).

[0074] Step 4: Multi-role Collaborative Design The planning institute completes the overall plan and the delineation of the "three lines" (water, electricity, gas, and land use rights) on the GIS side and then publishes it to CCMR. The architectural design institute loads the BIM-GIS integrated base map in Revit (loading LOD2 / LOD3 data of the corresponding area as needed based on the number of levels) to carry out architectural design. The municipal design institute conducts BIM design of municipal pipelines in Civil 3D.

[0075] During the design process, when the architectural design institute submitted a design proposal for a 30-story high-rise building (98m in height) to CCMR, CPCE automatically triggered a compliance check and found that the building's height exceeded the aviation height restriction area (80m at that location) by 18m, immediately marking the violation in red in the BIM model. The architectural design institute adjusted the proposal to 25 stories (78m in height), and after resubmitting it, it passed the compliance check.

[0076] During the collaborative design process, a total of 47 conflicts between architectural plans and the "three lines" control were detected and resolved, 23 spatial collisions between buildings and municipal pipelines were resolved, and 15 planning indicators exceeded the limits.

[0077] Step 5: Solution Comparison and Output Three alternative planning schemes for key areas were compared using a three-dimensional quantitative method. The indicators for each scheme were automatically extracted: Scheme A: plot ratio 2.8, building density 32%, green space ratio 35%; Scheme B: plot ratio 3.2, building density 28%, green space ratio 38%; Scheme C: plot ratio 2.5, building density 35%, green space ratio 30%. Simultaneously, solar radiation analysis was conducted. Scheme B had the shortest solar radiation duration on residential windowsills on the winter solstice at 1.8 hours (meeting the 2-hour standard requirement with a compliance rate of 92%), Scheme A at 95%, and Scheme C at 98%.

[0078] The review committee selected Scheme A as the recommended scheme after comprehensive comparison through three-dimensional sliding comparison and indicator dashboard.

[0079] Reference Figure 12 As shown in Example 2: BIM-GIS collaborative management and control application in the underground space development planning of a new district. Scenario Description: A new district (approximately 45 km²) is undergoing underground space development and utilization planning. This requires comprehensive consideration of BIM design schemes for underground construction projects such as integrated utility tunnels, underground parking garages, underground commercial facilities, and subway lines, while coordinating with the three-dimensional control requirements of surface ecological protection red lines and permanent basic farmland. Five parties are involved, including the planning unit, the integrated utility tunnel design unit, and the subway design unit.

[0080] Implementation steps: Step 1: BIM Modeling and Semantic Annotation of Underground Space Each design firm submitted a BIM model (LOD4 level) of the underground facilities, including: integrated utility tunnels (approximately 28km long, buried at depths of -8m to -12m), underground parking garages (5 locations, buried at depths of -6m to -15m), subway lines (2 lines, tunnels buried at depths of -15m to -25m), and underground commercial facilities (2 locations, buried at depths of -5m to -18m). The planning extension attributes (land use function, burial depth range, structural type, etc.) were labeled for each BIM element according to the TP-SMO model.

[0081] Step Two: Construction of the "Three-Line" Three-Dimensional Control System The focus is on constructing a three-dimensional control system for permanent basic farmland. Based on soil profile data for the area, the depth of the topsoil varies between 0.8m and 1.5m. The lower boundary of the control system is set at 3m below the bottom of the topsoil (i.e., the deepest point is 4.5m below the surface) to ensure the integrity of the topsoil and buffer layers.

[0082] Within the ecological protection red line control system, for a protected area involving a wetland park, the lower boundary of the control system is extended to 5 meters below the ground surface (to protect the groundwater level and the wetland root system).

[0083] Step 3: Collaborative Compliance Pre-inspection After each design firm submits its BIM design proposal on the collaborative platform, CPCE automatically performs a compliance pre-check and identifies the following conflicts: (1) From K3+200 to K3+580 (approximately 380m long), the top of the utility tunnel structure is buried at a depth of -7.5m, encroaching on the lower boundary of the permanent basic farmland control body (the bottom boundary of the cultivated layer at this location is -1.2m, and the lower boundary of the control body is -4.2m), with a conflict volume of approximately 2850m³. CPCE immediately marked the violation and suggested that the utility tunnel section be lowered to -15m to avoid the control body.

[0084] (2) The edge of the foundation pit (depth -22m) of a station on Metro Line 2 is only 18m away from the boundary of the ecological protection red line control body, which is less than the buffer zone distance requirement (50m). Although the foundation pit structure does not directly intrude into the control body, the construction impact may affect the wetland ecosystem. Marked as "Orange Warning".

[0085] (3) The underground commercial facility B and the integrated utility tunnel section have a spatial intersection at a depth of -8m to -12m, with a conflict volume of approximately 1,200m³. This is an interdisciplinary collision between the two BIM models.

[0086] Step 4: Coordinated Adjustment and Solution Optimization Regarding the conflict (1), the integrated utility tunnel design unit lowered the elevation of the conflict section of the utility tunnel to -16m, and after resubmission, it passed the compliance pre-inspection.

[0087] In response to the conflict (2), the subway design unit and the planning unit jointly reviewed the situation in the three-dimensional scene and decided to add a water-proof curtain (depth -30m) on the side of the station foundation pit close to the ecological protection red line, and to monitor the settlement of the wetland and the groundwater level in real time during the construction period.

[0088] In response to the conflict (3), the integrated utility tunnel design unit and the underground commercial design unit worked together to adjust the route of the utility tunnel to avoid the underground commercial facilities. After the adjustment, the length of the new utility tunnel was about 120m, and the investment increased by about 3.6 million yuan, but the spatial conflict was eliminated.

[0089] Step 5: Output of Results The final plan passed all compliance checks. The system automatically generated a 3D control map of the underground space (including the spatial location, burial depth, and spatial relationship with the "three lines" control body of each facility), as well as an audit log of the collaborative design process (including all change records, conflict detection results, adjustment plans, etc.), which served as technical support materials for planning approval.

[0090] Reference Figure 13 As shown, further, combining the above-mentioned BIM-GIS fusion-based three-dimensional collaborative design method for land spatial planning, a BIM-GIS fusion-based three-dimensional collaborative design system for land spatial planning is proposed, including: The semantic mapping module is used to construct a BIM-GIS semantic mapping ontology model to achieve bidirectional mapping between the semantics of BIM building components and the semantics of GIS geographic elements, including cross-scale geometric transformation and attribute inheritance chain maintenance. The multi-scale data management module is used to build an axis-aligned bounding box hierarchical tree index structure and manage the storage, scheduling and memory optimization of multi-scale BIM-GIS fusion data. The 3D control modeling module is used to parameterize ecological protection red lines, permanent basic farmland, and urban development boundaries into 3D control volumes and manage the control rule set. The compliance verification module is used to automatically perform height limit detection, setback distance detection, underground space intrusion detection, and land use compatibility detection when BIM design schemes are submitted or changed, and generate a 3D compliance report. The collaborative design module supports real-time collaborative editing by multiple roles, concurrent conflict resolution, and three-level conflict pre-detection. The scheme comparison module is used to automatically extract planning quantitative indicators and supports three-dimensional sliding comparison of multiple schemes, highlighting differences, and providing an overview of compliance.

[0091] The audit log module is used to record all design change operations, conflict detection results, and adjustment plans, generating a complete audit chain that includes change time, person making the change, change content, conflict type, and handling result; The indicator calculation engine is used to automatically calculate quantitative indicators for planning.

[0092] The scheme comparison module supports comparing up to 5 planning schemes side by side in the same 3D scene, and each view supports synchronous linkage or independent operation; The difference comparison mode uses color to mark the new buildings, demolished buildings, and modified buildings between the two schemes, and lists the differences including changes in area, height, and function.

[0093] In summary, the advantages of this invention are as follows: by constructing a semantic mapping ontology model and a cross-scale geometric mapping algorithm, the controllable conversion of BIM building component semantics to GIS geographic element semantics is achieved. During the conversion process, the complete transmission of key planning attributes is ensured through the attribute inheritance chain. Through the hierarchical tree index structure and the LOD scheduling strategy based on the dual constraints of line-of-sight and planning levels, adaptive switching of data details is achieved. By constructing a three-dimensional parametric control model and real-time compliance verification, the efficiency of planning scheme design is improved. The three-dimensional comparison of multiple schemes and the difference highlighting function enable decision-makers to intuitively and quickly grasp the differences between schemes, significantly improving decision-making efficiency.

[0094] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion, characterized in that, include: Based on the IFC standard and the CityGML standard, a BIM-GIS semantic mapping ontology model is constructed, which includes a BIM semantic layer, a GIS semantic layer and a mapping bridging layer. Obtain land and space planning and design schemes; Based on the mapping bridging layer, the planning hierarchy is divided to obtain the corresponding detailed level data; Based on the three-dimensional control volume of the control lines in the mapping bridging layer, the compliance verification of the land space planning and design scheme is carried out. Obtain information on changes to the scheme design, including information on participating parties and corresponding change operation information; Based on the scheme design change information, perform operational conflict detection and obtain the scheme change results; Based on the land and space planning design scheme, the system automatically extracts quantitative indicators for land parcel and area planning and compares the schemes.

2. The three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion as described in claim 1, characterized in that, The aforementioned BIM-GIS semantic mapping ontology model, based on the IFC and CityGML standards, includes a BIM semantic layer, a GIS semantic layer, and a mapping bridging layer. Specifically, it includes: Based on the attribute set extension mechanism of the IFC4 standard, a new planning-related attribute extension set is added to the IFC standard semantic framework. The planning-related attribute extension set is encapsulated in the attribute set definition format of the IFC standard and includes land use code field, plot ratio constraint field, building height limit field, setback distance field and the code field of the planning management unit to which it belongs. Bind the planning-related attribute extension set to the entity type corresponding to the IFC4 standard to generate BIM semantic layer feature definitions; Based on the national land spatial planning standards, business specification information is obtained, which includes the core control information dimensions of three types of control objects: ecological protection red line, permanent basic farmland, and urban development boundary. Among them, the core control information dimensions include four dimensions: control space scope association, control rule constraints, compliance judgment conditions and vertical elevation benchmark; Based on business specification information, obtain the control attributes corresponding to each core control information dimension. The control attributes include the unique identifier of control line elements, the set of control rule entries, the Boolean expression for compliance judgment, and the national elevation datum code. Based on the extracted control attributes, field-based abstract modeling is performed, and the unique identifier of the control line element is abstracted into the control line three-dimensional control volume associated field. The field data type is string, which stores the unique identifier of the associated three-dimensional control volume. The set of control rule entries is abstracted into a control rule set field. The field data type is a structured array, and each array element contains three subfields: rule type, constraint parameters, and applicable conditions. The compliance judgment Boolean expression is abstracted into a compliance judgment logical expression field. The field data type is string, and it stores the compliance judgment condition expression connected by Boolean logical operators. The national elevation datum code is abstracted into an elevation datum field, and the field data type is an enumeration type; Based on the application domain extension mechanism of CityGML3.0, the above fields are encapsulated into a set of management semantic extensions; Bind the control semantic extension set to CityGML3.0 to obtain the GIS semantic layer feature definition; Based on the feature definitions of the BIM semantic layer and the GIS semantic layer, a mapping bridging layer is constructed to define the geometric transformation rules and attribute transformation rules between BIM semantics and GIS semantics; Based on the BIM semantic layer, GIS semantic layer, and mapping bridging layer, a BIM-GIS semantic mapping ontology model is constructed.

3. The three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion as described in claim 2, characterized in that, The mapping bridging layer constructed based on the feature definitions of the BIM semantic layer and the GIS semantic layer specifically includes: Obtain the BIM models of buildings in the land spatial planning and design scheme; Based on the geometric vertex coordinates of all building components in the building BIM model, obtain the axis-aligned bounding box corresponding to each building component. The axis-aligned bounding box is determined by the extreme coordinates of the building component. Based on the axis-aligned bounding boxes corresponding to each building component, an axis-aligned bounding box hierarchy tree is constructed by aggregating from bottom to top based on spatial containment relationships. Based on the principle of minimum description length, the component set is optimally binary partitioned to obtain the building component group; The building component groups near the building's perimeter are labeled as the shell component set, and the building component groups located inside the building are labeled as the interior component set; Based on the shell component set, the corresponding continuous contour triangular mesh is obtained using the Alpha Shape algorithm; Based on continuous contour triangular mesh, the building space is voxelized, and the occupying voxels are marked as solid voxels. The moving cube algorithm is used to extract isosurfaces from the boundaries of solid voxels and reconstruct a simplified boundary representation shell. Based on the set of internal components, the components are grouped according to the floors to which they belong, and the weighted average net height of the horizontal projected area of ​​all internal spaces on the corresponding floors of each group is taken as the net height of that group. Based on the projected contours of each group of internal components on the horizontal plane, the Douglas-Peucker algorithm is used to simplify the contours and obtain simplified internal polygons. Using the simplified polygons corresponding to each group of internal components as the base and the net height of the group as the top height, a prism is generated, retaining the floor number, area and net height, to obtain the simplified internal space. Obtain roof type information, which indicates the roof type corresponding to the building; Based on the boundary representation of the outer shell, the roof surface is simplified based on the roof type information, and the wall surfaces are uniformly extracted as vertical facades to obtain the building block model. Based on the projected outline of the building block model on the horizontal plane, the Douglas-Peucker algorithm is used to simplify the outline and obtain simplified polygons. Based on the boundary representation of the outer shell, simplified height information is obtained by using the area-weighted average of each floor elevation. A simplified building model is obtained by generating a prism with a simplified polygon as the base and a simplified height as the top height. Based on the boundary representation of the outer shell, the simplified internal space, the building block model, and the simplified building model, obtain cross-scale geometric mapping information.

4. The three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion as described in claim 3, characterized in that, The construction of a mapping bridging layer based on the feature definitions of the BIM semantic layer and the GIS semantic layer also includes: The BIM model is simplified into a boundary representation shell and a simplified internal space as a first-level geometric simplification; the boundary representation shell is simplified into a building block model as a second-level geometric simplification; the building block model is simplified into a simplified building model as a third-level geometric simplification. Based on each level of geometric simplification process, an attribute inheritance chain node is established for each simplified geometry, recording the simplified geometry identifier, the source geometry identifier list, the inherited attribute list, the lost attribute list, and the reverse reconstruction prompts; Based on the attribute type of the building component, obtain attribute inheritance information; Among them, structural parameter attributes are directly mapped to the corresponding attributes of GIS building objects; For functional type attributes, the BIM functional type of building components is mapped to the land use and sea use classification code in GIS; All attribute inheritance chain nodes are linked together according to a simplified hierarchy to form a complete attribute tracing chain.

5. The three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion as described in claim 4, characterized in that, The process of dividing the planning hierarchy based on the mapping bridging layer and obtaining the corresponding detailed level data specifically includes: Based on the axis-aligned bounding box hierarchy tree, the planning level is allocated according to the tree depth to obtain the planning level corresponding to each node. The planning level includes the city area, central urban area, management unit, land parcel and building. Among them, if the tree depth corresponding to the node in the axis-aligned bounding box hierarchy tree is [0, 2], it is classified as a city area; if the tree depth is [3, 5], it is classified as a central urban area; if the tree depth is [6, 8], it is classified as a management unit; if the tree depth is [9, 11], it is classified as a plot of land; if the tree depth is greater than or equal to 12, it is classified as a building. Based on the simplified model in the cross-scale geometric mapping information, the details of the geometric data are divided to obtain the geometric data detail level information. Each level of the simplified model in the cross-scale geometric mapping information corresponds to a geometric data detail level. Based on the current camera position, obtain the distance from the center of the bounding box of each node to the camera, and determine the target detail level of each node according to the distance threshold; Based on the target level of detail for each node, load the geometric data corresponding to the level of detail. If a user-specified planning level exists, the corresponding node of that level will be forcibly loaded to the specified detail level, and the data detail level of other level nodes will be reduced.

6. The three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion as described in claim 5, characterized in that, The compliance verification of the land spatial planning and design scheme based on the three-dimensional control volume of the control lines in the mapping bridging layer specifically includes: Based on the needs of national land spatial planning, a three-dimensional control system for control lines is constructed, which includes a three-dimensional control system for ecological protection red lines, a three-dimensional control system for permanent basic farmland, and a three-dimensional control system for urban development boundaries. Based on the spatial coordinates of the BIM model, and using the axis-aligned bounding box hierarchy tree, obtain the corresponding management unit level nodes and the three-dimensional control volume of the control lines. The elevation of the highest point in the BIM model is compared point by point with the corresponding control line 3D control body height limit surface. If the elevation of the highest point exceeds the height limit surface, a height limit violation is generated, and the violation amount is the difference between the elevation of the highest point and the elevation of the height limit surface. Based on the minimum distance between the BIM model axis-aligned bounding box and the boundary of the control line 3D control body on the horizontal plane, if the minimum distance is less than the setback distance of the control line 3D control body, a setback violation will be generated. Based on the 3D collision detection between the underground part of the BIM model and the permanent basic farmland control body, if the underground part of the BIM model intrudes into the protection depth range of the cultivated layer, an underground space intrusion violation will be generated. The system compares the functional usage codes and control rule entries based on the BIM model to check whether the building functions meet the access requirements of the area, generates land use compatibility test results, and obtains compliance verification information for land space planning and design schemes.

7. The three-dimensional collaborative design method for land spatial planning based on BIM-GIS fusion as described in claim 6, characterized in that, The process of detecting operational conflicts based on scheme design change information and obtaining scheme change results specifically includes: Obtain the design change operations of each participant, and formalize each change operation into a triplet structure of operation type, operation target element reference path and operation parameters. The operation type includes three categories: creation, modification and deletion. Based on the distributed version management of the central collaborative model repository, operation transformation is performed on two concurrent operations to ensure that the final state is consistent after different clients execute in different orders; For two creation operations, if different features are created in the same location, they are both retained. If the same feature is created and the globally unique identifier is the same, the later one is marked as a conflict and the user is notified. For two modification operations, if different attributes of the same element are modified, the modifications are merged; if different values ​​of the same attribute are modified, the winning strategy is written last based on the priority of the participants. In the event of a conflict between deletion and modification operations, the deletion operation takes precedence, and the modification operation is marked as invalid and the modifier is notified.

8. A three-dimensional collaborative design system for land spatial planning based on BIM-GIS integration, used to implement the design method as described in any one of claims 1-7, characterized in that, include: The semantic mapping module is used to construct a BIM-GIS semantic mapping ontology model to achieve bidirectional mapping between the semantics of BIM building components and the semantics of GIS geographic elements, including cross-scale geometric transformation and attribute inheritance chain maintenance. The multi-scale data management module is used to build an axis-aligned bounding box hierarchical tree index structure and manage the storage, scheduling and memory optimization of multi-scale BIM-GIS fusion data. The 3D control modeling module is used to parameterize ecological protection red lines, permanent basic farmland, and urban development boundaries into 3D control volumes and manage the control rule set. The compliance verification module is used to automatically perform height limit detection, setback distance detection, underground space intrusion detection, and land use compatibility detection when BIM design schemes are submitted or changed, and generate a 3D compliance report. The collaborative design module supports real-time collaborative editing by multiple roles, concurrent conflict resolution, and three-level conflict pre-detection. The scheme comparison module is used to automatically extract planning quantitative indicators and supports three-dimensional sliding comparison of multiple schemes, highlighting differences, and providing an overview of compliance.

9. A three-dimensional collaborative design system for land spatial planning based on BIM-GIS fusion as described in claim 8, characterized in that, Also includes: The audit log module is used to record all design change operations, conflict detection results, and adjustment plans, generating a complete audit chain that includes change time, person making the change, change content, conflict type, and handling result; The indicator calculation engine is used to automatically calculate quantitative indicators for planning.

10. A three-dimensional collaborative design system for land spatial planning based on BIM-GIS fusion as described in claim 9, characterized in that, The scheme comparison module supports the side-by-side comparison of up to 5 planning schemes in the same 3D scene, and each view supports synchronous linkage or independent operation. The difference comparison mode uses color to mark the new buildings, demolished buildings, and modified buildings between the two schemes, and lists the differences including changes in area, height, and function.

Images