A building engineering design method and system based on BIM

Abstract

This invention discloses a BIM-based architectural engineering design method and system, relating to the field of architectural engineering design technology. The method includes: extracting the geometric features of each component of the target building, including bounding box coordinates, center point, dimensions, and principal direction vectors; calculating the average bounding box side length based on the geometric features; dividing the building space into grid cells and marking dense grid cells; offsetting components within the dense grid cells along three-dimensional directions to generate extended feature vectors; performing principal component rotation and multiple orthogonal projection on the extended feature vectors to generate a two-dimensional hash index; mapping components to index buckets and neighboring buckets; combining components pairwise within the index buckets; analyzing three-dimensional overlapping intervals to identify conflicting component pairs; statistically analyzing the overlapping volume of conflicting component pairs and their coefficient of variation; and marking component pairs with abnormal volumes to form conflict marker objects for BIM-driven component position adjustment. This solves the problem of missed detection of conflict candidate pairs in highly overlapping areas of the building space.

Description

Technical Field

[0001] This invention relates to the field of architectural engineering design technology, and more specifically, to a BIM-based architectural engineering design method and system. Background Technology

[0002] In modern architectural engineering design, Building Information Modeling (BIM) technology is widely used for 3D component modeling, construction simulation, and collaborative design. BIM technology can digitally manage the spatial location, dimensions, attributes, and connections of building components, providing visualization and information support for engineering design. It also supports component collision detection, construction conflict analysis, and engineering optimization decisions, thereby improving the accuracy and efficiency of architectural design.

[0003] In BIM-based architectural design, clash detection based on mesh generation is a commonly used technique. This method calculates the bounding box of each component, divides the building space into equal-sized 3D mesh cells based on the bounding box boundaries, and maps the components to the overlay mesh. Within each mesh, all components are paired to generate candidate clash pairs. Adjacent meshes are considered when necessary to avoid missed detections, and a hash set is used to remove duplicate candidate pairs, thus forming an initial set of conflicting component pairs, providing input for subsequent accurate clash analysis.

[0004] However, in areas with highly overlapping architectural spaces, even with mesh generation techniques, dense components within the mesh can still generate a large number of candidate conflict pairs. The close proximity or partial overlap of the bounding boxes of components within dense areas dramatically increases the number of conflict pairs, raising the computational burden of subsequent collision analysis and potentially leading to missed or false detections. This limits the efficiency and accuracy of traditional mesh generation methods under complex spatial conditions.

[0005] In computer vision, Locality Sensitive Hash (LSH) is used for fast retrieval of high-dimensional feature data. This method maps similar feature vectors into the same bucket using a hash function, thus avoiding full comparison. LSH significantly reduces the number of candidate combinations and improves matching efficiency when processing spatially dense and high-dimensional data. It offers insights for conflict detection in dense BIM component areas: LSH can be used to map component features to a two-dimensional or high-dimensional index space to quickly filter potential conflict candidate pairs.

[0006] Applying the LSH (Local Hierarchical Structure) approach to conflict detection in BIM still faces new challenges. After component features are mapped to index buckets, actual collisions may still occur between components in different buckets, indicating a risk of missed detections. Furthermore, the design of the hash function and the selection of index parameters are sensitive to the accuracy of candidate pair generation, requiring consideration of component dimensions, clearance, and spatial offset to ensure comprehensive capture of conflicts in densely populated areas. These issues are particularly prominent in architectural design scenarios with highly overlapping spaces.

[0007] In summary, existing BIM conflict detection methods suffer from an excessive number of candidate pairs and potential missed detections in densely populated areas. Furthermore, directly analogous to high-dimensional feature indexing methods in computer vision still lacks adaptability.

[0008] To address the above problems, this invention proposes a solution. Summary of the Invention

[0009] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a BIM-based architectural engineering design method and system, which solves the problem of missed detection of conflict candidate pairs in highly overlapping areas of architectural space by expanding the three-dimensional features of dense grid unit components and combining them with two-dimensional hash index mapping.

[0010] To achieve the above objectives, the present invention provides the following technical solution: A BIM-based architectural engineering design method includes: extracting the geometric features of components of the target building; determining the bounding box side length of the components based on the geometric features analysis, dividing the target building space into several grid units, and marking dense grid units; extending the geometric features of the components in the dense grid units along the three-dimensional direction based on a preset offset to generate extended feature vectors; performing principal component rotation alignment and multiple orthogonal projection on the extended feature vectors to generate a two-dimensional hash index; writing each component identifier into the corresponding index bucket according to the two-dimensional hash index, and simultaneously mapping it to adjacent index buckets to form a correspondence between index buckets and component identifiers; combining component identifiers in each index bucket in pairs, analyzing the overlapping interval in three-dimensional space based on the corresponding bounding box coordinates, identifying component pairs that overlap in all three directions as conflicting component pairs, and recording their overlapping intervals to form a set of conflicting component pairs; and using coefficient of variation analysis to analyze the overlapping volume of conflicting component pairs, marking component pairs with larger volume deviations from the overall level to form conflict marker objects, which are used for BIM-driven component position adjustment.

[0011] This invention also includes a BIM-based architectural engineering design system, comprising a component feature extraction module, a spatial grid division module, a feature expansion module, an index mapping module, a conflict candidate generation module, and a conflict marking module: The component feature extraction module is used to extract the geometric features of the target building's components; the spatial grid division module is used to determine the bounding box side length of the components based on the analysis of the component geometric features, and to divide the target building space into several grid units, marking dense grid units; the feature expansion module is used to expand the component geometric features in the dense grid units along the three-dimensional direction based on a preset offset, generating expanded feature vectors; the index mapping module is used to perform principal component rotation on the expanded feature vectors. The system employs multiple orthogonal projections to generate a two-dimensional hash index. Each component identifier is written into its corresponding index bucket based on this hash index and simultaneously mapped to neighboring index buckets, establishing a correspondence between index buckets and component identifiers. A conflict candidate generation module combines component identifiers within each index bucket in pairs. Based on the corresponding bounding box coordinates, it analyzes the overlapping intervals in three-dimensional space, identifying component pairs that overlap in all three directions as conflicting component pairs and recording their overlapping intervals to form a set of conflicting component pairs. A conflict marking module analyzes the overlapping volume of conflicting component pairs using the coefficient of variation, marking component pairs whose volumes deviate significantly from the overall level, thus creating conflict marking objects for BIM-driven component position adjustments.

[0012] The present invention also includes an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a BIM-based architectural engineering design method.

[0013] The present invention also includes a computer-readable storage medium storing a computer program that, when executed by a processor, implements a BIM-based architectural engineering design method.

[0014] The technical effects and advantages of the BIM-based architectural engineering design method and system of this invention are as follows: 1. This invention extends the geometric features of components in dense grid cells along three dimensions to generate extended feature vectors. It then combines principal component rotation alignment and multiple orthogonal projection to generate a two-dimensional hash index, enabling the mapping of components to index buckets and neighboring index buckets. This technology significantly reduces the risk of missed detections in highly overlapping areas of building space, improves the comprehensive coverage of conflict candidate pairs, and maintains processing efficiency. In practical applications, even with a large number of dense components, feature extension and neighboring index mapping can quickly screen potential conflict pairs, providing a complete and reliable data foundation for subsequent conflict analysis, thereby optimizing the spatial conflict identification effect of BIM models.

[0015] 2. This invention analyzes the coefficient of variation of the overlapping volumes of conflicting component pairs, marking component pairs with significantly larger volume deviations from the overall level to form conflict markers, which are then used to drive component position adjustments. This technology can automatically identify component pairs with abnormal spatial overlap, helping engineering designers prioritize key conflicts and achieve refined adjustments. Compared to traditional methods, this invention can highlight key conflicts while ensuring comprehensive candidate conflict coverage, improving the decision-making efficiency and construction controllability of BIM architectural engineering design, and reducing collision risks and rework probabilities during the construction phase. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a BIM-based architectural engineering design method according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, Figure 1 This invention presents a BIM-based architectural engineering design method, comprising the following steps: S1 extracts the geometric features of the target building's components.

[0019] In this embodiment, the extraction of the geometric features of the target building components specifically includes: Acquire the three-dimensional geometric data of all components in the building information model, including component boundary surfaces, node coordinates, and spatial connection relationships; Iterate through the geometric data of each component, calculate the minimum and maximum coordinates of all boundary points along the three-dimensional direction, generate the minimum bounding box of the component, and record the correspondence between the bounding box coordinates and the component identifier. Calculate the center point coordinates of each bounding box and associate the center coordinates with the corresponding component identifier; Extract the length, width, and height dimensions of each bounding box to form a dimension vector, and associate it with the corresponding component identifier; Analyze the main axes of each component in three-dimensional space, analyze the principal direction vector through geometric points, and associate the principal direction vector with the component identifier; The geometric features of a component are formed by combining its identifier, bounding box coordinates, center coordinates, dimension vector, and principal direction vector.

[0020] In this embodiment, the geometric features of the target building's components are extracted to obtain complete 3D geometric data for each component in the Building Information Model (BIM), including component boundary surfaces, node coordinates, and spatial connections. This enables accurate spatial positioning and dimensional analysis of each component in subsequent steps. Each component in the BIM consists of multiple boundary points and nodes in 3D space. These points collectively define the component's actual shape, size, and spatial relationships with other components. For example, a beam component may consist of eight corner points and several intermediate nodes. The coordinates of these points are used to generate the component's bounding box and center point.

[0021] In this embodiment, the minimum and maximum coordinates of all boundary points along the three-dimensional direction are calculated to provide a basis for generating the minimum bounding box of the component. Specifically, all boundary points of the component are traversed, and the coordinate values ​​of each point in the X, Y, and Z directions are compared. The minimum and maximum values ​​in each direction are recorded, and these values ​​define the boundary of the bounding box in space. For example, for a rectangular column component, if the minimum coordinate of its boundary point in the X direction is 1 meter and the maximum coordinate is 3 meters, then the length of the bounding box in the X direction is 2 meters; the same applies to the Y and Z directions. The bounding box generated in this way can completely enclose all geometric points of the component.

[0022] In this embodiment, the center coordinates of each bounding box are calculated to provide a reference for subsequent component alignment and spatial analysis. The center coordinates are obtained by adding the minimum and maximum coordinates in the X, Y, and Z directions and taking the midpoint. For example, if the minimum coordinate in the X direction is 1 meter and the maximum coordinate is 3 meters, then the center X coordinate is 2 meters. The center coordinates are not only used to represent the spatial position of the components, but also serve as a core component of the feature vector in subsequent mesh mapping and hash index generation.

[0023] In this embodiment, the length, width, and height dimensions of each bounding box are extracted to form a dimension vector, providing a quantitative basis for analyzing the spatial occupancy and interrelationships of components. The dimension vector is calculated by subtracting the minimum coordinate from the maximum coordinate. For example, if the aforementioned beam component has a length of 2 meters in the X direction, 0.5 meters in the Y direction, and 0.6 meters in the Z direction, then the dimension vector is [2, 0.5, 0.6]. The dimension vector plays a crucial role in component conflict analysis, mesh generation, and feature expansion.

[0024] In this embodiment, the principal axis of each component in three-dimensional space is analyzed to provide a basis for determining the directional characteristics of the component. The principal direction vector is obtained through geometric analysis of the spatial distribution of the component's boundary points. For example, for a beam component, its principal direction vector can be determined by connecting the nodes at both ends. The principal direction vector reflects the major axis direction of the component in space and has a guiding role for subsequent feature rotation, projection, and hash index mapping.

[0025] In this embodiment, the identifier, bounding box coordinates, center coordinates, size vector, and principal direction vector of each component are combined to form the component's geometric features, providing a complete data foundation for subsequent mesh generation, dense cell analysis, and collision detection. The geometric features of each component can be regarded as a comprehensive vector containing spatial location, size, and orientation information, which is used to support feature expansion, hash index generation, and accurate screening of candidate collision pairs.

[0026] It should be noted that in the process of generating the minimum bounding box and dimension vector, the boundary point coverage of the complex shape of the component needs to be considered, such as the oblique end face of the beam or the bend of the pipe. These boundary points must be included in the calculation to ensure that the bounding box can completely wrap the component.

[0027] It should be noted that when determining the main direction vector of a component, the distribution of multiple nodes can be statistically analyzed, and the direction with the most significant distribution can be taken as the main direction, while the secondary directions are retained for subsequent feature rotation and two-dimensional projection reference.

[0028] It should be noted that obtaining the center coordinates and dimension vectors is not only used to describe the spatial information of the component itself, but also provides basic parameters for the subsequent generation of perturbation vectors and extended feature vectors along the three-dimensional direction, ensuring that feature expansion can cover possible overlapping positions in dense areas.

[0029] It should be noted that the integrity of the component's geometric features directly affects the accuracy of subsequent mesh generation and hash index mapping. Therefore, all components must be fully traversed during the extraction process, and the complete recording of boundary points, nodes, and spatial connection relationships must be ensured.

[0030] It should be noted that the process of constructing the geometric features of components can be stored in matrix form. For example, the feature vector of each component can be stored as a row matrix, where the columns correspond to the minimum coordinates, maximum coordinates, center coordinates, size vector components, and principal direction vector components of X, Y, and Z, respectively, which facilitates subsequent batch processing and algorithm calculation.

[0031] It should be noted that the extraction of component geometric features not only provides a basis for bounding box generation and spatial mesh division, but also provides a unified data foundation for the generation of conflict candidate pairs, feature expansion and hash index mapping, enabling the data processing flow of the entire BIM building engineering design method to be executed efficiently and accurately.

[0032] The significance of this step lies in systematically and completely extracting the geometric features of the components, thereby achieving an accurate description of the spatial location, size, and orientation of the building components. This provides the necessary basic data for subsequent mesh generation, dense unit analysis, feature expansion, and conflict detection, thus ensuring the accuracy and reliability of the entire BIM building engineering design method in high-density, multi-component environments.

[0033] S2 determines the side length of the bounding box of the component based on the geometric feature analysis of the component, and divides the target building space into several grid units and marks the dense grid units.

[0034] In this embodiment, the step of determining the bounding box side length of the component based on the component's geometric features and dividing the target building space into several grid units, and marking dense grid units, specifically involves: Calculate the length, width, and height of the bounding box of all component geometric features, calculate the average value of each dimension, and then calculate the weighted variance to generate the spatial average side length. Based on the average side length of the space, the building space is divided into grid units of equal size along the three-dimensional direction, grid unit objects are generated and their spatial coordinate range is recorded; Traverse the geometric features of each component, determine the spatial coverage relationship between its bounding box and each mesh cell, and add the component identifier to the component list covering the mesh cells; The number of components in each grid cell is counted, and grid cells with a number exceeding a preset threshold are marked as dense grid cells, generating a list of dense grid cells.

[0035] In this embodiment, the bounding box length, width, and height of all component geometric features are statistically analyzed to provide basic data for analyzing component space occupancy and mesh generation. Specifically, all components are traversed, and the length, width, and height of each component's bounding box in the X, Y, and Z directions are extracted and summarized into three sets. For example, if a building has three beam components with X-direction lengths of 2 meters, 3 meters, and 2.5 meters respectively, then the X-direction length set is [2, 3, 2.5]. The same applies to the Y and Z directions.

[0036] In this embodiment, the average value of each dimension is calculated and weighted variance is added to generate the average side length of the space, providing a reference standard for the grid cell size. The average value is obtained by summing the bounding box lengths of each dimension and dividing by the total number of components. The weighted variance is calculated by summing the squares of the deviations of each component length from the average value, and then adjusting according to the number of components to obtain the overall spatial size distribution. These values ​​are used to determine the size of the grid, ensuring that the grid covers the space without being too sparse.

[0037] In this embodiment, the building space is divided into equally sized grid units along three dimensions, based on the average side length, providing structured spatial partitioning for component mapping and conflict candidate screening. Specifically, a continuous three-dimensional grid unit matrix is ​​generated in the X, Y, and Z directions, with the average side length as the spacing, starting from the smallest coordinate in the building space. Each grid unit records its spatial range. For example, if the X direction ranges from 0 meters to 10 meters, with an average side length of 2 meters, then 5 grid units are generated, each with a range of 0-2 meters, 2-4 meters, etc.

[0038] In this embodiment, the geometric features of each component are traversed to determine the spatial coverage relationship between its bounding box and each grid cell, providing data support for the component list of the grid cells. Specifically, the bounding box coordinates of the component are compared with the boundary of each grid cell. When the bounding box of the component intersects with the grid cell in the X, Y, and Z directions, the component identifier is added to the component list of that grid cell. For example, if a pipe spans two grid cells, the component identifier is added to the lists of both grid cells simultaneously.

[0039] In this embodiment, the number of components within each grid cell is counted, and grid cells with a number exceeding a preset threshold are marked as dense grid cells. This allows for focusing on high-density areas for subsequent feature expansion and conflict detection. The preset threshold can be determined based on building type or design experience; for example, setting it to 3 would mark grid cells containing three or more components as dense grid cells. This list of dense grid cells provides a basis for subsequent feature vector generation and conflict candidate pair screening.

[0040] It should be noted that the generation of the average side length of the space not only considers the average length of the components, but also should be adjusted according to the weighted variance to ensure that the grid division can take into account both the uniformity of spatial coverage and the resolution of dense areas, and avoid the omission or redundancy of candidate conflict pairs in some areas due to the grid being too small or too large.

[0041] It should be noted that when dividing the grid and marking dense cells, the geometry and spatial distribution of the component bounding boxes should be considered at the same time. For example, for long beams or pipes that span multiple grids, their identifiers should be added to all covering grid cells to ensure that no conflict candidate pairs in high-density areas are missed.

[0042] It should be noted that the grid cell object not only records the spatial coordinate range and the list of components it contains, but also records information such as component coverage ratio and component type. This provides additional attribute support for subsequent extended feature vector generation, hash index mapping, and collision detection, while improving the accuracy and efficiency of collision analysis in dense areas.

[0043] The significance of this step lies in dividing the building space into structured grid units by analyzing the geometric features of the components and generating the average side length of the space. At the same time, high-density areas are marked, providing a spatial positioning basis for subsequent extended feature vector generation, two-dimensional hash index mapping, and conflict candidate pair screening. This enables the BIM building engineering design method to efficiently and accurately identify potential conflicts in complex spatial environments.

[0044] S3 extends the geometric features of components in dense grid cells along the three-dimensional direction based on a preset offset to generate extended feature vectors.

[0045] In this embodiment, the step of extending the geometric features of components in the dense mesh cell along the three-dimensional direction based on a preset offset to generate an extended feature vector is specifically as follows: Traverse each dense grid cell, obtain the component geometric features of all components within it, including center coordinates, bounding box size and principal direction vector, and associate the component identifier with the original feature vector; Along the length, width, and height of the component bounding box, set the offset step size by dividing the minimum design clearance by two, and generate positive and negative offset sequences respectively; Based on the offset step size, a perturbation combination is generated in three spatial directions. By combining the original center coordinates with the offsets in each direction, a new center position is formed, while keeping the original bounding box size and main direction vector unchanged, and a set of perturbation vectors is generated. The original feature vector is merged with the disturbance vector to form an extended feature vector. The extended feature vector records the component identification, the center coordinates after disturbance, the size vector, and the main direction vector. The operation of expanding the feature vector is repeated for each component within the dense grid cell, generating perturbation vector combinations sequentially along the length, width, and height directions; The extended feature vectors generated by all components within the dense grid cell are aggregated to form an extended feature vector set.

[0046] In this embodiment, the geometric features of all components in the dense mesh cell are obtained to provide initial data for generating the extended feature vector. Specifically, each dense mesh cell is traversed, and the center coordinates, bounding box size, and principal direction vector of each component are extracted. This information is then associated with the corresponding component identifier. For example, if a pipe has center coordinates of (2.5, 3.0, 1.2) meters, a size vector of (0.5, 0.5, 3.0) meters, and a principal direction along the Z-axis, then the component identifier combined with these geometric features forms the initial feature vector.

[0047] In this embodiment, the offset step size is set by dividing the minimum design clearance by two along the length, width, and height directions of the component bounding box, providing precise control over component characteristic disturbances. The minimum design clearance is the minimum safe distance between components specified in the design specifications, for example, 50 mm, so the offset step size is 25 mm. Positive and negative offset sequences are generated in the X, Y, and Z directions respectively, for example, the offset in the X direction is +25 mm and -25 mm, and the same applies to the Y and Z directions, providing a numerical basis for subsequent disturbance combination generation.

[0048] In this embodiment, perturbation combinations are generated in three spatial directions based on the offset step size. A new center position is formed by combining the original center coordinates with the offsets in each direction, creating multiple candidate positions for each component. Specifically, the positive and negative offsets in the X, Y, and Z directions are triple-combined, for example, X direction +25 mm, Y direction -25 mm, Z direction +25 mm. The new center coordinates are the original center coordinates plus the offsets in each direction, while keeping the original bounding box size and main direction vector unchanged, generating a set of perturbation vectors.

[0049] In this embodiment, the original feature vector and the perturbation vector are merged to form an extended feature vector, providing redundant candidates for collision detection in dense areas. The extended feature vector records the component identifier, the perturbed center coordinates, the size vector, and the main direction vector. For example, if the original feature vector is (component identifier A, center (2.5, 3.0, 1.2), size (0.5, 0.5, 3.0), direction Z), after generating the perturbation vector, the new center coordinates are (2.525, 2.975, 1.225), then the extended feature vector is (component identifier A, center (2.525, 2.975, 1.225), size (0.5, 0.5, 3.0), direction Z).

[0050] In this embodiment, the operation of expanding feature vectors is repeated for each component within the dense grid cell, generating perturbation vector combinations sequentially along the length, width, and height directions, providing a complete spatial perturbation sample set for each component. Through the combination of positive and negative offsets in the three directions, each component may generate multiple sets of expanded feature vectors. For example, if there are two sets of positive and two sets of negative vectors in each of the three directions, then each component generates 8 sets of perturbation vector combinations, forming a cubic sample distribution.

[0051] In this embodiment, the extended feature vectors generated by all components within the dense grid cell are aggregated to form an extended feature vector set, which provides input data for subsequent principal component rotation alignment, multiple orthogonal projection, and two-dimensional hash index mapping. Each element in the set contains a component identifier, perturbed center coordinates, size vector, and principal direction vector, which can be directly used to construct conflict candidate pairs and neighbor bucket mappings.

[0052] It should be noted that the reason for setting the minimum design net distance divided by two as the offset step size is to generate enough disturbance samples while ensuring that there is no physical collision between components, so as to alleviate the problem of missed detection that may occur in dense areas of mesh division, and at the same time make the candidate conflict pairs more comprehensive.

[0053] It should be noted that the perturbation combination not only considers the positive and negative offsets in each direction, but also makes weighted adjustments according to the length, width and height ratio of the components. For example, longer beams generate more offset points along their length to increase the detection density of high-risk conflict areas and ensure that no conflict candidate pairs in dense areas are missed.

[0054] It should be noted that the center coordinates in the extended feature vector set maintain a correspondence with the center coordinates of the original components. Each component corresponds to multiple sets of perturbation centers in the set, with the size and main direction remaining unchanged. This allows both the original position and the perturbation position to be considered simultaneously in the subsequent two-dimensional hash index mapping, thereby improving the comprehensiveness and accuracy of collision detection.

[0055] The significance of this step lies in generating extended feature vectors by performing three-dimensional perturbation on the components in the dense grid cells. This provides sufficient spatial samples for subsequent conflict candidate pair generation, two-dimensional hash index mapping, and conflict detection in high-density areas of BIM, thereby effectively mitigating the risk of missed detection caused by high spatial overlap and grid division, and improving the accuracy and reliability of conflict detection in architectural engineering design.

[0056] S4. Perform principal component rotation alignment and multiple orthogonal projection on the extended feature vector to generate a two-dimensional hash index. Write each component identifier into the corresponding index bucket according to the two-dimensional hash index, and simultaneously map it to the neighboring index buckets to form a correspondence between index buckets and component identifiers.

[0057] In this embodiment, the process of performing principal component rotation alignment and multiple orthogonal projection on the extended feature vector to generate a two-dimensional hash index, and writing each component identifier into the corresponding index bucket according to the two-dimensional hash index, while simultaneously mapping it to neighboring index buckets to form a correspondence between index buckets and component identifiers, is as follows: The center coordinates of all extended feature vectors in the statistical dense grid cell are used to identify the direction with the largest variance in the three spatial dimensions, which is then taken as the first direction. Adjust the coordinates of each extended feature vector according to the first direction, so that the extended feature vectors are aligned along the first direction, and rotate the deviation of the main direction of the component to be consistent with the first direction. Based on the rotated and aligned coordinates, the direction perpendicular to the first direction is selected as the second direction. Each extended feature vector is projected onto the first and second directions to generate a two-dimensional hash index. According to the preset two-dimensional grid size, the two-dimensional hash index is mapped to the corresponding index bucket, and the component identifier contained in each index bucket is recorded; Offset mapping is performed on the two-dimensional hash index along the horizontal and vertical directions, and the component identifier is written to the adjacent index bucket, thus forming a correspondence between the index bucket and the component identifier.

[0058] In this embodiment, the center coordinates of all extended feature vectors in the dense grid cells are statistically analyzed to provide basic data for determining the principal direction and rotation alignment. Specifically, each extended feature vector is traversed, and its center coordinates are extracted. For example, the center coordinates of a component are (2.525, 2.975, 1.225) meters. The center coordinates of all components are formed into a three-dimensional coordinate set for analyzing the distribution in each spatial dimension.

[0059] In this embodiment, the direction with the largest variance in the three spatial dimensions is identified to provide a basis for determining the primary direction and rotation alignment. By analyzing the variance of all center coordinates in the X, Y, and Z directions, the dimension with the largest variance is selected as the first direction. For example, if the X direction has the largest variance in a group of components, then the first direction is set as the X-axis direction. This direction reflects the main distribution trend of dense components in space.

[0060] In this embodiment, the coordinates of each extended feature vector are adjusted according to the first direction to align the extended feature vectors along the first direction, while the deviation of the main direction of the component is rotated to be consistent with the first direction. Specifically, based on the main direction vector of each component, the deviation angle between it and the first direction is calculated, and the center of the component is rotated in space around an axis perpendicular to the first direction to make the main direction coincide with the first direction, ensuring that the components in the dense area have consistent orientation in projection and hash mapping.

[0061] In this embodiment, based on the rotated and aligned coordinates, a direction perpendicular to the first direction is selected as the second direction. Each extended feature vector is projected onto the first and second directions to generate a two-dimensional hash index. Specifically, the rotated center coordinates of each component are extracted along the first and second directions to form two-dimensional coordinates, for example, the projected coordinates are (X', Y'), which are used for subsequent two-dimensional mesh mapping.

[0062] In this embodiment, according to a preset two-dimensional grid size, the two-dimensional hash index is mapped to the corresponding index bucket, and the component identifier contained in each index bucket is recorded. For example, if the two-dimensional grid size is set to 0.5 meters × 0.5 meters, each two-dimensional coordinate is rounded and mapped to the grid number, and the component identifier is added to the corresponding index bucket list to form a preliminary index mapping relationship.

[0063] In this embodiment, the two-dimensional hash index is offset and mapped along the horizontal and vertical directions, and the component identifier is written into the neighboring index bucket, thus forming a correspondence between the index bucket and the component identifier. Specifically, each two-dimensional coordinate is added to or subtracted by one grid unit in the X and Y directions to obtain the neighboring grid number. The component identifier is then added to the neighboring index bucket list to ensure that even if the center of the component is close to the grid boundary, it can still be detected by the neighboring bucket, reducing the probability of missed detection.

[0064] It should be noted that the direction with the largest variance is used for the first direction selection. Its calculation takes into account the distribution of the center coordinates of all extended feature vectors. By statistically analyzing the dispersion of different directions, it reflects the spatial distribution trend, ensuring that the projection after rotation and alignment can more tightly aggregate components and improve the efficiency of hash index mapping.

[0065] It should be noted that the preset two-dimensional grid size in the two-dimensional hash index is a key parameter set manually. Its value needs to be determined in combination with the component size and density. For example, the grid size can be appropriately increased for large beams and pipes, and decreased for small components and high-density areas, in order to balance the number of index buckets and the collision detection coverage.

[0066] It should be noted that the offset range of the neighbor index bucket mapping can be dynamically adjusted according to the maximum size of the component and the minimum design clearance. For example, it can be offset by one grid unit in both the X and Y directions to ensure that the component can enter the neighbor bucket candidate set even if it is located at the edge of the grid, thereby further improving the conflict candidate coverage rate in dense areas.

[0067] The significance of this step lies in mapping the three-dimensional extended feature vector to the two-dimensional hash index through principal component rotation alignment and multiple orthogonal projection, forming a correspondence between index buckets and component identifiers, and alleviating the boundary omission problem of dense areas through neighbor bucket mapping, thus providing an efficient data structure foundation for subsequent generation of conflict component pairs and accurate spatial conflict detection.

[0068] S5. Combine the component identifiers in each index bucket in pairs, analyze the overlapping interval in three-dimensional space based on the corresponding bounding box coordinates, identify the component pairs that overlap in all three directions as conflicting component pairs, record their overlapping intervals, and form a set of conflicting component pairs.

[0069] In this embodiment, the component identifiers within each index bucket are paired, and the overlapping intervals in three-dimensional space are analyzed based on the corresponding bounding box coordinates. Component pairs that overlap in all three directions are identified as conflicting component pairs, and their overlapping intervals are recorded to form a set of conflicting component pairs. Specifically: The component identifiers within the index bucket are paired to form component pairs, and each component pair is associated with the corresponding bounding box coordinates. Extract the minimum and maximum coordinates of the bounding boxes of the two components in each component to form a three-dimensional spatial interval; Determine whether the spatial intervals of the two components in each component pair intersect in the X, Y, and Z directions. If they intersect in all three directions, mark the component pair as a conflicting component pair and record the minimum and maximum coordinates of the overlapping interval of the conflicting component pair. Traverse all component pairs in the index bucket and summarize the conflicting component pairs to form an initial set of conflicting component pairs; The initial set of conflicting component pairs in all index buckets is deduplicated to generate the final set of conflicting component pairs.

[0070] In this embodiment, the component identifiers within the index bucket are paired to provide basic data for generating potentially conflicting component pairs. Specifically, the component identifier list in each index bucket is traversed, and every two different components in the list are combined into a pair. For example, if there are components A, B, and C in the index bucket, component pairs (A,B), (A,C), and (B,C) are generated, and each component pair is associated with its corresponding bounding box coordinates for subsequent spatial intersection calculations.

[0071] In this embodiment, the minimum and maximum coordinates of the bounding boxes of the two components in each component pair are extracted to provide a basis for determining the three-dimensional spatial interval. Specifically, the minimum and maximum coordinates in the X, Y, and Z directions are obtained for the bounding box of each component. For example, the bounding box of component A has X direction coordinates of (1.0, 2.5) meters, Y direction coordinates of (0.5, 1.5) meters, and Z direction coordinates of (0.0, 1.0) meters, while the bounding box of component B has X direction coordinates of (2.0, 3.0) meters, Y direction coordinates of (1.0, 2.0) meters, and Z direction coordinates of (0.5, 1.5) meters. These coordinates are used to form the three-dimensional spatial interval data of the component pair.

[0072] In this embodiment, it is determined whether the spatial intervals of the two components in each component pair overlap in the X, Y, and Z directions, providing a standard for identifying conflicting component pairs. Specifically, the minimum and maximum coordinates in the X, Y, and Z directions are compared respectively. If the coordinate intervals of the two components overlap in the X direction, and the coordinate intervals in the Y and Z directions also overlap, then the component pair is marked as a conflicting component pair, and the minimum and maximum coordinates of the overlapping interval of the conflicting component pair are recorded. For example, the overlap of components A and B in the X direction is (2.0, 2.5) meters, in the Y direction is (1.0, 1.5) meters, and in the Z direction is (0.5, 1.0) meters.

[0073] In this embodiment, all component pairs within the index bucket are traversed, and conflicting component pairs are aggregated to form an initial set of conflicting component pairs, providing a candidate set for overall conflict detection. Specifically, each conflicting component pair is recorded in the set according to its identifier and overlapping range, for example, as [(component A, component B, (2.0, 2.5), (1.0, 1.5), (0.5, 1.0))]. The set contains all conflicting pairs detected in the current index bucket.

[0074] In this embodiment, the initial set of conflicting component pairs in all index buckets is deduplicated to generate the final set of conflicting component pairs, providing non-repeating conflict data for subsequent accurate analysis. Specifically, the initial set is traversed, and only one record is retained for the same component pair that appears multiple times, ensuring that each pair of conflicting components is unique in the final set. For example, if component pair (A, B) appears multiple times in multiple index buckets, only one record is retained.

[0075] It should be noted that the judgment of three-dimensional spatial intervals not only compares the boundary coordinates, but also needs to consider the component size, tolerance and clearance requirements. In actual calculations, a minimum clearance threshold can be set for the overlap condition to avoid misjudgment or missed detection caused by small gaps, thereby improving the accuracy of conflict detection.

[0076] It should be noted that the order of pairwise combinations within the index bucket can be generated according to the component identifier sorting or other stable methods to ensure the effectiveness of deduplication. Furthermore, in high-density areas, by reasonably selecting the index bucket size and neighbor bucket mapping, the number of candidate pairs can be effectively controlled, balancing computational load and detection accuracy.

[0077] The significance of this step lies in systematically generating pairwise combinations of components within the index bucket and analyzing their overlap in three-dimensional space to form a set of conflicting component pairs, providing a complete and accurate candidate data foundation for subsequent conflict volume assessment, marking, and BIM-driven component position adjustment.

[0078] S6 uses coefficient of variation analysis to identify overlapping volumes of conflicting component pairs, marking component pairs whose volumes deviate significantly from the overall level, thus creating conflict marker objects for use in BIM-driven component position adjustments.

[0079] In this embodiment, the overlapping volume of conflicting component pairs is analyzed based on the coefficient of variation. Component pairs with significantly larger volume deviations from the overall level are marked to form conflict marker objects, which are used for BIM-driven component position adjustment. Specifically: Extract the side lengths in the X, Y, and Z directions of the overlapping interval of each component pair in the conflict component pair set to form the overlapping volume; Calculate the overlap volume of all conflicting component pairs, and analyze the mean and standard deviation of the overlap volume; The coefficient of variation of the overlapping volume is generated based on the mean and standard deviation, and the overlapping volume of each component pair is compared with the coefficient of variation. When the overlap volume of a component pair deviates from the overall average level by more than a preset threshold, the component pair is marked as a conflict object, and the corresponding component identifier and overlap interval coordinates are recorded. Traverse the set of conflicting component pairs, mark all components that deviate from the threshold, and generate the final conflict mark object, which is used to drive component position adjustment in BIM.

[0080] In this embodiment, the side lengths in the X, Y, and Z directions of the overlapping interval of each component pair in the set of conflicting component pairs are extracted to provide basic data for calculating the overlapping volume. Specifically, for each conflicting component pair, the maximum coordinate in the X direction is subtracted from the minimum coordinate, the maximum coordinate in the Y direction is subtracted from the minimum coordinate, and the maximum coordinate in the Z direction is subtracted from the minimum coordinate. The three side lengths are then multiplied to generate the overlapping volume of the component pair. For example, if the overlapping interval of component pair A and B is (2.0, 2.5) meters in the X direction, (1.0, 1.5) meters in the Y direction, and (0.5, 1.0) meters in the Z direction, then the overlapping volume is 0.5 meters × 0.5 meters × 0.5 meters = 0.125 cubic meters.

[0081] In this embodiment, the overlapping volumes of all conflicting component pairs are statistically analyzed, and the average and standard deviation of the overlapping volumes are calculated to provide a data basis for subsequent calculation of the coefficient of variation. Specifically, the overlapping volumes of each component pair in the conflicting component pair set are summed and divided by the number of component pairs to obtain the average. At the same time, the sum of squares of the deviations of each volume from the average is calculated and divided by the square root of the number of component pairs to obtain the standard deviation. For example, if the overlapping volumes of three conflicting component pairs in the set are 0.125, 0.200, and 0.150 cubic meters, respectively, then the average is 0.158 cubic meters and the standard deviation is 0.038 cubic meters.

[0082] In this embodiment, the coefficient of variation of the overlapping volume is generated based on the mean and standard deviation, and the overlapping volume of each component pair is compared with the coefficient of variation to provide a criterion for identifying abnormal conflicts. Specifically, the mean is subtracted from the overlapping volume of each component pair and divided by the standard deviation to obtain the standardized deviation of the component pair. By comparing it with a preset deviation threshold, it is determined whether the component pair belongs to the conflict object with abnormal volume. For example, if the preset threshold is 1.5 times the standard deviation, component pairs that exceed this deviation will be marked.

[0083] In this embodiment, when the overlapping volume of a pair of components deviates from the overall average level by more than a preset threshold, the pair of components is marked as a conflict object, and the corresponding component identifier and overlapping interval coordinates are recorded to provide target data for BIM-driven component position adjustment. Specifically, an entry is created in the conflict object set to record the component pair identifier, the minimum and maximum overlapping coordinates in the X / Y / Z directions, and the calculated overlapping volume, ensuring that each abnormal conflict can be tracked and located.

[0084] It should be noted that the calculation of overlapping volume not only considers geometric dimensions, but also needs to be adjusted in conjunction with the shape of the component and installation tolerance. For example, at irregular components or pipeline intersections, the volume can be simplified or calculated in blocks to improve calculation efficiency and maintain the reliability of the results.

[0085] It should be noted that the preset deviation threshold can be customized according to the building type, component importance and safety specifications. The choice of threshold directly affects the number of conflict marked objects and the BIM location adjustment strategy. At the same time, it can be dynamically adjusted in the software to adapt to the accuracy requirements of different projects.

[0086] The significance of this step lies in the fact that by quantitatively analyzing the overlapping volume of conflicting components and combining it with the coefficient of variation for anomaly marking, it enables the accurate identification of components that may have serious spatial interference. This provides actionable target data for the subsequent automatic adjustment of component positions in the BIM system, thereby improving the safety and feasibility of architectural engineering design.

[0087] Example 2: The present invention also includes a BIM-based architectural engineering design system, comprising a component feature extraction module, a spatial grid division module, a feature expansion module, an index mapping module, a conflict candidate generation module, and a conflict marking module. The component feature extraction module is used to extract the geometric features of the components of the target building; The spatial grid generation module is used to determine the side length of the bounding box of a component based on the geometric features of the component, and to divide the target building space into several grid units and mark the dense grid units. The feature extension module is used to extend the geometric features of components in dense mesh cells along the three-dimensional direction based on a preset offset, and generate extended feature vectors. The index mapping module is used to perform principal component rotation alignment and multiple orthogonal projection on the extended feature vector to generate a two-dimensional hash index. Based on the two-dimensional hash index, each component identifier is written into the corresponding index bucket and simultaneously mapped to the neighboring index buckets to form a correspondence between index buckets and component identifiers. The conflict candidate generation module is used to combine the component identifiers in each index bucket in pairs, analyze the overlapping interval in three-dimensional space based on the corresponding bounding box coordinates, identify the component pairs that overlap in all three directions as conflict component pairs, record their overlapping intervals, and form a set of conflict component pairs. The conflict marker module is used to mark the overlapping volumes of conflicting component pairs based on coefficient of variation analysis. Component pairs whose volumes deviate significantly from the overall level are marked as conflict marker objects, which are then used to drive component position adjustments in BIM.

[0088] The present invention also includes an electronic device: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to execute a BIM-based architectural design method.

[0089] The present invention also includes a computer-readable storage medium storing a computer program that, when executed by a processor, implements a BIM-based architectural engineering design method.

[0090] In the embodiments provided by this invention, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0091] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0092] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0093] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0094] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0095] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0096] In the embodiments provided in this disclosure, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0097] It should be noted that, in this disclosure, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element limited by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0098] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A BIM-based architectural engineering design method, characterized in that, Includes the following steps: Extract the geometric features of the target building's components; The side length of the bounding box of the component is determined based on the analysis of the component's geometric features, and the target building space is divided into several grid units, with dense grid units marked. The geometric features of components in a dense grid cell are extended along the three-dimensional direction based on a preset offset to generate an extended feature vector. Principal component rotation alignment and multiple orthogonal projection are performed on the extended feature vector to generate a two-dimensional hash index. Each component identifier is written into the corresponding index bucket according to the two-dimensional hash index and simultaneously mapped to the neighboring index buckets to form a correspondence between index buckets and component identifiers. The component identifiers in each index bucket are paired up, and the overlapping intervals in three-dimensional space are analyzed based on the corresponding bounding box coordinates. Component pairs that overlap in all three directions are identified as conflicting component pairs, and their overlapping intervals are recorded to form a set of conflicting component pairs. Based on coefficient of variation analysis, the overlapping volumes of conflicting component pairs are used to identify component pairs whose volumes deviate significantly from the overall level, forming conflict marker objects that are used to drive component position adjustments in BIM.

2. The BIM-based architectural engineering design method according to claim 1, characterized in that, The extraction of the geometric features of the target building components specifically includes: Acquire the three-dimensional geometric data of all components in the building information model, including component boundary surfaces, node coordinates, and spatial connection relationships; Iterate through the geometric data of each component, calculate the minimum and maximum coordinates of all boundary points along the three-dimensional direction, generate the minimum bounding box of the component, and record the correspondence between the bounding box coordinates and the component identifier. Calculate the center point coordinates of each bounding box and associate the center coordinates with the corresponding component identifier; Extract the length, width, and height dimensions of each bounding box to form a dimension vector, and associate it with the corresponding component identifier; Analyze the main axes of each component in three-dimensional space, analyze the principal direction vector through geometric points, and associate the principal direction vector with the component identifier; The geometric features of a component are formed by combining its identifier, bounding box coordinates, center coordinates, dimension vector, and principal direction vector.

3. The BIM-based architectural engineering design method according to claim 2, characterized in that, The process involves determining the bounding box side length of a component based on its geometric features, dividing the target building space into several grid units, and marking dense grid units. Specifically: Calculate the length, width, and height of the bounding box of all component geometric features, calculate the average value of each dimension, and then calculate the weighted variance to generate the spatial average side length. Based on the average side length of the space, the building space is divided into grid units of equal size along the three-dimensional direction, grid unit objects are generated and their spatial coordinate range is recorded; Traverse the geometric features of each component, determine the spatial coverage relationship between its bounding box and each mesh cell, and add the component identifier to the component list covering the mesh cells; The number of components in each grid cell is counted, and grid cells with a number exceeding a preset threshold are marked as dense grid cells, generating a list of dense grid cells.

4. The BIM-based architectural engineering design method according to claim 3, characterized in that, The process of extending the geometric features of components in a dense mesh cell along three dimensions based on a preset offset to generate an extended feature vector is as follows: Traverse each dense grid cell, obtain the component geometric features of all components within it, including center coordinates, bounding box size and principal direction vector, and associate the component identifier with the original feature vector; Along the length, width, and height of the component bounding box, set the offset step size by dividing the minimum design clearance by two, and generate positive and negative offset sequences respectively; Based on the offset step size, a perturbation combination is generated in three spatial directions. By combining the original center coordinates with the offsets in each direction, a new center position is formed, while keeping the original bounding box size and main direction vector unchanged, and a set of perturbation vectors is generated. The original feature vector is merged with the disturbance vector to form an extended feature vector. The extended feature vector records the component identification, the center coordinates after disturbance, the size vector, and the main direction vector. The operation of expanding the feature vector is repeated for each component within the dense grid cell, generating perturbation vector combinations sequentially along the length, width, and height directions; The extended feature vectors generated by all components within the dense grid cell are aggregated to form an extended feature vector set.

5. The BIM-based architectural engineering design method according to claim 4, characterized in that, The process involves performing principal component rotation alignment and multiple orthogonal projection on the extended feature vector to generate a two-dimensional hash index. Each component identifier is then written into its corresponding index bucket based on this hash index, and simultaneously mapped to neighboring index buckets, forming a correspondence between index buckets and component identifiers. Specifically: The center coordinates of all extended feature vectors in the statistical dense grid cell are used to identify the direction with the largest variance in the three spatial dimensions, which is then taken as the first direction. Adjust the coordinates of each extended feature vector according to the first direction, so that the extended feature vectors are aligned along the first direction, and rotate the deviation of the main direction of the component to be consistent with the first direction. Based on the rotated and aligned coordinates, the direction perpendicular to the first direction is selected as the second direction. Each extended feature vector is projected onto the first and second directions to generate a two-dimensional hash index. According to the preset two-dimensional grid size, the two-dimensional hash index is mapped to the corresponding index bucket, and the component identifier contained in each index bucket is recorded; Offset mapping is performed on the two-dimensional hash index along the horizontal and vertical directions, and the component identifier is written to the adjacent index bucket, thus forming a correspondence between the index bucket and the component identifier.

6. The BIM-based architectural engineering design method according to claim 5, characterized in that, The process involves pairwise combination of component identifiers within each index bucket, analyzing the overlapping regions in three-dimensional space based on the corresponding bounding box coordinates, identifying component pairs that overlap in all three directions as conflicting component pairs, and recording their overlapping regions to form a set of conflicting component pairs. Specifically: The component identifiers within the index bucket are paired to form component pairs, and each component pair is associated with the corresponding bounding box coordinates. Extract the minimum and maximum coordinates of the bounding boxes of the two components in each component to form a three-dimensional spatial interval; Determine whether the spatial intervals of the two components in each component pair intersect in all three directions (X, Y, Z). If they intersect in all three directions, mark the component pair as a conflicting component pair and record the minimum and maximum coordinates of the overlapping interval of the conflicting component pair. Traverse all component pairs in the index bucket and summarize the conflicting component pairs to form an initial set of conflicting component pairs; The initial set of conflicting component pairs in all index buckets is deduplicated to generate the final set of conflicting component pairs.

7. The BIM-based architectural engineering design method according to claim 6, characterized in that, The overlapping volume of conflicting component pairs is analyzed based on the coefficient of variation. Component pairs with significantly larger volume deviations from the overall level are marked as conflict marker objects, which are used for BIM-driven component position adjustments. Specifically: Extract the side lengths in the X, Y, and Z directions of the overlapping interval of each component pair in the conflict component pair set to form the overlapping volume; Calculate the overlap volume of all conflicting component pairs, and analyze the mean and standard deviation of the overlap volume; The coefficient of variation of the overlapping volume is generated based on the mean and standard deviation, and the overlapping volume of each component pair is compared with the coefficient of variation. When the overlap volume of a component pair deviates from the overall average level by more than a preset threshold, the component pair is marked as a conflict object, and the corresponding component identifier and overlap interval coordinates are recorded. Traverse the set of conflicting component pairs, mark all components that deviate from the threshold, and generate the final conflict mark object, which is used to drive component position adjustment in BIM.

8. A system using the BIM-based architectural engineering design method as described in any one of claims 1-7, characterized in that, It includes a component feature extraction module, a spatial mesh generation module, a feature expansion module, an index mapping module, a conflict candidate generation module, and a conflict marking module. The component feature extraction module is used to extract the geometric features of the components of the target building; The spatial grid generation module is used to determine the side length of the bounding box of a component based on the geometric features of the component, and to divide the target building space into several grid units and mark the dense grid units. The feature extension module is used to extend the geometric features of components in dense mesh cells along the three-dimensional direction based on a preset offset, and generate extended feature vectors. The index mapping module is used to perform principal component rotation alignment and multiple orthogonal projection on the extended feature vector to generate a two-dimensional hash index. Based on the two-dimensional hash index, each component identifier is written into the corresponding index bucket and simultaneously mapped to the neighboring index buckets to form a correspondence between index buckets and component identifiers. The conflict candidate generation module is used to combine the component identifiers in each index bucket in pairs, analyze the overlapping interval in three-dimensional space based on the corresponding bounding box coordinates, identify the component pairs that overlap in all three directions as conflict component pairs, record their overlapping intervals, and form a set of conflict component pairs. The conflict marking module is used to mark the overlapping volumes of conflicting component pairs based on coefficient of variation analysis. Component pairs whose volumes deviate significantly from the overall level are marked as conflict marking objects, which are then used to drive component position adjustments in BIM.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the BIM-based architectural engineering design method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the BIM-based architectural engineering design method as described in any one of claims 1 to 7.

Images