Building decoration suspended ceiling design method and system based on BIM technology

Abstract

The invention belongs to the technical field of building information models (BIM) and digital construction, and discloses a building decoration suspended ceiling design method and system based on a BIM technology. The method comprises the following steps: acquiring main body structure data through laser scanning, reconstructing a three-dimensional model, and extracting connecting piece coordinates to generate a standardized data document; total station calibration and welding positioning are conducted on the basis of the coordinate data, and keel installation leveling and aluminum plate preprocessing are completed; and precise embedding of the aluminum plates is achieved by planning the installation sequence, and finally quality acceptance is conducted and fed back to the BIM model to form closed-loop optimization. According to the method, dynamic modeling driven by curvature characteristics and a differential geometric optimization algorithm are adopted, the problem of accumulated positioning errors of the suspended ceiling connecting piece under a complex building structure is solved, millimeter-level construction precision is achieved, construction efficiency is improved, material waste is reduced, and the method is particularly suitable for decorative suspended ceiling engineering of a large-span cantilever structure.

Description

Technical Field

[0001] The present invention belongs to the intersection of Building Information Modeling (BIM) technology and digital construction, and in particular relates to a method and system for designing architectural decorative ceilings based on BIM technology. Background Art

[0002] In traditional building decorative ceiling construction, the positioning of connectors mainly relies on manual measurement and two-dimensional drawing layout, which has three technical defects: 1. Poor adaptability of the main structure: When a building has complex geometric features such as curved surfaces and cantilevers, traditional total station layout cannot accurately capture changes in structural curvature (the measured average deviation is ±5mm), causing the welding position of the connector to deviate from the designed point; 2. Data link break: The BIM model and on-site construction use independent coordinate systems. The coordinate data exported from the model needs to be manually converted multiple times (at least 3 coordinate system transformations), and the cumulative error is magnified to more than ±8mm; 3. Uncontrolled dynamic deformation: The temperature deformation and load deformation (typical value 2-3mm / m) of the main structure during the construction phase were not fed back to the design model in real time, resulting in misalignment of the joints of the ceiling aluminum panels after installation.

[0003] Existing solutions use static BIM models to export coordinates, do not address the coupling effects of point cloud noise and structural curvature, and use robotic arms for automatic welding, but rely on preset trajectories and lack real-time deformation compensation. Summary of the Invention

[0004] The present invention provides a method and system for designing architectural decorative ceilings based on BIM technology, so as to solve the problem of how to realize high-precision connector coordinate extraction and millimeter-level construction positioning driven by the curvature characteristics of the building main structure based on a BIM model that integrates dynamic point cloud reconstruction and differential geometry optimization, and solve the problem of cumulative positioning errors of ceiling connectors under complex main structures.

[0005] In order to solve the above technical problems, the present invention provides a building decoration ceiling design method based on BIM technology, comprising: Obtain main structure data, reconstruct 3D model and extract connector coordinates to generate standardized connector coordinate data documents; The process of generating a standardized connector coordinate data document includes constructing a standardized data storage format; Obtain the coordinate data file of the connector, perform coordinate calibration and point matching, perform point-by-point positioning welding, ultrasonic flaw detection and position review, and generate a welding quality compliance list; Install the keel according to the position of the connectors, perform level calibration and bolt reinforcement, and generate a keel leveling completion report; Match the specifications and colors of the aluminum plates, perform surface dust removal and edge chamfering, store them by keel partitions, and generate an aluminum plate placement status table; Plan the installation sequence and insert the keel slots, perform joint alignment, fine-tuning and locking, and generate an aluminum panel installation completion report; Inspect the flatness and joint tightness of the aluminum plates, make local adjustments to the aluminum plate joints and fixings, generate a final acceptance report and feed it back to the 3D model database for closed-loop optimization.

[0006] Furthermore, the acquisition of main structure data, reconstruction of the three-dimensional model and extraction of connector coordinates include: Obtain laser scanning point cloud data and design drawings of the main structure, perform noise removal and precision reconstruction, and obtain 3D structure point cloud data; Extract the cantilever area and connector installation position information from the 3D structure point cloud data, perform mesh fitting processing, and obtain 3D model data; Extract connector coordinates and convert formats of 3D model data to generate standardized connector coordinate data documents.

[0007] Furthermore, the acquisition of the connector coordinate data file and the coordinate calibration and point matching include: Obtain the coordinate data file of the connector, combine it with the measurement parameters of the total station, perform coordinate calibration and point matching, and obtain the calibrated welding coordinate sequence; Based on the calibrated welding coordinate sequence, control the welding equipment to perform point-by-point positioning welding and generate a welding completion report; Perform ultrasonic flaw detection and position review on the welds in the welding completion report to generate a welding quality compliance list.

[0008] Furthermore, the keel is installed according to the position of the connector, and level meter calibration and bolt reinforcement are performed, including: Obtain welding quality compliance lists and keel design drawings, perform material sorting and transportation route planning, and obtain keel installation preparation data; Based on the keel installation preparation data, install the main keel and secondary keel in sequence according to the position of the connectors, and generate a preliminary keel installation report; Perform level meter calibration and bolt reinforcement on the keel in the preliminary keel installation report, and generate a keel leveling completion report.

[0009] Furthermore, the matching of aluminum plate specifications and color screening, and surface dust removal and edge chamfering processing include: Obtain the keel leveling completion report and aluminum plate design parameters, perform specification matching and color screening, and obtain an aluminum plate selection list; Perform surface dust removal and edge chamfering on the aluminum plates in the aluminum plate selection list to generate a pre-treated aluminum plate set; The pre-processed aluminum plate collection is classified and stored according to the keel partition number, and an aluminum plate placement status table is generated.

[0010] Furthermore, the planning of the installation sequence and the embedding of the keel slots, and the execution of seam alignment, fine-tuning, and locking processes include: Obtain the aluminum plate in-place status table and installation sequence diagram, perform aluminum plate grabbing and transportation path planning, and obtain the aluminum plate installation queue; Insert the aluminum panels into the keel slots according to the aluminum panel installation queue, perform joint alignment and temporary fixation, and generate a preliminary aluminum panel installation report; Fine-tune and lock the aluminum plates in the preliminary installation report and generate a completion report for the aluminum plate installation.

[0011] Furthermore, the above-mentioned detection of the flatness and joint tightness of the aluminum plate and local adjustment of the aluminum plate joints and fixings include: Obtain the aluminum panel installation completion report and design drawings, conduct joint tightness and flatness inspections, and generate an inspection problem list; Make local adjustments to the aluminum plate joints and fixings based on the inspection problem list and generate a rectification record sheet; Compare the rectification record sheet with the acceptance criteria, generate a final acceptance report and feed it back to the 3D model database for closed-loop optimization.

[0012] Furthermore, the extracting of the overhanging area and the connection piece installation position information from the three-dimensional structure point cloud data includes: Segment the overhang area based on the point cloud density threshold and extract the 3D coordinates of the connector installation location; The B-spline curve is used to fit the installation boundary of the connector to generate the grid topology.

[0013] Furthermore, the controlling the welding equipment to perform point-by-point positioning welding comprises: Generate robot arm motion trajectory according to welding coordinate sequence; Monitor welding current and temperature in real time and adjust welding parameters dynamically.

[0014] A building decoration ceiling design system based on BIM technology, used to implement any of the above methods, comprising: 3D modeling module, used to obtain main structure data, reconstruct 3D model and extract connector coordinates; Welding control module, used to obtain the coordinate data file of the connection parts, perform coordinate calibration and point matching; Keel installation module, used to install the keel according to the position of the connector and perform leveling and reinforcement; Aluminum plate processing module, used to match aluminum plate specifications and perform pre-processing and classification; Aluminum plate installation module, for planning installation sequence and completing embedded locking; Acceptance feedback module, for detecting aluminum plate installation quality and closing loop feedback data.

[0015] The key innovation points and beneficial effects of the present application include: (1) Curvature feature driven dynamic modeling (corresponding to S110 step) Through the improved joint registration equation ( Achieve millimeter-level alignment of point cloud and BIM drawing (error ≤1mm); Adaptive filtering algorithm based on Riemann curvature Eliminate 95% of scanning noise and improve reconstruction accuracy to 0.05mm (formula ②); (2) Cantilever boundary differential geometry extraction (corresponding to S120 step); Use Gaussian curvature energy field Identify cantilever area (formula ③); Extract installation boundary through Hamilton-Jacobi equation boundary evolution ( ) with 99.2% accuracy (formula ④); Normal-curvature joint optimization ( ) makes the installation surface fit degree reach 97° (formula ⑥); (3) Manifold optimization coordinate conversion (corresponding to S130 step); Gauss manifold projection ( ) extracts connector coordinates (formula ⑤); Lie group sparse optimization ( ) realizes construction coordinate conversion error ≤0.8mm (formula ⑧); (4) Convert acceptance data into differential geometry-group theory integrated mathematical entity (D→F→ ), realize millisecond-level closed loop of "detection→decision→feedback" (response delay <50ms), and lay the foundation of core algorithm of digital twin construction. (S630). BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A building decoration ceiling design method based on BIM technology provided by the embodiment of the present application; Figure 2 A building decoration ceiling design system based on BIM technology provided by the embodiment of the present application. DETAILED DESCRIPTION

[0017] Embodiment one: refer to Figure 1, is a process schematic diagram of a building decoration suspended ceiling design method based on BIM technology provided by the embodiment of the application, which can at least include steps S100-S600: S100, acquire main body structure data, reconstruct a three-dimensional model and extract connector coordinates, and generate a standardized connector coordinate data document.

[0018] S200, acquire the connector coordinate data document, perform coordinate calibration and point matching, execute point-by-point positioning welding, ultrasonic flaw detection and position review, and generate a welding quality compliance list.

[0019] S300, install the keel according to the connector position, perform level verification and bolt reinforcement, and generate a keel leveling completion report.

[0020] S400, match the aluminum plate specifications and color selection, and perform surface dust removal and edge chamfering treatment, store according to the keel partition classification, and generate an aluminum plate in-place state table.

[0021] S500, plan the installation sequence and embed the keel slot, perform joint alignment, fine adjustment and locking treatment, and generate an aluminum plate installation completion report.

[0022] S600, detect the flatness and tightness of the aluminum plate, locally adjust the aluminum plate joints and fixing parts, generate a final acceptance report and feed back to the three-dimensional model database for closed-loop optimization.

[0023] Step S100 at least includes steps S110-S130: S110, acquire laser scanning point cloud data and design drawings of the main body structure, perform noise removal and precision reconstruction, and obtain three-dimensional structure point cloud data.

[0024] Specifically, the laser scanning point cloud data of the main body structure is obtained by omnidirectional scanning of the main body structure of the construction site by a three-dimensional laser scanner. The three-dimensional laser scanner adopts pulse ranging principle, the scanning angle resolution is 0.001°, and the ranging accuracy reaches ±1mm. During the scanning process, the scanner sets up scanning stations at a preset station interval (usually 5-10 meters), and the point cloud data obtained at each station is automatically spliced by a target ball to form a complete building main body structure point cloud data set.

[0025] Further, the design drawings include building construction drawings and structure construction drawings, which are stored in DWG format. The design drawings are imported by BIM software, and the coordinate system is aligned with the laser scanning point cloud data. The coordinate system alignment adopts least squares method for registration, and selects not less than 4 feature points (such as column corners and beam ends) as registration reference to ensure that the spatial position error of the point cloud data and the design drawings is less than ±2mm.

[0026] Understandably, the noise removal and precision reconstruction process includes the following steps: first, a statistical outlier algorithm is used to remove noise from the point cloud, with a neighborhood radius of 50 mm and a standard deviation threshold of 1.5. Then, the point cloud is downsampled using a voxel grid filter, with a voxel size of 10 mm × 10 mm × 10 mm. Finally, a Poisson surface reconstruction algorithm is used to reconstruct the downsampled point cloud, with a reconstruction depth of 9 levels, to generate three-dimensional structured point cloud data with a topological structure. The three-dimensional structured point cloud data is stored in PLY format and contains vertex coordinates, normal vectors, and color information.

[0027] In another embodiment, the laser scanning point cloud data of the main structure is obtained by spatial data acquisition using a 3D laser scanner. The scanner is arranged at the construction site according to topological optimization to obtain the original point cloud data set. ,in Represents the Cartesian coordinates of the i-th scanning point , is the total amount of point cloud (usually Design drawings to parameterize surface models Indicates that The coordinates of the control points , is the B-spline basis function, are surface parameters.

[0028] Furthermore, an improved joint registration equation is designed, specifically the alignment of the point cloud and the design drawing is achieved by minimizing the energy functional: , in: To minimize the energy functional; is the rigid transformation matrix (including rotation and pan ); and They are the characteristic points of the point cloud and the drawing (such as column corners and beam ends); is the feature weight factor (taken as 0.35); is the number of feature points (≥4); This equation is solved by the Levenberg-Marquardt algorithm to ensure the registration error .

[0029] Furthermore, an improved topological noise filter is designed, specifically an adaptive filter based on Riemann curvature: , in: is adaptive filtering; for Neighborhood point set of (radius 50 mm); is the curvature tensor (calculated by principal component analysis); is a dynamic threshold (adaptive according to the standard deviation of 1.5); represents the Frobenius norm; Further construct the filtering decision function: , in, is the filtering decision function; dynamic threshold ,in is the mean, The algorithm eliminates outlier noise while preserving structural features, and outputs three-dimensional structural point cloud data after Poisson surface reconstruction. , the vertex normal vector accuracy is .

[0030] Technical effect: (1) Millimeter-level model reconstruction accuracy (corresponding to S110) By improving the joint registration equation and Riemann curvature filtering : Improved accuracy: point cloud registration error ≤ 1mm (traditional method ±5mm); Noise suppression: eliminates more than 95% of scanning noise and reconstructs the model surface with an accuracy of 0.05mm; Implementation verification: Actual measurements on a high-speed railway station project showed that the coordinate deviation of the cantilever beam endpoints was reduced from 4.3mm to 0.7mm; (2) Accurate extraction of complex structural features (corresponding to S120) Based on Gaussian curvature energy field and normal-curvature optimization ( ): Boundary recognition: The accuracy of overhang area extraction is 99.2% (traditional manual annotation is only 85%). Installation surface optimization: the angle between the normal direction of the connector and the structural surface is ≤3° (traditional method is 8°-15°); Project benefits: The number of custom connectors for the curved curtain wall project was reduced by 37%; (3) Lossless conversion of construction coordinates (corresponding to S130) Manifold projection of Lie group sparse optimization With Lie group sparse optimization (LSO) Conversion error: model coordinate → construction coordinate mapping error ≤ 0.8mm; Data integrity: SHA-256 encryption checksum to prevent data tampering (tamper detection rate 100%); Construction efficiency: total station data loading time from 45 minutes to 90 seconds.

[0031] S120, extract the overhanging area and the connecting piece installation position information from the three-dimensional structure point cloud data, perform grid fitting processing, and obtain three-dimensional model data.

[0032] Specifically, the extraction of the overhanging area is realized by a region growing algorithm. First, set a seed point (usually select the end point of the cantilever beam) in the three-dimensional structure point cloud data, set the curvature threshold to 0.05, and the normal vector angle threshold to 15°, and perform region growing to obtain the point cloud subset of the overhanging area. Then, the Alpha Shapes algorithm is used to extract the boundary of the point cloud subset of the overhanging area, and the Alpha value is set to 200mm to generate the boundary polygon of the overhanging area.

[0033] Further, the extraction of the connecting piece installation position information is realized by the following steps: according to the connecting piece arrangement interval (usually 800-1200mm) marked in the design drawing, set the installation point on the boundary polygon of the overhanging area at equal intervals; then based on the surface curvature of the three-dimensional structure point cloud data, fine-tune each installation point to ensure that the angle between the connecting piece installation surface and the main structure surface normal vector is not more than 5°.

[0034] Understandably, the grid fitting processing adopts the Delaunay triangulation algorithm. First, the boundary polygon of the overhanging area and the connecting piece installation point are taken as the constraint condition to generate a two-dimensional plane triangular grid; then the grid is optimized by the Laplace smoothing algorithm, the iteration number is set to 20 times, and the smoothing factor is set to 0.5; finally, the two-dimensional grid is mapped to three-dimensional space to generate three-dimensional model data containing the geometric characteristics of the overhanging area and the connecting piece installation position. The three-dimensional model data is stored in IFC format, containing geometric entities, attribute sets and relationship definitions.

[0035] In another embodiment, the overhanging area extraction based on curvature manifold segmentation defines a Gaussian curvature energy field: , wherein, is the Gaussian curvature energy field; is the principal curvature.

[0036] ​This formula calculates the Gaussian curvature (intrinsic curvature measure) at point p and is used to identify characteristic areas of cantilevered structures. >0 indicates a convex surface. <0 indicates a saddle surface.

[0037] Furthermore, the boundary of the cantilevered area is solved by the Hamilton-Jacobi equation: , in, is the level set function; is the pseudo time parameter; is the evolution velocity field, which is used to construct the driving force of boundary evolution; ; is the weight coefficient; is the curvature energy field gradient; is the function gradient; is the Euclidean norm.

[0038] This formula describes the dynamic evolution of the level set surface, making the zero-value interface converge to the real boundary of the cantilever structure.

[0039] Seed Point Set in At the maximum value (the end point of the cantilever beam), the gradient ascent method is used for optimization: , in, is the seed point coordinate; is the number of iterations; is the curvature energy field gradient; is the step length.

[0040] This formula uses the gradient ascent method to find the local maximum point in the curvature energy field and accurately locate the characteristic points of the cantilever structure (such as the sharp ends of the beams).

[0041] Furthermore, the differential geometry optimization of the connector installation points is performed, and the initial values ​​of the installation points are generated at equal intervals according to the design parameters: , in, is the coordinate of the initial installation point; is the arc length parameter; For installation spacing; Index of the mount point; is the parametric curve of the cantilever boundary.

[0042] This formula complies with building code requirements by generating initial installation points at equal intervals along the parameterized boundary curve (800–1200 mm spacing).

[0043] Optimize with curvature constraints: , Where: q is the optimization variable; Normal vector for the connector mounting surface; is the surface normal vector of the main structure; is the curvature tensor at the installation point; is the Frobenius norm; is the regularization coefficient; This formula can optimize the installation position so that: the normal of the connector is parallel to the normal of the structure (cross product → 0); the curvature change at the installation point is minimized (improving stability).

[0044] This optimization ensures that the normal angle of the mounting surface .

[0045] Furthermore, the grid fitting adopts the Delaunay decomposition on the Riemannian manifold: , in, is the final mesh; is a triangular face; is the number of patches; is the mesh vertex; is the Voronoi cell center; is the geodesic distance (calculated by the Fast Marching Method).

[0046] This formula constructs a triangular mesh on a Riemannian manifold that satisfies the Delaunay condition, ensuring that: vertex To the heart of one's own cell Shortest distance; The triangulation result meets the maximum and minimum angle criteria; Adapt to the changing characteristics of surface curvature.

[0047] S130 , extracting connector coordinates and converting the format of the three-dimensional model data to generate a standardized connector coordinate data document.

[0048] Specifically, connector coordinate extraction is achieved by traversing the connector installation entities in the 3D model data. For each connector installation entity, the center point coordinates (X, Y, Z) and normal vector (NX, NY, NZ) are extracted, and the connector type (e.g., L-type, T-type) and specifications (e.g., length, aperture) are recorded. The coordinate extraction accuracy reaches ±0.1mm, and the normal vector accuracy reaches ±0.1°.

[0049] Furthermore, the format conversion includes coordinate system conversion and data standardization. The coordinate system conversion converts the model coordinate system to the construction coordinate system. The conversion parameters include the translation vector (ΔX, ΔY, ΔZ) and the rotation matrix (3×3). The converted coordinate accuracy is maintained at ±1mm. The data standardization process structures the connector information according to a preset template. This includes: storing coordinate data using IEEE754 double-precision floating-point numbers; encoding attribute data in JSON format; and recording spatial relationships using a topological connection table.

[0050] Understandably, the standardized connector coordinate data file consists of the following: a file header records coordinate system information, conversion parameters, and generation time; a data volume stores the coordinates, normal vectors, and attribute data of each connector in partitions; and a checksum is generated using the CRC32 algorithm. The file is stored in binary format with a .ccd file extension and can be directly imported into a total station for construction stakeout. The file is digitally signed using the SHA-256 algorithm to ensure data integrity and traceability.

[0051] In another embodiment, coordinates are extracted via Grassmann manifold projection: , , in, is the coordinate of the center point of the connector; is the vertex set of the connector entity; is the coordinate of a single vertex; Projection depth provided for BIM models; is the unit normal vector of the connector installation surface; argmin is the solution operator of the minimization problem; is the normal vector to be optimized; is a Grassmann manifold; is the Euclidean norm; Transpose of the vector; is the vertex mean; Coordinate system transformation using screw theory: , in, is the rotation matrix; is the translation vector; Transpose the zero vector; is the matrix exponential function; is the Lie algebra parameter; is the generator of spiral motion; Parameter optimization: , in, To minimize the parameters; is a Lie algebra parameter vector; is a homogeneous transformation matrix; is a model control point coordinate; is a construction control point coordinate; is an L1 norm; is a regularization coefficient; Further, the data standardization storage format is: , wherein, is a standard data tensor; is a central coordinate; is a unit normal vector; is a type code; is a parameter vector; is the total number of connectors.

[0052] The check code is generated by an elliptic curve cryptography system to ensure data integrity.

[0053] The step S200 at least includes steps S210-S230: S210, obtain a connector coordinate data document, combine total station measurement parameters, perform coordinate calibration and point matching, and obtain a calibrated welding coordinate sequence.

[0054] Specifically, the connector coordinate data document is obtained by reading the standardized connector coordinate data document generated in S130. The connector coordinate data document is stored in binary format and includes a file header, a data body, and a check code. The file header records coordinate system information, conversion parameters, and generation time; the data body stores the coordinates, normal vectors, and attribute data of each connector according to partitions; and the check code is generated using the CRC32 algorithm. When reading, first verify the digital signature, then parse the file header to obtain the coordinate system conversion parameters, and finally load the connector coordinate information in the data body.

[0055] Further, the total station measurement parameters include instrument model, ranging accuracy, angle measurement accuracy, and prism constant. The total station uses a model with automatic target recognition function, the ranging accuracy is not less than ±1mm, and the angle measurement accuracy is not less than ±1". The coordinate calibration is achieved by establishing a construction control network, at least 4 control points are arranged on the construction site with a distance of not more than 50 meters, and the closed traverse measurement method is used for adjustment calculation, and the control point coordinate accuracy reaches ±2mm.

[0056] Understandably, point matching is accomplished through the following steps: first, converting the coordinates in the connector coordinate data file to a construction coordinate system; then, establishing a coordinate system consistent with the BIM model within the total station; and finally, performing point matching using the least squares method to calculate the actual construction coordinates of each connector. The calibrated welding coordinate sequence, containing the connector number, 3D coordinates, normal vector, and welding parameters, is stored in XML format and can be directly imported into a welding control system.

[0057] S220. Based on the calibrated welding coordinate sequence, control the welding equipment to perform point-by-point positioning welding and generate a welding completion report.

[0058] Specifically, the control of the welding equipment to perform point-by-point positioning welding is achieved through an industrial control computer. The industrial control computer reads the calibrated welding coordinate sequence, parses the coordinates and welding parameters of each connection, and then generates the motion trajectory of the welding robot arm. This motion trajectory is generated using a quintic polynomial interpolation algorithm, ensuring smooth robot arm movement and a trajectory accuracy of ±0.1mm.

[0059] Furthermore, the welding equipment includes a welding robot arm, a welding machine, and a shielding gas supply system. The welding robot arm is a six-axis linkage type with a repeatable positioning accuracy of ±0.05mm. The welding machine is a pulsed MIG welder with a welding current control accuracy of ±1A. The shielding gas is a mixture of argon and carbon dioxide in a ratio of 80%:20%. During welding, the welding robot arm precisely locates the connection part according to its motion trajectory, and the welding machine automatically completes the welding operation according to preset parameters.

[0060] Understandably, the generated welding completion report includes recording the actual welding coordinates of each joint, welding time, welding current and voltage parameters, and operator information. The welding completion report is stored in a database format and includes a preliminary assessment of welding quality. After each weld is completed, the system automatically takes a photo of the weld and compares it to the theoretical position in the BIM model. Welds with deviations exceeding ±1mm are marked as pending review points.

[0061] S230. Perform ultrasonic testing and position review on the welds in the welding completion report and generate a welding quality compliance list.

[0062] Specifically, ultrasonic flaw detection is performed using a digital ultrasonic flaw detector. The probe frequency is 5 MHz, and the detection sensitivity is 1 mm flat-bottom holes. During testing, the probe moves along the weld seam in 5 mm increments, fully scanning each weld. The ultrasonic flaw detector automatically records the location, amplitude, and depth of the defect echo and performs defect rating according to the GB / T 11345 standard.

[0063] Furthermore, position verification is performed using a total station. The points marked for verification in the welding completion report are precisely measured using a total station, achieving an accuracy of ±0.5mm. The measured coordinates are compared with the theoretical coordinates, and the positional deviation is calculated. Welds with a deviation within ±2mm are considered acceptable, while those exceeding ±2mm are marked as unacceptable.

[0064] Understandably, the welding quality compliance checklist is generated by comprehensively evaluating the ultrasonic testing results and position verification results. This checklist includes the serial number of each connection, the welding quality grade (categorized as excellent, good, acceptable, and unacceptable), the defect type (such as porosity, slag inclusion, and lack of fusion), and the position deviation value. This checklist is stored in a structured table format and can be directly imported into S310 for keel installation preparation. For unacceptable welds, the checklist includes corrective action requirements and re-inspection timelines.

[0065] Step S300 at least includes steps S310-S330: S310. Obtain the welding quality compliance list and keel design drawings, perform material sorting and transportation route planning, and obtain keel installation preparation data.

[0066] Specifically, the acquisition of the welding quality compliance list is achieved by reading the welding quality assessment file generated by S230. The welding quality compliance list is stored in a structured database format, and contains key parameters such as the number of each connector, welding quality grade, defect type and position deviation value. When reading, the data integrity check code is first verified, and then the file content is parsed, and the qualified welds are clustered and grouped according to the spatial position, and each group corresponds to a keel installation unit. The keel design drawing is a CAD file in DWG format, which contains information such as the specifications and dimensions of the main keel and secondary keel, the layout spacing and the connection node details. After the drawing is read, it is spatially matched with the welding quality compliance list to ensure that the design parameters are consistent with the actual weld point position.

[0067] Furthermore, the material sorting is completed through an intelligent warehouse management system. According to the list of materials in the keel design drawings, the system automatically generates a sorting task order, which includes the model, length, quantity and material requirements of the required keels. The intelligent warehouse management system uses RFID technology to identify material information, and the robotic arm grabs the keels of specified specifications according to the optimal path and places them on a dedicated transport rack. During the sorting process, the system scans the keel barcode in real time, compares and verifies it with the design parameters, and automatically removes keels with an error of more than ±1mm. The transport route planning is a path optimization model established based on the three-dimensional point cloud data of the construction site. It takes into account factors such as the working radius of the lifting equipment, the size of the transport channel, and the accessibility of the installation area, and uses a genetic algorithm to calculate the optimal transport route to ensure the safe lifting of heavy keels.

[0068] Understandably, the keel installation preparation data includes the following: an installation area division diagram, a materials delivery list, a hoisting schedule, and quality control points. The installation area division diagram divides the construction area into several installation units, each of which is annotated with a primary and secondary keel layout diagram and weld point location references; the materials delivery list details the keel model, quantity, and delivery time required for each installation unit; the hoisting schedule specifies the installation sequence and machinery deployment plan for each area; and the quality control points clearly define the installation tolerances and acceptance criteria. The keel installation preparation data is stored in a format recognizable by the BIM collaboration platform and synchronized to each construction terminal device via the cloud.

[0069] S320: Based on the keel installation preparation data, install the main keel and the secondary keel in sequence according to the positions of the connectors, and generate a preliminary keel installation report.

[0070] Specifically, the installation of the main keel is completed through the collaborative work of a tower crane and an installation robot. Before installation, the construction workers use AR glasses to view the three-dimensional positioning information in the keel installation preparation data and spray temporary marks at the connector positions. After the tower crane lifts the main keel to the specified height, the installation robot captures the connector mark through the visual recognition system and guides the keel end to precisely dock with the connector. During the docking process, the force feedback system monitors the contact pressure in real time. When the pressure value reaches the set threshold, it automatically triggers the fastening device to complete the temporary fixation. The allowable deviation of the main keel installation is controlled within the range of ±3mm in the length direction, ±2mm in the elevation direction, and ±2mm in the axis deviation.

[0071] Furthermore, the secondary keel installation is carried out after the main keel installation is accepted. The construction personnel retrieve the secondary keel layout diagram in the keel installation preparation data through the mobile terminal, and use a laser line projector to mark the secondary keel installation position on the installed main keel. The secondary keel is connected to the main keel through a special clip. During installation, a torque wrench is used to control the tightening force of the clip bolts to ensure a reliable connection. After each installation unit of the secondary keel is installed, preliminary leveling is performed immediately, and an aluminum alloy ruler is used to check the flatness of the keel surface. If the local deviation exceeds ±1.5mm / m, it is adjusted immediately. The entire installation process is recorded, and photos of key nodes are automatically uploaded to the quality management system.

[0072] Understandably, the preliminary keel installation report includes the following: an installation progress statistics table, quality inspection records, and an image database. The installation progress statistics table records the completion status of primary and secondary keels in each area in real time, comparing and analyzing them against the planned progress. The quality inspection records detail the measured deviation data for each installation unit, including parameters such as keel spacing, surface flatness, and joint width. The image database stores videos and photos of the entire installation process, categorized and indexed by timestamp and spatial location. The preliminary keel installation report is automatically generated through the BIM platform, supporting multi-dimensional data query and visualization.

[0073] S330. Perform level calibration and bolt reinforcement on the keel in the preliminary keel installation report, and generate a keel leveling completion report.

[0074] Specifically, the level calibration is carried out using an electronic level in conjunction with a total station. First, an electronic level with an accuracy of 0.02mm / m is used to conduct a general survey of the overall flatness of the keel, and then a total station is used to conduct a re-measurement at key locations. The total station is set up on the control point to measure the three-dimensional coordinates of the keel interference points, and the deviation values ​​are calculated by comparing them with the design coordinates. For key nodes such as cantilevered parts and corners, the distance between measurement points is no more than 1 meter; for conventional areas, the distance between measurement points is controlled within 2 meters. All measurement data is uploaded to the quality management platform in real time, and a deviation curve graph is automatically generated.

[0075] Furthermore, the bolt reinforcement is completed by an intelligent torque wrench system. According to the node requirements in the keel design drawings, the system automatically sets the torque value of each connection point. The intelligent torque wrench is connected to the management system via Bluetooth. After the construction personnel scan the QR code of the connection point, the wrench automatically loads the corresponding torque parameters. During the tightening process, the wrench records the torque-angle curve in real time, and automatically alarms when the set torque is reached. All tightening data is automatically uploaded to form a traceable quality record. For parts that require secondary tightening, the system sets an electronic fence, which automatically reminds the construction personnel to re-tighten after the specified time.

[0076] Understandably, the keel leveling completion report includes the following core contents: a leveling measurement data sheet, a bolt tightening record, and a final acceptance conclusion. The leveling measurement data sheet records the design elevation, measured elevation, and deviation values ​​of all measurement points, and calculates the overall flatness index. The bolt tightening record includes information such as the design torque, actual torque, tightening time, and operator for each connection point. The final acceptance conclusion integrates the leveling data and tightening quality to determine whether the keel system meets installation requirements. Once the keel leveling completion report is digitally signed and confirmed, it serves as a prerequisite for the installation of the S410 aluminum plate, and its data is directly linked to the quality control system of subsequent construction stages.

[0077] Step S400 at least includes steps S410-S430: S410. Obtain the keel leveling completion report and aluminum plate design parameters, perform specification matching and color screening, and obtain an aluminum plate selection list.

[0078] Specifically, the acquisition of the keel leveling completion report is achieved by reading the acceptance document generated by S330. The keel leveling completion report is stored in a structured database format, and contains key information such as leveling measurement data tables, bolt tightening records and final acceptance conclusions. When reading, the digital signature is first verified, and then the file content is parsed to extract the final elevation data, flatness index and allowable deviation range of each partition keel. The aluminum plate design parameters are a list of materials derived from the BIM model, including technical parameters such as aluminum plate model, specification size, surface treatment method and color number. The aluminum plate design parameters are spatially matched with the keel leveling completion report to ensure that the aluminum plate blocks correspond one-to-one with the keel partitions.

[0079] Furthermore, the specification matching is completed by an intelligent matching algorithm. According to the actual size of the keel partition and the standard specifications in the aluminum plate design parameters, the system automatically calculates the optimal aluminum plate arrangement plan. For non-standard areas, the system generates processing drawings of customized aluminum plates, marking the precise cutting size and opening position. The color screening uses a spectrophotometer to detect the aluminum plate sample and measure its color coordinate value L 、a 、b The color difference from the design requirements is controlled within the range of ΔE≤1.5. All selected aluminum plates are recorded with their batch number, color number and production date to ensure the color consistency of aluminum plates in the same area.

[0080] Understandably, the aluminum plate selection list includes the following: a zoning installation diagram, a standard plate usage table, a non-standard plate processing diagram, and material acceptance standards. The zoning installation diagram indicates the layout and joint locations of the aluminum plates in each area; the standard plate usage table details the number of aluminum plates used and the installation location of each type; the non-standard plate processing diagram provides detailed dimensions and process requirements for customized aluminum plates; and the material acceptance standards specify the allowable dimensional deviations and appearance quality requirements for aluminum plates. The aluminum plate selection list is stored in a format recognizable by the BIM collaboration platform and synchronized to the material procurement and processing departments via the cloud.

[0081] S420. Perform surface dust removal and edge chamfering treatment on the aluminum plates in the aluminum plate selection list to generate a pre-treated aluminum plate set.

[0082] Specifically, the surface dust removal is completed by an automatic cleaning production line. After entering the cleaning workshop, the aluminum plate is first cleaned by an ion wind curtain to remove surface dust, and then cleaned by a multi-stage roller brush cleaning system using ultra-fine fiber brush heads in combination with a special cleaning agent for deep cleaning. The cleaned aluminum plate is accepted by a laser dust detector, and the surface residual particulate matter is not greater than 0.3 pm in size and not more than 5 pieces per square centimeter. The cleaning process is carried out in a constant temperature and humidity environment to prevent oxidation of the aluminum plate surface.

[0083] Further, the edge chamfering process is realized by a numerical control machining center. According to the machining requirements in the aluminum plate selection list, the system automatically generates a machining program to control the milling cutter to accurately chamfer along the edge of the aluminum plate. The chamfer angle of the standard plate is 45°, and the width is 2 mm; the non-standard plate uses variable angle chamfering according to actual needs, and the chamfering precision is controlled within ±0.1 mm. After machining, a three-dimensional scanner is used to detect the chamfer quality to ensure that the chamfer size and angle meet the design requirements. All machining parameters and detection data are automatically recorded to form a traceable quality file.

[0084] Understandably, the pre-processed aluminum plate set includes the following: cleaning quality report, chamfering machining record, and pre-processing completion identification. The cleaning quality report records the cleaning parameters and detection results of each aluminum plate; the chamfering machining record includes machining time, equipment parameters, and quality inspection data; the pre-processing completion identification is a two-dimensional code label pasted on the back of the aluminum plate, and the entire pre-processing information of the aluminum plate can be obtained after scanning. The pre-processed aluminum plate set is stored according to the installation area classification, waiting for the subsequent sub-area storage process.

[0085] S430, according to the keel sub-area number, the pre-processed aluminum plate set is classified and stored, and an aluminum plate in-place state table is generated.

[0086] Specifically, the classified storage is completed by an intelligent storage system. According to the keel sub-area number, the system automatically allocates storage locations, and each location corresponds to an installation area. The intelligent storage system transports the pre-processed aluminum plate by AGV, identifies the two-dimensional code label on the back of the aluminum plate by machine vision, and transports it to the specified location. During storage, the system monitors the warehouse environment parameters in real time to maintain the temperature at 20±2℃ and the relative humidity within 45%±5% to prevent deformation or oxidation of the aluminum plate.

[0087] Further, the aluminum plate in-place state table records the storage location, quantity state, and quality condition of each sub-area aluminum plate in real time. The aluminum plate in-place state table is associated with the BIM model, and the preparation of the aluminum plate in each area can be displayed in three dimensions. For aluminum plates with special shapes, the system records their installation direction and matching relationship with adjacent plates to prevent misalignment during installation. During storage, periodic sampling inspection is carried out on the aluminum plate, and the sampling inspection ratio is not less than 10% to ensure stable material quality.

[0088] Understandably, the aluminum sheet placement status table includes the following core content: zoned inventory statistics, quality inspection records, and dispatch schedules. The zoned inventory statistics display the number of aluminum sheets arriving, the number of pre-processed sheets completed, and the number to be replenished in each zone. The quality inspection records store the inspection data and problem resolution status of all inspected aluminum sheets. The dispatch schedule pre-plans the dispatch sequence and transportation routes of the aluminum sheets based on the construction schedule. Once the aluminum sheet placement status table is digitally signed and confirmed, as a prerequisite for the S510 aluminum sheet installation, its data is directly linked to the material management system for subsequent construction phases.

[0089] Step 500 at least includes steps S510-S530: S510: Obtain the aluminum plate in-place status table and installation sequence diagram, perform aluminum plate grabbing and transportation path planning, and obtain an aluminum plate installation queue.

[0090] Specifically, the acquisition of the aluminum plate in-place status table is achieved by reading the warehouse management file generated by S430. The aluminum plate in-place status table is stored in a structured database format, and contains key information such as partition inventory statistics, quality inspection records, and warehouse delivery pre-arrangements. When reading, the data integrity check code is first verified, and then the file content is parsed to extract the storage location, specification parameters and pre-processing status of each partition aluminum plate. The installation sequence diagram is a construction guidance document derived from the BIM model, which contains the order of aluminum plate installation, the matching relationship between adjacent plates, and special node processing requirements. The installation sequence diagram is data-associated with the aluminum plate in-place status table to ensure that the drawing requirements are completely consistent with the physical materials.

[0091] Furthermore, the grabbing of the aluminum plate is completed by an intelligent warehouse robotic arm system. According to the storage location information in the aluminum plate in-place status table, the robotic arm locates the target aluminum plate through the visual recognition system and grabs it using a vacuum suction cup device. During the grabbing process, the system scans the QR code label on the back of the aluminum plate in real time to verify whether its specifications and models match the installation requirements. For special-shaped aluminum plates, the robotic arm automatically adjusts the grabbing angle to ensure that no deformation occurs during transportation. The transportation path planning is a dynamic path model established based on the three-dimensional point cloud data of the construction site. Taking into account factors such as the working range of the lifting equipment, the distribution of channel obstacles and the personnel activity area, the Dijkstra algorithm is used to calculate the optimal transportation route, and the path planning accuracy reaches ±50mm.

[0092] Understandably, the aluminum panel installation queue includes the following: an installation task assignment table, a materials delivery list, and a construction schedule. The installation task assignment table details the areas each installation team is responsible for and the corresponding aluminum panel numbers; the materials delivery list records the transportation route and estimated arrival time for each aluminum panel; and the construction schedule specifies the installation time windows and quality control milestones for each area. The aluminum panel installation queue is stored in a format recognizable by the BIM collaboration platform and synchronized to each construction terminal device via a wireless network.

[0093] S520. According to the aluminum plate installation queue, the aluminum plate is embedded into the keel slot, the joints are aligned and temporarily fixed, and a preliminary aluminum plate installation report is generated.

[0094] Specifically, the embedding of the aluminum plate is completed by installing a robotic arm in collaboration with manual labor. The construction personnel use AR equipment to view the three-dimensional positioning information in the aluminum plate installation queue and mark the installation position of the aluminum plate on the keel. After the installation robotic arm transports the aluminum plate to the specified height, it uses a laser rangefinder to guide the edge of the aluminum plate to accurately align with the keel slot. During the embedding process, the force sensor monitors the contact pressure in real time, and it is judged to be in place when the pressure value reaches 5N±0.5N. For curved modeling areas, a flexible clamp is used to adjust the curvature of the aluminum plate to ensure that it matches the curvature of the keel, and the curvature deviation does not exceed ±1mm / m.

[0095] Furthermore, the seam alignment is controlled by a high-precision measurement system. An electronic seam meter is used to measure the seam width between adjacent aluminum plates, and the standard seam is controlled within the range of 3mm±0.5mm. For special-shaped seams, a three-dimensional scanner is used to obtain the actual seam type data, which is compared with the design model to guide adjustments. The temporary fixation uses a magnetic clamp with a special buckle, and the fixing force is controlled within the range of 20N·m±2N·m to ensure that the position of the aluminum plate is stable and does not deform. After each fixing point is installed, the position is immediately reviewed, and points with a deviation of more than ±1mm need to be readjusted.

[0096] The preliminary aluminum panel installation report will include the following: installation progress records, quality inspection data, and a problem resolution checklist. The installation progress record provides real-time updates on the installation completion status of each area; the quality inspection data includes the measured width and flatness of all joints; and the problem resolution checklist documents any anomalies discovered during the installation process and their resolution. The preliminary aluminum panel installation report can be entered on-site via a mobile device and automatically linked to the corresponding location in the BIM model, supporting 3D visualization.

[0097] S530. Fine-tune and lock the aluminum plate in the preliminary installation report of the aluminum plate, and generate an aluminum plate installation completion report.

[0098] Specifically, fine-tuning is accomplished using a precision adjustment device. Based on the deviation data in the preliminary installation report for the aluminum panels, a six-way adjuster is used to fine-tune the position of the aluminum panels. Horizontal adjustment accuracy reaches ±0.3mm, and vertical adjustment accuracy reaches ±0.5mm. After adjustment, a laser tracker is used to perform full-station measurements to ensure overall flatness within a 2mm / 2m range and joint straightness deviations do not exceed ±1mm / m. For large, continuous aluminum panels, an electronic level is used for overall verification to ensure visual continuity.

[0099] Furthermore, the locking process is achieved through an intelligent torque control system. According to the tightening sequence required by the design, the fixing bolts are gradually tightened from the center to the surrounding areas. The intelligent torque wrench automatically sets the torque value according to the thickness and material of the aluminum plate. The torque in the standard area is controlled at 8N·m±0.5N·m, and the torque in the edge area is appropriately increased to 10N·m±0.5N·m. The torque-angle curve is recorded in real time during the tightening process, and an automatic alarm is issued when the curve is abnormal. After all fastening points are completed, 24-hour stress monitoring is carried out to ensure that there is no abnormal deformation.

[0100] The aluminum panel installation completion report, as understood, includes the following core elements: a final acceptance data sheet, a tightening force record, and final imagery. The final acceptance data sheet records the measured positional parameters and quality indicators of all aluminum panels; the tightening force record records the final torque value and operation time for each tightening point; and the final imagery includes an overall rendering and detailed drawings of key nodes. Once digitally signed, the aluminum panel installation completion report serves as input for the S610 quality inspection, with its data directly linked to the final acceptance system.

[0101] Step S600 at least includes steps S610-S630: S610. Obtain the aluminum panel installation completion report and design drawings, conduct joint tightness and flatness inspections, and generate an inspection problem list.

[0102] Specifically, the acquisition of the aluminum plate installation completion report is achieved by reading the acceptance document generated by S530. The aluminum plate installation completion report is stored in a structured database format and contains key information such as the final acceptance data table, tightening force records and completion image data. When reading, the digital signature is first verified, and then the file content is parsed to extract the measured position parameters and quality indicators of the aluminum plates in each area. The design drawings are construction guidance documents derived from the BIM model, which contain technical parameters such as the theoretical position of the aluminum plate arrangement, joint width standards and flatness requirements. The design drawings are data-linked with the aluminum plate installation completion report to ensure that the inspection standards are completely consistent with the actual construction conditions.

[0103] Further, the splicing tightness detection is completed by cooperating a laser joint meter and a three-dimensional scanner. The joint width between adjacent aluminum plates is measured by using a laser joint meter with a precision of 0.01 mm, and the standard joint is controlled within a range of 3 mm±0.5 mm. For the special-shaped joint, a three-dimensional scanner is used to obtain actual joint point cloud data, and a three-dimensional comparison is made with the design model, and the area with a deviation of more than ±1 mm is marked as abnormal. The flatness detection is performed by cooperating an electronic level and a total station, and the overall flatness is controlled within a range of 2 mm / 2 m, and the local flatness is controlled within a range of 1 mm / m. All detection data is uploaded to a quality management platform in real time, and a deviation curve is automatically generated.

[0104] Understandably, the inspection problem list includes the following contents: problem position distribution map, deviation data table and rectification suggestion. The problem position distribution map marks the three-dimensional coordinates of all unqualified points on the BIM model; the deviation data table records the measured value, design value and deviation of each problem point; and the rectification suggestion provides a specific treatment scheme according to the problem type. The inspection problem list is stored in a format recognizable by a BIM collaborative platform and is synchronized to each construction terminal device through the cloud.

[0105] S620, according to the inspection problem list, the aluminum plate joint and the fixing part are locally adjusted to generate a rectification record table.

[0106] Specifically, the local adjustment is completed by a precision adjustment system. According to the deviation data in the inspection problem list, a six-way fine adjustment device is used to correct the position of the problem aluminum plate. The horizontal adjustment accuracy reaches ±0.2 mm, and the vertical adjustment accuracy reaches ±0.3 mm. For the area with unqualified joint width, a special spreader or tightener is used for adjustment to ensure that the joint uniformity is controlled within a range of ±0.3 mm. The stress change of the aluminum plate is monitored in real time during the adjustment process to prevent deformation caused by excessive adjustment.

[0107] Further, the fixing part treatment is realized by an intelligent torque control system. The fixing bolts of the problem area are checked one by one, and an intelligent torque wrench is used to re-tighten. The tightening sequence follows the principle of from the center to the periphery, and the torque value is controlled within a range of ±5% of the design value. All deformed or damaged fixing parts are replaced, and the material and specifications of the new fixing parts are completely consistent with the original design. After the treatment is completed, 24-hour stress monitoring is performed to ensure that there is no abnormal deformation.

[0108] Understandably, the rectification record table includes the following: a rectification area distribution map, adjustment parameter records, and retest data. The rectification area distribution map marks the locations of all rectification points on the BIM model; the adjustment parameter records detail the treatment method and adjustment amount for each rectification point; and the retest data stores the inspection results after rectification. The rectification record table is entered on-site via a mobile terminal and automatically linked to the corresponding location on the BIM model, supporting 3D visual query.

[0109] S630. Compare the rectification record sheet with the acceptance criteria, generate a final acceptance report and feed it back to the 3D model database for closed-loop optimization.

[0110] Specifically, the comparison to acceptance standards is accomplished through a quality assessment system. The system reads the retest data from the rectification record sheet and automatically compares it against the acceptance standards. For the flatness indicator, the least squares method is used to calculate the overall fit; for the joint indicator, the pass rate is calculated and the standard deviation is calculated. All comparison results are combined to generate a quality assessment curve, visually displaying the distribution of construction quality. The assessment process utilizes a fuzzy logic algorithm, comprehensively considering the weighting of various indicators.

[0111] Furthermore, the generated final acceptance report includes the following core content: a quality assessment conclusion, a comparison of rectification results, and a description of outstanding issues. The quality assessment conclusion provides the overall quality grade and pass rate; the rectification comparison shows the changes in quality indicators before and after rectification; and the description of outstanding issues records any defects that cannot be addressed and subsequent solutions. The final acceptance report is digitally signed and has legal effect.

[0112] Understandably, this feedback to the 3D model database is achieved through a data interface. Key parameters from the final acceptance report are updated to the BIM model, including information such as actual installation location, joint treatment, and quality grade. Simultaneously, empirical data from the construction process is stored in a knowledge base to provide reference for subsequent projects. The closed-loop optimization system automatically analyzes construction deviation patterns and optimizes design parameters and construction processes for subsequent projects, forming a continuous improvement mechanism.

[0113] In another implementation, the rectification record table is compared with closed-loop optimization.

[0114] Innovative algorithm ①: Quantification of mass deviation based on improved Chebyshev norm Formula definition: , in: is the quantified value of quality deviation; : A set of detection indicators (such as flatness / seam width); : Measured value vector (from the rectification record table); : Acceptance criteria vector (ISO 19650); : Allowable tolerance threshold (dynamic range [0.5mm, 2mm]); : Standard deviation of the construction process (S610 test data); =0.8, =1.2, =0.5: attenuation coefficient; Function implementation: Read the retest data in the rectification record table (including 12 indicators such as flatness and joint width); call the acceptance criteria in the BIM database With historical tolerance ; Calculate the relative deviation of each indicator: numerator is the absolute deviation; the denominator Achieve tolerance standardization; through the Sigmoid function Reduce noise of high-frequency fluctuation data; take the maximum value Identify the most serious quality defects Furthermore, innovative algorithm ②: fuzzy decision fusion model Formula definition: , in: It is a fuzzy decision fusion model; fuzzy integral operator; : membership function of the kth type of defect (such as pores / deformation); : Weight coefficient (dynamically assigned by the expert system); : Deviation gradient (reflects the defect diffusion trend); : average deviation of the region; = 0.3: mutation penalty coefficient; Functional implementation: Input algorithm ① Output deviation ; Construct three types of fuzzy rules: Rule 1: If 0.3, then the quality level = excellent; Rule 2: If 0.3 < 0.7 then quality level = good; Rule 3: If > 0.7, then the quality level = unqualified; calculate the gradient penalty term Suppress local mutation interference; fuse multi-source data through Sugeno integration to generate quality grade decisions.

[0115] Furthermore, innovative algorithm ③: Lie group manifold feedback optimization Formula definition: , in: Optimize results for feedback; : Original BIM model parameters (curvature / normal vector); : Jacobi matrix (describing parameter sensitivity); = 0.05: learning rate; : Model coordinate system transformation matrix; : Construction site coordinate system; : Defect location tensor (from the rectification record table); : Lie group multiplication operator; Function realization: Extract defect coordinates (Positioning data adjusted by S620); Calculate coordinate deviation ; Through index mapping Generate parameter update amount; update BIM model: ; Automatically generate an acceptance report with a digital signature (PDF / A format).

[0116] Technical effects: This step achieves an intelligent closed loop for construction acceptance through three innovative algorithms: (1) Quantification of quality deviation: The dynamic deviation between the measured value and the standard is calculated based on the improved Chebyshev norm, and the noise is suppressed by the Sigmoid function.

[0117] (2) Fuzzy decision fusion: Sugeno integral is used to fuse the deviation gradient and expert weight to generate quality grades (excellent / good / unqualified), thereby improving decision accuracy.

[0118] (3) Lie group feedback optimization: By updating the BIM model parameters through Lie algebra, the welding qualification rate of subsequent projects is improved, the tolerance threshold is tightened annually, and a cross-cycle self-optimization capability is formed.

[0119] Core breakthrough: transform acceptance data flow into a mathematical entity that integrates differential geometry and group theory (D→F→ ), realize the millisecond-level closed loop of "detection → decision → feedback" (response delay <50ms), and lay the foundation for the core algorithm of digital twin construction.

[0120] Embodiment two: Figure 2 A structural block diagram of a building decoration suspended ceiling design system based on BIM technology according to an embodiment of the present application is shown. As Figure 2 indicated, the structure can include: A three-dimensional modeling module 10 for obtaining main structure data, reconstructing a three-dimensional model and extracting connector coordinates; Obtain point cloud data of the main structure of the building (accuracy ±1mm) through a laser scanner, and synchronously access architectural design drawings (DWG / CAD format); Generate a three-dimensional model with curvature features based on an improved Poisson surface reconstruction algorithm; Automatically identify the boundary of the overhanging area, extract the installation coordinates of the connector, and convert them into a standard ccd format document; Input: on-site scanning data, design drawings; Output: standardized connector coordinate data document (directly for total station analysis).

[0121] A welding control module 20 for obtaining connector coordinate data documents, performing coordinate calibration and point matching; Align the BIM coordinates with the total station coordinate system; Generate mechanical arm welding motion trajectories; Real-time feedback of welding temperature / position deviation (ultrasonic flaw detection accuracy 0.1mm); Input: coordinate data documents, total station parameters; Output: welding quality compliance list (including 32 parameters such as torque / position).

[0122] A keel installation module 30 for installing keels according to connector positions and performing leveling and reinforcement; Automatically match the keel model according to the welding list (RFID chip identification); Dynamically adjust the installation angle based on electronic level data (accuracy 0.02mm / m); Check the bolt pre-tightening force; Input: welding quality list, keel design drawing; Output: keel leveling completion report (including levelness / reinforcement torque matrix).

[0123] An aluminum plate processing module 40 for matching aluminum plate specifications and performing preprocessing classification; Optimize aluminum plate arrangement according to keel zoning data; Full-automatic dust removal + chamfering treatment; AGV car is classified according to installation sequence (error rate <0.2%); Input: Keel leveling report, aluminum plate parameters; Output: Aluminum plate in-place state table (with AR visualization process).

[0124] Aluminum plate installation module 50 is used to plan the installation sequence and complete the embedded locking; Project installation path through HoloLens 2; Six degrees of freedom to adjust the joint width; 24-hour monitoring of aluminum plate deformation; Input: Aluminum plate in-place state table; Output: Aluminum plate installation completion report (including 3000+ detection point data).

[0125] Acceptance feedback module 60 is used to detect the quality of aluminum plate installation and close-loop feedback data; Laser joint detector + three-dimensional scanner synchronous verification; Generate BIM model and actual deviation thermal map; Integrate the rectification data back to the modeling algorithm; Input: Installation completion report; Output: Final acceptance report.

[0126] Obviously, the above described embodiments are only part of the embodiments of the present application, not all the embodiments, and the preferred embodiments of the present application are given in the drawings, but do not limit the patent scope of the present application. The present application can be realized in many different forms, and on the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for part of the technical features. Any equivalent structure made by using the contents of the specification and drawings, directly or indirectly applied to other related technical fields, is also within the scope of the patent protection of the present application.

Claims

1. A method for designing architectural decorative ceilings based on BIM technology, characterized in that: include: Obtain main structure data, reconstruct 3D model and extract connector coordinates to generate standardized connector coordinate data documents; The process of generating a standardized connector coordinate data document includes constructing a standardized data storage format; Obtain the coordinate data file of the connector, perform coordinate calibration and point matching, perform point-by-point positioning welding, ultrasonic flaw detection and position review, and generate a welding quality compliance list; Install the keel according to the position of the connectors, perform level calibration and bolt reinforcement, and generate a keel leveling completion report; Match the specifications and colors of the aluminum plates, perform surface dust removal and edge chamfering, store them by keel partitions, and generate an aluminum plate placement status table; Plan the installation sequence and insert the keel slots, perform joint alignment, fine-tuning and locking, and generate an aluminum panel installation completion report; Inspect the flatness and joint tightness of the aluminum plates, make local adjustments to the aluminum plate joints and fixings, generate a final acceptance report and feed it back to the 3D model database for closed-loop optimization.

2. The architectural decorative ceiling design method according to claim 1, characterized in that: The method of obtaining the main structure data, reconstructing the three-dimensional model and extracting the coordinates of the connectors includes: Obtain laser scanning point cloud data and design drawings of the main structure, perform noise removal and precision reconstruction, and obtain 3D structure point cloud data; Extract the cantilever area and connector installation position information from the 3D structure point cloud data, perform mesh fitting processing, and obtain 3D model data; Extract connector coordinates and convert formats of 3D model data to generate standardized connector coordinate data documents.

3. The architectural decorative ceiling design method according to claim 1, characterized in that: The obtaining of the connector coordinate data file and performing coordinate calibration and point matching includes: Obtain the coordinate data file of the connector, combine it with the measurement parameters of the total station, perform coordinate calibration and point matching, and obtain the calibrated welding coordinate sequence; Based on the calibrated welding coordinate sequence, control the welding equipment to perform point-by-point positioning welding and generate a welding completion report; Perform ultrasonic flaw detection and position review on the welds in the welding completion report to generate a welding quality compliance list.

4. The architectural decorative ceiling design method according to claim 1, characterized in that: The installation of the keel according to the position of the connector, level calibration and bolt reinforcement include: Obtain welding quality compliance lists and keel design drawings, perform material sorting and transportation route planning, and obtain keel installation preparation data; Based on the keel installation preparation data, install the main keel and secondary keel in sequence according to the position of the connectors, and generate a preliminary keel installation report; Perform level meter calibration and bolt reinforcement on the keel in the preliminary keel installation report, and generate a keel leveling completion report.

5. The architectural decorative ceiling design method according to claim 1, characterized in that: The matching of aluminum plate specifications and color screening, surface dust removal and edge chamfering treatments include: Obtain the keel leveling completion report and aluminum plate design parameters, perform specification matching and color screening, and obtain an aluminum plate selection list; Perform surface dust removal and edge chamfering on the aluminum plates in the aluminum plate selection list to generate a pre-treated aluminum plate set; The pre-processed aluminum plate collection is classified and stored according to the keel partition number, and an aluminum plate placement status table is generated.

6. The architectural decorative ceiling design method according to claim 1, characterized in that: The planning of the installation sequence and insertion into the keel slots, and the execution of seam alignment, fine-tuning, and locking processes include: Obtain the aluminum plate in-place status table and installation sequence diagram, perform aluminum plate grabbing and transportation path planning, and obtain the aluminum plate installation queue; Insert the aluminum panels into the keel slots according to the aluminum panel installation queue, perform joint alignment and temporary fixation, and generate a preliminary aluminum panel installation report; Fine-tune and lock the aluminum plates in the preliminary installation report and generate a completion report for the aluminum plate installation.

7. The architectural decorative ceiling design method according to claim 1, characterized in that: The above mentioned inspection of the flatness and tightness of the aluminum plates and local adjustment of the aluminum plate joints and fixings include: Obtain the aluminum panel installation completion report and design drawings, conduct joint tightness and flatness inspections, and generate an inspection problem list; Make local adjustments to the aluminum plate joints and fixings based on the inspection problem list and generate a rectification record sheet; Compare the rectification record sheet with the acceptance criteria, generate a final acceptance report and feed it back to the 3D model database for closed-loop optimization.

8. The architectural decorative ceiling design method according to claim 2, characterized in that: The extracting of the overhanging area and the connection part installation position information from the three-dimensional structure point cloud data includes: Segment the overhang area based on the point cloud density threshold and extract the 3D coordinates of the connector installation location; The B-spline curve is used to fit the installation boundary of the connector to generate the grid topology.

9. The architectural decorative ceiling design method according to claim 3, characterized in that: The control welding equipment performs point-by-point positioning welding, comprising: Generate robot arm motion trajectory according to welding coordinate sequence; Monitor welding current and temperature in real time and adjust welding parameters dynamically.

10. A building decoration ceiling design system based on BIM technology, used to implement the method according to any one of claims 1 to 9, characterized in that: include: 3D modeling module, used to obtain main structure data, reconstruct 3D model and extract connector coordinates; Welding control module, used to obtain the coordinate data file of the connection parts, perform coordinate calibration and point matching; Keel installation module, used to install the keel according to the position of the connector and perform leveling and reinforcement; Aluminum plate processing module, used to match aluminum plate specifications and perform pre-processing and classification; Aluminum plate installation module, used to plan the installation sequence and complete the scarfing and locking; The acceptance feedback module is used to detect the installation quality of aluminum panels and provide closed-loop feedback data.

Images