A precise construction method for special-shaped curtain wall based on BIM and three-dimensional scanning

Abstract

The application provides a special-shaped curtain wall precise construction method based on BIM and three-dimensional scanning, and belongs to the technical field of digital construction of curtain wall engineering. In the method, a digital modeling of an adjustable connecting node is performed in a BIM model, geometric characteristics, adjustment degrees of freedom and records of the node are established, and a unique identifier is bound. Then, three-dimensional scanning point clouds of a main body structure are collected, accurate registration of the point clouds and the BIM model is realized through feature point pre-alignment and an ICP algorithm, a structural deviation field is constructed, and the target installation position and the target installation posture of the adjustable connecting node are derived accordingly. After the node is installed on site, pre-positioning adjustment is performed by using a reconfigurable adjustment mechanism, and the displacement and the angle of each adjustment degree of freedom are collected in real time by using a micro-sensing component to generate node pre-positioning data. The method realizes data connection in the whole process of design, construction, acceptance and operation and maintenance, and significantly improves the installation accuracy, construction controllability and operation and maintenance visualization level of the special-shaped curtain wall.

Description

Technical Field

[0001] This invention relates to the field of digital construction technology for curtain wall engineering, and in particular to a precise construction method for irregularly shaped curtain walls based on BIM and 3D scanning. Background Technology

[0002] With the continuous innovation of modern architectural design concepts, irregularly shaped curtain walls, due to their unique artistic expression and complex spatial geometry, have been widely used in large public buildings, commercial complexes, and landmark buildings. However, irregularly shaped curtain walls typically feature large curvature variations, diverse panel specifications, and complex spatial positioning, which places extremely high demands on construction and installation precision. Although Building Information Modeling (BIM) and 3D scanning technologies have been gradually applied to curtain wall projects, the following shortcomings still exist in actual construction and life-cycle management:

[0003] 1. Traditional construction of irregularly shaped curtain walls often relies on idealized BIM models. However, after the main structure is completed, the actual structure often exhibits complex deviation fields due to factors such as concrete pouring errors and settlement deformation. While existing technologies can use 3D scanning to acquire point clouds, they often lack a systematic method to automatically convert the "point cloud deviation field" into "node-level adjustment instructions." Construction workers still need to rely heavily on manual measurement and experience-based judgment to account for structural errors, resulting in low installation efficiency and difficulty in guaranteeing accuracy.

[0004] 2. Existing adjustable connection nodes for curtain walls are mostly purely mechanical structures (such as slotted embedded parts and gear-type adjusting parts), possessing only basic translational or rotational adjustment functions and lacking self-sensing and data recording capabilities. During pre-positioning and fine-tuning, the adjustment freedom (translation and rotation) of the nodes relies entirely on manual operation by workers, making it difficult to digitally record the displacement and angle changes during the adjustment process in real time. This results in the loss of core quality data during construction after installation, making it difficult to create traceable digital construction archives.

[0005] Therefore, we propose a precise construction method for irregularly shaped curtain walls based on BIM and 3D scanning. The description of the prior art in this specification is only for the purpose of helping to understand the technical background of this invention and does not constitute any admission of the prior art. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by providing a precise construction method for irregularly shaped curtain walls based on BIM and 3D scanning, thereby solving the technical problems mentioned in the background section.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A precise construction method for irregularly shaped curtain walls based on BIM and 3D scanning includes the following steps:

[0009] S1. Digitally model the adjustable connection nodes used for installing curtain wall panels; during the digital modeling process, establish a unique identifier for each adjustable connection node, and structurally store the adjustment freedom, adjustable range and recorded parameters of the connection node in the BIM model to form node-level digital construction information.

[0010] S2. After the main structure construction is completed, obtain the three-dimensional scanning point cloud data of the area to be installed, and spatially register the point cloud data with the BIM model; determine the target installation benchmark for each adjustable connection node, write the target installation benchmark into the BIM model, and form node-level installation guidance information.

[0011] S3. At the installation site, each adjustable connection node is deployed, and the position and attitude of the node are pre-positioned using a reconfigurable adjustment mechanism. The micro-sensor components monitor and temporarily store the pre-positioning parameters of the node in each degree of freedom of adjustment in real time, forming node-level pre-positioning data.

[0012] S4. Install the corresponding curtain wall panels onto the adjustable connection nodes, and use the reconfigurable adjustment mechanism to perform fine adjustments along each degree of freedom to make the curtain wall panels meet the predetermined assembly accuracy requirements; the micro-sensor components automatically collect and lock the final adjustment amount of the nodes in each degree of freedom to generate structured adjustment data.

[0013] S5. The structured adjustment data is written into the BIM model in real time through the on-site mobile terminal or communication module, so that the final adjustment status of each adjustable connection node forms a traceable installation record in the BIM model; during the reverse writing process, the node installation status, locking status and adjustment direction vector are bound to the corresponding node unique identifier as associated attributes, forming a node-level digital construction record.

[0014] S6. After all curtain wall panels are installed, the 3D scanning data is collected again and aligned with the BIM model after being reverse-written. The installation quality is checked based on the consistency between the final adjustment amount of the nodes and the actual posture of the curtain wall. The node-level digital construction record is used as the basic data for subsequent operation and maintenance, which is used to locate potential loose nodes, analyze local deformation trends, and guide structural reinforcement strategies.

[0015] S1 specifically includes: collecting structural information of adjustable connection nodes used for installing curtain wall panels, including the degree of freedom of adjustment of the reconfigurable adjustment mechanism, the adjustable range, and the recording parameter types of the micro-sensing components;

[0016] Based on the collected construction information, the digital features of adjustable connection nodes are defined in the BIM model, including the geometric features, adjustment features and recording features of the nodes, and a node-level digital description is formed.

[0017] A unique identifier is generated for each adjustable connection node, and this unique identifier is bound to the node's digital features, so that the node has stable identification attributes in the BIM model.

[0018] The adjustment degrees of freedom, adjustment direction, and adjustable range of the reconfigurable adjustment mechanism are stored in the BIM model in the form of structured fields, so that the adjustment parameters have traceability and recallability.

[0019] Based on the aforementioned binding and structured information, a digital construction model of adjustable connection nodes is generated in the BIM model, providing standardized node data for subsequent installation benchmark determination.

[0020] S2 specifically includes: after the main structure is completed, using a 3D scanning device to acquire structural point cloud data of the node installation area to form a scanning dataset containing the actual spatial morphology;

[0021] The collected scanned point cloud is spatially registered with the BIM model to obtain the structural deviation field, providing a spatial reference for determining the node installation benchmark.

[0022] Based on the registered deviation field, the target installation position of each adjustable connection node is derived so that the node position can adapt to the actual structural form.

[0023] Based on the design and assembly relationship of the curtain wall panels, determine the target installation posture of each adjustable connection node, including the rotation angle and normal orientation;

[0024] Write the target installation location and target installation posture of the node into the BIM model, so that the installation benchmark can form a visualized and callable node-level guidance information.

[0025] S3 specifically includes: based on the node installation benchmark in the BIM model, setting up adjustable connection nodes at the corresponding locations on the installation site to ensure that the node placement is consistent with the design benchmark;

[0026] The adjustable connection node is prepositioned along its degree of freedom using a reconfigurable adjustment mechanism so that the node position and attitude are close to the target installation reference.

[0027] During the pre-positioning process, the micro-sensing components monitor the pre-positioning parameters of the node in each degree of freedom of adjustment in real time, forming real-time pre-positioning data;

[0028] The pre-positioning parameters obtained from real-time monitoring are temporarily stored in the node's internal storage module or mobile terminal, enabling the pre-positioning parameters to have subsequent adjustment and comparison functions.

[0029] Based on the temporary parameters, node-level prepositioning data is generated, enabling the prepositioning status of nodes to participate in subsequent fine-tuning judgments in a structured form.

[0030] S4 specifically includes: installing the corresponding curtain wall panel to the assembly interface of the adjustable connection node, so that the panel is in an adjustable state and establishes a stable connection with the node;

[0031] The reconfigurable adjustment mechanism is used to perform fine adjustments along each degree of freedom to ensure that the curtain wall panels meet the assembly accuracy requirements, including positional accuracy and attitude accuracy.

[0032] Upon completion of fine-tuning, the micro-sensing components automatically monitor the final adjustment amount of the node in each degree of freedom of adjustment to obtain the final adjustment state of the node.

[0033] The final adjustment amount is latched and a node-level adjustment amount record is formed using the node's unique identifier;

[0034] The latched adjustment values ​​are organized into structured adjustment data to provide a standardized data format for subsequent writing into the BIM model.

[0035] S5 specifically includes: reading structured adjustment data through mobile terminals or communication components, enabling the adjustment data to be uploaded remotely;

[0036] The structured adjustment data is parsed to ensure its completeness, including the final adjustment amount, adjustment direction vector, and locking status.

[0037] The parsed adjustment data is bound to the unique identifier of the node, so that the adjustment data and the node identity are accurately matched;

[0038] Write the bound adjustment data into the BIM model so that the final installation status of the node forms a traceable installation record in the BIM model;

[0039] Based on the reverse-written adjustment data, node-level digital construction records are generated, providing a data foundation for quality tracking and subsequent maintenance.

[0040] S6 specifically includes: after all curtain wall panels are installed, a three-dimensional scanning point cloud of the curtain wall area is collected again to obtain as-built scanning data;

[0041] Spatially align the as-built scan data with the BIM model containing node adjustment data to form an as-built verification dataset.

[0042] Based on the completion verification data, analyze the consistency between the final adjustment amount of the nodes and the actual curtain wall posture, and determine whether the installation quality meets the requirements.

[0043] If the review results show local deviations, potential loose or abnormal nodes will be located based on the difference between the adjustment data in the BIM model and the as-built posture.

[0044] Based on digital construction records and as-built comparison data, it provides support for subsequent operation and maintenance, including node reinforcement strategies, displacement trend analysis, and safety assessment, thereby achieving long-term precise management.

[0045] The beneficial effects of this invention are as follows:

[0046] This invention defines the geometric features, adjustment degrees of freedom, and recording features of adjustable connection nodes in the BIM model, and combines this with micro-sensing components to achieve real-time acquisition of node adjustment amounts, making the nodes digital components with identifiable, calculable, and traceable attributes. Compared with traditional methods that rely on manual experience for adjustment, this invention can quantify the adjustment behavior of each degree of freedom of the node, standardizing and digitizing the curtain wall installation and adjustment process, and significantly improving the installation consistency of complex curved curtain walls.

[0047] This invention establishes a deviation field reflecting the true form of the main structure by registering 3D scanned point clouds with BIM models, and derives the target installation position and orientation of adjustable connection nodes. This eliminates the need for manual judgment in adjustment, allowing it to be automatically driven by the actual structural form. Unlike existing methods that rely solely on scan-based "report-style comparisons," this invention creates an automated closed loop in the chain of "scan results → node adjustment amount," effectively preventing structural deviations from being amplified or accumulated during curtain wall installation.

[0048] This invention, through the use of position error functions, attitude error functions, and an adjustment degree-of-freedom mapping model, introduces the curtain wall installation and adjustment process into a computable solution framework for the first time, transforming the determination of adjustment amounts from trial-and-error adjustments to mathematical solutions. The optimal adjustment amount is generated using the Jacobian matrix and the least squares algorithm, combined with real-time sensor calibration, achieving high-precision installation with panel position errors typically below 2mm and attitude errors below 0.5°, significantly improving the construction quality of irregularly shaped curtain walls.

[0049] This invention achieves real-time digital storage of node installation status through the parsing, binding, and attribute writing of structured adjustment data, transforming the BIM model from a simple design model into a "real-time construction model." This digital record can be used for construction acceptance, quality traceability, and third-party auditing, avoiding the industry pain points of missing adjustment records and inability to trace responsibility in traditional methods, and significantly improving the transparency of construction management.

[0050] Based on the deviation field, attitude residuals, and local gradient changes obtained from as-built scanning, this invention constructs a node loosening identification model to automatically locate nodes with abnormal installation conditions. Unlike traditional methods that rely on visual inspection or sampling, this invention can accurately identify nodes with minute displacements or attitude shifts throughout the entire curtain wall area, proactively detecting potential safety hazards and significantly improving detection efficiency and diagnostic accuracy during operation and maintenance.

[0051] This invention continuously incorporates digital construction records and as-built deviation fields into the BIM model, and constructs deformation trend models and node health levels, enabling the curtain wall system to possess long-term status visualization, health prediction, and maintenance decision-making capabilities. Compared to traditional maintenance methods that rely on manual experience, this invention can dynamically assess the structural health status during operation, promptly locate potential risk nodes, and achieve data-driven proactive maintenance, significantly improving the long-term safety and stability of irregularly shaped curtain walls. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of a precise construction method for irregularly shaped curtain walls based on BIM and 3D scanning according to the present invention.

[0053] Figure 2 This is a schematic diagram of the first BIM digital model of the irregularly shaped curtain wall panel in an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of the second BIM digital model of the irregular curtain wall panel in an embodiment of the present invention. Detailed Implementation

[0055] 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Example 1: As Figure 1 As shown, this embodiment provides a precise construction method for irregularly shaped curtain walls based on BIM and 3D scanning, including the following steps:

[0057] S1. Digital Modeling Steps for Intelligent Adjustable Connection Nodes: In the BIM model of the curtain wall project, the adjustable connection nodes used for installing curtain wall panels are digitally modeled. The adjustable connection nodes include reconfigurable adjustment mechanisms for achieving multi-degree-of-freedom fine-tuning and micro-sensor components for recording the adjustment status. During the digital modeling process, a unique identifier is established for each adjustable connection node, and the adjustment degrees of freedom, adjustable range, and recorded parameters of the connection node are structured and stored in the BIM model to form node-level digital construction information.

[0058] S2. Steps for determining the installation benchmark of connection nodes based on 3D scanning: After the main structure construction is completed, obtain the 3D scanning point cloud data of the area to be installed, and spatially register the point cloud data with the BIM model; based on the structural deviation results after registration, determine the target installation benchmark for each adjustable connection node, including the target spatial position of the node, the target installation posture, and the expected assembly relationship with the corresponding curtain wall panel, and write the target installation benchmark into the BIM model to form node-level installation guidance information.

[0059] S3. On-site installation and pre-positioning steps of adjustable connection nodes: Based on the target installation benchmark, each adjustable connection node is deployed on the installation site, and its reconfigurable adjustment mechanism is used to perform pre-positioning adjustment of the node position and attitude; during the pre-positioning process, the micro-sensing component monitors and temporarily stores the pre-positioning parameters of the node in each degree of freedom of adjustment in real time, forming node-level pre-positioning data.

[0060] S4. Assembly and Adjustment Recording Steps for Curtain Wall Panels: After the pre-positioning of the nodes is completed, the corresponding curtain wall panels are installed on the adjustable connection nodes, and fine-tuning is performed along each degree of freedom using the reconfigurable adjustment mechanism to ensure that the curtain wall panels meet the predetermined assembly accuracy requirements. When the fine-tuning is completed, the micro-sensor components automatically collect and lock the final adjustment amount of the nodes in each degree of freedom, and generate structured adjustment data by combining the unique node identifier.

[0061] S5. Real-time back-writing and digital storage steps for connection node adjustment data: The structured adjustment data is written into the BIM model in real time through the on-site mobile terminal or communication module, so that the final adjustment status of each adjustable connection node forms a traceable installation record in the BIM model; during the back-writing process, the node installation status, locking status and adjustment direction vector are bound to the corresponding node unique identifier as associated attributes to form a node-level digital construction record.

[0062] S6. Installation quality verification and subsequent maintenance support steps based on BIM model: After all curtain wall panels are installed, 3D scanning data is collected again and aligned with the reverse-written BIM model. The installation quality is verified based on the consistency between the final adjustment amount of the nodes and the actual curtain wall posture. After verification, the node-level digital construction records are used as the basic data for subsequent operation and maintenance. This data is used to locate potential loose nodes, analyze local deformation trends, and guide structural reinforcement strategies, thereby achieving long-term and accurate management of irregular curtain walls.

[0063] S1 specifically includes the following sub-steps:

[0064] S110, Sub-step for collecting node construction information: Collect the construction information of adjustable connection nodes used for installing curtain wall panels. The construction information includes at least:

[0065] (1) Structural parameters for a reconfigurable adjustment mechanism to achieve multi-degree-of-freedom adjustment, wherein the reconfigurable adjustment mechanism may adopt any of the following structural examples:

[0066] The slide matrix adjustment mechanism includes cross slides arranged along the X / Y direction and position locking holes for realizing translation adjustment;

[0067] A spherical universal adjustment mechanism, comprising a spherical shell and a hemispherical boss, is used to achieve pitch and rotation adjustment;

[0068] An eccentric sleeve-type adjustment mechanism, including an eccentric hole and an adjustment knob, is used to achieve fine position adjustment;

[0069] (2) The degree of freedom of adjustment and its adjustable range in conjunction with the reconfigurable adjustment mechanism, including at least three translational adjustment degrees of freedom in spatial directions and two rotational adjustment degrees of freedom;

[0070] (3) Basic parameters of the micro-sensing component used to record the adjustment amount of the node. The micro-sensing component can be any one of a magnetic encoder, Hall displacement sensor, or micro-resistive displacement sensor, used to collect displacement or angle of the adjustment degree of freedom in real time. Through the above collection, the minimum construction data required for subsequent modeling can be clearly realized.

[0071] It should be noted that the miniature sensing components (such as magnetic scales or Hall elements) are embedded in a package. A sealing groove is pre-milled on the side of the stationary base of the reconfigurable adjustment mechanism, and the sensor read head is fixed within this groove; a magnetic scale is attached to the side of the corresponding moving slider. The groove is covered with a dustproof cover made of stainless steel or engineering plastic to prevent concrete mortar or metal dust from interfering with the sensor readings.

[0072] S120, Node Digital Feature Definition Sub-step: Based on the collected node construction information, define the digital features of each adjustable connection node in the BIM model. The digital features shall include at least:

[0073] (1) The geometric features of the nodes are constructed using parameters such as three-dimensional coordinate points, local coordinate system, and connection surface normal;

[0074] (2) Node adjustment characteristics, including adjustment degree of freedom labels (e.g. , , , , ), Adjust the direction vector and its adjustable range;

[0075] (3) Node record features, including the sampling resolution, sampling range and record field name of the micro-sensing components; enabling technicians to directly generate adjustable node parameter models in the BIM environment.

[0076] S130, Node Unique Identifier Binding Sub-step: Generate a unique identifier (e.g., UUID format number) for each adjustable connection node, and bind the unique identifier to the node's geometric features, adjustment features, and record features in the BIM model to make the node's indexing method in the model stable and traceable.

[0077] S140, Structured Storage Sub-step for Node Adjustment Parameters: Store the adjustment degrees of freedom, adjustment direction vector, and adjustable range of each adjustable connection node in the BIM model as structured fields; the structured fields can be:

[0078] IFC property set (IfcPropertySet) format, or

[0079] The JSON field structure ({"Tx_range":[-10,10],"Ty_range":[-5,5],...}) ensures that the node adjustment parameters have a consistent data organization method, which can be read, modified or recorded in subsequent steps.

[0080] S150, Node Digital Construction Model Generation Sub-step: Based on the node digital features defined and stored in S120 and S140, generate a digital construction model of adjustable connected nodes, wherein the model includes at least:

[0081] (1) Local three-dimensional geometric representation of nodes;

[0082] (2) Parameterized expression of node adjustment degrees of freedom and adjustable range;

[0083] (3) Adjustment input interface field corresponding to the micro-sensing component;

[0084] (4) Node reference points used for subsequent installation datum determination (such as installation datum point, rotation center point, assembly interface surface).

[0085] The digital construction model can be directly used for subsequent installation benchmark derivation, adjustment recording, and BIM reverse writing, enabling adjustable connection nodes to have visual, calculable, and traceable construction capabilities in a virtual environment.

[0086] S2 specifically includes the following sub-steps:

[0087] S210, Sub-step for scanning point cloud acquisition: After the main structure construction is completed, use a 3D scanning device to acquire structural point cloud data of the node installation area;

[0088] The 3D scanning device can be a laser scanner or a structured light scanner, outputting 3D coordinate points. Point cloud dataset ;

[0089] The point cloud data covers the main structural surfaces and the layout area of ​​adjustable connection nodes in the curtain wall installation area.

[0090] The collected point cloud data is used to represent the spatial morphology of the actual constructed structure, providing basic data for subsequent registration and deviation calculation.

[0091] S220, Point Cloud and BIM Model Registration Sub-step: Match the collected point cloud data P with the target structure surface point set generated from the BIM model. Perform spatial registration;

[0092] The registration process includes at least the following steps:

[0093] (1) Feature point pre-alignment: Extract matching feature points from the point cloud P and the model point set Q, such as structural edges, corner points or embedded part locations, and use the feature point pairs to calculate the initial transformation matrix. .

[0094] (2) Iterative Closest Point (ICP) Fine Registration: As initial values, the transformation matrix is ​​solved iteratively using the ICP algorithm. (in For rotation matrix, (where the translation vector is used) makes the objective function

[0095]

[0096] To achieve convergence, where For point cloud points The corresponding nearest neighbor model point.

[0097] (3) Registration result output: Finally, the registration matrix T used to unify the coordinate system is obtained, so that the actual structure represented by the point cloud and the design structure expressed by the BIM model are established in the same coordinate system. Through the above registration, the subsequent deviation field calculation and node installation benchmark derivation are feasible.

[0098] S230, Sub-step for deducing the installation location of the node target: Based on the registered point cloud data and BIM model, calculate the deviation between the actual structure and the designed structure to form the structural deviation field D.

[0099] The calculation process of the deviation field includes:

[0100] (1) Install reference points for each node in the BIM model Find the nearest point in a point cloud ;

[0101] (2) Calculate the deviation vector: This indicates a local offset of the actual structure relative to the design position;

[0102] (3) Combine all deviation vectors A deviation field D is formed to reflect the deformation trend of the structural surface globally or locally; based on the deviation field, the target installation position is derived for each adjustable connection node. The target installation location is determined by the following formula:

[0103]

[0104] This allows the target installation location of the node to directly adapt to the actual structural form of the area.

[0105] S240, Node Target Installation Attitude Determination Sub-step: After determining the target installation position, analyze the actual attitude of the structural surface based on the local normal offset around the position in the deviation field, thereby deriving the target installation attitude of the adjustable connection node.

[0106] The installation posture includes at least the following:

[0107] (1) Connecting surface normal vector ;

[0108] (2) Corrected normal obtained from local gradient calculation of the deviation field ;

[0109] (3) The rotation angle compensation of the local patch is obtained based on the three-point method, and the target rotation angle of the node (such as the rotation amount around Rx, Ry, Rz) is obtained.

[0110] The final set of node target attitude parameters is formed:

[0111]

[0112] Used to guide attitude adjustment during node installation.

[0113] S250, Node Installation Baseline Write into BIM Model Sub-step: Enter the target installation location obtained in S230. The target installation attitude obtained from S240 Write the attribute fields of the corresponding node in the BIM model;

[0114] The fields can be stored in the form of an IFC property set (such as IfcPropertySet) or a JSON structure (such as {"position":[x,y,z],"normal":[nx,ny,nz],"rotation":[rx,ry,rz]}), where x, y, and z correspond to the target installation position of the node. The coordinate components rx, ry, rz correspond to the installation attitude of the target node. Rotational angular components ).

[0115] The written BIM model also includes:

[0116] (1) Unique identifier for the node;

[0117] (2) Node target installation position; (3) Node target installation attitude; (4) Reference direction and deviation compensation amount required for node adjustment.

[0118] This write operation creates installation reference information for the adjustable connection node, enabling subsequent node pre-positioning steps (S310–S350) to be completed under a unified digital coordinate system.

[0119] S3 specifically includes the following sub-steps:

[0120] S310, Node on-site layout sub-step: Based on the target installation position and target installation posture of the adjustable connection node written in the BIM model, the node is laid out to the corresponding on-site structural surface and initially fixed with connectors so that the node is in an adjustable state.

[0121] During the deployment process, ensure that the local coordinate system of the nodes corresponds to the coordinate system of the on-site structure to provide a unified spatial reference for subsequent pre-positioning adjustments.

[0122] S320, Node Pre-positioning Adjustment Sub-step: Pre-positioning adjustment of the adjustable connection node is performed using a reconfigurable adjustment mechanism along its adjustment degrees of freedom; the pre-positioning process refers to the following pre-positioning criteria:

[0123] (1) Basis for determining position error: based on the target installation position With the current structural point of the node Position error

[0124]

[0125] When position error Less than the pre-positioned threshold When the position is 3–5 mm (e.g., 3–5 mm), the pre-positioning requirement is met.

[0126] (2) Attitude error criterion: based on the target installation attitude. With the current pose of the node Calculate the attitude error vector based on the deviation:

[0127]

[0128] in These are the rotational angle deviation components of the node around the X, Y, and Z axes, respectively.

[0129] When all components of the attitude error are less than the attitude threshold When the angle is 1–2°, the attitude prepositioning is determined to meet the requirements. During the prepositioning adjustment process, the node is gradually brought closer to the above prepositioning criteria by adjusting the reconfigurable adjustment mechanism (such as a slide, spherical universal joint or eccentric sleeve).

[0130] S330, Real-time monitoring sub-step of pre-positioning parameters: During pre-positioning adjustment, miniature sensing components deployed inside the adjustable connection node collect the real-time adjustment amounts of the node in each degree of freedom of adjustment; the sensing components include at least one of the following types:

[0131] Magnetic encoder: Outputs angular displacement sample values; the sampling model is:

[0132]

[0133] in The measurement disturbance caused by noise, where This represents the true value of the actual angular displacement of the node.

[0134] Resistive displacement sensor: Outputs linear displacement sample values; sampling model is:

[0135]

[0136] in This represents the true value of the actual linear displacement of the node. Noise in linear displacement measurements;

[0137] Hall linear sensor: Outputs multi-degree-of-freedom displacement changes, which can be used to determine displacement trends. The above sampled values ​​are controlled by the node control unit at a fixed sampling period. (e.g., 20–50ms) are obtained to form a real-time parameter sequence for the node pre-positioning stage.

[0138] It should be added that, considering the large number of nodes distributed inside the curtain wall, the adjustable connection nodes do not have permanent power supplies. The nodes are equipped with passive near-field communication (NFC) interfaces or spring-loaded contact interfaces. During the S3 pre-positioning and S4 fine-tuning processes, the construction personnel use a handheld intelligent adjustment terminal (with its own battery) to approach or contact the node interface, using inductive coupling or physical contact to provide short-term power to the microcontroller unit (MCU) and sensors within the node and complete data reading. After adjustment, the node is powered off, and the data is latched in non-volatile memory (such as electrically erasable programmable read-only memory (EEPROM)).

[0139] S340, Pre-positioning parameter temporary storage sub-step: Temporarily store the real-time sampling data obtained in S330 according to the adjustment degrees of freedom, including:

[0140] (1) Real-time displacement of each translational degree of freedom ( , , );

[0141] (2) Real-time angular displacement of each rotational degree of freedom ( , );

[0142] (3) The sampled value when the prepositioning criterion was most recently met is used as a reference value in the subsequent fine-tuning stage.

[0143] The temporary data can be stored in the node's EEPROM or temporarily cached on the mobile terminal to ensure that the corresponding parameters can be called in subsequent fine-tuning steps.

[0144] S350, Node Pre-positioning Data Structure Generation Sub-step: Based on the pre-positioning parameters of each adjustment degree of freedom temporarily stored in S340, generate a node-level pre-positioning data structure; the structure can adopt JSON or IFC attribute set format, for example:

[0145] {

[0146] "NodeID":"UUID-xxxx",

[0147] "PrePos":[Tx_pre,Ty_pre,Tz_pre],

[0148] "PreAtt":[Rx_pre,Ry_pre],

[0149] "ErrorPos":e_pos,

[0150] "ErrorAtt":[Δθx,Δθy,Δθz]

[0151] }

[0152] Wherein: PrePos represents the displacement adjustment amount after node pre-positioning; its values ​​Tx_pre, Ty_pre, and Tz_pre correspond to the displacement components in the X, Y, and Z directions, respectively; PreAtt represents the angle adjustment amount after node pre-positioning; ErrorPos and ErrorAtt represent the residual error when the pre-positioning criterion converges, used as a reference for subsequent fine-tuning. The structured node pre-positioning data serves as the input for the next step of node fine-tuning in S410–S450, ensuring the complete traceability of the node adjustment process.

[0153] S4 specifically includes the following sub-steps:

[0154] S410, Curtain Wall Panel Assembly and Positioning Sub-step: Based on the pre-positioned state of the nodes, install the corresponding curtain wall panels to the assembly interface of the adjustable connection nodes, so that the panels are in a multi-degree-of-freedom adjustable state provided by the reconfigurable adjustment mechanism; during the assembly process, make the panels initially aligned with the assembly reference surface in the BIM model, so that a stable assembly relationship is established between the panels and the nodes, providing initial boundary conditions for subsequent fine-tuning steps.

[0155] S420, Node fine-tuning execution sub-step: After the panel is assembled in place, the reconfigurable adjustment mechanism is used along its adjustment degrees of freedom ( , , , , Fine-tuning is performed to ensure the panels achieve the target assembly accuracy. Node fine-tuning is adjusted based on the position error function and the attitude error function.

[0156] (1) Position error function: Let the current center point of the plate be ck, and the target installation position be vk′, then the position error is:

[0157]

[0158] when When the value approaches zero, it indicates that the plate position has reached the target requirement.

[0159] (2) Attitude error function: Let the current normal vector of the plate be... The target normal is The included angle error is:

[0160]

[0161] when When the value approaches zero, it indicates that the plate's posture is consistent with its target.

[0162] (3) Comprehensive error vector: The goal of node fine-tuning is to bring the comprehensive error vector E to a fine-tuning threshold.

[0163]

[0164] in It can be 1–2 mm. It can be 0.5–1°.

[0165] S430, Fine-tuning calculation and execution sub-step: In order to converge the error, the adjustment amount to be executed is solved by the mapping relationship between the node adjustment degree of freedom and the plate pose.

[0166] Adjustment Calculation Model (Jacobi Linear Approximation Model): Let the nodal adjustment degree of freedom vector be: Let the change in error be: The two approximately satisfy a linear mapping: ;

[0167] Where J is the Jacobian matrix between node adjustment and pose change, which can be determined by simulation before initial calibration or construction.

[0168] Least squares solution for adjustment:

[0169]

[0170] This solution process yields the precise adjustment amount required for each degree of freedom of adjustment, where the superscript... -1 indicates matrix transpose, and -1 indicates matrix inversion.

[0171] Execute adjustment: The solution obtained The input is fed into a reconfigurable adjustment mechanism (slide, spherical joint, eccentric sleeve, etc.), and the corresponding adjustment is performed by construction personnel or mechanical assistance to make the plate after node fine adjustment approach the target posture.

[0172] In detail: Taking an adjustment mechanism using a combination of a sliding groove and a ball joint as an example, a local coordinate system for the nodes is established. Let the translational displacement provided by the sliding groove be... The amount of rotation provided by the ball joint is Establish the geometric center of the plate. Regarding the kinematic forward function F of the moderating variable, i.e. In step S430, the system constructs a Jacobian matrix J specific to the node configuration by taking the partial derivatives of the kinematic equations with respect to each adjustment variable. For nodes with different configurations (as defined in S1), the BIM model stores corresponding kinematic equation templates.

[0173] S440, Fine-tuning Completion Status Monitoring Sub-step: During the fine-tuning process, the micro-sensing components collect the displacement and angle measurements in real time after fine-tuning, forming a sampling sequence:

[0174] Displacement sample value: ;

[0175] Angular displacement sampled values: ;

[0176] The sensor data is denoised after a one-time calibration (such as zero-point calibration or linear fitting calibration).

[0177]

[0178] in , This is the static deviation value of the system obtained through calibration.

[0179] When the calibrated real-time error vector satisfies:

[0180]

[0181] This indicates that the fine-tuning is complete.

[0182] S450, Structured Adjustment Data Generation Sub-step: After fine-tuning, the final adjustment amount, adjustment direction vector, and locking state of each adjustment degree of freedom are latched, and structured adjustment data is generated based on the unique node identifier; the structured adjustment data may adopt the following format:

[0183] {

[0184] "NodeID":"UUID-xxxx",

[0185] "AdjPos":[Tx_fin,Ty_fin,Tz_fin],

[0186] "AdjAtt":[Rx_fin,Ry_fin],

[0187] "CalibData":{"d_cal":...,"theta_cal":...},

[0188] "ErrResidual":{"Epos":...,"Eatt":...},

[0189] "LockState":"Locked"

[0190] }

[0191] The structured adjustment data will be written into the BIM model in subsequent steps S510–S550, serving as the core content of the digital construction record.

[0192] S5 specifically includes the following sub-steps:

[0193] S510, Adjustment Data Reading Sub-step: After node fine-tuning is completed, the structured adjustment data generated by each adjustable connection node is read through the field mobile terminal or node communication module; the structured adjustment data includes:

[0194] The final node position adjustment (AdjPos=[Tx_fin, Ty_fin, Tz_fin]);

[0195] The final attitude adjustment of the node (AdjAtt=[Rx_fin, Ry_fin]);

[0196] Sensor calibration data (CalibData);

[0197] Residual error (ErrResidual);

[0198] Node lock state (LockState);

[0199] Unique identifier for a node (NodeID).

[0200] The read data is cached on the mobile terminal in the form of timestamps to ensure the time sequence consistency of subsequent BIM reverse writing steps.

[0201] S520, Adjusting the data field parsing sub-step:

[0202] The structured adjustment data read from S510 is parsed according to a preset field format; the field format includes at least the following implementable structures:

[0203] (1) Example of JSON format parsing:

[0204] {

[0205] "NodeID":"UUID-xxxx",

[0206] "AdjPos":[Tx_fin,Ty_fin,Tz_fin],

[0207] "AdjAtt":[Rx_fin,Ry_fin],

[0208] "CalibData":{"d_cal":x1,"theta_cal":x2},

[0209] "ErrResidual":{"Epos":y1,"Eatt":y2},

[0210] "LockState":"Locked",

[0211] "Timestamp":"2025-12-05T10:00:00"

[0212] }

[0213] (2) Example of IFC attribute set parsing:

[0214] IfcPropertySingleValue("Tx_fin",IfcLengthMeasure)

[0215] IfcPropertySingleValue("Rx_fin",IfcPlaneAngleMeasure)

[0216] IfcPropertySingleValue("LockState",IfcText)

[0217] The parsing process identifies the above fields and converts them into parameter values ​​that BIM can recognize, preparing for subsequent binding of node attributes.

[0218] S530, Node Adjustment Data and Unique Identifier Binding Sub-step: Based on the NodeID in the adjustment data, retrieve the corresponding adjustable connection node object in the BIM model; bind the parsed adjustment data to this node object, with binding methods including at least:

[0219] Attribute set binding: Add an attribute set (such as Pset_AdjustRecord) under the node entity object (such as IfcBuildingElementProxy) of the BIM model, and write the parsed adjustment field into the attribute set.

[0220] Field bitwise binding rules:

[0221] "AdjPos" → Node position adjustment attribute (e.g., Pset_AdjustRecord.Tx_fin);

[0222] "AdjAtt" → Node pose adjustment attribute (e.g., Pset_AdjustRecord.Rx_fin);

[0223] “ErrResidual” → Installation error attribute;

[0224] "LockState" → Node state attribute;

[0225] "Timestamp" → Data synchronization time attribute.

[0226] Consistency check: Ensure that the bound data meets the following conditions:

[0227]

[0228] in The design range for the maximum allowable linear displacement of the node. Design range for the maximum allowable angular displacement of the node; This represents the Euclidean norm / modulus of the vector; if the validation fails, the write operation is rejected and an exception is displayed. This binding operation establishes a one-to-one mapping between the adjustment data and the node objects in the BIM model, enabling traceability of the data source and node identity.

[0229] S540, Sub-step for writing adjustment data to the BIM model: Synchronously write the bound adjustment data to the BIM model, so that each node entity object in the model has the actual installation status attribute. The writing method includes any of the following:

[0230] (1) Direct writing to local model: The attributes are written to the local IFC or RVT file by the mobile terminal through IFCAPI or BIM software plugin (such as RevitAPI).

[0231] (2) Real-time writing of cloud model: The mobile terminal submits the adjustment data to the BIM cloud platform through the network, and the server updates the attributes of the IFC model.

[0232] (3) The content to be written includes: final position adjustment amount; final attitude adjustment amount; calibration sensor data; error residual; node locking status; original acquisition timestamp.

[0233] (4) Ensure write consistency: Perform a data callback check before and after writing to ensure that the stored value is consistent with the written data and eliminate transmission errors.

[0234] This step upgrades the BIM model from a "design model" to a "construction reality model," giving it real-world node status.

[0235] In detail: To address the issue of unstable network signals at construction sites, the S5 process includes an offline caching mechanism. If the mobile terminal on site cannot connect to the BIM server in real time, the adjusted data will be automatically timestamped and encrypted and stored in the terminal's local database. Once the terminal moves to a network coverage area or connects to the office network, the backend service will automatically trigger batch data uploads and synchronized writing with the BIM model, ensuring eventual data consistency.

[0236] S550, Digital Construction Record Generation Sub-step: After the adjustment data of all nodes is written into the BIM model, a digital construction record is generated for subsequent quality tracking and operation and maintenance support.

[0237] The digital construction record includes:

[0238] (1) Node-level construction records

[0239] Record the following for each node: final value of adjustment degree of freedom; adjustment amount change curve (generated from sampling sequence); error convergence status; locking status; sampling timestamp; construction responsibility unit information.

[0240] (2) Model-level adjustment distribution map

[0241] Automatically generate color-coded maps in BIM to represent the spatial distribution of node adjustment amounts, which is used to identify areas of concentrated deviation.

[0242] (3) Traceable data chain: The data chain structure is as follows: NodeID → Adjustment data → BIM attribute set → As-built comparison data; ensuring that any subsequent operation and maintenance or inspection can be traced back to the actual construction status.

[0243] (4) Data export format: It can be exported as IFC, Excel, JSON or database format for third-party audit or supervision. This digital construction record serves as the core input for subsequent steps S610–S650 (completion comparison and operation and maintenance diagnosis), realizing digital management of the entire life cycle of the curtain wall.

[0244] S6 specifically includes the following sub-steps:

[0245] S610, As-built Scanning Data Acquisition Sub-step: After all curtain wall panels are installed, use the same or compatible 3D scanning equipment as the previous scan to acquire as-built point cloud data of the curtain wall area; the acquired as-built point cloud is represented as follows:

[0246]

[0247] The data collection coverage should include all adjustable connection node locations, plate edges, connection surface areas, and locations prone to local deformation to ensure the integrity of the as-built comparison. The collected as-built scan data will serve as the basis for subsequent quality verification and node loosening identification.

[0248] S620, Sub-step for comparing point cloud and BIM model as-built projects:

[0249] (1) Reuse of the design registration matrix The point cloud registration matrix determined before construction. Application in as-built point cloud:

[0250]

[0251] Align the as-built point cloud with the BIM model in the same coordinate system.

[0252] (2) Secondary fine registration (ICP refinement): In order to compensate for the slight sensing errors that may exist at the construction site, the ICP algorithm is used to refine the registration. Perform a second-order matching with the BIM model surface point set Q, and solve for the fine-tuning matrix. :

[0253]

[0254] Fine-tuning matrix The rotation amount is usually less than 0.5°, and the displacement amount is usually less than 2mm.

[0255] (3) As-built deviation field calculation: Installation reference points for nodes in the BIM model Points at corresponding locations in the as-built scan point cloud Perform deviation calculation:

[0256]

[0257] The completion deviation field is represented as follows:

[0258]

[0259] Deviation fields are used not only to verify quality, but also to determine whether there are loose nodes or localized plate deformation.

[0260] S630, Installation Quality Verification Sub-step: Perform the following verification on the as-built deviation site:

[0261] Position accuracy verification: Calculate the deviation of the completed position.

[0262]

[0263] If the following conditions are met: If the positional accuracy is satisfactory, then the positional accuracy is acceptable.

[0264] Attitude accuracy verification: Compare as-built normals With the target normal The included angle:

[0265]

[0266] If the following conditions are met: If the posture accuracy is satisfactory, then the posture accuracy is acceptable.

[0267] Consistency verification of adjustment status: The consistency of the node pose changes calculated from the as-built point cloud is compared with the node adjustment records (AdjPos, AdjAtt) in the BIM:

[0268]

[0269] when:

[0270]

[0271] If the adjustment status is consistent, it indicates that the node has not shifted or become loose. Those that do not meet the requirements proceed to the next step, S640, for loosening identification. The threshold for determining the consistency of node adjustment amounts; This is the current pose vector of the node calculated from the as-built point cloud. This is the final adjustment vector of the nodes recorded in the BIM model.

[0272] S640, Node Status Anomaly Location Sub-step: Compare the as-built deviation field with the adjustment records in the BIM model to identify potential loose nodes or deformation points.

[0273] Adjustment offset criterion (core criterion): If it exists: This indicates that the regulating mechanism has loosened in the reverse direction or that the plate has been displaced due to stress.

[0274] Criterion for the amplification trend of attitude residuals: If: This typically indicates insufficient node attitude maintenance or reduced connection stiffness. This refers to the final attitude residual of the nodes recorded in the BIM model during installation. This is the allowed attitude drift threshold.

[0275] Local gradient anomaly criterion (deformation identification): Calculate the local gradient around the node in the deviation field:

[0276]

[0277] like: This indicates that the area may experience plate deformation or local structural irregularities. Let $\mathbf{k+1}$ be the completion deviation vector of the (k+1)th node adjacent to node $k$. This is the threshold for local gradient anomalies.

[0278] Comprehensive loosening indicator function (innovation point):

[0279]

[0280] in , , These are the weighting coefficients.

[0281] when: If node k is marked as a "suspected loose node", the marked node will be automatically pushed to the operation and maintenance system and will be a key target for inspection. The threshold for determining overall loosening.

[0282] S650, Operation and Maintenance Decision Support Sub-step: Based on the digital construction record generated by S550 and the as-built comparison data obtained by S620–S640, support is provided for subsequent operation and maintenance.

[0283] Node health status label: Generate a health level for each node:

[0284] Level 0: No deviation, normal status;

[0285] Level 1: Minor displacement, re-inspection recommended;

[0286] Level 2: Significant offset, tightening is recommended;

[0287] Level 3: Severe offset, requiring immediate repair or replacement of nodes; health markers are written into the BIM model to form equipment-level "digital health records".

[0288] Deformation trend model construction: Based on the deviation field sequence obtained from each scan Constructing a deformation trend model:

[0289]

[0290] This is used to infer potential future plate warping, node loosening, or thermal deformation paths, providing a basis for long-term monitoring. It is the exponential smoothing coefficient (usually ranging from 0 to 1), used to adjust the weight of the current measurement value and the historical trend value.

[0291] Recommended operation and maintenance strategy: If a node or region shows an increasing trend of deviation, the system will automatically generate:

[0292] Adjustment recommendations; tightening procedures; panel unloading and reinstallation recommendations; risk warnings under wind load conditions; and can be exported as text, reports, or visual analysis charts.

[0293] like Figure 2 and Figure 3 As shown, in step S1, a parametric model of the irregularly shaped curtain wall is established using BIM software. The model accurately represents the complex curvature characteristics of the curtain wall as it changes with the main structure. The grid shown in the figure represents the independent curtain wall panels after parametric subdivision, with each grid unit corresponding to an actual glass or metal panel.

[0294] During the modeling process, the system automatically extracts the spatial coordinates of each edge node of the plate (i.e., the intersection of the grid shown in the figure) based on the surface shape illustrated, thereby determining the target installation position of the adjustable connection node. The area highlighted in the lower left corner of the figure indicates the specific construction unit from which the parameters are to be extracted. The system will read the four corner coordinates and normal vector of this unit as the reference data for fine-tuning in the subsequent S4 step.

[0295] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0296] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0297] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall, characterized in that, Includes the following steps: S1. Digitally model the adjustable connection nodes used for installing curtain wall panels; during the digital modeling process, establish a unique identifier for each adjustable connection node, and structurally store the adjustment freedom, adjustable range and recorded parameters of the connection node in the BIM model to form node-level digital construction information. S2. After the main structure construction is completed, obtain the three-dimensional scanning point cloud data of the area to be installed, and spatially register the point cloud data with the BIM model; Determine the target installation baseline for each adjustable connection node, write the target installation baseline into the BIM model, and form node-level installation guidance information; S3. At the installation site, each adjustable connection node is deployed, and the position and attitude of the node are pre-positioned using a reconfigurable adjustment mechanism. The micro-sensor components monitor and temporarily store the pre-positioning parameters of the node in each degree of freedom of adjustment in real time, forming node-level pre-positioning data. S4. Install the corresponding curtain wall panels onto the adjustable connection nodes, and use the reconfigurable adjustment mechanism to perform fine adjustments along each degree of freedom to make the curtain wall panels meet the predetermined assembly accuracy requirements; the micro-sensor components automatically collect and lock the final adjustment amount of the nodes in each degree of freedom to generate structured adjustment data. S5. Write the structured adjustment data into the BIM model in real time through the on-site mobile terminal or communication module, so that the final adjustment status of each adjustable connection node forms a traceable installation record in the BIM model. During the reverse writing process, the node installation status, locking status, and adjustment direction vector are used as associated attributes and bound to the corresponding node's unique identifier to form a node-level digital construction record.

2. The BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall according to claim 1, characterized in that, Also includes: S6. After all curtain wall panels are installed, the installation quality is checked by collecting 3D scanning data again and aligning it with the BIM model after it has been reverse-written, based on the consistency between the final adjustment amount of the nodes and the actual posture of the curtain wall. The node-level digital construction records will be used as the basis for subsequent operation and maintenance, to locate potential loose nodes, analyze local deformation trends, and guide structural reinforcement strategies.

3. The BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall according to claim 1, characterized in that, S1 specifically includes: Collect structural information on adjustable connection nodes used for installing curtain wall panels, including the degree of freedom of adjustment of the reconfigurable adjustment mechanism, the adjustable range, and the type of recorded parameters of the micro-sensor components; Based on the collected construction information, the digital features of adjustable connection nodes are defined in the BIM model, including the geometric features, adjustment features and recording features of the nodes, and a node-level digital description is formed. A unique identifier is generated for each adjustable connection node, and the unique identifier is bound to the node's digital features, so that the node has stable identification attributes in the BIM model.

4. The method for precise construction of irregularly shaped curtain walls based on BIM and 3D scanning according to claim 3, characterized in that, S1 also includes: The adjustment degrees of freedom, adjustment direction, and adjustable range of the reconfigurable adjustment mechanism are stored in the BIM model in the form of structured fields, so that the adjustment parameters have traceability and recallability. Based on stored structured fields, a digital construction model of adjustable connection nodes is generated in the BIM model, providing standardized node data for subsequent installation benchmark determination.

5. The BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall according to claim 1, characterized in that, S2 specifically includes: After the main structure is completed, the structural point cloud data of the node installation area is obtained using a 3D scanning device to form a scanning dataset containing the actual spatial morphology. The collected scanned point cloud is spatially registered with the BIM model to obtain the structural deviation field, providing a spatial reference for determining the node installation benchmark. Based on the registered deviation field, the target installation position of each adjustable connection node is derived so that the node position can adapt to the actual structural form. Based on the design and assembly relationship of the curtain wall panels, determine the target installation posture of each adjustable connection node, including the rotation angle and normal orientation; Write the target installation location and target installation posture of the node into the BIM model, so that the installation benchmark can form a visualized and callable node-level guidance information.

6. The BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall according to claim 1, characterized in that, S3 specifically includes: Based on the node installation benchmark in the BIM model, adjustable connection nodes are installed at the corresponding locations on the installation site to ensure that the node placement is consistent with the design benchmark. The adjustable connection node is prepositioned along its degree of freedom using a reconfigurable adjustment mechanism so that the node position and attitude are close to the target installation reference. During the pre-positioning process, the micro-sensing components monitor the pre-positioning parameters of the node in each degree of freedom of adjustment in real time, forming real-time pre-positioning data; The pre-positioning parameters obtained from real-time monitoring are temporarily stored in the node's internal storage module or mobile terminal, enabling the pre-positioning parameters to have subsequent adjustment and comparison functions. Node-level prepositioning data is generated based on the temporarily stored prepositioning parameters, enabling the prepositioning status of nodes to participate in subsequent fine-tuning judgments in a structured form.

7. The method for precise construction of irregularly shaped curtain walls based on BIM and 3D scanning according to claim 1, characterized in that, S4 specifically includes: Install the corresponding curtain wall panel to the assembly interface of the adjustable connection node, so that the panel is in an adjustable state and establishes a stable connection with the node; The reconfigurable adjustment mechanism is used to perform fine adjustments along each degree of freedom to ensure that the curtain wall panels meet the assembly accuracy requirements, including positional accuracy and attitude accuracy. Upon completion of fine-tuning, the micro-sensing components automatically monitor the final adjustment amount of the node in each degree of freedom of adjustment to obtain the final adjustment state of the node. The final adjustment amount is latched and a node-level adjustment amount record is formed using the node's unique identifier; The latched adjustment values ​​are organized into structured adjustment data to provide a standardized data format for subsequent writing into the BIM model.

8. The BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall according to claim 1, characterized in that, S5 specifically includes: Structured adjustment data can be read through mobile terminals or communication components, enabling the adjustment data to be uploaded remotely. The structured adjustment data is parsed to ensure its completeness, including the final adjustment amount, adjustment direction vector, and locking status. The parsed adjustment data is bound to the unique identifier of the node, so that the adjustment data and the node identity are accurately matched; Write the bound adjustment data into the BIM model so that the final installation status of the node forms a traceable installation record in the BIM model; Based on the reverse-written adjustment data, node-level digital construction records are generated, providing a data foundation for quality tracking and subsequent maintenance.

9. The BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall according to claim 2, characterized in that, S6 specifically includes: After all the curtain wall panels are installed, a three-dimensional point cloud scan of the curtain wall area is collected again to obtain the as-built scan data; Spatially align the as-built scan data with the BIM model containing node adjustment data to form an as-built verification dataset. Based on the completion verification data, the consistency between the final adjustment amount of the nodes and the actual curtain wall posture is analyzed, and it is determined whether the installation quality meets the requirements.

10. The BIM and three-dimensional scanning-based precise construction method for a special-shaped curtain wall according to claim 9, characterized in that, S6 also includes: If the review results show local deviations, potential loose or abnormal nodes will be located based on the difference between the adjustment data in the BIM model and the as-built posture. Based on digital construction records and as-built comparison data, it provides support for subsequent operation and maintenance, including node reinforcement strategies, displacement trend analysis, and safety assessment, thereby achieving long-term precise management.

Images