Virtual pre-assembly method and system for curtain wall prefabricated units

Abstract

The invention discloses a virtual pre-assembling method and system for a curtain wall prefabricated unit, particularly relates to the technical field of building information models and digital twinning, and is used for solving the problems that a virtual environment and physical reality are disjointed and collision detection is distorted due to the fact that an existing virtual pre-assembling method is based on an ideal geometric model. The method comprises the steps of generating a virtual model of each curtain wall prefabricated unit by obtaining design models and manufacturing data of a plurality of curtain wall prefabricated units, performing tolerance accumulation analysis on the plurality of virtual models under an assembly sequence to simulate transmission and accumulation of manufacturing errors in an assembly path through a connection interface, and identifying systematic assembly conflict risks. Evaluating a key tolerance transmission path which has the greatest influence on the overall assembling quality, and dynamically adjusting the assembling parameters and the assembling sequence of the virtual model; and simulating the splicing process of the curtain wall prefabricated unit in the virtual environment to generate a splicing effect picture, and verifying the splicing effect picture to output a verification report to realize accurate mapping of virtual pre-splicing and physical entities.

Description

Technical Field

[0001] This invention relates to the field of building information modeling and digital twin technology, and more specifically, to a virtual pre-assembly method and system for prefabricated curtain wall units. Background Technology

[0002] In the fields of Building Information Modeling (BIM) and digital construction, the virtual pre-assembly of curtain wall prefabricated units is a crucial process for ensuring construction accuracy. Current methods generally simulate assembly by reproducing the design model in a virtual space. This process essentially constructs a static, ideal digital model isolated from the physical entity. While this model can reflect the design intent, it is separated from the actual state of the physical units during manufacturing and assembly, failing to achieve a precise mapping and two-way interaction from the physical world to the digital space; in other words, it has not yet constituted an effective digital twin system.

[0003] Current virtual pre-assembly methods based on ideal geometric models suffer from a fundamental disconnect between the constructed virtual environment and physical reality because the core digital model fails to synchronize and integrate with the actual dimensional tolerances and morphological deviations generated during the manufacturing process. This disconnect manifests specifically in the distortion of collision detection logic: pseudo-compatible states deemed interference-free in the virtual environment may actually cause assembly interference on-site due to the accumulation of positive tolerances in actual components; conversely, pseudo-interference positions marked as conflicting in the model may possess safe physical assembly clearances due to the negative tolerances of actual components. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a virtual pre-assembly method and system for curtain wall prefabrication units to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A virtual pre-assembly method for prefabricated curtain wall units includes the following steps:

[0007] S1. Obtain design models and manufacturing data for multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information.

[0008] S2. Based on the design model and manufacturing data, generate a virtual model for each prefabricated curtain wall unit. The virtual model includes geometry and connection interfaces.

[0009] S3. Perform tolerance accumulation analysis on multiple virtual models under the assembly sequence to simulate the transmission and accumulation of manufacturing errors through the connection interface in the assembly path, so as to identify the risk of systematic assembly conflict.

[0010] S4. Assess the critical tolerance transfer path that has the greatest impact on overall assembly quality in the assessment of systemic assembly conflict risks;

[0011] S5. Based on the key tolerance transfer path, dynamically adjust the assembly parameters and assembly sequence of the virtual model. The assembly parameters include position offset, rotation angle and connection point coordinates.

[0012] S6. Using the adjusted assembly parameters, simulate the assembly process of the prefabricated curtain wall units in a virtual environment to generate an assembly effect diagram, and verify the assembly effect diagram to output a verification report.

[0013] Furthermore, design models and manufacturing data for multiple prefabricated curtain wall units are acquired. The manufacturing data includes actual dimensions and tolerance information, including:

[0014] Point cloud data of the surface of the prefabricated curtain wall unit was collected using an optical 3D scanning device;

[0015] The actual dimensions of the prefabricated curtain wall units are calculated based on point cloud data.

[0016] The actual dimensions are compared with the corresponding theoretical dimensions in the design model to generate tolerance information containing deviation values;

[0017] The actual dimensions and tolerance information are integrated to form manufacturing data.

[0018] Furthermore, based on the design model and manufacturing data, a virtual model of each prefabricated curtain wall unit is generated. The virtual model includes geometry and connection interfaces, including:

[0019] Input the actual dimensions from the manufacturing data into the design model, replacing the corresponding theoretical dimension parameters;

[0020] The geometry of the prefabricated curtain wall units was reconstructed based on the updated design model;

[0021] Update the mating surface geometry data of the connection interface based on the reconstructed geometry;

[0022] The model containing the updated geometry and connection interfaces is defined as a virtual model.

[0023] Furthermore, tolerance accumulation analysis is performed on multiple virtual models under assembly sequences to simulate the transmission and accumulation of manufacturing errors through connection interfaces in the assembly path, in order to identify systematic assembly conflict risks, including:

[0024] Each virtual model is processed sequentially based on a predefined assembly sequence;

[0025] For each virtual model, the process of manufacturing errors being transmitted from the previous virtual model to the current virtual model is simulated through the connection interface;

[0026] Accumulate the error effects of the preceding virtual model into the current virtual model to generate a cumulative error effect;

[0027] The mating status of the connection interface is evaluated based on the cumulative error effect. When the mating status indicates an assembly conflict, it is identified as a systemic assembly conflict risk.

[0028] Furthermore, for each virtual model, the process of simulating the transmission of manufacturing errors from the previous virtual model to the current virtual model through the connection interface includes: calculating the error transmission vector based on the cumulative error effect of the previous virtual model and the tolerance information of the current virtual model through the geometric relationship of the mating surfaces of the connection interface; and applying the error transmission vector to the position offset and rotation angle of the current virtual model to update its spatial attitude.

[0029] Furthermore, the key tolerance transfer paths that have the greatest impact on overall assembly quality in assessing systemic assembly conflict risks include:

[0030] Extract tolerance transfer paths from identified systemic assembly conflict risks;

[0031] Sensitivity analysis was performed on each tolerance transfer path, and the degree of change in overall assembly quality was observed by adjusting the tolerance parameter values ​​of the corresponding paths.

[0032] Based on the sensitivity analysis results, the degree of influence of each tolerance transfer path is ranked.

[0033] The tolerance transfer path with the highest degree of impact is identified as the critical tolerance transfer path.

[0034] Furthermore, sensitivity analysis is performed on each tolerance transfer path, including: for each tolerance transfer path, its tolerance parameter value is modified independently and the assembly sequence simulation is re-executed in the virtual environment; the response gradient of the overall assembly quality index relative to the change in tolerance parameter value is calculated; and the influence of the corresponding tolerance transfer path on the overall assembly quality is quantified based on the response gradient.

[0035] Furthermore, based on the critical tolerance transfer path, the assembly parameters and assembly sequence of the virtual model are dynamically adjusted. The assembly parameters include position offsets, rotation angles, and connection point coordinates, including:

[0036] Identify virtual models involved in key tolerance transfer paths as priority adjustment targets;

[0037] The assembly parameters of the priority adjustment objects are optimized and calculated to generate adjustment schemes for position offset, rotation angle and connection point coordinates;

[0038] Based on the adjustment plan, prioritize the updating of assembly parameters for the adjusted objects;

[0039] The assembly sequence is replanned based on the updated assembly parameters to generate an assembly order that minimizes cumulative error.

[0040] Furthermore, using the adjusted assembly parameters, the assembly process of the prefabricated curtain wall units is simulated in a virtual environment to generate an assembly rendering. The assembly rendering is then verified to output a verification report, including:

[0041] Assemble each virtual model sequentially in the virtual environment according to the adjusted assembly parameters and assembly sequence;

[0042] Real-time detection of interference between virtual models during assembly;

[0043] After assembly is completed, an assembly effect diagram showing the overall assembly status will be generated;

[0044] Perform structural stability verification and visual alignment checks on the assembled renderings;

[0045] A verification report containing an assembly quality assessment is generated based on the verification and inspection results.

[0046] On the other hand, the present invention provides a virtual pre-assembly system for curtain wall prefabrication units, comprising the following modules:

[0047] The data acquisition module is used to acquire design models and manufacturing data of multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information.

[0048] The model generation module is used to generate a virtual model of each prefabricated curtain wall unit based on the design model and manufacturing data. The virtual model includes geometry and connection interfaces.

[0049] The risk identification module is used to perform tolerance accumulation analysis on multiple virtual models under the assembly sequence, and to simulate the transmission and accumulation of manufacturing errors through the connection interface in the assembly path in order to identify systematic assembly conflict risks.

[0050] The path assessment module is used to evaluate the critical tolerance transfer path that has the greatest impact on overall assembly quality in the risk of systemic assembly conflict.

[0051] The dynamic adjustment module is used to dynamically adjust the assembly parameters and assembly sequence of the virtual model according to the key tolerance transfer path. The assembly parameters include position offset, rotation angle and connection point coordinates.

[0052] The virtual simulation module is used to simulate the assembly process of prefabricated curtain wall units in a virtual environment using adjusted assembly parameters to generate assembly effect diagrams, and to verify the assembly effect diagrams to output a verification report.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. By integrating actual dimensions and tolerance information from manufacturing data to construct a virtual model, a precise mapping from physical entities to digital space is achieved, effectively establishing a digital twin system. Through tolerance accumulation analysis, the transmission and accumulation of manufacturing errors in the assembly path are simulated, and systemic assembly conflict risks are dynamically identified. This allows for the realistic reproduction of deviation effects in the physical assembly process within a virtual environment. The virtual model generation and error simulation based on actual data overcome the problem of the disconnect between traditional ideal geometric models and physical reality, significantly improving the realism and reliability of virtual pre-assembly and providing an accurate data foundation for subsequent assembly optimization.

[0055] 2. By evaluating key tolerance transfer paths and dynamically adjusting assembly parameters and sequences, adaptive matching between the virtual model and physical assembly conditions was achieved. This effectively reduced assembly interference or gap anomalies caused by tolerance accumulation. The generation of assembly effect diagrams and the output of verification reports formed a closed-loop verification mechanism, ensuring the feasibility of the assembly scheme in practical applications. This not only improved the accuracy of assembly quality control but also achieved collaborative optimization of manufacturing and assembly processes through digital twin technology, providing an efficient and reliable pre-assembly solution for curtain wall engineering. Attached Figure Description

[0056] Figure 1 This is a flowchart of a virtual pre-assembly method for a prefabricated curtain wall unit according to the present invention;

[0057] Figure 2 This is a schematic diagram of the structure of a virtual pre-assembly system for a curtain wall prefabrication unit according to the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0059] Example 1: Figure 1 The present invention provides a virtual pre-assembly method for prefabricated curtain wall units, which includes the following steps:

[0060] S1. Obtain design models and manufacturing data for multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information.

[0061] S2. Based on the design model and manufacturing data, generate a virtual model for each prefabricated curtain wall unit. The virtual model includes geometry and connection interfaces.

[0062] S3. Perform tolerance accumulation analysis on multiple virtual models under the assembly sequence to simulate the transmission and accumulation of manufacturing errors through the connection interface in the assembly path, so as to identify the risk of systematic assembly conflict.

[0063] S4. Assess the critical tolerance transfer path that has the greatest impact on overall assembly quality in the assessment of systemic assembly conflict risks;

[0064] S5. Based on the key tolerance transfer path, dynamically adjust the assembly parameters and assembly sequence of the virtual model. The assembly parameters include position offset, rotation angle and connection point coordinates.

[0065] S6. Using the adjusted assembly parameters, simulate the assembly process of the prefabricated curtain wall units in a virtual environment to generate an assembly effect diagram, and verify the assembly effect diagram to output a verification report.

[0066] S1. Obtain design models and manufacturing data for multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information. The specific implementation is as follows:

[0067] In the process of acquiring design models and manufacturing data for multiple prefabricated curtain wall units, point cloud data of the surface of the prefabricated curtain wall units is first collected using an optical 3D scanning device. This optical 3D scanning device includes, but is not limited to, laser scanners or structured light scanners. The operation involves aligning the scanning device with the surface of the prefabricated curtain wall unit to be measured, emitting laser or structured light, and receiving reflected signals to acquire the three-dimensional coordinate information of the object's surface in a non-contact manner, thereby generating a dense point cloud dataset. Each point in the point cloud data contains three-dimensional spatial coordinate values, which are recorded in real time and stored as digital files by the scanning device's built-in sensors and data processing unit. The acquisition of point cloud data must ensure coverage... The entire outer surface and connecting areas of the prefabricated curtain wall unit are scanned to avoid data loss. In practice, the scanning resolution is set to, for example, 0.1 mm to ensure that the point cloud density is high enough to accurately reflect the subtle features of the unit surface. During the scanning process, ambient light and vibration interference need to be controlled to improve data quality. After the point cloud data is generated, noise points are removed through data preprocessing steps, such as filtering algorithms to remove outliers. The filtering threshold is set based on the point cloud density statistics. Specifically, the average distance between adjacent points in the point cloud is calculated, and points with a distance greater than twice the average distance are considered noise and removed. The preprocessed point cloud data is saved in a standard format, such as a PLY or XYZ file, for easy use in subsequent calculations.

[0068] When calculating the actual dimensions of prefabricated curtain wall units based on point cloud data, the key geometric features of the prefabricated curtain wall units, such as corner points and planar areas, are first extracted from the point cloud data. The actual dimensions are then calculated using a point cloud fitting algorithm, specifically the least squares method, which fits the point cloud data onto a pre-defined geometric model, such as fitting the point cloud onto a plane or curved surface, to calculate the length, width, and height dimensions. The specific implementation process involves identifying the set of points in the point cloud corresponding to the edges of the curtain wall units, iteratively calculating the distance from the point set to the geometric model, and adjusting the model parameters to minimize the sum of squared distances, thereby obtaining the actual dimension values. The calculation involves input... The input parameters include point cloud coordinates and geometric model type, and the output is the actual size value, such as the length value in millimeters. During the calculation process, a fitting accuracy threshold needs to be set. This threshold is set based on the manufacturing tolerance requirements of the curtain wall unit. For example, the fitting error is controlled within 0.5 mm to ensure dimensional accuracy. In addition, for units with complex shapes, the actual size calculation may include multiple sub-steps. For example, the size of the main plane is calculated first, and then the local size of the connection interface is calculated by projection method. All calculations are automatically executed by computer program. The program code is based on existing libraries such as Open3D or PCL to ensure that the calculation is repeatable and verifiable.

[0069] The process compares the actual dimensions with the corresponding theoretical dimensions in the design model to generate tolerance information containing deviation values. This process first extracts theoretical dimension parameters from the design model, typically a 3D model in BIM software. Theoretical dimensions include, for example, design length, design angles, and design coordinate values. The comparison is achieved by calculating the difference between the actual and theoretical dimensions. The deviation value is defined as the actual dimension minus the theoretical dimension, and the result is expressed numerically. For example, length deviation is represented by positive or negative values, indicating whether the actual dimension is too large or too small relative to the theoretical dimension. The generation of tolerance information includes setting an allowable deviation range, i.e., a tolerance zone, which is based on industry standards. As per the building curtain wall specifications, for example, for length dimensions, the tolerance zone is set to ±2 mm. During the comparison process, when the deviation value exceeds the tolerance zone, it is marked as out of tolerance, and the location and magnitude of the out-of-tolerance are recorded. When generating tolerance information, it is necessary to ensure that the comparison of each dimension corresponds one-to-one. For example, the dimension number in the design model is associated with the index in the actual dimension calculation by using an identifier. The tolerance information is finally stored as structured data, such as tables or JSON format, which includes deviation values, tolerance zones, and out-of-tolerance status. This process is implemented through scripts or software tools, such as using a Python script to read the design model and actual dimension data, perform the comparison, and output the tolerance information file.

[0070] The manufacturing data is formed by integrating actual dimensions and tolerance information. The integration process includes data fusion and format standardization. First, the actual dimension data and tolerance information data are associated by curtain wall unit number. For example, the actual dimension value and the corresponding tolerance information are merged into a record by using a unique unit identifier. The format of the manufacturing data is defined to include actual dimension field, tolerance information field, and metadata field. Metadata includes, for example, unit type, measurement time, and operator information. The integration operation is implemented through a database management system or custom program. For example, an SQL query is used to connect the actual dimension table and the tolerance information table to generate a manufacturing data table. The manufacturing data is stored in a common format, such as CSV or XML file, for easy access in subsequent steps. During the integration process, data consistency needs to be verified, such as checking the dimensional matching of actual dimensions and tolerance information to avoid data loss or misalignment. After the manufacturing data is generated, a verification step is taken to ensure integrity, such as calculating whether the number of data records matches the total number of curtain wall units. If they do not match, re-collection or recalculation is triggered. Finally, the manufacturing data is output as a digital file for subsequent virtual model generation steps. This integration process ensures that all information is traceable and accessible, supporting the coherent execution of the entire pre-assembly process.

[0071] S2. Based on the design model and manufacturing data, generate a virtual model for each prefabricated curtain wall unit. The virtual model includes geometry and connection interfaces. Specifically, the implementation is as follows:

[0072] In the process of generating a virtual model for each prefabricated curtain wall unit based on the design model and manufacturing data, the first step is to input the actual dimensions from the manufacturing data into the design model and replace the corresponding theoretical dimension parameters. The design model typically exists in a Building Information Modeling (BIM) format, such as using the IFC standard or Revit's native format. The actual dimension data comes from the manufacturing data file, which contains the actual length, width, and height values ​​of the prefabricated curtain wall unit. The input process is implemented using parametric modeling software, such as Autodesk. Revit's API or Rhino's Grasshopper component maps actual dimensional values ​​to corresponding parameter positions in the design model. The replacement operation is performed independently for each dimensional parameter. For example, the theoretical length value in the design model is replaced with the actual length value in the manufacturing data. During the replacement, it is necessary to ensure that the parameter name and unit are consistent, and the unit is uniformly in millimeters to avoid dimensional errors. The replacement process adopts a batch processing method. The script reads the manufacturing data file, traverses all design model instances of the curtain wall prefabricated units, and updates the parameter values ​​one by one. The script is written in an existing programming language such as Python and uses an existing library such as pyRevit to access the model. After the replacement is completed, all theoretical dimensional parameters in the design model are overwritten by the actual dimensional parameters, generating an updated design model, which serves as the basis for subsequent geometric reconstruction.

[0073] When reconstructing the geometry of the prefabricated curtain wall units based on the updated design model, the reconstruction process is executed through a parametric modeling engine. This engine automatically recalculates the shape and position of geometric elements according to the updated dimensional parameters. The core steps of geometric reconstruction include resolving geometric dependencies in the design model, such as constraints and associated conditions, and regenerating the 3D geometry using actual dimensional values. The reconstruction operation starts with basic geometric features, such as updating the main plane of the curtain wall unit first, and then gradually processing detailed features such as edge chamfers and holes. Geometric consistency must be maintained during the reconstruction process to ensure that all adjacent faces and parallel relationships meet the actual dimensional requirements. For example, when... When the actual length increases, the relevant width and height dimensions need to be adjusted proportionally. The reconstruction algorithm is based on boundary representation or solid geometry construction, both of which are existing computer-aided design methods. The reconstruction accuracy is controlled by setting a geometric tolerance threshold, which is based on the curtain wall manufacturing precision requirements. For example, the geometric deviation is controlled within 0.1 mm. The tolerance threshold is determined through experimental calibration. Specifically, different dimensional values ​​are applied in the test model, and the matching degree between the output geometry and the actual point cloud data is measured. The threshold that minimizes the average error is selected. The reconstructed geometry is saved as a 3D mesh or NURBS surface format to ensure consistency with the original design. Figure 1 It reflects the actual manufacturing status.

[0074] The mating surface geometry data of the connection interface is updated based on the reconstructed geometry. The connection interface is defined as the contact area between prefabricated curtain wall units used for assembly. The mating surface geometry data includes the position, normal vector, and curvature information of the mating surface. The update process first identifies the area in the reconstructed geometry corresponding to the connection interface, for example, by locating the interface boundary through labels or feature detection algorithms. The calculation of the mating surface geometry data is based on the vertex coordinates and patch data of the reconstructed geometry, for example, by fitting the plane equation of the mating surface using the least squares method and extracting the normal vector as a direction reference. The update operation includes adjusting the relative position and orientation of the mating surfaces to match the actual size changes. For example, when the unit size increases, the mating surfaces need to be translated accordingly to ensure assembly alignment. During the update process, the interference between the mating surfaces needs to be checked. Interference checking is achieved by calculating the distance between the surfaces. The distance threshold is set to, for example, 1 mm. If the distance exceeds the threshold, it is marked as needing further adjustment. The updated mating surface geometry data is integrated into the design model, replacing the original interface data, and the data integrity is ensured through verification steps, such as checking whether all interfaces have corresponding geometric definitions. If any are missing, recalculation is triggered.

[0075] The model containing the updated geometry and connection interfaces is defined as a virtual model. The virtual model is a digital representation that integrates the updated geometry and connection interface data. The definition process includes model format conversion and data encapsulation. The virtual model is saved in a lightweight format such as GLTF or OBJ for easy loading and processing in the virtual environment. Data encapsulation ensures that the geometry and connection interfaces are stored as a unified entity, for example, by linking the geometry file and the interface configuration file through metadata. After the virtual model is generated, its consistency with the manufacturing data is verified through a verification step, such as comparing the deviation of the dimensional values ​​in the virtual model with the actual dimensions. The allowable deviation range is set to, for example, 0.5 mm. If the deviation exceeds the range, the reconstruction step is re-executed. Finally, the virtual model is output as a digital file for subsequent tolerance accumulation analysis steps. This model, as a digital twin, accurately reflects the physical characteristics of the curtain wall prefabricated unit.

[0076] S3. Perform tolerance accumulation analysis on multiple virtual models under the assembly sequence to simulate the transmission and accumulation of manufacturing errors through connection interfaces in the assembly path, in order to identify systematic assembly conflict risks. The specific implementation is as follows:

[0077] When performing tolerance accumulation analysis on multiple virtual models under an assembly sequence, each virtual model must first be processed according to a predefined assembly sequence order. The assembly sequence is stored in list form, explicitly specifying the installation order of the prefabricated curtain wall units in the virtual environment. This sequence is determined by construction process requirements, such as installation logic from the bottom to the top or from the center to both sides. When processing each virtual model, the system loads the virtual model data sequentially according to the sequence. The virtual model data includes geometric shape data and connection interface data, originating from the virtual model files generated in previous steps. The sequential processing is implemented using a loop algorithm that traverses the assembly sequence list, accesses each virtual model instance in turn, and establishes temporary relationships between models to support subsequent error propagation analysis. During processing, the system maintains a status log table to track the processing progress and temporary data of each virtual model, ensuring strict adherence to the sequence order.

[0078] For each virtual model, the transmission process of manufacturing errors from the preceding virtual model to the current virtual model is simulated through the connection interface. This process uses the cumulative error effect of the preceding virtual model and the tolerance information of the current virtual model as input parameters. The cumulative error effect includes the positional offset and rotational angle deviation of the preceding model, while the tolerance information comes from the deviation values ​​in the manufacturing data. Error transmission simulation is based on the geometric relationship of the mating surfaces of the connection interface. The geometric data of the mating surfaces includes their spatial position and normal vector direction. The calculation of the error transmission vector is accomplished through a geometric projection method. Specifically, the cumulative error effect of the preceding virtual model is decomposed into components along the normal and tangential vector directions of the mating surfaces, and then the transmission ratio is calculated according to the contact type of the mating surfaces. For example, for a planar contact interface, the error transmission ratio in the normal vector direction is set to 1, and the error transmission ratio in the tangential vector direction is set to 0.5. The setting of the error transmission ratio is determined experimentally based on the physical characteristics of the contact surfaces. The calculated error transmission vector includes three translational components and three rotational components, corresponding to the error amounts of the six degrees of freedom.

[0079] The error propagation vector is applied to the position offset and rotation angle of the current virtual model to update its spatial pose. The position offset is in millimeters and includes three components: X-axis offset, Y-axis offset, and Z-axis offset. The rotation angle is in degrees and includes three components: rotation around the X-axis, rotation around the Y-axis, and rotation around the Z-axis. The application process is achieved through spatial transformations. Specifically, the translation component of the error propagation vector is vector-added with the original position offset of the current virtual model, and the rotation component of the error propagation vector is vector-added with the original rotation angle of the current virtual model. After the spatial pose is updated, the system recalculates the bounding box and collider of the current virtual model, which are used for subsequent interference detection. When applying the error propagation vector, unit consistency and coordinate system consistency must be considered. All calculations are performed in a unified global coordinate system to avoid introducing additional errors due to coordinate system transformation.

[0080] The cumulative error effect is generated by accumulating the error effects of previous virtual models into the current virtual model. The accumulation process employs an iterative algorithm, starting with the first virtual model in the assembly sequence, whose cumulative error effect is initialized to a zero vector. For each subsequent virtual model in the sequence, its cumulative error effect is obtained by weighted summation of the error effects of all preceding virtual models. The weighting coefficients are determined based on the type and location of the connection interfaces; for example, a weighting coefficient of 1.0 is set for primary load-bearing interfaces, and 0.7 for secondary connection interfaces. The weighting coefficients are set based on the importance assessment of the interfaces, which is determined through stress and structural analysis, with specific values ​​obtained through finite element simulation and experimental verification. The calculation of the cumulative error effect includes two parts: translational accumulation and rotational accumulation. Translational accumulation is achieved through vector addition, while rotational accumulation is achieved through quaternion interpolation or Euler angle transformation, ensuring the geometric correctness of the accumulation process.

[0081] The mating status of connection interfaces is assessed based on the cumulative error effect. When the mating status indicates assembly conflict, it is identified as a systemic assembly conflict risk. The mating status assessment is achieved by calculating the difference between the actual and theoretical gaps between connection interfaces. The actual gap is calculated based on a virtual model after updating the spatial attitude, while the theoretical gap is derived from design specifications. A mating status threshold is set during the assessment process. This gap threshold is determined based on the type of connector; for example, the gap threshold is set to 2 mm for bolted connections and 0.5 mm for welded connections. When the actual gap is less than the gap threshold, it is judged as a poor mating status and marked as an assembly conflict risk. The identification of systemic assembly conflict risks also needs to consider the propagation characteristics of the conflict; that is, a conflict at a single interface may trigger a chain reaction. Therefore, it is necessary to analyze the propagation path of the conflict in the assembly sequence. The finally identified systemic assembly conflict risks are recorded in a risk list, including information such as the conflict location, conflict type, and conflict severity. This information serves as input data for subsequent optimization and adjustments.

[0082] S4. Assess the critical tolerance transfer path that has the greatest impact on overall assembly quality in the assessment of systemic assembly conflict risks. The specific implementation is as follows:

[0083] In assessing the critical tolerance transfer paths that have the greatest impact on overall assembly quality in the evaluation of systemic assembly conflict risks, the first step is to extract each tolerance transfer path from the identified systemic assembly conflict risks. Systemic assembly conflict risks originate from the tolerance accumulation analysis results of previous steps and are stored in list form, including conflict location, conflict type, and conflict severity information. A tolerance transfer path is defined as a complete chain of manufacturing errors propagating through connection interfaces in the assembly sequence; each path consists of a series of consecutive virtual models and the connection interfaces between them. The extraction process is achieved by analyzing the propagation relationship of conflict risks. Specifically, it involves traversing each systemic assembly conflict risk record, identifying the virtual model sequence and connection interface sequence involved, thereby constructing a complete tolerance transfer path. The extraction operation uses a depth-first search method from graph theory, a current technique. This method uses virtual models as nodes and connection interfaces as edges to construct a directed graph model, and then traces the error source path backward from the conflict risk point. Each extracted tolerance transfer path is stored as a path description file, including the path number, a list of included virtual models, and a list of connection interfaces.

[0084] Sensitivity analysis was performed on each tolerance transfer path. The degree of change in overall assembly quality was observed by adjusting the tolerance parameter values ​​for the corresponding paths. Sensitivity analysis was conducted independently for each tolerance transfer path. First, the tolerance parameter values ​​corresponding to that path were read. These tolerance parameter values ​​were derived from tolerance information in the manufacturing data, including dimensional deviations and angular deviations. Adjustment was achieved by modifying the numerical values ​​of the tolerance parameters. The modification range was set as a percentage of the original tolerance parameter value, for example, increasing or decreasing the tolerance parameter value by 10%. The adjustment range was determined experimentally, with a typical range of ±20% of the original value. After adjustment, a complete assembly sequence simulation was re-executed in the virtual environment. The assembly sequence simulation process included loading all virtual models, assembling according to the assembly sequence, and detecting interference during the assembly process. The degree of change in overall assembly quality was quantified using overall assembly quality indicators, including assembly accuracy and interference levels. The assembly accuracy indicator was calculated as the average deviation between the actual assembly position and the theoretical assembly position, and the interference level indicator was calculated as the total detected interference volume.

[0085] For each tolerance transfer path, its tolerance parameter values ​​are modified independently, and the assembly sequence simulation is re-executed in the virtual environment. When modifying the tolerance parameter values, only the tolerance parameter values ​​corresponding to the current analysis path are changed; the tolerance parameter values ​​for other paths remain unchanged to ensure the independence of the analysis. When re-executing the assembly sequence simulation, the same virtual environment and assembly sequence parameters as the original simulation are used, but the modified tolerance parameter values ​​are applied. The simulation process includes initializing the virtual environment, loading the virtual model, executing assembly operations in sequence, and recording the assembly results. After each simulation, the response gradient of the overall assembly quality index relative to the change in tolerance parameter values ​​is calculated. The response gradient is obtained by calculating the ratio of the change in the overall assembly quality index to the change in the tolerance parameter values. Specifically, the response gradient equals the change in the overall assembly quality index divided by the change in the tolerance parameter values. The change in the overall assembly quality index is the index value after modification minus the original index value, and the change in the tolerance parameter values ​​is the modified tolerance parameter value minus the original tolerance parameter value.

[0086] After calculating the response gradient of the overall assembly quality index relative to the change in tolerance parameter values, the influence of the corresponding tolerance transfer path on the overall assembly quality is quantified based on the response gradient. The quantification process is achieved by mapping the response gradient value to an influence score, which is based on a percentage scale; the larger the absolute value of the response gradient, the higher the influence score. The specific mapping relationship is linear; for example, the influence score is zero when the absolute value of the response gradient is zero, and one hundred when the absolute value of the response gradient is at its maximum value. The maximum value is determined through historical data analysis. The influence score calculation also considers the length and complexity of the tolerance transfer path. The path length is defined as the number of virtual models contained in the path, and the complexity is assessed by the number of connection interface types in the path. The length and complexity of the tolerance transfer path are incorporated into the score calculation through weighting coefficients, which are determined by expert evaluation based on path characteristics.

[0087] Based on the sensitivity analysis results, the influence of each tolerance transfer path is ranked. The ranking process is achieved by comparing the influence scores of each tolerance transfer path, with paths having higher influence scores ranked higher. The ranking algorithm uses the existing quicksort algorithm, which recursively divides the path list into smaller and larger sublists and then recursively sorts the sublists. Stability conditions are set during the ranking process to ensure that paths with the same influence score maintain their original relative order. The ranking result is stored as an ordered list, including path numbers and corresponding influence scores, which serves as the basis for subsequently determining critical paths. After ranking, a verification step checks the consistency of the ranking logic, such as checking whether all paths have been correctly ranked; if any are missing, the sensitivity analysis is re-executed.

[0088] The tolerance transfer path with the highest impact ranking is identified as the critical tolerance transfer path. The selection process directly chooses the path with the highest impact score from the ranking results. If multiple paths have the same highest score, the shorter path is selected as the critical tolerance transfer path. The critical tolerance transfer path record includes detailed path information, such as the included virtual model, connection interfaces, and corresponding tolerance parameter values. After identification, the rationality of the critical tolerance transfer path is verified through a validation step. For example, it checks whether the path is indeed associated with significant systemic assembly conflict risks. Validation methods include path importance analysis and risk propagation analysis. The final critical tolerance transfer path is output as a file in a specified format for subsequent dynamic adjustment steps.

[0089] S5. Based on the critical tolerance transfer path, dynamically adjust the assembly parameters and assembly sequence of the virtual model. The assembly parameters include position offset, rotation angle, and connection point coordinates. Specifically, the implementation is as follows:

[0090] When dynamically adjusting the assembly parameters and sequence of virtual models based on the critical tolerance transfer path, the first step is to identify the virtual models involved in the critical tolerance transfer path as priority adjustment objects. The critical tolerance transfer path originates from the path data obtained in previous steps. This data is stored in a structured format, including the path number, a list of included virtual models, and a list of connection interfaces. The identification process is achieved by parsing the components of the critical tolerance transfer path. Specifically, it iterates through each connection interface in the path, extracts the virtual model instance associated with the interface, and marks the virtual model instance as a priority adjustment object. The identification operation uses set operation methods from existing technologies, such as difference and union operations, to remove duplicate models, ensuring that each virtual model is identified only once. The determination of priority adjustment objects also considers the positional weight of the virtual model in the assembly sequence. The positional weight is set based on the model's order in the sequence; for example, models earlier in the sequence have higher weights. The weight values ​​are calculated using a linear allocation method, specifically setting the weight of the first model to 1.0, and decreasing by 0.1 for subsequent models sequentially. After identification, the list of priority adjustment objects is stored as temporary data for subsequent optimization calculations.

[0091] The assembly parameters of the priority adjustment objects are optimized to generate adjustment schemes for position offsets, rotation angles, and connection point coordinates. The optimization calculation aims to minimize cumulative error, employing the gradient descent algorithm, a technique used in existing technologies. This algorithm finds the minimum value of the objective function by iteratively adjusting the assembly parameter values. Input parameters include the current assembly parameter values ​​of the priority adjustment objects and error data from the critical tolerance transfer path. The current assembly parameter values ​​are derived from the spatial pose data of the virtual model, and the error data originates from the cumulative error effects of previous steps. The core steps of the optimization calculation include initializing the assembly parameter values, calculating the gradient of the objective function, updating the parameter values, and checking the convergence condition. The objective function is defined as the sum of squares of the cumulative error, which is calculated by simulating the assembly process. The parameter update step size is based on a learning rate, determined through trial and error; for example, the initial learning rate is set to 0.01 and dynamically adjusted according to the convergence results. After the adjustment scheme is generated, it includes specific values ​​for position offsets, rotation angles, and connection point coordinates. Position offsets are in millimeters, rotation angles are in degrees, and connection point coordinates are represented by three-dimensional coordinates.

[0092] The assembly parameters of priority adjustment objects are updated based on the adjustment scheme. The update operation is implemented by modifying the data structure of the virtual model, specifically by writing the position offset, rotation angle, and connection point coordinate values ​​from the adjustment scheme into the corresponding fields of the virtual model. Position offset updates include assigning values ​​to three components: X-axis offset, Y-axis offset, and Z-axis offset. Rotation angle updates include assigning values ​​to three components: rotation around the X-axis, rotation around the Y-axis, and rotation around the Z-axis. Connection point coordinate updates involve recalculating the 3D coordinates of all connection points. During the update process, data consistency must be ensured. For example, it is checked whether the updated assembly parameters are within a reasonable range. A reasonable range is based on physical constraints; for example, the allowable range for position offset is ±10 mm, and the allowable range for rotation angle is ±5 degrees. The update operation is executed in batches by a script program. The script program traverses the list of priority adjustment objects, applies the adjustment scheme to each one, and records the update log. After the update is completed, the spatial pose and connection interface data of the virtual model are refreshed, reflecting the optimized state.

[0093] The assembly sequence is replanned based on the updated assembly parameters to generate an assembly order that minimizes cumulative error. The replanning process is based on the shortest path algorithm in graph theory, using virtual models as nodes and connecting interfaces as edges. The edge weight is defined as the expected cumulative error value under the updated assembly parameters. The expected cumulative error value is calculated by simulating the assembly process, using the updated assembly parameters and executing the assembly according to a temporary sequence. The assembly sequence generation employs a topological sorting algorithm from existing technologies, which considers the dependencies between virtual models to ensure the assembly order conforms to physical logic. Minimizing cumulative error is achieved by optimizing the order of models in the sequence, for example, placing models with large error contributions later in the sequence to reduce error accumulation. When generating the assembly order, a sequence evaluation threshold is set, based on historical assembly data; for example, the upper limit of cumulative error is set to 2 mm. If the threshold is exceeded, the sequence is readjusted. The final assembly order output is a sequence list, including model numbers and installation positions, for subsequent virtual assembly simulations.

[0094] S6. Using the adjusted assembly parameters, simulate the assembly process of the prefabricated curtain wall units in a virtual environment to generate an assembly effect diagram, and verify the assembly effect diagram to output a verification report. The specific implementation is as follows:

[0095] When simulating the assembly process of prefabricated curtain wall units in a virtual environment using adjusted assembly parameters to generate and verify assembly renderings, the virtual models are first assembled sequentially according to the adjusted assembly parameters and sequence. The adjusted assembly parameters are derived from dynamic adjustments made in previous steps, including specific values ​​for position offsets, rotation angles, and connection point coordinates. The assembly sequence is derived from a redesigned list of possible orders. The virtual environment is implemented using 3D modeling software platforms, such as Autodesk Revit or the Unity engine. This environment loads all virtual model data and applies the adjusted assembly parameters. The assembly process is controlled by a script. The script reads the assembly sequence list, instantiates the virtual models sequentially, and sets the spatial orientation of each model based on the position offset and rotation angle parameters. Connection point coordinates are used to accurately locate the interfaces between models, ensuring that the assembly position meets the actual assembly requirements. During assembly, the virtual environment updates the model position and orientation in real time, simulates physical assembly actions, and records the assembly status at each step.

[0096] Interference between virtual models is detected in real time during assembly. Interference detection is implemented using a collision detection algorithm, which utilizes existing techniques such as the separating axis theorem or boundary volume hierarchy method to calculate geometric overlap between models. The detection process is executed after each model position update, with input parameters including the geometric mesh data and spatial transformation matrix of the virtual model. Interference is quantified by calculating the minimum distance between models, based on vertex coordinates, using existing libraries such as Bullet Physics or Open Dynamics Engine. Real-time detection is configured with a detection frequency, e.g., 60 checks per second, to ensure timeliness. An interference threshold is set based on the connection interface type; for example, for bolted connections, the interference threshold is set to 0.5 mm. When a distance less than this threshold is detected, it is marked as an interference event. Interference event records include the interfering model number, interference location, and interference depth, and are used for subsequent analysis and reporting.

[0097] After assembly, a rendering of the overall assembled state is generated. This rendering is produced by a rendering engine that uses ray tracing or rasterization techniques to convert the 3D scene in the virtual environment into a 2D image. The rendering process sets viewpoint parameters, such as using isometric or perspective views, to fully display the assembled state. The rendering includes applying material maps, lighting settings, and background environment to enhance visualization. The generated rendering is saved in a standard image format, such as PNG or JPEG, with a resolution set to, for example, 1920 x 1080 pixels. The assembly rendering also includes annotation information, such as model numbers and key dimensions, for easy and intuitive evaluation of assembly quality.

[0098] Structural stability verification and visual alignment checks are performed on the assembled renderings. Structural stability verification is achieved through finite element analysis simulation, which applies static principles to calculate the stress distribution and deformation of the virtual model under gravity and wind loads. Input parameters include material properties such as elastic modulus and Poisson's ratio, and are based on the actual materials of the prefabricated curtain wall units. Stability verification sets a safety factor threshold; for example, the maximum allowable stress is set to 60% of the material's yield strength. Exceeding this threshold is marked as a stability risk. Visual alignment checks are implemented using image processing algorithms that compare the deviation between the actual alignment lines and the theoretical alignment lines in the assembled renderings. Alignment deviation is calculated using pixel coordinates, and alignment features are extracted using edge detection techniques such as the Canny operator. The visual alignment threshold is set based on human visual perception; for example, a visible deviation threshold of 2 pixels is set. Deviations exceeding this threshold are marked as poor alignment.

[0099] A validation report, including an assembly quality assessment, is generated based on the verification and inspection results. The report generation process first integrates the structural stability verification results and visual alignment inspection results, including all marked risk events and deviation data. The assembly quality assessment is achieved through a comprehensive scoring system, calculated based on the number of stability risks and the degree of alignment deviation using a weighted average method. Weights are set according to the importance of the risks; for example, the stability weight is set to 0.6, and the alignment weight to 0.4. The scoring threshold is set based on industry standards; for example, a passing score of 80 is set. The validation report is output as a structured document, such as a PDF, including an assessment summary, a detailed data list, and improvement recommendations. After report generation, a verification step ensures data integrity, such as checking whether all virtual models are covered, triggering re-verification if any are missing.

[0100] Example 2: Figure 2 A structural schematic diagram of a virtual pre-assembly system for curtain wall prefabricated units according to the present invention is provided. The virtual pre-assembly system for curtain wall prefabricated units includes the following modules:

[0101] The data acquisition module is used to acquire design models and manufacturing data of multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information.

[0102] The model generation module is used to generate a virtual model of each prefabricated curtain wall unit based on the design model and manufacturing data. The virtual model includes geometry and connection interfaces.

[0103] The risk identification module is used to perform tolerance accumulation analysis on multiple virtual models under the assembly sequence, and to simulate the transmission and accumulation of manufacturing errors through the connection interface in the assembly path in order to identify systematic assembly conflict risks.

[0104] The path assessment module is used to evaluate the critical tolerance transfer path that has the greatest impact on overall assembly quality in the risk of systemic assembly conflict.

[0105] The dynamic adjustment module is used to dynamically adjust the assembly parameters and assembly sequence of the virtual model according to the key tolerance transfer path. The assembly parameters include position offset, rotation angle and connection point coordinates.

[0106] The virtual simulation module is used to simulate the assembly process of prefabricated curtain wall units in a virtual environment using adjusted assembly parameters to generate assembly effect diagrams, and to verify the assembly effect diagrams to output a verification report.

[0107] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0108] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0109] 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 inventive 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.

[0110] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0113] 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 virtual pre-assembly method for prefabricated curtain wall units, characterized in that, Includes the following steps: S1. Obtain design models and manufacturing data for multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information. S2. Based on the design model and manufacturing data, generate a virtual model for each prefabricated curtain wall unit. The virtual model includes geometry and connection interfaces. S3. Perform tolerance accumulation analysis on multiple virtual models under the assembly sequence to simulate the transmission and accumulation of manufacturing errors through the connection interface in the assembly path, so as to identify the risk of systematic assembly conflict. S4. Assess the critical tolerance transfer path that has the greatest impact on overall assembly quality in the assessment of systemic assembly conflict risks; S5. Based on the key tolerance transfer path, dynamically adjust the assembly parameters and assembly sequence of the virtual model. The assembly parameters include position offset, rotation angle and connection point coordinates. S6. Using the adjusted assembly parameters, simulate the assembly process of the prefabricated curtain wall units in a virtual environment to generate an assembly effect diagram, and verify the assembly effect diagram to output a verification report.

2. The virtual pre-assembly method for a prefabricated curtain wall unit according to claim 1, characterized in that, Obtain design models and manufacturing data for multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information, including: Point cloud data of the surface of the prefabricated curtain wall unit was collected using an optical 3D scanning device; The actual dimensions of the prefabricated curtain wall units are calculated based on point cloud data. The actual dimensions are compared with the corresponding theoretical dimensions in the design model to generate tolerance information containing deviation values; The actual dimensions and tolerance information are integrated to form manufacturing data.

3. The virtual pre-assembly method for a prefabricated curtain wall unit according to claim 1, characterized in that, Based on the design model and manufacturing data, a virtual model is generated for each prefabricated curtain wall unit. The virtual model includes geometry and connection interfaces, including: Input the actual dimensions from the manufacturing data into the design model, replacing the corresponding theoretical dimension parameters; The geometry of the prefabricated curtain wall units was reconstructed based on the updated design model; Update the mating surface geometry data of the connection interface based on the reconstructed geometry; The model containing the updated geometry and connection interfaces is defined as a virtual model.

4. The virtual pre-assembly method for a prefabricated curtain wall unit according to claim 1, characterized in that, Tolerance accumulation analysis was performed on multiple virtual models under assembly sequence to simulate the transmission and accumulation of manufacturing errors through connection interfaces in the assembly path, in order to identify systematic assembly conflict risks, including: Each virtual model is processed sequentially based on a predefined assembly sequence; For each virtual model, the process of manufacturing errors being transmitted from the previous virtual model to the current virtual model is simulated through the connection interface; Accumulate the error effects of the preceding virtual model into the current virtual model to generate a cumulative error effect; The mating status of the connection interface is evaluated based on the cumulative error effect. When the mating status indicates an assembly conflict, it is identified as a systemic assembly conflict risk.

5. The virtual pre-assembly method for a prefabricated curtain wall unit according to claim 4, characterized in that, For each virtual model, the process of simulating the transmission of manufacturing errors from the previous virtual model to the current virtual model through the connection interface includes: calculating the error transmission vector based on the cumulative error effect of the previous virtual model and the tolerance information of the current virtual model through the geometric relationship of the mating surfaces of the connection interface; and applying the error transmission vector to the position offset and rotation angle of the current virtual model to update its spatial attitude.

6. The virtual pre-assembly method for a prefabricated curtain wall unit according to claim 1, characterized in that, The key tolerance transfer paths that have the greatest impact on overall assembly quality in assessing systemic assembly conflict risks include: Extract tolerance transfer paths from identified systemic assembly conflict risks; Sensitivity analysis was performed on each tolerance transfer path, and the degree of change in overall assembly quality was observed by adjusting the tolerance parameter values ​​of the corresponding paths. Based on the sensitivity analysis results, the degree of influence of each tolerance transfer path is ranked. The tolerance transfer path with the highest degree of impact is identified as the critical tolerance transfer path.

7. A virtual pre-assembly method for a prefabricated curtain wall unit according to claim 6, characterized in that, Sensitivity analysis was performed on each tolerance transfer path, including: for each tolerance transfer path, its tolerance parameter value was modified independently and the assembly sequence simulation was re-executed in the virtual environment; the response gradient of the overall assembly quality index relative to the change in tolerance parameter value was calculated; and the influence of the corresponding tolerance transfer path on the overall assembly quality was quantified based on the response gradient.

8. The virtual pre-assembly method for a prefabricated curtain wall unit according to claim 1, characterized in that, Based on the critical tolerance transfer path, the assembly parameters and assembly sequence of the virtual model are dynamically adjusted. The assembly parameters include position offsets, rotation angles, and connection point coordinates, including: Identify virtual models involved in key tolerance transfer paths as priority adjustment targets; The assembly parameters of the priority adjustment objects are optimized and calculated to generate adjustment schemes for position offset, rotation angle and connection point coordinates; Based on the adjustment plan, prioritize the updating of assembly parameters for the adjusted objects; The assembly sequence is replanned based on the updated assembly parameters to generate an assembly order that minimizes cumulative error.

9. A virtual pre-assembly method for a prefabricated curtain wall unit according to claim 1, characterized in that, Using the adjusted assembly parameters, the assembly process of the prefabricated curtain wall units is simulated in a virtual environment to generate an assembly rendering. The assembly rendering is then verified to output a verification report, including: Assemble each virtual model sequentially in the virtual environment according to the adjusted assembly parameters and assembly sequence; Real-time detection of interference between virtual models during assembly; After assembly is completed, an assembly effect diagram showing the overall assembly status will be generated; Perform structural stability verification and visual alignment checks on the assembled renderings; A verification report containing an assembly quality assessment is generated based on the verification and inspection results.

10. A virtual pre-assembly system for prefabricated curtain wall units, used to implement the virtual pre-assembly method for prefabricated curtain wall units as described in any one of claims 1-9, characterized in that, Includes the following modules: The data acquisition module is used to acquire design models and manufacturing data of multiple prefabricated curtain wall units. The manufacturing data includes actual dimensions and tolerance information. The model generation module is used to generate a virtual model of each prefabricated curtain wall unit based on the design model and manufacturing data. The virtual model includes geometry and connection interfaces. The risk identification module is used to perform tolerance accumulation analysis on multiple virtual models under the assembly sequence, and to simulate the transmission and accumulation of manufacturing errors through the connection interface in the assembly path in order to identify systematic assembly conflict risks. The path assessment module is used to evaluate the critical tolerance transfer path that has the greatest impact on overall assembly quality in the risk of systemic assembly conflict. The dynamic adjustment module is used to dynamically adjust the assembly parameters and assembly sequence of the virtual model according to the key tolerance transfer path. The assembly parameters include position offset, rotation angle and connection point coordinates. The virtual simulation module is used to simulate the assembly process of prefabricated curtain wall units in a virtual environment using adjusted assembly parameters to generate assembly effect diagrams, and to verify the assembly effect diagrams to output a verification report.

Images