Building steel structure in-depth design and virtual pre-assembly system based on BIM

Abstract

The present invention discloses a BIM-based building steel structure in-depth design and virtual pre-assembly system. By building a BIM system server to integrate multi-professional models, a multi-professional model integration platform is formed. On the platform, a three-dimensional model of the steel structure is created and refined, and design parameters and engineering data are input to form an accurate geometric representation. A geometric algorithm is used to automatically check the relative position and overlap between models, establish the association and dependency between models, automatically update the model, and perform collision detection. During the in-depth design process, the steel structure nodes are optimized, and drawings and material lists are generated. According to the generated drawings and lists, a trial assembly of the steel structure segmented components is carried out, and the assembly is simulated to form a virtual assembly contour model, which is then fitted and compared with the theoretical model, and the processing and assembly accuracy is analyzed to ensure the accuracy of the engineering installation and achieve zero-error installation. This method improves design efficiency, reduces construction errors, and is suitable for the in-depth design and pre-assembly of complex steel structure projects.

Description

Technical Field

[0001] The present invention relates to digital design and analysis of building steel structures, and in particular to a BIM-based in-depth design of building steel structures and a virtual pre-assembly system. Background Art

[0002] In the field of construction engineering, steel structure support frames are widely used due to their high strength, light weight and ease of construction. These support frames are mainly used to provide temporary or permanent structural support during the construction process to ensure construction safety and project quality. With the development of construction technology, the performance requirements for steel structure support frames are becoming higher and higher, including load-bearing capacity, stability, adjustment flexibility and the ability to adapt to different construction environments.

[0003] In the prior art, patent document CN 109267767 A proposes a building support frame structure, which improves the flexibility and stability of the support frame through designs such as rotatable main steel fixed feet and linked winding columns.

[0004] Patent document CN 109098471A introduces a steel structure support frame for construction engineering, which improves the adjustment capability and application range of the support frame through a slewing device and an adjustable load-bearing beam.

[0005] Patent document CN 109149470 A proposes a steel structure side wall bracket with adjustable support height. The bracket adjusts the support height through a hydraulic system, providing a solution for flexibly adjusting the support height.

[0006] However, these technologies may have limitations in terms of the stability, load-bearing capacity and ability to adapt to different construction conditions of the support frame. Detailed solutions for the rapid deployment of steel structures, height adjustment, and coordination with other construction equipment still need to be improved.

[0007] With the development of construction technology, steel structure design and construction increasingly rely on advanced computer simulation technologies such as finite element analysis (FEA) and building information modeling (BIM). These technologies can improve design accuracy, optimize construction plans, and predict potential problems in advance.

[0008] Among them, finite element analysis software, such as ANSYS and ABAQUS, has been widely used in stress analysis and modal analysis of steel structures. These software can accurately simulate the response of the structure under various loads. Construction simulation technology, including simulation of temporary support, material transportation, component installation, etc. during the construction process, can optimize the construction process and reduce construction risks.

[0009] Building Information Modeling (BIM) technologies, such as Revit and ArchiCAD, provide an integrated 3D model environment that enables designers and engineers to predict and resolve conflicts in design and construction before construction. BIM technology creates a digital model of a building, allowing designers, engineers, and contractors to better visualize the project before construction and coordinate during the construction process.

[0010] However, there may be limitations in the compatibility between BIM software and other engineering software in the prior art. Such compatibility issues may lead to data loss or errors during data exchange, requiring manual intervention to correct, which not only increases the workload but may also introduce additional errors.

[0011] In addition, data conversion between different software platforms may not be smooth enough, affecting the efficiency of collaborative work among multi-disciplinary teams. Moreover, the creation and maintenance of BIM models require highly specialized knowledge and skills. For the refined modeling of complex structures, such as spatial structures, bending and torsion components, and special-shaped components, designers need to have advanced operating skills and in-depth understanding. This may lead to high technical barriers for design and construction teams when applying BIM technology, limiting the popularity and depth of application of BIM technology. Summary of the invention

[0012] The purpose of the present invention is to provide a BIM-based building steel structure in-depth design and virtual pre-assembly system to overcome the deficiencies of the prior art.

[0013] Based on the first main aspect of the present invention, a BIM-based building steel structure in-depth design and virtual pre-assembly method is provided, comprising the following steps:

[0014] S1, build a BIM system server, integrate professional model software in the BIM system, import different professional models into the BIM system according to the construction logic and sequence using the international unified format IFC, and form a multi-professional model integration platform;

[0015] S2, creating a three-dimensional model of the steel structure in the multi-disciplinary model integration platform, inputting design parameters and engineering data to form an accurate geometric representation; based on the three-dimensional model, using a parametric modeling method to refine the geometric shape of each component;

[0016] S3, in the multi-professional model integration platform, the relative positions and overlaps between the models are checked by geometric algorithms, associations and dependencies are established between the models, the models are automatically updated according to the update amplitude formula, and a change propagation mechanism is constructed to achieve integration and collision detection of different professional models;

[0017] S4, using each professional model software in the multi-professional model integration platform to perform in-depth design to obtain a theoretical model for in-depth design; optimizing steel structure nodes in the theoretical model to generate in-depth design drawings and a material list;

[0018] S5, according to the design drawings and material lists generated by the multi-professional model integration platform, the measured three-dimensional coordinates of the control points of the steel structure segmented components are combined with the segmented three-dimensional trial assembly and the horizontal three-dimensional trial assembly to perform component trial assembly, and simulate the assembly in a computer to form a virtual assembly contour model of the segmented components;

[0019] S6, fit and compare the virtual assembly contour model with the theoretical model of the in-depth design, check and analyze the processing and assembly accuracy, and obtain the adjustment information required for modification; after correction, modification and simulated assembly, check the accuracy of the node connection complexity and precision requirements of the friction-type high-strength bolt connection manufacturing and processing of the component, so as to achieve the goal of zero error in the engineering installation.

[0020] As a further preferred solution, the professional model software includes one or a combination of engineering basic data analysis system, Luban Rebar 2017, Luban Civil Engineering 2017, Luban Installation 2017, Luban Site Layout 2017, Revit 2016, Tekla Structures 20.1, mobile APP-Luban View 4.1.0, SketchUp 2015, LUMION, and Pinming 3D construction planning software; the engineering basic data analysis system includes at least one or a combination of Luban Explorer, Luban Govern, Luban Plan, and Luban Works.

[0021] As a further preferred solution, the BIM system server is connected to an enterprise PDS main server, several enterprise PDS branch servers, several mobile terminals, and several computer terminals through at least one switch, and the switch is connected to the Internet through the enterprise main router.

[0022] As a further preferred embodiment, the parametric modeling method in step S2 includes using an intelligent assignment algorithm SCIA for assigning values ​​to complex spatial structures, bending-torsion components and special-shaped components. The intelligent assignment algorithm SCIA includes the steps of collecting and preprocessing design parameters and engineering data, classifying and grouping data, extracting key features from the data, calculating feature weights, establishing an intelligent assignment model, using a back-propagation algorithm to update weights and biases, and performing rationality checks and optimization on the assignment results.

[0023] As a further preferred solution, the design parameters and engineering data include three-dimensional coordinates, curvature, torsion angle, material, and strength requirements of complex spatial structures, bending and torsion components, and special-shaped components. The preprocessing of these data includes cleaning and standardization to remove outliers and noise data.

[0024] As a further preferred solution, the extracting key features from the data includes: extracting node coordinates, connection methods, and component length ratios from complex spatial structures; extracting curvature change rate, torsion angle range, and cross-sectional shape parameters from bending-torsion components; and extracting geometric shape parameters, material parameters, and strength parameters from special-shaped components; and, includes constructing a feature matrix X using the key features.

[0025] As a further preferred solution, the calculating feature weights includes calculating the covariance matrix C of the feature matrix:

[0026] ;

[0027] Where X is the feature matrix, is the transposed matrix of X, n is the number of samples;

[0028] Perform eigenvalue decomposition on the covariance matrix C and obtain the eigenvalue and the eigenvector :

[0029] = ;

[0030] Calculate the weight of each feature :

[0031] ;

[0032] Where m is the number of features; j is the index variable that traverses all feature values;

[0033] The weight of each feature Construct the weight matrix W.

[0034] As a further preferred solution, the intelligent assignment model is established, and its calculation formula is expressed as:

[0035] ;

[0036] Where x is the input feature vector, W is the weight matrix, b is the bias vector, f is the activation function, and y is the output assignment result.

[0037] As a further preferred solution, the back propagation algorithm is used to update the weights and biases using the following mean square error function:

[0038] ;

[0039] In the formula, is the actual value, is the model output value, and n is the number of samples.

[0040] As a further preferred solution, in step S3, checking the relative position and overlap between models by using a geometric algorithm in the multi-professional model integration platform includes the following steps:

[0041] In the multi-disciplinary model integration platform, each model is spatially positioned to ensure that each model is in its correct position; the reference point of each model is located using a spatial coordinate system to ensure that the spatial relationship between models is accurate;

[0042] The relative position and overlap between the models are checked by a geometric algorithm to ensure that there is no geometric conflict; in the geometric algorithm, the distance between the models is defined As a model and Model The distance between the closest points is calculated as:

[0043] ;

[0044] In the formula, and The models are and For the points in , ||pq|| is the Euclidean distance between the two points;

[0045] Establish associations and dependencies between models so that changes to one model can trigger automatic updates to related models; including: defining association metrics To measure the model and Model The correlation between them is calculated as:

[0046] ;

[0047] Where k is a constant used to adjust the sensitivity of the association measure to distance, and e is the base of the natural logarithm; The higher the value, the better the model and Model The stronger the correlation between them;

[0048] When the model When a change occurs, automatically calculate its correlation metrics with all other models , and automatically update the model according to the update amplitude formula:

[0049] ;

[0050] In the formula, It is a model The update range, is a proportional constant with a value range of 1-2; is the update threshold, triggering the model Updates;

[0051] Build a change propagation mechanism to ensure that changes to a model can be propagated consistently throughout the project model, using the change propagation coefficient formula:

[0052] ;

[0053] In the formula, is a proportional constant ranging from 1 to 2, used to adjust the intensity of change propagation; It is a model The sum of association measures with all other models.

[0054] Based on the second main aspect of the present invention, a BIM-based building steel structure in-depth design and virtual pre-assembly system is provided, which is used to implement the aforementioned BIM-based building steel structure in-depth design and virtual pre-assembly method, and the system at least includes:

[0055] BIM system server, used to integrate professional model software in the BIM system to form a multi-professional model integration platform;

[0056] The data acquisition module is used to collect various design parameters and engineering data of complex spatial structures, bending and torsion components and special-shaped components, including the three-dimensional coordinates, curvature, torsion angle, material, and strength requirements of the components, and clean and standardize these data to remove abnormal values ​​and noise data;

[0057] A data classification module is used to classify and group the data processed by the data acquisition module according to the type and function of the component;

[0058] A feature extraction module is used to extract key features from the data processed by the data classification module;

[0059] The weight calculation module is used to calculate the feature weight, including calculating the covariance matrix C of the feature matrix, performing eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors, and then calculating the weight of each feature;

[0060] Intelligent assignment module, which is used to establish an intelligent assignment model and use the back propagation algorithm to update weights and biases to minimize the error between the model output and the actual value, while also performing rationality checks and optimization on the assignment results;

[0061] The detailed design module is used to apply various professional model software to carry out detailed design, obtain the theoretical model of detailed design, optimize the steel structure nodes in the theoretical model, and generate detailed design drawings and material lists;

[0062] The virtual pre-assembly module is used to simulate the assembly in the computer to form a virtual assembly outline model of the segmented components according to the generated design drawings and material lists, and to fit and compare the virtual assembly outline model with the theoretical model of the in-depth design.

[0063] Based on the third main aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed, the aforementioned BIM-based building steel structure detailed design and virtual pre-assembly method is implemented.

[0064] According to a fourth main aspect of the present invention, there is provided an electronic device, comprising:

[0065] at least one processor;

[0066] a memory communicatively coupled to the at least one processor;

[0067] The memory stores a computer program, which, when executed by the at least one processor, enables the at least one processor to implement the aforementioned BIM-based building steel structure detailed design and virtual pre-assembly method.

[0068] Advantages and beneficial effects of the present invention:

[0069] The present invention uses the intelligent assignment algorithm SCIA to assign values ​​to complex spatial structures, bending and torsion components and special-shaped components. By collecting and processing various design parameters and engineering data, it can accurately consider various characteristics of components, such as node coordinates and connection methods of spatial structures, curvature and torsion angles of bending and torsion components, geometric shapes and material requirements of special-shaped components, etc. After a series of processing steps, including feature extraction, weight calculation, model establishment and optimization, an accurate assignment result is finally obtained, which improves the accuracy of component assignment in the in-depth design of steel structures.

[0070] The traditional method of assigning values ​​to complex components may require a lot of manual calculations and repeated adjustments, while the intelligent assignment algorithm of the present invention greatly reduces manual intervention through automated data processing and model training processes. Once the model is established and trained, a large number of components can be assigned values ​​quickly, improving the overall efficiency of the detailed design of steel structures. At the same time, the combination of parametric modeling methods and intelligent assignment algorithms makes it easier to modify and optimize components during the design process, further improving design efficiency.

[0071] By checking and optimizing the rationality of the assignment results, such as adjusting the size of components of complex spatial structures, adjusting the curvature and torsion angle of bending and torsion components, and checking the special geometric shapes and material requirements of special-shaped components, it can ensure that the assignment results meet the design specifications and engineering requirements. This helps to avoid structural safety hazards and design quality problems caused by unreasonable component assignments, and ensures the quality of the in-depth design of steel structures.

[0072] The intelligent value assignment algorithm of the present invention can perform targeted processing according to different types of components (spatial structures, bending and torsion components, special-shaped components), extract different key features and assign values. This flexibility enables the method to adapt to various complex steel structure design scenarios and improves the adaptability of the model to different engineering requirements. It can provide effective design support for both conventional steel structure projects and projects with special complex components. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A flow chart of a method for in-depth design and virtual pre-assembly of a building steel structure based on BIM in one embodiment of the present invention is shown;

[0074] Figure 2 The structure of the BIM system server in one embodiment of the present invention is shown. DETAILED DESCRIPTION

[0075] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings so that the purpose, features and advantages of the present invention can be more clearly understood. It should be understood that the following embodiments are not intended to limit the scope of the present invention, but are only intended to illustrate the essential spirit of the technical solution of the present invention.

[0076] In the following description, certain specific details are set forth for the purpose of illustrating the various disclosed embodiments to provide a thorough understanding of the various disclosed embodiments. However, those skilled in the relevant art will recognize that the embodiments may be practiced without one or more of these specific details. In other cases, well-known devices, structures, and techniques associated with the present application may not be shown or described in detail to avoid unnecessarily obscuring the description of the embodiments.

[0077] References throughout the specification to "one embodiment" or "an embodiment" indicate that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, the appearances of "in one embodiment" or "in an embodiment" in various places throughout the specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any manner in one or more embodiments.

[0078] Example 1

[0079] A certain project is a time sightseeing tower and a comprehensive parking building, which adopts pile raft foundation and has a total building height (including frame) of 99.9 meters. The building structure of the comprehensive parking building is a frame structure, the structural seismic fortification category is Class C, the building design reasonable use is 50 years, the building structure safety level is Level 2, and the total height is 18.4 meters; the foundation form adopts independent foundation and ground beam.

[0080] The project adopts the technical solution of the present invention to assist in construction. First, based on the design institute's construction drawings, calculation books and other relevant materials, the construction company builds a BIM system server, integrates professional model software in the BIM system, and imports different professional models into the BIM system according to the construction logic and sequence using the international unified format IFC, forming a multi-professional model integration platform based on a professional in-depth design software platform. By establishing a three-dimensional solid model and combining it with BIM, the model information sharing is realized, extending from the traditional "layout and drawing" to the entire construction process.

[0081] In the construction, BIM technology was used to deepen the design, and Tekla Structures (steel structure detailing design software) steel structure software was used for modeling and drawing. An information model was established based on the original design drawings, and the problems of the original design drawings were reviewed through the model, and the design was assisted in making design changes. After deepening the design drawings, it is more conducive to accurate material cutting and convenient on-site installation. The use of BIM technology to simulate construction first reduced the risk of installation and solved the difficulties in positioning the tower space angle, collision and complex nodes.

[0082] After the structural modeling is completed and reviewed, the software function is used to directly generate steel component parts drawings and assembly drawings, plane positioning and facade installation and other installation drawings to guide the steel structure production and on-site installation.

[0083] The steel structure CAM is used for auxiliary manufacturing, and BIM current simulation is used to avoid collisions between components and tower cranes. The underground steel column layer adopts one crane per layer; the tower group steel columns adopt one crane per two layers; the main tower steel columns adopt one crane per four layers. After leaving the factory, all steel columns, prefabricated stairs, steel beams and other materials will generate QR codes through the BIM system, enter relevant information and paste them on the corresponding components, so that the status of the component (transportation, installation, and all component information) can be located at any time. Use BIM technology to simulate construction, use BIM technology to simulate the layout of the on-site three-dimensional construction site, simulate the sequential hoisting of steel structure components, and simulate the coverage of tower cranes.

[0084] Specifically, the project adopts the BIM-based building steel structure in-depth design and virtual pre-assembly system of the present invention to be implemented through the following steps: Figure 1 As shown, first, build a BIM system server in the enterprise, such as Figure 2As shown, the BIM system server is connected to an enterprise PDS main server, two enterprise PDS sub-servers, several mobile terminals (mobile phone APPs), and several computer terminals (project department work computers) through at least one switch, and the switch is connected to the Internet through the enterprise main router. Professional model software is integrated in the BIM system, and models of different professions are imported into the BIM system according to the construction logic and sequence using the international unified format IFC to form a multi-professional model integration platform.

[0085] Among them, professional model software includes: basic engineering data analysis system (Luban Explorer, LubanGovern (New), Luban Plan (New), Luban Works), Tekla Structures20.1 (Lu 3banTrans_Tekla), Luban Rebar 2017, Luban Civil Engineering 2017, Luban Installation 2017, Luban Site Layout 2017, mobile APP-Luban View4.1.0, Revit2016 (LubanTrans_Revit), SketchUp 2015, LUMION, and Pinming 3D construction planning software.

[0086] The main structure of this project is prefabricated steel structure. Tekla software is used for in-depth design and processing table generation. Steel bar modeling software is used to form a 3D model and database. Civil engineering modeling software is used to form a 3D model and database. Electromechanical modeling software is used to form a 3D model and database. Construction site layout and masonry layout. Mobile phone terminals can export QR codes to support other APP scanning, upload short videos, document-related models and push them to relevant responsible persons. Revit is used for curtain wall and decoration modeling, site 3D simulation, 3D simulation construction, and 3D field layout simulation construction. Temporary facility engineering quantities can be obtained, and a 3D field layout model with data can be obtained.

[0087] In the present invention, by importing a multi-professional model integration platform (using the international unified format IFC for import), performing collision checks, internal roaming, viewing and analyzing model data, 5D display of progress plans, model and data viewing and sharing terminals, and photo and data attachment terminals, the integration and merging of multi-professional models is achieved.

[0088] After creating a three-dimensional model of the steel structure in the multi-professional model integration platform, the design parameters and engineering data are input to form an accurate geometric representation; based on the three-dimensional model, the geometric shape of each component is refined using a parametric modeling method. In the present invention, the parametric modeling method includes using an intelligent assignment algorithm SCIA for the assignment of complex spatial structures, bending and torsion components, and special-shaped components. The intelligent assignment algorithm SCIA includes the steps of collecting and preprocessing design parameters and engineering data, classifying and grouping data, extracting key features from data, calculating feature weights, establishing an intelligent assignment model, using a back-propagation algorithm to update weights and biases, and performing rationality checks and optimization on the assignment results.

[0089] Specifically, first, various design parameters and engineering data of complex spatial structures, bending-torsion components and special-shaped components are collected, including but not limited to the three-dimensional coordinates, curvature, torsion angle, material, and strength requirements of the components; these data are cleaned and standardized to remove outliers and noise data; these technologies belong to the existing technology and will not be introduced in detail here.

[0090] According to the type and function of the component, the data is classified and grouped to prepare for the subsequent assignment process.

[0091] Extract key features from the preprocessed data; for complex spatial structures, extract features such as node coordinates, connection methods, and component length ratios; for bending-torsion components, extract curvature change rate, torsion angle range, cross-sectional shape features, etc.; for special-shaped components, extract unique geometric shape parameters, special material requirements, etc.

[0092] Calculate feature weights: First, calculate the covariance matrix C of the feature matrix:

[0093] ;

[0094] Where X is the feature matrix and n is the number of samples.

[0095] Perform eigenvalue decomposition on the covariance matrix C and obtain the eigenvalue and the eigenvector :

[0096] = ;

[0097] Calculate the weight of each feature:

[0098] ;

[0099] Where m is the number of features; j is the index variable that traverses all feature values;

[0100] An intelligent assignment model is established, and its calculation formula is expressed as:

[0101] ;

[0102] Where x is the input feature vector (including weighted eigenvalues), W is the weight matrix, b is the bias vector, f is the activation function, and y is the output assignment result.

[0103] The back propagation algorithm is used to update the weights and biases to minimize the error between the model's output and the actual value. The error function uses the mean square error function:

[0104] ;

[0105] In the formula, is the actual value, is the model output value, and n is the number of samples.

[0106] Perform rationality check and optimization on the assignment results:

[0107] For the size adjustment of components of complex spatial structures, the following formula is used for verification:

[0108] ;

[0109] In the formula, is the original size, is the size adjustment value obtained by the intelligent assignment algorithm, It is the adjusted size.

[0110] The curvature and torsion angle adjustment of the bending-torsion member can be checked similarly:

[0111] ;

[0112] Where C is the original curvature, is the adjustment value, is the adjusted curvature; a similar formula is used for the torsion angle T.

[0113] For special-shaped components, check whether their special geometric shape parameters and material requirements meet the design specifications and engineering needs. If unreasonable assignment results are found, optimization algorithms such as genetic algorithms and simulated annealing algorithms can be used to adjust and optimize the assignment model until satisfactory results are obtained.

[0114] On the basis of the above, it is necessary to further check the relative position and overlap between models through geometric algorithms in the multi-professional model integration platform, establish associations and dependencies between models, automatically update the models according to the update amplitude formula, and build a change propagation mechanism to achieve the integration and collision detection of different professional models.

[0115] Specifically, first import the models of different disciplines into the multi-discipline model integration platform according to the construction logic and sequence, and perform spatial positioning to ensure that each model is in its correct position; then use the spatial coordinate system (x, y, z)(x, y, z) to locate the reference point of each model to ensure the accuracy of the spatial relationship between models.

[0116] The relative position and overlap between the models are checked by a geometric algorithm to ensure that there is no geometric conflict; in the geometric algorithm, the distance between the models is defined As a model and Model The distance between the closest points is calculated as:

[0117] ;

[0118] In the formula, and The models are and For the points in , ||pq|| is the Euclidean distance between the two points;

[0119] Establish associations and dependencies between models so that changes to one model can trigger automatic updates to related models; including: defining association metrics To measure the model and Model The correlation between them is calculated as:

[0120] ;

[0121] Where k is a constant used to adjust the sensitivity of the association measure to distance, and e is the base of the natural logarithm; The higher the value, the better the model and Model The stronger the correlation between them;

[0122] When the model When a change occurs, automatically calculate its correlation metrics with all other models , and automatically update the model according to the update amplitude formula:

[0123] ;

[0124] In the formula, It is a model The update range, is a proportional constant with a value range of 1-2; is the update threshold, triggering the model Updates;

[0125] Build a change propagation mechanism to ensure that changes to a model can be propagated consistently throughout the project model, using the change propagation coefficient formula:

[0126] ;

[0127] In the formula, is a proportional constant ranging from 1 to 2, used to adjust the intensity of change propagation; It is a model The sum of association measures with all other models.

[0128] After the above multi-disciplinary model integration platform is built, the professional model software in the multi-disciplinary model integration platform can be used to carry out in-depth design and obtain a theoretical model for in-depth design; and the steel structure nodes can be optimized in the theoretical model to generate in-depth design drawings and material lists.

[0129] Specifically, first start each professional model software, such as structural analysis software, 3D drawing software, etc., in the multi-professional model integration platform, and ensure that all necessary design parameters and engineering data have been entered and verified. Use the modeling tools of the BIM software to build a theoretical model of the steel structure based on the design parameters and engineering data. Including accurate 3D representation of major structural components such as beams, columns, and trusses. Detailed design and optimization of the steel structure nodes in the theoretical model. Including adjustments to the node geometry, connection methods, and material selection to ensure the stability and safety of the structure.

[0130] Use the automation function of BIM software to generate detailed design drawings and material lists from the theoretical model. These drawings and lists include the size, material grade, quantity and location of each component. Design verification is carried out through BIM software, including structural analysis, load testing and safety inspection to ensure that the design meets all relevant building codes and standards. Based on the verification results, the theoretical model is modified and optimized as necessary. This involves adjusting the size of components, changing the connection method or updating material specifications.

[0131] Among them, as a feasible implementation method, constructing a theoretical model of the steel structure includes the following steps:

[0132] Open the software in the multi-disciplinary model integration platform, such as Revit or Tekla Structures, and create a new project file. According to the specific requirements of the project, set the project's geographical location, material library, load conditions and other parameters. Establish a global coordinate system in the BIM software to ensure the correct position and orientation of all components.

[0133] Build column models: Use the Structural Column tool to draw a 3D model of the column based on the size and location of the design drawing. Specify the material properties of the column, such as steel grade, cross-sectional shape (rectangular, circular, etc.). Apply appropriate column foot and column capital details to ensure proper connection with the foundation and beams.

[0134] Model beams: Use the Structural Beam tool to create 3D models of primary and secondary beams based on your design requirements. Determine beam cross-sectional dimensions, material properties, and load carrying capacity. Place beams in the model, ensuring they are properly connected to columns and accounting for beam overlap and penetration.

[0135] Model a truss: Use the Truss or Steel Structure tools to create a 3D model of a truss, which is typically used for a roof or bridge structure. Set the geometry of the truss, including the size and spacing of the top chord, bottom chord, and web members. Specify the truss's materials and connection details, such as welds or bolts.

[0136] Detailed design: For node connections, use the "Node" tool to create detailed 3D node models, including bolts, welds, and other connectors. For complex nodes, collision checks and design optimization are required to ensure the integrity of the structure and the feasibility of construction.

[0137] Perform parametric modeling on each component and refine the geometry of each component, such as adding holes, grooves or other construction details. Apply construction process parameters, such as welding preparation, bolt hole reservation, etc. Finally, use the analysis tools of BIM software to perform structural analysis on the constructed model, such as load analysis, deflection analysis, etc. Adjust the size and position of the components according to the analysis results to ensure that the model meets the design specifications and safety requirements.

[0138] After the above work is completed, a detailed component list, material list and construction drawings are generated. The 3D view of the model is exported for construction briefing and reference during the construction process.

[0139] According to the design drawings and material lists generated by the above multi-disciplinary model integration platform, the measured three-dimensional coordinates of the control points of the steel structure segmented components are used to conduct component trial assembly by combining segmented three-dimensional trial assembly and horizontal three-dimensional trial assembly. The assembly is simulated in the computer to form a virtual assembly contour model of the segmented components.

[0140] In the specific implementation, 3D scanning technology or total station and other equipment are used to measure the control points of the steel structure segment components to obtain their accurate 3D coordinate data. In the BIM software environment of the multi-professional model integration platform, a 3D model of the steel structure segment components is created based on the measured 3D coordinate data, and the design drawings and material lists are imported to ensure the consistency between the model and the actual components. The virtual assembly function of the BIM software is used to perform a three-dimensional trial assembly of the steel structure segment components. This includes simulating the placement of the components in their designed positions in the computer and checking the accuracy of their spatial positioning and direction.

[0141] For some large or complex components, a horizontal stereo test assembly is performed in the BIM software to check the matching accuracy and connection details of the components. The assembly process is simulated in the computer to form a virtual assembly contour model of the segmented components. This can be done through the animation or simulation function of the BIM software to show the dynamic process of component assembly. At the same time, during the trial assembly process, relevant assembly data and problem feedback are collected to make necessary adjustments to the model. This may involve modifying the size of the component, adjusting the connection method, or optimizing the assembly sequence. Then, the virtual assembly contour model is analyzed for accuracy to ensure that the processing and assembly accuracy meets the design and construction requirements. This is achieved through the collision detection and dimensioning functions of the BIM software.

[0142] After obtaining the theoretical model of the in-depth design and the virtual assembly contour model in the above steps, the virtual assembly contour model is fitted and compared with the theoretical model of the in-depth design, the processing and assembly accuracy is checked and analyzed, and the adjustment information required for modification is obtained; after correction, modification and simulated assembly, the node connection complexity and precision requirements of the friction-type high-strength bolt connection manufacturing and processing of the component are checked to achieve the goal of zero error in the engineering installation.

[0143] The fitting and comparison between the virtual assembly outline model and the theoretical model of the in-depth design includes the following steps:

[0144] First, the virtual assembly contour model and the theoretical model are preliminarily aligned using the iterative closest point (ICP) algorithm in three-dimensional space. The ICP algorithm achieves alignment by minimizing the distance between two sets of point clouds, and its iterative formula can be expressed as: , where P is the original point cloud, P′ is the aligned point cloud, and t is the translation vector.

[0145] Based on the alignment, the best fitting plane or line is found by minimizing the sum of square errors. ), parameters of the fitted line Calculated by the following formula: ;

[0146] ;

[0147] ;

[0148] ;

[0149] Where n is the number of data points and ∑ represents the summation symbol. It refers to fitting a straight line equation on a two-dimensional plane: y=kx+b or z=kx+b, where k is the slope of the line and b is the y-intercept.

[0150] In the present invention, further, by constructing and calculating the determination coefficient R 2 , which measures the percentage of total variation explained by the model, thereby quantifying the quality of the fit.

[0151] R 2 The calculation formula is:

[0152] ;

[0153] Among them, SSres is the residual sum of squares and SStot is the total sum of squares.

[0154] The residual sum of squares is the sum of the squares of the differences between the actual observed values ​​and the values ​​predicted by the model. The calculation formula is:

[0155] ;

[0156] Where: y i is the actual value of the ith observation. is the predicted value of the ith observation, that is, the model is based on the independent variable x i The calculated value of the dependent variable y. n is the total number of observations.

[0157] The total sum of squares is the sum of the squares of the differences between the actual observations and the mean of the observations. It is calculated as:

[0158] ;

[0159] Where: y i is the actual value of the ith observation. is the mean of all observations. n is the total number of observations.

[0160] Through the above calculation, R 2 The closer the value is to 1, the better the model fit is.

[0161] As a further preferred solution, the size of the fitting error is evaluated by calculating the root mean square error (RMSE) and the mean absolute error (MAE):

[0162] ;

[0163] ;

[0164] Among them, y i is the actual value of the ith observation. is the predicted value for the ith observation.

[0165] After the fitting comparison is completed, the BIM software is used to check the accuracy and analyze the processing and assembly accuracy. This includes measuring the deviation of the steel structure connection points, such as the misalignment of the rod axis at the node, the misalignment of the interface and other key parameters to ensure that the requirements of the "Code for Acceptance of Steel Structure Construction Quality" GB50205 are met. According to the adjustment information obtained, the virtual assembly model is corrected and modified as necessary. After the modification, the simulation assembly is re-performed to verify whether the adjustment meets the requirements. This step requires multiple iterations until all connection points reach the expected accuracy.

[0166] For the nodes where friction-type high-strength bolts are connected and processed, special inspections are carried out to ensure that the complexity and precision requirements of the connection are met. This includes detailed design and simulation of the connectors to verify their feasibility in actual construction. After the simulated assembly is confirmed to be correct, a final verification is carried out to ensure that the assembly accuracy of all components meets the zero error requirement. This step is the key to ensuring construction quality and can reduce rework and adjustments during on-site construction.

[0167] The matters not described in detail in the present invention are all known technologies to those skilled in the art.

[0168] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A BIM-based building steel structure in-depth design and virtual pre-assembly method, characterized in that: The steps include: S1, build a BIM system server, integrate professional model software in the BIM system, import different professional models into the BIM system according to the construction logic and sequence using the international unified format IFC, and form a multi-professional model integration platform; S2, creating a three-dimensional model of the steel structure in the multi-discipline model integration platform, inputting design parameters and engineering data to form an accurate geometric representation; Based on the three-dimensional model, a parametric modeling method is used to refine the geometric shape of each component; S3, in the multi-professional model integration platform, the relative positions and overlaps between the models are checked by geometric algorithms, associations and dependencies are established between the models, the models are automatically updated according to the update amplitude formula, and a change propagation mechanism is constructed to achieve integration and collision detection of different professional models; S4, applying each professional model software in the multi-professional model integration platform to perform in-depth design to obtain a theoretical model of in-depth design; Optimize the steel structure nodes in the theoretical model and generate detailed design drawings and material lists; S5, according to the design drawings and material lists generated by the multi-professional model integration platform, the measured three-dimensional coordinates of the control points of the steel structure segmented components are combined with the segmented three-dimensional trial assembly and the horizontal three-dimensional trial assembly to perform component trial assembly, and simulate the assembly in a computer to form a virtual assembly contour model of the segmented components; S6, fitting and comparing the virtual assembly contour model with the theoretical model of the in-depth design, checking and analyzing the processing and assembly accuracy, and obtaining the adjustment information required for modification; after correction, modification and simulated assembly, checking the accuracy of the node connection complexity and precision requirements of the friction type high-strength bolt connection manufacturing and processing of the component, so that the engineering installation can achieve the purpose of zero error; In step S2, the parametric modeling method includes using an intelligent value assignment algorithm SCIA for assigning values ​​to complex spatial structures, bending and torsion components, and special-shaped components. The intelligent value assignment algorithm SCIA includes the steps of collecting and preprocessing design parameters and engineering data, classifying and grouping data, extracting key features from data, calculating feature weights, establishing an intelligent value assignment model, using a back-propagation algorithm to update weights and biases, and performing rationality checks and optimization on the value assignment results. The design parameters and engineering data include three-dimensional coordinates, curvature, torsion angle, material, and strength requirements of complex spatial structures, bending and torsion components, and special-shaped components. The preprocessing of these data includes cleaning and standardization to remove abnormal values ​​and noise data; The extracting key features from the data includes: extracting node coordinates, connection methods, and component length ratios from complex spatial structures, extracting curvature change rates, torsion angle ranges, and cross-sectional shape parameters from bending and torsion components, and extracting geometric shape parameters, material parameters, and strength parameters from special-shaped components; and constructing a feature matrix X from the key features; The calculation of feature weights includes calculating the covariance matrix C of the feature matrix: ； Where X is the feature matrix, is the transposed matrix of X, n is the number of samples; Perform eigenvalue decomposition on the covariance matrix C and obtain the eigenvalue and the eigenvector : = ; Calculate the weight of each feature : ； Where m is the number of features; j is the index variable that traverses all eigenvalues; The weight of each feature Construct the weight matrix W; The intelligent assignment model is established, and its calculation formula is expressed as: ； Where x is the input feature vector, W is the weight matrix, b is the bias vector, f is the activation function, and y is the output assignment result; The back propagation algorithm is used to update weights and biases using the following mean square error function: ； In the formula, is the actual value, is the model output value, and n is the number of samples.

2. The BIM-based building steel structure in-depth design and virtual pre-assembly method according to claim 1 is characterized in that: The professional model software includes one or a combination of the engineering basic data analysis system, Luban Rebar 2017, Luban Civil Engineering 2017, Luban Installation 2017, Luban Field Layout 2017, Revit 2016, Tekla Structures 20.1, mobile APP-Luban View 4.1.0, SketchUp 2015, LUMION, and Pinming 3D construction planning software; the engineering basic data analysis system includes at least one or a combination of Luban Explorer, Luban Govern, Luban Plan, and Luban Works; The BIM system server is connected to an enterprise PDS main server, several enterprise PDS sub-servers, several mobile terminals, and several computer terminals through at least one switch, and the switch is connected to the Internet through the enterprise main router.

3. The BIM-based building steel structure in-depth design and virtual pre-assembly method according to claim 1 is characterized in that: In step S3, checking the relative position and overlap between models by using a geometric algorithm in the multi-professional model integration platform includes the following steps: In the multi-disciplinary model integration platform, each model is spatially positioned to ensure that each model is in its correct position; the reference point of each model is located using a spatial coordinate system to ensure that the spatial relationship between models is accurate; The relative position and overlap between the models are checked by a geometric algorithm to ensure that there is no geometric conflict; in the geometric algorithm, the distance between the models is defined As a model and Model The distance between the closest points is calculated as: ； In the formula, and The models are and For the points in , ||pq|| is the Euclidean distance between the two points; Establish associations and dependencies between models so that changes to one model can trigger automatic updates to related models; including: defining association metrics To measure the model and Model The correlation between them is calculated as: ； Where k is a constant used to adjust the sensitivity of the association measure to distance, and e is the base of the natural logarithm; The higher the value, the better the model and Model The stronger the correlation between them; When the model When a change occurs, automatically calculate its correlation metrics with all other models , and automatically update the model according to the update amplitude formula: ； In the formula, Δ j It is a model The update range, is a proportional constant with a value range of 1-2; is the update threshold, triggering the model Updates; Build a change propagation mechanism to ensure that changes to a model can be propagated consistently throughout the project model, using the change propagation coefficient formula: ； In the formula, is a proportional constant ranging from 1 to 2, used to adjust the intensity of change propagation; It is a model The sum of association measures with all other models.

4. A BIM-based building steel structure in-depth design and virtual pre-assembly system, used to implement the BIM-based building steel structure in-depth design and virtual pre-assembly method as claimed in any one of claims 1 to 3, characterized in that: The system includes at least: BIM system server, used to integrate professional model software in the BIM system to form a multi-professional model integration platform; The data acquisition module is used to collect various design parameters and engineering data of complex spatial structures, bending and torsion components and special-shaped components, including the three-dimensional coordinates, curvature, torsion angle, material, and strength requirements of the components, and clean and standardize these data to remove abnormal values ​​and noise data; A data classification module is used to classify and group the data processed by the data acquisition module according to the type and function of the component; A feature extraction module is used to extract key features from the data processed by the data classification module; The weight calculation module is used to calculate the feature weight, including calculating the covariance matrix C of the feature matrix, performing eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors, and then calculating the weight of each feature; Intelligent assignment module, which is used to establish an intelligent assignment model and use the back propagation algorithm to update weights and biases to minimize the error between the model output and the actual value, while also performing rationality checks and optimization on the assignment results; The detailed design module is used to apply various professional model software to carry out detailed design, obtain the theoretical model of detailed design, optimize the steel structure nodes in the theoretical model, and generate detailed design drawings and material lists; The virtual pre-assembly module is used to simulate the assembly in the computer to form a virtual assembly outline model of the segmented components according to the generated design drawings and material lists, and to fit and compare the virtual assembly outline model with the theoretical model of the in-depth design.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed, the BIM-based building steel structure in-depth design and virtual pre-assembly method as described in any one of claims 1 to 3 is implemented.

6. An electronic device, characterized in that: include: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores a computer program, which, when executed by the at least one processor, enables the at least one processor to implement the BIM-based building steel structure in-depth design and virtual pre-assembly method as described in any one of claims 1-3.

Images