Steel truss girder digital pre-assembly method and system based on three-dimensional laser scanning

Abstract

The invention relates to a steel truss girder digital pre-assembly method and system based on three-dimensional laser scanning, and belongs to the technical field of constructional engineering. The method comprises the following steps: calculating a three-dimensional coordinate and a scanning angle of a scanning observation station based on a steel truss girder design model and field environment point cloud data to obtain observation station layout position parameters; controlling a three-dimensional laser scanner to scan the steel truss girder section by section according to the arrangement position parameters of the observation station, and acquiring original point cloud data of each section of the steel truss girder; performing spatial alignment and noise reduction processing on the original point cloud data to obtain standardized point cloud data; based on the standardized point cloud data, feature parameters of all the segments are segmented and extracted, digital splicing is carried out, and a steel truss girder three-dimensional point cloud model is generated; based on the steel truss girder design model and the steel truss girder three-dimensional point cloud model, the displacement deviation value and the angle deviation value between the segments are calculated, and an assembly error result is obtained; and generating a construction correction parameter set based on the assembly error result. The assembling precision of the steel truss girder can be improved.

Description

Technical Field

[0001] The present application relates to the field of construction engineering technology, and in particular to a digital pre-assembly method and system for steel trusses based on three-dimensional laser scanning. Background Art

[0002] In the traditional steel structure construction process, steel trusses are widely used in the construction of large buildings such as bridges, factories, warehouses, etc. due to their good load-bearing capacity and structural stability. Steel trusses are usually composed of multiple segments or units, which are prefabricated in the factory and then transported to the construction site for assembly. The accuracy of the assembly process is crucial because any assembly error may directly affect the stability, durability and safety of the entire steel structure. For example, if the connection nodes of the steel trusses deviate from the design requirements, it may lead to uneven stress distribution of the structure, increase the fatigue load of the structure, and may even cause structural failure, causing serious safety hazards.

[0003] At present, in the actual assembly process of steel trusses, due to the errors between the design model and the actual construction, especially the deviations of the node positions and angles, the assembly may be inaccurate, which not only affects the safety of the structure, but also may cause engineering rework, increase costs and time. Therefore, how to automatically calculate and correct these deviations through digital means to improve the assembly accuracy of steel trusses is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In order to improve the assembly accuracy of steel trusses, the present application provides a digital pre-assembly method and system for steel trusses based on three-dimensional laser scanning.

[0005] In a first aspect, the present application provides a digital pre-assembly method for steel trusses based on three-dimensional laser scanning, which adopts the following technical solutions: A digital pre-assembly method for steel trusses based on three-dimensional laser scanning, the method comprising: Based on the steel truss design model and the on-site environmental point cloud data, the three-dimensional coordinates and scanning angles of the scanning station are calculated through the spatial coverage algorithm to obtain the station layout position parameters; According to the measurement station layout position parameters, control the three-dimensional laser scanner to scan the steel truss segment by segment, and collect the original point cloud data of each segment of the steel truss; Performing spatial alignment and noise reduction processing on the original point cloud data to obtain standardized point cloud data; Based on the standardized point cloud data, the characteristic parameters of each segment are segmented and extracted, and digital splicing is performed to generate a three-dimensional point cloud model of the steel truss; Based on the steel truss design model and the three-dimensional point cloud model of the steel truss, the displacement deviation and the angle deviation between the segments are calculated to obtain the assembly error result; A construction correction parameter set is generated based on the assembly error result.

[0006] By adopting the above technical solutions, combining the design model with the actual data on site, using 3D laser scanning technology, efficient spatial coverage algorithm, accurate point cloud data processing and error analysis, an efficient steel truss pre-assembly solution is realized. Through accurate station layout, standardized point cloud data processing, comprehensive error calculation and generation of correction parameter sets, the high precision and high efficiency of steel trusses can be ensured during the assembly process. This technical solution not only improves the quality of engineering construction, but also effectively reduces costs and time, and optimizes the overall process of steel structure construction.

[0007] Optionally, based on the steel truss design model and the on-site environmental point cloud data, the steps of calculating the three-dimensional coordinates and scanning angles of the scanning station by a spatial coverage algorithm to obtain the station layout position parameters include: Based on the steel truss design model, the length of each segment of the steel truss and the distribution characteristics of the connection nodes are obtained, and the maximum effective scanning distance between adjacent measuring stations is calculated; Initialize the station position and scanning angle, and configure the spatial coverage constraints; Based on the spatial coverage constraint and the maximum effective scanning distance, the layout positions and scanning angles of the measuring stations are iteratively optimized to minimize the number of measuring stations and ensure that all sections of the steel truss are covered, thereby obtaining optimized measuring station layout position parameters.

[0008] By adopting the above technical solution, based on the steel truss design model and the on-site environmental point cloud data, combined with the spatial coverage algorithm, the optimization of the measurement station layout is achieved. By minimizing the number of required measurement stations while ensuring that all segments can be fully scanned, it not only reduces costs and time, but also improves construction accuracy and efficiency, providing an effective solution for the digital scanning of steel trusses.

[0009] Optionally, based on the standardized point cloud data, the steps of segmenting and extracting characteristic parameters of each segment and digitally splicing the segment to generate a three-dimensional point cloud model of the steel truss include: Based on the standardized point cloud data, the bolt hole center points and flange edge line features of each segment of the steel truss are segmented and extracted; Taking the center point of the bolt hole as the registration reference, locating the segment splicing direction according to the flange edge line features, and calculating the rigid body transformation matrix between each pair of adjacent segments based on the ICP algorithm; Registering and splicing each pair of adjacent segment point clouds according to the rigid body transformation matrix; The three-dimensional model is reconstructed according to the aligned and spliced ​​adjacent segment point clouds to generate the three-dimensional point cloud model of the steel truss, and the actual position parameters of each segment in the global coordinate system are output.

[0010] By adopting the above technical solutions, based on feature constraints and algorithm optimization, it can effectively resist noise and occlusion interference, greatly reducing the need for manual intervention. The final generated three-dimensional point cloud model and posture parameters can be seamlessly connected with BIM, CNC machine tools and other systems, realizing the digitalization of the entire chain from design to construction and inspection, providing a high-precision, low-cost and standardized solution for steel structure projects.

[0011] Optionally, the step of calculating the rigid body transformation matrix between each pair of adjacent segments based on the ICP algorithm includes: Get the bolt hole center point spacing and flange edge line direction vector between each pair of adjacent segments; Based on a preset spacing standard, a set of matching point pairs whose bolt hole spacing errors are less than a preset error threshold are selected, and an initial angle compensation value is calculated according to the flange edge line direction vector; According to the spacing deviation set of the bolt hole matching point pair set, a weighted residual function is constructed in combination with the flange edge line direction vector; The rotation parameters are initialized based on the weighted residual function and the initial angle compensation value, the optimal rotation matrix and translation vector are solved by the singular value decomposition iterative optimization algorithm, the calculation is stopped when the error change rate or the number of iterations meets the termination condition, and the rigid body transformation matrix is ​​output.

[0012] By adopting the above technical solutions, the improved ICP algorithm based on feature constraints has achieved high precision and high efficiency in steel truss segment registration. Through dynamic weighted residual function and robustness optimization, the ability to resist noise and interference has been significantly improved, which can not only meet the requirements of high-precision assembly, but also be suitable for complex construction sites and support the rapid construction of large-scale projects.

[0013] Optionally, the step of calculating the displacement deviation and angle deviation between segments based on the steel truss design model and the steel truss three-dimensional point cloud model to obtain the assembly error result includes: Converting the CAD geometric data of the steel truss design model into a point cloud format to generate a theoretical point cloud model; Obtaining the design coordinates of the center points of the bolt holes of each segment in the theoretical point cloud model and the actual coordinates in the three-dimensional point cloud model of the steel truss; Calculate the Euclidean distance deviation between the design coordinates and the actual coordinates to obtain the displacement deviation of each segment; Obtaining a theoretical flange edge line direction vector of each pair of adjacent segments in the theoretical point cloud model and an actual flange edge line direction vector in the three-dimensional point cloud model of the steel truss; Calculating the angle deviation between the theoretical flange edge line direction vector and the actual flange edge line direction vector to obtain the angle deviation of each pair of adjacent segments; An assembly error result is generated according to the displacement deviation and the angle deviation.

[0014] By adopting the above technical solution, the error comparison between the steel truss design model and the actual 3D point cloud model is carried out, and the displacement deviation and angle deviation of each segment are accurately calculated, thereby generating assembly error data. This technical solution effectively identifies assembly errors caused by construction, measurement or material deviations, and provides an accurate basis for adjustment, ensuring high precision and high efficiency of the steel truss assembly process.

[0015] Optionally, after the step of generating a construction correction parameter set based on the assembly error result, the method further includes: Angle correction instructions and displacement correction instructions are generated according to the construction correction parameter set; wherein the angle correction instructions are used to adjust the lifting angle of each steel truss segment to perform angle compensation, and the displacement correction instructions are used to adjust the alignment deviation of each steel truss bolt hole to perform displacement adjustment.

[0016] By adopting the above technical solution, the installation accuracy of each segment of the steel truss can be accurately adjusted. The angle correction instruction ensures that the lifting angle of each segment is compensated, thereby avoiding angle errors during the assembly process, while the displacement correction instruction is used to adjust the alignment deviation of the bolt holes to ensure that each segment can be accurately docked. These correction instructions provide clear operating instructions for on-site construction, greatly improving the accuracy and construction efficiency of steel truss assembly and reducing the risk of rework caused by errors.

[0017] In a second aspect, the present application provides a digital pre-assembly system for steel trusses based on three-dimensional laser scanning, which adopts the following technical solutions: A digital pre-assembly system for steel trusses based on three-dimensional laser scanning, comprising: The station layout module is used to calculate the three-dimensional coordinates and scanning angles of the scanning station through the spatial coverage algorithm based on the steel truss design model and the on-site environmental point cloud data, and obtain the station layout position parameters; A point cloud data acquisition module, used to control the three-dimensional laser scanner to scan the steel truss segment by segment according to the measurement station layout position parameters, and acquire the original point cloud data of each segment of the steel truss; A point cloud data processing module is used to perform spatial alignment and noise reduction on the original point cloud data to obtain standardized point cloud data; A segment splicing module is used to segment and extract characteristic parameters of each segment based on the standardized point cloud data and perform digital splicing to generate a three-dimensional point cloud model of the steel truss; An assembly error analysis module is used to calculate the displacement deviation and angle deviation between segments based on the steel truss design model and the three-dimensional point cloud model of the steel truss to obtain an assembly error result; A correction parameter determination module is used to generate a construction correction parameter set based on the assembly error result.

[0018] Optionally, the system also includes a correction instruction generation module, which is used to generate angle correction instructions and displacement correction instructions based on a construction correction parameter set; the angle correction instructions are used to adjust the lifting angle of each steel truss segment for angle compensation, and the displacement correction instructions are used to adjust the alignment deviation of each steel truss bolt hole for displacement adjustment.

[0019] In a third aspect, the present application provides a computer device, which adopts the following technical solution: A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the methods in the first aspect.

[0021] In summary, the present application includes at least one of the following beneficial technical effects: combining the steel truss design model and the on-site environmental point cloud data, the digital pre-assembly of the steel truss is realized by using three-dimensional laser scanning technology. By accurately calculating the layout position of the measuring station, performing segmented scanning and collecting point cloud data, a high-precision three-dimensional point cloud model of the steel truss is generated after spatial alignment, noise reduction processing and digital splicing. The displacement and angular deviation between the segments are further calculated to obtain the assembly error results, and a construction correction parameter set is generated to ensure high precision and high efficiency of the assembly process. This technical solution significantly improves the accuracy of steel truss assembly, reduces manual measurement errors and rework, and improves construction efficiency. It has broad application significance, especially in the precise control of pre-assembly and on-site construction of complex steel structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a first process diagram of a method for digital pre-assembly of steel trusses based on three-dimensional laser scanning in one of the embodiments of the present application.

[0023] Figure 2 This is a segmented scan of a steel truss according to one of the embodiments of the present application.

[0024] Figure 3 This is a three-dimensional modeling diagram of the pre-assembled steel trusses of one embodiment of the present application.

[0025] Figure 4It is a second flow chart of a method for digital pre-assembly of steel trusses based on three-dimensional laser scanning in one of the embodiments of the present application.

[0026] Figure 5 It is a third process diagram of a method for digital pre-assembly of steel trusses based on three-dimensional laser scanning in one of the embodiments of the present application.

[0027] Figure 6 It is a fourth process diagram of a method for digital pre-assembly of steel trusses based on three-dimensional laser scanning in one of the embodiments of the present application.

[0028] Figure 7 It is a fifth process diagram of a method for digital pre-assembly of steel trusses based on three-dimensional laser scanning in one of the embodiments of the present application. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-7 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0030] The embodiment of the present application discloses a digital pre-assembly method for steel trusses based on three-dimensional laser scanning.

[0031] Reference Figure 1 , a digital pre-assembly method of steel trusses based on three-dimensional laser scanning, comprising: Step S101, based on the steel truss design model and the on-site environment point cloud data, the three-dimensional coordinates and scanning angles of the scanning station are calculated by a spatial coverage algorithm to obtain the station layout position parameters; The steel truss design model is a digital model generated by CAD software, which contains the geometric dimensions and node coordinates of the steel truss. This model serves as the benchmark data for pre-assembly accuracy to ensure that it can be compared and verified with the actual structural data during the assembly process. The on-site environment point cloud data includes terrain, obstacles, construction equipment locations, etc.

[0032] In the embodiment of the present application, the station layout scheme must meet the requirements of covering all sections of the steel truss and minimizing the number of stations. The spatial coverage algorithm calculates the optimal scanning station position and angle by analyzing the design model and the point cloud data of the on-site environment. The algorithm takes into account the layout of each section of the steel truss and the visualization conditions of the on-site environment to ensure that each component of the steel truss can be covered to the greatest extent during measurement.

[0033] It can be understood that by calculating the measurement station layout position parameters, the number and position of scanning stations are optimized, making the measurement coverage area more comprehensive, while reducing the number of measurement stations as much as possible, reducing costs and time.

[0034] Step S102, according to the measurement station layout position parameters, control the three-dimensional laser scanner to scan the steel truss segment by segment, and collect the original point cloud data of each segment of the steel truss; Among them, the 3D laser scanner starts to scan the steel truss in sections according to the calculated measuring station position parameters. Each scanning area covers a part of the steel truss segment, and the point cloud data is collected by the laser scanner. The point cloud data is a set of discrete points that describe the surface morphology of an object in three-dimensional space, and each point has spatial coordinate information. Point cloud data is the basis for digital description of the physical morphology of the steel truss, and can be used for subsequent modeling and error analysis.

[0035] Reference Figure 2 During the scanning process, the scanner will collect data segment by segment according to the predetermined scanning angle and position. Figure 2 The following is a segmented scan of a steel truss. The scanning process needs to take into account the complex shape and environmental conditions of the steel truss, so it is usually scanned multiple times from different angles to ensure that each segment can be fully covered. The accuracy and density of the scan are crucial for subsequent data processing. High-density point cloud data can more accurately restore the three-dimensional shape of the steel truss, thereby providing data support for accurate assembly.

[0036] Step S103, performing spatial alignment and noise reduction processing on the original point cloud data to obtain standardized point cloud data; Among them, since the data collected by the 3D laser scanner at different measuring stations may have errors, and the original point cloud data may contain noise (such as caused by environmental factors or scanning errors), spatial alignment and noise reduction processing are required in the data processing stage. The point cloud data from different measuring stations are aligned according to a unified coordinate system to ensure that all point cloud data are in the same position when merged. For example, the ICP algorithm can be used to minimize the errors between different point cloud data sets so that the data can be merged as much as possible.

[0037] At the same time, noise reduction processing is to remove inaccurate points in the data caused by environmental interference (such as rain, dust, etc.) and problems with the scanning equipment itself. Noise reduction technology includes the application of filters to remove outliers and data that does not conform to physical laws to ensure the purity and accuracy of the data set. This process can effectively improve the quality of the data, making subsequent splicing and modeling work more reliable.

[0038] Step S104, based on the standardized point cloud data, segment and extract characteristic parameters of each segment and perform digital splicing to generate a three-dimensional point cloud model of the steel truss; Among them, after the data processing is completed, the standardized point cloud data needs to be segmented and feature extracted. Steel trusses usually have multiple segments, and the shape and features of each segment need to be extracted. These features include information such as the center point of the bolt hole and the edge line features of the flange. Through point cloud segmentation technology, the point cloud data of different segments are separated to obtain an independent point cloud data set for each segment. These segmented point cloud data need to be digitally spliced.

[0039] Reference Figure 3 The stitching process involves accurately aligning multiple scanned point cloud data sets to construct a complete 3D point cloud model. Figure 3 The figure shows the 3D modeling of the steel truss after pre-assembly. During the assembly process, commonly used algorithms include ICP algorithm and other point cloud registration algorithms to ensure that there is no significant error at the joints between point clouds. The final generated 3D point cloud model of the steel truss can highly restore the actual shape of the steel truss, providing data support for subsequent accuracy verification and error correction.

[0040] Step S105, based on the steel truss design model and the steel truss three-dimensional point cloud model, calculating the displacement deviation and angle deviation between the segments to obtain an assembly error result; Among them, error calculation is performed based on the design model of the steel truss and the point cloud model obtained by 3D laser scanning. The design model defines the ideal position and angle of each segment of the steel truss, while the point cloud model reflects the results obtained by actual measurement. By comparing the coordinate differences and angle differences between the design model and the point cloud model, the displacement deviation and angle deviation between each segment can be calculated. These errors reflect the deviations caused by factors such as construction, measurement errors or material deviations during the assembly process.

[0041] Specifically, the error calculation process usually involves comparing the design model with the point cloud data segment by segment, and calculating the displacement and angle error of each node. The displacement deviation refers to the deviation in spatial position between the two models, while the angle deviation reflects the rotation error of the structural segment. Through this process, the error distribution of each segment of the steel truss during assembly can be clearly seen.

[0042] Step S106, generating a construction correction parameter set based on the assembly error result.

[0043] Specifically, based on the assembly error results, the system will generate a detailed set of construction correction parameters, including coordinate corrections and angle compensation values. The coordinate corrections represent the position adjustment of each segment during the installation process, while the angle compensation is to correct the installation error caused by angle deviation. These correction parameters will serve as an operating guide on the construction site, instructing construction personnel on how to adjust the specific installation position and angle of each segment of the steel truss to ensure that the design accuracy is achieved during the assembly process.

[0044] In the above implementation, the design model is combined with the actual data on site, and an efficient steel truss pre-assembly solution is realized by using 3D laser scanning technology, efficient spatial coverage algorithm, accurate point cloud data processing and error analysis. Through accurate station layout, standardized point cloud data processing, comprehensive error calculation and generation of correction parameter sets, the high precision and high efficiency of steel trusses can be ensured during the assembly process. This technical solution not only improves the quality of engineering construction, but also effectively reduces costs and time, and optimizes the overall process of steel structure construction.

[0045] Reference Figure 4 As an implementation method of step S101, based on the steel truss design model and the on-site environment point cloud data, the three-dimensional coordinates and scanning angles of the scanning station are calculated by a spatial coverage algorithm to obtain the station layout position parameters, including: Step S201, obtaining the length of each segment and the connection node distribution characteristics of the steel truss based on the steel truss design model, and calculating the maximum effective scanning distance between adjacent measuring stations; Among them, in the measurement of steel trusses, the length of each segment and the distribution characteristics of the connection nodes are the key factors in determining the layout of the measuring stations. The steel truss design model usually contains the geometric dimensions and node coordinates of each segment. This information helps to understand the relative position and connection method between the segments. The maximum effective scanning distance refers to the farthest distance that a measuring station can effectively cover, which is usually determined by the maximum scanning range, viewing angle and actual position of the segment of the laser scanner. In order to avoid scanning blind spots caused by the distance between the measuring stations being too far, the maximum effective scanning distance between adjacent measuring stations must be calculated. Specifically, the maximum effective scanning distance should be calculated based on the distance and connection method between each pair of segments to ensure that there are no dead angles between each measuring station.

[0046] Step S202, initializing the station position and scanning angle, and configuring the spatial coverage constraint conditions; Among them, according to the geometric structure of the steel truss design model and the on-site environmental data, the station position and scanning angle are initially set. The station position should be selected where the maximum scanning area can be covered, generally at the two ends or main connection nodes of the steel truss. The configuration of the scanning angle takes into account the field of view angle of the scanner of the station, and the scanning angle should usually be determined according to the relative position between adjacent segments to ensure that there is no scanning blind spot.

[0047] At the same time, spatial coverage constraints need to be configured. The most common constraint is that the scanning areas of every two adjacent measuring stations have a certain overlap rate, which is usually set to 30% to 50%. The purpose of this is to ensure that each segment is scanned multiple times, obtain sufficient measurement data from different angles, improve data accuracy, and ensure no blind spots.

[0048] Step S203, based on the spatial coverage constraint and the maximum effective scanning distance, iteratively optimize the layout position and scanning angle of the measuring station, minimize the number of measuring stations and ensure that all sections of the steel truss are covered, and obtain the optimized measuring station layout position parameters.

[0049] The goal of iterative optimization is to minimize the number of measuring stations while ensuring that all segments of the steel truss are effectively scanned and that the overlap of the scanning area meets the predetermined requirements. This optimization process can use iterative algorithms, such as genetic algorithms, particle swarm optimization algorithms (PSO) or other global optimization methods. In each iteration, based on the scanning area of ​​the current measuring station, it is calculated whether all segments are covered. If there are segments that are not covered, the position of the measuring station or the scanning angle is adjusted until the coverage requirements are met and the number of measuring stations is minimized.

[0050] In addition, during the optimization process, it is necessary to ensure that the scanning angle and position of each measuring station can meet the requirements of the maximum effective scanning distance, and at the same time take into account the on-site environmental factors (such as obstacles and measuring station visibility) to dynamically adjust the measuring station layout. The final generated measuring station layout plan will contain the specific three-dimensional coordinates and scanning angles of each measuring station.

[0051] For example, assuming that 5 measuring stations are initially deployed, the number of measuring stations is reduced to 4 through iterative adjustment during the optimization process, and it is ensured that each segment can be fully scanned and the overlapping area of ​​each pair of adjacent measuring stations reaches 30%. This optimization scheme reduces the number of measuring stations and improves the scanning efficiency.

[0052] In the above implementation, based on the steel truss design model and the on-site environmental point cloud data, combined with the spatial coverage algorithm, the optimization of the measurement station layout is achieved. By minimizing the number of required measurement stations while ensuring that all segments can be fully scanned, it not only reduces costs and time, but also improves construction accuracy and efficiency, providing an effective solution for the digital scanning of steel trusses.

[0053] Reference Figure 5 As an implementation method of step S104, based on the standardized point cloud data, the steps of segmenting and extracting characteristic parameters of each segment and digitally splicing to generate a three-dimensional point cloud model of a steel truss include: Step S301, based on the standardized point cloud data, segment and extract the bolt hole center points and flange edge line features of each segment of the steel truss; Specifically, important features related to the steel truss structure are extracted from the standardized point cloud data: the center point of the bolt hole and the edge line features of the flange. First, the extraction of the center point of the bolt hole is achieved through point cloud clustering and cylindrical surface fitting. Point cloud clustering uses the Euclidean distance clustering algorithm (such as DBSCAN) to separate the dense points in the bolt hole area and exclude background noise and irrelevant point cloud data. Then, the RANSAC (random sampling consensus) algorithm is used to fit the bolt hole with a cylindrical surface, calculate the axis equation of the bolt hole, and determine the exact position of the bolt hole by solving the intersection of the axis and the segment end face.

[0054] In addition, for the feature extraction of the flange edge line, the local normal vector of the point cloud data is first calculated, and the boundary points are detected based on the mutation of the normal vector. When the angle of the boundary point changes by more than 30°, it is regarded as the flange edge. Then, the least squares method is used to fit these boundary points to obtain the straight line or broken line of the flange edge, and the direction vector is further extracted.

[0055] Step S302, using the center point of the bolt hole as the alignment reference, locating the segment splicing direction according to the flange edge line features, and calculating the rigid body transformation matrix between each pair of adjacent segments based on the ICP algorithm; After extracting the bolt hole center point and flange edge line features from the point cloud data, the next step is to align adjacent segments through feature matching. First, the bolt hole center point is used as a hard constraint to build the correspondence between segments. In this way, the one-to-one correspondence between each registered reference point (bolt hole center) between the two segments is ensured.

[0056] Next, based on the improved ICP algorithm, the weighted least squares method is used for optimization to calculate a rigid body transformation matrix (including the rotation matrix R and the translation vector T) to align the source point cloud data with the target point cloud data. During the optimization process, the errors between the bolt holes and the flange direction errors are given different weights to ensure that the accuracy of the bolt hole docking is prioritized.

[0057] Exemplarily, in the registration process of adjacent segments A and B, after 8 iterations, the error converges to 0.4 mm, and the corresponding rotation matrix R and translation vector T are output.

[0058] Step S303, registering and splicing each pair of adjacent segment point clouds according to the rigid body transformation matrix; Among them, each source point cloud is transformed into the coordinate system of the target point cloud through the rotation matrix R and the translation vector T to achieve accurate registration. At this time, the overlapping area between adjacent segments needs to be optimized to eliminate redundant data and retain key information. For the overlapping area, voxel grid downsampling (voxel size is 2mm) is used to balance the point cloud density, remove excessive duplicate points, and ensure that high-precision feature points are retained.

[0059] Step S304, reconstructing a three-dimensional model based on the aligned and spliced ​​adjacent segment point clouds, generating a three-dimensional point cloud model of the steel truss, and outputting the actual position and posture parameters of each segment in the global coordinate system.

[0060] Specifically, after completing the point cloud registration and stitching, the next step is to generate a 3D model using a surface reconstruction algorithm. For example, Poisson reconstruction is an implicit surface generation method based on point cloud normal vectors, which can convert the stitched point cloud data into a closed triangular mesh model. In this process, the surface details of the bolt holes and flange edge areas are optimized through local subdivision to ensure the accuracy of key structural parts.

[0061] Furthermore, after the 3D model is generated, the actual pose parameters of each segment need to be calculated. This includes mapping the endpoint coordinates of each segment to the global coordinate system, calculating their positions in the global coordinate system (Xg, Yg, Zg), and obtaining the segment rotation angles (α, β, γ) through Euler angle calculation.

[0062] In the above implementation, based on feature constraints and algorithm optimization, it can effectively resist noise and occlusion interference, greatly reducing the need for manual intervention. The final generated three-dimensional point cloud model and posture parameters can be seamlessly connected with BIM, CNC machine tools and other systems, realizing full-chain digitization from design to construction and testing, providing high-precision, low-cost and standardized solutions for steel structure projects.

[0063] Reference Figure 6 As an implementation of step S302, the step of calculating the rigid body transformation matrix between each pair of adjacent segments based on the ICP algorithm includes: Step S401, obtaining the bolt hole center point spacing and flange edge line direction vector between each pair of adjacent segments; Among them, in the assembly process of steel trusses, bolt holes and flange edges are key connection points, and the center point of the bolt hole and the direction vector of the flange edge line are used to accurately determine the registration relationship between segments. First, the point cloud data of the steel truss segment is obtained using 3D laser scanning technology, and the bolt hole area is separated from the point cloud using a density clustering algorithm (such as DBSCAN); then, the RANSAC (random sampling consistency) algorithm is used to fit the bolt hole cylindrical surface, calculate the equation of the bolt hole axis, and calculate the spacing between the center points of the bolt holes of adjacent segments to obtain the actual spacing data.

[0064] For the extraction of flange edge lines, the boundary points in the point cloud are first fitted using the least squares method to calculate the flange direction vector of the steel truss segment. This direction vector reflects the main direction of the flange and is the basis for calculating the segment splicing direction and rotation.

[0065] Step S402, based on a preset spacing standard, screening a set of matching point pairs whose bolt hole spacing errors are less than a preset error threshold, and calculating an initial angle compensation value according to the flange edge line direction vector; Among them, through the calculation of the distance between the center points of the bolt holes and the flange direction vector, the next step is to screen out valid matching point pairs. First, according to the preset design spacing (for example, 200mm) and the allowable error threshold (such as ±2mm), by calculating the actual spacing between the center points of the bolt holes of adjacent segments, a set of matching point pairs that meet the requirements is screened out. The purpose of this step is to filter out abnormal points caused by point cloud noise or occlusion to ensure the high quality of the registration process.

[0066] In addition, the initial angle compensation value is calculated according to the angle between the flange direction vectors. The compensation value can be used to adjust the rotation angle, thereby reducing the number of iterations required in the subsequent registration process and further improving the calculation efficiency.

[0067] Step S403, constructing a weighted residual function based on the spacing deviation set of the bolt hole matching point pair set and the flange edge line direction vector; Among them, after feature matching, a weighted residual function is constructed to further optimize the registration process. The weight of each matching point pair is dynamically assigned according to the deviation between the actual distance of the bolt hole center point and the designed value. Specifically, matching point pairs with large deviations will be assigned lower weights to reduce their impact on the final registration results. This can effectively suppress the negative impact caused by outliers or noise.

[0068] In addition, the flange direction alignment constraint needs to be added to the residual function, which can ensure the angle consistency between segments during splicing and avoid the accumulation of angle errors. By combining the distance error and the direction error, a weighted residual function is constructed to make the registration process more accurate.

[0069] Step S404, initialize the rotation parameters based on the weighted residual function and the initial angle compensation value, solve the optimal rotation matrix and translation vector through the singular value decomposition iterative optimization algorithm, stop the calculation when the error change rate or the number of iterations meets the termination condition, and output the rigid body transformation matrix.

[0070] Among them, after constructing the weighted residual function, the optimal rigid body transformation matrix is ​​solved by the optimization algorithm. In this process, the covariance matrix is ​​first calculated by singular value decomposition (SVD) to analytically solve the rotation matrix R and translation vector T. In each iteration, the rotation and displacement are optimized according to the weighted residual function until the error converges. Exemplarily, during the iterative optimization process, if the error change rate is less than 0.5% or the number of iterations exceeds 20 times, the calculation is stopped and the final rigid body transformation matrix is ​​output.

[0071] In the above implementation, the improved ICP algorithm based on feature constraints achieves high precision and high efficiency in steel truss segment registration. Through dynamic weighted residual function and robustness optimization, the ability to resist noise and interference is significantly improved, which can not only meet the requirements of high-precision assembly, but also be suitable for complex construction site environments and support the rapid construction of large-scale projects.

[0072] Reference Figure 7 As an implementation method of step S105, based on the steel truss design model and the steel truss three-dimensional point cloud model, the displacement deviation and angle deviation between segments are calculated to obtain the assembly error result, including: Step S501, converting the CAD geometric data of the steel truss design model into a point cloud format to generate a theoretical point cloud model; Among them, the steel truss design model is usually generated by CAD software (such as AutoCAD or Revit), which contains the geometric information of all segments, including node coordinates, segment dimensions, angles, etc. These design data are mainly expressed as vector data, which is not suitable for direct comparison with actual point cloud data. Therefore, it is first necessary to convert these CAD data into point cloud format to generate a theoretical point cloud model. The point cloud is composed of a set of discrete points in space, each point corresponds to a spatial position, and can accurately describe the geometric shape of the surface or boundary of an object. After the design model is converted into a point cloud format, it can be compared with the point cloud data obtained by 3D laser scanning to identify and calculate errors.

[0073] Specifically, each point in the CAD model (such as the nodes and key points of the steel truss) will be converted into point cloud coordinates to form a theoretical "ideal" point cloud model, which is used as the basis for subsequent comparison and error calculation.

[0074] Step S502, obtaining the design coordinates of the center points of the bolt holes of each segment in the theoretical point cloud model and the actual coordinates in the three-dimensional point cloud model of the steel truss; Among them, when assembling steel trusses, the center points of the bolt holes are the key positioning basis because they determine the relative position between each segment. The center points of the bolt holes in each segment in the design model have predetermined ideal coordinates, which are represented in the theoretical point cloud model. The three-dimensional point cloud model is obtained from the actual site through three-dimensional laser scanning technology. It contains the coordinates of the bolt holes of each segment in the actual environment. In order to calculate the assembly error, it is first necessary to extract the coordinates of the center points of these bolt holes in the theoretical point cloud model and the actual point cloud model, and compare the position differences of the corresponding points in the two models.

[0075] Step S503, calculating the Euclidean distance deviation between the design coordinates and the actual coordinates to obtain the displacement deviation of each segment; Among them, by comparing the ideal coordinates of the bolt holes in the design model with the actual coordinates of the bolt holes in the actual point cloud model, the displacement deviation of each bolt hole can be calculated. The displacement deviation can be calculated by the Euclidean distance formula, which represents the spatial difference between the actual bolt hole position and the designed bolt hole position.

[0076] Step S504, obtaining the theoretical flange edge line direction vector of each pair of adjacent segments in the theoretical point cloud model and the actual flange edge line direction vector in the three-dimensional point cloud model of the steel truss; Among them, in the process of splicing steel trusses, the butt joint of the flange edge line is an important reference point. The direction of each pair of adjacent segment flanges in the design model is usually preset by the designer during modeling and expressed as a direction vector in the theoretical point cloud model. Each pair of adjacent segments in the actual point cloud model will also have an actual measured flange direction vector.

[0077] Step S505, calculating the angle deviation between the theoretical flange edge line direction vector and the actual flange edge line direction vector to obtain the angle deviation of each pair of adjacent segments; Among them, by calculating the angle between the direction vectors of the flanges of adjacent segments in the design model and the actual point cloud model, the angle deviation can be obtained, and the angle deviation reflects the rotation error between the splicing surfaces. In some embodiments, the angle deviation can be calculated using the dot product formula between the direction vectors. By accurately calculating the angle deviation, the error of inconsistent angles of the splicing surfaces can be accurately identified.

[0078] Step S506, generating an assembly error result according to the displacement deviation and the angle deviation.

[0079] The displacement deviation and angle deviation of each segment are summarized and integrated to generate the assembly error result, which can intuitively reflect the overall error caused by factors such as construction, measurement error or material deviation during the assembly process. For example, if the displacement deviation of a segment is 0.3mm and the angle deviation is 0.2°, the assembly error data will include these two pieces of information and will be used for the subsequent generation of correction instructions.

[0080] In the above implementation, the error comparison between the steel truss design model and the actual three-dimensional point cloud model is performed to accurately calculate the displacement deviation and angle deviation of each segment, thereby generating assembly error data. This technical solution effectively identifies assembly errors caused by construction, measurement or material deviations, and provides an accurate basis for adjustment, ensuring high precision and high efficiency in the steel truss assembly process.

[0081] As a further implementation of the pre-assembly method, after the step of generating a construction correction parameter set based on the assembly error result, it also includes: generating an angle correction instruction and a displacement correction instruction according to the construction correction parameter set.

[0082] Among them, the angle correction instruction is used to adjust the lifting angle of each steel truss segment to perform angle compensation, and the displacement correction instruction is used to adjust the alignment deviation of the bolt holes of each steel truss to perform displacement adjustment.

[0083] In the above implementation, the installation accuracy of each segment of the steel truss is precisely adjusted. The angle correction instruction ensures that the lifting angle of each segment is compensated, thereby avoiding angle errors during the assembly process, and the displacement correction instruction is used to adjust the alignment deviation of the bolt holes to ensure that each segment can be accurately docked. These correction instructions provide clear operating instructions for on-site construction, greatly improving the accuracy and construction efficiency of steel truss assembly, and reducing the risk of rework caused by errors.

[0084] The embodiment of the present application also discloses a digital pre-assembly system for steel trusses based on three-dimensional laser scanning.

[0085] A digital pre-assembly system for steel trusses based on three-dimensional laser scanning, comprising: The station layout module is used to calculate the three-dimensional coordinates and scanning angles of the scanning station through the spatial coverage algorithm based on the steel truss design model and the on-site environmental point cloud data, and obtain the station layout position parameters; The point cloud data acquisition module is used to control the 3D laser scanner to scan the steel truss segment by segment according to the measurement station layout position parameters, and acquire the original point cloud data of each segment of the steel truss; The point cloud data processing module is used to perform spatial alignment and noise reduction on the original point cloud data to obtain standardized point cloud data; The segment splicing module is used to segment and extract the characteristic parameters of each segment based on the standardized point cloud data and perform digital splicing to generate a three-dimensional point cloud model of the steel truss; The assembly error analysis module is used to calculate the displacement deviation and angle deviation between segments based on the steel truss design model and the steel truss 3D point cloud model to obtain the assembly error result; The correction parameter determination module is used to generate a construction correction parameter set based on the assembly error results.

[0086] As a further implementation of the pre-assembly system, it also includes: A correction instruction generation module, used for generating angle correction instructions and displacement correction instructions according to a construction correction parameter set; Among them, the angle correction instruction is used to adjust the lifting angle of each steel truss segment to perform angle compensation, and the displacement correction instruction is used to adjust the alignment deviation of the bolt holes of each steel truss to perform displacement adjustment.

[0087] A digital pre-assembly system for steel trusses based on three-dimensional laser scanning in an embodiment of the present application can implement any of the above-mentioned pre-assembly methods, and the specific working process of each module in the pre-assembly system can refer to the corresponding process in the above-mentioned method embodiment.

[0088] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0089] The embodiment of the present application also discloses a computer device.

[0090] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a digital pre-assembly method for steel trusses based on three-dimensional laser scanning as described above is implemented.

[0091] The embodiment of the present application also discloses a computer-readable storage medium.

[0092] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the above-mentioned methods for digital pre-assembly of steel trusses based on three-dimensional laser scanning.

[0093] Among them, computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus or device; the program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0094] It should be noted that in the above embodiments, the description of each embodiment has different emphases, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0095] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A digital pre-assembly method for steel trusses based on three-dimensional laser scanning, characterized in that: The method comprises: Based on the steel truss design model and the on-site environmental point cloud data, the three-dimensional coordinates and scanning angles of the scanning station are calculated through the spatial coverage algorithm to obtain the station layout position parameters; According to the measurement station layout position parameters, control the three-dimensional laser scanner to scan the steel truss segment by segment, and collect the original point cloud data of each segment of the steel truss; Performing spatial alignment and noise reduction processing on the original point cloud data to obtain standardized point cloud data; Based on the standardized point cloud data, the characteristic parameters of each segment are segmented and extracted, and digital splicing is performed to generate a three-dimensional point cloud model of the steel truss; Based on the steel truss design model and the three-dimensional point cloud model of the steel truss, the displacement deviation and the angle deviation between the segments are calculated to obtain the assembly error result; A construction correction parameter set is generated based on the assembly error result.

2. The method for digital pre-assembly of steel trusses based on three-dimensional laser scanning according to claim 1, characterized in that: Based on the steel truss design model and the on-site environmental point cloud data, the three-dimensional coordinates and scanning angles of the scanning station are calculated by the spatial coverage algorithm, and the steps of obtaining the station layout position parameters include: Based on the steel truss design model, the length of each segment of the steel truss and the distribution characteristics of the connection nodes are obtained, and the maximum effective scanning distance between adjacent measuring stations is calculated; Initialize the station position and scanning angle, and configure the spatial coverage constraints; Based on the spatial coverage constraint and the maximum effective scanning distance, the layout positions and scanning angles of the measuring stations are iteratively optimized to minimize the number of measuring stations and ensure that all sections of the steel truss are covered, thereby obtaining optimized measuring station layout position parameters.

3. The method for digital pre-assembly of steel trusses based on three-dimensional laser scanning according to claim 1, characterized in that: Based on the standardized point cloud data, the steps of segmenting and extracting characteristic parameters of each segment and digitally splicing to generate a three-dimensional point cloud model of a steel truss include: Based on the standardized point cloud data, the bolt hole center points and flange edge line features of each segment of the steel truss are segmented and extracted; Taking the center point of the bolt hole as the registration reference, locating the segment splicing direction according to the flange edge line features, and calculating the rigid body transformation matrix between each pair of adjacent segments based on the ICP algorithm; Registering and splicing each pair of adjacent segment point clouds according to the rigid body transformation matrix; The three-dimensional model is reconstructed according to the aligned and spliced ​​adjacent segment point clouds to generate the three-dimensional point cloud model of the steel truss, and the actual position parameters of each segment in the global coordinate system are output.

4. The method for digital pre-assembly of steel trusses based on three-dimensional laser scanning according to claim 3 is characterized in that: The steps of calculating the rigid body transformation matrix between each pair of adjacent segments based on the ICP algorithm include: Get the bolt hole center point spacing and flange edge line direction vector between each pair of adjacent segments; Based on a preset spacing standard, a set of matching point pairs whose bolt hole spacing errors are less than a preset error threshold are selected, and an initial angle compensation value is calculated according to the flange edge line direction vector; According to the spacing deviation set of the bolt hole matching point pair set, a weighted residual function is constructed in combination with the flange edge line direction vector; The rotation parameters are initialized based on the weighted residual function and the initial angle compensation value, the optimal rotation matrix and translation vector are solved by the singular value decomposition iterative optimization algorithm, the calculation is stopped when the error change rate or the number of iterations meets the termination condition, and the rigid body transformation matrix is ​​output.

5. The method for digital pre-assembly of steel trusses based on three-dimensional laser scanning according to claim 4, characterized in that: Based on the steel truss design model and the steel truss three-dimensional point cloud model, the steps of calculating the displacement deviation and angle deviation between segments and obtaining the assembly error result include: Converting the CAD geometric data of the steel truss design model into a point cloud format to generate a theoretical point cloud model; Obtaining the design coordinates of the center points of the bolt holes of each segment in the theoretical point cloud model and the actual coordinates in the three-dimensional point cloud model of the steel truss; Calculate the Euclidean distance deviation between the design coordinates and the actual coordinates to obtain the displacement deviation of each segment; Obtaining a theoretical flange edge line direction vector of each pair of adjacent segments in the theoretical point cloud model and an actual flange edge line direction vector in the three-dimensional point cloud model of the steel truss; Calculating the angle deviation between the theoretical flange edge line direction vector and the actual flange edge line direction vector to obtain the angle deviation of each pair of adjacent segments; An assembly error result is generated according to the displacement deviation and the angle deviation.

6. A digital pre-assembly method for steel trusses based on three-dimensional laser scanning according to any one of claims 1 to 5, characterized in that: After the step of generating a construction correction parameter set based on the assembly error result, the method further includes: Angle correction instructions and displacement correction instructions are generated according to the construction correction parameter set; wherein the angle correction instructions are used to adjust the lifting angle of each steel truss segment to perform angle compensation, and the displacement correction instructions are used to adjust the alignment deviation of each steel truss bolt hole to perform displacement adjustment.

7. A digital pre-assembly system for steel trusses based on three-dimensional laser scanning, characterized in that: The system comprises: The station layout module is used to calculate the three-dimensional coordinates and scanning angles of the scanning station through the spatial coverage algorithm based on the steel truss design model and the on-site environmental point cloud data, and obtain the station layout position parameters; A point cloud data acquisition module, used to control the three-dimensional laser scanner to scan the steel truss segment by segment according to the measurement station layout position parameters, and acquire the original point cloud data of each segment of the steel truss; A point cloud data processing module is used to perform spatial alignment and noise reduction on the original point cloud data to obtain standardized point cloud data; A segment splicing module is used to segment and extract characteristic parameters of each segment based on the standardized point cloud data and perform digital splicing to generate a three-dimensional point cloud model of the steel truss; An assembly error analysis module is used to calculate the displacement deviation and angle deviation between segments based on the steel truss design model and the three-dimensional point cloud model of the steel truss to obtain an assembly error result; A correction parameter determination module is used to generate a construction correction parameter set based on the assembly error result.

8. The digital pre-assembly system for steel trusses based on three-dimensional laser scanning according to claim 7 is characterized in that: The system also includes a correction instruction generation module, which is used to generate angle correction instructions and displacement correction instructions according to the construction correction parameter set; the angle correction instructions are used to adjust the lifting angle of each steel truss segment to perform angle compensation, and the displacement correction instructions are used to adjust the alignment deviation of each steel truss bolt hole to adjust the displacement.

9. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 6.

Images