Digital manufacturing method and system for bracket joint in steel structure column

Abstract

The invention discloses a digital manufacturing method and system for a bracket node in a steel structure column, and relates to the related field of digital manufacturing of building components, and the method comprises the steps: collecting three-dimensional point cloud data of a bracket node connection region in a target steel structure column; stress concentration analysis is conducted in the vertical direction and the horizontal direction, and a key welding point set and a bolt hole site set are marked; lightweight design parameters of the bracket supporting structure are introduced, load stability analysis is conducted, and first assembly parameters are configured; traversing M bracket nodes in the target steel structure column to obtain a first assembly parameter to an Mth assembly parameter, and configuring a digital manufacturing instruction sequence in combination with the three-dimensional point cloud data; and inputting the data into a numerical control machining center, and driving multi-axis linkage equipment to manufacture and manage the bracket node. The technical problems that existing bracket joint manufacturing is low in efficiency and insufficient in precision and stability are solved, and the technical effects that the manufacturing efficiency is improved, the manufacturing precision and stability are improved, and powerful support is provided for energy-saving building material production are achieved.

Description

Technical Field

[0001] The present application relates to the field of digital manufacturing of building components, and in particular to a digital manufacturing method and system for corbel nodes in steel structure columns. Background Art

[0002] In today's society, with the rapid development of the construction industry and the increasing demands for structural safety and durability, energy-saving building materials and efficient manufacturing processes have become a key focus of the industry. Especially during the construction of steel structures, the manufacturing accuracy and stability of joints are directly related to the safety and durability of the entire structure. Corbel joints, as key connections in steel structures, are particularly important due to their manufacturing quality. Currently, the manufacturing of corbel joints in steel columns primarily relies on traditional manual measurement and welding processes. This method is not only inefficient but also relies on the worker's experience and skill level, often resulting in insufficient manufacturing accuracy and uneven welding quality, making it difficult to meet the high precision and high stability requirements of modern steel structures.

[0003] In the current related technologies, the manufacturing of corbel nodes in steel structure columns has technical problems such as low efficiency, insufficient precision and stability. Summary of the Invention

[0004] This application provides a digital manufacturing method and system for corbel nodes in steel structure columns. It first collects three-dimensional point cloud data of the corbel nodes in the steel structure columns, then performs stress concentration analysis and marks key welding points and bolt hole sets, then introduces lightweight design parameters to perform load stability analysis, configures assembly parameters, and then traverses all corbel nodes. Digital manufacturing instructions are generated based on the three-dimensional point cloud data, and finally the instructions are input into a CNC machining center to perform digital manufacturing management of the corbel nodes. Technical means such as these achieve the technical effect of improving manufacturing efficiency, enhancing manufacturing accuracy and stability, and providing strong support for the production of energy-saving building materials.

[0005] The present application provides a digital manufacturing method for a corbel node in a steel structure column, comprising: collecting three-dimensional point cloud data of a connection area of ​​a corbel node in a target steel structure column; performing stress concentration analysis in a vertical direction and a horizontal direction based on the connection area, marking a first key welding point set and a first bolt hole position set, and marking a second key welding point set and a second bolt hole position set; introducing lightweight design parameters of a corbel support structure, performing a load stability analysis based on the first key welding point set and the first bolt hole position set, and the second key welding point set and the second bolt hole position set, and configuring a first assembly parameter; traversing M corbel nodes in the target steel structure column, obtaining a first assembly parameter, a second assembly parameter, and up to an Mth assembly parameter, and configuring a digital manufacturing instruction sequence in combination with the three-dimensional point cloud data; inputting the digital manufacturing instruction into a CNC machining center, and driving a multi-axis linkage device to perform digital manufacturing management on the M corbel nodes.

[0006] In a possible implementation, the following processing is performed: a fiber Bragg grating sensor is deployed at the end of the corbel support structure to collect strain spectrum data under dynamic load; based on the strain spectrum data under dynamic load, the proportion of low-frequency vibration energy in the strain signal is extracted; if the proportion of low-frequency vibration energy exceeds a preset safety threshold, a structural reinforcement instruction is triggered.

[0007] In a possible implementation, the following processing is performed: the connection area includes a top plane and a side plane; based on the top plane and the side bevel of the connection area, a stress concentration analysis is performed from a vertical direction, and a first key welding point set and a first bolt hole position set are marked; based on the side plane and the side bevel of the connection area, a stress concentration analysis is performed from a horizontal direction, and a second key welding point set and a second bolt hole position set are marked.

[0008] In a possible implementation, the first assembly parameters are configured and the following processing is performed: discrete element simulation is performed based on the first key welding point set and the first bolt hole position set, and the plastic deformation distribution of the corbel node under the vertical load is output; according to the plastic deformation distribution under the vertical load, the welding sequence and the bolt preload combination are optimized to determine the first unidirectional assembly parameters; based on the first unidirectional assembly parameters under the vertical load and the second unidirectional assembly parameters under the horizontal load, the first assembly parameters are determined.

[0009] In a possible implementation, the following processing is performed: the first unidirectional assembly parameter belongs to a first unidirectional assembly parameter set; under a horizontal load, the spacing parameter of the second bolt hole position set is corrected by the friction energy dissipation coefficient to determine a first stability coefficient, and at the same time, the welding process parameters of the second key welding point set are corrected by the welding residual stress to determine a second stability coefficient; using the first stability coefficient and the second stability coefficient, the first unidirectional assembly parameter set including the welding current gradient and the bolt torque sequence is configured.

[0010] In a possible implementation, the M corbel nodes in the target steel structure column are traversed to obtain the first assembly parameter, the second assembly parameter, and up to the Mth assembly parameter, and the digital manufacturing instruction sequence is configured in combination with the three-dimensional point cloud data, and the following processing is also performed: the first corbel node among the M corbel nodes is used to jointly encode the three-dimensional point cloud data and the first assembly parameter to determine a first joint encoding vector; based on the M corbel nodes, M joint encoding vectors are configured and input into a generative adversarial network to obtain a manufacturing error probability distribution map, the input layer of the adversarial network receives the M joint encoding vectors, and the output layer of the adversarial network generates the manufacturing error probability distribution map; wherein, the welding heat input parameter is mapped to the welding heat influence superposition layer under the composite working condition area, and the bolt preload error is mapped to the bolt hole edge superposition layer under the composite working condition area.

[0011] In a possible implementation, the following processing is performed: the hidden layer of the adversarial network is composed of residual convolution units, and a cross-layer feature index relationship is established to fuse multi-scale manufacturing error features; the generator of the adversarial network The training process uses the adversarial loss function , constraining the KL divergence between the generated error distribution and the measured error data x to be less than the divergence error threshold.

[0012] In a possible implementation, the following processing is performed: the adversarial loss function ;in, Used to characterize the distribution of measured error data The expectation of , x is the measured error data, is the measured error data distribution, It refers to the judgment result of the discriminator D on the measured error data x. To characterize the generator Prior distribution of input The expectation of z is the generator The input sample, refers to the generator Prior distribution of the input, Is the discriminator D to the generator Generated error data The judgment result of .

[0013] In a possible implementation, a multi-axis linkage device is driven to perform digital manufacturing management on the M corbel nodes, and the following processing is performed: based on the manufacturing error probability distribution map, a reverse compensation cutting trajectory is inserted into the CNC machining path; the machining priority of each of the M corbel nodes is matched, and the task scheduling sequence of the multi-axis linkage device is optimized.

[0014] The present application also provides a digital manufacturing system for a corbel node in a steel structure column, comprising: a three-dimensional point cloud data acquisition module for collecting three-dimensional point cloud data of a connection area of ​​a corbel node in a target steel structure column; a stress concentration analysis module for performing stress concentration analysis in a vertical direction and a horizontal direction based on the connection area, marking a first key welding point set and a first bolt hole position set, and marking a second key welding point set and a second bolt hole position set; an assembly parameter configuration module for introducing lightweight design parameters of a corbel support structure, performing load stability analysis based on the first key welding point set and the first bolt hole position set, and the second key welding point set and the second bolt hole position set, and configuring a first assembly parameter; a digital manufacturing instruction sequence configuration module for traversing M corbel nodes in the target steel structure column, obtaining a first assembly parameter, a second assembly parameter, and up to an Mth assembly parameter, and configuring a digital manufacturing instruction sequence in combination with the three-dimensional point cloud data; a digital manufacturing management module for inputting the digital manufacturing instruction into a CNC machining center, and driving a multi-axis linkage device to perform digital manufacturing management on the M corbel nodes.

[0015] The proposed digital manufacturing method and system for a corbel node in a steel column first collects three-dimensional point cloud data of the connection area of ​​the corbel node in a target steel column. Based on the connection area, stress concentration analysis is performed in both the vertical and horizontal directions. A first set of key welding points and a first set of bolt hole locations are annotated, and a second set of key welding points and a second set of bolt hole locations are annotated. Then, lightweight design parameters for the corbel support structure are introduced. A load stability analysis is performed based on the first set of key welding points and the first set of bolt hole locations, and the second set of key welding points and the second set of bolt hole locations. First assembly parameters are configured, and then M corbel nodes in the target steel column are traversed to obtain first, second, and Mth assembly parameters. A digital manufacturing instruction sequence is configured based on the three-dimensional point cloud data. Finally, the digital manufacturing instruction sequence is input into a CNC machining center to drive a multi-axis linkage device to perform digital manufacturing management on the M corbel nodes. This achieves the technical effect of improving manufacturing efficiency, enhancing manufacturing precision and stability, and providing strong support for the production of energy-saving building materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A schematic flow chart of a digital manufacturing method for a corbel node in a steel structure column provided in an embodiment of the present application.

[0018] Figure 2 A schematic structural diagram of a digital manufacturing system for a corbel node in a steel structure column provided in an embodiment of the present application.

[0019] Description of the accompanying drawings: three-dimensional point cloud data acquisition module 10, stress concentration analysis module 20, assembly parameter configuration module 30, digital manufacturing instruction sequence configuration module 40, digital manufacturing management module 50. DETAILED DESCRIPTION

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0021] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0022] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0023] The embodiment of the present application provides a digital manufacturing method for the corbel node in the steel structure column, such as Figure 1 As shown, the method includes: Step S100: collecting three-dimensional point cloud data of the connection area of ​​the corbel node at the target steel structure column.

[0024] Specifically, a 3D laser scanner is used to scan the corbel joint connection area of ​​the target steel column to obtain high-precision 3D point cloud data. Specifically, a high-precision 3D laser scanner, such as the Leica HDS7000 or FARO Focus3D, is used. These devices can capture 3D coordinate information of the object surface with millimeter-level accuracy. The scanner is placed in a suitable position to perform a full-scale scan of the corbel joint connection area. The scanner emits a laser beam, which is reflected by the object surface and then received by the scanner. The 3D coordinate points on the object surface are obtained by calculating the time difference or phase difference in the laser's round-trip travel. The collected point cloud data is imported into professional 3D modeling software (such as Rhinoceros or AutoCAD), where it is denoised, filtered, and spliced ​​to generate a complete 3D point cloud model. Each point in the 3D point cloud data contains X, Y, and Z coordinate information, which describes the shape and size of the object.

[0025] Step S200 , based on the connection area, stress concentration analysis is performed in the vertical direction and the horizontal direction respectively, and a first key welding point set and a first bolt hole position set are marked, and a second key welding point set and a second bolt hole position set are marked.

[0026] Specifically, finite element analysis (FEA) software is used to perform stress concentration analysis on the connection area of ​​the corbel joint, identifying areas where uneven stress distribution occurs when the structure is subjected to load. These areas represent weak links in the structure and require special attention and reinforcement. Critical weld points and bolt hole locations are then annotated using an automated algorithm. Specifically, finite element analysis software such as ANSYS or ABAQUS is used, as these software are capable of performing stress analysis on complex structures. The 3D point cloud model generated in step S100 is imported into the FEA software to establish a finite element model. Material properties (such as the elastic modulus and Poisson's ratio of the steel) and boundary conditions (such as the node's fixed constraints and load application method) are defined. FEA analysis is then run to calculate the vertical and horizontal stress distribution in the connection area. The software's built-in post-processing capabilities are used to identify stress concentration areas and extract the coordinates of critical weld points and bolt hole locations. Based on the stress distribution results, an automated algorithm (such as a rule-based or machine learning algorithm) annotates the first set of critical weld points and bolt hole locations, and the second set of critical weld points and bolt hole locations. The critical welding point set is a collection of key locations in the structure that require welding, determined based on stress distribution and design requirements. The welding quality of these points directly affects the strength and stability of the structure. The bolt hole set is a collection of hole locations for installing bolts, determined according to structural design requirements. The accuracy and location of these holes are crucial to the reliability of the structural connection.

[0027] For example, in an actual steel column corbel joint, finite element analysis revealed that the four corners and two symmetrical points in the middle of the connection area are stress concentration areas. These points are automatically marked as critical welding points. Furthermore, bolt hole locations are marked near these critical points according to design specifications to ensure the reliability of the welded and bolted connections.

[0028] In a possible implementation, step S200 further includes step S210, and the connection area includes a top plane and a side plane. Specifically, the three-dimensional point cloud data of the collected corbel node connection area is segmented and fitted by three-dimensional point cloud data processing software, and the top plane and side plane are extracted. Specifically, point cloud processing software, such as CloudCompare or MeshLab, is used to segment the three-dimensional point cloud data, and the top plane and side plane of the connection area are extracted respectively. The software divides the point cloud data into different areas through a clustering algorithm or a segmentation method based on geometric features. The segmented point cloud data is plane fitted, and the equations of the top plane and the side plane are calculated using the least squares method or other fitting algorithms. The fitting results can be intuitively displayed in a three-dimensional view as a basis for subsequent stress concentration analysis.

[0029] Step S220: Perform a stress concentration analysis in the vertical direction based on the top plane and side bevels of the connection area, and annotate the first set of key weld points and the first set of bolt hole locations. Specifically, using finite element analysis software, the geometric information of the top plane and side bevels is combined to perform a stress concentration analysis in the vertical direction, and the first set of key weld points and the first set of bolt hole locations are annotated. Specifically, the geometric information of the top plane and side bevels extracted in step S210 is imported into finite element analysis software, such as ANSYS or ABAQUS, to establish a finite element model. In the model, material properties (such as the elastic modulus and Poisson's ratio of the steel) and boundary conditions (such as the fixed constraints of the nodes and the load application method) are defined. A finite element analysis is run to calculate the stress distribution in the vertical direction of the connection area. The software generates a stress distribution cloud map that visually displays the stress concentration areas. Using the software's post-processing function, stress concentration areas are identified, and the coordinate information of the key weld points and bolt hole locations is extracted. These points are located in the stress peak area and require special reinforcement.

[0030] For example, in a vertical stress analysis of a steel column corbel joint, finite element analysis results revealed significant stress concentration at the intersection of the top plane and the side bevel. The analysis software automatically marked five critical welding points and four bolt hole locations in this area, and the coordinates of these points were recorded for subsequent generation of manufacturing instructions.

[0031] Step S230, based on the side plane and the side bevel of the connection area, perform stress concentration analysis in the horizontal direction, and mark the second key welding point set and the second bolt hole position set. Specifically, using finite element analysis software, combined with the geometric information of the side plane and the side bevel, perform stress concentration analysis in the horizontal direction, and mark the second key welding point set and the second bolt hole position set. Specifically, in the finite element analysis software, adjust the boundary conditions and load application direction of the model to make it suitable for stress analysis in the horizontal direction. Rerun the finite element analysis to calculate the stress distribution of the connection area in the horizontal direction. Similarly, generate a stress distribution cloud map to intuitively display the stress concentration area. Through the post-processing function of the software, identify the stress concentration area in the horizontal direction, and extract the coordinate information of the key welding points and bolt holes. The marking method of these points is similar to that of the vertical direction, but the focus is on the stress distribution in the horizontal direction.

[0032] For example, in a horizontal stress analysis of a steel column corbel joint, finite element analysis results revealed significant stress concentration at the intersection of the side plane and the side bevel. The analysis software automatically marked three key welding points and two bolt hole locations in this area, and the coordinates of these points were also recorded for subsequent manufacturing instruction generation.

[0033] Step S300 , introducing lightweight design parameters of the corbel support structure, performing load stability analysis based on the first key welding point set and the first bolt hole position set, and the second key welding point set and the second bolt hole position set, and configuring first assembly parameters.

[0034] Specifically, lightweight design parameters are used to reduce structural weight while ensuring structural strength and stability by optimizing material selection and structural shape. Numerical analysis methods are used to analyze the load stability of the corbel support structure and configure assembly parameters, combining these lightweight design parameters. Specifically, these parameters include optimizing material selection (such as high-strength steel) and structural shape (such as using hollow sections or lattice structures). In finite element analysis software, the corbel support structure model is modified based on the lightweight design parameters, and a new load stability analysis is performed to examine the structure's stability under various loads (such as deadweight, wind loads, and seismic loads) to ensure that the structure will not fail during operation. By analyzing loads under different operating conditions (such as deadweight, wind loads, and seismic loads), stability indicators (such as buckling coefficient and safety factor) of the corbel support structure are calculated. Based on the analysis results, assembly parameters for weld points and bolt hole locations are determined. These parameters, such as welding current, voltage, welding time, and bolt tightening torque, are used to control assembly operations such as welding and bolting during the manufacturing process.

[0035] For example, in the steel columns of a high-rise building, the corbel joints are made of high-strength steel and designed as hollow sections. Load stability analysis determined that a 10mm diameter welding rod, 200A welding current, 25V voltage, and 10 seconds welding time were used at key welding points. Furthermore, M20 bolts were used in the bolt holes, with a tightening torque of 200N·m.

[0036] In one possible implementation, after configuring the first assembly parameters, step S300 further includes step S310, performing a discrete element simulation based on the first set of key weld points and the first set of bolt hole locations to output the plastic deformation distribution of the corbel node under vertical load. Specifically, discrete element method (DEM) simulation software is used to simulate and analyze the plastic deformation of the corbel node under vertical load, combining the geometric information of the first set of key weld points and the first set of bolt hole locations. Specifically, discrete element simulation software, such as PFC (Particle Flow Code) or EDEM, is used. These software programs can simulate the microscopic particle behavior of materials under load, thereby deriving the macroscopic plastic deformation distribution. In the discrete element simulation software, a discrete element model is established based on the geometric information of the corbel node (including the first set of key weld points and the first set of bolt hole locations). Material properties of the corbel node (such as elastic modulus, Poisson's ratio, yield strength, etc.) are input into the model. A vertical load is applied to the top of the model to simulate the load conditions under actual working conditions. The load can be a uniformly distributed load or a concentrated load, depending on the design requirements. Upon starting the simulation, the software calculates the plastic deformation distribution of the corbel joint under vertical load based on material properties and loading conditions. The simulation results are presented as a deformation cloud map, visually displaying the degree of deformation in each area. After the simulation is complete, the plastic deformation distribution data of the corbel joint under vertical load is extracted, including information such as the area of ​​maximum deformation and deformation gradient. This data serves as the basis for subsequent optimization of the welding sequence and bolt preload.

[0037] For example, in a discrete element simulation of a steel column corbel joint, the software simulated the deformation of the corbel joint under vertical load. The results showed that the area near the first critical weld point cluster exhibited significant deformation, with a maximum deformation of 2 mm. This deformation data was recorded and used to optimize the subsequent welding sequence and bolt preload.

[0038] Step S320: Based on the plastic deformation distribution under vertical load, the welding sequence and bolt preload combination are optimized to determine the first unidirectional assembly parameters. Specifically, an optimization algorithm (such as a genetic algorithm or particle swarm optimization algorithm) is used in conjunction with the plastic deformation distribution data obtained from discrete element simulation to optimize the welding sequence and bolt preload combination. Specifically, an optimization algorithm such as a genetic algorithm (GA) or particle swarm optimization (PSO) is selected. These algorithms can find the optimal welding sequence and bolt preload combination through an iterative search. An optimization objective function is defined to minimize structural deformation and stress concentration. The objective function can include the root mean square value of deformation, maximum stress, etc. The parameter ranges for the welding sequence and bolt preload are initialized. For example, the welding sequence can be from center to edge or from edge to center, and the bolt preload can vary within a certain range. The optimization algorithm is run, and through multiple iterations, the welding sequence and bolt preload combination are gradually adjusted until the optimal solution that meets the objective function is found. After the optimization is complete, the first unidirectional assembly parameters, including the optimal welding sequence and bolt preload combination, are determined. These parameters are used in the subsequent generation of digital manufacturing instructions.

[0039] For example, during the optimization process, the genetic algorithm, through multiple generations of iteration, ultimately determined that the optimal welding sequence was from center to edge with a bolt preload of 150 kN. These parameters were recorded as the first unidirectional assembly parameters.

[0040] Step S330 determines the first assembly parameters based on the first unidirectional assembly parameters under vertical load and the second unidirectional assembly parameters under horizontal load. Specifically, the second unidirectional assembly parameters are determined by performing a similar discrete element simulation analysis on the corbel node under horizontal load to optimize the welding sequence and bolt preload combination. The first unidirectional assembly parameters under vertical load (such as welding sequence and bolt preload) and the second unidirectional assembly parameters under horizontal load are integrated. These parameters include welding current, voltage, welding time, bolt tightening torque, etc. The integrated parameters are comprehensively evaluated through finite element analysis or other numerical analysis methods to ensure that the structural performance of the corbel node meets the design requirements under multi-directional loads. Based on the comprehensive analysis results, the assembly parameters are fine-tuned to achieve optimal structural performance. For example, the welding current or bolt preload needs to be adjusted to balance the effects of loads in different directions. Finally, the first assembly parameters are determined, and these parameters are used for subsequent digital manufacturing instruction generation.

[0041] For example, after comprehensive analysis, the final first assembly parameters were determined as follows: welding sequence from center to edge, welding current 200A, voltage 25V, welding time 10 seconds, bolt preload 150kN, and tightening torque 200N·m. These parameters were input into the CNC machining center for the manufacture of the corbel joint.

[0042] In one possible implementation, step S320 further includes step S321, where the first unidirectional assembly parameters belong to a first unidirectional assembly parameter set. Specifically, the first unidirectional assembly parameters (such as welding sequence, welding current, voltage, welding time, bolt preload, etc.) optimized in step S320 are stored in a database or parameter file to form the first unidirectional assembly parameter set. The parameter set can be a table or structured data file, with each row or record corresponding to the assembly parameters of a corbel joint.

[0043] For example, the optimized parameters for the first unidirectional assembly include a center-to-edge welding sequence, a welding current of 200A, a voltage of 25V, a welding time of 10 seconds, and a bolt preload of 150kN. These parameters are stored in a database table called "Vertical_Assembly_Parameters," with each column corresponding to a parameter and each row corresponding to a corbel node.

[0044] Step S322, under horizontal load, correct the spacing parameters of the second bolt hole set by the friction energy dissipation coefficient to determine the first stability coefficient, and at the same time, correct the welding process parameters of the second key welding point set by the welding residual stress to determine the second stability coefficient. Specifically, in the finite element analysis software, establish a corbel node model under horizontal load, and input the friction energy dissipation coefficient (a coefficient reflecting the friction energy dissipation capacity of the material between the contact surfaces, determined according to the material and contact conditions). Run the simulation to analyze the stress distribution and deformation of the bolt connection area under horizontal load. According to the simulation results, adjust the spacing parameters of the second bolt hole set to optimize the stability of the structure. Calculate the adjusted structural stability index and determine the first stability coefficient, which reflects the stability of the bolt connection area under horizontal load. The first stability coefficient can be calculated by comparing the stability index before and after adjustment. For example, if the maximum stress after adjustment is reduced, the stability coefficient can be expressed as: first stability coefficient = , this coefficient is greater than 1, indicating that the adjusted design is more stable.

[0045] In the finite element analysis software, a welding model of the corbel node is established, and the welding residual stress distribution is input (the residual stress caused by thermal cycling during welding is obtained through thermal-structural coupling analysis). Run the simulation to analyze the influence of welding residual stress on the second critical welding point set. According to the simulation results, adjust the welding process parameters (such as welding current, voltage, welding speed, etc.) to reduce the influence of welding residual stress and optimize the stability of the structure. Calculate the stability index of the adjusted welding area and determine the second stability coefficient, which reflects the stability of the welding area under horizontal load. The second stability coefficient can be calculated by comparing the stability index before and after adjustment. For example, if the residual stress is reduced after adjustment, the stability coefficient can be expressed as: Second stability coefficient = , this coefficient is greater than 1, indicating that the adjusted welding process is more conducive to the stability of the structure.

[0046] For example, under horizontal loads, finite element analysis revealed that adjusting the spacing of the second bolt hole set from 150 mm to 160 mm significantly improved the stability of the structure. The calculated first stability coefficient was 1.2, indicating that the adjusted structure's stability under horizontal loads was 20% higher than the original design.

[0047] For example, finite element analysis revealed that adjusting the welding current from 200A to 180A and the welding speed from 50mm / min to 60mm / min significantly reduced weld residual stress. The calculated secondary stability coefficient was 1.15, indicating that the weld zone's stability under horizontal loads after the adjustments was 15% higher than the original design.

[0048] In step S323, the first and second stability coefficients are used to configure a first unidirectional assembly parameter set comprising a welding current gradient and a bolt torque sequence. Specifically, an optimization objective function for the welding current gradient and the bolt torque sequence is determined based on the first and second stability coefficients. This objective function aims to maximize structural stability. The welding current gradient and the bolt torque sequence are optimized using an optimization algorithm (such as a genetic algorithm or a particle swarm optimization algorithm). After optimization is complete, a first unidirectional assembly parameter set comprising the welding current gradient and the bolt torque sequence is generated. The optimized first unidirectional assembly parameter set is stored in a database or parameter file for subsequent use.

[0049] For example, based on a first stability factor of 1.2 and a second stability factor of 1.15, the genetic algorithm optimizes the welding current gradient from 200 A to 180 A, and the bolt torque sequence from 200 N·m to 220 N·m. These parameters are stored in a database table called "Horizontal_Assembly_Parameters," with each column corresponding to a parameter and each row corresponding to a corbel node.

[0050] Step S400, traverse the M corbel nodes in the target steel structure column, obtain the first assembly parameter, the second assembly parameter, and up to the Mth assembly parameter, and configure a digital manufacturing instruction sequence in combination with the three-dimensional point cloud data.

[0051] Specifically, a programming algorithm traverses all corbel nodes in the target steel structure column, combining 3D point cloud data and assembly parameters to generate a digital manufacturing instruction sequence. Specifically, programming languages ​​such as Python or C++ are used in conjunction with a database management system (such as MySQL or SQLite) to store and manage corbel node data. The corbel node information stored in the database is traversed to extract the key welding point set, bolt hole location set, and assembly parameters for each node. Based on the 3D point cloud data and assembly parameters, digital manufacturing instructions are generated for each corbel node. These manufacturing instructions include the motion path and welding parameter settings of the welding robot, as well as the machining path and parameters of the CNC machining equipment. The manufacturing instructions for all corbel nodes are sorted according to the machining order to generate a complete digital manufacturing instruction sequence, which is used to guide CNC machining centers and welding robots in manufacturing operations. The welding robot is an automated welding device whose motion path and welding parameters are controlled through programming to achieve efficient and precise welding operations.

[0052] For example, for a steel column with 10 corbel nodes, a programming algorithm extracts data for each node sequentially, generating corresponding manufacturing instructions. The manufacturing instructions for the first node include the welding robot's motion path (from coordinate points A to B and then to C), welding parameters (current 200A, voltage 25V), and the CNC machine's bolt hole machining path (from coordinate points D to E). These instructions are arranged in node number order, forming a complete instruction sequence for subsequent use by the CNC machining center.

[0053] In one possible implementation, M corbel nodes in the target steel structure column are traversed to obtain first, second, and up to the Mth assembly parameter. These parameters are then combined with the 3D point cloud data to configure a digital manufacturing instruction sequence. Step S400 further includes step S410, where the 3D point cloud data and the first assembly parameter are jointly encoded using the first of the M corbel nodes to determine a first joint encoding vector. Specifically, the 3D point cloud data of the first corbel node and the first assembly parameter are jointly encoded using a data fusion algorithm to generate a comprehensive encoding vector. Specifically, the 3D point cloud data of the first corbel node is subjected to dimensionality reduction, such as by principal component analysis (PCA) to extract key features. Simultaneously, the first assembly parameters (such as welding current, voltage, and bolt preload) are standardized for fusion with the point cloud data. The preprocessed 3D point cloud data and the first assembly parameter are jointly encoded using a data fusion algorithm (such as a weighted average method or a feature fusion method used in deep learning). The generated joint coding vector is a fixed-length vector that contains the geometric information and assembly parameter information of the corbel node. The generated first joint coding vector is stored in a database or data file for subsequent use.

[0054] For example, for the first corbel node, PCA was used to reduce the dimensionality of the 3D point cloud data to 10 key features. The first assembly parameters (welding current 200A, voltage 25V, bolt preload 150kN) were also standardized. This data was then fused using a weighted average method to generate a joint encoding vector of length 15. This vector was stored in a database table named "Joint_Encoding_Vector," with each column corresponding to a feature or parameter and each row corresponding to a corbel node.

[0055] In step S420, M joint encoding vectors are configured based on the M corbel nodes and input into a generative adversarial network to obtain a manufacturing error probability distribution map. The adversarial network's input layer receives the M joint encoding vectors, and the adversarial network's output layer generates the manufacturing error probability distribution map. The welding heat input parameter is mapped to the weld heat effect superposition layer under the composite working condition area, and the bolt preload error is mapped to the bolt hole edge superposition layer under the composite working condition area. Specifically, the M corbel nodes are jointly encoded to generate M joint encoding vectors. These vectors contain the geometric information and assembly parameter information of each corbel node. A generative adversarial network (GAN) is constructed, comprising a generator and a discriminator. The generator is responsible for generating the manufacturing error probability distribution map, while the discriminator is responsible for evaluating the authenticity of the generated map. The GAN is trained using historical manufacturing data until the generator can generate realistic manufacturing error probability distribution maps. The GAN's input layer receives the M joint encoding vectors. Each vector contains the geometric information and assembly parameter information of the corbel node. The output layer of the GAN generates a manufacturing error probability distribution map. This map maps the welding heat input parameters to the welding heat effect superposition layer under the composite working condition area, and maps the bolt preload error to the bolt hole edge superposition layer under the composite working condition area. Through the generated manufacturing error probability distribution map, the possible error areas in the welding heat affected zone and the bolt hole edge are analyzed to provide optimization suggestions for the subsequent manufacturing process. Among them, the welding heat effect superposition layer is the area where material properties change due to heat input during the welding process, and the distribution of welding heat effects is displayed through layering. The bolt hole edge superposition layer is the deformation area of ​​the bolt hole edge caused by the preload error, and the distribution of bolt hole edge errors is displayed through layering.

[0056] For example, for M = 10 corbel nodes, 10 joint encoding vectors are generated. These vectors are fed into the trained GAN to generate a probability distribution map of manufacturing errors. The map shows that the maximum error in the heat-affected zone (HAZ) is 5 mm, primarily concentrated in areas with high welding current. The maximum error at the bolt hole edge is 3 mm, primarily concentrated in areas with low bolt preload. This error distribution information is used to optimize subsequent manufacturing instruction sequences.

[0057] In one possible implementation, step S420 further includes step S421, wherein the hidden layers of the adversarial network are composed of residual convolutional units (RCUs), and a cross-layer feature index relationship is established to fuse multi-scale manufacturing error features. Specifically, Residual Convolutional Units (ResConvs) are used in the hidden layers of the Generative Adversarial Network (GAN), and multi-scale manufacturing error features are fused through cross-layer feature index relationships. The RCU is an improved convolutional neural network structure that addresses the vanishing gradient problem in deep network training by introducing residual connections. Each RCU consists of two convolutional layers, with batch normalization and ReLU activation functions used in between. After passing through the convolutional layers, the input data is directly added to the input data to form a residual connection. In the hidden layers of the GAN generator and discriminator, RCUs replace traditional convolutional layers. This structure can better extract and preserve the feature information of the input data while reducing the vanishing gradient problem during training. In the hidden layers, convolution kernels of different scales are used to extract multi-scale features of the input data. For example, 3×3, 5×5, and 7×7 convolution kernels are used to extract local, medium, and global features, respectively. Features of different scales are fused through cross-layer feature index relationships. Specifically, feature maps from lower layers are transferred to higher layers through upsampling or skip connections, where they are concatenated or added to the feature maps from higher layers. This cross-layer fusion preserves more detailed information while improving feature robustness.

[0058] In the input data, manufacturing error features may exist at different scales. For example, errors in the heat-affected zone of a weld may manifest as local thermal stress distributions, while errors at the edge of a bolt hole may manifest as global geometric deviations. Using residual convolutional units and cross-layer feature indexing, these manufacturing error features at different scales are fused to generate a comprehensive feature representation. This feature representation better reflects the distribution of manufacturing errors and provides more accurate input for subsequent error prediction.

[0059] Step S422: Generator of the adversarial network The training process uses the adversarial loss function , constraining the KL divergence between the generated error distribution and the measured error data x to be less than the divergence error threshold. Specifically, during the training of the GAN generator, an adversarial loss function is used, and the similarity between the generated error distribution and the measured error data is constrained by the Kullback-Leibler (KL) divergence. The adversarial loss function is the core of GAN training, which trains the generator by minimizing the difference between the data generated by the generator and the real data. The specific form is: ,in, Used to characterize the distribution of measured error data The expectation of , x is the measured error data, is the measured error data distribution, It refers to the judgment result of the discriminator D on the measured error data x. To characterize the generator Prior distribution of input The expectation of z is the generator The input sample, refers to the generator Prior distribution of the input, Is the discriminator D to the generator Generated error data The loss function trains the generator by minimizing the difference between the data generated by the generator and the real data. The goal of the generator is to generate data that is as realistic as possible, making it difficult for the discriminator to distinguish between the generated data and the real data; the goal of the discriminator is to distinguish between the generated data and the real data as accurately as possible. During the training of the generator, the KL divergence constraint is used to constrain the similarity between the generated error distribution and the measured error data. The specific method is to add a KL divergence term to the adversarial loss function so that the KL divergence between the generated error distribution and the measured error data is less than a preset divergence error threshold. During the training process, an optimization algorithm, such as Adam or RMSprop, is used to update the parameters.

[0060] Step S500: input the digital manufacturing instruction into a CNC machining center to drive a multi-axis linkage device to perform digital manufacturing management on the M bracket nodes.

[0061] Specifically, digital manufacturing instructions are input into a CNC machining center (a type of automated machining equipment that uses a computer control system to drive multi-axis linkage equipment to machine the workpiece according to preset machining paths and parameters). The corbel joint is then digitally manufactured using multi-axis linkage equipment (machinery with multiple motion axes, such as three or five axes, capable of high-precision machining of complex workpieces). Specifically, a CNC machining center (such as one from DMG MORI or MAZAK) and a welding robot (such as one from ABB or KUKA) are used. The generated digital manufacturing instructions are then imported into the CNC machining center's control system. The CNC machining center drives the multi-axis linkage equipment (such as a five-axis machining center) according to the instructions to machine the corbel joint, including cutting, drilling, and milling. The welding robot then performs welding operations according to the instructions, completing the welds at key points according to the preset motion paths and welding parameters. During the manufacturing process, sensors and monitoring systems monitor the equipment's operating status and machining quality in real time to ensure accuracy and reliability.

[0062] For example, in an automated production workshop, after receiving digital manufacturing instructions, a five-axis machining center begins machining the first corbel node. The machining center's cutting tools cut and drill the corbel node according to a preset path, while a welding robot simultaneously welds key welds according to the instructions. The entire manufacturing process is monitored in real time by the workshop's monitoring system. If an anomaly is detected (such as excessive welding current or excessive machining deviation), the system automatically issues an alarm and pauses processing to ensure manufacturing quality. The present embodiment of the application utilizes a method that first collects three-dimensional point cloud data of the corbel nodes in steel structural columns, then performs a stress concentration analysis and labels key weld points and bolt hole locations. Lightweight design parameters are then introduced for load stability analysis, assembly parameters are configured, and all corbel nodes are traversed. Digital manufacturing instructions are generated based on the three-dimensional point cloud data. Finally, the instructions are input into the CNC machining center for digital manufacturing management of the corbel nodes. These technical measures improve manufacturing efficiency, enhance manufacturing precision and stability, and provide strong support for the production of energy-saving building materials.

[0063] In one possible implementation, a multi-axis linkage device is driven to perform digital manufacturing management on the M corbel nodes, and step S500 further includes step S510, inserting a reverse compensation cutting trajectory into the CNC machining path based on the manufacturing error probability distribution map. Specifically, by analyzing the manufacturing error probability distribution map, areas with higher error probabilities are identified. These areas may be caused by welding heat effects or bolt preload errors. For the identified high error probability areas, reverse compensation cutting trajectories are planned. These trajectories are based on the original machining path and are offset in the opposite direction by a certain distance to offset potential errors. By modifying the CNC program or using CAM software, the planned reverse compensation cutting trajectory is inserted into the CNC machining path. Before actual machining, the CNC machining path after the compensation trajectory is inserted is simulated and verified to ensure that the compensation trajectory can effectively offset the error and will not affect the machining quality of other areas. If necessary, adjust the compensation trajectory.

[0064] For example, suppose a manufacturing error probability distribution map reveals a high probability of error in a particular weld area. To offset this error, a reverse compensating cutting path can be inserted into the CNC machining path for that area. This path is offset 0.2 mm in the opposite direction of the weld area from the original path. This approach automatically corrects potential weld errors during the actual machining process.

[0065] Step S520, matching the processing priority of each of the M corbel nodes, and optimizing the task scheduling sequence of the multi-axis linkage equipment. Specifically, the processing priority of each corbel node is evaluated based on factors such as the geometric complexity, processing difficulty, and assembly requirements of the corbel node. The task scheduling sequence of the multi-axis linkage equipment is optimized using an optimization algorithm (such as a genetic algorithm, a particle swarm optimization algorithm, etc.). The goal is to minimize the total processing time while ensuring the processing quality. Based on the optimization results, a final task scheduling sequence is generated. This sequence specifies the order and time for the multi-axis linkage equipment to process each corbel node. Processing tasks are executed according to the generated task scheduling sequence, and the equipment operating status is monitored in real time during the processing to ensure processing quality and efficiency.

[0066] For example, suppose that among M corbel nodes, node A has the highest geometric complexity and requires priority processing. Nodes B and C are relatively easy to process and can be scheduled for later processing. An optimization algorithm generates a task scheduling sequence that prioritizes processing node A, followed by nodes B and C. This scheduling approach ensures the processing quality of critical nodes while improving overall production efficiency.

[0067] In one possible implementation, the method further includes: deploying a fiber grating sensor at the end of the corbel support structure to collect strain spectrum data under dynamic load; extracting the proportion of low-frequency vibration energy in the strain signal based on the strain spectrum data under dynamic load; if the proportion of low-frequency vibration energy exceeds a preset safety threshold, triggering a structural reinforcement instruction.

[0068] Specifically, fiber Bragg grating (FBG) sensors are deployed at key locations on the corbel support structure, such as ends or areas of stress concentration. These sensors can accurately measure changes in structural strain under load. A data acquisition system, compatible with the FBG sensors, collects the strain signals output by the sensors in real time. This system includes a FBG interrogator, a data acquisition card, and a computer. Spectral analysis of the acquired strain signals is performed using Fourier transforms or other spectral analysis methods to extract the frequency components within the strain signals. Low-frequency vibration components, defined as components with frequencies below a certain threshold (e.g., 1 Hz), are extracted from the strain signal spectrum. The energy of the low-frequency components is integrated and divided by the total energy to calculate the percentage of the low-frequency vibration energy in the total energy. Low-frequency vibration is closely related to the stability and safety of the structure. Based on the structural design requirements and safety standards, a preset safety threshold, for example, 15%, is set. The low-frequency vibration energy percentage is monitored in real time and compared with the preset threshold. When the low-frequency vibration energy percentage exceeds the threshold, structural reinforcement instructions are automatically triggered. According to the structural strengthening instructions, take corresponding strengthening measures, such as adding supports, adjusting load distribution, etc., to improve the stability and safety of the structure.

[0069] For example, fiber Bragg grating sensors were deployed at the corbel joints of steel columns in a high-rise building to monitor strain changes under wind loads. A data acquisition system collected strain signals in real time and performed spectral analysis. The analysis revealed that low-frequency vibration energy accounted for 20%, exceeding the preset safety threshold of 15%. This automatically triggered a structural reinforcement order. Based on this order, additional support measures were implemented to improve structural stability and safety.

[0070] In the above, refer to Figure 1 The digital manufacturing method of the corbel node in the steel structure column according to the embodiment of the present invention is described in detail. Figure 2 A digital manufacturing system for a corbel node in a steel structure column according to an embodiment of the present invention is described.

[0071] The digital manufacturing system for steel column corbel joints according to an embodiment of the present invention is designed to address the technical issues of low efficiency, insufficient precision, and insufficient stability in existing steel column corbel manufacturing, thereby improving manufacturing efficiency, enhancing manufacturing precision, and stability, and providing strong support for the production of energy-saving building materials. The digital manufacturing system for steel column corbel joints includes a three-dimensional point cloud data acquisition module 10, a stress concentration analysis module 20, an assembly parameter configuration module 30, a digital manufacturing instruction sequence configuration module 40, and a digital manufacturing management module 50.

[0072] A three-dimensional point cloud data acquisition module 10 is used to collect three-dimensional point cloud data of the connection area of ​​the corbel node in the target steel structure column; a stress concentration analysis module 20 is used to perform stress concentration analysis in the vertical and horizontal directions based on the connection area, mark the first key welding point set and the first bolt hole position set, and mark the second key welding point set and the second bolt hole position set; an assembly parameter configuration module 30 is used to introduce the lightweight design parameters of the corbel support structure, perform load stability analysis based on the first key welding point set and the first bolt hole position set, and the second key welding point set and the second bolt hole position set, and configure the first assembly parameter; a digital manufacturing instruction sequence configuration module 40 is used to traverse the M corbel nodes in the target steel structure column, obtain the first assembly parameter, the second assembly parameter, and up to the Mth assembly parameter, and configure the digital manufacturing instruction sequence in combination with the three-dimensional point cloud data; a digital manufacturing management module 50 is used to input the digital manufacturing instruction into the CNC machining center and drive the multi-axis linkage equipment to perform digital manufacturing management on the M corbel nodes.

[0073] Among them, the system may further include: a strain spectrum data acquisition module for deploying a fiber grating sensor at the end of the corbel support structure to collect strain spectrum data under dynamic load; a low-frequency vibration energy ratio extraction module for extracting the low-frequency vibration energy ratio in the strain signal based on the strain spectrum data under dynamic load; a structure reinforcement instruction triggering module for triggering a structure reinforcement instruction if the low-frequency vibration energy ratio exceeds a preset safety threshold.

[0074] The specific configuration of the stress concentration analysis module 20 will be described in detail below. As described above, the stress concentration analysis module 20 may further include: a vertical stress concentration analysis unit for performing a stress concentration analysis in a vertical direction based on the top plane and the side bevel of the connection region, annotating a first set of key weld points and a first set of bolt hole locations; and a horizontal stress concentration analysis unit for performing a stress concentration analysis in a horizontal direction based on the side plane and the side bevel of the connection region, annotating a second set of key weld points and a second set of bolt hole locations.

[0075] The specific configuration of the assembly parameter configuration module 30 will be described in detail below. As described above, to configure the first assembly parameter, the assembly parameter configuration module 30 may further include: a discrete element simulation unit for performing discrete element simulation based on the first set of key welding points and the first set of bolt hole locations, and outputting the plastic deformation distribution of the corbel node under vertical load; a first unidirectional assembly parameter determination unit for optimizing the welding sequence and bolt preload combination based on the plastic deformation distribution under vertical load to determine the first unidirectional assembly parameter; and a first assembly parameter determination unit for determining the first assembly parameter based on the first unidirectional assembly parameter under vertical load and the second unidirectional assembly parameter under horizontal load.

[0076] Among them, the first unidirectional assembly parameter determination unit may further include: a stability coefficient determination subunit used for the first unidirectional assembly parameter belonging to the first unidirectional assembly parameter set, under horizontal load, correcting the spacing parameters of the second bolt hole position set by the friction energy consumption coefficient to determine the first stability coefficient, and at the same time, correcting the welding process parameters of the second key welding point set by the welding residual stress to determine the second stability coefficient; the first unidirectional assembly parameter set configuration subunit is used to use the first stability coefficient and the second stability coefficient to configure the first unidirectional assembly parameter set including the welding current gradient and the bolt torque sequence.

[0077] The specific configuration of the digital manufacturing instruction sequence configuration module 40 will be described in detail below. As described above, the M corbel nodes in the target steel structure column are traversed to obtain the first assembly parameter, the second assembly parameter, and the Mth assembly parameter, and the digital manufacturing instruction sequence is configured in combination with the three-dimensional point cloud data. The digital manufacturing instruction sequence configuration module 40 may further include: a joint encoding unit for using the first corbel node of the M corbel nodes to jointly encode the three-dimensional point cloud data and the first assembly parameter to determine a first joint encoding vector; a manufacturing error probability distribution map generation unit for configuring M joint encoding vectors based on the M corbel nodes, and inputting them into a generative adversarial network to obtain a manufacturing error probability distribution map, wherein the input layer of the adversarial network receives the M joint encoding vectors, and the output layer of the adversarial network generates the manufacturing error probability distribution map, wherein the welding heat input parameter is mapped to the welding heat effect superposition layer under the composite working condition area, and the bolt preload error is mapped to the bolt hole edge superposition layer under the composite working condition area.

[0078] The manufacturing error probability distribution map generation unit may further include: an adversarial network construction subunit for the hidden layer of the adversarial network to be composed of residual convolution units, to establish a cross-layer feature index relationship to fuse multi-scale manufacturing error features; a training subunit for the generator of the adversarial network The training process uses the adversarial loss function , constraining the KL divergence between the generated error distribution and the measured error data x to be less than the divergence error threshold.

[0079] The training subunit may further include: an adversarial loss function construction component for the adversarial loss function ,in, Used to characterize the distribution of measured error data The expectation of , x is the measured error data, is the measured error data distribution, It refers to the judgment result of the discriminator D on the measured error data x. To characterize the generator Prior distribution of input The expectation of z is the generator The input sample, refers to the generator Prior distribution of the input, Is the discriminator D to the generator Generated error data The judgment result of .

[0080] The specific configuration of the digital manufacturing management module 50 will be described in detail below. As described above, the multi-axis linkage device is driven to perform digital manufacturing management on the M corbel nodes. The digital manufacturing management module 50 may further include: a reverse compensation cutting trajectory insertion unit for inserting reverse compensation cutting trajectories into the NC machining path based on the manufacturing error probability distribution map; and a task scheduling sequence optimization unit for matching the machining priority of each of the M corbel nodes to optimize the task scheduling sequence of the multi-axis linkage device.

[0081] The digital manufacturing system for the corbel node in the steel structure column provided by the embodiment of the present invention can execute the digital manufacturing method for the corbel node in the steel structure column provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0082] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0083] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A digital manufacturing method for a corbel node in a steel structure column, characterized in that: The method comprises: Collect 3D point cloud data of the connection area of ​​the corbel node at the target steel structure column; Based on the connection area, stress concentration analysis is performed in the vertical direction and the horizontal direction respectively, and a first key welding point set and a first bolt hole position set are marked, and a second key welding point set and a second bolt hole position set are marked; Introducing lightweight design parameters of the corbel support structure, performing load stability analysis based on the first key welding point set and the first bolt hole position set, and the second key welding point set and the second bolt hole position set, and configuring first assembly parameters; Traversing M corbel nodes in the target steel structure column, obtaining a first assembly parameter, a second assembly parameter, and up to an Mth assembly parameter, and configuring a digital manufacturing instruction sequence in combination with the three-dimensional point cloud data; The digital manufacturing instructions are input into a CNC machining center to drive a multi-axis linkage device to perform digital manufacturing management on the M bracket nodes.

2. The digital manufacturing method of the corbel node in the steel structure column according to claim 1, characterized in that: The method comprises: Deploy a fiber Bragg grating sensor at the end of the corbel support structure to collect strain spectrum data under dynamic load; According to the strain spectrum data under dynamic load, the proportion of low-frequency vibration energy in the strain signal is extracted; If the proportion of the low-frequency vibration energy exceeds a preset safety threshold, a structural reinforcement instruction is triggered.

3. The digital manufacturing method of the corbel joint in the steel structure column according to claim 1, characterized in that: The connection area includes a top plane and a side plane; Based on the top plane and the side bevel of the connection area, a stress concentration analysis is performed in the vertical direction, and a first key welding point set and a first bolt hole position set are marked; Based on the side plane and the side bevel of the connection area, a stress concentration analysis is performed in the horizontal direction, and a second key welding point set and a second bolt hole position set are marked.

4. The digital manufacturing method of the corbel joint in the steel structure column according to claim 3, characterized in that: Configuring first assembly parameters, the method includes: Performing discrete element simulation based on the first key welding point set and the first bolt hole position set to output the plastic deformation distribution of the corbel node under the vertical load; According to the distribution of plastic deformation under vertical load, the welding sequence and bolt preload combination are optimized to determine the first unidirectional assembly parameters; The first assembly parameter is determined based on the first unidirectional assembly parameter under vertical load and the second unidirectional assembly parameter under horizontal load.

5. The digital manufacturing method of the corbel joint in the steel structure column according to claim 4, characterized in that: The first unidirectional assembly parameter belongs to a first unidirectional assembly parameter set; Under horizontal load, the spacing parameters of the second bolt hole position set are corrected by the friction energy dissipation coefficient to determine the first stability coefficient, and at the same time, the welding process parameters of the second key welding point set are corrected by the welding residual stress to determine the second stability coefficient; The first stability coefficient and the second stability coefficient are used to configure a first unidirectional assembly parameter set including a welding current gradient and a bolt torque sequence.

6. The digital manufacturing method of the corbel joint in the steel structure column according to claim 5, characterized in that: Traversing M corbel nodes in the target steel structure column, obtaining first assembly parameters, second assembly parameters, and up to Mth assembly parameters, and configuring a digital manufacturing instruction sequence in combination with the three-dimensional point cloud data, the method further includes: Using a first corbel node among the M corbel nodes, jointly encoding the three-dimensional point cloud data and the first assembly parameter to determine a first joint encoding vector; Based on the M corbel nodes, M joint coding vectors are configured and input into a generative adversarial network to obtain a manufacturing error probability distribution map, wherein an input layer of the adversarial network receives the M joint coding vectors, and an output layer of the adversarial network generates the manufacturing error probability distribution map; Among them, the welding heat input parameters are mapped to the welding heat influence superposition layer under the composite working condition area, and the bolt preload error is mapped to the bolt hole edge superposition layer under the composite working condition area.

7. The digital manufacturing method of the corbel joint in the steel structure column according to claim 6, characterized in that: The hidden layer of the adversarial network is composed of residual convolution units, which establishes a cross-layer feature index relationship to fuse multi-scale manufacturing error features; The generator of the adversarial network The training process uses the adversarial loss function , constraining the KL divergence between the generated error distribution and the measured error data x to be less than the divergence error threshold.

8. The digital manufacturing method for the corbel joint in the steel structure column according to claim 7, characterized in that: The adversarial loss function ; in, Used to characterize the distribution of measured error data The expectation of , x is the measured error data, is the measured error data distribution, It refers to the judgment result of the discriminator D on the measured error data x. To characterize the generator Prior distribution of input The expectation of z is the generator The input sample, refers to the generator Prior distribution of the input, Is the discriminator D to the generator Generated error data The judgment result of .

9. The digital manufacturing method for a corbel joint in a steel structure column according to claim 6, characterized in that: Driving a multi-axis linkage device to perform digital manufacturing management on the M bracket nodes, the method includes: inserting a reverse compensation cutting trajectory into a numerical control machining path based on the manufacturing error probability distribution map; The processing priority of each of the M corbel nodes is matched to optimize the task scheduling sequence of the multi-axis linkage device.

10. The digital manufacturing system of the corbel joint in the steel structure column is characterized by: The system is used to implement the digital manufacturing method of the corbel node in the steel structure column according to any one of claims 1 to 9, and the system comprises: 3D point cloud data acquisition module, used to collect 3D point cloud data of the connection area of ​​the corbel node at the target steel structure column; A stress concentration analysis module is used to perform stress concentration analysis in the vertical direction and the horizontal direction based on the connection area, mark the first key welding point set and the first bolt hole position set, and mark the second key welding point set and the second bolt hole position set; an assembly parameter configuration module, configured to introduce lightweight design parameters of the corbel support structure, perform load stability analysis based on the first key welding point set and the first bolt hole position set, and the second key welding point set and the second bolt hole position set, and configure first assembly parameters; a digital manufacturing instruction sequence configuration module, configured to traverse the M corbel nodes in the target steel structure column, obtain a first assembly parameter, a second assembly parameter, and up to an Mth assembly parameter, and configure a digital manufacturing instruction sequence in combination with the three-dimensional point cloud data; The digital manufacturing management module is used to input the digital manufacturing instructions into the CNC machining center and drive the multi-axis linkage equipment to perform digital manufacturing management on the M bracket nodes.

Images