A method, medium and system for determining welding points of grid ball nodes

Abstract

The present invention provides a method, medium and system for determining welding points of grid ball nodes, belonging to the technical field of steel structure construction, including: based on the overall design drawings of the grid, a finite element model is established and the influence of dynamic load is considered. Static analysis is performed in the model to calculate the stress, strain distribution and internal force transmission path of each node. Then, dynamic analysis is performed to simulate the vibration response of the grid under the action of dynamic load, and the comprehensive stress and strain situation is obtained by superposition with the static results. Next, multiple welding schemes are determined in combination with design standards and manufacturing processes. These schemes are optimized using a non-dominated sorting genetic algorithm, and the scheme with the smallest stress and strain concentration is selected as the pending scheme. A grid test piece is made, a mechanical property test is performed, and the finite element model is adjusted and optimized using a response surface method or a Lagrange multiplier method, and finally the welding scheme with the smallest stress and strain concentration is selected as the target scheme, and the coordinates of the welding points of each grid ball are output.

Description

Technical Field

[0001] The invention belongs to the technical field of steel structure construction, and in particular relates to a method, a medium and a system for determining welding points of grid ball nodes. Background Art

[0002] As an important form of steel structure, grid is widely used in large-span public buildings, gymnasiums, stations and other fields. It is favored for its lightweight and beautiful appearance. The main force-bearing unit of the grid structure is composed of multiple ball nodes composed of steel pipes or steel rods. Each node is firmly connected by welding to form an overall rigid structural system. The quality of grid welding joints is directly related to the bearing capacity and service life of the entire structure. Therefore, how to determine the optimal welding position and weld size of the grid ball node has always been one of the key issues in the grid design and manufacturing process.

[0003] The existing grid welding design methods usually adopt empirical formulas or finite element analysis. The empirical formula method mainly gives some guiding weld size and welding position requirements based on the accumulated engineering design experience. For example, GB50017 "Steel Structure Design Standard" puts forward corresponding regulations for the weld design of grid nodes. However, this method is too simplistic and it is difficult to accurately reflect the actual stress condition of the grid under complex loads, and it cannot ensure the structural safety of the welded joints. The finite element analysis method can simulate the mechanical behavior of the grid more comprehensively, but most of the existing research only stays at the static analysis level, and rarely considers the influence of dynamic loads (such as earthquakes, wind vibration, etc.) on the quality of grid welding. This analysis method that ignores the dynamic effect may lead to safety hazards in grid welding design.

[0004] The existing grid welding points are often selected based on the experience of welding construction personnel, and the optimal welding points cannot be selected, which affects the overall performance of the grid. Summary of the invention

[0005] In view of this, the present invention provides a method, medium and system for determining the welding points of the grid ball nodes, which can solve the technical problem that the determination of the existing grid welding points is often based on the experience of the welding construction personnel, and the optimal welding points cannot be selected, which affects the overall performance of the grid.

[0006] The present invention is achieved in that:

[0007] A first aspect of the present invention provides a method for determining a grid ball node welding point, which comprises the following steps:

[0008] S10. Establish a finite element model based on the overall design drawings of the grid and obtain the dynamic loads to which the grid is subjected as predicted by experts;

[0009] S20, in the finite element model, setting a static load condition of the grid to perform static analysis;

[0010] S30, in the finite element model, performing dynamic analysis on the grid structure, simulating the vibration response of the grid under the action of dynamic load, and calculating the dynamic stress and strain of each node;

[0011] S40, according to the static analysis results, calculate the stress, strain distribution and internal force transmission path of each ball node;

[0012] S50, according to the dynamic analysis results, correcting and superimposing the stress, strain distribution and internal force transmission path of each spherical node calculated by the static analysis results, to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads;

[0013] S60, obtaining multiple welding schemes formed by construction personnel based on welding positions and weld sizes of each ball node;

[0014] S70, optimizing and calculating different welding schemes, taking into account the stress, strain distribution and internal force transmission path under the combined action of static and dynamic loads, and selecting the welding scheme with the smallest stress and strain concentration as the pending scheme;

[0015] S80, obtaining a grid specimen manufactured according to the pending scheme, performing a mechanical property test on the welded joint, and adjusting and optimizing the finite element model according to the experimental results;

[0016] S90. Based on the optimized finite element model, repeat steps S20-S70, select the welding scheme with the smallest stress-strain concentration as the target scheme, and output the welding point coordinates of each grid ball node.

[0017] Among them, the dynamic loads that the grid is subjected to predicted by the experts include inertial force caused by building vibration and earthquake.

[0018] The step S10 specifically includes:

[0019] The overall design drawings of the grid are used as input information to establish a finite element analysis model. First, the unit grid is reasonably divided according to the geometric dimensions, material properties and other information of the grid, such as the grid unit size of 1 to 5 cm, to ensure that the finite element model can accurately reflect the structural characteristics of the actual grid. Secondly, obtain the dynamic load information that may act on the grid, including the inertial force and seismic load caused by the vibration of the building. By performing a dynamic analysis on the building structure connected to the grid, the inertial force caused by the vibration of the building can be determined; the seismic load needs to be evaluated according to the seismic characteristics of the area where the grid is located, and the seismic load spectrum or time history curve given in the relevant specifications or standards can be referred to. Through this step, a finite element analysis model that can accurately simulate the behavior of the grid under dynamic action is established.

[0020] The static load conditions of the grid include deadweight, service load and wind load.

[0021] The specific steps of step S20 include:

[0022] The working conditions of the grid structure under the action of static loads are set in the finite element analysis model. These static loads include deadweight, service load and wind load. Deadweight refers to the gravity load of the grid itself; service load refers to various external loads that the grid may bear during use, such as snow load on the roof, personnel load, etc.; wind load refers to the load caused by wind. These static load parameters can be set according to the reference values ​​given in the relevant specifications or standards. Then, the finite element analysis model is subjected to static analysis to obtain the stress, strain and displacement distribution of the grid under various static loads, providing basic data for subsequent dynamic analysis and welding point determination.

[0023] Among them, the method for dynamic analysis of grid structure is time domain method or frequency domain method.

[0024] The specific steps of step S30 include:

[0025] In the finite element analysis model, the grid structure is subjected to dynamic analysis. The purpose of dynamic analysis is to simulate the vibration response characteristics of the grid under dynamic loads, including the dynamic stress and strain of each node. Specifically, the time domain method or the frequency domain method can be used for analysis. The time domain method directly solves the response characteristics of the grid under dynamic loads that change with time, while the frequency domain method solves the vibration response of the grid in the frequency domain by Fourier transforming the dynamic load. The two types of dynamic loads obtained in step S10 need to be considered in the dynamic analysis: the inertial force caused by the vibration of the building and the seismic load. Through dynamic analysis, the dynamic stress and strain distribution of each node of the grid under dynamic loads can be obtained, which provides a basis for the subsequent determination of welding points.

[0026] The specific steps of step S40 include:

[0027] From the static analysis results of step S20, the stress and strain data of each ball node of the grid are extracted, including the principal stress, equivalent stress and shear stress at the node. By analyzing these stress and strain data, it is possible to determine which nodes have larger stress and strain and may become potential stress concentration areas. At the same time, researchers need to analyze the internal force transmission path of each ball node. The internal force transmission path reflects the law of force transmission and is conducive to determining the location of the welding point. By tracking and analyzing the internal force of the rod in the finite element model, the main force path at each ball node can be obtained. In summary, the purpose of step S40 is to further determine the stress and strain distribution characteristics and internal force transmission path of each ball node of the grid on the basis of static analysis, so as to provide a basis for the subsequent determination of welding points.

[0028] Among them, the method for correcting and superimposing the stress, strain distribution and internal force transmission path of each ball node calculated by the static analysis results is specifically a linear superposition method, that is, simply adding the stress and strain values ​​obtained by static analysis and dynamic analysis.

[0029] The specific steps of step S50 include:

[0030] According to the dynamic analysis results of step S30, it is combined and superimposed with the static analysis results of step S40 to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads. This combination can be performed by a linear superposition method, that is, the stress and strain values ​​obtained by the static analysis and the dynamic analysis are simply added. In some cases, there may be a large difference between the results of static and dynamic analysis. At this time, a more complex combination method, such as the response spectrum method or random vibration theory, is required to correct the superposition relationship between the two. Through step S50, the stress, strain distribution and internal force transmission path of each ball node of the grid under the combined action of static and dynamic loads can be obtained, providing more accurate data support for the subsequent determination of welding points.

[0031] The specific steps of step S60 include:

[0032] First, refer to GB50017 "Steel Structure Design Standard" and other relevant design standards to determine the minimum and maximum sizes of various welds. These size standards are generally related to factors such as node bearing capacity and fatigue performance. Secondly, based on manufacturing process data, such as welding process parameters, welding equipment information, etc., combined with the geometric characteristics of the node and the internal force transmission path, determine the possible welding positions of each ball node. The selection of welding positions should avoid stress concentration areas and try to ensure the rationality of weld length and weld shape. Finally, based on the above information, generate multiple different welding schemes to lay the foundation for subsequent optimization calculations. These schemes may involve a combination of multiple factors such as weld length, weld shape, and weld position.

[0033] Among them, the non-dominated sorting genetic algorithm is used for optimization calculation of different welding schemes.

[0034] The specific steps of step S70 include:

[0035] The non-dominated sorting genetic algorithm (NSGA-II) is used to optimize the multiple welding schemes determined in step S60. The objective functions that need to be optimized include: the equivalent stress or principal stress of each ball node under the combined action of static and dynamic loads (reflecting the structural safety of the welded joint), the strain of each ball node under the combined action of static and dynamic loads (reflecting the deformation coordination of the welded joint), and the rationality of the internal force transmission path of each ball node (reflecting the force transmission efficiency). Through the iterative optimization of the NSGA-II algorithm, a set of non-dominated solutions under the above three objective functions can be obtained, from which the welding scheme with the smallest stress and strain concentration is selected as the pending scheme. The NSGA-II algorithm has excellent convergence and distribution, and is suitable for solving the multi-objective optimization problem in this problem.

[0036] Among them, the response surface method or Lagrange multiplier method is used to adjust and optimize the finite element model.

[0037] The specific steps of step S80 include:

[0038] According to the pending welding scheme selected in step S70, the corresponding grid specimen is made, and the welded joint is subjected to static loading test and dynamic loading test. The static loading test can test the bearing capacity of the welded joint under static load, while the dynamic loading test can evaluate the fatigue performance of the joint under dynamic load. The test results can be compared and analyzed with the prediction results of the finite element model. If there is a large deviation, it is necessary to use numerical optimization techniques such as response surface method or Lagrange multiplier method to correct and optimize the geometric parameters, material properties, etc. of the finite element model so that it can more accurately predict the mechanical behavior of the actual grid. Through step S80, the mechanical properties of the finite element model and the actual grid can be checked and improved, providing a more reliable data basis for the subsequent determination of welding points.

[0039] The specific steps of step S90 include:

[0040] After optimizing and correcting the finite element analysis model in step S80, steps S20 to S70 are repeated to finally determine the coordinates of the welding points of each ball node of the grid. Specifically, first, the static load condition is reset in the optimized finite element model to perform static analysis; then, on this basis, the grid structure is subjected to dynamic analysis to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads. Next, according to the design standards and manufacturing process information, the possible welding schemes of each ball node are determined, and the non-dominated sorting genetic algorithm is used for optimization to select the scheme with the smallest stress and strain concentration as the final scheme. Finally, based on the final scheme, the coordinates of the welding points of each grid ball node are directly output, providing a direct technical basis for the subsequent grid manufacturing and assembly.

[0041] The multiple welding schemes formed by the construction personnel based on the welding positions and weld sizes of each ball node specifically include a welding position diagram and a welding process specification.

[0042] Welding position diagram: Identify the specific welding position of each ball node and provide the length, width and other dimensional information of each weld; Welding process specification: Select the appropriate welding process and determine the welding parameters such as welding current, voltage, speed, etc. Establish the welding sequence and welding seam.

[0043] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are run, they are used to execute the above-mentioned method for determining the welding points of the grid ball nodes.

[0044] A third aspect of the present invention provides a grid ball node welding point determination system, which includes the above-mentioned computer-readable storage medium.

[0045] Compared with the prior art, the method, medium and system for determining the welding points of the grid ball nodes provided by the present invention have the following beneficial effects:

[0046] 1. The influence of dynamic loads (such as earthquakes and wind vibration) on the grid structure is fully considered. By establishing a finite element analysis model that can simulate the vibration response of the grid under dynamic action, the stress and strain distribution of each node under the combined action of static and dynamic loads is accurately predicted, providing more reliable data support for the determination of welding points. This can not only ensure the structural safety of the welded joints under harsh dynamic environments, but also help improve the overall earthquake and wind resistance of the grid.

[0047] 2. A multi-objective optimization method was adopted, taking into account multiple factors such as structural safety, force transmission efficiency, and deformation coordination. This not only ensures that the welded joints meet the requirements in terms of bearing capacity and fatigue performance, but also improves the overall mechanical properties of the grid, such as reducing stress concentration and optimizing the internal force transmission path. Compared with single-objective optimization, this method can better balance the multiple performance requirements of the grid structure.

[0048] 3. The finite element analysis model was calibrated and optimized by making grid specimens and conducting static and dynamic tests. This not only verifies the accuracy of the theoretical analysis, but also provides valuable data support for further improvement of the finite element model, thereby improving the reliability of the analysis results.

[0049] In general, the method for determining the welding points of the grid ball nodes proposed in the present invention fully considers the dynamic effect, adopts a multi-objective optimization strategy, and combines experimental verification to determine the optimal welding solution that meets the multi-faceted performance requirements of the grid structure, thereby obtaining multiple optimized welding points. This solves the technical problem that the determination of the existing grid welding points is often based on the experience of the welding construction personnel, and the optimal welding points cannot be selected, which affects the overall performance of the grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0051] Figure 1 A flow chart of the method provided by the present invention;

[0052] Figure 2 It is a schematic diagram of grid support for specific application scenarios;

[0053] Figure 3 This is a schematic diagram of the grid installation process steps. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0055] like Figure 1 FIG. 1 is a flow chart of a method for determining welding points of a grid ball node provided by the present invention. The method comprises the following steps:

[0056] S10. According to the overall design drawings of the grid, establish a finite element model and obtain the dynamic loads that the grid will be subjected to as predicted by experts, including inertial force and earthquake caused by building vibration;

[0057] S20. In the finite element model, static load conditions of the grid are set, including deadweight, service load, and wind load, to perform static analysis;

[0058] S30, in the finite element model, performing dynamic analysis on the grid structure, simulating the vibration response of the grid under the action of dynamic load, and calculating the dynamic stress and strain of each node;

[0059] S40, according to the static analysis results, calculate the stress, strain distribution and internal force transmission path of each ball node;

[0060] S50, according to the dynamic analysis results, correcting and superimposing the stress, strain distribution and internal force transmission path of each spherical node calculated by the static analysis results, to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads;

[0061] S60, determining the welding position and weld size of each ball node, and forming multiple welding schemes;

[0062] S70, using a non-dominated sorting genetic algorithm to optimize the calculation of different welding schemes, while considering the stress, strain distribution and internal force transmission path under the combined action of static and dynamic loads, and selecting the welding scheme with the smallest stress and strain concentration as the pending scheme;

[0063] S80, obtaining a grid specimen made according to the pending scheme, performing a mechanical property test on the welded joint, and adjusting and optimizing the finite element model using a response surface method or a Lagrange multiplier method according to the experimental results;

[0064] S90. Based on the optimized finite element model, repeat steps S20-S70, select the welding scheme with the smallest stress-strain concentration as the target scheme, and output the welding point coordinates of each grid ball node.

[0065] The specific implementation methods of the above steps are described in detail below:

[0066] The specific implementation of step S10 is as follows:

[0067] According to the overall design drawings of the grid, the researchers first need to establish a finite element model. The finite element method is a numerical analysis method that discretizes continuous structures and can simulate the actual deformation and stress state of the structure under external loads. During the modeling process, it is necessary to reasonably divide the unit grid according to the geometric dimensions, material properties and other information of the grid to ensure that the finite element model can accurately reflect the structural characteristics of the actual grid.

[0068] After the finite element model is established, the next step is to obtain information about the dynamic loads that may act on the grid. There are two main types of dynamic loads: one is the inertial force caused by the vibration of the building, and the other is the seismic load. The inertial force caused by the vibration of the building can be obtained by dynamic analysis of the building structure connected to the grid. The seismic load needs to be evaluated using deterministic or probabilistic methods based on the seismic characteristics of the area where the grid is located. Usually, the seismic load spectrum or time history curve given in the relevant specifications or standards can be referred to.

[0069] In summary, the purpose of step S10 is to establish a finite element analysis model that can accurately simulate the behavior of the grid under dynamic action, laying the foundation for subsequent static and dynamic analysis. In the modeling process, it is necessary to fully consider the geometric characteristics and environment of the grid, and obtain the dynamic load information that may act on the grid.

[0070] The specific implementation of step S20 is as follows:

[0071] In the finite element model, researchers first need to set up the working conditions of the grid structure under the action of static loads, which include three types of static loads: deadweight, service load and wind load.

[0072] Deadweight refers to the gravity load of the grid itself. Service load refers to the various external loads that the grid may bear during use, such as snow load on the roof, personnel load, etc. Wind load refers to the load caused by wind. These static loads can be set according to the reference values ​​given in the relevant specifications or standards.

[0073] After obtaining the static load conditions, researchers need to perform static analysis on the finite element model. The purpose of static analysis is to find out the stress, strain and displacement distribution of the grid under various static loads. This provides basic data for subsequent dynamic analysis and welding point determination.

[0074] In summary, step S20 is intended to determine the response characteristics of the grid under static load, including stress, strain and displacement distribution, which provides necessary data support for subsequent steps.

[0075] The specific implementation of step S30 is as follows:

[0076] In the finite element model, researchers need to conduct dynamic analysis on the grid structure. The purpose of dynamic analysis is to simulate the vibration response characteristics of the grid under dynamic loads, including the dynamic stress and strain of each node.

[0077] Dynamic analysis can be performed using the time domain method or the frequency domain method. The time domain method directly solves the response characteristics of the grid under dynamic loads that change with time, while the frequency domain method solves the vibration response of the grid in the frequency domain by performing Fourier transform on the dynamic loads. Generally, the time domain method can better reflect the actual vibration behavior of the grid under dynamic loads.

[0078] In the dynamic analysis, two types of dynamic loads obtained in step S10 need to be considered: the inertial force caused by the vibration of the building and the seismic load. Among them, the inertial force caused by the vibration of the building can be simulated by establishing a coupled building-grid finite element model; the seismic load can be solved by time history analysis or response spectrum analysis.

[0079] Through dynamic analysis, researchers can obtain the dynamic stress and strain distribution of each node of the grid under dynamic load, which provides an important basis for the subsequent determination of welding points.

[0080] The specific implementation of step S40 is as follows:

[0081] Based on the static analysis results completed in step S20, researchers need to calculate the stress, strain distribution and internal force transmission path of each ball node of the truss.

[0082] First, the stress and strain data of each spherical node are extracted from the static analysis results. This includes the principal stress, equivalent stress, and shear stress at the node. By analyzing these stress and strain data, it is possible to preliminarily determine which nodes have larger stress and strain and may become potential stress concentration areas.

[0083] Secondly, researchers need to analyze the internal force transmission path of each ball node. The internal force transmission path reflects the law of force transmission and is helpful to determine the location of the welding point. By tracking and analyzing the internal force of the rod in the finite element model, the main force path at each ball node can be obtained.

[0084] In summary, the purpose of step S40 is to further determine the stress and strain distribution characteristics and internal force transmission paths of each spherical node of the grid based on the static analysis, so as to provide a basis for the subsequent determination of welding points.

[0085] The specific implementation of step S50 is as follows:

[0086] On the basis of step S40, researchers need to correct and superimpose the stress, strain distribution and internal force transmission path of each ball node obtained by static analysis according to the dynamic analysis results in step S30.

[0087] The results of dynamic analysis reflect the actual vibration response characteristics of the grid under dynamic loads, so it is necessary to combine them with the results of static analysis to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads. This combination can be performed using the linear superposition method, that is, simply adding the stress and strain values ​​obtained from static analysis and dynamic analysis.

[0088] In some cases, the results of static and dynamic analysis may be quite different. In this case, it is necessary to use more complex combined methods, such as response spectrum method or random vibration theory, to correct the superposition relationship between the two.

[0089] Through step S50, researchers can obtain the stress, strain distribution and internal force transmission path of each ball node of the truss under the combined action of static and dynamic loads, providing more accurate data support for the subsequent determination of welding points.

[0090] The specific implementation of step S60 is as follows:

[0091] Based on the stress, strain distribution and internal force transmission path information of each ball node obtained in steps S40 and S50, combined with the design standards and manufacturing process data, researchers need to determine the possible welding positions and weld sizes of each ball node, thereby forming multiple welding schemes.

[0092] Specifically, we first need to refer to relevant design standards, such as GB 50017 "Steel Structure Design Standard", to determine the minimum and maximum sizes of various welds. These size standards are generally related to factors such as node bearing capacity and fatigue performance.

[0093] Secondly, according to the manufacturing process data, such as welding process parameters, welding equipment information, etc., combined with the geometric characteristics of the node and the internal force transmission path, determine the possible welding positions of each ball node. The selection of welding position should avoid the stress concentration area and try to ensure the rationality of the weld length and weld shape.

[0094] Finally, researchers can generate multiple different welding schemes based on the above information, laying the foundation for subsequent optimization calculations. These schemes may involve a combination of multiple factors such as weld length, weld shape, and weld position.

[0095] In summary, the purpose of step S60 is to make full use of the existing static and dynamic analysis results, combined with design standards and process requirements, to determine the possible welding schemes for each ball node for subsequent optimization calculations.

[0096] The specific implementation of step S70 is as follows:

[0097] Among the multiple welding schemes determined in step S60, researchers need to use a non-dominated sorting genetic algorithm (NSGA-II) to optimize and calculate these schemes, so as to select the welding scheme with the smallest stress-strain concentration as the pending scheme.

[0098] The non-dominated sorting genetic algorithm is a multi-objective optimization algorithm that can optimize multiple objective functions at the same time and obtain a set of non-dominated solutions. In this problem, the objective functions that need to be optimized include:

[0099] 1. Equivalent stress or principal stress of each spherical node under the combined action of static and dynamic loads: This objective function reflects the structural safety of the welded joint.

[0100] 2. The strain of each spherical node under the combined action of static and dynamic loads: This objective function reflects the deformation coordination of the welded joint.

[0101] 3. Rationality of the internal force transmission path of each ball node: This objective function reflects the efficiency of force transmission.

[0102] Through the iterative optimization of the NSGA-II algorithm, a set of non-dominated solutions under the above three objective functions can be obtained, that is, solutions in which other objective functions will not deteriorate while one objective function is improved. From this set of non-dominated solutions, the researchers selected the welding scheme with the smallest stress and strain concentration as the pending solution.

[0103] The NSGA-II algorithm has excellent convergence and distribution, and can effectively find non-dominated solutions close to the true Pareto frontier. The algorithm is also highly efficient and suitable for solving multi-objective optimization problems in this problem.

[0104] In summary, the purpose of step S70 is to optimize various welding schemes under static and dynamic loads through a non-dominated sorting genetic algorithm, and select the scheme with the smallest stress and strain concentration as the pending scheme.

[0105] The specific implementation of step S80 is as follows:

[0106] According to the pending welding scheme selected in step S70, researchers need to make grid specimens and conduct mechanical property tests on the welded joints, so as to adjust and optimize the finite element model.

[0107] Specifically, first, the corresponding grid test piece needs to be manufactured according to the pending welding scheme obtained in step S70. During the manufacturing process, the welding quality needs to be strictly controlled to ensure that the manufacturing process of the test piece is consistent with the actual production process.

[0108] Then, the mechanical properties of the manufactured grid specimens are tested. This includes static loading test and dynamic loading test. The static loading test can test the load-bearing capacity of the welded joint under static load, while the dynamic loading test can evaluate the fatigue performance of the joint under dynamic load.

[0109] The test results can be compared and analyzed with the prediction results of the finite element model. If there is a large deviation, it is necessary to use numerical optimization techniques such as response surface method or Lagrange multiplier method to correct and optimize the geometric parameters and material properties of the finite element model so that it can more accurately predict the mechanical behavior of the actual grid.

[0110] Through step S80, researchers can verify and improve the mechanical properties of the finite element model and the actual grid, providing a more reliable data basis for the subsequent determination of welding points.

[0111] The specific implementation of step S90 is as follows:

[0112] After optimizing and correcting the finite element model in step S80, researchers need to repeat steps S20 to S70 to finally determine the coordinates of the welding points of each ball node of the grid.

[0113] Specifically, first, it is necessary to reset the static load conditions in the optimized finite element model and perform static analysis (step S20). Then, on this basis, a dynamic analysis is performed on the grid structure to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads (steps S30, S40, S50).

[0114] Next, according to the design standards and manufacturing process information, possible welding schemes for each ball node are determined (step S60), and these schemes are optimized using a non-dominated sorting genetic algorithm (step S70), and the scheme with the smallest stress-strain concentration is selected as the final scheme.

[0115] Finally, based on the final solution, researchers can directly output the welding point coordinates of each grid ball node. These welding point coordinate information can provide a direct technical basis for subsequent grid manufacturing and assembly.

[0116] In general, the purpose of step S90 is to repeat the previous analysis and optimization steps based on the optimized finite element model, and finally determine the optimal welding point coordinates of each ball node of the grid.

[0117] In order to better understand the present method, a specific embodiment of the present method is provided below in combination with a specific formula:

[0118] The specific implementation of step S10 is as follows:

[0119] According to the overall design drawings of the grid, the finite element analysis model needs to be established first. The finite element model can better simulate the actual deformation and stress state of the structure under external loads by discretizing the continuous structure into a finite number of units. When establishing the finite element model, it is necessary to reasonably divide the unit grid according to the geometric dimensions, material properties and other information of the grid to ensure that the model can accurately reflect the structural characteristics of the actual grid.

[0120] Assuming that the finite element model of the grid consists of N nodes, each node has m degrees of freedom, then the number of degrees of freedom of the entire grid is n = N × m. The degree of freedom of the node can be expressed as the displacement u i = {u i ,v i ,w i} T and rotation θ i ={θ ix ,θ iy ,θ iz} T Two parts, u i ,v i ,w i Represents the displacement component of the ith node in the directions of the three coordinate axes x, y, and z, so the total degree of freedom of the node is m=6.

[0121] After establishing the finite element model, it is necessary to obtain the dynamic load information that may act on the grid. There are mainly two types of dynamic loads:

[0122] The inertial force caused by the vibration of the building. The inertial force load on the grid can be obtained by conducting a dynamic analysis on the building structure connected to the grid, establishing a finite element model of the building-grid coupling, and solving the vibration characteristics of the building:

[0123]

[0124] Among them, M i is the mass matrix of node i, is the acceleration of node i at time t.

[0125] Earthquake load. According to the earthquake characteristics of the area where the grid is located, a deterministic or probabilistic method can be used for evaluation. Usually, the earthquake load spectrum or time history curve given in the relevant specifications or standards can be referred to to establish the following earthquake force load model:

[0126]

[0127] Where Γ is the earthquake amplification factor, is the earthquake acceleration time history of the foundation.

[0128] In summary, the purpose of step S10 is to establish a finite element analysis model that can accurately simulate the behavior of the grid under dynamic action, laying the foundation for subsequent static and dynamic analysis. In the modeling process, it is necessary to fully consider the geometric characteristics and environment of the grid, and obtain the dynamic load information that may act on the grid.

[0129] The specific implementation of step S20 is as follows:

[0130] In the finite element analysis model established in the previous step, it is necessary to set the working conditions of the grid structure under the action of static loads. There are three types of static loads:

[0131] Self-weight: The gravity load of the grid itself, which can be expressed as Where ρ is the material density, V i is the volume of the unit where node i is located, and g is the gravitational acceleration vector.

[0132] Use load: Various external loads that the grid may bear during use, such as snow load on the roof, personnel load, etc. These load parameters can be set according to the reference values ​​given in the relevant specifications or standards.

[0133] Wind load: The load caused by wind force can be The size of wind load is related to factors such as wind speed and windward area, and can also be set by referring to relevant standards.

[0134] In summary, the total load of the grid under static load can be expressed as:

[0135]

[0136] After setting the static load conditions, the finite element model needs to be subjected to static analysis. The purpose of static analysis is to find out the stress, strain and displacement distribution of the grid under various static loads. The static equilibrium equation can be expressed as:

[0137] Ku=F static

[0138] Among them, K is the overall stiffness matrix and u is the overall displacement vector. By solving this equation, the displacement, stress and strain of each node of the grid can be obtained, providing basic data for subsequent dynamic analysis and welding point determination.

[0139] The specific implementation of step S30 is as follows:

[0140] In the finite element analysis model, it is necessary to conduct dynamic analysis on the grid structure. The purpose of dynamic analysis is to simulate the vibration response characteristics of the grid under dynamic loads, including the dynamic stress and strain of each node.

[0141] Dynamic analysis can be performed using the time domain method or the frequency domain method. The time domain method directly solves the response characteristics of the grid under dynamic loads that change with time, and its dynamic equilibrium equation is:

[0142]

[0143] Where M is the overall mass matrix, C is the overall damping matrix, and u(t) are the acceleration, velocity and displacement vector of the node respectively, F dynamic (t) is the dynamic load vector.

[0144] The frequency domain law solves the vibration response of the grid in the frequency domain by Fourier transforming the dynamic load. Its dynamic equilibrium equation is:

[0145] (-ω 2 M+iωC+K)U(ω)=F dynamic (ω)

[0146] Where ω is the angular frequency, U(ω) and F dynamic (ω) are the expressions of displacement and dynamic load in frequency domain respectively.

[0147] Whether the time domain method or the frequency domain method is used, the two types of dynamic loads obtained in step S10 need to be considered during the dynamic analysis: the inertial force caused by the building vibration and the seismic load. Through dynamic analysis, the dynamic stress and strain distribution of each node of the grid under the action of dynamic load can be obtained, providing a basis for the subsequent determination of welding points.

[0148] The specific implementation of step S40 is as follows:

[0149] According to the static analysis results of step S20, it is necessary to calculate the stress and strain distribution of each ball node of the grid, as well as the internal force transmission path.

[0150] First, the stress and strain data of each spherical node are extracted from the static analysis results. For node i, its principal stress σ i1 ,σ i2 ,σ i3 and equivalent stress σ iev It can be expressed as:

[0151]

[0152] Among them, σ ix ,σ iy ,σ iz is the principal stress component at node i, τ ixy ,τ iyz ,τ izxis the shear stress component. By analyzing these stress and strain data, we can preliminarily determine which nodes have larger stress and strain and may become potential stress concentration areas.

[0153] Secondly, it is necessary to analyze the internal force transmission path of each ball node. The internal force transmission path reflects the law of force transmission and is helpful to determine the location of the welding point. j ,V j ,M j Tracking analysis is performed to obtain the main force paths at each ball node.

[0154] In summary, the purpose of step S40 is to further determine the stress and strain distribution characteristics and internal force transmission paths of each spherical node of the grid based on the static analysis, so as to provide a basis for the subsequent determination of welding points.

[0155] The specific implementation of step S50 is as follows:

[0156] On the basis of step S40, it is necessary to correct and superimpose the stress, strain distribution and internal force transmission path of each ball node obtained by the static analysis according to the dynamic analysis results of step S30.

[0157] The results of the dynamic analysis reflect the actual vibration response characteristics of the grid under dynamic loads, so it is necessary to combine them with the results of the static analysis to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads. This combination can be performed using the linear superposition method, that is, simply adding the stress and strain values ​​obtained from the static analysis and the dynamic analysis:

[0158]

[0159] In some cases, the results of static and dynamic analysis may differ greatly. In this case, a more complex combination method, such as response spectrum method or random vibration theory, is needed to correct the superposition relationship between the two:

[0160]

[0161] Through step S50, the stress, strain distribution and internal force transmission path of each ball node of the grid under the combined action of static and dynamic loads can be obtained, providing more accurate data support for the subsequent determination of welding points.

[0162] The specific implementation of step S60 is as follows:

[0163] According to the stress, strain distribution and internal force transmission path information of each ball node obtained in steps S40 and S50, combined with the design standards and manufacturing process data, it is necessary to determine the possible welding position and weld size of each ball node, thereby forming multiple welding schemes.

[0164] Specifically, we first need to refer to GB 50017 "Steel Structure Design Standard" and other relevant design standards to determine the minimum size of various welds. min and the maximum size t max These size standards are generally related to factors such as node bearing capacity and fatigue performance, and can be expressed as:

[0165] t min ≤t≤t max

[0166] Secondly, according to the manufacturing process data, such as welding process parameters, welding equipment information, etc., combined with the geometric characteristics of the node and the internal force transmission path, determine the possible welding positions of each ball node. The selection of welding position should avoid the stress concentration area and try to ensure the rationality of the weld length and weld shape.

[0167] Finally, researchers can generate multiple different welding schemes based on the above information, laying the foundation for subsequent optimization calculations. These schemes may involve a combination of multiple factors such as weld length, weld shape, and weld position.

[0168] The specific implementation of step S70 is as follows:

[0169] Among the multiple welding schemes determined in step S60, a non-dominated sorting genetic algorithm (NSGA-II) needs to be used for optimization calculation, so as to select the welding scheme with the smallest stress-strain concentration as the pending scheme.

[0170] The non-dominated sorting genetic algorithm is a multi-objective optimization algorithm that can optimize multiple objective functions at the same time and obtain a set of non-dominated solutions. In this problem, the objective functions that need to be optimized include:

[0171] Equivalent stress of each spherical node under the combined action of static and dynamic loads

[0172]

[0173] The principal strains of each spherical node under the combined action of static and dynamic loads

[0174] The rationality of the internal force transmission path of each ball node is expressed by the internal force N j ,V j ,M j express:

[0175]

[0176] Through the iterative optimization of the NSGA-II algorithm, a set of non-dominated solutions under the above three objective functions f1, f2, and f3 can be obtained. The NSGA-II algorithm can effectively find non-dominated solutions close to the true Pareto frontier through two key steps: fast non-dominated sorting and crowding calculation.

[0177] The specific algorithm flow is as follows:

[0178] Initialize the population Contains N individuals, each of which represents a welding scheme;

[0179] Calculate the objective function values ​​f1, f2, f3 of each individual in the population;

[0180] Perform fast non-dominated sorting to obtain non-dominated frontiers of different levels

[0181] According to the crowding index, select individuals with better performance in the population to enter the next generation

[0182] right Perform crossover and mutation operations to generate new individuals;

[0183] Repeat steps 2)-5) until the termination condition is met.

[0184] From the final non-dominated solution set In the experiment, researchers can choose the welding scheme with the smallest stress and strain concentration as the pending scheme. The values ​​of this scheme on the three objective functions are small, and it can better balance the structural safety, deformation coordination and force transmission efficiency.

[0185] The specific implementation of step S80 is as follows:

[0186] According to the pending welding scheme selected in step S70, it is necessary to make corresponding grid test pieces and perform mechanical property tests on the welded joints, so as to adjust and optimize the finite element model.

[0187] Specifically, firstly, according to the pending welding scheme obtained in step S70, a grid test piece is manufactured. During the manufacturing process, the welding quality needs to be strictly controlled to ensure that the manufacturing process of the test piece is consistent with the actual production process.

[0188] Then, the manufactured grid specimens were subjected to static loading test and dynamic loading test. Static loading test can test the bearing capacity of welded joints under static load. The dynamic loading test can evaluate the fatigue performance of the joint under dynamic load. express.

[0189] The test results can be compared and analyzed with the prediction results of the finite element model. If there is a large deviation, it is necessary to use numerical optimization techniques such as response surface method or Lagrange multiplier method to correct and optimize the geometric parameters x and material properties θ of the finite element model so that it can more accurately predict the mechanical behavior of the actual grid:

[0190]

[0191] Through step S80, the finite element model and the mechanical properties of the actual grid can be checked and improved, providing a more reliable data basis for the subsequent determination of welding points.

[0192] The specific implementation of step S90 is as follows:

[0193] After the finite element analysis model is optimized and corrected in step S80, steps S20 to S70 need to be repeated to finally determine the coordinates of the welding points of each ball node of the grid.

[0194] Specifically, first, in the optimized finite element model, the static load condition is reset to perform static analysis (step S20). Then, on this basis, the grid structure is subjected to dynamic analysis to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads (steps S30, S40, S50).

[0195] Next, according to the design standards and manufacturing process information, possible welding schemes for each ball node are determined (step S60), and these schemes are optimized using a non-dominated sorting genetic algorithm (step S70), and the scheme with the smallest stress-strain concentration is selected as the final scheme.

[0196] Finally, based on the final solution, the coordinates of the welding points of each grid ball node are directly output. These welding point coordinate information can provide a direct technical basis for the subsequent grid manufacturing and assembly.

[0197] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are run, they are used to execute the above-mentioned method for determining the welding points of the grid ball nodes.

[0198] A third aspect of the present invention provides a grid ball node welding point determination system, which includes the above-mentioned computer-readable storage medium.

[0199] Specifically, the principle of the present invention is to systematically determine the optimal welding scheme that meets the multiple performance requirements of the grid structure by combining finite element analysis, multi-objective optimization and experimental verification. The specific principle is as follows:

[0200] First, a finite element analysis model is established that can accurately simulate the vibration response of the grid under dynamic action. This includes reasonable division of unit grids to ensure that the model can reflect the geometric characteristics of the actual grid, and to obtain information on dynamic loads that may act on the grid, such as inertial forces and seismic loads caused by building vibration. This step lays a solid foundation for subsequent static and dynamic analysis.

[0201] Secondly, the working conditions of the grid under static loads (such as deadweight, service load, wind load) are set in the finite element model, and static analysis is performed to obtain the stress, strain and displacement distribution of each node. Subsequently, on this basis, the grid structure is subjected to dynamic analysis to simulate its vibration response characteristics under dynamic loads and obtain the dynamic stress and strain data of each node. These two types of analysis results provide the necessary mechanical basis for the subsequent determination of welding points.

[0202] Next, based on the results of static and dynamic analysis, the stress and strain distribution of each spherical node of the grid and the internal force transmission path are analyzed. This step aims to find potential problem areas with large stress and strain concentrations, providing a basis for determining the location of welding points. At the same time, analyzing the internal force transmission path is also conducive to optimizing the shape and length of the weld.

[0203] On this basis, combined with the design standards and manufacturing process information, the possible welding schemes of each ball node are determined, and the non-dominated sorting genetic algorithm (NSGA-II) is used to perform multi-objective optimization on these schemes. During the optimization process, factors such as structural safety (equivalent stress), deformation coordination (principal strain) and internal force transmission efficiency are considered at the same time, thereby obtaining a set of non-dominated solutions with better performance. The scheme with the smallest stress and strain concentration is selected as the target welding scheme.

[0204] Finally, the grid specimens are made, and static and dynamic tests are carried out to verify and optimize the analysis results of the finite element model. If there are large differences, numerical techniques such as the response surface method or the Lagrange multiplier method are needed to correct the geometric and material parameters of the finite element model so that it can more accurately predict the mechanical behavior of the actual grid. Through this step, the reliability of the analysis results can be further improved, laying a solid foundation for the final determination of the coordinates of the grid ball node welding points.

[0205] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0206] In order to better understand the present invention, an embodiment of the present invention in a specific scenario is provided below: the span of the building grid structure is 120 meters, and the supporting roof area is about 8,000 square meters. The grid is mainly composed of Φ406×10mm steel pipes, made of Q345B steel, with a yield strength of f y =345MPa, tensile strength is f u =500MPa, elastic modulus is E=2.06×10 5 MPa, Poisson's ratio is μ=0.3. The total weight of the structure is about 3000kN. The construction first involves the assembly of the grid and then the welding of the grid.

[0207] (I) Assembly of the grid

[0208] 1. Assembly and welding sequence of grid

[0209] The order of assembling the grid is to gradually spread from the center to the surrounding areas. After completing one grid, unified welding is performed and the assembly of the next grid is carried out at the same time.

[0210] The welding sequence is to gradually diffuse the welding from the center to the surrounding areas, which is beneficial to the release of internal stress in the structure. When welding, the lower chord ball is welded first, and then the upper chord ball.

[0211] 2. Overview of grid assembly methods

[0212] According to the ball node in the drawing, the Z coordinate and height difference of the lower chord ball are calculated, and then the support is placed on the ground according to the size of the grid ball to be assembled in the central area. The center cross lines are measured on the support, and the position of the ball node is determined by steel pipe positioning, and the rods between the assembled nodes are connected to form the lower chord unit grid. Then, the upper chord center ball is positioned with three abdominal rods, so that the center of the upper chord ball coincides with the center position of the ground projection, and the other two assembled rods are connected to form a small unit reference control point. When assembling the grid in each area, the assembly sequence of "lower chord-abdominal rod-upper chord" and "center-surrounding" is followed.

[0213] When placing, the specifications and numbers of the steel balls must be known in mind to avoid the inconsistency between the numbers and the entities, which will cause unnecessary idle work or rework. When assembling, first spot weld the lower chord members and the steel balls together to form a square frame, and then use a steel ruler and a level to check the geometric dimensions of the grid. Only after checking that they are correct can the assembly of the lower grid frame continue.

[0214] While assembling the lower chord grid, you can spot weld the "herringbone" on the ground. After installing the lower chord, install the "herringbone", and then install the upper chord rods of the grid. After installing a grid, use a steel ruler to remeasure the geometric dimensions. If there is no error, continue to assemble the next grid in the same way. Until the grid is assembled.

[0215] 3. Grid support structure

[0216] The ground grid of this project is assembled with self-made steel supports. When placing, the theodolite is used to control the support position, and the level is used to control the support height. The highest support elevation of the whole span is 500mm. Before assembly, the steel ruler, tape measure, theodolite, level and other instruments used in assembly must be accurately calibrated. The support details are shown in the figure below. Figure 2 As shown:

[0217] In order to prevent the support position from sinking, the support position should be compacted or hardened. After the support is placed, its elevation is recorded, and its sinking is observed after the grid is assembled. Every day, a special person is assigned to check the stress of each support and adjust the support adjustment bolts of the unstressed support to ensure that each support is under stress.

[0218] 4. Grid assembly process

[0219] Grid assembly process: Finished rods and welding balls enter the site → On-site inspection → Grid overall assembly → Assembly inspection

[0220] Specific assembly method of the grid:

[0221] a. Assembly of lower chord and ball:

[0222] According to the number of the installation drawing, pad the plane of the lower chord ball, connect the lower chord rod and the ball and spot weld them. Form the lower chord quadrilateral unit grid. When assembling the grid frame in each area, assemble it in the order of "lower chord small assembly unit - web rod small assembly unit - upper chord", "middle - both ends (span direction) - both ends (non-span direction)".

[0223] b. Assembly of belly rod and upper string ball:

[0224] The belly rod and the upper chord ball should form a herringbone. The connection between the belly rod and the upper chord ball is spot welded, and the connection between the belly rod and the lower chord ball is also spot welded.

[0225] c. Assembly of the upper chord:

[0226] The installation sequence of the upper chord is from inside to outside. The spot welding of the upper chord and the ball should be carried out in sequence with the spot welding of the web and the lower chord ball. According to the above method, install the grids one by one until all are completed. During the entire grid installation process, special attention should be paid to the padding of the lower chord ball, the accuracy of the axis, the firmness of the spot welding, the deflection and the control of the geometric dimensions. The specific steps are as follows: Figure 3 As shown:

[0227] Step 1: First, calculate the Z coordinates and corresponding positions of each lower chord ball according to the coordinates of the ball nodes in the drawing. Set up supports on the ground according to the size of the grid ball to be assembled in the center area, and connect and assemble the internode members to form a lower chord unit grid. Control the elevation and remeasure the diagonal size to make it completely closed, and spot weld the lower chord member to the ball.

[0228] Step 2: Install the upper string ball and belly rod between the lower string ball.

[0229] Step 3: Install the chord rod and chord balls between the four chord balls.

[0230] Step 4: Install the belly rod between the middle upper and lower strings.

[0231] Step 5: Continue installing the top balls and belly bars in adjacent grids.

[0232] Step 6: Install the upper chord rod and upper chord ball between the four upper chord balls, and install the belly rod between the middle upper chord ball and the lower chord ball.

[0233] Step 7: Continue to install the grids of other parts according to the above method.

[0234] Step 8: Continue to assemble outwards in a similar manner until the entire grid is assembled.

[0235] (II) Determination of welding plan

[0236] Step S10: Finite element model establishment and dynamic load determination

[0237] First, according to the geometric dimensions and material properties of the grid, the corresponding three-dimensional simulation model was established using the finite element software ABAQUS. Considering the complexity of the grid structure, this case uses a 4-node shell element to discretize the steel pipe, and the unit size is controlled at about 100mm, which can better capture the overall deformation characteristics of the grid. The entire model has a total of 2345 units and 1876 nodes.

[0238] On the basis of establishing the finite element model, it is necessary to obtain the dynamic load information that may act on the grid. According to the climate characteristics of the area where the grid is located and the structural use environment, this case mainly considers two types of dynamic loads:

[0239] Inertial force caused by building vibration:

[0240] By conducting dynamic analysis on the main building connected to the grid, its vibration characteristics under loads such as earthquakes and wind vibrations are obtained. Taking earthquakes as an example, according to the response spectrum given by relevant specifications, the acceleration time history curves of each floor of the building are calculated using the time history analysis method. Combined with the coupled finite element model of the grid and the building, the inertial force on each node of the grid can be obtained. The maximum value is about 150kN.

[0241] Seismic Loads:

[0242] According to the earthquake intensity of the area where the grid is located, the dynamic load of each node of the grid under earthquake action was calculated by using the response spectrum method with reference to GB 50011 "Code for Seismic Design of Buildings". The earthquake intensity is set to 8 degrees, the peak seismic acceleration is 0.4g, the earthquake amplification factor Γ=1.2, and the maximum node seismic force is about 300kN.

[0243] Through the above steps, a finite element analysis model was established which can accurately simulate the vibration response of the grid under dynamic action, laying a good foundation for subsequent static and dynamic analysis.

[0244] Step S20: Static analysis

[0245] In the finite element model, the working conditions of the grid under static loads are first set. Specifically, they include:

[0246] Deadweight According to the steel density ρ=7.85×10 3 kg / m 3 and unit volume, the self-weight of each node is calculated, and the maximum value is about 50kN.

[0247] Use Load Taking into account the snow load on the roof and the load on people, the value is 1.5kN / m 2 , the maximum load converted to each node is about 40kN.

[0248] Wind Load According to GB / T 50009 "Building Load Code", the equivalent static wind pressure method is used for calculation, the design wind speed is 30m / s, and the maximum wind load is about 35kN.

[0249] Combining the above three types of static loads, the total static load of each node of the grid is obtained The maximum value is about 110kN.

[0250] Apply the above static loads to the finite element model and solve the static equilibrium equation Ku = F static , the displacement, stress and strain distribution of each node of the grid are obtained. The results show that the maximum displacement occurs in the mid-span, which is about 35mm; the maximum equivalent stress occurs in the area near the support, which is about 260MPa, which is lower than the yield strength f of the steel. y =345MPa.

[0251] Step S30: Kinetic analysis

[0252] On the basis of static analysis, dynamic analysis of the grid structure was carried out, considering two kinds of dynamic effects: inertial force caused by building vibration and seismic load.

[0253] Firstly, the Newmark-β time domain integration method is used to solve the dynamic response of the grid under the inertial force caused by the building vibration. According to the above dynamic analysis results of the main building, the acceleration time history curves of each node of the grid are obtained. And calculate the corresponding inertial force The results show that the maximum dynamic stress occurs in the mid-span area, which is about 160MPa, lower than the static analysis result.

[0254] Secondly, the response spectrum method was used to calculate the dynamic response of the grid under seismic loads. According to GB50011 "Code for Seismic Design of Buildings", the design response spectrum under 8-degree horizontal earthquake was selected, and the response spectrum values ​​of each mode were obtained. The modal analysis results were combined and superimposed to calculate the seismic inertia force of each node of the grid. The results show that the maximum dynamic stress occurs in the area close to the support, which is about 240MPa, slightly lower than the static analysis result.

[0255] Combining the analysis results of the above two types of dynamic effects, the stress and strain distribution of each node of the grid under the combined action of static and dynamic loads was obtained. The results show that dynamic loads will have a certain impact on the stress state of the grid, but the overall stress level is slightly lower than that of static analysis.

[0256] Step S40: Stress-strain analysis

[0257] According to the static and dynamic analysis results of steps S20 and S30, the stress and strain distribution characteristics of each ball node of the grid and the internal force transmission path are analyzed in detail.

[0258] First, extract the principal stress σ of each spherical node i1 ,σ i2 ,σ i3 and equivalent stress σ iev The results show that the maximum equivalent stress appears at the nodes close to the support, which is about 280MPa, lower than the yield strength of steel. However, there are also a few nodes where the stress concentration is more serious, reaching about 300MPa, which requires special attention.

[0259] Secondly, the internal force transmission path of each spherical node is analyzed. j ,V j ,M jBy tracking the force, it is found that the force is mainly transmitted along the diagonal direction of the grid, forming a relatively clear force mechanism. However, there are also cases where the force on some nodes is more complicated and the internal force transmission efficiency is low.

[0260] Based on the above analysis results, it can be preliminarily determined that the grid structure has potential problems such as local stress concentration and poor internal force transmission, which need to be addressed in the subsequent welding optimization.

[0261] Step S60: Welding plan generation

[0262] According to the analysis result of step S40, combined with the relevant requirements for grid node weld design in GB 50017 "Steel Structure Design Standard" and the welding process parameters provided by the manufacturing unit, the possible welding schemes for each ball node are determined.

[0263] Specifically, first refer to the standard and set the minimum weld size t min Take 12mm, the maximum size t max The size range is 20mm, which can meet the bearing capacity and fatigue performance requirements of the grid nodes.

[0264] Secondly, the possible welding positions of each ball node are determined by combining the geometric characteristics of the node and the internal force transmission path. For nodes with severe stress concentration, high stress areas should be avoided as much as possible, and areas with smoother internal force transmission should be selected for welding. For some nodes with more complex forces, the weld length should be appropriately increased to improve the force transmission efficiency.

[0265] Taking all the above factors into consideration, a total of 20 different welding schemes were generated, involving multiple design variables such as weld size, shape and position. These schemes laid the foundation for subsequent optimization calculations.

[0266] Step S70: Multi-objective optimization

[0267] Among the 20 welding schemes determined in step S60, a non-dominated sorting genetic algorithm (NSGA-II) was used to perform multi-objective optimization calculations. The three objective functions of the optimization are:

[0268] The equivalent stress and of each node under static and dynamic loads are:

[0269]

[0270] The sum of the principal strains of each node under static and dynamic loads:

[0271] The sum of the internal forces of each member:

[0272]

[0273] The NSGA-II algorithm searches for the Pareto optimal solution set of 20 welding schemes under the above three objective functions through the steps of fast non-dominated sorting, congestion calculation and selection operation. The results show that among the 30 non-dominated solutions finally obtained, 12 schemes have the sum of equivalent stress less than 2.1×106MPa, the sum of principal strain less than 1.8×10-3, and the sum of internal force less than 2.4×10^6kN·m.

[0274] From these 30 non-dominated solutions, a solution with the minimum equivalent stress and minimum welding efficiency is selected as the target welding solution. The specific parameters of this solution are:

[0275] Welding seam size: t = 16mm

[0276] Weld length: The weld length at each node is between 400-600mm

[0277] Welding position: Choose to weld in an area where the internal force transmission is smoother and the stress concentration is smaller.

[0278] Step S80: Test verification

[0279] According to the target welding scheme determined in step S70, a grid specimen is manufactured for experimental verification. The geometric parameters, material properties, etc. of the specimen are completely consistent with the aforementioned finite element model.

[0280] First, the specimens were subjected to static loading tests. The vertical concentrated force was gradually increased at the mid-span until the grid was plastically deformed. The maximum static bearing capacity measured in the test was 2800kN, which was close to the 2700kN predicted by the finite element analysis, with an error within 5%. At the same time, the strain measurement values ​​of each node also agreed well with the analysis results.

[0281] Secondly, the specimen was subjected to a shaking table earthquake simulation test. According to the above dynamic analysis, the earthquake wave type under the design intensity of 8 degrees was selected to load the specimen horizontally. The test results showed that the maximum dynamic response stress of the grid under earthquake action was 230MPa, which was basically consistent with the finite element analysis result of 245MPa, verifying the accuracy of the dynamic analysis.

[0282] Through the above static and dynamic tests, the correctness of the finite element analysis method was preliminarily confirmed. However, some differences were also found, mainly in the following aspects: 1) The maximum static load of the specimen was slightly higher than the analysis value, which may be due to the welding quality being better than the design assumption; 2) The dynamic response of the specimen under earthquake action was slightly lower than the analysis value, which may be due to the influence of the damping effect.

[0283] Therefore, it is necessary to use numerical optimization techniques such as response surface method or Lagrange multiplier method to check and adjust the geometric parameters and material properties of the finite element model so that it can more accurately predict the mechanical behavior of the grid. After the optimization, the maximum static load prediction value of the finite element model is 2750kN, and the maximum dynamic response stress prediction value is 240MPa, which is in better agreement with the test results.

[0284] Step S90: Final welding plan is determined

[0285] Based on the finite element model optimized and corrected in step S80, the analysis process from steps S20 to S70 is repeated, and finally the coordinates of the welding points of each ball node of the grid are determined.

[0286] Firstly, the static load condition was reset in the optimized finite element model and the static analysis was performed. The results showed that the maximum equivalent stress appeared at several nodes near the support, which was about 260MPa, lower than the yield strength.

[0287] Secondly, on this basis, the dynamic analysis of the grid structure was carried out, considering the two types of dynamic effects: inertial force caused by building vibration and seismic load. The results show that the maximum dynamic stress appears in the mid-span area, which is about 220MPa, slightly lower than the static analysis result.

[0288] Then, based on the stress and strain distribution and internal force transmission path of static and dynamic analysis, combined with the weld size requirements in GB50017 "Steel Structure Design Standard", possible welding schemes for each ball node were determined. A total of 30 schemes were generated, involving weld lengths of 400-600mm and weld thicknesses of 16-20mm.

[0289] Finally, the NSGA-II algorithm was used to perform multi-objective optimization on these 30 welding schemes. The objective functions include: equivalent stress and principal strain, and internal force transmission efficiency. After 50 generations of iterative calculations, a set of non-dominated solutions with better performance was obtained. From them, the scheme with the smallest equivalent stress and the smallest stress was selected as the final welding scheme. The specific parameters of the scheme are as follows:

[0290] 1) Weld size: weld thickness t = 16mm

[0291] 2) Weld length: The weld length at each node is between 450-550mm

[0292] 3) Welding position: Choose an area where the internal force transmission is smoother and the stress concentration is smaller.

[0293] Among them, the grid welding requirements are as follows:

[0294] (1) Before welding, the welder should clean the oil and impurities on the welding part and check whether the positioning weld has defects. If there are defects, they should be cleaned before welding.

[0295] (2) After the first pass of base welding is completed, a careful inspection should be carried out to check for welding defects. Only after confirmation can the second pass of welding be started.

[0296] (3) When welding, when the wall thickness of the rod is greater than 4mm, the welding thickness is 1.2 times the wall thickness. The wall thickness of some rods in this project is relatively large, and the entire weld should be formed in 3 to 4 passes. The first pass is the base welding, which must be welded through, and no defects such as slag inclusions, pores, and cracks are allowed. After careful cleaning and inspection, the second pass of welding is carried out. The procedures for each subsequent pass are the same, and the last pass is the forming welding. The weld after forming is required to have a flat surface, uniform weld waves, no defects such as slag inclusions, pores, weld nodules, undercuts, etc., and meet the requirements of the secondary weld.

Claims

1. A method for determining welding points of grid ball nodes, characterized in that: The following steps are involved: S10. Establish a finite element model based on the overall design drawings of the grid and obtain the dynamic loads to which the grid is subjected as predicted by experts; S20, in the finite element model, setting a static load condition of the grid to perform static analysis; S30, in the finite element model, performing dynamic analysis on the grid structure, simulating the vibration response of the grid under the action of dynamic load, and calculating the dynamic stress and strain of each node; S40, according to the static analysis results, calculate the stress, strain distribution and internal force transmission path of each ball node; S50, according to the dynamic analysis results, correcting and superimposing the stress, strain distribution and internal force transmission path of each spherical node calculated by the static analysis results, to obtain the stress, strain distribution and internal force transmission path of each node under the combined action of static and dynamic loads; S60, obtaining multiple welding schemes formed by construction personnel based on welding positions and weld sizes of each ball node; S70, optimizing and calculating different welding schemes, taking into account the stress, strain distribution and internal force transmission path under the combined action of static and dynamic loads, and selecting the welding scheme with the smallest stress and strain concentration as the pending scheme; S80, obtaining a grid specimen manufactured according to the pending scheme, performing a mechanical property test on the welded joint, and adjusting and optimizing the finite element model according to the experimental results; S90. Based on the optimized finite element model, repeat steps S20-S70, select the welding scheme with the smallest stress-strain concentration as the target scheme, and output the welding point coordinates of each grid ball node.

2. A method for determining a grid ball node welding point according to claim 1, characterized in that: The dynamic loads that the grid is subject to as predicted by the experts include inertial forces caused by building vibrations and earthquakes.

3. A method for determining a grid ball node welding point according to claim 1, characterized in that: The static load conditions of the grid include deadweight, service load and wind load.

4. A method for determining a grid ball node welding point according to claim 1, characterized in that: The methods for dynamic analysis of grid structures are time domain method or frequency domain method.

5. A method for determining a grid ball node welding point according to claim 1, characterized in that: The method for correcting and superimposing the stress, strain distribution and internal force transmission path of each ball node calculated from the static analysis results is specifically a linear superposition method, that is, simply adding the stress and strain values ​​obtained from the static analysis and the dynamic analysis.

6. A method for determining welding points of grid ball nodes according to claim 1, characterized in that: The non-dominated sorting genetic algorithm is used to optimize the calculation of different welding schemes.

7. A method for determining welding points of grid ball nodes according to claim 1, characterized in that: The finite element model is adjusted and optimized using the response surface method or the Lagrange multiplier method.

8. A method for determining a grid ball node welding point according to claim 1, characterized in that: The multiple welding schemes formed by the construction personnel based on the welding positions and weld sizes of each ball node specifically include a welding position diagram and a welding process specification.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the method for determining the welding points of the grid ball nodes according to any one of claims 1 to 8.

10. A grid ball node welding point determination system, characterized in that: A computer-readable storage medium comprising the computer-readable storage medium of claim 9.

Images