A steel structure component welding method and system based on robot welding

Abstract

The application belongs to the technical field of welding, and provides a steel structure component welding method and system based on robot welding, which first constructs a structure model of the steel structure component; then, according to the structure model, a locating and welding track is planned; specifically, weld seam information that needs to be welded is determined from the structure model; according to the structure model and the weld seam information, a motion track is planned in a preset motion control strategy, a locating process strategy and a welding process strategy; finally, according to the planned locating and welding track, the robot is controlled to weld the steel structure component; through the structure model, and according to the structure model and the weld seam information, a motion track is planned in a preset motion control strategy, a locating process strategy and a welding process strategy, the dependence on machine vision technology is reduced, an effective welding track is obtained, and the working stability of the whole welding process is improved.

Description

Technical Field

[0001] This invention belongs to the field of welding technology, and in particular relates to a welding method and system for steel structure components based on robot welding. Background Technology

[0002] The steel structure industry has a large demand for welding and high requirements for flexibility, which has always been a challenge for robotic welding. The flexibility requirements of the steel structure industry are mainly reflected in two aspects: first, the requirement for changeover efficiency. The shape and size of steel components are based on the actual needs of the owner's site, making it a completely non-standard customized industry. This requires changeover speeds close to mass production speeds to meet the demands of the steel structure industry, which is completely impossible to achieve with traditional robotic welding solutions that use tooling positioning and teach programming for large-scale welding. Second, the requirement for workpiece dimensional deviation compensation. Steel components are generally large in size and have relatively low assembly precision. During welding, a weld seam positioning function is necessary to achieve stable welding.

[0003] The inventors discovered that most of the current industry uses machine vision to address the flexibility needs of the steel structure industry, but the high cost and poor stability of vision technology have been hindering the promotion of this solution and preventing the acquisition of effective welding trajectories. Summary of the Invention

[0004] To address the aforementioned problems, this invention proposes a welding method and system for steel structure components based on robot welding. This invention plans motion trajectories using a structural model and based on the structural model and weld information, within preset motion control strategies, positioning process strategies, and welding process strategies. This reduces reliance on machine vision technology, yields effective welding trajectories, and improves the overall stability of the welding process.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0006] In a first aspect, the present invention provides a welding method for steel structure components based on robot welding, comprising:

[0007] Obtain the structural model of the steel structure components;

[0008] Based on the structural model, the positioning and welding trajectories are planned; specifically, the weld information to be welded is determined from the structural model; based on the structural model and the weld information, the motion trajectory is planned within the preset motion control strategy, positioning process strategy, and welding process strategy.

[0009] Based on the planned positioning and welding trajectory, the robot is controlled to weld steel structure components.

[0010] Furthermore, the steel structure components are photographed as a whole to obtain an overall point cloud model of the components.

[0011] Furthermore, various component node types are preset in the offline programming software; the preset node is selected according to the node type of the actual component, and the corresponding dimensions are input, and a component containing the node is generated in the software; the preset node types and input dimensions are continuously selected and combined according to the node type and dimensions of the actual component to obtain the three-dimensional model of the actual component.

[0012] Furthermore, the dimensions of the steel column base, ribs, corbels, and connecting plates in the steel structure components are obtained; a steel column model is established based on the flange width, web width, flange thickness, web thickness, inner radius, and column length; a pre-set column base node is selected, and the column base model on the steel column is obtained based on the length, width, thickness of the column base at one end of the steel column, the number of triangular ribs on the column base, the spacing of the triangular ribs, the length of the triangular ribs, the width of the triangular ribs, the chamfer of the triangular ribs, and the thickness of the triangular ribs; a pre-set rib node is selected, and the ribs are obtained based on the distance of the ribs at one end of the steel column, the height of the ribs, the thickness of the ribs, and the thickness of the ribs. By adjusting the chamfer dimensions, a model of the rib plate on the steel column is obtained. The shape and size of the rib plate are copied, and only the position dimensions are changed to obtain multiple rib plate models on the steel column. A pre-set corbel node is selected, and a corbel model is obtained based on the distance of the corbel to one end of the steel column, the length, width, slope, outer contour plate thickness, and the position and plate thickness of multiple rib plates inside the corbel. A pre-set connecting plate node is selected, and a connecting plate model is obtained based on the distance between the connecting plate and one end of the steel column, the distance between the connecting plate and the edge of the steel column, the width of the connecting plate, and the thickness of the connecting plate. The shape and size of the connecting plate are copied, and only the position dimensions are changed to obtain multiple connecting plate models.

[0013] Furthermore, the positioning process strategy includes selecting one or more component assembly deviation types based on the actual site conditions and the phenomenon of different assembly accuracies in different sites; the assembly deviation types include conventional deviation, longitudinal deviation, and dimensional deviation.

[0014] Furthermore, when using the conventional deviation positioning method, three positioning points are made at one end of the weld in the XYZ directions to calculate the position of the endpoint. Then, the weld is automatically extended to the other end according to the weld length to obtain the actual welding trajectory.

[0015] For a bifacial fillet weld perpendicular to the component, three finding points are established in the XYZ directions at one end of the weld to calculate the position of the endpoint. Then, the weld is automatically extended to the other end according to the weld length to obtain the actual welding trajectory. For a trifacial fillet weld perpendicular to the component, three finding points are established in the XYZ directions at the intersection of the three welds to calculate the intersection position. Then, the weld is extended from the intersection position to the end of the three welds according to the weld length to obtain the actual welding trajectory. For a tetrafacial fillet weld perpendicular to the component, three finding points are established in each of the XYZ directions at two intersection points to calculate the two endpoints of one weld. Then, the weld is extended from the two endpoints to the end of the other four welds according to the weld length to obtain the actual welding trajectory. For a pentahedral fillet weld perpendicular to the component, three finding points are established in each of the XYZ directions at four intersection points to calculate the two endpoints of four welds. Then, the weld is extended from the two endpoints to the end of the other four welds according to the weld length to obtain the actual welding trajectory.

[0016] Furthermore, when using the longitudinal deviation positioning strategy, the number of positioning points is reduced based on the conventional deviation positioning strategy.

[0017] Furthermore, when employing a dimensional deviation positioning strategy, three positioning points are established at one end of the weld in the XYZ directions to calculate the position of the endpoint. Then, extending the weld length to the other end, three positioning points are established in the XYZ directions at the other end to calculate the position of the other endpoint. Connecting the two endpoints yields the actual welding trajectory. For a two-sided fillet weld, three positioning points are established at one end of the weld in the XYZ directions to calculate the position of the endpoint. Then, extending the weld length to the other end, three positioning points are established in the XYZ directions at the other end to calculate the position of the other endpoint. Connecting the two endpoints yields the actual welding trajectory. For a three-sided fillet weld, three positioning points are established at the intersection of the three welds in the XYZ directions to calculate the intersection position. Then, extending the other three welds to the end, three positioning points are established in the XYZ directions at each of the other three end points to calculate the position of the other endpoint. Calculate the positions of the other three endpoints, and connect the intersecting positions to the other three endpoints to obtain the actual welding trajectory. For a four-sided fillet weld, at the two intersection points, make three finding points in each of the XYZ directions to calculate the two endpoints of one weld. Then, extend the other four welds to the end, and at the other four end points, make three finding points in each of the XYZ directions to calculate the positions of the other four endpoints. Connect the two endpoints of the middle weld to the other four endpoints to obtain the actual welding trajectory. For a five-sided fillet weld, at the four intersection points, make three finding points in each of the XYZ directions to calculate the two endpoints of four welds. Then, extend the four welds to the end, and at the other four end points, make three finding points in each of the XYZ directions to calculate the positions of the other four endpoints. Connect the endpoints of the middle four welds to the other four endpoints to obtain the actual welding trajectory.

[0018] Furthermore, the motion control strategy includes the robot's external axis type, robot coordinate system orientation, robot travel range calibration, and robot movement speed; the welding process strategy includes welding current, welding speed, arc swing mode, welding power supply working mode, weld offset mode, and weld offset distance.

[0019] Secondly, the present invention also provides a welding system for steel structure components based on robotic welding, comprising:

[0020] The data acquisition module is configured to acquire the structural model of the steel structure components;

[0021] The planning module is configured to: plan the positioning and welding trajectory based on the structural model; determine the weld information to be welded from the structural model; and plan the motion trajectory based on the structural model and the weld information, within a preset motion control strategy, positioning process strategy, and welding process strategy.

[0022] The control module is configured to control the robot to weld steel structural components according to the planned positioning and welding trajectory.

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0024] 1. This invention first constructs a structural model of the steel structure component; then, based on the structural model, it plans the positioning and welding trajectories; specifically, it determines the weld information to be welded from the structural model; based on the structural model and the weld information, it plans the motion trajectory within a preset motion control strategy, positioning process strategy, and welding process strategy; finally, it controls the robot to weld the steel structure component according to the planned positioning and welding trajectories; by using the structural model, and by planning the motion trajectory within the preset motion control strategy, positioning process strategy, and welding process strategy based on the structural model and the weld information, the reliance on machine vision technology is reduced, an effective welding trajectory is obtained, and the working stability of the entire welding process is improved;

[0025] 2. In this invention, the actual welding trajectory corresponding to different numbers of fillet welds is determined under three conditions: conventional deviation, longitudinal deviation, and dimensional deviation. This results in an effective welding trajectory and improves the working stability of the entire welding process.

[0026] 3. Traditional parametric modeling methods sometimes generate a model by providing all parameters at once, which is inflexible, unsuitable for complex structural models, and results in low model accuracy. Other methods provide only the most basic parameters to generate a simple primitive, which is then combined into a complex model, resulting in a large workload. To address these issues, this invention pre-defines various component node types in offline programming software during model creation. Pre-define nodes are selected according to the actual component's node type, and corresponding dimensions are input. The software then generates a component containing that node. By continuously selecting and combining pre-define node types and input dimensions based on the actual component's node type and dimensions, a 3D model of the actual component is obtained. Compared to providing all parameters at once, which is only applicable to one type of model, this method greatly improves flexibility, allowing for the creation of complex models of various forms while maintaining model accuracy. Compared to providing a simple primitive at a time, this method can provide a relatively complex node at once, reducing workload and providing a reliable model foundation for the intelligent process of robotic welding of steel structure components. Attached Figure Description

[0027] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.

[0028] Figure 1 This is a flowchart of Embodiment 1 of the present invention. Detailed Implementation

[0029] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0030] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0031] Example 1:

[0032] Currently, most industries use machine vision to address the flexibility needs of the steel structure industry. However, the high cost and poor stability of vision technology have been hindering the promotion of this solution and preventing the acquisition of effective welding trajectories.

[0033] To address the aforementioned issues, this embodiment provides a welding method for steel structure components based on robot welding. First, a structural model of the steel structure component is constructed. Then, based on the structural model, positioning and welding trajectories are planned. Specifically, the weld information to be welded is determined from the structural model. Based on the structural model and the weld information, a motion trajectory is planned within a preset motion control strategy, positioning process strategy, and welding process strategy. Finally, the robot is controlled to weld the steel structure component according to the planned positioning and welding trajectories. By using the structural model and planning the motion trajectory within the preset motion control strategy, positioning process strategy, and welding process strategy based on the structural model and the weld information, the reliance on machine vision technology is reduced, an effective welding trajectory is obtained, and the overall stability of the welding process is improved.

[0034] Specifically, this embodiment reduces reliance on machine vision technology and improves the overall stability of the process. Compared to existing pure vision recognition components or vision + 3D model approaches, this embodiment provides three methods for obtaining the construction model, and the specific methods differ from existing technologies. Compared to existing line laser positioning or 3D vision positioning methods, this embodiment uses offline programming to generate touch positioning trajectories to solve the stability problem of vision positioning; specifically including:

[0035] S1. Obtain the structural model of the steel structure components;

[0036] S2. Plan the positioning and welding trajectory based on the structural model;

[0037] S3. Execute actions by sending control codes to the robot or connecting to the robot control API.

[0038] In step S1, optionally, when obtaining the structural model, it can be modeled in other 3D software, exported, and then imported into offline programming software; or, the component can be photographed as a whole to obtain the overall point cloud model of the component; in the latter case, the node size can be manually input according to the preset node type to parameterize and build the component model.

[0039] One approach involves creating a model using other 3D software, exporting the model, and then importing it into offline programming software. Different methods are required depending on the modeling software used. Optional methods include:

[0040] To model using Tekla, a secondary Tekla plugin is needed to convert the internal model data of Tekla into data that can be accepted by offline programming software. This data includes 3D model information and weld location information.

[0041] Alternatively, you can use mechanical engineering software such as SolidWorks, UG, and ProE to model the model and directly export the STP format model to offline programming software. This data only contains 3D model information.

[0042] Among them, taking a picture of the entire component to obtain the overall point cloud model is done by using an "eye-out-of-hand" approach. The camera is fixed on a ground rail and installed directly above the workpiece, facing the workpiece. Line laser scanning or structured light multi-point photography is used to obtain the overall point cloud.

[0043] One method for parametrically creating component models involves pre-setting component node types in offline programming software. The operator simply selects a pre-set node according to the actual component's node type and inputs the corresponding dimensions; the software will then generate a component containing that node. By continuously selecting and combining pre-set node types and inputting dimensions according to the actual component's node type and dimensions, a 3D model of the actual component can be quickly obtained.

[0044] For example, a typical steel column component often includes a column base, ribs, corbels, and connecting plates. You can first input the dimensions of the steel column body, such as the flange width, web width, flange thickness, web thickness, inner corner radius, and column length of an H-beam, thus creating a steel column. Then, select a pre-defined column base node and input the position and shape dimensions of the column base, such as the length, width, and thickness of the column base plate at one end of the steel column, the number of triangular stiffeners on the column base, the spacing of the triangular stiffeners, the length, width, chamfer size, and thickness of the triangular stiffeners, thus creating a column base on the steel column. Next, select a pre-defined rib node and input the position and shape dimensions of the rib, such as the distance of the rib from one end of the steel column, the height of the rib, the thickness of the rib, and the chamfer size of the rib, thus creating a rib on the steel column. Generally, a component will have multiple ribs; you can copy ribs. To quickly add multiple ribs, you can modify only the position dimensions of the plate. Alternatively, you can select a pre-defined corbel node and input its position and shape dimensions, such as the distance between the corbel and one end of the steel column, whether the corbel is present on one or both sides of the steel column, its length and width, whether it has a slope, the outer plate thickness, the number of ribs inside the corbel, and the rib positions and thicknesses. This will give you one or a pair of corbels. Similarly, you can select a pre-defined connection plate node and input its position and shape dimensions, such as the distance between the connection plate and one end of the steel column, the distance between the connection plate and the edge of the steel column, the width, and the thickness. Generally, a component will have multiple connection plates; you can copy the shape dimensions of the connection plate and only modify its position dimensions to quickly add multiple connection plates.

[0045] S2. Planning the positioning and welding trajectory based on the structural model involves first locating the weld seam to be welded from the structural model. If the structural model comes from Tekla or the software's built-in parametric modeling, it will contain weld seam information. If it comes from other 3D software or point cloud models obtained from visual photography, deep learning methods are needed to identify the weld seam, followed by manual verification of the weld seam information. Based on the model information and weld seam information, and according to pre-set motion control strategies, positioning process strategies, and welding process strategies, motion trajectory planning is performed. The most important strategy is the positioning process strategy, which uses a touch-based positioning method. The software automatically provides the positioning point, direction, and number of positioning points based on the node type.

[0046] Motion control strategies mainly include the robot's external axis type, robot coordinate system orientation, robot travel range calibration, and robot movement speed. Among these, the robot's external axis type includes single-axis ground-rail type, single-axis suspended type, single-axis gantry type, two-axis suspended type, two-axis gantry type, and three-axis gantry type.

[0047] Welding process strategies mainly include welding current, welding speed, arc swinging method, welding power supply working mode, weld offset method, and weld offset distance.

[0048] The positioning process strategy mainly includes, based on the actual site conditions and the varying assembly accuracy in different sites, selecting one or more component assembly deviation types, including conventional deviation, longitudinal deviation, and dimensional deviation, in three forms:

[0049] The conventional deviation positioning strategy is designed for general working conditions, assuming that the perpendicularity of the workpiece assembly meets welding requirements and no positioning compensation is needed; it also assumes that the dimensional accuracy of the workpiece itself meets welding requirements and no positioning compensation is needed. Therefore, the following positioning method is adopted:

[0050] For a double fillet weld that is perpendicular to the component itself, there is only one weld. As long as three finding points are made in the XYZ directions at one end of the weld, the position of the endpoint can be calculated. Then, based on the weld length in the model, it can automatically extend to the other end to obtain the actual welding trajectory.

[0051] For a three-sided fillet weld that is perpendicular to the component itself, there are three welds, but all three welds are reduced to one point. As long as three finding points are made in the XYZ directions at the point where the three welds intersect, the position where the three welds intersect can be calculated. Then, according to the length of the weld in the model, the actual welding trajectory is obtained by extending from this endpoint to the end of the three welds.

[0052] For a four-sided fillet weld that is perpendicular to the component itself, there are five welds, two of which are three-sided intersections. Three finding points are made in each of the three directions XYZ at the two intersections. The two endpoints of one of the welds can be calculated. Then, according to the length of the weld in the model, the weld is extended from these two endpoints to the ends of the other four welds to obtain the actual welding trajectory.

[0053] For a pentagonal fillet weld that is perpendicular to the component itself, there are eight welds, four of which are three-sided intersections. By making three finding points in each of the four intersections in the XYZ directions, the two ends of four of the welds can be calculated. Then, according to the length of the weld in the model, the actual welding trajectory is obtained by extending from these two ends to the ends of the other four welds.

[0054] The longitudinal deviation positioning strategy is designed for applications requiring high assembly precision and low material deformation. It assumes that the workpiece assembly perpendicularity meets welding requirements, eliminating the need for positioning compensation; it also assumes that the workpiece's dimensional accuracy meets welding requirements, further eliminating the need for positioning compensation; and it assumes that the workpiece is not bent and the base material has high dimensional consistency. Therefore, it adopts a method that reduces the number of positioning points compared to the conventional deviation positioning strategy. While conventional deviation positioning involves checking all three directions (X, Y, and Z) for each positioning point, longitudinal deviation only checks the X direction, with the Y and Z directions assumed to meet welding requirements.

[0055] The dimensional deviation positioning strategy is designed for situations with poor assembly accuracy and large material deformation. It assumes that problems exist at all positions of the workpiece, requiring positioning compensation. Therefore, the following positioning method is adopted:

[0056] For a double fillet weld, which consists of only one weld, three finding points are made in the XYZ directions at one end of the weld to calculate the position of the endpoint. Then, based on the weld length in the model, the weld is extended to the other end, and three finding points are made in the XYZ directions at the other end to calculate the position of the other endpoint. Connecting the two endpoints gives the actual welding trajectory.

[0057] For a three-sided fillet weld, there are three welds, but all three welds converge at a single point. By making three finding points in the XYZ directions at the point where the three welds intersect, the position of the intersection of the three welds can be calculated. Then, as the other three welds extend to their ends, three finding points are made in the XYZ directions at the other three end points to calculate the positions of the other three endpoints. Connecting the intersection point and the other three endpoints respectively yields the actual welding trajectory.

[0058] For a four-sided fillet weld, there are five welds, two of which are three-sided intersections. At the two intersections, three finding points are made in each of the three directions (XYZ) to calculate the two endpoints of one of the welds. Then, the other four welds are extended to the ends, and three finding points are made in each of the four end points in the XYZ directions to calculate the positions of the other four endpoints. By connecting the two endpoints of the middle weld and the other four endpoints, the actual welding trajectory is obtained.

[0059] For a pentagonal fillet weld, there are eight welds, four of which are three-sided intersections. At each of the four intersections, three finding points are made in the XYZ directions to calculate the two ends of four of the welds. Then, as the four welds extend to the ends, three finding points are made in the XYZ directions at the other four end points to calculate the positions of the other four ends. By connecting the ends of the four middle welds with the other four ends, the actual welding trajectory is obtained.

[0060] S3. Execute actions by sending control codes to the robot or connecting to the robot control API.

[0061] Optionally, the software has built-in multiple robot post-processing logics, which can generate control code instructions adapted to the brand of robot based on the previously planned trajectory. These instructions can be sent to the robot teach pendant via network communication, or the code file can be copied to the robot teach pendant via USB flash drive for execution.

[0062] Some robot brands support direct API control, allowing users to control the robot's movements directly on a host computer using offline programming software, eliminating the need for separate operation on a teach pendant.

[0063] Example 2:

[0064] This embodiment provides a welding system for steel structure components based on robotic welding, including:

[0065] The data acquisition module is configured to acquire the structural model of the steel structure components;

[0066] The planning module is configured to: plan the positioning and welding trajectory based on the structural model; determine the weld information to be welded from the structural model; and plan the motion trajectory based on the structural model and the weld information, within a preset motion control strategy, positioning process strategy, and welding process strategy.

[0067] The control module is configured to control the robot to weld steel structural components according to the planned positioning and welding trajectory.

[0068] The working method of the system is the same as that of the robot-based welding method for steel structure components in Embodiment 1, and will not be repeated here.

[0069] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A welding method for steel structure components based on robot welding, characterized in that, include: Obtain the structural model of the steel structure components; Based on the structural model, plan the positioning and welding trajectories; Specifically, the weld information to be welded is determined from the structural model; based on the structural model and the weld information, the motion trajectory is planned in the preset motion control strategy, positioning process strategy and welding process strategy. Based on the planned positioning and welding trajectory, the robot is controlled to weld steel structure components; The positioning process strategy includes selecting one or more component assembly deviation types based on the actual site conditions and the phenomenon of different assembly accuracies in different sites; the assembly deviation types include conventional deviation, longitudinal deviation, and dimensional deviation. When using the conventional deviation positioning method, three positioning points are made at one end of the weld in the XYZ directions to calculate the position of the endpoint. Then, the weld is automatically extended to the other end according to the weld length to obtain the actual welding trajectory. For a double fillet weld that is perpendicular to the component itself, three finding points are made in the XYZ directions at one end of the weld to calculate the position of the endpoint. Then, the weld is automatically extended to the other end according to the weld length to obtain the actual welding trajectory. For a three-sided fillet weld that is perpendicular to the component itself, three finding points are made in the XYZ directions at the point where the three welds intersect. The position where the three welds intersect is calculated. Then, the actual welding trajectory is obtained by extending from the intersection point to the end of the three welds according to the length of the weld. For a four-sided fillet weld that is perpendicular to the component itself, three finding points are made in each of the three directions XYZ at the two intersection points to calculate the two endpoints of one of the welds. Then, the actual welding trajectory is obtained by extending from the two endpoints to the ends of the other four welds according to the length of the weld. For a pentahedral fillet weld that is perpendicular to the component itself, three search points are made in each of the four intersection points through the XYZ directions. The two ends of four of the welds are calculated. Then, the actual welding trajectory is obtained by extending from the two ends to the ends of the other four welds according to the length of the weld.

2. The welding method for steel structure components based on robot welding as described in claim 1, characterized in that, Take a complete photograph of the steel structure components to obtain an overall point cloud model of the components.

3. The welding method for steel structure components based on robot welding as described in claim 1, characterized in that, Various component node types are preset in the offline programming software; select the preset node according to the node type of the actual component and input the corresponding dimensions to generate a component containing the node in the software; continuously select the preset node types and input dimensions according to the node type and dimensions of the actual component to arrange and combine them to obtain the three-dimensional model of the actual component.

4. The welding method for steel structure components based on robot welding as described in claim 3, characterized in that, Obtain the dimensions of the steel column base, ribs, corbels, and connecting plates in the steel structure components; establish a steel column model based on the flange width, web width, flange thickness, web thickness, inner fillet size, and column length; select a pre-set column base node, and obtain the column base model on the steel column based on the length, width, thickness of the column base at one end of the steel column, the number of triangular stiffeners on the column base, the spacing of the triangular stiffeners, the length of the triangular stiffeners, the width of the triangular stiffeners, the chamfer of the triangular stiffeners, and the thickness of the triangular stiffeners; select a pre-set rib node, and obtain the rib model based on the distance of the ribs at one end of the steel column, the height of the ribs, the thickness of the ribs, and the chamfer of the ribs. To obtain a model of the ribs on the steel column, copy the shape and size of the ribs, changing only the position dimensions, to obtain multiple rib models on the steel column. Select a pre-set corbel node, and based on the distance of the corbel to one end of the steel column, the length, width, slope, outer contour plate thickness, and the position and plate thickness of multiple ribs inside the corbel, obtain a corbel model. Select a pre-set connecting plate node, and based on the distance between the connecting plate and one end of the steel column, the distance between the connecting plate and the edge of the steel column, the width of the connecting plate, and the thickness of the connecting plate, obtain a connecting plate model. Copy the shape and size of the connecting plate, changing only the position dimensions, to obtain multiple connecting plate models.

5. The welding method for steel structure components based on robot welding as described in claim 1, characterized in that, When using the longitudinal deviation positioning strategy, the number of positioning points is reduced based on the conventional deviation positioning strategy.

6. The welding method for steel structure components based on robot welding as described in claim 1, characterized in that, When employing a dimensional deviation positioning strategy, three positioning points are established at one end of the weld in the XYZ directions to calculate the position of the endpoint. Then, extending the weld length to the other end, three positioning points are established in the XYZ directions at that end to calculate the position of the other endpoint. Connecting the two endpoints yields the actual welding trajectory. For a two-sided fillet weld, three positioning points are established at one end of the weld in the XYZ directions to calculate the position of the endpoint. Then, extending the weld length to the other end, three positioning points are established in the XYZ directions at that end to calculate the position of the other endpoint. Connecting the two endpoints yields the actual welding trajectory. For a three-sided fillet weld, three positioning points are established at the intersection of the three welds in the XYZ directions to calculate the intersection position. Then, extending the other three welds to their ends, three positioning points are established in the XYZ directions at each of the other three end points to calculate the position of the other endpoint. The positions of the other three endpoints are connected to the intersection points and the other three endpoints to obtain the actual welding trajectory. For a four-sided fillet weld, three finding points are made in each of the XYZ directions at the two intersection points to calculate the two endpoints of one weld. Then, as the other four welds extend to the end, three finding points are made in each of the XYZ directions at the other four end points to calculate the positions of the other four endpoints. The two endpoints of the middle welds and the other four endpoints are then connected to obtain the actual welding trajectory. For a five-sided fillet weld, three finding points are made in each of the XYZ directions at the four intersection points to calculate the two endpoints of four welds. Then, as the four welds extend to the end, three finding points are made in each of the XYZ directions at the other four end points to calculate the positions of the other four endpoints. The endpoints of the middle four welds and the other four endpoints are then connected to obtain the actual welding trajectory.

7. The welding method for steel structure components based on robot welding as described in claim 1, characterized in that, The motion control strategy includes the robot's external axis type, robot coordinate system orientation, robot travel range calibration, and robot movement speed; the welding process strategy includes welding current, welding speed, arc swing mode, welding power supply working mode, weld offset mode, and weld offset distance.

8. A welding system for steel structure components based on robotic welding, characterized in that, A welding method for steel structure components based on robot welding as described in any one of claims 1-7, comprising: The data acquisition module is configured to acquire the structural model of the steel structure components. The planning module is configured to: plan the positioning and welding trajectory based on the structural model; determine the weld information to be welded from the structural model; and plan the motion trajectory based on the structural model and the weld information, within a preset motion control strategy, positioning process strategy, and welding process strategy. The control module is configured to control the robot to weld steel structural components according to the planned positioning and welding trajectory.

Images