A method for automatically constructing a finite element simulation model of a welded structure

Abstract

The application discloses a kind of automatic construction methods of welding structure finite element simulation model, comprising the following steps: constructing weld three-dimensional model in original assembly model;Original assembly model and weld three-dimensional model are merged into integral model by Boolean operation;Integral model is cut and separated out weld module and base material module, each weld is individually as a component, and base material is as a component as a whole;Tetrahedral mesh is automatically divided, wherein weld module and base material module grid unit are separately divided respectively;Gap is set in the overlapping structure area of base material module to realize node separation;In the position of splicing point, base material grid unit on the two sides of weld shares unit node.The application utilizes the strong adaptability of tetrahedral mesh to complex structure, predefines weld module and base material module, realizes the transformation of welding simulation model from manual creation to automatic creation, and realizes the automatic division of mesh, and weld and base material grid node are automatically matched.

Description

Technical Field

[0001] This invention belongs to the field of welding finite element simulation model construction, and more specifically, relates to an automatic construction method for welding structure finite element simulation model. Background Technology

[0002] The straightening process after welding deformation of large components is time-consuming and labor-intensive, hindering the improvement of production efficiency. Optimizing welding specifications using numerical simulation is an effective means of controlling welding deformation. Constructing a finite element model is a prerequisite and necessary step for simulation calculations. Improving the efficiency of finite element model construction helps shorten the welding simulation cycle.

[0003] When constructing simulation models for welding large structural components, the weld seams are complex and mutually interfering, and the 3D models of welded structures often lack weld seam models, requiring the manual creation of solid or mesh element models of the weld seams. Model construction typically begins with manual meshing of a specific weld seam region, gradually transitioning to other regions. During mesh generation, it is crucial to ensure the quality of the mesh elements near the weld seam. Furthermore, the weld seam and the base material on both sides often need to share element nodes to simulate the interatomic connections between the weld seam and the base material.

[0004] For finite element models of large welded structures, the number of mesh elements is usually in the millions and is not necessarily symmetrically distributed. The mesh elements of the weld model affect each other, and attention must also be paid to the quality of mesh elements near the weld. This inevitably leads to a long time to build the mesh model, which can take several days or even more than ten days. The modeling efficiency is particularly low, which is the main obstacle to shortening the simulation cycle. Summary of the Invention

[0005] The purpose of this invention is to address the above shortcomings by providing an automatic construction method for finite element simulation models of welded structures. This method utilizes the strong adaptability of tetrahedral meshes to complex structures and pre-sets weld seam and base material modules to achieve rapid and automatic construction of welding simulation models.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0007] An automatic construction method for finite element simulation models of welded structures includes the following steps:

[0008] Construct a 3D model of the weld in the original assembly model;

[0009] Boolean operations are used to merge the original assembly model and the weld 3D model into a single model.

[0010] The overall model is cut and separated into weld seam modules and base material modules. Each weld seam is treated as a separate component, and the base material as a whole is treated as a separate component.

[0011] The tetrahedral mesh is automatically generated, with the weld module and the base material module mesh cells being generated separately.

[0012] Gap is set in the overlapping structural area of ​​the parent material module to achieve node separation;

[0013] At the splicing point, the base material mesh elements on both sides of the weld share the same element node.

[0014] Preferably, the method further includes pre-qualifying the impact on the original assembly model, simplifying or deleting components or structural features that have little impact on the simulation results; the components that have little impact on the simulation results include, but are not limited to, bolts, nuts, threaded seats and reinforcing ribs, and the structural features that have little impact on the simulation results include small surfaces or small holes of parts, such as small fillets, small chamfers, small bosses, small grooves, wire holes, etc.

[0015] Preferably, the construction of the three-dimensional weld model includes:

[0016] For groove welds, when re-editing the groove parameters, the groove feature is deleted to achieve weld groove filling, and a weld solid module is generated according to the design requirements.

[0017] For fillet welds, the weld cross-section is simplified to a triangle;

[0018] For butt joints, the weld transitions smoothly on both sides, and the weld reinforcement feature is removed.

[0019] Preferably, the method further includes exporting the overall model in a specific format and then importing it into the finite element mesh generation software Hypermesh. The overall model is recognized as a single entity, and the model length unit is kept consistent with the actual unit by scaling the model.

[0020] Preferably, the method further includes, after the overall model is cut and separated, using the solid cutting function of the software Hypermesh, cutting and separating the overlapping areas in the parent material module and setting them as two independent components. The overlapping areas include two types of overlap: one is the overlapping area between boards, and the other is the overlapping area between boards butt joints or corner joints.

[0021] Preferably, the automatic tetrahedral mesh generation is achieved through the Hypermesh software's automatic tetrahedral solid mesh generation function, gradually building the overall simulation model starting from the weld module. The mesh size is jointly determined by the weld size, structural complexity, and CPU computing power.

[0022] Preferably, the gap setting in the overlapping area of ​​the parent material module is to address the defect of shared unit nodes in the overlapping area of ​​the parent material module. The gap is set by the Detach function to achieve node separation, and the gap distance is 1% to 2% of the grid size.

[0023] Preferably, the splicing point positions include pre-weld spot welding positions and positioning welding positions. At the splicing point positions, the base material mesh units on both sides of the weld are treated with shared nodes to match the change in overall structural stiffness after the pre-weld assembly splicing is completed.

[0024] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0025] The model construction method described in this invention utilizes the strong adaptability of tetrahedral meshes to complex structures, pre-sets weld seam modules and base material modules, and realizes the transformation of welding simulation model creation from manual creation to automatic creation. When the mesh is automatically divided, the weld seam and base material mesh nodes are automatically matched.

[0026] This invention uses Boolean addition to merge the assembly model and the weld model into a whole model. In this way, during automatic mesh generation, the weld and the base materials on both sides are treated as a whole, which enables the automatic mesh sharing of nodes.

[0027] In order to facilitate the application of boundary conditions to weld elements individually, this invention requires that the weld entity be cut and separated from the overall model before meshing, and a separate mesh model of the weld module be created to prevent the appearance of mesh elements that span the interface between the weld and the base material.

[0028] In order to simulate the stiffness in actual welded structures, the tetrahedral mesh elements of this invention require further processing. In the mesh elements at the spot welding and reinforcement welding positions, shared nodes are used to improve the matching degree between the model and reality. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the automatic construction of the finite element simulation model of the welded structure described in this invention.

[0030] Figure 2 This is a triaxial view of the initial assembly model of the front fork described in the embodiment;

[0031] Figure 3 This is a triaxial view of the front fork assembly model after cleaning, as described in the embodiment.

[0032] Figure 4 This is a triaxial view of the U-shaped plate described in the embodiment;

[0033] Figure 5 This is a triaxial view of the ear plate described in the embodiment;

[0034] Figure 6 This is a triaxial view of the inner panel described in the embodiment;

[0035] Figure 7 This is a triaxial view of the outer panel described in the embodiment;

[0036] Figure 8 This is the overall model after the weld seam module and the base material module are combined as described in the embodiment;

[0037] In the diagram: 10. Front fork; 10-1. U-shaped plate; 10-1-1. U-shaped plate bevel; 10-2. Ear plate; 10-2-1. Ear plate bevel; 10-3. Inner plate; 10-4. Outer plate; 10-5. Threaded seat; 10-6. Pipe clamp seat; 10-7. Wiring harness support; 10-8. Fillet weld between ear plate and inner side of U-shaped plate; 10-9. Fillet weld between ear plate and outer side of U-shaped plate; 10-10. Circular fillet weld between ear plate and inner plate; 10-11. Circular fillet weld between ear plate and outer plate. Detailed Implementation

[0038] Preferred embodiments of the present invention will now be described in more detail with reference to the accompanying drawings and specific examples.

[0039] Example 1

[0040] like Figure 2 The model shown is of a certain excavator boom front fork 10, made of S355D steel. Figure 4 U-shaped plate 10-1 shown Figure 5 Ear plate 10-2 shown Figure 6 Inner panel 10-3 shown Figure 7 The fork 10-4 is shown as consisting of an outer panel 10-4, a threaded seat 10-5, a pipe clamp seat 10-6, and a wiring harness support 10-7, all connected by welds. To calculate the welding deformation of the fork 10 using numerical simulation, a finite element simulation model needs to be constructed, the process of which is as follows: Figure 1 As shown, the specific implementation steps are as follows:

[0041] 1) Assembly model simplification

[0042] Import the assembly model of the front fork 10 into the software SpaceClaim, such as Figure 2 As shown, the 3D model was cleaned step by step. First, the threaded seat 10-5, pipe clamp seat 10-6, and wire harness support 10-7, which had a smaller impact on welding deformation, were cleaned. The cleaning results are shown below. Figure 3 As shown in the figure, point K represents the butt joint feature of U-shaped plate 10-1 and outer panel 10-4.

[0043] 2) Weld seam module construction

[0044] The weld module is divided into two parts: bevel welds and fillet welds. For the construction of the bevel weld module, as follows... Figure 4 and Figure 5As shown, the U-shaped plate 10-1 and the ear plate 10-2 contain weld bevels. The U-shaped plate bevel 10-1-1 feature is deleted. However, the weld of the ear plate bevel 10-2-1 is not actually filled; the weld entity needs to be constructed according to the actual bevel filling degree. For the fillet weld module construction, there are two methods: first, create a weld triangular cross-section and stretch the cross-section along the weld direction to form a fillet weld; second, after the overall model is formed, form the fillet weld at the weld location using the chamfering function of the software SpaceClaim. Ignoring the influence of short welds on the overall structural welding deformation, the butt joint feature at point K in the fork 10 is deleted.

[0045] 3) Transform into a holistic model

[0046] Using the "combination" function (Boolean addition) of the software SpaceClaim, the U-shaped plate 10-1, ear plate 10-2, inner plate 10-3, and outer plate 10-4 are gradually transformed into the entire front fork 10, i.e., the part mode. The weld seam is then integrated with the base materials on both sides to facilitate automatic matching of element nodes during subsequent automatic mesh generation, ultimately forming a complete model including the weld seam module. Figure 8 As shown, 10-8 is the inner fillet weld between the ear plate and the U-shaped plate, with a weld leg of 8mm; 10-9 is the outer bevel fillet weld between the ear plate and the U-shaped plate, with a weld leg of 12mm; 10-10 is the circumferential fillet weld between the ear plate and the inner plate, with a weld leg of 12mm; 10-11 is the circumferential fillet weld between the ear plate and the outer plate, with a weld leg of 12mm.

[0047] 4) Model export and import

[0048] Save the overall model file of the front fork 10 as a file with the .x_t extension, and import the file into the Hypermesh software. The overall model will be recognized as a single entity. Check the default length unit of the imported model and keep the model length unit consistent with the actual unit by scaling the model. The overall model of the front fork 10 is axisymmetric. Cut it along the axis and take half of it for subsequent model processing and mesh generation. After the mesh generation is completed, use the reflection function to obtain the other half of the mesh elements to reduce the adverse effect of mesh element asymmetry on the simulation results.

[0049] 5) Weld seam model cutting

[0050] Using the solid cutting function of the software Hypermesh, the weld module and the base material module in the overall model are separated by cutting and separating them using three methods: point, line, and surface. The inner fillet weld 10-8 between the ear plate and the U-shaped plate, the outer bevel fillet weld 10-9 between the ear plate and the U-shaped plate, the circumferential fillet weld 10-10 between the ear plate and the inner plate, and the circumferential fillet weld between the ear plate and the outer plate are each treated as a separate component, while the base material as a whole is treated as a single component. The outer bevel fillet weld 10-9 between the ear plate and the U-shaped plate is composed of both the bevel weld portion and the fillet weld portion.

[0051] 6) Separation of overlapping regions

[0052] Using the solid cutting function of the Hypermesh software, overlapping areas in the parent material module are cut and separated. There are two types of overlap: first, the overlapping area between plates, such as the overlapping area between the inner plate 10-3 and the outer plate 10-4 on the side of the ear plate 10-2; second, the overlapping area between plates that are butt joints or corner joints, such as the beveled blunt edge of the welded joint between the ear plate 10-2 and the U-shaped plate 10-1, which is also an overlapping area. After cutting and separating the components that make up the parent material module, independent components are set for each component. This makes it easier to determine the mesh element of the node to be separated based on the geometric solid boundary when separating nodes later.

[0053] 7) Automatic mesh generation

[0054] Using the Hypermesh software's automatic tetrahedral mesh generation function, the overall simulation model was gradually built starting from the weld module. Since the minimum weld leg is 8mm, the mesh size was set to 6mm. After mirror symmetry processing, the overall model of the front fork 10 had 395,964 element meshes. In this model, the overlapping area of ​​the inner plate 10-3 and outer plate 10-4 on the side of the ear plate 10-2 shares nodes, and the bevel edge of the weld joint between the ear plate 10-2 and the U-shaped plate 10-1 also shares nodes. This does not match the actual situation and node separation is required.

[0055] 8) Separation of nodes in overlapping regions

[0056] The overlapping areas of the inner plate 10-3 and the outer plate 10-4 on the side of the ear plate 10-2, and the blunt edge of the weld joint bevel of the ear plate 10-2 and the U-shaped plate 10-1, are all in a shared node state. The node needs to be separated by setting the gap through the Detach function, and the gap distance is set to 0.1mm.

[0057] 9) The positions of the splicing points are the same.

[0058] At the pre-weld spot welding and tack welding locations, the base material mesh elements on both sides of the weld are treated with shared nodes, so that all base material components of the model are connected through shared nodes to match the overall structural stiffness of the model when assembling and splicing before welding.

[0059] The purpose of model construction in this embodiment is to simulate welding deformation. In some other embodiments, the model construction method described in this invention may also be applied to the simulation calculation of welding thermal process and welding stress and strain.

[0060] In some other embodiments, the order of “weld module construction” and “converting to an overall model” in the model construction steps can be interchanged, and the order of “weld model cutting” and “overlapping area separation” can be interchanged.

[0061] In some other embodiments, the steps "assembly model simplification", "weld module construction" and "converting to an overall model" can be implemented not only in the software SpaceClaim, but also in the software Pro / E and SolidWorks.

[0062] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and principles of the described embodiments, and these modifications and variations should also be considered within the scope of protection of the present invention.

Claims

1. A method for automatically constructing a finite element simulation model of a welded structure, characterized in that, Includes the following steps: Construct a 3D model of the weld in the original assembly model; The construction of the three-dimensional model of the weld includes: For groove welds, when re-editing the groove parameters, the groove feature is deleted to achieve weld groove filling, and a weld solid module is generated according to the design requirements. For fillet welds, the weld cross-section is simplified to a triangle; For butt joints, the weld transitions smoothly on both sides, and the weld reinforcement feature is removed. Boolean operations are used to merge the original assembly model and the weld 3D model into a single model. The overall model is cut and separated into weld seam modules and base material modules. Each weld seam is treated as a separate component, and the base material as a whole is treated as a separate component. The tetrahedral mesh is automatically generated, with the weld module and the base material module mesh cells being generated separately. Gap is set in the overlapping structural area of ​​the parent material module to achieve node separation; At the splice point, the base material mesh elements on both sides of the weld share the same element node; The method also includes saving the overall model as a file with the .x_t extension, and then importing it into finite element mesh generation software. The overall model is recognized as a single entity, and the model length unit is kept consistent with the actual unit by scaling the model. After the overall model is cut and separated, the overlapping areas in the parent material module are cut and separated and set as two independent components. The overlapping areas include two types of overlap: one is the overlapping area between boards, and the other is the overlapping area between boards butt or corner joints. The splicing points include the pre-welding spot welding points and the positioning welding points.

2. The automatic construction method for finite element simulation model of welded structures according to claim 1, characterized in that, The method also includes pre-qualitatively assessing the impact on the original assembly model, simplifying or deleting components or structural features that have a minor impact on the simulation results; the components that have a minor impact on the simulation results include, but are not limited to, bolts, nuts, threaded seats and reinforcing ribs, and the structural features that have a minor impact on the simulation results include small faces or small holes of parts.

3. The automatic construction method for finite element simulation model of welded structures according to claim 1, characterized in that, The automatic generation of the tetrahedral mesh is a step-by-step process that starts with the weld module to build the overall simulation model. The mesh size is determined by the weld size, structural complexity, and CPU computing power.

4. The method for automatically constructing a finite element simulation model of a welded structure according to claim 1, characterized in that, The gap distance between overlapping areas in the parent material module is 1% to 2% of the grid size.

Images