Welding fixture parameterized design method and device, electronic equipment and storage medium
By constructing a heterogeneous multimodal graph and introducing a semantic reasoning engine and a physical calculation engine, a parametric design scheme for welding fixtures is generated, which solves the problem of low design efficiency in the existing technology and realizes efficient and accurate fixture design.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- FFT PRODION SYST SHANGHAI
- Filing Date
- 2026-04-03
- Publication Date
- 2026-08-04
AI Technical Summary
Existing welding fixture designs rely on engineers' experience, resulting in low design efficiency.
By constructing a heterogeneous multimodal graph and combining a semantic reasoning engine and a physical computing engine, a target parameterized fixture layout scheme is generated, taking into account the coupling relationship between geometry, process and environment, thereby improving design accuracy and efficiency.
It significantly improves the efficiency and accuracy of welding fixture design, suppresses design results that do not conform to the laws of engineering physics, and enhances the engineering feasibility of automatically generated solutions.
Smart Images
Figure CN121980711B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computer technology, and more specifically, to a parametric design method, apparatus, electronic device, and storage medium for welding fixtures. Background Technology
[0002] In the automotive and equipment manufacturing industries, welding fixtures are key process equipment connecting product design and actual production. They are used to reliably position, stably clamp, and suppress welding deformation during the welding process, thereby ensuring the geometric accuracy and structural quality of the welded car body or components. Currently, the design of welding fixtures mainly relies on experienced design engineers who manually conceive, model, and assemble the fixture structure using computer-aided design software, and gradually complete the design through repeated verification and modification. This design mode is highly dependent on engineering experience and individual ability, resulting in low design efficiency. Summary of the Invention
[0003] The problem addressed by this invention is how to improve the design efficiency of welding fixtures.
[0004] To address the above problems, this invention provides a parametric design method, apparatus, electronic device, and storage medium for welding fixtures.
[0005] In a first aspect, the present invention provides a parametric design method for welding fixtures, comprising: Heterogeneous multimodal graphs are constructed based on multimodal information, wherein the multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data; The degree-of-freedom constraint matrix and welding variation energy constraint field are constructed using the heterogeneous multimodal graph. The heterogeneous multimodal graph, the degree-of-freedom constraint matrix, and the welding variation energy constraint field are encoded as different semantic levels to obtain the constraint condition context representation; By utilizing the semantic reasoning engine and the physical calculation engine through the constraint context representation, the target parameterized fixture layout scheme is obtained.
[0006] Optionally, the step of constructing a heterogeneous multimodal map based on multimodal information includes: The welding process planning data and the production line environment data are analyzed respectively to obtain the process node set and the environment node set; By performing topological structure analysis on the three-dimensional geometric model data, a set of geometric nodes is obtained; The heterogeneous multimodal map is obtained by using the set of geometric nodes, the set of process nodes, and the set of environmental nodes.
[0007] Optionally, the heterogeneous multimodal map includes: , Wherein, G represents the heterogeneous multimodal spectrum. For the set of geometric nodes, The set of process nodes, Let E be the set of environmental nodes, and E be the topological adjacency relationship between the nodes.
[0008] Optionally, the step of constructing the degree-of-freedom constraint matrix and the welding variation form energy constraint field through the heterogeneous multimodal spectrum includes: The degree-of-freedom constraint matrix is constructed using the localization feature information of the heterogeneous multimodal map; Based on the inherent strain prediction algorithm, the inherent strain tensor is obtained from the heterogeneous multimodal spectrum; The inherent strain tensor includes: , in, Let the intrinsic strain tensor be... For plastic strain, For thermal strain, For other strains; The welding deformation energy constraint field is constructed using the inherent strain tensor.
[0009] Optionally, constructing the welding deformation energy constraint field using the inherent strain tensor includes: The welding deformation energy density is obtained through the inherent strain tensor. The welding deformation energy density includes: , in, The weld deformation energy density, C is the transpose of the inherent strain tensor, and C is the energy density matrix of the welding deformation. The welding deformation energy constraint field is constructed using the welding deformation energy density; The welding deformation energy confinement field includes: , Wherein, U is the welding deformation energy constraint field. For spatial domain.
[0010] Optionally, constructing the degree-of-freedom constraint matrix using the localization feature information of the heterogeneous multimodal map includes: Multiple constraint screws are obtained based on the positioning feature information of the heterogeneous multimodal map, wherein the constraint screws are applied to the rigid body of the workpiece, and the positioning feature information is used to represent the positioning element and clamping element information; The constraint screw includes: , Where S is the constraint spiral. ω is the angular velocity vector, and V is the linear velocity vector; The degree-of-freedom constraint matrix is obtained by using all the aforementioned constraint spirals; The degree-of-freedom constraint matrix includes: , Where J is the degree-of-freedom constraint matrix, There are m constraint spirals.
[0011] Optionally, the step of using a semantic reasoning engine and a physical computing engine to obtain a target parameterized fixture layout scheme through the constraint context representation includes: Based on the multimodal large model, the semantic reasoning engine is used to perform hierarchical reasoning through the contextual representation of the constraint conditions to obtain the fixture layout scheme; Based on the obstacle avoidance potential energy field and the structural stiffness objective function, the physical calculation engine is used to adaptively adjust the fixture layout scheme to obtain the target parameterized fixture layout scheme.
[0012] Secondly, the present invention provides a parametric design device for welding fixtures, comprising: a heterogeneous multimodal graph construction module, used to construct a heterogeneous multimodal graph based on multimodal information, wherein the multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data; The constraint acquisition module is used to construct a degree-of-freedom constraint matrix and a welding variation form energy constraint field through the heterogeneous multimodal graph. The constraint context representation acquisition module is used to encode the heterogeneous multimodal graph, the degree-of-freedom constraint matrix and the welding deformation energy constraint field as different semantic levels to obtain the constraint context representation. The target parameterized fixture layout scheme acquisition module is used to obtain the target parameterized fixture layout scheme by utilizing the semantic reasoning engine and the physical calculation engine through the constraint context representation.
[0013] Thirdly, the present invention provides an electronic device, including a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the parametric design method for welding fixtures as described in the first aspect when executing the computer program.
[0014] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the parametric design method for welding fixtures as described in the first aspect.
[0015] The beneficial effects of the parametric design method, apparatus, electronic device, and storage medium for welding fixtures of the present invention are as follows: A heterogeneous multimodal graph is constructed based on multimodal information. This graph comprehensively considers the coupling relationship between geometry, process, and environment, significantly improving the consistency between fixture layout and actual manufacturing conditions. By constructing a degree-of-freedom constraint matrix and a welding variable energy constraint field through the heterogeneous multimodal graph, physical effects are transformed into calculable constraint fields, thereby improving fixture design accuracy. The heterogeneous multimodal graph, degree-of-freedom constraint matrix, and welding variable energy constraint field are encoded as different semantic levels to obtain constraint condition context representations. These constraint condition context representations guide the semantic reasoning engine and physical calculation engine in reasoning, resulting in a target parametric fixture layout scheme. This effectively suppresses design results that may not conform to engineering physical laws during the artificial intelligence generation process, improves the engineering feasibility of automatically generated schemes, and significantly enhances the efficiency of welding fixture design. Attached Figure Description
[0016] Figure 1 This is a flowchart illustrating a parametric design method for welding fixtures according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a parametric design device for welding fixtures according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation
[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.
[0018] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.
[0019] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.
[0020] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".
[0021] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.
[0022] like Figure 1 As shown in the figure, an embodiment of the present invention provides a parametric design method for welding fixtures, comprising: Step 110: Construct a heterogeneous multimodal map based on multimodal information, wherein the multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data.
[0023] Specifically, the multimodal information includes three-dimensional geometric model data of the vehicle body or components to be welded, welding process planning data, and production line environmental information. This multimodal information undergoes unified analysis and standardization to construct a heterogeneous multimodal graph containing geometric topology information, process semantic information, and operating condition constraint information. This heterogeneous multimodal graph, by establishing different types of nodes and their relationships, achieves a unified expression of the geometric features of parts, weld point distribution, assembly relationships, and spatial constraints.
[0024] In some more specific embodiments, boundary representation parsing is performed on the 3D geometric model data to obtain surfaces, boundaries, and topological adjacency relationships. The heterogeneous multimodal graph includes a set of geometric nodes, a set of process nodes, and a set of environment nodes, and the relationships between different types of nodes are encoded using a graph neural network. The geometric node set is used to represent surface patches, hole features, and their normal information; the process node set is used to represent weld point locations, welding sequences, and welding parameters; and the environment node set is used to represent the reachable space of the welding torch, equipment interference areas, and station boundary conditions. A graph attention network is used to weighted model the dependencies between the geometric node set and the process node set to explicitly express the influence of the weld point location on surrounding geometric features and welding deformation sensitivity.
[0025] Step 120: Construct a degree-of-freedom constraint matrix and a welding variation energy constraint field using the heterogeneous multimodal graph.
[0026] Specifically, kinematic modeling of the fixture positioning and clamping scheme is performed based on helix theory. Each positioning and clamping element is equivalent to a constraint helix applied to the rigid body of the workpiece, and a degree-of-freedom constraint matrix is constructed from all constraint helices to determine the completeness of the degree-of-freedom constraints on the workpiece under welding conditions. The rank of the degree-of-freedom constraint matrix is used to determine whether the workpiece achieves complete positioning. When the rank of the degree-of-freedom constraint matrix is less than 6, it is determined to be in an under-constrained state. When it is determined to be in an under-constrained state, the system further calculates the motion helix corresponding to the under-constrained direction and maps the motion helix to structured constraint information or natural language prompts. The structured constraint information or natural language prompts are fed back to the multimodal large model to trigger supplementary reasoning of the fixture positioning or clamping elements. A coupled model of the temperature field, stress-strain field, and material microstructure evolution field during the welding process is established in conjunction with welding process parameters, and a welding deformation form energy constraint field is constructed based on the coupled model to characterize the sensitivity of different spatial regions to welding deformation. The welding variable potential energy constraint field is established based on the inherent strain method or the equivalent thermal strain method, and the welding heat input is converted into an equivalent variable potential energy distribution through spatial integration. The high potential energy region is given a higher clamping arrangement weight to guide the clamping unit to be preferentially arranged near the high potential energy region.
[0027] Step 130: The heterogeneous multimodal graph, the degree-of-freedom constraint matrix, and the welding variation energy constraint field are encoded as different semantic levels to obtain the constraint condition context representation.
[0028] In some more specific embodiments, the heterogeneous multimodal graph, the degree-of-freedom constraint matrix, and the welding deformation energy constraint field are encoded as context representations at different semantic levels. The structural semantic context, encoded from the heterogeneous multimodal graph, describes "what parts, welds, and spatial obstacles are present." The kinematic constraint context, encoded from the degree-of-freedom constraint matrix, describes "which degrees of freedom are constrained." The physical field constraint context, encoded from the welding deformation energy constraint field, describes "which regions are most sensitive to deformation."
[0029] Step 140: Using the semantic reasoning engine and the physical calculation engine, the target parameterized fixture layout scheme is obtained through the constraint context representation.
[0030] Specifically, a multimodal large model is used for collaborative reasoning of fixture design schemes. This multimodal large model uses heterogeneous multimodal maps and explicit physical constraint fields as input context. Through hierarchical reasoning, it sequentially completes the functional layout planning of the fixture, the selection of key structural forms, and the preliminary determination of main design parameters. The reasoning process is jointly constrained by positioning freedom constraints, welding deformation energy distribution, and spatial accessibility conditions to avoid generating design schemes that do not conform to engineering physics laws. For the initial functional layout and structural scheme of the fixture, a dynamic obstacle avoidance potential energy function containing the welding torch motion envelope and a structural objective function reflecting the overall stiffness requirements of the fixture are constructed. Under the joint drive of the aforementioned potential energy fields, the fixture structure is adaptively generated and optimized, ensuring that the generated structure meets stiffness and stability requirements while effectively avoiding welding equipment and the surrounding environment, reducing the risk of structural interference. The obtained structural results are parametrically reconstructed, transforming the generated discrete structure into a parametric 3D model with a complete feature history tree. Based on a standard parts library, functional components in the generated structure are instantiated and matched or parametrically modeled. Finally, 2D engineering drawings and a bill of materials conforming to engineering drawing specifications are automatically output.
[0031] In some more specific embodiments, the multimodal large model performs hierarchical collaborative reasoning under explicit physical constraints to generate the functional layout scheme, structural configuration, and initial values of key parameters of the fixture. The hierarchical collaborative reasoning of the multimodal large model includes at least positioning reference layout reasoning, clamping scheme reasoning, and structural avoidance and configuration reasoning. The multimodal large model uses the welding torch trajectory and its safety envelope as spatial constraints. The dynamic obstacle avoidance potential energy field is constructed by calculating the directed distance field between the fixture structure and the welding torch trajectory; when the distance is less than a preset safety threshold, the obstacle avoidance potential energy value at the corresponding position increases significantly. Generative optimization employs generative adversarial networks, variational autoencoders, or neural distance field models, and incorporates structural flexibility, interference penalty, and connectivity continuity as components of the joint loss function. Parametric reconstruction includes skeleton extraction of the generated structure and fitting skeleton segments to standard or non-standard structural features with parameter-driven cross-sections. Parametric reconstruction further includes shape similarity matching between the generated structure and an enterprise standard parts library; when the similarity meets a preset threshold, the corresponding standard parts model is used for instantiation and replacement.
[0032] In this embodiment, a heterogeneous multimodal graph is constructed based on multimodal information. This graph comprehensively considers the coupling relationship between geometry, process, and environment, significantly improving the consistency between fixture layout and actual manufacturing conditions. A degree-of-freedom constraint matrix and a welding variable energy constraint field are constructed using the heterogeneous multimodal graph, transforming physical effects into computable constraint fields, thereby improving fixture design accuracy. The heterogeneous multimodal graph, degree-of-freedom constraint matrix, and welding variable energy constraint field are encoded as different semantic levels to obtain constraint context representations. These constraint context representations guide the semantic reasoning engine and the physical calculation engine to derive the target parameterized fixture layout scheme. This effectively suppresses design results that may not conform to engineering physics laws during the artificial intelligence generation process, improves the engineering feasibility of automatically generated schemes, and significantly enhances the efficiency of welding fixture design.
[0033] Optionally, the step of constructing a heterogeneous multimodal map based on multimodal information includes: The welding process planning data and the production line environment data are analyzed respectively to obtain the process node set and the environment node set; By performing topological structure analysis on the three-dimensional geometric model data, a set of geometric nodes is obtained; The heterogeneous multimodal map is obtained by using the set of geometric nodes, the set of process nodes, and the set of environmental nodes.
[0034] Specifically, the system receives 3D digital model data of the vehicle body or components and parses its topology using boundary representation. Specifically, the geometric model of the part is represented as a geometric diagram. ,include: , in, Let be the set of geometric nodes, representing the geometric feature surface, and the set of edges. This refers to the topological adjacency relationship between geometric features.
[0035] Each geometric node It has the following eigenvectors: , in, Let be the surface area. The main curvature feature. It is the normal vector. This refers to the distance from the feature to the assembly datum. The welding process document is parsed into the set of process nodes. : , Each solder joint node It must contain at least the following attributes: , in, Let these be the spatial coordinates of the solder joint. For plate thickness combination, For welding current, For welding pressure, This serves as an index for the welding sequence. Associative edges are established between weld point nodes and their respective geometric surface nodes to represent the coupling relationship between the weld heat-affected zone and the geometric features.
[0036] In this optional embodiment, welding process planning data, production line environment data, and three-dimensional geometric model data are respectively parsed into process nodes, environment nodes, and geometric nodes, and organically integrated into a unified heterogeneous graph. This enables joint modeling of multi-dimensional semantics in the manufacturing scenario and significantly enhances the system's ability to understand complex production contexts.
[0037] Optionally, the heterogeneous multimodal map includes: , Wherein, G represents the heterogeneous multimodal spectrum. For the set of geometric nodes, The set of process nodes, Let E be the set of environmental nodes, and E be the topological adjacency relationship between the nodes.
[0038] Specifically, the system uses a graph attention mechanism to aggregate information from different types of nodes, and its message passing process can be represented as follows: , in, Let be the hidden state (i.e., the updated representation vector) of node i at layer l+1. For activation function, Let be the attention weight of node j to node i, used to measure the importance of solder joints or environmental constraints to geometric nodes. Let be the learnable weight matrix of the l-th layer. The hidden state (input representation) of node j at layer l is used to measure the importance of solder joints or environmental constraints to geometric nodes.
[0039] In this optional embodiment, by constructing the aforementioned heterogeneous node set and defining cross-modal edge relationships such as "geometry-weld point" and "weld point-welding gun trajectory," a unified heterogeneous multimodal graph is formed. This graph is used to explicitly express the geometric topology, process semantics, and spatial constraint information involved in fixture design. Based on the heterogeneous multimodal graph information, subsequent degree-of-freedom constraint matrices and welding variation form energy constraint fields can be constructed.
[0040] Optionally, the step of constructing the degree-of-freedom constraint matrix and the welding variation form energy constraint field through the heterogeneous multimodal spectrum includes: The degree-of-freedom constraint matrix is constructed using the localization feature information of the heterogeneous multimodal map; Based on the inherent strain prediction algorithm, the inherent strain tensor is obtained from the heterogeneous multimodal spectrum; The inherent strain tensor includes: , in, Let the intrinsic strain tensor be... For plastic strain, For thermal strain, For other strains; The welding deformation energy constraint field is constructed using the inherent strain tensor.
[0041] Specifically, after the freedom constraint matrix is constructed and the clamping state is confirmed, the system further constructs the welding deformation energy constraint field. The system employs a fast prediction algorithm based on inherent strain. The complex welding transient process is transformed into an equivalent thermal load, and the initial deformation distribution of the workpiece in its free state is calculated. The welding heat input Q is used to calculate the temperature field distribution. Based on the inherent strain method, the plastic strain and phase transformation strain caused by the welding process are uniformly expressed as the inherent strain tensor.
[0042] Optionally, constructing the welding deformation energy constraint field using the inherent strain tensor includes: The welding deformation energy density is obtained through the inherent strain tensor. The welding deformation energy density includes: , in, The weld deformation energy density, C is the transpose of the inherent strain tensor, and C is the energy density matrix of the welding deformation. The welding deformation energy constraint field is constructed using the welding deformation energy density; The welding deformation energy confinement field includes: , Wherein, U is the welding deformation energy constraint field. For spatial domain.
[0043] Specifically, the welding deformation potential energy density is defined based on the stress-strain relationship. Integrating over the entire workpiece volume, the welding deformation potential energy constraint field is obtained. This potential energy field serves as an explicit constraint for subsequent clamping arrangement and structure generation, guiding the system to arrange auxiliary clamping units in high-potential-energy regions.
[0044] In this optional embodiment, the temperature field, stress-strain field, and material microstructure evolution field during the welding process are incorporated into a unified multiphysics constraint framework. The welding deformation control requirements are considered during the fixture structure generation stage, thus achieving the coordinated implementation of structural design and physical verification.
[0045] Optionally, constructing the degree-of-freedom constraint matrix using the localization feature information of the heterogeneous multimodal map includes: Multiple constraint screws are obtained based on the positioning feature information of the heterogeneous multimodal map, wherein the constraint screws are applied to the rigid body of the workpiece, and the positioning feature information is used to represent the positioning element and clamping element information; The constraint screw includes: , Where S is the constraint spiral. ω is the angular velocity vector, and V is the linear velocity vector; The degree-of-freedom constraint matrix is obtained by using all the aforementioned constraint spirals; The degree-of-freedom constraint matrix includes: , Where J is the degree-of-freedom constraint matrix, There are m constraint spirals.
[0046] Specifically, the workpiece is considered a rigid body, and its instantaneous motion can be represented by a constraint screw. Each constraint applied to the workpiece by a positioning or clamping element can be represented as a constraint screw. It satisfies the reciprocal condition with the permissible motion spiral, including: , in, For constrained screws transpose, To allow the movement of the helix, the direction of free movement that the workpiece can still achieve under this constraint is represented. Subsequently, the positioning freedom constraint matrix is constructed and determined. Assuming the system currently has m positioning / clamping elements, the freedom constraint matrix is constructed by calculating the matrix rank, including: , Specifically, when r is less than 6, the system is considered to be in an under-positioned state; when r equals 6, the workpiece is fully constrained. The system further solves for the unconstrained motion direction through singular value decomposition and feeds this degree of freedom direction back to the semantic reasoning engine as explicit constraint information to guide the correction of the positioning scheme.
[0047] Optionally, the step of using a semantic reasoning engine and a physical computing engine to obtain a target parameterized fixture layout scheme through the constraint context representation includes: Based on the multimodal large model, the semantic reasoning engine is used to perform hierarchical reasoning through the contextual representation of the constraint conditions to obtain the fixture layout scheme; Based on the obstacle avoidance potential energy field and the structural stiffness objective function, the physical calculation engine is used to adaptively adjust the fixture layout scheme to obtain the target parameterized fixture layout scheme.
[0048] Specifically, the semantic reasoning engine performs hierarchical reasoning based on a multimodal large model to generate fixture layout schemes; the physics calculation engine applies obstacle avoidance potential energy and stiffness objective function constraints to the generated structure and adaptively adjusts the structure. Based on the kinematic constraint context, the model determines whether the current constraint matrix is full rank and, in the case of underconstraint, infers the types and locations of newly added positioning points. Combining the welding deformation potential energy constraint field, it prioritizes the inference and placement of auxiliary clamping or support units in high potential energy regions to suppress welding deformation. Based on the welding torch trajectory and equipment envelope information in the heterogeneous multimodal graph, it infers the spatial orientation of the fixture columns and support structures to avoid interfering with the reachable path of the welding torch.
[0049] In some more specific embodiments, the structure undergoes skeleton extraction and primitive fitting, resulting in the instantiation and engineering output of a standard parts library. For the generated irregular structure, skeletonization is first performed, extracting the central path line. Through geometric fitting, the path line is transformed into standard CAD features with cross-sectional parameters (such as channel steel specifications and pipe wall thickness) to fit the design intent. For the processed structure, the system employs 3D shape fingerprint matching technology to calculate the similarity between the generated unit and the enterprise's standard parts library (positioning pins, L-shaped adapter blocks, etc.). When the similarity is high, the system performs automated instance replacement, preserving the parametric offset attributes of mounting holes and adjusting shims. The final generated model contains a complete hierarchical tree structure (such as "positioning module - base module - fastener group"), and automatically annotates dimensions based on Model Definition (MBD) technology, directly generating 2D engineering drawings and a Bill of Materials (BOM) that conform to enterprise specifications.
[0050] like Figure 2 As shown in the figure, an embodiment of the present invention provides a parametric design device for welding fixtures, comprising: The heterogeneous multimodal graph construction module 10 is used to construct a heterogeneous multimodal graph based on multimodal information, wherein the multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data; The constraint acquisition module 20 is used to construct a degree-of-freedom constraint matrix and a welding deformation energy constraint field through the heterogeneous multimodal graph. The constraint context representation acquisition module 30 is used to encode the heterogeneous multimodal graph, the degree of freedom constraint matrix and the welding deformation energy constraint field as different semantic levels to obtain the constraint context representation. The target parameterized fixture layout scheme acquisition module 40 is used to obtain the target parameterized fixture layout scheme by utilizing the semantic reasoning engine and the physical calculation engine through the constraint context representation.
[0051] The welding fixture parametric design device of this embodiment is used to implement the welding fixture parametric design method as described above. Its advantages over the prior art are the same as the advantages of the welding fixture parametric design method over the prior art, and will not be repeated here.
[0052] like Figure 3 As shown, an electronic device 300 provided in this embodiment of the invention includes a memory 310 and a processor 320; the memory 310 is used to store a computer program; the processor 320 is used to implement the parametric design method for welding fixtures as described above when the computer program is executed.
[0053] Alternatively, an electronic device 300 includes a memory 310 and a processor 320 coupled to the memory 310; the memory 310 is configured to store a computer program; and the processor 320 is configured to perform the following operations when the computer program is executed: Heterogeneous multimodal graphs are constructed based on multimodal information, wherein the multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data; The degree-of-freedom constraint matrix and welding variation energy constraint field are constructed using the heterogeneous multimodal graph. The heterogeneous multimodal graph, the degree-of-freedom constraint matrix, and the welding variation energy constraint field are encoded as different semantic levels to obtain the constraint condition context representation; By utilizing the semantic reasoning engine and the physical calculation engine through the constraint context representation, the target parameterized fixture layout scheme is obtained.
[0054] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the parametric design method for welding fixtures as described above.
[0055] Alternatively, a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following operations: Heterogeneous multimodal graphs are constructed based on multimodal information, wherein the multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data; The degree-of-freedom constraint matrix and welding variation energy constraint field are constructed using the heterogeneous multimodal graph. The heterogeneous multimodal graph, the degree-of-freedom constraint matrix, and the welding variation energy constraint field are encoded as different semantic levels to obtain the constraint condition context representation; By utilizing the semantic reasoning engine and the physical calculation engine through the constraint context representation, the target parameterized fixture layout scheme is obtained.
[0056] The present invention will now be described an electronic device 300 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic device 300 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 300 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.
[0057] Electronic device 300 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or a computer program loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.
[0058] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. In this application, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention according to actual needs. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units.
[0059] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.
Claims
1. A parametric design method for welding fixtures, characterized in that, include: A heterogeneous multimodal map is constructed based on multimodal information, wherein the multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data, including: The welding process planning data and the production line environment data are analyzed respectively to obtain the process node set and the environment node set; By performing topological structure analysis on the three-dimensional geometric model data, a set of geometric nodes is obtained; The heterogeneous multimodal map is obtained through the set of geometric nodes, the set of process nodes, and the set of environmental nodes; The degree-of-freedom constraint matrix and welding deformation energy constraint field are constructed using the heterogeneous multimodal graph, including: The degree-of-freedom constraint matrix is constructed using the localization feature information of the heterogeneous multimodal map, including: Multiple constraint screws are obtained based on the positioning feature information of the heterogeneous multimodal map, wherein the constraint screws are applied to the rigid body of the workpiece, and the positioning feature information is used to represent the positioning element and clamping element information; The constraint screw includes: ， Where S is the constraint spiral. ω is the angular velocity vector, and V is the linear velocity vector; The degree-of-freedom constraint matrix is obtained by using all the aforementioned constraint spirals; The degree-of-freedom constraint matrix includes: ， wherein J is the degree of freedom constraint matrix, are m constraint helices; Based on the inherent strain prediction algorithm, the inherent strain tensor is obtained from the heterogeneous multimodal spectrum; The inherent strain tensor includes: ， in, Let the intrinsic strain tensor be... For plastic strain, For thermal strain, For other strains; The welding deformation energy constraint field is constructed using the inherent strain tensor; The heterogeneous multimodal graph, the degree-of-freedom constraint matrix, and the welding variation energy constraint field are encoded as different semantic levels to obtain the constraint condition context representation; By utilizing the semantic reasoning engine and the physical calculation engine through the constraint context representation, the target parameterized fixture layout scheme is obtained.
2. The parametric design method for welding fixtures according to claim 1, characterized in that, The heterogeneous multimodal map includes: ， Wherein, G represents the heterogeneous multimodal spectrum. For the set of geometric nodes, The set of process nodes, Let E be the set of environmental nodes, and E be the topological adjacency relationship between the nodes.
3. The parametric design method for welding fixtures according to claim 1, characterized in that, The construction of the welding deformation energy constraint field through the inherent strain tensor includes: The welding deformation energy density is obtained through the inherent strain tensor. The welding deformation energy density includes: ， in, The weld deformation energy density, C is the transpose of the inherent strain tensor, and C is the energy density matrix of the welding deformation. The welding deformation energy constraint field is constructed using the welding deformation energy density; The welding deformation energy confinement field includes: ， Wherein, U is the welding deformation energy constraint field. For spatial domain.
4. The parametric design method for welding fixtures according to claim 1, characterized in that, The process of obtaining a target parameterized fixture layout scheme by utilizing a semantic reasoning engine and a physical computing engine through the constraint context representation includes: Based on the multimodal large model, the semantic reasoning engine is used to perform hierarchical reasoning through the contextual representation of the constraint conditions to obtain the fixture layout scheme; Based on the obstacle avoidance potential energy field and the structural stiffness objective function, the physical calculation engine is used to adaptively adjust the fixture layout scheme to obtain the target parameterized fixture layout scheme.
5. A parametric design device for welding fixtures, characterized in that, include: A heterogeneous multimodal mapping construction module is used to construct a heterogeneous multimodal mapping based on multimodal information. The multimodal information includes three-dimensional geometric model data of the body parts to be welded, welding process planning data, and production line environment data, including: The welding process planning data and the production line environment data are analyzed respectively to obtain the process node set and the environment node set; By performing topological structure analysis on the three-dimensional geometric model data, a set of geometric nodes is obtained; The heterogeneous multimodal map is obtained through the set of geometric nodes, the set of process nodes, and the set of environmental nodes; The constraint acquisition module is used to construct a degree-of-freedom constraint matrix and a welding deformation form energy constraint field through the heterogeneous multimodal graph, including: The degree-of-freedom constraint matrix is constructed using the localization feature information of the heterogeneous multimodal map, including: Multiple constraint screws are obtained based on the positioning feature information of the heterogeneous multimodal map, wherein the constraint screws are applied to the rigid body of the workpiece, and the positioning feature information is used to represent the positioning element and clamping element information; The constraint screw includes: ， Where S is the constraint spiral. ω is the angular velocity vector, and V is the linear velocity vector; The degree-of-freedom constraint matrix is obtained by using all the aforementioned constraint spirals; The degree-of-freedom constraint matrix includes: ， Where J is the degree-of-freedom constraint matrix, There are m constraint screws; Based on the inherent strain prediction algorithm, the inherent strain tensor is obtained from the heterogeneous multimodal spectrum; The inherent strain tensor includes: ， in, Let the intrinsic strain tensor be... For plastic strain, For thermal strain, For other strains; The welding deformation energy constraint field is constructed using the inherent strain tensor; The constraint context representation acquisition module is used to encode the heterogeneous multimodal graph, the degree-of-freedom constraint matrix and the welding deformation energy constraint field as different semantic levels to obtain the constraint context representation. The target parameterized fixture layout scheme acquisition module is used to obtain the target parameterized fixture layout scheme by utilizing the semantic reasoning engine and the physical calculation engine through the constraint context representation.
6. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the parametric design method for welding fixtures as described in any one of claims 1 to 4 when executing the computer program.
7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the parametric design method for welding fixtures as described in any one of claims 1 to 4.