A method for analyzing the structural performance of sheet metal parts based on multi-scale modeling

Abstract

The present invention provides a method for analyzing the structural performance of sheet metal parts based on multi-scale modeling. The method includes the following steps: Step S1, perform high-precision scanning on the solid sheet metal part to obtain the point cloud data of the sheet metal part and perform preprocessing; Step S2, import the preprocessed point cloud data into reverse engineering software for surface reconstruction to generate a three-dimensional digital model of the sheet metal part, and perform feature recognition and parameterization processing to extract key dimensional parameters; Step S3, perform mesh division on the parameterized three-dimensional digital model, use the multi-scale modeling method to generate finite element mesh models with different densities, and perform mesh encryption in key areas; Step S4, define loads and boundary conditions for the multi-scale model, select material property parameters, and use the finite element analysis method to perform structural performance simulation analysis on the multi-scale model under multiple working conditions. The present invention can comprehensively evaluate performance indicators such as the strength, stiffness, and deformation of sheet metal parts under different loads and boundary conditions.

Description

Technical Field

[0001] The present invention relates to a method for analyzing the structural performance of sheet metal parts, specifically to a method for analyzing the structural performance of sheet metal parts based on multi-scale modeling, belonging to the technical field of sheet metal part structure analysis. Background Art

[0002] Sheet metal parts are a type of component widely used in modern manufacturing, mainly manufactured through processes such as stamping, bending, and stretching of metal sheets. For sheet metal part products such as electrical equipment cabinets and electronic chassis, conducting structural performance analysis is very important for evaluating their reliability and safety. In many cases, the replacement of such products is relatively fast, and the original design data is incomplete, making it impossible to conduct structural analysis based on the original design documents. On the other hand, such products may encounter design changes, repairs, and modifications during actual use, resulting in differences between their structures and the original designs. In these cases, it is challenging to conduct structural performance analysis of sheet metal parts based on reverse engineering.

[0003] The structural performance analysis of the prior art is usually based on three-dimensional modeling. Since the local product features of such sheet metal parts are rich, full-size high-precision modeling will result in an excessive amount of model data. Therefore, it is prone to cause great difficulties in subsequent finite element mesh generation and solution. The operations of surface fitting and detail simplification have the effect of reducing the model complexity, but for stress and strain concentration regions, such as welded joints, bolt connections, and stiffeners, this simplified model is difficult to accurately describe the local geometric details and mechanical behaviors, affecting the accuracy of the structural performance analysis of key parts.

[0004] Therefore, for sheet metal part products such as electrical equipment cabinets and electronic chassis, there is an urgent need for an efficient and flexible multi-scale modeling method to conduct structural performance analysis on them. Summary of the Invention

[0005] Based on the above background, the purpose of the present invention is to provide a method for analyzing the structural performance of sheet metal parts based on multi-scale modeling to solve the problems described in the background art.

[0006] To achieve the above invention purpose, the present invention provides the following technical solutions:

[0007] A method for analyzing the structural performance of sheet metal parts based on multi-scale modeling, the method comprising the following steps:

[0008] Step S1, using three-dimensional scanning technology to perform high-precision scanning on the solid sheet metal part to obtain the point cloud data of the sheet metal part, and preprocessing the point cloud data;

[0009] Step S2: Import the preprocessed point cloud data into reverse engineering software for surface reconstruction to generate a 3D digital model of the sheet metal part, and perform feature recognition and parameterization on the 3D digital model to extract key dimensional parameters;

[0010] Step S3: Mesh the parameterized 3D digital model, use the multi-scale modeling method to generate finite element mesh models with different densities, and encrypt the mesh in key areas to obtain a multi-scale model of the sheet metal part;

[0011] Step S4: Define loads and boundary conditions for the multi-scale model, select material property parameters, and use the finite element analysis method to perform structural performance simulation analysis on the multi-scale model under multiple working conditions.

[0012] Preferably, step S1 specifically includes:

[0013] Step S11: Select a suitable 3D scanning device and determine scanning parameters according to the size, shape complexity, and required accuracy of the sheet metal part;

[0014] Step S12: Perform surface treatment on the solid sheet metal part and fix the solid sheet metal part in an appropriate position;

[0015] Step S13: Scan the solid sheet metal part from multiple angles and directions through the 3D scanning device to obtain the point cloud data of the solid sheet metal part;

[0016] Step S14: Preprocess the obtained point cloud data, and the preprocessing includes point cloud filtering, point cloud registration, point cloud downsampling, and point cloud smoothing.

[0017] Preferably, step S2 specifically includes:

[0018] Step S21: Import the preprocessed point cloud data into reverse engineering software;

[0019] Step S22: Perform surface reconstruction on the imported point cloud data to generate a 3D digital model of the sheet metal part;

[0020] Step S23: Perform post-processing operations such as model repair, model simplification, and model fairing on the reconstructed 3D digital model;

[0021] Step S24: Perform feature recognition on the post-processed 3D digital model, extract key geometric features such as planes, round holes, bends, and stamping of the sheet metal part, classify, label, and organize the identified features, and establish a complete feature tree;

[0022] Step S25: Parametrize the identified features, extract key dimension parameters such as planar dimensions, round hole diameters, bending radii, stamping depths, etc., transform the geometric information of the features into parametric dimension constraints and relationships, and establish a parametric three-dimensional digital model.

[0023] Preferably, step S3 specifically includes:

[0024] Step S31: Use an intelligent feature recognition method based on machine learning to perform geometric cleaning and simplification on the three-dimensional digital model;

[0025] Step S32: Mesh the simplified three-dimensional digital model to generate a global-scale finite element mesh model, and define the material property parameters of the sheet metal part;

[0026] Step S33: Identify the stress concentration areas in the bending area, punching area and welding area of the sheet metal part, perform local geometry extraction on the stress concentration areas to generate independent local sub-models, define local coordinate systems, and establish geometric coupling relationships between the local sub-models and the global-scale finite element mesh model;

[0027] Step S34: Use high-order element types and refined mesh sizes to densify the mesh of the local sub-model;

[0028] Step S35: Perform error estimation, evaluate the quality and convergence of the current finite element mesh, and adopt an adaptive mesh generation strategy driven by an evolutionary algorithm to adjust the mesh density and element size in areas with large errors according to the key geometric feature information extracted by the intelligent feature recognition method. Repeat the error estimation and adaptive mesh densification process until the preset convergence accuracy threshold is reached;

[0029] Step S36: Embed the local sub-model into the global-scale finite element mesh model to form a multi-scale model of the sheet metal part.

[0030] Preferably, step S31 specifically includes:

[0031] Step S311: Collect a certain number of three-dimensional digital models of sheet metal parts of different types as the original data set; manually label the features of each three-dimensional digital model of the sheet metal part, identify key features such as the bending area, punching area and welding area to form a labeled data set; convert the labeled three-dimensional digital model of the sheet metal part into a mesh model and perform mesh cleaning and repair; perform normalization processing on the mesh model, convert it into a voxel representation, and generate a three-dimensional voxel mesh;

[0032] Step S312: Construct a 3D CNN network model for extracting multi-scale geometric features from a 3D voxel grid. The 3D CNN network model includes 2 convolutional layers, 2 pooling layers, 2 fully connected layers, a multi-scale feature extraction and fusion module, and an attention module;

[0033] Step S313: Input the 3D voxel grid labeled in Step S311 into the 3D CNN network model, train the model, and optimize the network parameters to minimize the feature prediction error; Use the trained 3D CNN network model to extract features from the 3D digital model of the sheet metal part to obtain the feature vector representation of each voxel; Convert the extracted voxel feature vectors into a graph representation, with each voxel as a node of the graph, and establish edge connections between adjacent voxels;

[0034] Step S314: Construct a GNN network model for feature recognition and segmentation in the graph domain. The GNN network model includes 2 graph convolutional layers, 2 graph attention layers, 2 graph pooling layers, 1 global pooling layer, and 1 fully connected layer;

[0035] Step S315: Input the graph constructed in Step S313 into the GNN network model, train the model, and optimize the network parameters to minimize the feature classification error; Use the trained GNN network model to perform feature recognition and segmentation on the 3D digital model of the sheet metal part to obtain the feature category of each voxel;

[0036] Step S316: According to the feature segmentation result of the GNN network model, perform geometric cleaning and simplification on the 3D digital model; For ordinary regions, adopt the edge collapse algorithm to merge adjacent patches and simplify the mesh complexity; For key feature regions, retain the original geometric topology and do not simplify; For symmetric or periodic structures, identify and extract the smallest repeating unit, only retain the geometric information of the smallest repeating unit, and impose periodic constraints on the unit boundaries to generate a simplified partial model to reduce the computational scale.

[0037] Preferably, in the three-dimensional CNN network model, the first convolutional layer is used to perform a convolutional operation on the input data; the first pooling layer is used to perform a max-pooling operation on the output of the first convolutional layer to halve the size of the feature map; the multi-scale feature extraction and fusion module is used to first perform convolutional and pooling operations on the output of the first pooling layer in three branches in parallel, then perform upsampling on the outputs of the three branches, and finally splice the upsampling results of the three branches in the feature dimension to obtain a fused multi-scale feature representation; the second convolutional layer is used to perform a convolutional operation on the multi-scale feature; the attention module is used to perform feature recalibration and weight adjustment on the output of the second convolutional layer to generate an attention weight map with the same size as the input feature map, and then multiply the attention weight map by the output of the second convolutional layer to obtain an attention-enhanced feature map; the second pooling layer is used to perform a max-pooling operation on the attention-enhanced feature map to halve the size of the feature map; the first fully-connected layer is used to flatten the output of the second pooling layer to obtain a high-dimensional feature vector, and map the feature vector to a high-dimensional hidden representation through a fully-connected operation; the second fully-connected layer is used to map the output of the first fully-connected layer to a K-dimensional output vector through a fully-connected operation, where K is the number of feature categories.

[0038] Preferably, in the GNN network model, the first graph convolutional layer is used to perform a graph convolutional operation on the input data; the first graph attention layer is used to perform feature aggregation and update on the output of the first graph convolutional layer; the first graph pooling layer is used to perform a pooling operation on the output of the first graph attention layer; the second graph convolutional layer is used to perform a graph convolutional operation on the output of the first graph pooling layer; the second graph attention layer is used to perform feature aggregation and update on the output of the second graph convolutional layer; the second graph pooling layer is used to perform a pooling operation on the output of the second graph attention layer; the global pooling layer is used to perform a global pooling operation on the output of the second graph pooling layer to aggregate the graph-level features into a vector representation to obtain a global feature vector; the fully-connected layer is used to map the output of the global pooling layer to the number of feature categories K through a fully-connected operation to obtain the probability distribution of each category.

[0039] Preferably, step S36 specifically includes:

[0040] Establish a position constraint relationship based on the coupling of node degrees of freedom, apply a displacement compatibility equation at the interface between the local sub-model and the global-scale finite element mesh model to ensure the continuity of displacement and rotation angle, and define a load transfer mechanism between the local sub-model and the global-scale finite element mesh model to achieve two-way interaction of multi-scale information, thereby forming a multi-scale model of the sheet metal part.

[0041] Preferably, step S4 specifically includes:

[0042] Step S41: Analyze the static load, dynamic load, and impact load encountered by the sheet metal part during actual use. For each load condition, determine the magnitude, direction, and acting position of the load; apply the simplified and equivalent load conditions to the corresponding positions of the multi-scale model, define boundary conditions such as displacement constraints and symmetry conditions, and simulate the connection and support methods between the sheet metal part and the surrounding structures.

[0043] Step S42: According to the material type of the sheet metal part, select an elastoplastic model or an anisotropic model, and input the elastic modulus, Poisson's ratio, yield strength, and hardening curve of the material.

[0044] Step S43: Select an implicit solver or an explicit solver to match different load conditions and material nonlinear characteristics, set the time step, iteration convergence criterion, and stress failure criterion, and perform nonlinear finite element analysis on the multi-scale model to obtain the displacement field, stress field, and strain field of the sheet metal part under different conditions.

[0045] Step S44: Evaluate the strength, stiffness, and stability of the sheet metal part under different conditions, and analyze the fracture failure mode and fatigue failure position of the sheet metal part under different conditions.

[0046] Compared with the prior art, the present invention has the following advantages:

[0047] A method for analyzing the structural performance of a sheet metal part based on multi-scale modeling according to the present invention, through the integration of 3D scanning, reverse engineering, multi-scale modeling, and finite element analysis, supports the simulation analysis of structural performance under multiple conditions, can comprehensively evaluate performance indicators such as the strength, stiffness, and deformation of the sheet metal part under different loads and boundary conditions. By defining key regions and performing mesh encryption, it is possible to specifically analyze the local stress and strain concentration of the sheet metal part and evaluate potential damage and failure risks. Detailed implementation mode

[0048] The following will further specifically illustrate the technical solutions of the present invention through specific embodiments. It should be understood that the implementation of the present invention is not limited to the following embodiments, and any form of modification and / or change made to the present invention will fall within the protection scope of the present invention.

[0049] In the present invention, unless otherwise specified, all parts and percentages are in weight units, and the equipment and raw materials used can be purchased from the market or are commonly used in the art. The methods in the following embodiments, unless otherwise specified, are conventional methods in the art. The components or equipment in the following embodiments, unless otherwise specified, are general standard parts or components known to those skilled in the art, and their structures and principles can all be known to those skilled in the art through technical manuals or obtained through conventional experimental methods.

[0050] The following provides a detailed description of the embodiments of the present invention. In the following detailed description, for the sake of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present invention. However, one or more embodiments can also be implemented by those skilled in the art without these specific details.

[0051] An embodiment of the present invention discloses a method for analyzing the structural performance of sheet metal parts based on multi-scale modeling. The method includes the following steps:

[0052] Step S1: Use three-dimensional scanning technology to perform high-precision scanning on the solid sheet metal part to obtain the point cloud data of the sheet metal part, and preprocess the point cloud data;

[0053] Step S2: Import the preprocessed point cloud data into reverse engineering software for surface reconstruction to generate a three-dimensional digital model of the sheet metal part, and perform feature recognition and parameterization on the three-dimensional digital model to extract key dimension parameters;

[0054] Step S3: Perform mesh division on the parameterized three-dimensional digital model, use the multi-scale modeling method to generate finite element mesh models with different densities, and perform mesh encryption in key areas to obtain the multi-scale model of the sheet metal part;

[0055] Step S4: Define loads and boundary conditions for the multi-scale model, select material property parameters, and use the finite element analysis method to perform structural performance simulation analysis on the multi-scale model under multiple working conditions.

[0056] This method provides a new solution for the structural performance analysis of sheet metal parts by integrating three-dimensional scanning, reverse engineering, multi-scale modeling, and finite element analysis. Through three-dimensional scanning technology and reverse engineering software, it realizes the rapid conversion from a solid sheet metal part to a digital model, reducing the workload of manual modeling. The multi-scale modeling method uses meshes with different densities in different regions, reducing the computational scale in non-critical regions while ensuring the accuracy of key regions, and balancing the computational scale and analysis efficiency as a whole. This method supports structural performance simulation analysis under multiple working conditions, can comprehensively evaluate performance indicators such as strength, stiffness, and deformation of sheet metal parts under different loads and boundary conditions, and can specifically analyze the local stress and strain concentration of sheet metal parts by defining key regions and performing mesh encryption, and evaluate potential damage and failure risks.

[0057] Step S1 specifically includes:

[0058] Step S11: Select a suitable three-dimensional scanning device, for example, a structured light three-dimensional scanner, a laser scanner, or a handheld scanner, and determine scanning parameters such as scanning resolution, scanning distance, and scanning angle according to the size, shape complexity, and required accuracy of the sheet metal part;

[0059] Step S12: Perform surface treatment on the solid sheet metal part. For example, spray a light-shielding agent or attach reflective marking points to improve the scanning quality and feature recognition ability of the surface. Fix the solid sheet metal part in an appropriate position to ensure stability during the scanning process;

[0060] Step S13: Use a 3D scanning device to scan the solid sheet metal part from multiple angles and directions to obtain the point cloud data of the solid sheet metal part;

[0061] Step S14: Preprocess the obtained point cloud data. The preprocessing includes point cloud filtering, point cloud registration, point cloud downsampling, and point cloud smoothing. Statistical filtering method or radius filtering method can be used for point cloud filtering, ICP algorithm can be used for point cloud registration, uniform sampling method or curvature sampling method can be used for point cloud downsampling, and moving least squares method can be used for point cloud smoothing.

[0062] In steps S11 - S14, in this embodiment, 3D scanning technology is used to quickly and efficiently obtain the high-precision point cloud data of the solid sheet metal part. The preprocessed point cloud data can better reflect the geometric features and detailed information of the solid sheet metal part.

[0063] Step S2 specifically includes:

[0064] Step S21: Import the preprocessed point cloud data into reverse engineering software such as Geomagic Design X, CATIA, SolidWorks, etc.;

[0065] Step S22: Perform surface reconstruction on the imported point cloud data to generate a 3D digital model of the sheet metal part. Use surface fitting tools such as mesh fitting and NURBS surface fitting in the reverse engineering software to convert the discrete point cloud data into a continuous surface model. According to the geometric features and shape complexity of the sheet metal part, select fitting algorithms such as the least squares method and B-spline surface to obtain the 3D digital model;

[0066] Step S23: Perform post-processing operations of model repair, model simplification, and model fairing on the reconstructed 3D digital model. Model repair is to repair defects such as small holes, gaps, and self-intersections in the model. Model simplification is to remove unnecessary detailed features such as tiny protrusions and holes. Model fairing is to smooth the surface to reduce noise and jaggedness;

[0067] Step S24: Perform feature recognition on the post-processed 3D digital model, extract key geometric features such as planes, round holes, bends, and stamping of the sheet metal part, classify, label, and organize the recognized features, and establish a complete feature tree;

[0068] Step S25: Parametrize the identified features, extract key dimension parameters such as planar dimensions, round hole diameters, bending radii, stamping depths, etc., transform the geometric information of the features into parametric dimension constraints and relationships, and establish a parametric three-dimensional digital model.

[0069] In steps S21 - S25, in this embodiment, the preprocessed point cloud data is transformed into a parametric three-dimensional digital model, realizing the conversion from discrete data to a continuous model, and providing a digital basis for subsequent finite element analysis and structural optimization.

[0070] Step S3 specifically includes:

[0071] Step S31: Use an intelligent feature recognition method based on machine learning to perform geometric cleaning and simplification on the three-dimensional digital model, detect and extract the key geometric features of the sheet metal part, retain the main geometric features and topological relationships of the sheet metal part, remove the detailed features with little impact on the structural performance, and for symmetric or periodic structures, identify and extract the minimum repeating unit to generate a simplified partial model to reduce the computational scale;

[0072] Step S32: Perform mesh division on the simplified three-dimensional digital model to generate a global-scale finite element mesh model, and define the material property parameters of the sheet metal part;

[0073] Step S33: Identify the stress concentration areas in the bending area, punching area, and welding area of the sheet metal part, perform local geometry extraction on the stress concentration areas to generate independent local sub-models, define local coordinate systems, and establish geometric coupling relationships between the local sub-models and the global-scale finite element mesh model;

[0074] Step S34: Use high-order element types and refined mesh sizes to perform mesh encryption on the local sub-models;

[0075] Step S35: Perform error estimation, evaluate the quality and convergence of the current finite element mesh, and according to the key geometric feature information extracted by the intelligent feature recognition method, adopt an adaptive mesh division strategy driven by an evolutionary algorithm to adjust the mesh density and element size in areas with large errors, and repeat the error estimation and adaptive mesh encryption process until the preset convergence accuracy threshold is reached;

[0076] Step S36: Embed the local sub-models into the global-scale finite element mesh model, establish position constraint relationships based on node degree-of-freedom coupling, apply displacement compatibility equations on the interfaces between the local sub-models and the global-scale finite element mesh model to ensure the continuity of displacements and rotations, and define the load transfer mechanism between the local sub-models and the global-scale finite element mesh model to achieve two-way interaction of multi-scale information, forming a multi-scale model of the sheet metal part.

[0077] In steps S31 - S36, in this embodiment, machine learning and evolutionary algorithms are introduced in key links to achieve full - process intelligent processing from geometric processing to mesh generation. It can adaptively identify key geometric features at different scales and adaptively encrypt local meshes according to the feature recognition results.

[0078] Among them, step S31 specifically includes:

[0079] Step S311: Collect a certain number of three - dimensional digital models of sheet metal parts of different types as the original data set; manually label the features of each three - dimensional digital model of sheet metal parts, identify key features such as the bending area, punching area, and welding area, etc., to form a labeled data set; convert the labeled three - dimensional digital model of sheet metal parts into a mesh model and perform mesh cleaning and repair to eliminate noise and topological errors; perform normalization processing on the mesh model, scale it to a unified size range, and convert it into a voxel representation to generate a 64×64×64 three - dimensional voxel mesh;

[0080] Step S312: Construct a three - dimensional CNN network model for extracting multi - scale geometric features from the three - dimensional voxel mesh. The three - dimensional CNN network model includes 2 convolutional layers, 2 pooling layers, 2 fully - connected layers, a multi - scale feature extraction and fusion module, and an attention module; the input data is a 64×64×64 three - dimensional voxel mesh. The first convolutional layer performs a 3×3×3 convolution operation on the input data and outputs 32 feature maps; the first pooling layer performs a 2×2×2 max - pooling operation on the output of the first convolutional layer to halve the size of the feature maps; the multi - scale feature extraction and fusion module first performs convolutional and pooling operations on the output of the first pooling layer in three branches in parallel. The first branch performs a 3×3×3 convolution operation and a 2×2×2 max - pooling operation, the second branch performs a 3×3×3 convolution operation and a 4×4×4 max - pooling operation, and the third branch performs a 3×3×3 convolution operation and an 8×8×8 max - pooling operation. Then, perform up - sampling on the outputs of the three branches, and finally splice the up - sampling results of the three branches in the feature dimension to obtain a fused multi - scale feature representation; the second convolutional layer performs a 3×3×3 convolution operation on the multi - scale features and outputs 128 feature maps; the attention module performs feature recalibration and weight adjustment on the output of the second convolutional layer to generate an attention weight map with the same size as the input feature map, and then multiply the attention weight map by the output of the second convolutional layer to obtain an attention - enhanced feature map; the second pooling layer performs a 2×2×2 max - pooling operation on the attention - enhanced feature map to halve the size of the feature maps; the first fully - connected layer flattens the output of the second pooling layer to obtain a high - dimensional feature vector, and maps the feature vector to a 1024 - dimensional hidden representation through a fully - connected operation; the second fully - connected layer maps the output of the first fully - connected layer to a K - dimensional output vector through a fully - connected operation, where K is the number of feature categories;

[0081] Step S313: Input the three-dimensional voxel grid labeled in step S311 into the three-dimensional CNN network model, train the model, and optimize the network parameters to minimize the feature prediction error; use the trained three-dimensional CNN network model to extract features from the three-dimensional digital model of the sheet metal part to obtain the feature vector representation of each voxel; convert the extracted voxel feature vectors into a graph representation, with each voxel as a node of the graph, and establish edge connections between adjacent voxels.

[0082] Step S314: Construct a GNN network model for feature recognition and segmentation in the graph domain. The GNN network model includes 2 graph convolutional layers, 2 graph attention layers, 2 graph pooling layers, 1 global pooling layer, and 1 fully connected layer; the input data is the node feature matrix converted from the output of the three-dimensional CNN network model. The first graph convolutional layer performs graph convolution operations on the input data to update the feature representation of each node, and the output feature dimension is 256; the first graph attention layer performs feature aggregation and update on the output of the first graph convolutional layer, and the output feature dimension is 256; the first graph pooling layer performs a pooling operation on the output of the first graph attention layer to reduce the resolution and computational complexity of the graph, and the pooling ratio is 0.5; the second graph convolutional layer performs graph convolution operations on the output of the first graph pooling layer, and the output feature dimension is 128; the second graph attention layer performs feature aggregation and update on the output of the second graph convolutional layer, and the output feature dimension is 128; the second graph pooling layer performs a pooling operation on the output of the second graph attention layer to further reduce the resolution and computational complexity of the graph, and the pooling ratio is 0.5; the global pooling layer performs a global pooling operation on the output of the second graph pooling layer to aggregate the graph-level features into a vector representation to obtain a 128-dimensional global feature vector; the fully connected layer maps the output of the global pooling layer to the number of feature categories K through a fully connected operation to obtain the probability distribution of each category.

[0083] Step S315: Input the graph constructed in step S313 into the GNN network model, train the model, and optimize the network parameters to minimize the feature classification error; use the trained GNN network model to perform feature recognition and segmentation on the three-dimensional digital model of the sheet metal part to obtain the feature categories of each voxel.

[0084] Step S316: According to the feature segmentation result of the GNN network model, perform geometric cleaning and simplification on the three-dimensional digital model; for ordinary regions, use the edge collapse algorithm to merge adjacent patches and simplify the mesh complexity; for key feature regions, retain the original geometric topology and do not simplify; for symmetric or periodic structures, identify and extract the smallest repeating unit, only retain the geometric information of the smallest repeating unit, and apply periodic constraints on the unit boundary to generate a simplified partial model to reduce the computational scale.

[0085] In steps S311 - S316, in this embodiment, deep learning technology is utilized to construct an intelligent feature recognition method based on 3D CNN and GNN, and through it, the key geometric features of sheet metal parts are automatically extracted and recognized, as well as the model simplification process is guided. Considering that electrical equipment cabinets and electronic chassis usually have the characteristics of being relatively large in size, containing multiple functional areas and components, having a relatively regular geometric structure, having a large number of symmetrical and repetitive structures, and the distribution of key features such as bending areas, punching areas, welding areas, etc. is relatively concentrated, in this embodiment, a multi-scale feature fusion mechanism is introduced into the 3D CNN network to better capture geometric features at different scales, and an attention mechanism is introduced to better focus on the key feature areas of sheet metal parts. The GNN network and the 3D CNN network form a collaborative feature extraction and recognition framework.

[0086] Among them, step S35 specifically includes:

[0087] Step S351: Extract the stress components on each element of the current finite element mesh. Adopt the Zienkiewicz-Zhu error estimation method, use the stress values at the element nodes to recover a smooth stress field, calculate the energy norm error between the recovered stress field and the finite element solution on each element, and perform normalization processing on the energy norm errors of all elements to obtain the relative error distribution.

[0088] Step S352: Define the optimization objective of mesh generation as minimizing the overall error and the number of network elements. Design a fitness function, adopt a genetic algorithm, randomly generate an initial mesh generation scheme that meets the constraint conditions as the initial population, repeat the selection operation, crossover operation, and mutation operation, continuously update the population until the maximum number of iterations is reached or the fitness converges, and output the optimal mesh generation scheme, which is used as the basis for mesh refinement.

[0089] Step S353: According to the Zienkiewicz-Zhu error estimation results, determine the elements with a relative error greater than 10% as the candidate areas for mesh refinement, refine these elements, and reduce the element size to 5 mm. Further reduce the element size to 2 mm in the key feature areas to improve the local accuracy.

[0090] Step S354: Re-perform finite element analysis on the mesh after encryption adjustment to obtain an updated stress field. Repeat steps S351 - S353 until the overall relative error is less than 5%, or the preset maximum number of iterations is reached.

[0091] In steps S351 - S354, considering error estimation, mesh quality, and computational efficiency comprehensively, through evolutionary algorithms and intelligent strategies, the rapid optimization of finite element meshes is achieved, reducing the workload of manual adjustment and repeated trial calculations, and improving the efficiency of finite element analysis.

[0092] Step S4 specifically includes:

[0093] Step S41: Analyze the static load, dynamic load, and impact load encountered by the sheet metal part during actual use. For each load condition, determine the magnitude, direction, and acting position of the load; apply the simplified and equivalent load conditions to the corresponding positions of the multi-scale model, define boundary conditions such as displacement constraints and symmetry conditions, and simulate the connection and support methods between the sheet metal part and the surrounding structure;

[0094] Step S42: According to the material type of the sheet metal part, select an elastoplastic model or an anisotropic model, and input the elastic modulus, Poisson's ratio, yield strength, and hardening curve of the material;

[0095] Step S43: Select an implicit solver or an explicit solver to match different load conditions and material nonlinear characteristics, set the time step, iteration convergence criterion, and stress failure criterion, and perform nonlinear finite element analysis on the multi-scale model to obtain the displacement field, stress field, and strain field of the sheet metal part under different conditions;

[0096] Step S44: Evaluate the strength, stiffness, and stability of the sheet metal part under different conditions, and analyze the fracture failure mode and fatigue failure position of the sheet metal part under different conditions.

[0097] Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention. It should be noted that for those of ordinary skill in the art of this technology, without departing from the principle of the present invention, several improvements and modifications can still be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A sheet metal structural performance analysis method based on multi-scale modeling, characterized in that: The method comprises the following steps: Step S1, using three-dimensional scanning technology to perform high-precision scanning on the physical sheet metal part, obtain point cloud data of the sheet metal part, and pre-process the point cloud data; Step S2, importing the pre-processed point cloud data into reverse engineering software for surface reconstruction, generating a three-dimensional digital model of the sheet metal part, and performing feature recognition and parameterization processing on the three-dimensional digital model to extract key dimensional parameters; Step S3, meshing the parameterized three-dimensional digital model, generating finite element mesh models of different densities using a multi-scale modeling method, and performing mesh encryption in key areas to obtain a multi-scale model of the sheet metal part; Step S4, defining loads and boundary conditions for the multi-scale model, selecting material property parameters, and using a finite element analysis method to perform structural performance simulation analysis on the multi-scale model under multiple working conditions; Wherein, the step S3 specifically includes: Step S31, using an intelligent feature recognition method based on machine learning to geometrically clean and simplify the three-dimensional digital model; Step S32, meshing the simplified three-dimensional digital model to generate a global scale finite element mesh model, and defining material property parameters of the sheet metal part; Step S33, identifying stress concentration areas in the bending area, punching area and welding area of ​​the sheet metal part, performing local geometric extraction on the stress concentration area, generating an independent local sub-model, defining a local coordinate system, and establishing a geometric coupling relationship between the local sub-model and the global scale finite element mesh model; Step S34, using a high-order unit type and a refined grid size to encrypt the grid of the local sub-model; Step S35: perform error estimation, evaluate the quality and convergence of the current finite element mesh, and use an adaptive meshing strategy driven by an evolutionary algorithm based on key geometric feature information extracted by the intelligent feature recognition method to adjust the mesh density and unit size for areas with large errors, and repeat the error estimation and adaptive mesh encryption process until a preset convergence accuracy threshold is reached; Step S36, embedding the local sub-model into the global scale finite element mesh model to form a multi-scale model of the sheet metal part; the step S31 specifically includes: Step S311, collecting a certain number of three-dimensional digital models of sheet metal parts of different types as original data sets; manually annotating the features of each three-dimensional digital model of the sheet metal part, identifying the key features of the bending area, the punching area and the welding area, and forming a data set with labels; converting the annotated three-dimensional digital model of the sheet metal part into a mesh model and performing mesh cleaning and repair; normalizing the mesh model, converting it into a voxel representation, and generating a three-dimensional voxel mesh; Step S312: constructing a three-dimensional CNN network model for extracting multi-scale geometric features from a three-dimensional voxel grid, wherein the three-dimensional CNN network model includes two convolutional layers, two pooling layers, two fully connected layers, a multi-scale feature extraction fusion module, and an attention module; Step S313, input the three-dimensional voxel grid annotated in step S311 into the three-dimensional CNN network model, train the model, and optimize the network parameters to minimize the feature prediction error; use the trained three-dimensional CNN network model to extract features from the three-dimensional digital model of the sheet metal part to obtain a feature vector representation of each voxel; convert the extracted voxel feature vector into a graph representation, with each voxel as a node of the graph, and establishing edge connections between adjacent voxels; Step S314: construct a GNN network model for feature recognition and segmentation in the graph domain, wherein the GNN network model includes 2 graph convolution layers, 2 graph attention layers, 2 graph pooling layers, 1 global pooling layer and 1 fully connected layer; Step S315, input the graph constructed in step S313 into the GNN network model, train the model, and optimize the network parameters to minimize the feature classification error; use the trained GNN network model to perform feature recognition and segmentation on the three-dimensional digital model of the sheet metal part to obtain the feature category of each voxel; Step S316: According to the feature segmentation results of the GNN network model, the three-dimensional digital model is geometrically cleaned and simplified; for ordinary areas, the edge collapse algorithm is used to merge adjacent faces to simplify the mesh complexity; for key feature areas, the original geometric topological structure is retained without simplification; for symmetrical or periodic structures, the minimum repeating unit is identified and extracted, only the geometric information of the minimum repeating unit is retained, and periodic constraints are imposed on the unit boundaries to generate simplified partial models to reduce the scale of calculation.

2. The sheet metal structural performance analysis method based on multi-scale modeling according to claim 1, characterized in that: The step S1 specifically includes: Step S11, selecting a suitable 3D scanning device, and determining scanning parameters according to the size, shape complexity and required accuracy of the sheet metal part; Step S12, performing surface treatment on the solid sheet metal part and fixing the solid sheet metal part in an appropriate position; Step S13, scanning the physical sheet metal part at multiple angles and in multiple directions by means of a three-dimensional scanning device to obtain point cloud data of the physical sheet metal part; Step S14: preprocessing the acquired point cloud data, including point cloud filtering, point cloud registration, point cloud downsampling and point cloud smoothing.

3. The sheet metal structural performance analysis method based on multi-scale modeling according to claim 1, characterized in that: The step S2 specifically includes: Step S21, importing the pre-processed point cloud data into reverse engineering software; Step S22, performing surface reconstruction on the imported point cloud data to generate a three-dimensional digital model of the sheet metal part; Step S23, performing post-processing operations of model repair, model simplification and model smoothing on the reconstructed three-dimensional digital model; Step S24, performing feature recognition on the post-processed three-dimensional digital model, extracting the plane, circular hole, bend, and stamping key geometric features of the sheet metal part, classifying, annotating, and organizing the recognized features, and establishing a complete feature tree; Step S25, perform parameterization processing on the identified features, extract key dimensional parameters such as plane size, circular hole diameter, bending radius, and punching depth, convert the geometric information of the features into parameterized dimensional constraints and relationships, and establish a parameterized three-dimensional digital model.

4. The sheet metal structural performance analysis method based on multi-scale modeling according to claim 1, characterized in that: In the three-dimensional CNN network model, the first convolution layer is used to perform a convolution operation on the input data; the first pooling layer is used to perform a maximum pooling operation on the output of the first convolution layer to halve the size of the feature map; the multi-scale feature extraction fusion module is used to first perform convolution and pooling operations of three branches on the output of the first pooling layer in parallel, then upsample the outputs of the three branches, and finally splice the upsampled results of the three branches in the feature dimension to obtain a fused multi-scale feature representation; The second convolutional layer is used to perform convolution operations on multi-scale features; the attention module is used to recalibrate and adjust the weights of the output of the second convolutional layer to generate an attention weight map with the same size as the input feature map, and then multiply the attention weight map with the output of the second convolutional layer to obtain the feature map after attention enhancement; The second pooling layer is used to perform the maximum pooling operation on the feature map after attention enhancement to halve the size of the feature map; the first fully connected layer is used to flatten the output of the second pooling layer to obtain a high-dimensional feature vector, and map the feature vector to a high-dimensional hidden representation through a fully connected operation; The second fully connected layer is used to map the output of the first fully connected layer to a K-dimensional output vector through a fully connected operation, where K is the number of feature categories.

5. The sheet metal structural performance analysis method based on multi-scale modeling according to claim 1, characterized in that: In the GNN network model, the first graph convolution layer is used to perform graph convolution operations on input data; the first graph attention layer is used to perform feature aggregation and update on the output of the first graph convolution layer; the first graph pooling layer is used to perform pooling operations on the output of the first graph attention layer; the second graph convolution layer is used to perform graph convolution operations on the output of the first graph pooling layer; the second graph attention layer is used to perform feature aggregation and update on the output of the second graph convolution layer; the second graph pooling layer is used to perform pooling operations on the output of the second graph attention layer; the global pooling layer is used to perform global pooling operations on the output of the second graph pooling layer, aggregate graph-level features into a vector representation, and obtain a global feature vector; the fully connected layer is used to map the output of the global pooling layer to the number of feature categories K through a fully connected operation, and obtain the probability distribution of each category.

6. The sheet metal structural performance analysis method based on multi-scale modeling according to claim 1, characterized in that: The step S36 specifically includes: A position constraint relationship based on the coupling of node degrees of freedom is established, and the displacement coordination equation is applied on the interface between the local sub-model and the global-scale finite element mesh model to ensure the continuity of displacement and rotation. The load transfer mechanism between the local sub-model and the global-scale finite element mesh model is defined to achieve two-way interaction of multi-scale information, thereby forming a multi-scale model of sheet metal.

7. The sheet metal structural performance analysis method based on multi-scale modeling according to claim 1, characterized in that: The step S4 specifically includes: Step S41, analyzing the static load, dynamic load and impact load encountered by the sheet metal in actual use, and determining the size, direction and action position of each load condition; applying the simplified and equivalent load condition to the corresponding position of the multi-scale model, defining displacement constraints and symmetry condition boundary conditions, and simulating the connection and support mode between the sheet metal and the surrounding structure; Step S42: Select an elastic-plastic model or an anisotropic model according to the material type of the sheet metal part, and input the elastic modulus, Poisson's ratio, yield strength and hardening curve of the material; Step S43, selecting an implicit solver or an explicit solver to match different load conditions and material nonlinear characteristics, setting the time step, iterative convergence criterion and stress failure criterion, performing nonlinear finite element analysis on the multi-scale model, and obtaining the displacement field, stress field and strain field of the sheet metal under different conditions; Step S44: Evaluate the strength, stiffness and stability of the sheet metal parts under different working conditions, and analyze the fracture failure mode and fatigue failure position of the sheet metal parts under different working conditions.