Shearing plate design method based on generative topology and related equipment

Abstract

The invention discloses a shear plate design method based on generative topology and related equipment, and relates to the field of metal plate shearing. Comprising the following steps: acquiring a design condition, encoding the design condition into a condition vector, inputting the condition vector into a generative model, and generating a candidate topological configuration; geometric reconstruction and gridding processing are carried out on the candidate topological configuration, and energy consumption evaluation indexes such as a hysteretic curve area, an equivalent damping ratio and a rigidity degradation rate are calculated based on cyclic load finite element analysis; comparing the energy consumption evaluation index with a preset criterion, and determining a target shear plate structure in combination with a rollback optimization mechanism; and an additive manufacturing printing path is constructed based on the target structure, the printing path and process parameters are adaptively adjusted through thermal process prediction, a manufacturing file meeting temperature gradient constraints is output for forming manufacturing, and rapid generation of a shear plate topological structure, energy consumption performance closed-loop optimization and manufacturing process cooperative control are achieved.

Description

Technical Field

[0001] This invention relates to the field of metal plate shearing, and more specifically to a shearing plate design method and related equipment based on generative topology. Background Technology

[0002] Topology optimization technology aims to optimize the material distribution in a structure under given design space, load conditions and constraints in order to achieve design goals such as lightweight, high strength and high performance. It has been widely used in mechanical, architectural and aerospace fields.

[0003] Existing topology optimization methods mainly include density-based and evolutionary algorithm-based methods. These methods typically introduce a large number of design variables and iteratively approximate the optimal structural form. While they can achieve theoretically sound optimization results, they generally suffer from large computational scale, numerous iterations, and long solution cycles in practical applications, making them unsuitable for engineering applications requiring batch processing and rapid design. Furthermore, the structural forms obtained through traditional topology optimization often exhibit problems such as blurred boundaries and numerous local micro-holes, necessitating geometric reconstruction and process modifications in subsequent stages, thereby increasing the design cycle and potentially leading to a decline in structural performance.

[0004] With the development of additive manufacturing technology, its layer-by-layer forming method has improved the manufacturing feasibility of complex structures to a certain extent. However, the following shortcomings still exist in the existing technology: on the one hand, there is a lack of effective coordination mechanism between topology optimization results and additive manufacturing process parameters, making it difficult to balance structural performance and forming quality; on the other hand, in the design process of components such as shear plates that are subjected to cyclic loads, existing topology optimization methods usually take stiffness or mass as the main optimization target, lacking comprehensive constraints on factors such as energy dissipation performance, fatigue performance, and the thermal effects of the forming process. Summary of the Invention

[0005] The technical problem to be solved by this invention is that the shearing design results of existing metal plate shearing optimization methods are difficult to meet mechanical performance requirements while taking into account manufacturing stability and service reliability. The purpose is to provide a shearing plate design method and related equipment based on generative topology, which solves the problem of how to achieve rapid generation of topology structure while ensuring the cyclic energy consumption performance of shearing plate, and how to coordinate and optimize it with additive manufacturing process.

[0006] This invention is achieved through the following technical solution:

[0007] A clipboard design method based on generative topology includes:

[0008] Obtain the design conditions for the shear plate based on generative topology. The design conditions include at least: geometric parameters, boundary connection parameters, cyclic load spectrum parameters, upper volume constraint, and minimum feature size constraint for additive manufacturing.

[0009] Based on the design conditions, the design conditions are encoded into condition vectors and input into the generative model to generate at least one candidate topology configuration;

[0010] Geometric reconstruction and meshing are performed on each candidate topology, followed by cyclic load finite element analysis to obtain a set of energy consumption evaluation indicators. This set of indicators includes at least the area under the hysteresis curve. Equivalent damping ratio and secant stiffness degradation rate ;

[0011] The energy consumption evaluation index set is compared with a preset energy consumption criterion, the preset energy consumption criterion including at least the following: , , When the preset energy consumption criterion is not met, the topology configuration and / or energy consumption evaluation index set are regenerated until the preset energy consumption criterion is met, and the target shear plate structure is obtained.

[0012] An additive manufacturing printing path is constructed based on the target shear plate structure, and the maximum temperature gradient of the printing process is obtained based on a thermal process prediction model. ;

[0013] Will With preset temperature gradient threshold Comparison: When If necessary, adjust at least one process parameter and / or path parameter, and recalculate. until ;when At that time, the output is a printing path and manufacturing file that meet the temperature gradient constraints, which are used to form and manufacture the target shear plate structure.

[0014] Preferably, the generative model is a conditional generative adversarial network (cGAN), and the conditional vector includes at least the geometric dimension parameter, the cyclic load spectrum parameter, and the volume upper limit constraint.

[0015] Preferably, the discriminator of the cGAN is a multi-scale discriminator, which simultaneously constrains the global connectivity structure and local detail features of the candidate topology.

[0016] Preferably, the geometric reconstruction and meshing process includes: performing region growth segmentation on the binary pixel image of the candidate topology, extracting the boundary using an edge detection operator, and performing curve fitting on the boundary to obtain a closed vector contour, and then generating a finite element mesh based on the closed vector contour; the edge detection operator is the Sobel operator, and the curve fitting is the Bezier curve fitting.

[0017] Preferably, after obtaining the finite element mesh, the method further includes: using the maximum equivalent stress under cyclic loading. The closed vector profile is optimized with the goal of minimization to obtain the optimized structure for the cyclic load finite element analysis.

[0018] Preferably, the , , At least one of them is determined using a relative benchmark threshold, which is determined based on the corresponding energy consumption evaluation index of the initial design shear plate.

[0019] Preferably, the The threshold is within a preset range, and when When the adjustment includes at least one of the following: reducing energy density, increasing the scanning interval, adjusting the scanning order to print regions with lower predicted temperatures first and then regions with higher predicted temperatures, or setting the interlayer dwell time; when When printing, the path is output according to the shortest path at the same level.

[0020] This invention also provides a clipboard design system based on generative topology, used to implement the aforementioned clipboard design method based on generative topology, comprising:

[0021] The design condition acquisition unit is used to acquire the design conditions of the shear plate based on generative topology. The design conditions include at least geometric dimension parameters, boundary connection parameters, cyclic load spectrum parameters, volume upper limit constraints, and additive manufacturing minimum feature size constraints.

[0022] A condition encoding unit is used to encode the design conditions into a condition vector;

[0023] A generative topology generation unit is used to input the condition vector into a generative model to generate at least one candidate topology configuration.

[0024] A geometric reconstruction and meshing unit is used to perform geometric reconstruction and meshing processing on the candidate topological configuration to obtain a structural model for finite element analysis; wherein the geometric reconstruction and meshing processing includes at least region segmentation, boundary extraction and curve fitting of the pixel representation of the candidate topological configuration to form a closed vector contour, and generating a finite element mesh based on the closed vector contour;

[0025] The simulation evaluation unit is used to perform cyclic load finite element analysis based on the structural model and calculate the energy consumption evaluation index set, which includes at least the hysteresis curve area A, the equivalent damping ratio ζe, and the secant stiffness degradation rate rd.

[0026] The fallback control unit is used to compare the set of energy consumption evaluation indicators with preset energy consumption criteria, wherein the preset energy consumption criteria include at least the following: , , When the comparison result does not meet the preset energy consumption criterion, the generative topology generation unit is controlled to regenerate the candidate topology configuration and / or the geometric reconstruction and meshing unit and the cyclic energy consumption simulation evaluation unit are controlled to re-evaluate the updated candidate topology configuration until the preset energy consumption criterion is met to determine the target shear plate structure.

[0027] The printing path construction unit is used to construct an additive manufacturing printing path based on the target shear plate structure and generate a manufacturing file.

[0028] The thermal process prediction unit is used to predict the thermal process of the printing process based on the printing path and preset process parameters, and to obtain the maximum temperature gradient Gmax of the printing process.

[0029] Adaptive adjustment unit, used to adjust With preset temperature gradient threshold Comparison: When If necessary, adjust at least one process parameter and / or path parameter, and recalculate. until ;when At that time, the output is a printing path and manufacturing file that meet the temperature gradient constraints, which are used to form and manufacture the target shear plate structure.

[0030] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the clipboard design method based on generative topology as described above.

[0031] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the clipboard design method based on generative topology as described above.

[0032] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0033] This invention enables rapid generation of shear plate topologies based on generative models, avoiding the extensive iterative calculations required in traditional topology optimization, significantly reducing computational costs, and improving structural design efficiency. It is suitable for batch and rapid design scenarios. Using cyclic load energy dissipation performance as the core constraint, this invention evaluates the hysteresis curve area, equivalent damping ratio, and stiffness degradation rate, and combines this with a backoff optimization mechanism to achieve closed-loop optimization design oriented towards energy dissipation performance, effectively ensuring the stability of the structure's mechanical performance under cyclic loading conditions.

[0034] This invention improves the continuity and manufacturability of the topology by performing deterministic reconstruction from pixels to vectors and then to finite element meshes, combined with a shape optimization method based on cyclic stress constraints, thereby reducing the problems of micro-holes and boundary ambiguity. In the additive manufacturing stage, an adaptive adjustment mechanism for path and process parameters based on temperature gradient thresholds is introduced to achieve synergistic optimization of the topology design results and the manufacturing process, improving forming stability and finished product consistency.

[0035] This invention is applicable to the optimization design of shear plate structures subjected to cyclic loads, and can achieve a synergistic improvement in structural energy dissipation performance and manufacturing feasibility while ensuring lightweight and mechanical performance. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0037] Figure 1 A flowchart illustrating the construction of a topology optimization model based on a generative topology clipboard design method;

[0038] Figure 2 The flowchart shows the operation of the topology optimization model based on the generative topology clipboard design method.

[0039] Figure 3 This is a diagram of the generator structure based on the topology optimization model.

[0040] Figure 4 This is a diagram of the discriminator structure based on the topology optimization model. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0042] Example 1

[0043] A clipboard design method based on generative topology, such as Figure 1-2 As shown, it includes:

[0044] Obtain the design conditions for the shear plate based on generative topology. The design conditions include at least: geometric parameters, boundary connection parameters, cyclic load spectrum parameters, upper volume constraint, and minimum feature size constraint for additive manufacturing.

[0045] Based on the design conditions, the design conditions are encoded into condition vectors and input into the generative model to generate at least one candidate topology configuration;

[0046] Geometric reconstruction and meshing are performed on each candidate topology, followed by cyclic load finite element analysis to obtain a set of energy consumption evaluation indicators. This set of indicators includes at least the area under the hysteresis curve. Equivalent damping ratio and secant stiffness degradation rate Among them, the area of ​​the hysteresis curve The equivalent damping ratio is calculated by integral calculation based on the force-displacement curve of the complete loading and unloading cycle. The secant stiffness degradation rate is determined based on the equivalent energy method. The stiffness was calculated based on the change of secant stiffness under adjacent loading cycles.

[0047] The energy consumption evaluation index set is compared with a preset energy consumption criterion, the preset energy consumption criterion including at least the following: , , When the preset energy consumption criterion is not met, the topology configuration and / or energy consumption evaluation index set are regenerated until the preset energy consumption criterion is met, or the preset maximum number of iterations is reached and the iteration is terminated to obtain the target shear plate structure.

[0048] An additive manufacturing printing path is constructed based on the target shear plate structure, and the maximum temperature gradient of the printing process is obtained based on a thermal process prediction model. The thermal process prediction model can be any one of the following: finite element thermal analysis model, simplified heat conduction model, or prediction model trained based on historical data.

[0049] Will With preset temperature gradient threshold Comparison: When If necessary, adjust at least one process parameter and / or path parameter, and recalculate. until ;when At that time, the output is a printing path and manufacturing file that meet the temperature gradient constraints, which are used to form and manufacture the target shear plate structure.

[0050] In this embodiment, energy dissipation evaluation parameters such as hysteresis area, equivalent stiffness, and equivalent damping ratio are calculated for each candidate topology configuration, and compared with the corresponding parameters of the initial design shear plate. The improvement factor of the hysteresis area relative to the initial design shear plate is set to 1.2 to 2.0 times, more preferably 1.4 to 1.8 times; the equivalent damping ratio is not less than 15%, more preferably 18% to 25%; and the secant stiffness degradation rate for three consecutive cycles at the design displacement amplitude is not greater than 20%, more preferably not greater than 10% to 15%. When a candidate topology configuration does not meet the above criteria, the generative topology optimization results are re-acquired and shape optimization processing is performed before re-evaluation, until the energy dissipation performance criteria are met or the preset number of iterations is reached. The equivalent stiffness is calculated using secant stiffness and is used to determine the secant stiffness degradation rate.

[0051] In this embodiment, the preferred values ​​are: hysteresis curve area increase factor of 1.5 times, equivalent damping ratio of 20%, and secant stiffness degradation rate controlled within 15%.

[0052] Through the above evaluation and rollback optimization process, while ensuring that the energy consumption performance of the shear plate meets the design requirements, the edges of the structure are smoothed, thereby improving the stability and consistency of subsequent additive manufacturing.

[0053] Based on the target shear plate structure that meets the above criteria, manufacturing documents are generated, and the structure is formed and manufactured using metal additive manufacturing equipment to obtain a solid shear plate component that meets the mechanical performance requirements.

[0054] In this embodiment, the initial design shear plate is a regular plate-type shear component without topology optimization, and its material parameters, dimensional parameters, and connection method are consistent with the target shear plate. The generative model is a conditional generative adversarial network (cGAN), and the conditional vector includes at least the geometric dimensional parameters, the cyclic load spectrum parameters, and the volume upper limit constraint. The discriminator of the cGAN is a multi-scale discriminator to simultaneously constrain the global connectivity structure and local detail features of the candidate topology configuration.

[0055] The construction and training process of the clipboard-generated topology optimization model is as follows: Figure 1As shown, firstly, an initial shape scheme is set within a preset design space, and constraints such as size, volume fraction, or mass upper limit are given. Based on the optimization criterion (OC), topology optimization of the shear plate structure is performed to form training samples and establish a shear plate topology optimization result database. Subsequently, a conditional generative design model (e.g., cGAN) is constructed, encoding the design conditions as condition vectors as input. The generative design model is trained to learn the mapping relationship between "optimization conditions and topology results." When the convergence criterion of the training process meets preset requirements, the shear plate generative topology optimization model is output.

[0056] The process of generating shear plate structures, reconstructing geometry, evaluating performance, and outputting additive manufacturing based on a trained generative topology optimization model is as follows: Figure 2 As shown, the shear plate optimization conditions are input into the shear plate generative topology optimization model, and an image representation of the shear plate topology is output. The image representation is then converted to obtain a vector representation of the topological components, and a mesh file for finite element analysis is generated accordingly. Based on preset loads and boundary conditions, finite element analysis is performed on the structure, and energy dissipation performance indicators are calculated. When the energy dissipation performance does not meet the preset criteria, the topology configuration is regenerated and / or the rough edges are shaped and re-evaluated until the criteria are met. For target structures that meet the criteria, the additive manufacturing printing path is further planned, and manufacturing files (e.g., LS files containing printing path information) are generated to complete the forming and manufacturing of the shear plate components.

[0057] In this embodiment, the geometric reconstruction and meshing process includes: performing region growth segmentation on the binary pixel image of the candidate topology, extracting the boundary using an edge detection operator, and performing curve fitting on the boundary to obtain a closed vector contour, and then generating a finite element mesh based on the closed vector contour; the edge detection operator is the Sobel operator, and the curve fitting is the Bezier curve fitting.

[0058] Specifically, the optimization conditions are input into the trained generative model, and a high-resolution pixel map of the clipboard topology optimization configuration is output.

[0059] The pixel image is segmented using a region growing algorithm to identify pixel regions with similar colors and interconnectedness as the same pixel region. The Sobel edge detection operator is used to extract the boundaries between each pixel region, and the extracted boundary curves are fitted with Bezier curves to eliminate jagged pixel boundaries and form closed vector paths. The corresponding color information of the original pixel image is filled into the closed vector paths to generate a vector image file containing geometric attribute parameters.

[0060] Based on the vector image file, the topology optimization structure is discretized using mesh generation software to generate a mesh file for finite element analysis. In the finite element analysis environment, material parameters, load conditions, and boundary constraints are set for the mesh file. Shape optimization is performed on the rough edge region of the topology optimization result, with the objective function being to minimize the maximum equivalent stress of the structure under cyclic loading conditions.

[0061] After shape optimization, the node coordinate information of the optimized structure is exported. For the same layer of forming material, a genetic algorithm is used to plan the printing path to obtain the printing trajectory with the shortest path length. The genetic algorithm is a commonly used intelligent optimization algorithm in this field, and its encoding method, selection strategy, crossover method, mutation method, and termination condition can be set according to specific application scenarios. The fitness function of the genetic algorithm aims to minimize the total length of the printing path in the same layer, and constrains path continuity and consistency of start and end points.

[0062] In this embodiment, after obtaining the finite element mesh, the method further includes performing the above-described shape optimization process to obtain an optimized structure for the cyclic load finite element analysis.

[0063] In this embodiment, the , , At least one of them is determined using a relative benchmark threshold, which is determined based on the corresponding energy consumption evaluation index of the initial design shear plate.

[0064] In this embodiment, the The threshold is within a preset range, and when When the adjustment includes at least one of the following: reducing energy density, increasing the scanning interval, adjusting the scanning order to print regions with lower predicted temperatures first and then regions with higher predicted temperatures, or setting the interlayer dwell time; when When printing, the path is output according to the shortest path at the same level.

[0065] like Figure 3As shown, in this embodiment, the generator of the generative model is used to generate a high-resolution pixel image (e.g., a binary image or a grayscale density image) of the clipboard topology under conditional vector constraints. The generator includes a conditional encoder and a generative network: the conditional encoder performs multi-layer feature encoding on the conditional vector and fuses it with a random noise vector (random latent variable) before inputting it into the generative network; the generative network uses a transposed convolutional structure with progressive upsampling to map the fused features to a topological image of the target resolution, and preferably sets residual blocks and skip connections during the upsampling process to enhance feature transfer stability and suppress local artifacts; furthermore, the generative network preferably includes an attention enhancement module (e.g., an ECA module) to improve the ability to express key structural details. The generator output preferably uses a nonlinear activation function to constrain the output range in order to obtain a thresholdable topological representation. The number of network layers, channels, convolutional kernel parameters, and activation / normalization configuration can be adjusted according to the target resolution and computing resources.

[0066] like Figure 4 As shown, in this embodiment, the discriminator employs a multi-scale structure to simultaneously constrain the global structural rationality and local detail features of candidate topological configurations. The multi-scale discriminator includes multiple single-scale discriminators at different scales; each single-scale discriminator receives the topological image at its corresponding scale and its matching conditional information, and outputs a true / false discrimination result. The input images at different scales can be obtained by performing average pooling or downsampling on the original input image. The single-scale discriminator preferably consists of multiple convolutional units, each of which may include convolutional layers, normalization layers, and nonlinear activation layers, and outputs a discrimination probability or discrimination score at the end. Through joint optimization of the multi-scale discrimination results, the trained generator can generate topological configurations that simultaneously satisfy global connectivity constraints and local detail constraints.

[0067] This invention addresses the problems of high computational cost, long design cycle, and disconnect between structural performance constraints and manufacturing processes in existing shear plate topology optimization design methods. It proposes a shear plate design method based on the collaborative optimization of generative topology and additive manufacturing.

[0068] This invention uses the geometric dimensions, boundary connection parameters, cyclic load spectrum parameters, and volume upper limit constraints of the shear plate as design conditions. By constructing a condition-generative topology generation model, it achieves rapid generation of the shear plate topology configuration, avoiding the large number of iterative solutions in traditional density-based or evolutionary algorithm-based topology optimization, thus improving the efficiency of structural design.

[0069] In the topology generation stage, this invention introduces an energy dissipation performance evaluation mechanism based on cyclic load finite element analysis. It quantifies and calculates key indicators such as the hysteresis curve area, equivalent damping ratio, and stiffness degradation rate of candidate topology configurations, and combines them with a backtracking optimization strategy to form a closed-loop design process. This ensures that the generated structure meets the requirements of lightweighting while also taking into account good cyclic energy dissipation performance and mechanical stability.

[0070] Regarding manufacturing constraints, this invention further uses the thermal process prediction results in the additive manufacturing process as design constraints. By adaptively adjusting the printing path and process parameters, the topology design results and the forming manufacturing process are optimized in synergy, which effectively reduces the risk of residual stress and forming defects, and improves the consistency and reliability of the finished product structure.

[0071] This invention integrates generative topology generation, performance evaluation, shape optimization, and manufacturing process control to construct a holistic design methodology system for the cyclic service performance and manufacturing feasibility of shear plates, which is suitable for high-performance shear plate structure design scenarios involving mass production and rapid manufacturing.

[0072] Example 2

[0073] A clipboard design system based on generative topology, used to implement the clipboard design method based on generative topology of Embodiment 1, characterized in that it includes:

[0074] The design condition acquisition unit is used to acquire the design conditions of the shear plate based on generative topology. The design conditions include at least geometric dimension parameters, boundary connection parameters, cyclic load spectrum parameters, volume upper limit constraints, and additive manufacturing minimum feature size constraints.

[0075] A condition encoding unit is used to encode the design conditions into a condition vector;

[0076] A generative topology generation unit is used to input the condition vector into a generative model to generate at least one candidate topology configuration.

[0077] A geometric reconstruction and meshing unit is used to perform geometric reconstruction and meshing processing on the candidate topological configuration to obtain a structural model for finite element analysis; wherein the geometric reconstruction and meshing processing includes at least region segmentation, boundary extraction and curve fitting of the pixel representation of the candidate topological configuration to form a closed vector contour, and generating a finite element mesh based on the closed vector contour;

[0078] The simulation evaluation unit is used to perform cyclic load finite element analysis based on the structural model and calculate the energy consumption evaluation index set, which includes at least the hysteresis curve area A, the equivalent damping ratio ζe, and the secant stiffness degradation rate rd.

[0079] The fallback control unit is used to compare the set of energy consumption evaluation indicators with preset energy consumption criteria, wherein the preset energy consumption criteria include at least the following: , , When the comparison result does not meet the preset energy consumption criterion, the generative topology generation unit is controlled to regenerate the candidate topology configuration and / or the geometric reconstruction and meshing unit and the cyclic energy consumption simulation evaluation unit are controlled to re-evaluate the updated candidate topology configuration until the preset energy consumption criterion is met to determine the target shear plate structure.

[0080] The printing path construction unit is used to construct an additive manufacturing printing path based on the target shear plate structure and generate a manufacturing file.

[0081] The thermal process prediction unit is used to predict the thermal process of the printing process based on the printing path and preset process parameters, and to obtain the maximum temperature gradient Gmax of the printing process.

[0082] Adaptive adjustment unit, used to adjust With preset temperature gradient threshold Comparison: When If necessary, adjust at least one process parameter and / or path parameter, and recalculate. until ;when At that time, the output is a printing path and manufacturing file that meet the temperature gradient constraints, which are used to form and manufacture the target shear plate structure.

[0083] Example 3

[0084] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the clipboard design method based on generative topology as described in Embodiment 1.

[0085] Example 4

[0086] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the clipboard design method based on generative topology as described in Example 1.

[0087] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0089] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0090] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0091] Those skilled in the art will understand that all or part of the steps in the above facts and methods can be implemented by a program instructing related hardware. The program or the program described therein can be stored in a computer-readable storage medium. When the program is executed, it includes the following steps: at this time, the corresponding method steps are introduced. The storage medium can be ROM / RAM, magnetic disk, optical disk, etc.

[0092] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A clipboard design method based on generative topology, characterized in that, include: Obtain the design conditions for the shear plate based on generative topology. The design conditions include at least: geometric parameters, boundary connection parameters, cyclic load spectrum parameters, upper volume constraint, and minimum feature size constraint for additive manufacturing. Based on the design conditions, the design conditions are encoded into condition vectors and input into the generative model to generate at least one candidate topology configuration; Geometric reconstruction and meshing are performed on each candidate topology, followed by cyclic load finite element analysis to obtain a set of energy consumption evaluation indicators. This set of indicators includes at least the area under the hysteresis curve. Equivalent damping ratio and secant stiffness degradation rate ; The energy consumption evaluation index set is compared with a preset energy consumption criterion, the preset energy consumption criterion including at least the following: , , When the preset energy consumption criterion is not met, the topology configuration and / or energy consumption evaluation index set are regenerated until the preset energy consumption criterion is met, and the target shear plate structure is obtained. An additive manufacturing printing path is constructed based on the target shear plate structure, and the maximum temperature gradient of the printing process is obtained based on a thermal process prediction model. ; Will With preset temperature gradient threshold Comparison: When If necessary, adjust at least one process parameter and / or path parameter, and recalculate. until ;when At that time, the output is a printing path and manufacturing file that meet the temperature gradient constraints, which are used to form and manufacture the target shear plate structure.

2. The clipboard design method based on generative topology according to claim 1, characterized in that, The generative model is a conditional generative adversarial network (cGAN), and the conditional vector includes at least the geometric dimension parameter, the cyclic load spectrum parameter, and the volume upper limit constraint.

3. The clipboard design method based on generative topology according to claim 2, characterized in that, The discriminator of the cGAN is a multi-scale discriminator, which simultaneously constrains the global connectivity structure and local detail features of the candidate topology.

4. The clipboard design method based on generative topology according to claim 1, characterized in that, The geometric reconstruction and meshing process includes: performing region growth segmentation on the binary pixel image of the candidate topology, extracting the boundary using an edge detection operator, and performing curve fitting on the boundary to obtain a closed vector contour, and then generating a finite element mesh based on the closed vector contour; the edge detection operator is the Sobel operator, and the curve fitting is the Bezier curve fitting.

5. The clipboard design method based on generative topology according to claim 1, characterized in that, After obtaining the finite element mesh, the method further includes: using the maximum equivalent stress under cyclic loading. The closed vector profile is optimized with the goal of minimization to obtain the optimized structure for the cyclic load finite element analysis.

6. The clipboard design method based on generative topology according to claim 1, characterized in that, The , , At least one of them is determined using a relative benchmark threshold, which is determined based on the corresponding energy consumption evaluation index of the initial design shear plate.

7. The clipboard design method based on generative topology according to claim 1, characterized in that, The The threshold is within a preset range, and when When adjusting, the adjustment includes at least one of the following: reducing energy density, increasing scanning interval, adjusting the scanning order to print the area with lower predicted temperature first and then the area with higher predicted temperature, or setting interlayer dwell time; when When printing, the path is output according to the shortest path at the same level.

8. A clipboard design system based on generative topology, used to implement the clipboard design method based on generative topology as described in claims 1-7, characterized in that, include: The design condition acquisition unit is used to acquire the design conditions of the shear plate based on generative topology. The design conditions include at least geometric dimension parameters, boundary connection parameters, cyclic load spectrum parameters, volume upper limit constraints, and additive manufacturing minimum feature size constraints. A condition encoding unit is used to encode the design conditions into a condition vector; A generative topology generation unit is used to input the condition vector into a generative model to generate at least one candidate topology configuration. A geometric reconstruction and meshing unit is used to perform geometric reconstruction and meshing processing on the candidate topological configuration to obtain a structural model for finite element analysis; wherein the geometric reconstruction and meshing processing includes at least region segmentation, boundary extraction and curve fitting of the pixel representation of the candidate topological configuration to form a closed vector contour, and generating a finite element mesh based on the closed vector contour; The simulation evaluation unit is used to perform cyclic load finite element analysis based on the structural model and calculate the energy consumption evaluation index set, which includes at least the hysteresis curve area A, the equivalent damping ratio ζe, and the secant stiffness degradation rate rd. The fallback control unit is used to compare the set of energy consumption evaluation indicators with preset energy consumption criteria, wherein the preset energy consumption criteria include at least the following: , , When the comparison result does not meet the preset energy consumption criterion, the generative topology generation unit is controlled to regenerate the candidate topology configuration and / or the geometric reconstruction and meshing unit and the cyclic energy consumption simulation evaluation unit are controlled to re-evaluate the updated candidate topology configuration until the preset energy consumption criterion is met to determine the target shear plate structure. The printing path construction unit is used to construct an additive manufacturing printing path based on the target shear plate structure and generate a manufacturing file. The thermal process prediction unit is used to predict the thermal process of the printing process based on the printing path and preset process parameters, and to obtain the maximum temperature gradient Gmax of the printing process. Adaptive adjustment unit, used to adjust With preset temperature gradient threshold Comparison: When If necessary, adjust at least one process parameter and / or path parameter, and recalculate. until ;when At that time, the output is a printing path and manufacturing file that meet the temperature gradient constraints, which are used to form and manufacture the target shear plate structure.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the clipboard design method based on generative topology as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the clipboard design method based on generative topology as described in any one of claims 1 to 7.

Images