A generative design method for fundamental frequency of sheet metal parts filled with 3D printing lattice

Abstract

The present invention discloses a generative design method for the fundamental frequency of a sheet member filled with a 3D printing lattice, the method comprising: establishing a structure-equivalent stiffness database of a lattice unit cell, and a unit cell structure-equivalent stiffness inverse design model; simplifying the sheet entity into a shell, and setting boundary conditions for it; performing iterative calculations according to the target fundamental frequency of the sheet to obtain the target equivalent stiffness; obtaining the unit cell structure corresponding to the target equivalent stiffness, and filling it into the sheet; using dynamic finite element analysis simulation software to perform dynamic simulation calculations on the sheet filled with the lattice to obtain its fundamental frequency value; if the calculated fundamental frequency value is judged to be a design value, it indicates that the design of the sheet is completed, otherwise the above steps are continued after correcting and optimizing the structure-equivalent stiffness database, the unit cell structure-equivalent stiffness inverse design model or the target equivalent stiffness, wherein the design value is the fundamental frequency value corresponding to the target equivalent stiffness. The present invention realizes the rapid generative design of the fundamental frequency of sheet members.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing generative design, and in particular to a generative design method for the fundamental frequency of a 3D printed dot-matrix filled plate member. Background Art

[0002] The generative design of structural parts is to automatically generate the internal structure through artificial intelligence methods according to the user-specified contour shape, size, boundary conditions and performance requirements, and verify that the structure meets the application requirements through finite element simulation and experimental testing. The design process relies less on the designer's basic knowledge and design experience, and does not require the more cumbersome operation process in traditional design. It has fast generation speed, low cost and high accuracy. Therefore, generative design is an inevitable trend in the future development of structural design, and is a hot direction for artificial intelligence-enabled structural design, with important research and development value.

[0003] As a basic component, sheet metal is widely used in the industrial field. It not only plays an important role in basic livelihood industries such as construction, furniture, packaging, automobiles, and ships, but also plays a vital role in high-tech fields such as electronics, electrical appliances, and aerospace. High-performance electronic devices and precision instruments have increasingly higher requirements for the vibration reduction and heat transfer performance of sheet metal; high-speed moving tools such as spacecraft also have extremely high requirements for the lightweight, high strength, and high temperature resistance of sheet metal. Therefore, the directional improvement of the performance of sheet metal is an urgent need for industrial development.

[0004] As a porous structure, the lattice can achieve lightweight and performance optimization of sheet metal parts. At present, the lattice filling method of sheet metal parts is still based on the experience of the designer, and the appropriate cell filling is selected subjectively. The structural parts generated in this way are time-consuming, have many operation processes, and have large redundancy in weight and performance. The lattice sheet metal parts generated by intelligent algorithms not only significantly reduce the weight of the structure and enhance its mechanical properties, but also meet personalized design needs and improve material utilization and production efficiency. The present invention takes the fundamental frequency problem of vibration often encountered when sheet metal parts are used on sports tools as a representative of the performance of sheet metal parts, and proposes a generative design method for the fundamental frequency of sheet metal parts filled with lattices. Summary of the invention

[0005] The purpose of the present invention is to provide a generative design method for the fundamental frequency of a 3D printed dot-filled sheet material in view of the deficiencies in the prior art.

[0006] The object of the present invention is achieved through the following technical solution: A generative design method for the fundamental frequency of a sheet material filled with 3D printing lattice, comprising the following steps:

[0007] (1) Establish a structure-equivalent stiffness database of lattice unit cells and its corresponding unit cell structure-equivalent stiffness inverse design model;

[0008] (2) Simplify the plate entity into a shell and set the plate boundary conditions according to the service requirements of the plate;

[0009] (3) Perform iterative calculation based on the target fundamental frequency of the plate to obtain the target equivalent stiffness;

[0010] (4) According to the unit cell structure-equivalent stiffness inverse design model, the unit cell structure corresponding to the target equivalent stiffness is obtained, and it is filled into the plate according to the boundary conditions of the plate;

[0011] (5) Using dynamic finite element analysis simulation software, a dynamic simulation calculation is performed on the plate after the lattice filling obtained in step (4) to obtain its fundamental frequency value;

[0012] (6) Determine whether the fundamental frequency value obtained in step (5) is the design value. If the fundamental frequency value obtained in step (5) is the design value, the design of the plate is completed; otherwise, the structure-equivalent stiffness database, the unit cell structure-equivalent stiffness inverse design model or the target equivalent stiffness in step (3) in the optimization step (1) is corrected, and steps (4) to (6) are repeated; wherein the design value is the fundamental frequency value corresponding to the target equivalent stiffness in step (3).

[0013] Furthermore, the step (1) specifically includes the following sub-steps:

[0014] (1.1) Design 8 vertices, 12 edge points and 6 surface points of the lattice unit cell;

[0015] (1.2) Use the 8 vertices of the lattice unit cell to connect into a simple cubic structure;

[0016] (1.3) The 12 edge points and 6 surface points of the lattice unit cell are collected as the feature point set of the lattice unit cell, and the position matrix and connection relationship matrix of the feature point set are set;

[0017] (1.4) Constructing a unit cell structure dataset with different volume fractions based on the position matrix and connection relationship matrix of the feature point set;

[0018] (1.5) Use the Comsol simulation software to simulate the mechanical properties of each unit cell structure in the unit cell structure data set to obtain the equivalent stiffness matrix of each unit cell structure;

[0019] (1.6) Based on all unit cell structures and their corresponding equivalent stiffness matrices, a structure-equivalent stiffness database and a corresponding unit cell structure-equivalent stiffness inverse design model are constructed.

[0020] Furthermore, in the structure-equivalent stiffness database, the equivalent stiffness of the unit cell sample is isotropic; the equivalent stiffness value of the unit cell sample covers the range of 0MPa~1000Mpa.

[0021] Furthermore, in the structure-equivalent stiffness database, the porosity of the unit cell samples covers the range of 50% to 90%.

[0022] Furthermore, in the structure-equivalent stiffness database, the unit cell sample includes a unit cell structure and a corresponding equivalent stiffness matrix, and the unit cell structure includes two structural characteristic data: the unit cell side length and the volume fraction.

[0023] Furthermore, the unit cell structure-equivalent stiffness inverse design model specifically includes:

[0024] The equivalent stiffness is input into the unit cell structure-equivalent stiffness inverse design model, and the implicit coding vector is first obtained through the implicit vector predictor; the implicit vector predictor adopts the encoding part in the variational autoencoder network architecture and is implemented by a multi-layer perceptron; the input of the implicit vector predictor also includes the data labels of the unit cell side length and volume fraction;

[0025] The implicit coding vector output by the implicit vector predictor is input into the unit cell structure generation model to obtain multiple unit cell structures of different types and structural property error labels; wherein, the unit cell structure generation model includes an implicit diffusion model, and the implicit diffusion model adopts the architecture of the diffusion model.

[0026] Furthermore, the simplification of the plate entity into a shell is specifically achieved by adopting a layering method, in which the performance of the upper surface skin part, the lower surface skin part and the internal lattice part of the plate are assigned to the shell unit as independent layers.

[0027] Furthermore, the boundary conditions of the plate refer to constraints and preloads, and constraints and forces are set at specified positions to limit the filling area of ​​the unit cell structure in the plate.

[0028] Furthermore, the step (3) specifically includes the following sub-steps:

[0029] (3.1) Apply an initial value of equivalent stiffness to the plate, and use dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate corresponding to the equivalent stiffness;

[0030] (3.2) Analyze the fundamental frequency error between the fundamental frequency of the plate obtained by calculation and the target fundamental frequency of the plate. If the fundamental frequency error exceeds the allowable error range and the calculated fundamental frequency is less than the target fundamental frequency, increase the equivalent stiffness and use the dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate again; if the fundamental frequency error exceeds the allowable error range and the calculated fundamental frequency is greater than the target fundamental frequency, reduce the equivalent stiffness and use the dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate again;

[0031] (3.3) Analyze the fundamental frequency error between the calculated fundamental frequency of the plate and the target fundamental frequency of the plate again. If the fundamental frequency error decreases again, continue to increase the value of the equivalent stiffness; if the fundamental frequency error increases again, continue to decrease the value of the equivalent stiffness;

[0032] (3.4) Repeat steps (3.2) to (3.3) to change the value of the equivalent stiffness until the calculated fundamental frequency error is within the allowable error range, and determine the value of the equivalent stiffness at this time as the target equivalent stiffness.

[0033] Furthermore, the dynamic finite element analysis simulation software includes ansys software, abaqus software, and comsol software.

[0034] The beneficial effects of the present invention are as follows: the present invention reduces the number of finite elements and greatly speeds up the fundamental frequency iterative calculation by using shell unit equivalent plate entities; the present invention converts the lattice into equivalent solid units by calculating equivalent properties, reduces the amount of lattice performance calculation and speeds up the calculation; the present invention adopts a method for automatic iterative calculation of plate equivalent stiffness, reduces the calculation time, makes efficient generative design possible, and realizes the rapid generative design of the fundamental frequency of plate parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of the generative design method of the fundamental frequency of a sheet member filled with 3D printing lattice of the present invention;

[0036] Figure 2 It is a schematic diagram of modeling of a 1 / 8 unit cell structure of the present invention;

[0037] Figure 3 It is a schematic diagram of the structure of a complete unit cell after three mirror images of the present invention;

[0038] Figure 4 It is a flow chart of constructing the structure-equivalent stiffness database of the lattice unit cell of the present invention;

[0039] Figure 5 is a flow chart of determining the target equivalent stiffness of a plate according to the present invention;

[0040] Figure 6 It is the architecture flow chart of the unit cell structure-equivalent stiffness inverse design model of the present invention. DETAILED DESCRIPTION

[0041] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0042] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0043] It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0044] The present invention is described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the features of the following embodiments and implementations can be combined with each other.

[0045] See also Figure 1 The generative design method of the fundamental frequency of a plate member filled with 3D printing lattice of the present invention specifically comprises the following steps:

[0046] (1) Establish a lattice unit cell structure-equivalent stiffness database and its corresponding unit cell structure-equivalent stiffness inverse design model. The lattice unit cell refers to the unit cell of the lattice in the plate. Figure 4 As shown, it specifically includes the following sub-steps:

[0047] (1.1) Design 8 vertices, 12 edge points and 6 surface points of the lattice unit cell, such as Figure 2 shown.

[0048] (1.2) Use the eight vertices of the lattice unit cell to connect into a simple cubic structure, such as Figure 2 As shown, the eight vertices have a fixed connection method, which is connected to form a simple cubic structure. Figure 2 In the form of a cubic frame.

[0049] (1.3) The 12 edge points and 6 surface points of the lattice unit cell are taken as the feature point set of the lattice unit cell, and the position matrix and connection relationship matrix of the feature point set are set. The shape of the position matrix tensor is (26,3), which represents the position of the 26 points in the 1 / 8 cell in the three-dimensional coordinate system.

[0050] Furthermore, an edge point can move on the corresponding edge with one degree of freedom, and the minimum distance from the vertex is not less than 1 / 10 of the edge length and not less than 2 mm. A surface point can move on the corresponding surface with two degrees of freedom, and the minimum distance from the vertex and edge point is not less than 2 mm.

[0051] (1.4) A unit cell structure dataset with different volume fractions is constructed based on the position matrix and connection relationship matrix of the feature point set. An example of a unit cell structure is Figure 3 shown.

[0052] like Figure 2 As shown in , the connection between points should not produce isolated rods, that is, if a point on the surface is connected to other points, it is necessary to connect at least two points at the same time. Figure 2 When the 1 / 8 cell shown is mirrored three times, we get Figure 3 The complete unit cell shown has some vertices overlapped, some points on the edges overlapped, and some points on the faces overlapped, so that some edges also overlapped. After three mirrorings, the length of the horizontally placed rods in the cell should not exceed 6mm, otherwise it will be unfavorable for 3D printing.

[0053] (1.5) Comsol simulation software is used to simulate the mechanical properties of each unit cell structure in the unit cell structure data set to obtain the equivalent stiffness matrix of each unit cell structure. The tensor shape of the equivalent stiffness matrix is ​​(6,6) and is represented by Voigt notation. Voigt notation is mainly used to simplify tensor calculations, especially in continuous medium mechanics. It makes complex tensor operations more intuitive and easier to manage by matrixing tensors.

[0054] It should be understood that the Comsol simulation software is an existing commercial software developed based on the Linux system and automatically calls the code. It can perform high-throughput simulation of the mechanical properties of each unit cell structure. The equivalent stiffness matrix corresponding to each unit cell structure can be obtained through simulation.

[0055] (1.6) Based on all unit cell structures and their corresponding equivalent stiffness matrices, a structure-equivalent stiffness database and a corresponding unit cell structure-equivalent stiffness inverse design model are constructed.

[0056] Furthermore, the number of samples in the structure-equivalent stiffness database is over 100,000. In the structure-equivalent stiffness database, the equivalent stiffness of the unit cell sample is isotropic; the equivalent stiffness value of the unit cell sample covers the range of 0MPa~1000Mpa; the porosity of the unit cell sample covers the range of 50%~90%; the unit cell sample includes the unit cell structure and the corresponding equivalent stiffness matrix, and the unit cell structure contains two structural characteristic data: the unit cell edge length and volume fraction.

[0057] Furthermore, the unit cell structure-equivalent stiffness inverse design model is developed based on a deep learning framework, and its specific functions are: the equivalent stiffness value can be specified to generate a unit cell structure that meets the stiffness value requirements, that is, the equivalent stiffness value is input into the unit cell structure-equivalent stiffness inverse design model, and corresponding multiple unit cell structures of different types can be obtained.

[0058] like Figure 6 As shown, the equivalent stiffness is input into the unit cell structure-equivalent stiffness inverse design model, and firstly, an implicit coding vector is obtained through an implicit vector predictor, wherein the implicit vector predictor adopts the coding part in the variational autoencoder (VAE) network architecture, which is implemented by a multi-layer perceptron, and the input of the implicit vector predictor also includes data labels of the unit cell side length and volume fraction; the implicit coding vector output by the implicit vector predictor is input into the unit cell structure generation model to obtain multiple unit cell structures of different types and structural property error labels, wherein the unit cell structure generation model includes an implicit diffusion model, and the implicit diffusion model adopts the architecture of the diffusion model.

[0059] It should be understood that the VAE network is a variational autoencoder, which consists of an encoder and a decoder. In this embodiment, the encoder part is used to construct an implicit vector predictor. The diffusion model is an unsupervised generative model used for probability density function modeling.

[0060] (2) Simplify the plate entity into a shell, and set the plate boundary conditions according to the service requirements of the plate. The shell includes the top structure, the lattice area of ​​the middle layer, and the bottom structure. After the unit cell structure is determined, the determined unit cell structure can be directly filled into the lattice area of ​​the middle layer according to the boundary conditions of the plate. The boundary conditions of the plate refer to the constraints and preloads. Constraints and forces are set at specified locations to limit the filling area of ​​the unit cell structure in the plate.

[0061] Furthermore, the plate entity is simplified into a shell, which is specifically achieved by adopting a layering method, and the performance of the upper surface skin part, the lower surface skin part and the internal lattice part of the plate are assigned to the shell unit as independent layers.

[0062] It should be understood that the layering method is a conventional method for shelling a solid body.

[0063] (3) Perform iterative calculation based on the target fundamental frequency of the plate to obtain the target equivalent stiffness. Figure 5 As shown, it specifically includes the following sub-steps:

[0064] (3.1) Apply an initial value of equivalent stiffness to the plate, and use dynamic finite element analysis simulation software such as ANSYS, ABAQUS, COMSO, etc. to calculate the fundamental frequency of the plate corresponding to the equivalent stiffness.

[0065] It should be understood that when calculating the fundamental frequency of the plate based on the value of the equivalent stiffness, existing commercial simulation software can be used as the calculation software for equivalent properties, that is, existing dynamic finite element analysis simulation software, such as ANSYS software, ABAQUS software, COMSOL software, etc.

[0066] (3.2) Analyze the fundamental frequency error between the calculated fundamental frequency of the plate and the target fundamental frequency of the plate. If the fundamental frequency error exceeds the allowable error range and the calculated fundamental frequency is smaller than the target fundamental frequency, increase the equivalent stiffness and use the dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate again. If the fundamental frequency error exceeds the allowable error range and the calculated fundamental frequency is larger than the target fundamental frequency, reduce the equivalent stiffness and use the dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate again.

[0067] It should be noted that the target fundamental frequency is related to the constraints in the boundary conditions, and different constraints produce different plate fundamental frequencies.

[0068] (3.3) The fundamental frequency error between the calculated fundamental frequency of the plate and the target fundamental frequency of the plate is analyzed again. If the fundamental frequency error decreases again, the value of the equivalent stiffness continues to increase; if the fundamental frequency error increases again, the value of the equivalent stiffness continues to decrease.

[0069] (3.4) Repeat steps (3.2) to (3.3) to change the value of the equivalent stiffness until the calculated fundamental frequency error is within the allowable error range, and determine the value of the equivalent stiffness at this time as the target equivalent stiffness.

[0070] It should be noted that, through the above iterative calculation process, the automatic iterative calculation of the equivalent stiffness of the plate with a specified fundamental frequency can be realized, and the single iteration time is in the order of seconds. In the iterative calculation process of step (3), it is necessary to consider the size effect of the lattice filling, and set the conversion coefficient of the equivalent stiffness of the plate filled with a single cell. The conversion coefficient is added here because the calculation of the equivalent stiffness of the single cell structure assumes that there are countless single cells connected around the single cell to form an infinitely large lattice, and then the equivalent stiffness of the single cell is calculated by calculating the equivalent stiffness of the lattice. But in fact, in the actual plate, the number of filled single cells may be very limited, which theoretically does not meet the prerequisite for the calculation of the equivalent stiffness matrix mentioned above, so the equivalent stiffness calculated by the commercial software must be multiplied by the conversion coefficient.

[0071] (4) According to the unit cell structure-equivalent stiffness inverse design model, the unit cell structure corresponding to the target equivalent stiffness is obtained, and it is filled into the plate according to the boundary conditions of the plate.

[0072] It should be understood that since the plate entity is simplified into a shell in step (2), the simplified shell is a multi-layer structure, and here the determined unit cell structure can be directly filled into the middle layer of the shell. In addition, the unit cell structure-equivalent stiffness inverse design model can generate multiple different types of unit cell structures according to a specified equivalent stiffness. Therefore, the subsequent steps require dynamic simulation of different types of unit cell structures for further judgment.

[0073] (5) Use dynamic finite element analysis simulation software such as Comsol software to perform dynamic simulation calculations on the plate after filling the lattice obtained in step (4) to obtain its fundamental frequency value.

[0074] It should be noted that in the current step (5), the dynamic simulation calculation of the plate after filling the lattice is performed in the form of a solid model rather than a simplified shell form.

[0075] (6) Determine whether the fundamental frequency value obtained in step (5) is the design value. If the fundamental frequency value obtained in step (5) is the design value, the design of the plate is completed; otherwise, modify the structure-equivalent stiffness database in step (1), the unit cell structure-equivalent stiffness inverse design model or the target equivalent stiffness in step (3), and repeat steps (4) to (6). The design value is the fundamental frequency value corresponding to the target equivalent stiffness in step (3).

[0076] In summary, the present invention reduces the number of finite elements and greatly speeds up the iterative calculation of the fundamental frequency by using the equivalent plate entity of the shell unit; the present invention converts the lattice into an equivalent solid unit by calculating the equivalent properties, reduces the amount of lattice performance calculation and speeds up the calculation; the present invention adopts a method for automatic iterative calculation of the equivalent stiffness of the plate, reduces the calculation time, makes efficient generative design possible, and realizes the rapid generative design of the fundamental frequency of the plate part.

[0077] The above is only a preferred implementation case of the present invention and does not limit the present invention in any form. Although the implementation process of the present invention is described in detail above, for those familiar with the art, they can still modify the technical solutions recorded in the above examples, or replace some of the technical features therein with equivalents. All modifications, equivalent replacements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A generative design method for fundamental frequency of a sheet material filled with 3D printing lattice, characterized in that: The following steps are involved: (1) Establish a structure-equivalent stiffness database of lattice unit cells and its corresponding unit cell structure-equivalent stiffness inverse design model; (2) Simplify the plate entity into a shell and set the plate boundary conditions according to the service requirements of the plate; (3) performing iterative calculation according to the target fundamental frequency of the plate to obtain the target equivalent stiffness; the step (3) specifically includes the following sub-steps: (3.1) Apply an initial value of equivalent stiffness to the plate, and use dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate corresponding to the equivalent stiffness; (3.2) Analyze the fundamental frequency error between the fundamental frequency of the plate obtained by calculation and the target fundamental frequency of the plate. If the fundamental frequency error exceeds the allowable error range and the calculated fundamental frequency is less than the target fundamental frequency, increase the equivalent stiffness and use the dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate again; if the fundamental frequency error exceeds the allowable error range and the calculated fundamental frequency is greater than the target fundamental frequency, reduce the equivalent stiffness and use the dynamic finite element analysis simulation software to calculate the fundamental frequency of the plate again; (3.3) Analyze the fundamental frequency error between the calculated fundamental frequency of the plate and the target fundamental frequency of the plate again. If the fundamental frequency error decreases again, continue to increase the value of the equivalent stiffness; if the fundamental frequency error increases again, continue to decrease the value of the equivalent stiffness; (3.4) Repeat steps (3.2) to (3.3) to change the value of the equivalent stiffness until the calculated fundamental frequency error is within the allowable error range, and determine the value of the equivalent stiffness at this time as the target equivalent stiffness; (4) According to the unit cell structure-equivalent stiffness inverse design model, the unit cell structure corresponding to the target equivalent stiffness is obtained, and it is filled into the plate according to the boundary conditions of the plate; (5) using dynamic finite element analysis simulation software to perform dynamic simulation calculation on the plate after filling the lattice obtained in step (4) to obtain its fundamental frequency value; (6) Determine whether the fundamental frequency value obtained in step (5) is the design value. If the fundamental frequency value obtained in step (5) is the design value, the design of the plate is completed; otherwise, the structure-equivalent stiffness database, the unit cell structure-equivalent stiffness inverse design model or the target equivalent stiffness in step (3) in the optimization step (1) is corrected, and steps (4) to (6) are repeated; wherein the design value is the fundamental frequency value corresponding to the target equivalent stiffness in step (3).

2. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1, characterized in that: The step (1) specifically includes the following sub-steps: (1.1) Design the 8 vertices, 12 edge points and 6 surface points of the lattice unit cell; (1.2) Use the 8 vertices of the lattice unit cell to connect into a simple cubic structure; (1.3) The 12 edge points and 6 surface points of the lattice unit cell are collected as the feature point set of the lattice unit cell, and the position matrix and connection relationship matrix of the feature point set are set; (1.4) Constructing a unit cell structure data set with different volume fractions based on the position matrix and connection relationship matrix of the feature point set; (1.5) Use the Comsol simulation software to simulate the mechanical properties of each unit cell structure in the unit cell structure data set to obtain the equivalent stiffness matrix of each unit cell structure; (1.6) Based on all unit cell structures and their corresponding equivalent stiffness matrices, a structure-equivalent stiffness database and a corresponding unit cell structure-equivalent stiffness inverse design model are constructed.

3. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1 or 2, characterized in that: In the structure-equivalent stiffness database, the equivalent stiffness of the unit cell sample is isotropic; the equivalent stiffness value of the unit cell sample covers the range of 0MPa to 1000Mpa.

4. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1 or 2, characterized in that: In the structure-equivalent stiffness database, the porosity of the unit cell samples ranges from 50% to 90%.

5. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1 or 2, characterized in that: In the structure-equivalent stiffness database, the unit cell sample includes a unit cell structure and a corresponding equivalent stiffness matrix, and the unit cell structure includes two structural characteristic data: the unit cell side length and the volume fraction.

6. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1 or 2, characterized in that: The unit cell structure-equivalent stiffness inverse design model specifically includes: The equivalent stiffness is input into the unit cell structure-equivalent stiffness inverse design model, and the implicit coding vector is first obtained through the implicit vector predictor; the implicit vector predictor adopts the encoding part in the variational autoencoder network architecture and is implemented by a multi-layer perceptron; the input of the implicit vector predictor also includes the data labels of the unit cell side length and volume fraction; The implicit coding vector output by the implicit vector predictor is input into the unit cell structure generation model to obtain multiple unit cell structures of different types and structural property error labels; wherein, the unit cell structure generation model includes an implicit diffusion model, and the implicit diffusion model adopts the architecture of the diffusion model.

7. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1, characterized in that: The simplification of the plate entity into a shell is specifically achieved by adopting a layering method, in which the performance of the upper surface skin part, the lower surface skin part and the internal lattice part of the plate are assigned to the shell unit as independent layers.

8. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1, characterized in that: The boundary conditions of the plate refer to constraints and preloads, and constraints and forces are set at specified locations to limit the filling area of ​​the unit cell structure in the plate.

9. The generative design method for fundamental frequency of a sheet material filled with 3D printing lattice according to claim 1, characterized in that: The dynamic finite element analysis simulation software includes ansys software, abaqus software and comsol software.

Images