A method for optimizing the design of heterogeneous lattices with customized mechanical responses

Abstract

The present invention discloses a method for optimizing the design of a heterogeneous lattice with customized mechanical response, which belongs to the field of heterogeneous lattice modeling and simulation. Aiming at the correlation between heterogeneous porous materials, the present invention adopts a design method of a heterogeneous lattice, designs different cells according to the combination rule, and greatly enriches the design space of the lattice configuration; adopts a machine learning model to quickly and accurately obtain the relationship between the lattice configuration and its mechanical response; through the mechanical responses of all configurations predicted by the machine learning model, by constructing a database of configuration responses, it is possible to realize the reverse design of heterogeneous lattice materials according to the target response, significantly improve the optimization design efficiency, and meet the engineering requirements for different mechanical properties of porous materials. The present invention is applicable to the fields of construction, automobiles, aerospace, etc., and porous materials with specific target responses are obtained through the optimization design of heterogeneous lattices.

Description

Technical Field

[0001] The present invention relates to an optimization design method for a lattice material microstructure, and in particular to an optimization design method for a heterogeneous lattice with customized mechanical response, and belongs to the field of modeling and simulation. Background Art

[0002] The mechanical properties of porous materials are closely related to their microstructures, so the mechanical properties of porous materials can be regulated by adjusting their microstructures. Current research mainly focuses on studying the mechanical properties of porous materials with a periodic single configuration. Due to the singleness of their microstructures, the mechanical response of porous materials is single. Porous materials with a single configuration are difficult to meet the complex needs of practical engineering applications. The usual practice is to design corresponding microstructures to meet engineering needs. Studies have shown that by designing heterogeneous porous materials, that is, destroying the periodicity of porous materials, the adjustable range of their mechanical properties is significantly increased. Therefore, porous materials with heterogeneous microstructures provide unlimited possibilities in terms of failure modes, deformation energy absorption, etc.

[0003] The molding quality of 3D printed lattice materials is affected by different factors, such as printing process, printing direction, etc. The widely used metal 3D printing technologies currently include selective laser melting and selective electron beam melting. The surface quality of the specimens obtained by selective laser melting technology is good, but there are pores inside. The specimens printed by selective electron beam melting technology have almost no pores inside, but the surface quality is poor. Therefore, the specimens obtained by 3D printing inevitably have geometric defects. Studies have shown that the existence of geometric defects will cause the actual mechanical properties of the lattice material to deviate from its design value. Considering the mechanical influence of geometric defects on 3D printed lattice materials is of great significance for practical applications.

[0004] Although some studies have explored the mechanical properties of porous materials with heterogeneous microstructures, the research ideas are similar to the previous ones, that is, designing a porous material with heterogeneous microstructures and using quasi-static experiments or numerical simulations to study its mechanical properties. However, currently only a single porous material is studied independently, ignoring the correlation between heterogeneous porous materials, and no reverse design is performed for the target response, which is greatly limited in actual engineering applications and cannot meet the actual needs of engineering. Summary of the invention

[0005] In view of the correlation between heterogeneous porous materials, the main purpose of the present invention is to provide a heterogeneous lattice optimization design method with customized mechanical response. Under the premise of introducing heterogeneous microstructures, the corresponding porous materials are reversely designed according to the target response to meet the engineering requirements for different mechanical properties of porous materials.

[0006] The objective of the present invention is achieved through the following technical solutions:

[0007] The present invention discloses a method for optimizing the design of heterogeneous lattices with customized mechanical responses. By selecting lattice materials with different mechanical properties, they are combined to obtain a series of lattice materials with heterogeneous microstructures. Artificial intelligence is combined to obtain a substitute model that can predict lattice materials, and finally the corresponding lattice microstructure is obtained according to the target with the help of the substitute model. The corresponding porous material is reversely designed according to the target response to meet the engineering requirements for different mechanical properties of porous materials.

[0008] The present invention discloses a method for optimizing the design of a heterogeneous lattice with customized mechanical response, which specifically includes the following steps:

[0009] Step 1: Select two lattice materials with different mechanical properties and obtain their basic unit cells;

[0010] Step 2: Obtain a series of lattice materials with heterogeneous structures according to the combination rules;

[0011] a) Based on a basic lattice material configuration, different numbers of another configuration are selected to replace it, thereby obtaining a heterogeneous lattice. This heterogeneous lattice design method is also applicable to the combination of multiple cell types.

[0012] b) Determine the combination method and calculate the number of possible configurations according to the combination formula. This method mainly considers the combination method within a two-dimensional surface.

[0013] Step 3, numerically simulating the mechanical response of the heterogeneous lattices with different configurations obtained in step 2;

[0014] a) Establishing the geometric model of heterogeneous lattice;

[0015] b) Meshing the geometric model;

[0016] c) Create a material model for the base material, build a cross-section model, and assign it to the model;

[0017] d) Since large deformation simulation is a strongly nonlinear problem, the explicit solver of the finite element software is used to solve it. Establish the explicit solution analysis step;

[0018] e) Create contact properties in the calculation model;

[0019] f) Apply quasi-static loads and set boundary conditions;

[0020] g) Submit the calculation.

[0021] Step 4: Obtain a machine learning model for predicting lattice materials;

[0022] a) Based on the numerical simulation results in step 3, a database consisting of configurations and stress-strain curves is established.

[0023] b) Select a suitable machine learning model and build the corresponding model.

[0024] c) Match the configuration information and the corresponding mechanical response one by one and prepare a database for training the model.

[0025] d) Divide the database into training set, validation set and test set, and use the training set and validation set to train and evaluate the machine learning model.

[0026] Step 5: Obtain the configuration of the heterogeneous lattice according to the target response to meet the target requirements of the project for the performance of porous materials.

[0027] a) Based on the machine model obtained in step 4, predict the mechanical responses of all possible configurations, obtain the responses of all configurations, and establish a new database.

[0028] b) Based on the database obtained in a), obtain the upper and lower limits of the mechanical response.

[0029] c) Based on the database obtained in a), obtain the possible laws of mechanical response.

[0030] d) Design target responses based on the possible rules obtained in c).

[0031] e) For the target response, search the database for the response with the smallest relative error with the design target, and return its configuration information to meet the engineering target requirements for the performance of porous materials.

[0032] Preferably, the method further includes step six: obtaining a heterogeneous lattice configuration according to the target response obtained in step five, and applying it to the automotive and aerospace fields to significantly improve the deformation energy absorption performance of porous materials, thereby improving the protective heat insulation capabilities and lifespan of automotive and aerospace.

[0033] Beneficial results:

[0034] 1. The present invention discloses a method for optimizing the design of a heterogeneous lattice with customized mechanical response. The method adopts a design method for a heterogeneous lattice, and different cells are designed according to a combination rule, which greatly enriches the design space of the lattice configuration.

[0035] 2. The present invention discloses a method for optimizing the design of heterogeneous lattices with customized mechanical responses, which uses a machine learning model to quickly and accurately obtain the relationship between the lattice configuration and its mechanical response.

[0036] 3. The present invention discloses a method for optimizing the design of heterogeneous lattices with customized mechanical responses. By predicting the mechanical responses of all configurations through a machine learning model and building a database of configuration responses, it is possible to achieve reverse design of heterogeneous lattice materials according to target responses, thereby significantly improving the efficiency of optimization design. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 There are two unit cells with different deformation modes;

[0038] Figure (a) shows an octagonal lattice, and Figure (b) shows a rhombic dodecahedron;

[0039] Figure 2 is a schematic diagram of the finite element model;

[0040] Figure 3 It is a neural network structure diagram;

[0041] Figure 4 This is a comparison chart between the numerical simulation prediction results and the experimental results;

[0042] Figure 5 This is a comparison chart between the finite element prediction results and the artificial neural network prediction results;

[0043] Figure 6 The following are the prediction results of four typical curves;

[0044] Figure 7 This is a flow chart of a method for optimizing the design of a heterogeneous lattice with customized mechanical response disclosed in the present invention. DETAILED DESCRIPTION

[0045] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. The technical problems solved by the technical solution of the present invention and the beneficial effects are also described. It should be noted that the described embodiments are only intended to facilitate the understanding of the present invention and do not have any limiting effect on it.

[0046] In the automotive and aerospace fields, there is an urgent need for porous materials with specific target responses, but the design space of traditional design solutions is very limited and cannot meet this demand. This embodiment is in accordance with a heterogeneous lattice optimization design method with customized mechanical response disclosed by the present invention. According to the demand targets with different characteristics, the configuration of the lattice material is designed in a targeted manner so that the mechanical response of the designed lattice material can meet the target demand. For example, in the automotive field, when a person collides with a car, in order to achieve the purpose of protecting pedestrians, a lattice material with two stages of platform stress can be designed, which can achieve energy absorption in a longer energy absorption stroke. At the same time, since the amplitude of the first stage of platform stress is small, the purpose of protecting pedestrians can be fully achieved. In addition, when a car collides with a guardrail, a high and stable energy absorption platform with a high stress amplitude is required. At this time, the configuration of the lattice material can be designed so that the lattice material exhibits a stable high stress platform segment to achieve the purpose of absorbing a large amount of energy. In order to be able to cope with both working conditions at the same time, a stress-strain curve with a continuous linear strengthening section can be designed, so that when the strain is small, the stress amplitude of the lattice material is low, and as the deformation continues to increase, the stress continues to increase, achieving the purpose of absorbing a large amount of energy. In response to the above different needs, the design method is combined to design a variety of lattice material configurations with a stable platform section, two different platform sections, and a linear strengthening platform section in the stress-strain curve. Perform lattice design of porous materials.

[0047] The specific steps are as follows:

[0048] Laser stereolithography technology is used with resin as the base material to print a lattice material with a heterogeneous microstructure composed of a combination of rhombic dodecahedron and octagonal lattice.

[0049] like Figure 1 As shown, the cells of the above two lattice materials are combined to design and print four lattice materials, two of which are rhombic dodecahedron and octagonal lattice materials with a single structure, and the other two are designed heterogeneous lattice materials, referred to as heterogeneous 1 and heterogeneous 2.

[0050] The possible configurations used to solve the problem are as follows: The total possible combination configurations are 65,536. It can be seen here that a simple combination of two cell elements can greatly expand the design space of lattice materials.

[0051] For the microstructure of the printed heterogeneous lattice, the commercial finite element software ABAQUS is used to establish its geometric model, such as Figure 2As shown. Considering the balance between numerical simulation solution accuracy and time, beam elements are used for modeling and solving. The rods between any two nodes are discretized into 7 beam elements. The isotropic hardening model is used to characterize the mechanical behavior of the matrix material (tough2000). The general contact model is used to consider the friction between the rods and the friction between the heterogeneous lattice and the indenter. Its contact property is normal hard contact, and the penalty function method is used to consider the friction in the tangential direction. Constrain all degrees of freedom of the lower indenter and set their values ​​to 0. Except for the translational degrees of freedom in the loading direction, all other degrees of freedom of the upper indenter are set to 0. Set a constant compression velocity for the upper indenter to simulate the quasi-static load in the experiment. Create a solution task and submit the calculation.

[0052] The predicted results of stress-strain curves of four printed heterogeneous lattice configurations are compared with the experimental results. Figure 4 shown.

[0053] In addition to the printed configurations, a certain number of configurations are selected from all possible configurations for corresponding numerical simulations.

[0054] The solution results of all the above finite element models are processed to obtain their stress-strain curves. The configuration of the heterogeneous lattice is bound to the corresponding stress-strain curve results to form a configuration-response database. The database is divided into a training set, a validation set, and a test set.

[0055] This embodiment uses an artificial neural network as a machine learning model. The neural network has five hidden layers, one input layer, and one output layer. The model of the neural network is as follows Figure 3 As shown. The proposed neural network is trained with the training set and validation set in the above database. Then the trained neural network is evaluated. In this embodiment, the stress-strain prediction results of a certain configuration are compared with the prediction results of the neural network using the commercial finite element software ABAQUS to determine the prediction effect and generalization ability of the neural network. Figure 5 As shown in the figure, it can be seen that the trained neural network can accurately reproduce the prediction results of the finite element method, and thus can replace the finite element model to quickly and accurately predict the mechanical behavior of heterogeneous lattice configurations.

[0056] Use the trained artificial neural network to predict the mechanical response of all configurations, and establish a configuration-response database of the configuration predicted by the artificial neural network. According to the established database, the upper and lower limits of the stress-strain curves that can be obtained for all configurations are obtained. On the basis of further understanding of the database, a reasonable target response is designed. A reasonable target response includes the stress amplitude of a single point and the change trend of the entire stress-strain curve. For the proposed reasonable target response, the mean square error between the stress-strain curve in the database and the target response is calculated, and the configuration with the smallest error is returned as the recommended configuration.

[0057] In order to verify the effectiveness of the alternative configurations provided, the mechanical response of the configurations obtained from the database was verified using quasi-static compression experiments and finite element simulations, and the results are shown in Figure 6. In order to achieve different demand goals in engineering practice, the characteristics of the stress-strain curve of the lattice material are designed here, such as having a platform, having two platforms of different lengths, and having a linear strengthening section. The results show that the configuration of the heterogeneous lattice can be customized for different target responses.

[0058] The results of the above embodiments show that the heterogeneous lattice design method with customized mechanical response can be used to design the configuration of the lattice material in a targeted manner according to different demand targets, and the mechanical properties of the designed lattice material can meet the target requirements. This method can achieve the precise design of heterogeneous lattice microstructures with specific target responses.

[0059] The specific description above further illustrates the purpose, technical solutions and beneficial effects of the invention in detail. 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 in the scope of protection of the present invention.

Claims

1. A method for optimizing the design of heterogeneous lattices with customized mechanical responses, characterized in that: The following steps are included: Step 1: Select two lattice materials with different mechanical properties and obtain their basic unit cells; Step 2: Obtain a series of lattice materials with heterogeneous structures according to the combination rules; Step 3, numerically simulating the mechanical response of the heterogeneous lattices with different configurations obtained in step 2; Step 4: Obtain a machine learning model for predicting lattice materials; a) establishing a database composed of configurations and stress-strain curves according to the numerical simulation results in step 3; b) selecting a suitable machine learning model and establishing a corresponding model; the suitable machine learning model is an artificial neural network machine learning model, the neural network has five hidden layers, an input layer, and an output layer; from the first hidden layer to the fifth hidden layer, the number of neurons in the five hidden layers is 500, 400, 300, 200 and 100, respectively, and the configuration information and the corresponding mechanical response are matched one by one to prepare a database for training the model; c) Divide the database into training set, validation set and test set, and use the training set and validation set to train and evaluate the machine learning model; Step 5: Obtain the configuration of the heterogeneous lattice according to the target response to meet the target requirements of the engineering on the performance of the porous material; The implementation method of step five is: a) predicting the mechanical responses of all possible configurations based on the machine model obtained in step 4, obtaining the responses of all configurations, and establishing a new database; b) obtaining the upper and lower limits of the mechanical response according to the database obtained in a); c) obtaining possible laws of mechanical response based on the database obtained in a); d) Design target responses based on the possible rules obtained in c); e) For the target response, search the database for the response with the smallest relative error with the design target, and return its configuration information to meet the engineering target requirements for the performance of porous materials.

2. The method for optimizing the design of a heterogeneous lattice with customized mechanical response according to claim 1, characterized in that: It also includes step six: obtaining the configuration of the heterogeneous lattice according to the target response obtained in step five, and applying it to the automotive and aerospace fields to significantly improve the deformation energy absorption performance of the porous material, thereby improving the protective heat insulation capabilities and lifespan of the automotive and aerospace industries.

3. A method for optimizing the design of a heterogeneous lattice with customized mechanical response according to claim 1 or 2, characterized in that: The implementation method of step 2 is: a) On the basis of a basic lattice material configuration, different numbers of another configuration are selected to replace it, thereby obtaining a heterogeneous lattice; the heterogeneous lattice design method is also applicable to the combination of multiple cell types; b) Determine the combination method and calculate the number of possible configurations according to the combination formula. This method mainly considers the combination method within a two-dimensional surface.

4. The method for optimizing the design of a heterogeneous lattice with customized mechanical response according to claim 3, characterized in that: The implementation method of step three is: a) Establishing the geometric model of heterogeneous lattice; b) Meshing the geometric model; c) Create a material model for the base material, build a cross-section model, and assign it to the model; d) Since large deformation simulation is a strongly nonlinear problem, the explicit solver of the finite element software is used to solve it; an explicit solution analysis step is established; e) Create contact properties in the calculation model; f) Apply quasi-static loads and set boundary conditions; g) Submit the calculation.

Images