Design method of impact-resistant part and establishment method and system of design model of impact-resistant part

Abstract

The invention relates to a design method of an anti-impact part and an establishment method and system of a design model of the anti-impact part, and belongs to the technical field of part design. The establishing method comprises the following steps: obtaining a sub-model of a unit cell structure parameter to a lattice structure stress-strain curve based on stress-strain curves under different lattice structures of an anti-impact part and a deep learning model; and on the basis of the relationship between the stress-strain curves of the anti-impact part under different lattice structures and the absorbed energy of the corresponding anti-impact part and the deep learning model, obtaining a sub-model of the lattice structure stress-strain curves for resisting the absorbed energy of the anti-impact part. In the process of designing the anti-impact part, the relation between the anti-impact performance of the anti-impact part and the structural parameters can be considered, and the relation between the anti-impact performance of the anti-impact part and the absorbed energy can also be considered, so that the finally obtained anti-impact part has high impact resistance under the condition of absorbing impact energy of a specific numerical value. And the light weight of the anti-impact part is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of parts design, and in particular relates to a design method for impact-resistant parts and a method and system for establishing a design model thereof. Background Art

[0002] As a critical impact-resistant component of aircraft engines, the turbine case is designed to absorb the impact energy of broken blades through its structural design, containing the fragments within the case to protect the aircraft and passengers. In other words, the turbine case must be able to contain a broken blade during engine operation. Furthermore, as a key performance parameter of aircraft engines, modern aircraft engine design focuses on finding lightweight materials to reduce the overall weight of the engine and thus improve the thrust-to-weight ratio.

[0003] Traditional turbine casings are usually solid structures, and materials usually use metal and composite materials. For turbine casings, the high temperature characteristics of their working environment put forward special requirements for turbine casing materials. Traditional turbine casings are usually designed and processed with GH4169 materials. However, due to the high density of metal materials, traditional turbine casings are heavy, usually accounting for 10%-30% of the overall weight of the turbine.

[0004] Most existing turbine containment cases are monolithic structures. To account for the ductility of the turbine case and the high-temperature characteristics of its operating environment, high-temperature alloys are often used as processing materials. However, high-temperature alloys have a higher density and are heavier than other high-temperature resistant materials. Furthermore, high-temperature alloys have a poor ability to absorb the kinetic energy of rotor fragments. To achieve the required containment requirements, the turbine containment case typically needs to be designed to be larger, which further increases its weight, making the engine heavier. Furthermore, existing turbine containment cases rely on manual iterative calculations to determine their final geometry and dimensions. The manufacturing process relies on traditional forging and machining methods, which is time-consuming and labor-intensive.

[0005] When applying lattice materials to aircraft engines, the primary consideration is their containment capacity. Due to the extremely high rotational speeds of aircraft engine rotors, experimental investigations of lattice materials under such high-speed impacts with ruptured rotors are extremely challenging. Secondly, once impact resistance has been verified, the more complex issues arise in selecting the lattice structure model and designing the overall turbine casing structure to achieve optimal weight reduction and reinforcement. Applying additively manufactured lattice sandwich structures to aircraft engine turbine casings primarily presents these two challenges. The mechanical properties of a lattice structure are primarily determined by its internal design, and varying structural parameters can significantly alter the performance of the resulting lattice. Therefore, given a given impact energy, designing the lattice structure's unit cell form to minimize overall weight while fully absorbing the energy of the ruptured rotor impact is both the key and the challenge. Summary of the Invention

[0006] In order to solve the above problems, the present invention provides a design method for impact-resistant parts and a method and system for establishing a design model thereof.

[0007] A first object of the present invention is to provide a method for establishing a design model for an impact-resistant part, wherein the design model includes a sub-model of unit cell structure parameters versus lattice structure stress-strain curve and a sub-model of the lattice structure stress-strain curve versus energy absorption of the impact-resistant part;

[0008] The establishment method comprises:

[0009] Based on the stress-strain curves of impact-resistant parts under different lattice structures and the deep learning model, a sub-model of the stress-strain curve of the lattice structure with respect to the unit cell structure parameters is obtained;

[0010] Based on the relationship between the stress-strain curves of different lattice structures of impact-resistant parts and the absorbed energy of the corresponding impact-resistant parts, as well as the deep learning model, a sub-model of the lattice structure stress-strain curve for the absorbed energy of the impact-resistant parts is obtained.

[0011] In a specific embodiment of the present invention, obtaining stress-strain curves of the impact-resistant component under different lattice structures includes:

[0012] Construct the first unit cell size lattice structure model of impact-resistant parts;

[0013] Based on the compression test of the solid parts and the compression numerical simulation test of the first unit cell size lattice structure model, the numerical simulation model of the impact-resistant parts is modified;

[0014] According to the numerical simulation test of the modified numerical simulation model of the impact-resistant part, the stress-strain curves of the impact-resistant part under different lattice structures are obtained.

[0015] In a specific embodiment of the present invention, the stress-strain curves under different lattice structures of the impact-resistant parts and the deep learning model are used to obtain a sub-model of the stress-strain curve of the lattice structure based on the unit cell structure parameters, including:

[0016] The dataset was constructed using stress-strain curves as input features of the deep learning model and different single-cell structural parameters as output features.

[0017] Train, test, and validate deep learning models based on datasets;

[0018] The verified deep learning model is used as a sub-model of the stress-strain curve of the lattice structure based on the unit cell structure parameters.

[0019] In a specific embodiment of the present invention, obtaining the relationship between the stress-strain curves of the impact-resistant component under different lattice structures and the absorbed energy of the corresponding impact-resistant component includes:

[0020] Construct a numerical simulation model of lattice impact fracture of impact-resistant parts;

[0021] According to the numerical simulation test of the impact fracture numerical simulation model, the relationship between the stress-strain curves under different lattice structures and the absorbed energy of the corresponding impact-resistant parts is obtained.

[0022] In a specific embodiment of the present invention, the construction of a numerical simulation model of lattice impact fracture of impact-resistant parts includes:

[0023] Keeping the total weight unchanged, a geometric model of a lattice impact-resistant part filled with different unit cell types was established;

[0024] Filling lattices in a geometric model of an impact-resistant part filled with lattices of different unit cell types, wherein the filled lattices are homogenized constitutive models constructed according to stress-strain curves under different lattice structures;

[0025] After the filling is complete, a numerical simulation model of the lattice impact fracture of the impact-resistant part is obtained.

[0026] In a specific embodiment of the present invention, the relationship between the stress-strain curves under different lattice structures of the impact-resistant part and the absorbed energy of the corresponding impact-resistant part, and the deep learning model, are used to obtain a sub-model of the lattice structure stress-strain curve for the absorbed energy of the impact-resistant part, including:

[0027] The dataset was constructed using absorbed energy as the input feature of the deep learning model and stress-strain curves under different lattice structures as the output feature.

[0028] Train, test, and validate deep learning models based on datasets;

[0029] The verified deep learning model is used as a sub-model of the lattice structure stress-strain curve to absorb energy of impact-resistant parts.

[0030] A second object of the present invention is to provide a method for designing an impact-resistant part, comprising:

[0031] According to the lattice structure stress-strain curve in the above-mentioned design model, the sub-model of the energy absorption of the anti-impact part and the optimal energy absorption value, the stress-strain curve corresponding to the optimal energy absorption is obtained, and the curve is used as the target stress-strain curve;

[0032] According to the unit cell structure parameters in the above design model, the sub-model of the lattice structure stress-strain curve and the target stress-strain curve, the optimal unit cell structure parameters corresponding to the target stress-strain curve are obtained;

[0033] Complete the design of impact-resistant parts based on the optimal unit cell structure parameters.

[0034] In a specific embodiment of the present invention, the unit cell structural parameters include connecting rod length, connecting rod diameter and unit cell relative density.

[0035] A third object of the present invention is to provide a system for establishing a design model for impact-resistant parts, the design model comprising a sub-model of unit cell structure parameters versus lattice structure stress-strain curves and a sub-model of lattice structure stress-strain curve versus energy absorption of the impact-resistant part;

[0036] The establishment system includes a first learning module and a second learning module;

[0037] The first learning module is used to obtain a sub-model of the stress-strain curve of the lattice structure based on the stress-strain curves of the impact-resistant part under different lattice structures and the deep learning model;

[0038] The second learning module is used to obtain a sub-model of the lattice structure stress-strain curve for the energy absorption of the impact-resistant part based on the relationship between the stress-strain curve under different lattice structures of the impact-resistant part and the absorbed energy of the corresponding impact-resistant part, as well as a deep learning model.

[0039] In a specific embodiment of the present invention, the first learning module includes a first submodule, a second submodule and a third submodule;

[0040] The first submodule is used to construct a data set using stress-strain curves as input features of a deep learning model and different single-cell structural parameters as output features;

[0041] The second submodule is used to train, test and verify the deep learning model based on the data set;

[0042] The third submodule is used to use the verified deep learning model as a submodel of the stress-strain curve of the lattice structure with respect to the unit cell structure parameters.

[0043] Beneficial effects of the present invention:

[0044] The design method of the impact-resistant parts of the present invention and the method and system for establishing the design model thereof, the design method of the present invention utilizes a deep learning method to construct two design sub-models - a sub-model of the unit cell structure parameters versus the lattice structure stress-strain curve and a sub-model of the lattice structure stress-strain curve versus the energy absorbed by the impact-resistant parts. In the process of designing the impact-resistant parts, the design model can consider both the relationship between the impact resistance performance and structural parameters of the impact-resistant parts and the relationship between the impact resistance performance and the absorbed energy of the impact-resistant parts. The final impact-resistant parts take into account both impact resistance and lightweight design, and achieve lightweighting of the impact-resistant parts while absorbing a specific value of impact energy.

[0045] The present invention also applies the above-mentioned design model and design method to the design of additively manufactured lattice sandwich containment casing, simplifies the design process, and obtains a containment casing structure with optimal energy absorption characteristics, better containment performance, and lighter weight; and has passed the simulation part test. After the test, the containment casing completed the containment assessment task, and compared with the traditional containment casing, the weight was reduced by 15%.

[0046] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 A flow chart showing a method for establishing a design model for an impact-resistant part according to an embodiment of the present invention is shown;

[0049] Figure 2 A flow chart showing a method for designing an impact-resistant part according to an embodiment of the present invention is shown;

[0050] Figure 3 A flow chart of a method for designing a lattice sandwich casing for additive manufacturing according to an embodiment of the present invention is shown;

[0051] Figure 4 A schematic diagram of a quasi-static compression test and a dynamic compression test of a solid component according to an embodiment of the present invention and corresponding stress-strain curves are shown;

[0052] Figure 5 A diagram showing a numerical simulation model of a wheel disc rupture according to an embodiment of the present invention is shown;

[0053] Figure 6 A framework diagram of a system for establishing a design model for an impact-resistant part according to an embodiment of the present invention is shown;

[0054] In the figure: 10, first learning module; 20, second learning module. DETAILED DESCRIPTION

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0056] like Figure 1 As shown, according to some embodiments of the present invention, a method for establishing a design model for an impact-resistant part is provided, wherein the design model includes a sub-model of unit cell structure parameters versus lattice structure stress-strain curve and a sub-model of lattice structure stress-strain curve versus energy absorption of the impact-resistant part;

[0057] The establishment method comprises:

[0058] S1. Based on the stress-strain curves of impact-resistant parts under different lattice structures and the deep learning model, a sub-model of the stress-strain curve of the lattice structure is obtained based on the unit cell structure parameters;

[0059] S2. Based on the relationship between the stress-strain curves of different lattice structures of impact-resistant parts and the absorbed energy of the corresponding impact-resistant parts, as well as the deep learning model, a sub-model of the lattice structure stress-strain curve and the absorbed energy of the impact-resistant parts is obtained.

[0060] In some embodiments of the present invention, in step S1, the unit cell structural parameters include connecting rod length, connecting rod diameter and unit cell relative density.

[0061] In certain embodiments of the present invention, in step S1, obtaining stress-strain curves of the impact-resistant component under different lattice structures includes:

[0062] A1. Construct the first unit cell size lattice structure model of the impact-resistant parts;

[0063] A2. Based on the compression test of the physical part and the compression numerical simulation test of the first unit cell size lattice structure model, the numerical simulation model of the impact-resistant parts is modified;

[0064] A3. Based on the numerical simulation test of the modified numerical simulation model of the impact-resistant part, the stress-strain curves of the impact-resistant part under different lattice structures are obtained.

[0065] In certain embodiments of the present invention, in step A1, the first unit cell size lattice structure model is a typical unit cell size lattice structure model corresponding to an impact-resistant part.

[0066] In some embodiments of the present invention, step A2 includes:

[0067] A2-1. Compare the actual stress-strain curve obtained from the compression test of a solid part of the first unit cell-sized lattice structure model with the simulated stress-strain curve obtained from the compression numerical simulation test, wherein the solid part is printed by an additive manufacturing method based on the first unit cell-sized lattice structure model;

[0068] A2-2. Determine whether the comparison result meets the error requirement.

[0069] A2-3. Modify the compression numerical model according to the judgment result that does not meet the error;

[0070] A2-4. Based on the compression numerical simulation test of the modified compression numerical model, a secondary simulation stress-strain curve is obtained;

[0071] A2-5. Repeat the steps of "comparing the entity stress-strain curve with the secondary simulation stress-strain curve", "judging whether the comparison result satisfies the error", "correcting the compression numerical model based on the judgment result that does not satisfy the error", and "obtaining the secondary simulation stress-strain curve based on the compression numerical simulation test of the corrected compression numerical model" until the result of the error judgment is satisfied, that is, repeat steps A2-1, A2-2, A2-3 and A2-4 until the result of the error judgment is satisfied.

[0072] In certain embodiments of the present invention, the modification in step A2-3 is specifically:

[0073] The adjustment parameters include three: "Young's modulus, initial yield strength, and subsequent yield hardening modulus of the matrix material." The curve obtained from the lattice structure compression simulation is processed to obtain the equivalent Young's modulus, initial yield strength, and subsequent yield hardening modulus of the lattice structure compression. These are then compared with the equivalent Young's modulus, initial yield strength, and subsequent yield hardening modulus obtained from the experimental curve to adjust the parameters. For example, if the equivalent Young's modulus obtained from the simulation is lower than that obtained from the test, the Young's modulus of the matrix material is increased, and so on, the initial yield strength and subsequent yield hardening modulus of the matrix material are adjusted.

[0074] In some embodiments of the present invention, step A3 includes:

[0075] According to the numerical simulation test of the modified numerical simulation model of the impact-resistant part, the stress-strain curves of the impact-resistant part under different lattice structures are obtained;

[0076] The numerical simulation test process is the same as the parameter adjustment of the correction process.

[0077] In some embodiments of the present invention, step S1 includes:

[0078] S1-1. Use stress-strain curves as input features of the deep learning model and different single-cell structural parameters as output features to construct a dataset;

[0079] S1-2. Train, test, and validate the deep learning model based on the dataset;

[0080] S1-3. Use the verified deep learning model as a sub-model of the stress-strain curve of the lattice structure based on the unit cell structure parameters.

[0081] In certain embodiments of the present invention, for example, the deep learning model in step S1 is one of the existing deep learning models, and the existing deep learning models are convolutional neural networks (CNN), generative adversarial networks (GAN), etc.

[0082] Step S1-2 is the training, testing and verification operations of the deep learning model well known in the technical field and will not be repeated here.

[0083] In certain embodiments of the present invention, in step S2, obtaining the relationship between the stress-strain curves of the impact-resistant component under different lattice structures and the absorbed energy of the corresponding impact-resistant component includes:

[0084] B1. Construct a numerical simulation model of lattice impact fracture of impact-resistant parts;

[0085] B2. Based on the numerical simulation test of the impact fracture numerical simulation model, the relationship between the stress-strain curves under different lattice structures and the absorbed energy of the corresponding impact-resistant parts is obtained.

[0086] In some embodiments of the present invention, step B1 includes:

[0087] B1-1. Keeping the total weight constant, establish a geometric model of a lattice impact-resistant part filled with different unit cell types;

[0088] B1-2. Filling a lattice in a geometric model of an impact-resistant part filled with different unit cell types, wherein the filled lattice is a homogenized constitutive model constructed based on stress-strain curves under different lattice structures;

[0089] B1-3. After the filling is complete, a numerical simulation model of the lattice impact fracture of the impact-resistant part is obtained.

[0090] In some embodiments of the present invention, step S2 includes:

[0091] S2-1. Using absorbed energy as the input feature of the deep learning model and stress-strain curves under different lattice structures as the output feature, a dataset was constructed.

[0092] S2-2. Train, test, and validate the deep learning model based on the dataset;

[0093] S2-3. Use the verified deep learning model as a sub-model of the lattice structure stress-strain curve to absorb energy of impact-resistant parts.

[0094] In certain embodiments of the present invention, for example, the deep learning model in step S2 is one of existing deep learning models, such as convolutional neural networks (CNNs) and generative adversarial networks (GANs). The deep learning model in step S2 may be the same as or different from the deep learning model in step S1.

[0095] Step S2-2 is the training, testing and verification operation of the deep learning model well known in the technical field and will not be repeated here.

[0096] like Figure 2 As shown, a design method for an impact-resistant part according to some embodiments of the present invention includes:

[0097] X1. Based on the stress-strain curve of the lattice structure in the design model of the above embodiment, the sub-model of the energy absorption of the anti-impact part and the optimal energy absorption value, obtain the stress-strain curve corresponding to the optimal energy absorption;

[0098] Specific process:

[0099] Input the optimal energy absorption value into the sub-model of the lattice structure stress-strain curve for energy absorption of impact parts, and obtain the corresponding stress-strain curve;

[0100] Among them, the optimal energy absorption value is related to the unit cell type. Different unit cell types have different relative densities, which leads to changes in the weight of the impact-resistant parts. The optimal energy absorption means that the energy absorption per unit mass (SEA) of the impact-resistant parts reaches the maximum value;

[0101] At the same time, each stress-strain curve corresponds to a specific unit cell type, so the energy absorption and stress-strain curve form a one-to-one correspondence.

[0102] X2. Based on the sub-model of the lattice structure stress-strain curve and the target stress-strain curve, the single cell structure parameters corresponding to the target stress-strain curve are obtained according to the unit cell structure parameters in the design model of the above embodiment. The specific process is as follows:

[0103] By inputting the target stress-strain curve into the sub-model of the unit cell structure parameter versus lattice structure stress-strain curve, the corresponding single cell structure parameters can be obtained;

[0104] X3. Complete the design of impact-resistant parts based on the optimal unit cell structure parameters.

[0105] The design method of the impact-resistant parts of the embodiment of the present invention is applicable to the additive manufacturing of lattice sandwich casings, as well as other impact-resistant fields, such as human body protection and high-speed and high-energy fragment protection.

[0106] like Figure 3 As shown in FIG, the entire process of the additive manufacturing lattice sandwich core-enclosed casing design method (including the process of building a design model) using the additive manufacturing lattice sandwich core-enclosed casing as an example includes:

[0107] After determining the specific lattice structure type, a typical unit cell lattice structure model was established. Finite element analysis was used to obtain stress-strain curves under quasi-static and dynamic compression simulations. Simultaneously, additive manufacturing was used to produce solid parts of typical unit cell sizes, and the simulation results were experimentally verified. Once the error requirements were met, numerical simulation was used to obtain stress-strain curves for a large number of different unit cell sizes. Machine learning methods were used to determine the relationship between lattice structure parameters and stress-strain curves.

[0108] While maintaining a constant gross weight, the containment casing was filled with lattice structures of varying sizes, and numerical simulations of wheel disc fracture were performed. A homogenized constitutive model for the lattice was constructed based on quasi-static and dynamic compression simulations. Deep learning methods were then used to determine the relationship between the lattice stress-strain curve and energy absorption, yielding the stress-strain curve corresponding to optimal energy absorption. By combining the relationship between the lattice structural parameters and the stress-strain curve, the optimal unit cell structural parameters were determined, ultimately resulting in a sandwich lattice containment casing with optimal energy absorption.

[0109] In the design process of the additive manufacturing lattice sandwich casing, the physical part compression test is a quasi-static compression test of the physical part and a dynamic compression test of the physical part, and the compression numerical simulation test is a quasi-static compression numerical simulation test and a dynamic compression numerical simulation test;

[0110] For the additively manufactured lattice sandwich casing, the geometric model of the lattice impact-resistant part filled with different unit cell types is a lattice sandwich casing model filled with different unit cell types;

[0111] For the containment casing, its impact fracture numerical simulation model corresponds to the wheel disc fracture numerical simulation model, and the corresponding fracture numerical simulation test is the wheel disc fracture simulation test.

[0112] For the additively manufactured lattice sandwich containment receiver, the core adopts a lattice structure and the shell adopts a high-ductility high-temperature alloy material. Through additive manufacturing, the material properties can be fully utilized, the containment can be improved, the weight of the containment ring can be reduced, and the manufacturing cost can be reduced.

[0113] In the design process of the additive manufacturing lattice sandwich core casing, the schematic diagram of the quasi-static compression test and dynamic compression test of the solid part used in step A2 and the corresponding stress-strain curve diagram are as follows: Figure 4 As shown in the figure, the numerical simulation model of the wheel disk rupture is as follows Figure 5 shown.

[0114] According to the above example, the sandwich lattice containment casing with the best energy absorption was printed into a simulation part by the additive manufacturing method. After the test, the containment casing completed the containment assessment task and reduced the weight by 15% compared with the traditional containment casing. 。

[0115] like Figure 6 As shown, a system for establishing a design model of an impact-resistant part according to some embodiments of the present invention includes a sub-model of unit cell structure parameters versus lattice structure stress-strain curve and a sub-model of lattice structure stress-strain curve versus energy absorption of the impact-resistant part;

[0116] The establishment system includes a first learning module 10 and a second learning module 20;

[0117] The first learning module 10 is used to obtain a sub-model of the stress-strain curve of the lattice structure based on the stress-strain curves of the impact-resistant part under different lattice structures and the deep learning model;

[0118] The second learning module 20 is used to obtain a sub-model of the lattice structure stress-strain curve for the energy absorption of the impact-resistant part based on the relationship between the stress-strain curve under different lattice structures of the impact-resistant part and the absorbed energy of the corresponding impact-resistant part, as well as a deep learning model.

[0119] In some embodiments of the present invention, the first learning module 10 includes a first submodule, a second submodule, and a third submodule;

[0120] The first submodule is used to construct a data set using stress-strain curves as input features of a deep learning model and different single-cell structural parameters as output features;

[0121] The second submodule is used to train, test and verify the deep learning model based on the data set;

[0122] The third submodule is used to use the verified deep learning model as a submodel of the stress-strain curve of the lattice structure with respect to the unit cell structure parameters.

[0123] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for establishing a design model for an impact-resistant part, characterized in that: The design model includes a sub-model of unit cell structure parameters versus lattice structure stress-strain curve and a sub-model of lattice structure stress-strain curve versus energy absorption of impact-resistant parts; The establishment method comprises: Based on the stress-strain curves of impact-resistant parts under different lattice structures and the deep learning model, a sub-model of the stress-strain curve of the lattice structure with respect to the unit cell structure parameters is obtained; Based on the relationship between the stress-strain curves of different lattice structures of impact-resistant parts and the absorbed energy of the corresponding impact-resistant parts, as well as the deep learning model, a sub-model of the lattice structure stress-strain curve for the absorbed energy of the impact-resistant parts is obtained.

2. The method for establishing a design model of an impact-resistant part according to claim 1, characterized in that: Obtaining stress-strain curves of the impact-resistant component under different lattice structures includes: Construct the first unit cell size lattice structure model of impact-resistant parts; Based on the compression test of the solid parts and the compression numerical simulation test of the first unit cell size lattice structure model, the numerical simulation model of the impact-resistant parts is modified; According to the numerical simulation test of the modified numerical simulation model of the impact-resistant part, the stress-strain curves of the impact-resistant part under different lattice structures are obtained.

3. The method for establishing a design model of an impact-resistant part according to claim 1, wherein: The stress-strain curves of different lattice structures of the impact-resistant parts and the deep learning model are used to obtain a sub-model of the stress-strain curve of the lattice structure based on the unit cell structure parameters, including: The dataset was constructed using stress-strain curves as input features of the deep learning model and different single-cell structural parameters as output features. Train, test, and validate deep learning models based on datasets; The verified deep learning model is used as a sub-model of the stress-strain curve of the lattice structure based on the unit cell structure parameters.

4. The method for establishing a design model of an impact-resistant part according to claim 1, wherein: Obtaining the relationship between the stress-strain curves of the impact-resistant parts under different lattice structures and the absorbed energy of the corresponding impact-resistant parts includes: Construct a numerical simulation model of lattice impact fracture of impact-resistant parts; According to the numerical simulation test of the impact fracture numerical simulation model, the relationship between the stress-strain curves under different lattice structures and the absorbed energy of the corresponding impact-resistant parts is obtained.

5. The method for establishing a design model of an impact-resistant component according to claim 4, characterized in that: The method of constructing a numerical simulation model of lattice impact fracture of impact-resistant parts comprises: Keeping the total weight unchanged, a geometric model of a lattice impact-resistant part filled with different unit cell types was established; Filling lattices in a geometric model of an impact-resistant part filled with lattices of different unit cell types, wherein the filled lattices are homogenized constitutive models constructed according to stress-strain curves under different lattice structures; After the filling is complete, a numerical simulation model of the lattice impact fracture of the impact-resistant part is obtained.

6. A method for establishing a design model for an impact-resistant component according to any one of claims 1 to 5, characterized in that: The relationship between the stress-strain curves under different lattice structures of the impact-resistant parts and the absorbed energy of the corresponding impact-resistant parts, and the deep learning model, is used to obtain a sub-model of the lattice structure stress-strain curve for the absorbed energy of the impact-resistant parts, including: The dataset was constructed using absorbed energy as the input feature of the deep learning model and stress-strain curves under different lattice structures as the output feature. Train, test, and validate deep learning models based on datasets; The verified deep learning model is used as a sub-model of the lattice structure stress-strain curve to absorb energy of impact-resistant parts.

7. A method for designing impact-resistant parts, characterized in that: include: According to the sub-model of the lattice structure stress-strain curve in the design model according to any one of claims 1 to 6 for absorbing energy of the impact-resistant part, and the optimal energy absorption value, a stress-strain curve corresponding to the optimal energy absorption is obtained, and the curve is used as the target stress-strain curve; According to any one of claims 1 to 6, the unit cell structure parameters in the design model are compared with the sub-model of the lattice structure stress-strain curve, and the target stress-strain curve to obtain the optimal unit cell structure parameters corresponding to the target stress-strain curve; Complete the design of impact-resistant parts based on the optimal unit cell structure parameters.

8. The method for designing an impact-resistant component according to claim 7, characterized in that: The unit cell structural parameters include connecting rod length, connecting rod diameter and unit cell relative density.

9. A system for establishing a design model for impact-resistant parts, characterized in that: The design model includes a sub-model of unit cell structure parameters versus lattice structure stress-strain curve and a sub-model of lattice structure stress-strain curve versus energy absorption of impact-resistant parts; The establishment system includes a first learning module and a second learning module; The first learning module is used to obtain a sub-model of the stress-strain curve of the lattice structure based on the stress-strain curves of the impact-resistant part under different lattice structures and the deep learning model; The second learning module is used to obtain a sub-model of the lattice structure stress-strain curve for the energy absorption of the impact-resistant part based on the relationship between the stress-strain curve under different lattice structures of the impact-resistant part and the absorbed energy of the corresponding impact-resistant part, as well as a deep learning model.

10. The system for establishing a design model of an impact-resistant part according to claim 9, characterized in that: The first learning module includes a first submodule, a second submodule and a third submodule; The first submodule is used to construct a data set using stress-strain curves as input features of a deep learning model and different single-cell structural parameters as output features; The second submodule is used to train, test and verify the deep learning model based on the data set; The third submodule is used to use the verified deep learning model as a submodel of the stress-strain curve of the lattice structure with respect to the unit cell structure parameters.

Images