A voronoi type lattice reliability optimization method and device

Abstract

This invention discloses a method and apparatus for optimizing the reliability of Voronoi lattice structures, belonging to the field of lightweight structural design. The method includes: dividing the component to be optimized into elements, and determining the number of seed points in each element based on the division results; sequentially inputting the number of seed points in all elements into a pre-trained Voronoi lattice structural optimization model, and outputting the optimal distribution of seed points in each element; wherein, the structural optimization model is established based on element mass, the number of seed points, and the element load-bearing failure probability, the element load-bearing failure probability is determined based on a preset performance distribution proxy model, and the performance distribution proxy model is determined based on the finite element simulation results of the Voronoi lattice geometric model; and generating optimal components that meet preset requirements based on the optimal distribution of seed points in each element. This invention can effectively improve the reliability of Voronoi lattice structures.

Description

Technical Field

[0001] This invention relates to the field of lightweight structural design technology, and in particular to a method and apparatus for optimizing the reliability of Voronoi type lattice. Background Technology

[0002] Voronoi lattices are generated from randomly distributed seed points. Even if the number of seed points is fixed, the distribution of seed point positions is different each time they are randomly generated, resulting in significant and unavoidable random differences in the final lattice structure (arrangement of rods, connection method) and its mechanical properties (such as elastic modulus and strength).

[0003] In related technologies, traditional lattice optimization design methods typically simplify the problem into deterministic optimization. This simplification ignores the aforementioned randomness, leading to significant risks in the actual manufacturing and application of the optimization results.

[0004] Therefore, there is an urgent need for a Voronoi-type lattice reliability optimization method and apparatus to solve the above-mentioned technical problems. Summary of the Invention

[0005] This invention provides a method and apparatus for optimizing the reliability of Voronoi-type lattice structures, which can effectively improve the reliability of Voronoi-type lattice structures. The technical solution is as follows: On the one hand, a reliability optimization method for Voronoi type lattice is provided, the method comprising: The parts to be optimized are divided into units, and the number of seed points in each unit is determined based on the division results; The number of seed points in all units is sequentially input into a pre-trained Voronoi-type lattice structure optimization model, and the optimal distribution of seed points in each unit is output. The structure optimization model is established based on the unit mass, the number of seed points, and the unit load failure probability. The unit load failure probability is determined based on a preset performance distribution proxy model, which is determined based on the finite element simulation results of the Voronoi-type lattice geometric model. Based on the optimal distribution of seed points within each unit, the optimal components that meet the preset requirements are generated.

[0006] On the other hand, a Voronoi-type lattice reliability optimization device is provided, the device comprising: The partitioning module is used to divide the parts to be optimized into units and determine the number of seed points in each unit based on the partitioning results. The optimization module is used to sequentially input the number of seed points in all units into a pre-trained Voronoi-type lattice structure optimization model and output the optimal distribution of seed points in each unit. The structure optimization model is established based on unit mass, number of seed points and unit load failure probability. The unit load failure probability is determined based on a preset performance distribution proxy model. The performance distribution proxy model is determined based on the finite element simulation results of the Voronoi-type lattice geometric model. The generation module is used to generate the optimal component that meets the preset requirements based on the optimal distribution of seed points within each unit.

[0007] On the other hand, a computer device is provided, the computer device including a memory and a processor, the memory for storing computer programs, and the processor for executing the computer programs stored in the memory to implement the steps of the Voronoi type lattice reliability optimization method described above.

[0008] On the other hand, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements the steps of the Voronoi-type lattice reliability optimization method described above.

[0009] On the other hand, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the Voronoi-type lattice reliability optimization method described above.

[0010] The technical solution provided by this invention offers at least the following beneficial effects: It innovatively considers the randomness of the Voronoi lattice during optimization and adds reliability constraints to component optimization, ensuring the reliability of component load-bearing capacity. Furthermore, to improve optimization efficiency, a surrogate model is constructed to quickly predict the random distribution of lattice performance, transforming the originally time-consuming and lengthy randomness analysis process into instantaneous computation. This method successfully transforms an advanced structure with inherent randomness into an engineering component that can be reliably and efficiently designed and applied, which is of great value in promoting the practical application of lattice structures in high-performance fields such as aerospace and automotive. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1This is a flowchart of a Voronoi-type lattice reliability optimization method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a Voronoi-type dot matrix reliability optimization provided in an embodiment of the present invention; Figure 3 This is a structural diagram of a Voronoi-type dot matrix reliability optimization device provided in an embodiment of the present invention; Figure 4 This is a hardware architecture diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0014] As mentioned earlier, traditional methods treat Voronoi lattices as deterministic structures, ignoring the inherent performance fluctuations due to the random distribution of seed points. This results in structures with theoretically superior performance but actual load-bearing capacity (i.e., reliability) falling far short of design expectations.

[0015] Based on this, the concept of the present invention is to optimize the distribution of seed points by constructing a proxy model for the performance distribution of Voronoi lattice, to consider randomness in lightweight design, improve structural reliability, and avoid the risks of traditional deterministic optimization.

[0016] The specific implementation of the above concept is described below.

[0017] Please refer to Figure 1 This invention provides a Voronoi-type lattice reliability optimization method, which includes: Step 100: Divide the parts to be optimized into units and determine the number of seed points in each unit based on the division results; Step 102: Input the number of seed points in all units into the pre-trained Voronoi lattice structure optimization model in sequence, and output the optimal distribution of seed points in each unit; wherein, the structure optimization model is established based on the unit mass, the number of seed points and the unit load failure probability, the unit load failure probability is determined based on the preset performance distribution proxy model, and the performance distribution proxy model is determined based on the finite element simulation results of the Voronoi lattice geometric model; Step 104: Generate the optimal component that meets the preset requirements based on the optimal distribution of seed points within each unit.

[0018] In this embodiment of the invention, the randomness of the Voronoi lattice is innovatively considered during optimization, and reliability constraints are added to the component optimization to ensure the reliability of the component's load-bearing capacity. Simultaneously, to improve optimization efficiency, a surrogate model is constructed to quickly predict the random distribution of lattice performance, transforming the originally time-consuming and lengthy randomness analysis process into instantaneous computation. This method successfully transforms an advanced structure with inherent randomness into an engineering component that can be reliably and efficiently designed and applied, which has significant value in promoting the practical application of lattice structures in high-performance fields such as aerospace and automotive.

[0019] The following description Figure 1 The execution method for each step is shown.

[0020] First, for step 100, the parts to be optimized are divided into units, and the number of seed points in each unit is determined based on the division results.

[0021] In embodiments of the present invention, such as Figure 2 As shown, the given components are divided into units, specifically cubic units similar to pixel blocks. Each cubic unit is a dot matrix design unit. Then, a trained dot matrix structure optimization model is used to calculate the number of seed points within each dot matrix design unit. The Voronoi dot matrix structure generated under this set of seed point numbers is used to fill the interior of the components.

[0022] Then, for step 102, the number of seed points in all units is sequentially input into the pre-trained Voronoi type lattice structure optimization model, and the optimal distribution of seed points in each unit is output.

[0023] In this embodiment of the invention, the structural optimization model is established based on the unit mass, the number of seed points, and the unit bearing failure probability. The unit bearing failure probability is determined based on a preset performance distribution proxy model, which is determined based on the finite element simulation results of the Voronoi lattice geometric model.

[0024] In this embodiment of the invention, the performance distribution proxy model is determined based on the finite element simulation results of the Voronoi-type lattice geometric model, including: Based on the preset target number of seed points, Latin hypercube sampling is repeated multiple times to randomly generate multiple sets of seed point distribution structures, as well as the Voronoi lattice geometric model corresponding to each seed point distribution structure.

[0025] Specifically, in 10*10*10 mm 3Within the region, the Latin hypercube method is used to randomly select... n There are several seed points used to generate Voronoi lattices. Generally, the more seed points there are, the denser the distribution, and the more members the Voronoi lattice generated based on the seed points contains, resulting in better performance, but also higher density.

[0026] Because the point selection method is random, the distribution of seed points and the generated Voronoi lattice are also random each time. For example, when the number of seed points in the lattice is... n In this case, 1000 sets of random seed points are generated, and these 1000 sets of seed points are different. Each of these 1000 sets of seed points is then used to generate 1000 corresponding Voronoi-type lattices. These 1000 lattices have different member distributions and structural properties.

[0027] Furthermore, finite element simulations were performed on all Voronoi-type lattice geometric models to obtain the mechanical performance parameters of each lattice geometric model.

[0028] Specifically, finite element simulations were performed on the performance of these 1000 sets of lattice structures. The quasi-static compression condition was simulated by constraining and fixing the lower surface of the generated cubic lattice, while applying a low-speed, uniform displacement to the upper surface. d Finally, the support reaction force on the lower surface is read. F And the energy absorption of the dot matrix.

[0029] The energy absorption caused by the destruction of the lattice was calculated. : in, S Let be the area of ​​the lower surface.

[0030] The elastic modulus of the lattice was calculated. : in, l This is the distance between the upper and lower surfaces.

[0031] The intensity of the lattice was calculated. : in, This represents the reaction force on the lower surface when the lattice is destroyed.

[0032] Furthermore, the mean and variance of the performance distribution corresponding to the number of target seed points are calculated based on the mechanical performance parameters.

[0033] After obtaining 1000 sets of data related to lattice performance, since the performance typically follows a Gaussian distribution, the mean lattice performance can be calculated using the following formula. : The variance of lattice performance is calculated using the following formula. : in, Indicates the first of 1000 dot matrix groups i Performance parameters of the lattice x These can be the elastic modulus, mean, and energy absorption. Thus, the mean and variance can be used to describe the randomness of the lattice properties.

[0034] Finally, based on the mean and variance of the performance distribution corresponding to different seed point numbers, a proxy model for the performance distribution corresponding to the target seed point number is established.

[0035] A surrogate model, also called an approximation model, uses a simpler model to approximate a complex model. This method sacrifices some accuracy but significantly improves efficiency. Since calculating the mean and variance of the lattice performance with any number of seed points requires repeated calculations, resulting in low computational efficiency, this step improves the computational efficiency of lattice randomness by establishing a surrogate model to approximate the original model.

[0036] In this embodiment of the invention, the performance distribution proxy model is established as follows: the mean and variance of the performance distribution corresponding to multiple sets of seed point numbers are obtained; the seed point numbers, the mean of the performance distribution, and the variance of the performance distribution are fitted respectively to establish the performance distribution proxy model of the Voronoi type lattice geometric model.

[0037] Specifically, the number of seed points required to predict lattice performance is determined to be between 5 and 40. Sample points are then selected within this range. Seed point numbers of 5, 10, ..., 40 are used as approximate fitting sample points. The steps for generating mechanical performance parameters are repeated for each of these seed point numbers to calculate the response corresponding to the selected sample points, i.e., the random distribution of lattice performance. Now, several sample points and their responses are known. These are then fitted using a Chebyshev polynomial to obtain a surrogate model of the random distribution of lattice performance. At this point, calculating the random distribution of lattice performance no longer requires the inefficient and tedious process described above; it only requires calculating the number of seed points. nBy incorporating the surrogate model, the performance distribution of the generated point matrix can be quickly predicted with any number of seed points within the range of 5-40.

[0038] In this embodiment of the invention, determining the unit load-bearing failure probability according to the performance distribution proxy model includes: inputting the number of seed points of the unit into the performance distribution proxy model, and outputting the mean and variance of the performance distribution of the unit under different seed point distribution structures; based on the mean and variance of the performance distribution of the unit, counting the number of times the components corresponding to the unit experience load-bearing failure; and calculating the ratio of the number of load-bearing failures to the total number of lattice structures to obtain the unit load-bearing failure probability.

[0039] Specifically, based on the number of seed points n This generates 1000 performance combinations of different design units, allowing for the calculation of the component lattice structure performance corresponding to each design unit combination. Within these 1000 data sets, the number of times a component experiences load-bearing failure is recorded. k So, the number of seed points n The probability of failure of the generated lattice structure for: .

[0040] In this embodiment of the invention, the optimized structural model includes: the optimization objective is to minimize the mass of the components; the optimization variable is the number of seed points within each design unit; the optimization constraint is that the load-bearing reliability of the components is greater than 95%, i.e., the failure probability is less than 5%. The structural optimization model is established using the following formula: In the formula, This is a vector representing the number of seed points within each unit. This refers to the unit mass calculated when the seed point vector is n. This represents the failure probability of the unit.

[0041] For step 104, the optimal component that meets the preset requirements is generated based on the optimal distribution of seed points in each unit.

[0042] For example, such as Figure 2 As shown, the above steps were used to design a lattice structure inside a solid three-point bending beam. The beam was divided into 8 design unit regions, and the number of seed points was optimized. The results show that the load-bearing reliability of the three-point bending beam is greater than 95%, and the mass is reduced by 76%. In contrast, the reliability of the three-point bending beam obtained by the conventional Voronoi lattice design method is only 54%, far less than the reliability requirement. Moreover, the surrogate model calculation significantly improves efficiency, reducing the optimization process from 32 hours to 4 minutes.

[0043] Please refer to Figure 3This invention provides a Voronoi-type lattice reliability optimization device, which includes: The partitioning module 300 is used to partition the parts to be optimized into units and determine the number of seed points in each unit based on the partitioning results. The optimization module 302 is used to sequentially input the number of seed points in all units into a pre-trained Voronoi-type lattice structure optimization model and output the optimal distribution of seed points in each unit; wherein, the structure optimization model is established based on the unit mass, the number of seed points and the unit load failure probability, the unit load failure probability is determined based on a preset performance distribution proxy model, and the performance distribution proxy model is determined based on the finite element simulation results of the Voronoi-type lattice geometric model; The generation module 304 is used to generate the optimal component that meets the preset requirements based on the optimal distribution of seed points in each unit.

[0044] In this embodiment of the invention, the performance distribution proxy model is determined based on the finite element simulation results of the Voronoi-type lattice geometric model, including: The Latin hypercube sampling process is repeated multiple times based on the preset target number of seed points to randomly generate multiple sets of seed point distribution structures and the corresponding Voronoi lattice geometric model for each seed point distribution structure. Finite element simulations were performed on all Voronoi-type lattice geometric models to obtain the mechanical performance parameters of each lattice geometric model. Calculate the mean and variance of the performance distribution corresponding to the number of target seed points based on the mechanical performance parameters; Based on the mean and variance of the performance distribution corresponding to different seed point numbers, a proxy model for the performance distribution corresponding to the target seed point number is established.

[0045] In this embodiment of the invention, the mechanical performance parameters include elastic modulus, lattice strength, and lattice failure energy absorption.

[0046] In this embodiment of the invention, establishing a proxy model for the performance distribution corresponding to the target number of seed points based on the mean and variance of the performance distribution corresponding to different seed point numbers includes: Obtain the mean and variance of the performance distribution corresponding to multiple sets of seed points; By fitting the seed point count, the mean of the performance distribution, and the variance of the performance distribution, a proxy model for the performance distribution of the Voronoi type lattice geometric model is established.

[0047] In this embodiment of the invention, determining the unit bearer failure probability based on the performance distribution proxy model includes: Input the number of seed points of the unit into the performance distribution proxy model, and output the mean and variance of the performance distribution of the unit under different seed point distribution structures; Based on the mean and variance of the performance distribution of the unit, the number of times the load-bearing failure of the corresponding component of the unit is counted; The ratio of the number of load-bearing failures to the total number of lattice structures is used to obtain the unit load-bearing failure probability.

[0048] In this embodiment of the invention, the structural optimization model is established using the following formula: In the formula, This is a vector representing the number of seed points within each unit. For the seed point vector is n The unit mass is calculated at that time; This represents the failure probability of the unit.

[0049] It should be noted that the Voronoi-type lattice reliability optimization device provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the Voronoi-type lattice reliability optimization device and the Voronoi-type lattice reliability optimization method embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0050] Embodiments of this application also provide a computer device, please refer to... Figure 4 The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, at least one program, code set or instruction set being loaded and executed by the processor to implement the Voronoi type lattice reliability optimization method provided in the above method embodiments.

[0051] Embodiments of this application also provide a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the Voronoi-type dot matrix reliability optimization method provided in the above-described method embodiments.

[0052] Embodiments of this application also provide a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium and executes the computer program, causing the computer device to perform any of the Voronoi-type lattice reliability optimization methods described in the above embodiments.

[0053] For ease of description, the above systems or devices are described separately as various modules or units based on their functions. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware components.

[0054] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0055] Finally, it should be noted that in this document, relational terms such as first, second, third, and fourth are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0056] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for optimizing the reliability of a Voronoi type lattice, characterized in that, The method includes: The parts to be optimized are divided into units, and the number of seed points in each unit is determined based on the division results; The number of seed points in all units is sequentially input into a pre-trained Voronoi-type lattice structure optimization model, and the optimal distribution of seed points in each unit is output. The structure optimization model is established based on the unit mass, the number of seed points, and the unit load failure probability. The unit load failure probability is determined based on a preset performance distribution proxy model, which is determined based on the finite element simulation results of the Voronoi-type lattice geometric model. Based on the optimal distribution of seed points within each unit, the optimal components that meet the preset requirements are generated.

2. The method of claim 1, wherein, Based on the finite element simulation results of the Voronoi-type lattice geometry model, a performance distribution proxy model is determined, including: The Latin hypercube sampling process is repeated multiple times based on the preset target number of seed points to randomly generate multiple sets of seed point distribution structures and the corresponding Voronoi lattice geometric model for each seed point distribution structure. Finite element simulations were performed on all Voronoi-type lattice geometric models to obtain the mechanical performance parameters of each lattice geometric model. Calculate the mean and variance of the performance distribution corresponding to the number of target seed points based on the mechanical performance parameters; Based on the mean and variance of the performance distribution corresponding to different seed point numbers, a proxy model for the performance distribution corresponding to the target seed point number is established.

3. The method of claim 2, wherein, The mechanical performance parameters include elastic modulus, lattice strength, and lattice failure energy absorption.

4. The method of claim 2, wherein, The step of establishing a proxy model for the performance distribution corresponding to the target number of seed points based on the mean and variance of the performance distribution corresponding to different seed point numbers includes: Obtain the mean and variance of the performance distribution corresponding to multiple sets of seed points; By fitting the seed point count, the mean of the performance distribution, and the variance of the performance distribution, a proxy model for the performance distribution of the Voronoi type lattice geometric model is established.

5. The method of claim 1, wherein, The failure probability of the unit bearing is determined based on the performance distribution proxy model, including: Input the number of seed points of the unit into the performance distribution proxy model, and output the mean and variance of the performance distribution of the unit under different seed point distribution structures; Based on the mean and variance of the performance distribution of the unit, the number of times the load-bearing failure of the corresponding component of the unit is counted; The ratio of the number of load-bearing failures to the total number of lattice structures is used to obtain the unit load-bearing failure probability.

6. The method of claim 1, wherein, The structural optimization model is established using the following formula: wherein is a vector of the number of seed points in each cell; is the cell quality calculated when the seed point vector is n is the cell quality calculated when the seed point vector is is the failure probability carried by the cell.

7. A Voronoi-type lattice reliability optimization apparatus, characterized by, The device includes: The partitioning module is used to divide the parts to be optimized into units and determine the number of seed points in each unit based on the partitioning results. The optimization module is used to sequentially input the number of seed points in all units into a pre-trained Voronoi-type lattice structure optimization model and output the optimal distribution of seed points in each unit. The structure optimization model is established based on unit mass, number of seed points and unit load failure probability. The unit load failure probability is determined based on a preset performance distribution proxy model. The performance distribution proxy model is determined based on the finite element simulation results of the Voronoi-type lattice geometric model. The generation module is used to generate the optimal component that meets the preset requirements based on the optimal distribution of seed points within each unit.

8. A computer device, comprising: The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to implement the steps of the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1-6.

Images