Topological optimization method and device of frame material, electronic equipment and storage medium

Abstract

The invention provides a topological optimization method and device for a frame material, electronic equipment and a storage medium. The method comprises the following steps: according to a preset grid seed size, performing grid division on a geometric model of a frame to obtain a plurality of discrete frame units; configuring a frame material for each frame unit of all the frame units, and constructing a finite element model; the finite element model comprises physical characteristics and a target function taking the minimum average flexibility of the topological optimization structure as an optimization target; the target function comprises the macroscopic density of each frame unit of all the frame units and constraint conditions of the whole volume of the topological optimization structure; the macroscopic density comprises a variable mapped by the microscopic density of the frame material; based on the target function and a global optimization algorithm, global optimization is carried out on the average flexibility corresponding to each frame unit of all the frame units, and a topological optimization structure after frame material optimization is obtained; the optimized topological optimization structure comprises the minimum average flexibility of the topological optimization structure.

Description

Technical Field

[0001] The present application relates to the technical field of topology optimization, and in particular to a topology optimization method, device, electronic device and storage medium for frame materials. Background Art

[0002] Currently, topology optimization is an effective design method for seeking high-performance, lightweight and multifunctional innovative structures.

[0003] Among the many topology optimization methods, the level set method can accurately track boundary evolution and has the advantage of obtaining clear and smooth structural boundaries. However, when the level set method uses the radial basis function (RBF) for decoupling and optimizes the level set function in a local range, the accuracy of the final topology optimization results is reduced. Summary of the invention

[0004] The present application provides an improved topology optimization method, device, electronic device and storage medium for frame materials.

[0005] The present application provides a topology optimization method for a frame material, comprising:

[0006] According to the preset grid seed size, the geometric model of the frame is meshed to obtain multiple discrete frame units;

[0007] A frame material is configured for each of all the frame units, and a finite element model is constructed; the finite element model includes: physical characteristics, and an objective function with the minimized average flexibility of the topology optimized structure as an optimization target; the objective function includes the macro density of each of all the frame units, and the constraint conditions of the entire volume of the topology optimized structure; the macro density includes variables mapped by the micro density of the frame material; the micro density presents a periodic distribution in the macro density;

[0008] Based on the objective function and the global optimization algorithm, the average flexibility corresponding to each frame unit of all frame units is globally optimized to obtain a topological optimization structure after the frame material is optimized; the optimized topological optimization structure includes the minimized average flexibility of the topological optimization structure.

[0009] Furthermore, based on the objective function and the global optimization algorithm, the average flexibility corresponding to each frame unit of all frame units is globally optimized to obtain a topological optimization structure after the frame material is optimized, including:

[0010] Determine the average compliance of each frame unit relative to the macro density of the material;

[0011] Based on the change, the sensitivity of each frame unit of all frame units is obtained; the greater the sensitivity of each frame unit, the greater the macro density of each frame unit;

[0012] According to the size of the sensitivity, a global optimization algorithm is used to globally optimize the average flexibility corresponding to each frame unit of all frame units to obtain the optimized topology optimization structure.

[0013] Further, determining the change in average flexibility of each frame unit relative to the macroscopic density of the material includes:

[0014] Derivative the macroscopic density corresponding to each frame unit of all frame units to obtain the change;

[0015] According to the sensitivity, the global optimization algorithm is used to globally optimize the average flexibility corresponding to each frame unit of all frame units to obtain the optimized topology optimization structure, including:

[0016] Using a global optimization algorithm, sorting the sensitivity of each frame unit of all frame units according to the size of the sensitivity, and determining a first frame area corresponding to a high sensitivity and a second frame area corresponding to a low sensitivity; wherein the first frame area is retained and the second frame area is deleted;

[0017] Determining whether the volumes of all first frame regions meet the constraint conditions;

[0018] If it is achieved, the optimized topological optimized structure is obtained;

[0019] The first frame area is a frame area formed by a plurality of frame units having a sensitivity greater than a preset threshold; the second frame area is a frame area formed by a plurality of frame units having a sensitivity less than a preset threshold.

[0020] Furthermore, the global optimization algorithm includes a genetic algorithm; each frame unit is an independent individual of the population corresponding to the genetic algorithm; the population number of the frame units remains unchanged;

[0021] Based on the objective function and the global optimization algorithm, the average flexibility corresponding to each frame unit of all frame units is globally optimized to obtain a topological optimization structure after the frame material is optimized, including:

[0022] During the iterative cycle of the genetic algorithm, the current iteration executes the following steps A to C:

[0023] Step A, performing a crossover operation and a mutation operation of a genetic operation on each frame unit in turn to obtain a genetic frame unit;

[0024] Step B, determining the sensitivity of the inherited frame unit;

[0025] Step C, sorting the sensitivities of the inherited frame units, selecting a specified number of frame units according to the sorting of the sensitivities, determining a first frame region corresponding to high sensitivity, and determining a second frame region corresponding to low sensitivity; wherein the first frame region is retained and the second frame region is deleted; the specified number is the product of the evolution rate of the genetic algorithm and the population size; the evolution rate decreases as the number of iterations increases;

[0026] Step D, determining whether the volumes of all first frame regions meet the constraint conditions;

[0027] If not, return to step A to step D for loop iteration;

[0028] If so, and when the average flexibility corresponding to each frame unit satisfies the convergence condition, the iterative process is terminated to obtain the optimized topology optimization structure.

[0029] Furthermore, step C includes: starting from the direction of maximum sensitivity to the direction of decreasing sensitivity, selecting a specified number of frame units as the first frame area corresponding to high sensitivity; starting from the direction of minimum sensitivity to the direction of increasing sensitivity, selecting a specified number of frame units as the second frame area corresponding to low sensitivity.

[0030] Furthermore, the global optimization algorithm includes a non-dominated sorting genetic algorithm; and step C includes: using the non-dominated sorting genetic algorithm to perform non-dominated sorting on the sensitivities of the inherited frame units, determining the frame units on the non-dominated frontier as the first frame region and the frame region corresponding to the sensitivities of the inherited frame units other than the first frame region as the second frame region.

[0031] Furthermore, the step A comprises:

[0032] Randomly select any two frame units as parents in the design domain, randomly select any two crossover positions, perform a crossover operation on the parents based on the crossover rate, and obtain the crossover frame units; the crossover operator of the crossover operation is executed according to the two-point crossover operator;

[0033] A mutation operation is performed on the binary code of each crossed frame unit according to the mutation rate to obtain a mutated frame unit as the inherited frame unit; the mutation operator of the mutation operation adopts basic bit mutation.

[0034] Furthermore, the iterative cycle process of the genetic algorithm includes:

[0035] In each iteration step, for the first frame region, as the sensitivity decreases from high to low, the crossover rate decreases each time and iterates to a minimum value, and the first frame region is retained; and

[0036] For the second frame region, as the sensitivity decreases from high to low, the mutation rate increases to a maximum value in each iteration, and the second frame region is deleted.

[0037] The present application provides a topology optimization device for implementing the above-mentioned topology optimization method for frame materials, the topology optimization device comprising:

[0038] A meshing module, used to mesh the geometric model of the frame according to a preset mesh seed size to obtain a plurality of discrete frame units;

[0039] A model building module is used to configure a frame material for each of all frame units and build a finite element model; the finite element model includes: physical characteristics, and an objective function with the minimized average flexibility of the topology optimization structure as an optimization target; the objective function includes the macro density of each frame unit of all frame units and the constraint conditions of the entire volume of the topology optimization structure; the macro density includes variables mapped by the micro density of the frame material; the micro density presents a periodic distribution in the macro density;

[0040] A global optimization module is used to globally optimize the average flexibility corresponding to each frame unit of all frame units based on the objective function and the global optimization algorithm to obtain a topological optimization structure after the frame material is optimized; the optimized topological optimization structure includes the minimized average flexibility of the topological optimization structure.

[0041] The present application provides an electronic device, comprising one or more processors, for implementing any of the methods described above.

[0042] The present application provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method described in any one of the above items is implemented.

[0043] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0044] In some embodiments, the topology optimization method of the frame material of the present application includes an objective function that uses the minimization of the average flexibility of the topology optimization structure as the optimization target through a finite element model, and the objective function includes dual scales such as macro density and micro density. Therefore, the dual-scale planning optimization makes the topology optimization structure after the frame material is optimized more accurate. At the same time, the global optimization algorithm is combined with the dual-scale topology. Through the advantages of the global optimization algorithm in global search, it is possible to avoid the inability to fully utilize all possible design spaces caused by local optimization in related technologies, thereby enhancing the utilization of the design space and improving the accuracy of the topology optimization structure after the frame material is optimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a schematic flow chart of a topology optimization method for a vehicle frame material according to an embodiment of the present application;

[0046] Figure 2 Shown Figure 1 The topology optimization method for the frame material shown includes a schematic diagram of binary coding of multiple discrete frame units;

[0047] Figure 3 Shown Figure 1 The topology optimization method for the frame material shown includes a schematic diagram of the frame;

[0048] Figure 4 The figure shows a schematic diagram of the structure of a topology optimization device for a vehicle frame material provided in an embodiment of the present application;

[0049] Figure 5 Shown is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of the present specification. Instead, they are merely examples of devices and methods consistent with some aspects of one or more embodiments of the present specification as detailed in the appended claims.

[0051] It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0052] In order to solve the technical problem that the above-mentioned related technologies reduce the accuracy of the final topology optimization results, in the embodiment of the present application, the finite element model includes an objective function that takes the minimization of the average flexibility of the topology optimization structure as the optimization target, and the objective function includes dual scales such as macro density and micro density. Therefore, the dual-scale planning optimization makes the topology optimization structure after the frame material is optimized more accurate. At the same time, the global optimization algorithm is combined with the dual-scale topology. Through the advantages of the global optimization algorithm in global search, it is possible to avoid the inability to fully utilize all possible design spaces caused by local optimization in related technologies, thereby enhancing the utilization of the design space and improving the accuracy of the topology optimization structure after the frame material is optimized.

[0053] Figure 1 The following is a flow chart of the topology optimization method for the frame material according to an embodiment of the present application.

[0054] like Figure 1 As shown, the topology optimization method of the frame material may include but is not limited to the following steps 110 to 130:

[0055] Step 110 , meshing the geometric model of the frame according to a preset mesh seed size to obtain a plurality of discrete frame units.

[0056] The preset grid seed size can be based on user needs. The preset grid seed size is used to indicate the fineness of the grid. According to the preset grid seed size, the geometric model of the frame is meshed to obtain a plurality of discrete grid units as a plurality of discrete frame units.

[0057] It should be noted that the vehicle frame in this article may be, but is not limited to, one or more of a vehicle frame of an aircraft tractor, a vehicle frame of an aircraft maintenance platform, and a vehicle frame of an airport cargo tractor.

[0058] The aircraft tractor is a ground vehicle used to push and pull an aircraft. The frame supports the traction system and carries the weight of the aircraft to ensure the movement of the aircraft on the ground.

[0059] The aircraft maintenance platform is used to provide a lifting platform for support when maintaining an aircraft, and the frame is used to fix the platform and carry maintenance personnel and equipment.

[0060] The above-mentioned airport cargo tractor is a vehicle used to transport cargo and luggage in and outside the aircraft cargo hold, and the frame provides load-bearing and stability.

[0061] In some examples, the aircraft tractor may be the frame of an airport luggage tractor. The frame of the airport luggage tractor is a key component that bears the weight of the vehicle body and the load of the luggage trailer. The design of the frame needs to have sufficient structural strength and durability to cope with long-term high-load operation requirements. The present application uses a genetic algorithm to perform global optimization and explore a variety of design schemes to create a lighter and more load-bearing frame structure. While ensuring the lightweight of the frame, its stress distribution and material utilization are optimized, thereby improving the overall performance of the tractor and extending the service life of the frame. In addition, by expanding the design space, the present application can design more innovative frame structures, break the limitations of traditional designs, and achieve higher performance.

[0062] Step 120, configure the frame material for each frame unit of all frame units and construct a finite element model; the finite element model includes: physical characteristics, and an objective function with minimizing the average flexibility of the topologically optimized structure as the optimization target; the objective function includes the macro density of each frame unit of all frame units and the constraints of the entire volume of the topologically optimized structure; the macro density includes variables mapped to the micro density of the frame material; the micro density is periodically distributed in the macro density.

[0063] It should be noted that in the process of mapping the micro density of the above frame material to the macro density, the idea of ​​homogenization is mainly used, that is, the microstructure is distributed in the macro structure, that is, the microstructure is a substructure of the macro structure, and the microstructure is periodically distributed in the macro structure. In this way, the macro displacement field is a function of the microstructure density. Since the scale of the microstructure is smaller than that of the macro structure, the microstructure can be equivalent to a stiffness tensor as the unit stiffness matrix calculation of the macro structure unit, which is the process from micro to macro.

[0064] In the process of topology optimization, the finite element model (FEM) is the core analysis tool for numerical simulation and evaluation of the mechanical behavior of material distribution in the design space. Topology optimization aims to find the optimal structural layout under the condition of meeting specific performance requirements (such as load-bearing capacity, stiffness, minimum weight, etc.) by changing the shape of the structure and the distribution of internal materials.

[0065] It should be noted that the above-mentioned physical properties are used to represent the physical properties of the frame, such as material, mechanical properties, etc.

[0066] Next, the above macro density is used to represent the macro structure of the frame unit. The above micro density is used to represent the micro structure of the frame unit.

[0067] In addition, the above objective function is ultimately to achieve the optimization of minimizing the average flexibility of the topology optimization structure.

[0068] It should also be noted that the constraints of the entire volume of the above-mentioned topological optimization structure are used to express the specified target volume fraction so that the final frame can achieve the required stiffness and strength requirements.

[0069] Specifically, in dual-scale optimization, the macro density of each frame unit of all frame units has a coupling relationship with the micro density. In the homogenization method, the size of the micro density of each frame unit of all frame units is often considered to be much smaller than the macro density size, and the micro density presents a periodic distribution in the macro density. Within the linear elastic range, the displacement field of the macro density can be expanded as:

[0070] u ∈ (x)=u0(x,y)+∈u1(x,y)+∈ 2 u2(x,y)+…,y=x / ∈ (1)

[0071] In the formula, u v (x) represents the displacement field of the macro density of each frame unit, x represents the global macro variable such as macro density, y represents the local micro variable such as micro density, u0(x, y), u1(x, y) and u2(x, y) represent the 0th order variation term, 1st order variation term and 2nd order variation term of the displacement field respectively, ∈ is a value much less than 1, which represents the ratio of the macro scale to the micro scale.

[0072] When considering the first-order term of the asymptotic expansion of the previous formula, the macroscopic stiffness tensor can be expressed as:

[0073]

[0074] Where D H represents the macroscopic stiffness tensor, Ω m represents the area of ​​the microscopic density of each frame unit (under two-dimensional conditions), D represents the elastic tensor of local variation, represents the linearly independent unit strain field, ε(u m ) represents the unknown strain field of microscopic density, which can be solved by the linear elastic equilibrium equation:

[0075]

[0076] Where, ν m The virtual displacement of each frame unit’s microscopic density, H per represents the displacement space, indicates existence, d indicates the dimension of the vector space, Represents a d-dimensional vector space as the range of the displacement space.

[0077] The above finite element model can be expressed as:

[0078]

[0079] In the formula, St represents the constraint condition, and denote macro design variables and micro variables respectively, represents the macroscopic density, represents the microscopic density, i represents the i-th unit of all macroscopic frames (also called frame units), j represents the j-th unit of the microstructure, N represents the total number of frame microscopic units, M represents the total number of frame macroscopic units, T represents the transpose of the force vector, Min represents the minimization of the average flexibility, represents the objective function, C represents the optimization target of minimizing the average flexibility of the topology optimization structure, F and U represent the external force vector and node displacement vector of the macro density, and the macro density Ω m is discretized into M macro units, and the micro density Ω m is discretized into N microscopic units. M With G M are the volume constraints of the macro unit and the micro unit, V M With V m are the target volume fractions of macroscopic density and microscopic density, respectively, and υ0 is the solid volume fraction of each iteration process. In order to eliminate the singularity in the calculation process, and It is regarded as a positive number infinitely close to 0, and the value in this paper is 0.001.

[0080] Step 130, based on the objective function and the global optimization algorithm, globally optimize the average flexibility corresponding to each frame unit of all frame units to obtain a topologically optimized structure after the frame material is optimized; the optimized topologically optimized structure includes the minimized average flexibility of the topologically optimized structure.

[0081] Topology optimization is essentially a process of determining how materials are distributed in a given design space through mathematical optimization methods in order to maximize or minimize certain performance indicators (such as stiffness, strength, weight, etc.). It does not directly give the final geometry or specific structural design, but provides a structural layout (Topology) based on optimization objectives and constraints.

[0082] The global priority algorithm in this article is used to optimize all frame units globally. The above global optimization algorithm may include but is not limited to one or more of a particle swarm optimization (PSO) algorithm, an ant colony algorithm (ACO) and a genetic algorithm.

[0083] In the embodiment of the present application, the adaptability and flexibility of the frame design are enhanced: the baggage tractor works in different ground conditions and weather environments, and the frame must have good adaptability. The present application uses topology optimization technology to adjust the frame according to different working conditions (such as wet ground, snowy weather, or high temperature environment), optimize its structural shape and material selection, and ensure that the frame can still maintain efficient operation under changing operating environments.

[0084] For example, specific parts of the frame can be optimized according to the changes in dynamic loads to enhance its fatigue resistance and impact resistance, thereby improving its flexibility and adaptability under different working conditions. Through such optimized design, the frame can better cope with frequent starts and stops, different load distributions and complex ground conditions, improving the overall operating efficiency of the tractor.

[0085] Combination Figure 1 As shown, the above step 130 can be implemented by using at least any one of the following optional embodiments:

[0086] In an optional first embodiment of the above step 130, in the first step, the average flexibility of each frame unit is determined relative to the change of the macroscopic density of the material.

[0087] The second step is to obtain the sensitivity of each frame unit of all frame units based on the change; the greater the sensitivity of each frame unit, the greater the macro density of each frame unit.

[0088] The sensitivity of each of the above-mentioned frame units is the congestion degree of each of the frame units calculated by the genetic algorithm.

[0089] The third step is to use a global optimization algorithm to globally optimize the average flexibility corresponding to each frame unit of all frame units according to the size of the sensitivity to obtain the optimized topology optimization structure.

[0090] In this regard, the first step may include, but is not limited to, taking the derivative of the macroscopic density corresponding to each frame unit of all frame units to obtain the change.

[0091] Specifically, for topology optimization, sensitivity analysis is needed to guide the search direction of the global optimization algorithm. At the macro scale, the average flexibility depends on the material properties of each frame unit in the structure. Let F be a load vector that is independent of the design variables, so the i-th frame unit The derivative of the average flexibility with respect to the change in material density is:

[0092]

[0093] Among them, U represents the displacement matrix of the frame structure, K- 1 represents the inverse matrix of the global stiffness matrix of the frame structure, and K represents the global stiffness matrix of the frame structure.

[0094] For the proposed optimization framework, the sensitivity expression of the average compliance to the microscopic variables is:

[0095]

[0096] Among them, u i represents the unit displacement of the i-th unit of the macrostructure, represents the unit displacement transpose of the i-th unit of the macrostructure, k i represents the unit stiffness matrix of the i-th unit of the microstructure, B represents the strain displacement matrix, Ω i represents the design space of the macrostructure, B T represents the transpose of the strain displacement matrix, D H represents the homogenized elastic tensor, and T in this paper represents the transpose.

[0097] The elastic tensor of microscopic density is expressed by the material interpolation scheme in the improved SIMP (Solid Isotropic Material with Penalization) method, and the specific form is:

[0098] D m =[c+(x m ) p (1-c)]D0 (7)

[0099] Where D m represents the interpolated elastic tensor for microscopic density, p represents the penalty coefficient, c = 1e-9, to avoid the singularity of the matrix, p is the penalty factor, and D0 is the constitutive elastic tensor of the material. For microscopic density, the homogenized elastic tensor can be expressed as:

[0100]

[0101] Where D H represents the homogenized elastic tensor, Represents the micro-unit test displacement field.

[0102] For micro variables The derivative is:

[0103]

[0104] Where p represents the penalty factor, represents the actual displacement of the jth unit of the microstructure, represents the design space of the jth unit of the microstructure.

[0105] The sensitivity of the average flexibility is:

[0106]

[0107] Among them, Ω i represents the design space of the macrostructure, represents the actual displacement of the jth unit of the microstructure, u i represents the unit displacement of the i-th unit of the macrostructure.

[0108] In the BESO (Bi-directional Evolutionary Structural Optimization, genetic algorithm structural optimization method) method, the sensitivity α of the j-th frame unit is j It can be defined as:

[0109]

[0110] The sensitivity filtering strategy is used to correct the sensitivity. For example, the sensitivity of the empty unit is first derived through the sensitivity value of the frame unit (also called the physical unit) within the filtering radius, and then the empty unit is deleted. This not only effectively reduces the checkerboard phenomenon, but also derives the sensitivity of the empty unit through the sensitivity value of the physical unit within the filtering radius, thereby realizing the addition and deletion of units. The filtering strategy adopted in this paper is shown in the following formula:

[0111]

[0112] In the formula, α j represents the sensitivity of the j-th frame unit after filtering, represents the sensitivity of the kth frame unit, and N represents the total number of units that need to be filtered. jk ) is the weight factor, r jk represents the distance from the center of node j to the center of node k. The weight factor can be expressed as:

[0113]

[0114] Among them, rmin Indicates the filter radius.

[0115] The sensitivity value is corrected by using a convergence strategy that considers historical iteration information. The specific operation is to calculate the average sensitivity of the current iteration and the previous iteration. Use this as the updated value for sensitivity:

[0116]

[0117] In the formula, α j , g With α j , g-1 are the sensitivity values ​​of the g-th and g-1-th iterations of the j-th frame unit, respectively, g represents the current iteration in the number of iterations, and g-1 represents the previous iteration in the number of iterations. Finally, the influence of each frame unit on the structural performance is measured based on this sensitivity value.

[0118] The third step may include but is not limited to the following steps (1) to (4):

[0119] (1) Using a global optimization algorithm, sort the sensitivity of each frame unit of all frame units according to the sensitivity, and determine a first frame area corresponding to high sensitivity and a second frame area corresponding to low sensitivity; wherein the first frame area is retained and the second frame area is deleted. wherein the first frame area is a frame area formed by a plurality of frame units having a sensitivity greater than a preset threshold; and the second frame area is a frame area formed by a plurality of frame units having a sensitivity less than a preset threshold.

[0120] The above step (1) can be implemented by using at least one of the following optional implementation methods:

[0121] In an optional implementation, the sensitivity of each frame unit of all the frame units is sorted from high to low, so as to determine a first frame area corresponding to high sensitivity and a second frame area corresponding to low sensitivity.

[0122] In another optional implementation, the sensitivity of each frame unit of all the frame units is sorted from low to high, so as to determine a first frame area corresponding to high sensitivity and a second frame area corresponding to low sensitivity.

[0123] (2) Determine whether the volumes of all first frame regions meet the constraint conditions.

[0124] (3) If achieved, the optimized topological optimized structure is obtained.

[0125] (4) If not, continue global optimization using a global optimization algorithm until the volumes of all first frame regions meet the constraint conditions, and then obtain the optimized topology optimized structure.

[0126] Use global optimization algorithm to continue global optimization. You can use particle swarm optimization algorithm or ant colony algorithm for global optimization.

[0127] Example 1: First, each frame unit of all frame units is used as each particle. Secondly, based on each particle, a particle swarm optimization algorithm is used to perform global optimization to achieve information exchange between particles to help global search.

[0128] Example 2: First, each frame unit of all frame units is used as each ant. Second, based on each ant, the ant colony algorithm is used to perform global optimization to guide the search process between multiple paths. In this way, the efficiency is higher and the accuracy is higher.

[0129] In an optional second embodiment of step 130, the global optimization algorithm includes a genetic algorithm; each frame unit is an independent individual of the population corresponding to the genetic algorithm; and the population size of the frame units remains unchanged. The genetic algorithm may include, but is not limited to, a standard genetic algorithm and a non-dominated sorting genetic algorithm (NSGA). ， Non-dominated Sorting Genetic Algorithm) one or more.

[0130] In this regard, during the iterative cycle of the genetic algorithm, the current iteration performs the following steps A to C:

[0131] Step A, performing the crossover operation and mutation operation of the genetic operation on each frame unit in turn to obtain the inherited frame unit.

[0132] It should be noted that, in the iterative cycle process of the genetic algorithm in the above step A, as the number of iterations increases, the crossover rate and the mutation rate are gradually punished to extreme values, for example, the crossover rate is gradually punished to a maximum value, and at the same time, the mutation rate is gradually punished to a minimum value, which is described in detail as follows:

[0133] In each iteration step, for the first frame region, as the sensitivity decreases from high to low, the crossover rate decreases each time and iterates to a minimum value, and the first frame region is retained; and

[0134] For the second frame region, as the sensitivity decreases from high to low, the mutation rate increases to a maximum value in each iteration, and the second frame region is deleted.

[0135] In addition, the above step A may include the following steps [1] and [2]:

[0136] [1] Randomly select any two frame units as parents in the design domain, randomly select any two crossover positions, perform a crossover operation on the parents based on the crossover rate, and obtain the crossover frame units; the crossover operator of the crossover operation is executed according to the two-point crossover operator.

[0137] The specific process of the above cross operation is as follows: Step 1) to Step 3):

[0138] 1) Parent selection: Select two parent individuals from the current population based on the frame unit through some selection mechanism (such as roulette selection, tournament selection, etc.).

[0139] 2) Determine whether to perform crossover: Determine whether to perform crossover operation based on the crossover rate.

[0140] ①. Generate a random number (assuming this number is in the range [0, 1]).

[0141] ② If the random number is less than or equal to the crossover rate (random_number<=crossover_rate), a crossover operation is performed; otherwise, the parent individual is directly copied to the next generation.

[0142] ③. The crossover rate usually takes a value between 0 and 1. For example, if the crossover rate is 0.8, there is an 80% probability that a crossover operation will be performed and a 20% probability that no crossover will be performed.

[0143] 3) Perform crossover: If a crossover operation is performed, the genes of the parent individuals are combined according to a specific crossover method (such as single-point crossover, two-point crossover, uniform crossover, etc.) to generate a frame unit.

[0144] Return to the frame unit: the generated child will replace part of the original parent's position and enter the next generation.

[0145] [2] Perform a mutation operation on the binary code of each crossed frame unit according to the mutation rate to obtain a mutated frame unit as the inherited frame unit; the mutation operator of the mutation operation adopts basic bit mutation.

[0146] The specific process of the above mutation operation is as follows: Step 1] to Step 3]:

[0147] 1], Select individuals: The mutation operation is usually performed in the frame unit generated by crossover, or sometimes directly in the parent generation. Each frame unit may be mutated.

[0148] 2] Determine whether to mutate: For each individual, decide whether to mutate the gene based on the mutation rate.

[0149] 1. For each gene locus, generate a random number (assumed to be in the range of [0, 1]).

[0150] 2. If the random number is less than the mutation rate (random_number<=mutation_rate), the gene locus mutates (for example, the gene value is reversed, changed to another value, etc.); if the random number is greater than the mutation rate, the gene locus remains unchanged.

[0151] 3. The mutation rate is usually low, such as 0.01, 0.05, etc., because the mutation is to maintain the diversity of the population based on the frame unit and avoid premature convergence to the local optimal solution.

[0152] 3] Perform mutation: Perform mutation operations on the individual's genes according to the mutation rate. The mutation method can be to randomly change a certain value of the gene (such as changing the 1 of a binary gene to 0, or adding or subtracting the value of an integer gene within a certain range, etc.).

[0153] Return the mutated individual: The mutated individual will become part of the next generation and enter the frame unit-based population.

[0154] The global optimization algorithm of the related art directly adds or deletes materials according to the size of the unit sensitivity. In the embodiment of the present application, the diversity of the frame units can be increased by crossover operations and mutation operations on the frame units.

[0155] In the embodiment of the present application, the dual-scale topology optimization process can automatically adjust the search strategy, such as crossover, mutation rate and selection mechanism, according to the progress of the optimization process to achieve efficient optimization. Such algorithm design is the key to solving large-scale and complex optimization problems.

[0156] Step B, determining the sensitivity of the inherited frame unit.

[0157] Step C, sorting the sensitivities of the inherited frame units, selecting a specified number of frame units according to the sorting of sensitivities, determining a first frame region corresponding to high sensitivity, and determining a second frame region corresponding to low sensitivity; wherein the first frame region is retained and the second frame region is deleted; the specified number is the product of the evolution rate of the genetic algorithm and the population size; the evolution rate decreases as the number of iterations increases.

[0158] Step D: Determine whether the volumes of all first frame regions meet the constraint conditions.

[0159] Step E: If not, return to execute steps A to D for loop iteration.

[0160] Step F: If yes, and the average flexibility corresponding to each frame unit meets the convergence condition, the iterative process is terminated to obtain the optimized topology optimized structure.

[0161] Furthermore, the average flexibility corresponding to each frame unit satisfies the convergence condition, which may include but is not limited to the following: the convergence error between the average flexibility corresponding to each frame unit in the current iteration and the average flexibility corresponding to each frame unit in the previous iteration remains unchanged or is less than a target convergence error, such as less than 0.01.

[0162] In this regard, during the topology optimization process, the redistribution of materials and the update calculation of the objective function are continuously performed. The optimization process is terminated when the optimization reaches the specified target volume fraction and meets the following convergence conditions. The convergence error is mainly calculated by the objective function value.

[0163]

[0164] In the formula, C represents the objective function value, k represents the current number of iterations, and E is 5, which means that the optimization results of 5 consecutive iterations have a certain stability. τ is the allowed convergence error, k represents the current number of iterations, k-j+1 represents the current number of iterations minus the loop index plus 1, kE-j+1 represents the current number of iterations minus the error iteration number E minus the loop index plus 1, and E represents the selected error iteration number.

[0165] In the embodiments of the present application, the global search capability of the genetic algorithm is combined with the fine control of the dual-scale topology optimization, which can effectively explore and optimize the macroscopic morphology and microscopic features of the structural design to adapt to complex load conditions and manufacturing constraints.

[0166] Moreover, the above-mentioned genetic algorithm is capable of performing global search in a wider design space, which helps to generate more diversified and innovative optimization solutions for the frame unit, thereby providing better design solutions than traditional methods and increasing the diversity of design solutions.

[0167] Continue to combine Figure 1 As shown, the above step C can be implemented in the following optional manner:

[0168] In an optional manner, 1) starting from the direction of maximum sensitivity to the direction of decreasing sensitivity, a specified number of frame units are selected as the first frame area corresponding to high sensitivity. And, 2) starting from the direction of minimum sensitivity to the direction of increasing sensitivity, a specified number of frame units are selected as the second frame area corresponding to low sensitivity.

[0169] In another optional manner, the global optimization algorithm may include, but is not limited to, a non-dominated sorting genetic algorithm. Using the non-dominated sorting genetic algorithm, the sensitivities of the inherited frame units are non-dominatedly sorted, and the frame units at the non-dominated frontier are determined as the first frame region, and the frame region corresponding to the sensitivities of the inherited frame units other than the first frame region is determined as the second frame region. In this way, the sensitivities at the same frontier level are crossed in the genetic operation.

[0170] It should be noted that the above NSGA-II (Non-dominated Sorting Genetic Algorithm II) maintains the diversity of the population by dividing the population into different non-dominated layers. When comparing each individual, not only its target value is considered, but also whether other individuals are dominated. Non-dominated sorting divides individuals into different levels (frontiers), ensuring that the competition between individuals within each front is small and there is a clear relationship between the advantages and disadvantages of each front.

[0171] In the embodiment of the present application, in order to merge the dual-scale optimization algorithm based on BESO (structural optimization method of genetic algorithm) with the NSGA-II algorithm, the topology optimization model needs to be reconstructed. The topology optimization problem requires discretization of units and finite element analysis, while NSGA-II analyzes individuals in the population. Therefore, in the fusion process, it is necessary to establish a mathematical model in which the two algorithms are mapped to each other to establish a corresponding mathematical model. First, the structure is divided into discrete units. According to the set grid seed size, a certain number of overall units can be obtained. Each frame unit is regarded as an independent individual for analysis in the NSGA-II algorithm. The number of populations remains unchanged during the iteration process to ensure that the number of individuals in the entire population is always equal to the number of frame units in the design domain. The uniform elastic stiffness tensor and its derivative of microscopic density are derived according to the homogenization theory. The results obtained by calculating the microscopic density guide the assembly of the macroscopic density unit stiffness matrix to achieve the coupling of macroscopic density and microscopic density.

[0172] Through the calculation of the objective function, the entire population based on the frame unit is guided to perform fast non-dominated sorting. Individuals with higher levels in the distribution mean higher sensitivity values, which are more likely to be retained in subsequent iterations. Individuals with lower levels mean lower sensitivity, and will be gradually eliminated according to the rule of survival of the fittest. This algorithm also incorporates the BESO operator to implement the material addition and deletion strategy. During the optimization process, BESO can convert mistakenly deleted materials back into solid materials, ensuring the stability of the optimization results.

[0173] Compared with the global optimization algorithm of the related art that directly adds and deletes materials according to the size of the unit sensitivity, in the embodiment of the present application, after the introduction of NSGA-II, the binary coding of the individual is to perform crossover and mutation operations with a certain probability, and the individuals with low sensitivity will have a certain probability to be saved in this iteration cycle to prevent the accidental deletion of the unit. The main contents of the proposed algorithm are introduced in detail below.

[0174] 1. Algorithm initialization

[0175] like Figure 2 As shown in the figure, each discrete unit is represented by a binary string of length 6. The initial string is defined in this article as an array of all "1", which means that all units are solid materials. The change of binary characters is only performed in the crossover mutation operation. During the optimization iteration process, the array of all 0s (empty units) means the units to be deleted in this iteration. The individuals composed of binary numbers "0" and "1" (called solid units, and the frame units have some materials) will also be saved in this iteration. Deleting only the individuals in the array of all "0" can ensure that several units with lower sorting levels are retained in the optimization process and enter the next iteration cycle, which can effectively prevent the accidental deletion of units.

[0176] It is worth noting that the deleted elements do not mean complete deletion. They also participate in the calculation of objective functions and sensitivity in the optimization iterations. There is a certain chance that they will be transformed from hollow elements to solid elements in the subsequent optimization iterations.

[0177] 2. Objective function calculation and sensitivity value ranking

[0178] After initializing the optimization model, the stiffness matrix and elastic tensor matrix of the micro-unit are calculated and assembled according to the homogenization theory. The objective function value and the corresponding sensitivity are calculated and used as the basis for fast non-dominated sorting. The sensitivity between each frame unit is compared to establish the dominant and non-dominated relationship and sort the levels. For individuals of the same level, the sensitivity R is calculated, and the expression is:

[0179]

[0180] In the formula, α max , α min Respectively represent the maximum and minimum sensitivity in this iteration cycle, α i+1 represents the sensitivity calculated for the i+1th frame unit, α i-1 Represents the sensitivity calculated for the i-1th frame unit.

[0181] 3. Crossover mutation operator

[0182] The crossover mutation operator plays a very important role in NSGA-II, which enriches the diversity of the population. The crossover mutation operator proposed in this paper adopts a semi-random method, which avoids the checkerboard phenomenon and non-convergence under the full random operator to a certain extent. Different from the related technology NSGA-II algorithm, this algorithm does not change the population size during the iteration process, because in topology optimization, the overall number of units of the structure is unchanged, and there is a transformation between solid units and void units. In order to ensure the diversity of the population, the crossover mutation operator only has a transformation between 0 and 1. The crossover operator is executed according to the two-point crossover operator, and the mutation operator adopts basic bit mutation. Each individual consists of a binary string of length 6, and a certain mutation rate is set for each binary code. In order to ensure the rapid convergence of the algorithm, this paper proposes a double penalty operator.

[0183] The double penalty operator is mainly used in the crossover mutation operator. First, according to the sensitivity threshold calculated in each generation, the population is divided into high-sensitivity areas and low-sensitivity areas. Two individuals are randomly selected as parents in the design domain, and two crossover positions are randomly selected to perform crossover operations on the parents. In order to accelerate convergence, the algorithm in this paper gradually penalizes the crossover rate to a certain maximum value as the number of iterations increases during the optimization process.

[0184] The mutation operator mutates the binary code of each individual according to a certain mutation rate. Here, a double penalty operator for the mutation rate is proposed. When the optimization algorithm is iterated, as the number of iterations increases, the crossover rate and mutation rate are gradually penalized to a certain maximum value, and their maximum value is limited to 1. c is the crossover rate, p c , min is the minimum crossover rate, p c , max is the maximum crossover rate, p m is the mutation rate, p m，min is the minimum mutation rate, p m，max is the maximum mutation rate, prg is the index for controlling the crossover mutation rate, and ph is the penalty coefficient, which is 3 in this paper.

[0185] p c =p c , min +(p c , max -p c , min )prg ph (17)

[0186] p m =p m，min +(p m，max -p m，min )prgph (18)

[0187] In each iteration step, for high-sensitivity areas, as the sensitivity decreases from high to low, the mutation rate (the probability of binary code mutation to 1) is gradually punished to a minimum value. At this time, individuals with higher sensitivity have a greater probability of having more 1s and are more likely to be retained during the optimization process.

[0188] For low-sensitivity areas, as the sensitivity decreases from high to low, the mutation rate (the probability of binary code mutation to 0) is gradually punished to a certain maximum value. At this time, individuals with lower sensitivity have a greater probability of having more 0s and are more likely to be deleted during the optimization process. It is worth noting that individuals of the same level have the same mutation rate. Through the double penalty mechanism, the rapid convergence of the optimization process and a clearer and more stable topological structure are guaranteed.

[0189]

[0190] In the formula, p high，m is the variation rate of the frame unit contained in the first frame area (also called the high sensitivity area), p low，m is the variation rate of the frame unit contained in the second frame area (also called the low-sensitivity area), r high，max is the maximum value of the individual level in the high sensitivity area, r high，min is the minimum value of the individual level in the high sensitivity area. Similarly, r low，max With r low，min are the maximum and minimum values ​​of the individual levels in the low sensitivity area. high，m The initial value is p m，max , p low，m The initial value is p m，min .

[0191] 4. Generate the final manufacturable frame structure

[0192] The topology optimized structure after frame material optimization is subjected to shape optimization, size optimization and frame geometry modeling to generate the final manufacturable frame structure.

[0193] Combination Figure 3 As shown, the above-mentioned final manufacturable frame structure may be applied to the frame of a luggage tractor, and this article is in the optimized design of the frame of the luggage tractor.

[0194] Specifically, the optimized frame structure can not only reduce the overall weight, but also maintain the high load-bearing capacity and fatigue resistance of the frame. The less sensitive areas will gradually reduce the use of materials through the penalty mechanism of mutation rate, thereby achieving lightweight design, while the more sensitive areas will retain more materials to enhance their load-bearing and impact resistance.

[0195] In the embodiments of the present application, both theoretical optimization goals and actual manufacturing limitations are taken into account, so that the final design is both efficient and easy to manufacture, increasing the practical application value of the design scheme, thereby improving the practicability and manufacturability of the design scheme.

[0196] Typically, topology optimization describes the optimization problem through an objective function and constraints. The objective function is usually to minimize the weight of the structure, maximize the stiffness, or improve specific performance (such as energy absorption, heat dissipation). In the genetic algorithm or topology optimization process, the conversion of hollow elements and solid elements is achieved through optimization iterations. Formulatively, the optimization goal drives the change of the structure through sensitivity analysis (usually involving stress, strain, or material distribution).

[0197] In actual manufacturing, it is necessary to consider the processability of the material and the feasibility of assembly. For example, complex cavities or overly small structures may be difficult to manufacture or too costly in reality. Therefore, the results of topology optimization often incorporate manufacturing constraints, such as minimum feature size restrictions, manufacturing process constraints (such as the resolution of 3D printing or the process limitations of traditional casting). These manufacturing constraints are reflected by adjusting the design variables and penalty functions in topology optimization. For example, a certain volume of material is forced to be retained to avoid structures that are too small or cannot be manufactured.

[0198] In theoretical topology optimization, void elements (i.e., areas without material) can optimize the lightweighting of structures. However, in actual manufacturing, completely voids or extremely small gaps increase the difficulty of manufacturing. For example, void areas may lead to insufficient material support, unstable structures, or high processing costs during manufacturing. Converting these void elements into solid elements can solve these problems, making the design more manufacturable.

[0199] Solid elements refer to the parts that retain actual material in the optimized structure. These areas are easy to process, assemble and form during manufacturing. Solid elements reduce complex hollow structures and avoid common problems in the manufacturing process, such as difficult mold making and difficult support material removal. For the tractor frame, solid elements ensure that the key parts of the frame have sufficient strength and stability, while also ensuring operability during the manufacturing process.

[0200] The design of the solid unit is directly related to the feasibility of manufacturing. In the tractor frame, the solid unit often represents the main load-bearing part of the frame. These parts not only need to be strong enough to bear the weight of the luggage cart and the towing load, but also need to conform to the actual processing technology. For example, using traditional welding, casting or modern 3D (three-dimensional) printing technology, the solid unit can effectively support the processing requirements in the manufacturing process, such as the strength requirements of the welded joints or the stability of the mold during the casting process.

[0201] Through optimized design, the solid element is closely integrated with the manufacturing process. For example, the designed solid element can be directly suitable for casting or welding, and the shape and thickness can be adjusted to meet the capabilities of the manufacturing equipment. For the tractor frame, this means solid design at key stress-bearing parts to ensure the strength of the structure when towing heavy objects, while reducing the complexity of the manufacturing and assembly process.

[0202] Based on the same application concept as the above method, the present application embodiment also provides a topology optimization device for frame materials, such as Figure 4 As shown, the topology optimization device may include the following modules:

[0203] A meshing module 31 is used to mesh the geometric model of the frame according to a preset mesh seed size to obtain a plurality of discrete frame units;

[0204] The model building module 32 is used to configure the frame material for each of the frame units of all the frame units and build a finite element model; the finite element model includes: physical characteristics, and an objective function with the minimized average flexibility of the topological optimization structure as the optimization target; the objective function includes the macro density of each frame unit of all the frame units and the constraint conditions of the entire volume of the topological optimization structure; the macro density includes variables mapped by the micro density of the frame material; the micro density presents a periodic distribution in the macro density;

[0205] The global optimization module 33 is used to perform global optimization on the average flexibility corresponding to each frame unit of all frame units based on the objective function and the global optimization algorithm to obtain a topological optimization structure after the frame material is optimized; the optimized topological optimization structure includes the minimized average flexibility of the topological optimization structure.

[0206] In the embodiments of the present application, a genetic algorithm is combined with a dual-scale topology optimization method, which takes into account the advantages of topology optimization in determining the most effective distribution of materials to improve structural performance and the powerful ability of genetic algorithms in global search to solve complex optimization problems. This method optimizes the structural design by optimizing at both the macro and micro scales, taking into account both the overall structural layout and fine-tuning local details. This dual-scale method is particularly suitable for advanced structures that require high performance and material utilization because it can explore possible solutions in a wider design space and provide innovative and practical design solutions. Although the integration of genetic algorithms into dual-scale topology optimization faces challenges in computational cost and optimization efficiency, with the advancement of computing technology and the continuous development of algorithm optimization technology, the method of the embodiments of the present application is expected to promote innovation in design and manufacturing in many fields such as aerospace, automobile manufacturing, and mechanical design.

[0207] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and the same technical effect can be achieved, which will not be repeated here.

[0208] An embodiment of the present application provides an electronic device, comprising the above-mentioned topology optimization device for the frame material.

[0209] The method provided in the embodiment of the present application can be applied to electronic devices. Specifically, the electronic device can be: a desktop computer, a portable computer, an intelligent mobile terminal, a server, a PDA (Personal Digital Assistant) and a handheld terminal, etc. Among them, the PDA can include an industrial PDA and a consumer PDA. Any electronic device that can implement the embodiment of the present application belongs to the protection scope of the present application and is not limited here.

[0210] Figure 5 Shown is a schematic structural diagram of an electronic device 50 provided in an embodiment of the present application.

[0211] like Figure 5 As shown, the electronic device 50 includes one or more processors 51 for implementing the topology optimization method of the frame material as described above.

[0212] In some embodiments, the electronic device 50 may include a storage medium 59. For example, the computer-readable storage medium may store a program that can be called by the processor 51, and may include a non-volatile storage medium. In some embodiments, the electronic device 50 may include a memory 58 and an interface 57. In some embodiments, the electronic device 50 may also include other hardware according to actual applications.

[0213] The computer-readable storage medium of the embodiment of the present application stores a program thereon, which, when executed by the processor 51, is used to implement the topology optimization method for the frame material described above.

[0214] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0215] The present application also provides a computer program, which is stored in a computer-readable storage medium, such as Figure 5 The computer program is stored in the storage medium 59, and when the processor executes the computer program, it causes the processor 51 to execute the method described above.

[0216] The present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer-readable storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0217] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

[0218] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the phrase "includes a ..." defines an element, and does not exclude the presence of other identical elements in the process, method, commodity or device including the element.

Claims

1. A topological optimization method for a vehicle frame material, characterized in that: include: According to the preset grid seed size, the geometric model of the frame is meshed to obtain multiple discrete frame units; configuring frame materials for each of the frame units of all the frame units, and constructing a finite element model; The finite element model includes: physical characteristics, and an objective function with minimization of average flexibility of the topology optimized structure as an optimization target; The objective function includes the macroscopic density of each frame unit of all frame units and the constraint conditions of the entire volume of the topology optimization structure; the macroscopic density includes variables mapped by the microscopic density of the frame material; the microscopic density presents a periodic distribution in the macroscopic density; Based on the objective function and the global optimization algorithm, the average flexibility corresponding to each frame unit of all frame units is globally optimized to obtain a topological optimization structure after the frame material is optimized; the optimized topological optimization structure includes the minimized average flexibility of the topological optimization structure.

2. The topology optimization method of the frame material according to claim 1, characterized in that: Based on the objective function and the global optimization algorithm, the average flexibility corresponding to each frame unit of all frame units is globally optimized to obtain a topological optimization structure after the frame material is optimized, including: Determine the average compliance of each frame unit relative to the macro density of the material; Based on the change, the sensitivity of each frame unit of all frame units is obtained; the greater the sensitivity of each frame unit, the greater the macro density of each frame unit; According to the size of the sensitivity, a global optimization algorithm is used to globally optimize the average flexibility corresponding to each frame unit of all frame units to obtain the optimized topology optimization structure.

3. The topology optimization method of the frame material according to claim 2, characterized in that: Determining the average flexibility of each frame unit relative to the change in the macroscopic density of the material includes: Derivative the macroscopic density corresponding to each frame unit of all frame units to obtain the change; According to the sensitivity, the global optimization algorithm is used to globally optimize the average flexibility corresponding to each frame unit of all frame units to obtain the optimized topology optimization structure, including: Using a global optimization algorithm, sorting the sensitivity of each frame unit of all frame units according to the size of the sensitivity, and determining a first frame area corresponding to a high sensitivity and a second frame area corresponding to a low sensitivity; wherein the first frame area is retained and the second frame area is deleted; Determining whether the volumes of all first frame regions meet the constraint conditions; If it is achieved, the optimized topological optimized structure is obtained; The first frame area is a frame area formed by a plurality of frame units having a sensitivity greater than a preset threshold; the second frame area is a frame area formed by a plurality of frame units having a sensitivity less than a preset threshold.

4. The topology optimization method of the frame material according to claim 1, characterized in that: The global optimization algorithm includes a genetic algorithm; each frame unit is an independent individual of the population corresponding to the genetic algorithm; the population number of the frame units remains unchanged; Based on the objective function and the global optimization algorithm, the average flexibility corresponding to each frame unit of all frame units is globally optimized to obtain a topological optimization structure after the frame material is optimized, including: During the iterative cycle of the genetic algorithm, the current iteration executes the following steps A to C: Step A, performing a crossover operation and a mutation operation of a genetic operation on each frame unit in turn to obtain a genetic frame unit; Step B, determining the sensitivity of the inherited frame unit; Step C, sorting the sensitivities of the inherited frame units, selecting a specified number of frame units according to the sorting of the sensitivities, determining a first frame region corresponding to high sensitivity, and determining a second frame region corresponding to low sensitivity; wherein the first frame region is retained and the second frame region is deleted; the specified number is the product of the evolution rate of the genetic algorithm and the population size; the evolution rate decreases as the number of iterations increases; Step D, determining whether the volumes of all first frame regions meet the constraint conditions; If not, return to step A to step D for loop iteration; If so, and when the average flexibility corresponding to each frame unit satisfies the convergence condition, the iterative process is terminated to obtain the optimized topology optimization structure.

5. The topology optimization method of the frame material according to claim 4, characterized in that: The step C comprises: starting from the direction of maximum sensitivity to the direction of decreasing sensitivity, selecting a specified number of frame units as a first frame area corresponding to high sensitivity; starting from the direction of minimum sensitivity to the direction of increasing sensitivity, selecting a specified number of frame units as a second frame area corresponding to low sensitivity.

6. The topology optimization method for a vehicle frame material according to claim 4, characterized in that: The global optimization algorithm includes a non-dominated sorting genetic algorithm; the step C includes: using the non-dominated sorting genetic algorithm to non-dominated sort the sensitivities of the inherited frame units, determining the frame units on the non-dominated frontier as the first frame area and the frame area corresponding to the sensitivities of the inherited frame units other than the first frame area as the second frame area.

7. The topology optimization method for a vehicle frame material according to claim 4, characterized in that: The step A comprises: Randomly select any two frame units as parents in the design domain, randomly select any two crossover positions, perform a crossover operation on the parents based on the crossover rate, and obtain the crossover frame units; the crossover operator of the crossover operation is executed according to the two-point crossover operator; A mutation operation is performed on the binary code of each crossed frame unit according to the mutation rate to obtain a mutated frame unit as the inherited frame unit; the mutation operator of the mutation operation adopts basic bit mutation.

8. The topology optimization method for a vehicle frame material according to claim 4, characterized in that: The iterative cycle process of the genetic algorithm includes: In each iteration step, for the first frame region, as the sensitivity decreases from high to low, the crossover rate decreases each time and iterates to a minimum value, and the first frame region is retained; and For the second frame region, as the sensitivity decreases from high to low, the mutation rate increases to a maximum value in each iteration, and the second frame region is deleted.

9. A topology optimization device, characterized in that: A method for topological optimization of a vehicle frame material according to any one of claims 1 to 8, wherein the topological optimization device comprises: A meshing module, used to mesh the geometric model of the frame according to a preset mesh seed size to obtain a plurality of discrete frame units; A model building module is used to configure a frame material for each of all frame units and build a finite element model; the finite element model includes: physical characteristics, and an objective function with the minimized average flexibility of the topology optimization structure as an optimization target; the objective function includes the macro density of each frame unit of all frame units and the constraint conditions of the entire volume of the topology optimization structure; the macro density includes variables mapped by the micro density of the frame material; the micro density presents a periodic distribution in the macro density; A global optimization module is used to perform global optimization on the average flexibility corresponding to each frame unit of all frame units based on the objective function and the global optimization algorithm to obtain a topological optimization structure after the frame material is optimized; the optimized topological optimization structure includes the minimized average flexibility of the topological optimization structure.

10. An electronic device, characterized in that: The method comprises one or more processors for implementing the topology optimization method of the frame material according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: A program is stored thereon, and when the program is executed by a processor, the topology optimization method for the frame material as described in any one of claims 1-8 is implemented.

Images