Structural configuration generation method, device and equipment and readable storage medium

Abstract

The invention discloses a structure configuration generation method, device and equipment and a readable storage medium, and is applied to the technical field of computers, and the method comprises the steps: carrying out the decoupling processing of a global grid model, and obtaining an order reduction model and a structure optimization model; applying an external load and a boundary condition to the reduced-order model, and solving to obtain a displacement load; and inputting the displacement load, the design constraint and the manufacturing constraint into a pre-trained structure configuration optimization model to obtain a generated optimal structure configuration. According to the method, the calculation scale of the structure configuration optimization model is greatly reduced, the calculation efficiency is improved, and the grid resolution in the to-be-optimized target structure area is improved, so that the high-fidelity structure configuration is generated; unnecessary materials can be reduced on the premise that the safety of a train is not influenced; on the premise of keeping the geometric features of the target structure region to be optimized similar, a high-fidelity structure configuration can be generated for any size, external load and boundary conditions, and the universality in a global parameter space is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a structure configuration generation method, device and equipment and readable storage medium. BACKGROUND

[0002] In order to obtain high-fidelity and clear structure configuration, the traditional structure configuration generation method often relies on improving the grid resolution in the finite element model, which undoubtedly leads to a sharp expansion of the calculation scale and a significant increase in resource consumption. In recent years, with the wide application of machine learning methods, a method has appeared that uses machine learning methods to replace the complex calculation process in the traditional iterative optimization to achieve rapid generation of structure configuration. However, this method has certain defects, such as when dealing with complex external load and boundary conditions, large grid scale, and the requirement of clear and smooth result boundary in actual engineering problems, the training sample generation process itself still has a very high calculation cost; and this method is usually strongly dependent on specific problem settings, and its actual application is also limited to the model type and boundary condition range used for training.

[0003] Therefore, how to improve the calculation efficiency and generalization ability of structure configuration generation and reduce the calculation cost is a technical problem that needs to be solved at present. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a structure configuration generation method, device, equipment and readable storage medium, which solves the problems of low calculation efficiency and poor generalization ability of structure configuration generation, and high calculation cost in the prior art.

[0005] To solve the above technical problems, the present application provides a structure configuration generation method, comprising: decoupling processing a global grid model to obtain a reduced-order model and a structure optimization model; applying external load and boundary conditions to the reduced-order model and solving to obtain displacement load; inputting the displacement load, design constraints and manufacturing constraints into a pre-trained structure configuration optimization model to obtain a generated optimal structure configuration; Wherein, the global grid model is a finite element model of the whole vehicle body structure; the reduced-order model includes a non-design space area, a load application area and a boundary constraint area; the structure optimization model includes a target structure area to be optimized and a grid transition area; the optimal structure configuration is the optimal structure configuration of the target structure area to be optimized.

[0006] Optionally, applying external load and boundary conditions to the reduced-order model and solving to obtain displacement load, comprising: applying the external load and the boundary condition to the reduced order model to obtain a total stiffness matrix of the reduced order model and equivalent nodal forces applied on each node in the reduced order model; calculating displacement vectors of all nodes in the reduced order model based on the total stiffness matrix and the equivalent nodal forces; selectively screening the displacement vectors of all nodes to obtain the displacement vectors at the interface between the reduced order model and the structural optimization model as the displacement load.

[0007] Optionally, before decoupling the global mesh model to obtain the reduced order model and the structural optimization model, the method further comprises: obtaining a training data set containing various training samples; the training samples include a reference optimal structure configuration and displacement loads for training, design constraints for training and manufacturing constraints for training corresponding to the reference optimal structure configuration; performing normalization processing on the displacement loads for training, the design constraints for training and the manufacturing constraints for training to obtain normalized input data; inputting the input data into an initial structure configuration optimization model to output a predicted optimal structure configuration; performing loss calculation according to the predicted optimal structure configuration and the reference optimal structure configuration to obtain a loss value; optimizing the initial structure configuration optimization model based on the loss value until the loss value meets a preset value or the number of iterations of training meets a maximum number of iterations, and stopping training to obtain the trained structure configuration optimization model.

[0008] Optionally, the loss calculation according to the predicted optimal structure configuration and the reference optimal structure configuration to obtain a loss value comprises: calculating a density field loss based on the predicted optimal structure configuration and the reference optimal structure configuration; calculating a physical constraint loss based on the predicted optimal structure configuration and the reference optimal structure configuration; calculating the loss value based on the density field loss and the physical constraint loss.

[0009] Optionally, obtaining a training data set containing various training samples comprises: determining external loads for training, boundary conditions for training, design constraints for training and manufacturing constraints for training according to a design space of a target structure to be optimized; generate a parameter combination set covering a parameter space based on the external load for training, the boundary condition for training, the design constraint for training and the manufacturing constraint for training; the parameter combination set contains N parameter combinations; apply the external load for training and the boundary condition for training to the reduced order model based on each of the parameter combinations, and solve the reduced order model to obtain the displacement load for training; obtain the optimal structure configuration of the reference based on the design constraint for training, the manufacturing constraint for training and the displacement load for training; construct the training data set based on the design constraint for training, the manufacturing constraint for training and the displacement load for training, and the optimal structure configuration of the reference.

[0010] Optionally, obtaining the optimal structure configuration of the reference based on the design constraint for training, the manufacturing constraint for training and the displacement load for training comprises: applying the displacement load for training to the boundary nodes of the structure optimization model to maximize the flexibility of the structure optimization model as an optimization objective; taking the material density of the elements in the target structure region to be optimized as a variable; taking the design constraint for training and the manufacturing constraint for training as constraint conditions; obtaining the optimal material density of each element in the target structure region to be optimized as the optimal structure configuration of the reference by using a gradient optimization algorithm to iteratively solve based on the optimization objective, the variable and the constraint conditions.

[0011] Optionally, after inputting the displacement load, the design constraint and the manufacturing constraint into the pre-trained structure configuration optimization model to obtain the generated optimal structure configuration, the method further comprises: when the optimal structure configuration does not meet the preset requirement, taking the external load, the boundary condition, the design constraint and the manufacturing constraint as new samples; training the trained structure configuration optimization model using the new samples until a training termination condition is met, and updating the trained structure configuration optimization model.

[0012] The application further provides a structure configuration generation device, comprising: a decoupling module configured to decouple a global grid model to obtain a reduced order model and a structure optimization model; a solving module configured to apply an external load and a boundary condition to the reduced order model and solve the reduced order model to obtain a displacement load; The generating module is configured to input the displacement load, the design constraint, and the manufacturing constraint into a pre-trained structure configuration optimization model to obtain a generated optimal structure configuration. The global grid model is a finite element model of the whole structure of the vehicle body. The reduced-order model includes a non-design space area, a load application area, and a boundary constraint area. The structure optimization model includes a target structure area to be optimized and a grid transition area. The optimal structure configuration is an optimal structure configuration of the target structure area to be optimized.

[0013] The application further provides a structure configuration generation device, which comprises: A memory configured to store a computer program. A processor configured to execute the computer program to implement the structure configuration generation method.

[0014] The application further provides a computer readable storage medium, which stores computer executable instructions.

[0015] The application has the following beneficial effects: the global grid model is decoupled to obtain a reduced-order model and a structure optimization model; external loads and boundary conditions are applied to the reduced-order model to obtain displacement loads; the displacement loads, the design constraint, and the manufacturing constraint are input into a pre-trained structure configuration optimization model to obtain a generated optimal structure configuration; the global grid model is a finite element model of the whole structure of the vehicle body; the reduced-order model includes a non-design space area, a load application area, and a boundary constraint area; and the structure optimization model includes a target structure area to be optimized and a grid transition area.

[0016] In addition, the application further provides a structure configuration generation apparatus, device, and readable storage medium, which also have the above beneficial effects. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only a part of the present application, and for those skilled in the art, other drawings can be obtained based on the provided drawings without any creative effort.

[0018] Figure 1 A flow chart of a structure configuration generation method provided by the embodiment of the present application; Figure 2 A global grid model division schematic diagram provided by the embodiment of the present application; Figure 3 A reduced order model schematic diagram provided by the embodiment of the present application; Figure 4 A structure optimization model schematic diagram provided by the embodiment of the present application; Figure 5 A structure configuration generation device structural schematic diagram provided by the embodiment of the present application; Figure 6 A structure configuration generation device structural schematic diagram provided by the embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort are within the protection scope of the present application.

[0020] Traditional structural configuration generation methods often rely on increasing mesh resolution to obtain high-fidelity, clear structural configurations, which inevitably leads to a sharp increase in computational scale and resource consumption. In recent years, many studies have attempted to use machine learning methods to establish an end-to-end mapping relationship from optimization parameters to structural configurations, replacing the complex computational process in traditional iterative optimization and achieving rapid configuration prediction. However, these end-to-end machine learning methods still require a large number of optimization results covering different external loads, boundary conditions, design constraints, and manufacturing constraints as training samples during the model training phase. Especially when dealing with practical engineering problems with complex external loads and boundary conditions, large mesh sizes, and requirements for clear and smooth boundaries in optimization results, the sample generation process itself still has extremely high computational costs. Furthermore, existing machine learning-assisted optimization methods are usually heavily dependent on specific problem settings, and their practical applications are mostly limited to the type of model and boundary conditions used for training. Once the input conditions (such as external loads / boundary conditions) exceed the parameter distribution during training, the structural configurations generated by the model are prone to significantly deviating from physical reality, thus their generalization ability is clearly limited.

[0021] To address the aforementioned issues, this invention provides a physically intelligent-driven end-to-end structural configuration generation method. By decoupling the global grid model and generalizing the displacement load mapping, it overcomes the computational bottleneck of high-precision structural configuration optimization, significantly improving the optimization efficiency of structural configurations. Based on the trained structural configuration optimization model, while maintaining similar geometric features in the design domain (referring to the target structural area to be optimized), it can generate high-fidelity, clear, and smooth structural configurations for arbitrary sizes, external loads, and boundary conditions (including non-sample working conditions), achieving universal prediction in the global parameter space.

[0022] Please refer to the details. Figure 1 , Figure 1 A flowchart illustrating a structural configuration generation method provided in an embodiment of the present invention. The method may include: S101: Decouple the global mesh model to obtain a reduced-order model and a structural optimization model.

[0023] The execution subject of the embodiment is a terminal. The embodiment does not limit the type of the terminal, as long as the operation of the structure configuration generation method can be completed. The method can be applied to a train structure configuration optimization scene. It needs to be pointed out that, according to the structure performance requirement, a global grid model is built, and the global grid model is a finite element model of the whole train body structure. In simple terms, first, the performance requirements that the train body needs to meet are determined, and then a global grid model covering the key structure of the whole train is built to simulate and calculate whether these performance requirements are met. The performance requirements determine which components the global grid model needs to include and how fine the grid needs to be, so that the subsequent calculation can accurately determine whether the structure meets the safety, stiffness and other requirements. The global grid model is decoupled into a reduced-order model and a structure optimization model, and the reduced-order model and the structure optimization model also refer to the finite element model. For the convenience of understanding, the reduced-order model can also be called a reduced-order region, and the structure optimization model can also be called a structure optimization region. For details, refer to Figure 2 、 Figure 3 and Figure 4 , Figure 2 is a global grid model division schematic diagram provided by the embodiment of the application; Figure 3 is a reduced-order model schematic diagram provided by the embodiment of the application; Figure 4 is a structure optimization model schematic diagram provided by the embodiment of the application. The model decoupling division needs to avoid the junction of the two regions, and the grid node density at the junction is large, so the division is implemented at the junction of the low-resolution grid (which refers to the reduced-order model) and the transition grid. The reduced-order model covers the non-design space area, the load application area and the boundary constraint area, and uses a low-resolution grid; the structure optimization area includes a high-resolution grid target structure area to be optimized and a grid transition area. That is, the target structure area to be optimized, the grid transition area and the reduced-order area, and the grid resolution of the three areas decreases in turn.

[0024] S102: Apply external loads and boundary conditions to the reduced-order model and solve to obtain displacement loads.

[0025] It needs to be pointed out that the external loads and boundary conditions in the step are the external loads and boundary conditions applied to the global grid model; the displacement loads solved in the step refer to the displacement loads at the junction of the reduced-order model and the structure optimization model, and the displacement loads are in the form of vectors.

[0026] Further, the above-mentioned application of external loads and boundary conditions to the reduced-order model and the solving to obtain displacement loads can specifically include: Step 11: Apply external loads and boundary conditions to the reduced-order model to obtain the total stiffness matrix of the reduced-order model and the equivalent node force applied to each node in the reduced-order model; Step 12: Based on the total stiffness matrix and the equivalent node force, the displacement vector of all nodes in the reduced-order model is calculated. Step 13: Selectively screen from all the displacement vectors of the nodes to obtain the displacement vectors at the interface between the reduced order model and the structure optimization model as the displacement load.

[0027] Specifically, the reduced order region is calculated independently, and the displacement results at the interface with the structure optimization model are extracted. The specific steps are as follows: the external load (which can include concentrated force, distributed force, body force and moment, etc.) and boundary conditions applied by the global grid model are applied to the reduced order model; the reduced order model is solved and analyzed, and U S is obtained according to the following balance equation. The displacement results of all nodes at the interface are extracted, and the extracted displacement results can be represented by the vector {Uinterface} as the displacement load.

[0028] The balance equation is: [K S ]×{U S}={F S}。

[0029] Where K S is the total stiffness matrix of the reduced order model; U S is the displacement vector of all nodes in the reduced order model; F S is the equivalent node force vector applied to the nodes of the reduced order model.

[0030] The vector representation of the displacement load {Uinterface} is: {Uinterface}=[T]×{U S}。

[0031] Where T is a selection matrix, that is, a matrix composed of all nodes at the interface.

[0032] S103: Input the displacement load, design constraint and manufacturing constraint into the pre-trained structure configuration optimization model to obtain the generated optimal structure configuration.

[0033] The displacement load obtained in step 102, and the obtained design constraint and manufacturing constraint are taken as the input of the structural configuration optimization model, and a high-resolution structural configuration with clear and smooth boundary is directly output by the structural configuration optimization model. It should be noted that the method is for structural configuration optimization of the target structure region to be optimized, and therefore, the high-resolution structural configuration corresponds to the high-resolution structural configuration of the target structure region to be optimized, that is, the optimal structural configuration is the optimal structural configuration of the target structure region to be optimized. In this step, the displacement load, the design constraint and the manufacturing constraint are input into the pre-trained structural configuration optimization model, the model generates the optimal material density field of the target structure region to be optimized, represents the optimal structural configuration of the target structure region to be optimized, and realizes the structural configuration optimization of the target structure region to be optimized. The target structure region to be optimized in this embodiment can be a train body component region such as a car frame beam, a car body covering member, and a column.

[0034] Before being input into the structural configuration optimization model, the design constraint, the manufacturing constraint and the displacement load can also be normalized and then input into the structural configuration optimization model. The structural configuration optimization model in this embodiment is not specifically limited, and the structural optimization model refers to a machine learning model. For example, it can be a neural network model, or it can also be a convolutional neural network model.

[0035] Further, the training process of the structural configuration optimization model is described below, which can specifically include: Step 21: obtaining a training data set containing various training samples; the training samples include a reference optimal structural configuration and a displacement load for training, a design constraint for training and a manufacturing constraint for training corresponding to the reference optimal structural configuration.

[0036] The displacement load, the design constraint and the manufacturing constraint obtained in this step are all used for model training, and the optimal structural configuration is used for reference. The model training in this embodiment is supervised training.

[0037] Specifically, the above obtaining of the training data set containing various training samples can include: Step 211: determining the external load for training, the boundary condition for training, the design constraint for training and the manufacturing constraint for training according to the design space of the target structure to be optimized.

[0038] In this step, according to the design space (such as design requirements, spatial structure, etc.) of the target structure to be optimized, the variation range and combination of the external load F (such as the magnitude and position of the concentrated force, pressure, torque, etc.), the boundary condition B (such as the position and size of the displacement, torsion, etc.), the design constraint D (such as the upper and lower limits of the volume constraint, modal constraint, mass constraint, etc.), and the manufacturing constraint M (such as the draft constraint, symmetry constraint, extrusion constraint, etc.) acting on the global grid model are determined.

[0039] Step 212: Based on the external load for training, the boundary condition for training, the design constraint for training, and the manufacturing constraint for training, a set of parameter combinations covering the parameter space is generated using the Latin hypercube method; the set of parameter combinations contains N parameter combinations.

[0040] This step is based on the training data determined in step 211. The Latin hypercube method is used to generate N sets of parameter combinations covering the parameter space. The formula is as follows: .

[0041] Wherein, is the set of parameter combinations; is the th parameter combination; is the th external load for training F, the th boundary condition for training B, the th design constraint for training D, and the th manufacturing constraint for training M, respectively.

[0042] Step 213: Based on each set of parameter combinations, the external load for training and the boundary condition for training are applied to the reduced-order model, and the displacement load for training is obtained by solving.

[0043] Specifically, according to each set of parameter , the data pair is generated, and first and are applied to the reduced-order model, and the th displacement load for training is obtained by solving. The specific solving process can refer to the contents of steps 11 to 13 in step 102.

[0044] Step 214: Based on the design constraint for training, the manufacturing constraint for training, and the displacement load for training, the reference optimal structure configuration is obtained.

[0045] Specifically, , , The input into the physical constraint driven configuration optimization framework results in the optimal structural configuration of the reference . .

[0046] Further, the above-mentioned design constraints for training, manufacturing constraints for training, and displacement loads for training, to obtain the optimal structural configuration of the reference, can specifically include: Step 2141: Apply the displacement load for training to the boundary nodes of the structural optimization model to maximize the compliance of the structural optimization model as the optimization objective; Step 2142: Take the element material density within the target structure optimization region as a variable; Step 2143: Take the design constraints for training and the manufacturing constraints for training as constraint conditions; Step 2144: Based on the optimization objective, the variable, and the constraint condition, use a gradient optimization algorithm to iteratively solve to obtain the optimal material density of each element within the target structure optimization region as the optimal structural configuration of the reference.

[0047] Specifically, the displacement load for training can be calculated based on the external load for training and the boundary condition for training using the displacement load solving method provided in steps 11 to 13 in step S102, and is applied to the boundary nodes of the structural optimization model as a forced displacement boundary condition to maximize the compliance as the optimization objective. The compliance C is calculated according to the following formula: .

[0048] wherein, is the stiffness sub-matrix of the target structure optimization region; T represents vector transposition; and the in the formula refers to the displacement load for training.

[0049] At the same time, the design constraints for training (such as volume fraction, mass, modal, stress, or displacement constraints, etc.) and the manufacturing constraints for training (such as draft, member size, symmetry, or periodicity constraints, etc.) are applied, i.e. constraint conditions are restricted. The relative density of the element material within the target structure optimization region [0, 1] (e is the element index) is taken as a continuous design variable, and the correlation between the density and the element stiffness is established through a material property interpolation model. Based on the global stiffness matrix assembled by the element stiffness matrix, a gradient optimization algorithm is used to iteratively solve the nonlinear constraint optimization problem: under the satisfaction of the optimization objective and the condition constraints, the density field of the target structure optimization region is updated to maximize the compliance , until the optimal material density field ​(i.e., reference density, i.e., the optimal structural configuration of the reference), characterizing the optimal structural configuration of the target structural region to be optimized.

[0050] Material property interpolation models include: .

[0051] in, The element stiffness matrix; To meet The penalty function; This is the stiffness matrix of the solid element.

[0052] Step 215: Based on the design constraints, manufacturing constraints, and displacement loads used for training, as well as the corresponding optimal structural configuration for reference, a training dataset is constructed.

[0053] Based on the aforementioned design constraints, manufacturing constraints, and displacement loads used for training, as well as the corresponding optimal structural configuration for reference, a training dataset is constructed. Furthermore, the training dataset can be divided into a training set and a validation set in a certain ratio (e.g., 8:2) for model parameter learning and generalization ability evaluation, respectively.

[0054] Step 22: Normalize the displacement load, design constraints, and manufacturing constraints used for training to obtain normalized input data.

[0055] Specifically, the model input data in the training dataset can be normalized. , , Map to the [-1,1] interval to eliminate the impact of differences in parameter magnitudes on model training.

[0056] Step 23: Input the input data into the initial structural configuration optimization model and output the predicted optimal structural configuration.

[0057] Specifically, a convolutional neural network model framework is built as a structural configuration optimization model. The network parameters are initialized, and the normalized displacement load, design constraints, and manufacturing constraints are input. The output is the predicted optimal structural configuration.

[0058] Step 24: Calculate the loss based on the predicted optimal structural configuration and the reference optimal structural configuration to obtain the loss value.

[0059] Specifically, the loss calculation process is not limited in the embodiment. Further, the loss value is obtained by calculating the loss according to the predicted optimal structure configuration and the reference optimal structure configuration, which can specifically include: calculating the density field loss based on the predicted optimal structure configuration and the reference optimal structure configuration; calculating the physical constraint loss based on the predicted optimal structure configuration and the reference optimal structure configuration; and calculating the loss value based on the density field loss and the physical constraint loss.

[0060] Specifically, the model training process can refer to the following process: The total loss function is as follows: Loss = a x L data1 + b x L data2 .

[0061] Wherein, a and b are weight coefficients, used to balance the influence of two parts of loss (weight coefficients are dynamically adjusted according to the training stage).

[0062] L data1 is the density field loss, and the calculation method is as follows: .

[0063] Wherein, M is the total number of all target structure region units to be optimized in the batch; is the predicted density, i.e. the predicted optimal structure configuration; is the reference density, i.e. the reference optimal structure configuration.

[0064] It is known that the finite element discretization is obtained . Wherein, is the structure stiffness matrix, is the input displacement load, is the corresponding node force vector, and the physical constraint loss L data2 can be expressed as: .

[0065] Wherein, is the structure stiffness of the structure configuration optimization model, which is directly related to the material density field predicted by the structure configuration optimization model, and the density value affects the element stiffness through the interpolation function; is the node force vector obtained by solving the structure optimization model.

[0066] Step 25: Based on the loss value, the initial structure configuration optimization model is optimized until the loss value meets the preset value or the iteration number of training meets the maximum iteration number, and the training is stopped, and a trained structure configuration optimization model is obtained.

[0067] After a training iteration is completed, the performance of the structure configuration optimization model can be evaluated by a validation set in the training data set, and the number of iterations is adjusted. Until the training meets the preset value of the loss function or meets the maximum number of iterations, the structure configuration optimization model capable of end-to-end generating a structure is output.

[0068] In order to solve the problem that the model lacks a continuous evolution mechanism and cannot dynamically adjust and improve performance when encountering new samples or new constraints, which limits its applicability and flexibility in actual engineering. Further, after the displacement load, design constraint and manufacturing constraint are input into the pre-trained structure configuration optimization model to obtain the generated optimal structure configuration, the following can be included: when the optimal structure configuration does not meet the preset requirements, the external load, boundary condition, design constraint and manufacturing constraint are used as new samples; the trained structure configuration optimization model is trained until the training termination condition is met, and the trained structure configuration optimization model is updated.

[0069] Specifically, the structure configuration optimization model in the embodiment can drive the structure configuration optimization model to perform dynamic optimization through incremental learning. For example, when it is found that the optimal structure configuration predicted by the structure configuration optimization model does not meet the requirements, such as the difference with the result solved by the physical constraint driven configuration optimization framework is large, the manufacturing path is not available or a completely new optimization constraint combination is encountered, the system automatically takes this case as a new sample to perform dynamic optimization of the structure configuration optimization model, so that it can continuously improve and adapt to new requirements.

[0070] The application embodiment provided by the application decouples the global grid model to obtain a reduced order model and a structure optimization model; external loads and boundary conditions are applied to the reduced order model to obtain displacement loads by solving; the displacement loads, design constraints and manufacturing constraints are input into a pre-trained structure configuration optimization model to obtain a generated optimal structure configuration; wherein the global grid model is a finite element model of the whole structure of the vehicle body; the reduced order model includes a non-design space area, a load application area and a boundary constraint area; the structure optimization model includes a target structure area to be optimized and a grid transition area. The method decouples the complex global grid model into two parts of the reduced order model and the structure optimization model, and converts the complex external loads and boundary conditions into an intermediate variable of displacement loads, which greatly reduces the calculation scale of the structure configuration optimization model and improves the calculation efficiency. In this way, the grid resolution in the target structure area to be optimized can also be improved to generate a high-fidelity and smooth structure configuration, which can reduce unnecessary materials and make the train body lighter without affecting the safety of the train; and under the premise of maintaining similar geometric characteristics of the target structure area to be optimized, a high-fidelity and smooth structure configuration can be generated for any size (referring to adjustable size parameters of the target structure limited by design constraints and manufacturing constraints), external loads and boundary conditions, realizing universality in the global parameter space; and the method also integrates an incremental learning mechanism to autonomously optimize the prediction results for new constraint combinations and continuously improve the model adaptability and evolution ability.

[0071] The structure configuration generation device provided by the application embodiment is described below, and the structure configuration generation device described below can be correspondingly referred to the structure configuration generation method described above.

[0072] For details, please refer to Figure 5 , Figure 5 The structure diagram of the structure configuration generation device provided by the application embodiment can include: The decoupling module 100 is configured to decouple the global grid model to obtain a reduced order model and a structure optimization model; The solving module 200 is configured to apply external loads and boundary conditions to the reduced order model and solve to obtain displacement loads; The generation module 300 is configured to input the displacement loads, design constraints and manufacturing constraints into a pre-trained structure configuration optimization model to obtain a generated optimal structure configuration; The global grid model is a finite element model of the whole structure of the vehicle body; the reduced order model includes a non-design space area, a load application area and a boundary constraint area; the structure optimization model includes a target structure area to be optimized and a grid transition area; and the optimal structure configuration is the optimal structure configuration of the target structure area to be optimized.

[0073] Based on the above embodiment, the solving module 200 can comprise: An acquisition unit is configured to apply the external load and the boundary condition to the reduced order model to obtain a total stiffness matrix of the reduced order model and equivalent node forces applied to each node in the reduced order model; A displacement vector calculation unit is configured to calculate displacement vectors of all nodes in the reduced order model based on the total stiffness matrix and the equivalent node forces; A selection unit is configured to selectively filter the displacement vectors of all nodes to obtain the displacement vectors at the intersection between the reduced order model and the structural optimization model as the displacement load.

[0074] Based on the above embodiment, the structural configuration generation device can further comprise: A training data set acquisition module is configured to acquire a training data set containing various training samples; the training samples include a reference optimal structural configuration and a displacement load for training, a design constraint for training, and a manufacturing constraint for training corresponding to the reference optimal structural configuration; A normalization module is configured to normalize the displacement load for training, the design constraint for training, and the manufacturing constraint for training to obtain normalized input data; An input-output module is configured to input the input data into an initial structural configuration optimization model and output a predicted optimal structural configuration; A loss calculation module is configured to calculate a loss value based on the predicted optimal structural configuration and the reference optimal structural configuration; An optimization module is configured to optimize the initial structural configuration optimization model based on the loss value until the loss value meets a preset value or the number of iterations meets a maximum number of iterations, at which point the training is stopped, and the trained structural configuration optimization model is obtained.

[0075] Based on the above embodiment, the training data set acquisition module can comprise: A determination unit is configured to determine an external load for training, a boundary condition for training, a design constraint for training, and a manufacturing constraint for training based on a design space of a target structure to be optimized; A parameter set generation unit is configured to generate a parameter combination set covering a parameter space based on the external load for training, the boundary condition for training, the design constraint for training, and the manufacturing constraint for training using a Latin hypercube method; the parameter combination set contains N parameter combinations; a displacement load calculation unit configured to apply the external load for training and the boundary condition for training to the reduced-order model based on each combination of the parameter groups, and solve the reduced-order model to obtain the displacement load for training; an optimal structure configuration generation unit configured to obtain the reference optimal structure configuration based on the design constraint for training, the manufacturing constraint for training, and the displacement load for training; a training data set construction unit configured to construct the training data set based on the design constraint for training, the manufacturing constraint for training, and the displacement load for training, and the corresponding reference optimal structure configuration.

[0076] Based on the above embodiments, the optimal structure configuration generation unit can include: an optimization objective determination subunit configured to apply the displacement load for training to boundary nodes of the structure optimization model to maximize flexibility of the structure optimization model as an optimization objective; a variable determination subunit configured to take material density of a cell in the target structure region as a variable; a constraint condition determination subunit configured to take the design constraint for training and the manufacturing constraint for training as constraint conditions; a solving subunit configured to iteratively solve, based on the optimization objective, the variable, and the constraint conditions, using a gradient optimization algorithm to obtain optimal material density of each cell in the target structure region as the reference optimal structure configuration.

[0077] Based on the above embodiments, the loss calculation module can include: a density field loss calculation unit configured to calculate a density field loss based on the predicted optimal structure configuration and the reference optimal structure configuration; a physical constraint loss calculation unit configured to calculate a physical constraint loss based on the predicted optimal structure configuration and the reference optimal structure configuration; a loss value calculation unit configured to calculate the loss value based on the density field loss and the physical constraint loss.

[0078] Based on the above embodiments, the structure configuration generation apparatus can further include: a new sample generation module configured to, when the optimal structure configuration does not meet a preset requirement, take the external load, the boundary condition, the design constraint, and the manufacturing constraint as a new sample; a model optimization module configured to train the trained structure configuration optimization model using the new sample until a training termination condition is met, and update the trained structure configuration optimization model.

[0079] It should be noted that the modules and units in the above structural configuration generation device can be changed in sequence without affecting the logic.

[0080] The structural configuration generation device provided by the embodiment of the application is used for decoupling the global grid model through the decoupling module 100 to obtain a reduced order model and a structural optimization model; the solving module 200 is used for applying external loads and boundary conditions to the reduced order model and solving to obtain displacement loads; and the generation module 300 is used for inputting the displacement loads, design constraints and manufacturing constraints into the pre-trained structural configuration optimization model to obtain a generated optimal structural configuration. The device decouples the complex global grid model into two parts of the reduced order model and the structural optimization model, and converts the complex external loads and boundary conditions into an intermediate variable of displacement loads, which greatly reduces the calculation scale of the structural configuration optimization model and improves the calculation efficiency. In this way, the grid resolution in the target structure area to be optimized can also be improved, a high-fidelity and smooth structural configuration is generated, unnecessary materials can be reduced to make the train body lighter without affecting the safety of the train; and under the premise of maintaining similar geometric characteristics of the target structure area to be optimized, a high-fidelity and smooth structural configuration can be generated for any size (referring to adjustable size parameters of the target structure, limited by design constraints and manufacturing constraints), external loads and boundary conditions, realizing universality in the global parameter space; and the device also integrates an incremental learning mechanism, which can autonomously optimize the prediction results for new constraint combinations and continuously improve the model adaptability and evolution ability.

[0081] The structural configuration generation device provided by the embodiment of the application is introduced as follows, and the structural configuration generation device described below can be correspondingly referred to the structural configuration generation method described above.

[0082] Please refer to Figure 6 , Figure 6 The structural configuration generation device provided by the embodiment of the application can include: The memory 10 is used for storing a computer program; The processor 20 is used for executing the computer program to realize the above structural configuration generation method.

[0083] The memory 10, the processor 20 and the communication interface 31 can communicate with each other through the communication bus 32.

[0084] In the embodiment of the application, the memory 10 stores one or more programs, and the program can include program code including computer operation instructions. In the embodiment of the application, the memory 10 can store programs for realizing the following functions: The global grid model is decoupled to obtain a reduced model and a structure optimization model; The external load and boundary conditions are applied to the reduced model, and the displacement load is obtained by solving the reduced model; The displacement load, design constraints and manufacturing constraints are input into the pre-trained structure configuration optimization model to obtain a generated optimal structure configuration; The global grid model is a finite element model of the whole structure of the vehicle body; the reduced model includes a non-design space area, a load application area and a boundary constraint area; the structure optimization model includes a target structure area to be optimized and a grid transition area; and the optimal structure configuration is an optimal structure configuration of the target structure area to be optimized.

[0085] In a possible implementation, the memory 10 can include a program storage area and a data storage area, where the program storage area can store an operating system and application programs required by at least one function, and the like; and the data storage area can store data created during use.

[0086] In addition, the memory 10 can include a read-only memory and a random access memory, and provide instructions and data for the processor. A part of the memory can also include an NVRAM. The memory stores an operating system and operation instructions, executable modules or data structures, or a subset thereof, or an extended set thereof, where the operation instructions can include various operation instructions for implementing various operations. The operating system can include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0087] The processor 20 can be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic device, and the processor 20 can be a microprocessor or any conventional processor. The processor 20 can invoke a program stored in the memory 10.

[0088] The communication interface 31 can be an interface of a communication module, used for connecting with other devices or systems.

[0089] Of course, it should be noted that, Figure 6 The structures shown do not constitute a limitation on the structure configuration generation device in the embodiments of the present application, and in actual applications, the structure configuration generation device can include more or fewer components than Figure 6 those shown, or combine certain components.

[0090] The computer-readable storage medium provided in the embodiments of the present application is described below, and the computer-readable storage medium described below can be mutually referred to the structure configuration generation method described above.

[0091] The application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the structure configuration generation method.

[0092] The computer readable storage medium can include a U disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and various storage medium capable of storing program codes.

[0093] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts of each embodiment can be referred to each other. For the device disclosed by the embodiments, since it corresponds to the method disclosed by the embodiments, the description is relatively simple, and the related parts can be referred to the method part.

[0094] The skilled person can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present text can be realized in electronic hardware, computer software or combination of both. In order to clearly show the interchangeability of hardware and software, the composition and steps of each example have been described in the above description. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0095] Finally, it should be noted that in the present text, the relationship such as first and second is only used to distinguish one entity or operation from another entity or operation, and does not necessarily require or imply any actual relationship or order between the entities or operations. Moreover, the term "include", "contain" or any other variant is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0096] The structural configuration generation method, device, equipment and computer readable storage medium provided by the present application are described in detail above, the principle and implementation mode of the present application are described by applying specific examples in this paper, and the above example description is only used to help understand the method of the present application and its core idea; at the same time, for the general technical personnel in the art, according to the idea of the present application, the specific implementation mode and application range will be changed, and the above description should not be understood as the limitation of the present application.

Claims

1. A method for generating structural configurations, characterized in that, include: The global mesh model is decoupled to obtain a reduced-order model and a structure optimization model; External loads and boundary conditions are applied to the reduced-order model, and the displacement loads are obtained by solving the model. The displacement load, design constraints, and manufacturing constraints are input into a pre-trained structural configuration optimization model to obtain the generated optimal structural configuration. The global mesh model is a finite element model of the overall vehicle structure; the reduced-order model includes a non-design space region, a load application region, and a boundary constraint region; the structural optimization model includes a target structure region to be optimized and a mesh transition region; and the optimal structural configuration is the optimal structural configuration of the target structure region to be optimized.

2. The structural configuration generation method according to claim 1, characterized in that, External loads and boundary conditions are applied to the reduced-order model, and the displacement loads are obtained by solving the model, including: The external load and the boundary conditions are applied to the reduced-order model to obtain the total stiffness matrix of the reduced-order model and the equivalent nodal forces applied to each node in the reduced-order model. Based on the total stiffness matrix and the equivalent nodal forces, the displacement vectors of all nodes in the reduced-order model are calculated. Selectively filter the displacement vectors of all nodes to obtain the displacement vector at the intersection of the reduced-order model and the structural optimization model, and use it as the displacement load.

3. The structural configuration generation method according to any one of claims 1 to 2, characterized in that, Before decoupling the global mesh model to obtain the reduced-order model and the structure optimization model, the following steps are also included: Obtain a training dataset containing various training samples; the training samples include a reference optimal structural configuration and the displacement load, design constraints, and manufacturing constraints corresponding to the reference optimal structural configuration for training. The displacement load, design constraints, and manufacturing constraints used for training are normalized to obtain normalized input data. The input data is fed into the initial structural configuration optimization model, and the predicted optimal structural configuration is output. The loss is calculated based on the predicted optimal structural configuration and the reference optimal structural configuration to obtain the loss value; The initial structural configuration optimization model is optimized based on the loss value until the loss value meets the preset value or the number of training iterations meets the maximum number of iterations. Then the training stops, and the trained structural configuration optimization model is obtained.

4. The structural configuration generation method according to claim 3, characterized in that, The loss is calculated based on the predicted optimal structural configuration and the reference optimal structural configuration to obtain the loss value, including: Based on the predicted optimal structural configuration and the reference optimal structural configuration, the density field loss is calculated. Based on the predicted optimal structural configuration and the reference optimal structural configuration, the physical constraint loss is calculated. The loss value is calculated based on the density field loss and the physical constraint loss.

5. The structural configuration generation method according to claim 3, characterized in that, Obtain a training dataset containing various training samples, including: Based on the design space of the target structure to be optimized, determine the external load for training, the boundary conditions for training, the design constraints for training, and the manufacturing constraints for training. Based on the external load used for training, the boundary conditions used for training, the design constraints used for training, and the manufacturing constraints used for training, a set of parameter combinations covering the parameter space is generated using the Latin hypersol method; the set of parameter combinations contains N parameter combinations. Based on each set of parameter combinations, the external load and boundary conditions for training are applied to the reduced-order model, and the displacement load for training is obtained by solving the problem. Based on the design constraints, manufacturing constraints, and displacement loads used for training, the optimal structural configuration for the reference is obtained. The training dataset is constructed based on the design constraints, manufacturing constraints, and displacement loads used for training, as well as the corresponding optimal structural configuration of the reference.

6. The structural configuration generation method according to claim 5, characterized in that, Based on the design constraints, manufacturing constraints, and displacement loads used for training, the optimal structural configuration for the reference is obtained, including: The displacement load used for training is applied to the boundary nodes of the structural optimization model, with the optimization objective being to maximize the flexibility of the structural optimization model. The unit material density within the target structural region to be optimized is used as a variable; The design constraints and manufacturing constraints used for training are used as constraint conditions. Based on the optimization objective, the variables, and the constraints, the optimal material density of each unit in the target structural region is obtained by iteratively solving the gradient optimization algorithm, which serves as the optimal structural configuration for reference.

7. The structural configuration generation method according to claim 1, characterized in that, After inputting the displacement load, design constraints, and manufacturing constraints into a pre-trained structural configuration optimization model to obtain the generated optimal structural configuration, the process further includes: When the optimal structural configuration does not meet the preset requirements, a new sample is obtained based on the external load, the boundary conditions, the design constraints, and the manufacturing constraints. The trained structural configuration optimization model is trained using the new samples until the training termination condition is met, and then the trained structural configuration optimization model is updated.

8. A structural configuration generation device, characterized in that, include: The decoupling module is used to decouple the global mesh model to obtain a reduced-order model and a structure optimization model; The solver module is used to apply external loads and boundary conditions to the reduced-order model and solve for the displacement loads. The generation module is used to input the displacement load, design constraints and manufacturing constraints into a pre-trained structural configuration optimization model to obtain the generated optimal structural configuration. The global mesh model is a finite element model of the overall vehicle structure; the reduced-order model includes a non-design space region, a load application region, and a boundary constraint region; the structural optimization model includes a target structure region to be optimized and a mesh transition region; and the optimal structural configuration is the optimal structural configuration of the target structure region to be optimized.

9. A structural configuration generation device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the structural configuration generation method as described in any one of claims 1 to 7.

10. A readable storage medium, characterized in that, The readable storage medium stores computer-executable instructions, which, when loaded and executed by a processor, implement the steps of the structural configuration generation method as described in any one of claims 1 to 7.

Images