Notch optimization method and device for woven composite expandable structure and medium

Abstract

The invention provides a cut optimization method and device for a woven composite deployable structure and a medium, and relates to the technical field of data processing, and the method comprises the steps: building a search model, and determining a material optimization variable of the woven composite deployable structure, so as to obtain an initial structure; calculating an ABD matrix by using a deep neural network model based on the initial structure to generate initial structure parameters; presetting an initial notch in the initial structure, establishing a finite element model based on initial structure parameters to determine a structure optimization variable, and performing notch optimization on the initial notch iteration by adopting a bidirectional progressive structure optimization model according to the structure optimization variable; and updating the initial structure based on the material optimization variable under the search model, performing notch optimization by using the updated structure, performing iteration until a predetermined optimization target is met, generating a target structure with a target notch, and obtaining the target structure with the target notch. The problem that an existing woven composite deployable structure lacks an incision optimization method which considers balance of a woven composite material, an incision rigid target and a flexible target at the same time is solved.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to a method, a device and a medium for optimizing the incision of a braided composite deployable structure. Background Art

[0002] Deployable Composite Thin-Walled Shell with Cutouts (DCTWS) structures are widely used in space applications because they can be compactly folded while maintaining sufficient stiffness when deployed. These structures usually have cutouts in thin laminates to prevent material failure, making the cutout design critical for optimizing DCTWS performance. Material safety, deployment drive capability, deformation resistance, and buckling stability are important mechanical performance indicators of the structure under folding, deployment, and load-bearing conditions, respectively, and are significantly affected by the number, size, and shape of the cutouts.

[0003] The complete service process of DCTWS involves issues such as large geometric deformation, composite material failure, deployment dynamics and structural stability. These complex behaviors pose challenges to the construction of the incision structure. Its design faces the sharp problem of the contradiction between rigidity and flexibility goals: flexibility makes it easy to fold and reduce material damage; rigidity enhances deployment performance, load-bearing capacity and structural stability.

[0004] Most existing incision designs consider the parameter adjustment of the incision size or shape. Even though some studies have optimized the balance between the incision rigidity target and the flexibility target, they lack consideration of woven composite materials, which makes it impossible to significantly improve the folding and unfolding performance, easily increases the risk of failure, and has poor optimization effect. Summary of the invention

[0005] In order to overcome the above-mentioned technical defects, the purpose of the present invention is to provide a method, device and medium for optimizing the incision of a braided composite deployable structure, so as to solve the problem that the existing incision optimization of braided composite deployable structures lacks an optimization method that simultaneously considers the balance of braided composite materials, incision rigidity targets and flexibility targets.

[0006] The present invention discloses a method for optimizing the incision of a braided composite deployable structure, comprising:

[0007] Establishing a search model to determine the material optimization variables of the braided composite deployable structure to obtain an initial structure;

[0008] Calculating an ABD matrix based on the initial structure using a deep neural network model to generate initial structural parameters;

[0009] Predetermining an initial cut in the initial structure, establishing a finite element model based on the initial structural parameters to determine structural optimization variables, and iteratively optimizing the initial cut using a bidirectional progressive structural optimization model according to the structural optimization variables, wherein the structural optimization variables include bending stiffness, critical buckling load, and failure index;

[0010] The initial structure is updated based on the material optimization variables under the search model, and the cutout is optimized with the updated structure, and the process is repeated until a predetermined optimization target is met, thereby generating a target structure with a target cutout.

[0011] Preferably, the predetermined optimization objectives include: for each structure composed of different material optimization variables, after the target incision is determined by the incision optimization of the bidirectional progressive structural optimization model, the bending stiffness, critical buckling load and failure index aggregation of the target incision corresponding to each structure under the failure constraint are maximized.

[0012] Preferably, a third generation non-dominated sorting genetic algorithm is used to establish the search model.

[0013] Preferably, the material optimization variables include material parameters, geometric parameters and ply parameters;

[0014] The material parameters include the major and minor axis dimensions of the yarn cross section and the yarn spacing dimensions;

[0015] The set parameters include the number of warp / weft yarns and the yarn interlacing position;

[0016] The ply parameters include the number of plies and the single-layer ply laying angle.

[0017] Preferably, Latin hypercube sampling is used to determine the initial structure and / or update the initial structure based on the material optimization variables.

[0018] Preferably, the calculating the ABD matrix based on the initial structure using a deep neural network model to generate initial structural parameters includes:

[0019] Obtain various material optimization variables corresponding to the initial structure;

[0020] Based on the various material optimization variables of the initial structure, microscopic RVE homogenization analysis and mesoscopic RVE homogenization analysis are performed to calculate the ABD matrix corresponding to the initial structure to generate initial structure parameters.

[0021] Preferably, the deep neural network model is pre-built and / or trained offline.

[0022] Preferably, under the bidirectional progressive structural optimization model, the initial cutout is divided into a number of units, and the bending stiffness, linear buckling analysis, and sensitivity of the failure index corresponding to each unit are determined;

[0023] The initial cutout is optimized by adding or reducing units according to the sensitivity, until the bending stiffness, critical buckling load and failure index of the target cutout corresponding to each structure are maximized under the failure constraint.

[0024] The present invention also provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any step of the incision optimization method when executing the computer program.

[0025] The present invention also provides a computer storage medium, wherein a computer program is stored on the storage medium, and when executed by a processor, the steps of any one of the incision optimization methods are implemented.

[0026] Compared with the prior art, the above technical solution has the following beneficial effects:

[0027] The present application provides a method, device and medium for optimizing the incision of a braided composite deployable structure. The method determines the material optimization variables through a search model, determines the initial structure, generates the corresponding ABD matrix in combination with a deep neural network model, and then performs incision iterative optimization based on the initial structure using a finite element model combined with a bidirectional progressive structural optimization model with a predetermined optimization target to obtain the target incision of the initial structure. The initial structure is then updated and the incision iterative optimization is continued until the predetermined optimization target is reached. The search model and the bidirectional progressive structural optimization model converge, thereby forming a synchronous optimum for the braided composite material design and the incision structure design, that is, simultaneously introducing material optimization and structural optimization to form a nested double-layer iterative optimization model as a whole, so as to solve the problem that the existing braided composite deployable structure incision optimization lacks an optimization method that simultaneously considers the balance of the braided composite material, the incision rigidity target and the flexibility target. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A flow chart of the incision optimization method of Embodiment 1 of the incision optimization method, device and medium of the braided composite deployable structure of the present invention;

[0029] Figure 2 It is a schematic diagram of a specific process of the incision optimization method in Embodiment 1 of the incision optimization method, device and medium of the braided composite deployable structure of the present invention;

[0030] Figure 3A schematic diagram of a processing process of generating initial structural parameters by calculating an ABD matrix using a search model and a deep neural network in a first embodiment of a method, device and medium for optimizing a cut of a braided composite deployable structure according to the present invention;

[0031] Figure 4 This is a reference diagram of a braided composite deployable structure in Embodiment 1 of the incision optimization method, device and medium of the braided composite deployable structure described in the present invention. DETAILED DESCRIPTION

[0032] The advantages of the present invention are further described below in conjunction with the accompanying drawings and specific embodiments.

[0033] 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 the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0034] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. The singular forms of "a", "said" and "the" used in this disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0035] It should be understood that although the terms first, second, third, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present disclosure, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0036] In the description of the present invention, it is necessary to understand that the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0037] In the description of the present invention, unless otherwise specified and limited, it should be noted that the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a mechanical connection or an electrical connection, or it can be the internal connection between two components. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0038] In the following description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present invention, and have no specific meanings. Therefore, "module" and "component" can be used interchangeably.

[0039] Embodiment 1: This embodiment discloses a cutout optimization method for a braided composite deployable structure. For illustration, the composite structure is a foldable and deployable composite material with cutouts, which may be a DCTWS (Deployable Composite Thin-Walled Shell with Cutouts), or a RA-WCTH right-angle braided composite thin-walled tubular hinge structure, or other composite structures with cutouts that are deployable and foldable, such as Figure 4 As illustrated, the "structure" described in this embodiment is the composite structure with an incision that can be expanded and folded, the "woven composite" includes but is not limited to being composed of multiple parts (the two cylindrical shells mentioned below) / multiple materials (the yarns and layers mentioned below), and the "incision" is the area where the incision is formed.

[0040] In this implementation method, a nested two-layer iterative optimization model is constructed (the search model described below is iterated in the outer layer, and the incision optimization model (the finite element model described below and the bidirectional progressive structural optimization model) is iterated in the inner layer). The outer global search algorithm is combined with the agent model (deep neural network model, DNN) to dynamically optimize the material and structural parameters of the woven composite laminate, generate the ABD matrix and import the incision optimization model (specifically the (DCTWS macro) finite element model), and the inner layer performs incision iterative optimization until the optimization target converges and meets the constraints, thereby forming the synchronous optimality of the woven composite material design and the macroscopic structure incision design.

[0041] It can be understood that the outer network of the double-layer iterative optimization model of this embodiment is used to optimize the parameters of the woven composite material, and the inner network is used to optimize the incision structure. The models under the outer network and the inner network do not operate independently of each other, but operate in a related manner. Specifically, the parameters of the woven composite material will affect the bending stiffness, critical buckling load and failure index calculated by the finite element model under the layer-by-layer network, thereby affecting the iterative optimization results of the incision structure performed by the bidirectional progressive structural optimization model. After meeting the target incision iteratively output by the bidirectional progressive structural optimization model, it is also necessary to meet the optimization of the woven composite material parameters under the outer network, thereby achieving the optimization of the material and incision structure.

[0042] For details, see Figure 1-Figure 3 ,include:

[0043] S10: establishing a search model to determine material optimization variables of the braided composite deployable structure to obtain an initial structure;

[0044] In this embodiment, the above-mentioned material optimization variables include single-sided material parameters, geometric parameters and laying parameters, etc.; the material parameters include but are not limited to the major and minor axis dimensions of the yarn cross-section, the yarn spacing dimensions, etc.; the collective parameters include but are not limited to the number of warp / weft yarns, the yarn interlacing position, etc.; the laying parameters include but are not limited to the number of layers, the single-layer laying angle, etc., and the material optimization variables can be increased or decreased according to the woven composite material in the application scenario.

[0045] Specifically, Figure 3 As shown, in this embodiment, the third generation non-dominated sorting genetic algorithm (NSGA-III) is preferably used to establish a search model. NSGA-III is a multi-objective optimization algorithm that maintains the diversity of the population through non-dominated sorting and microhabitat technology. As an option, simulation analysis can be performed to determine the parts of the braided composite material (laminate) that are prone to failure when subjected to axial loads, thereby improving the optimization effect of the search model.

[0046] Furthermore, in this embodiment, it is necessary to optimize the incision structure under different materials to determine the best material and incision structure. Therefore, it is necessary to first determine an initial structure (determined by each material optimization variable), and then use Latin hypercube sampling (such as Figure 3 The Latin hypercube sampling is a method of approximating random sampling from a multivariate parameter distribution, which has the characteristics of uniform stratification and can obtain sample values ​​at the tail with less sampling, which makes it particularly effective in processing large-scale data, and uniformly samples in multiple dimensions (this embodiment and in each material optimization variable), reduces data correlation, and thus improves the reliability of the results.

[0047] S20: Calculating an ABD matrix based on the initial structure using a deep neural network model to generate initial structural parameters;

[0048] In this embodiment, the following incision optimization (structural optimization) is performed under the inner layer network, which needs to consider the folding and unfolding performance of the incision. Specifically, in this embodiment, its performance is characterized by bending stiffness, critical buckling load and failure index, and the above parameters are calculated based on the ABD matrix. The ABD matrix is ​​used to describe the elastic response of the composite laminate when it is loaded, which includes the tensile, shear and bending properties of the material in the plane, and is the basis for structural analysis.

[0049] In this scheme, if Figure 3 As shown in the figure, based on the microscopic characteristics and arrangement rules of woven composite materials, a micro-mesoscopic series homogenization model of parameterized representative volume units is constructed, and a proxy model is constructed (i.e., mapping is applied) in combination with a deep neural network (DNN) to achieve rapid prediction of the ABD matrix of laminates under different weaving forms, material parameters and geometric parameters.

[0050] Specifically, the ABD matrix is ​​calculated based on the initial structure using a deep neural network model to generate initial structure parameters, including: obtaining various material optimization variables corresponding to the initial structure; performing microscopic RVE homogenization analysis and mesoscopic RVE homogenization analysis based on the various material optimization variables of the initial structure to calculate the ABD matrix corresponding to the initial structure to generate initial structure parameters.

[0051] Micro RVE homogenization analysis involves predicting the macroscopic mechanical properties of materials from a microscopic scale. At the microscopic scale, each material point is considered to be the center of the RVE. The purpose of micro RVE homogenization analysis is to make the strain energy of the RVE equivalent to the strain energy of a uniform material by applying specified unit displacement boundary conditions or force boundary conditions to the RVE, so as to obtain equivalent properties; meso RVE homogenization analysis focuses on the mesoscopic structure of the material, establishes the incremental nonlinear constitutive relationship of the RVE, and expresses the incremental forces and coupling moments from the external medium of the RVE on the RVE boundary. These forces and moments are expressed in terms of the incremental translation and rotational displacements of the surrounding particles of the RVE, the elastic stiffness of the current deformed mesostructure RVE, and the incremental dissipative friction condensed to the surrounding particles of the RVE.

[0052] In the prior application of the present applicant, it is disclosed that: a mesoscale RVE for weaving a composite material is generated by composite material modeling software; and the generated mesoscale RVE is imported into ABAQUS finite element analysis software; an ABD matrix plug-in is run in the ABAQUS finite element analysis software, material parameters are input in the ABD matrix plug-in, and material directions are specified for the yarns; the material parameters include yarn parameters and matrix parameters; periodic boundary conditions are set by the ABD matrix plug-in to apply 6 periodic boundary conditions to the mesoscale RVE, thereby realizing 6 variations of the mesoscale RVE; the variation data of the 6 variations of the mesoscale RVE are analyzed by the ABAQUS finite element analysis software, thereby calculating the ABD matrix of the thin woven composite material; that is, a method for determining the ABD matrix based on microscopic RVE homogenization and mesoscopic RVE homogenization is specifically disclosed, or other existing disclosed methods for determining the ABD matrix can be used to replace the aforementioned method; in the present application, the above-mentioned method of determining the ABD matrix is ​​implemented by using the fitting mapping of a deep neural network, that is, the input and output of the aforementioned method are used as the input and output of the deep neural network model of this embodiment, and it is implemented by network layer fitting mapping.

[0053] It is understandable that a micro-macroscale tandem homogenization calculation program can be established to batch calculate the ABD matrices of different structures to form an output data set. Therefore, in this embodiment, a DNN is used to fit the mapping relationship between input and output to complete the establishment of the deep neural network model and perform batch ABD matrix calculations for the initial structure / the updated structure described below.

[0054] As described above, in this embodiment, the deep neural network model implements the above-mentioned ABD matrix calculation based on input and output fitting mapping. It can be understood that the deep neural network model can be pre-built and / or trained offline. It can be pre-built and autonomously called through the interface when applied in this embodiment, thereby improving processing efficiency without occupying large processing resources.

[0055] S30: predetermine an initial cut in the initial structure, establish a finite element model based on the initial structural parameters to determine structural optimization variables, and iteratively optimize the initial cut using a bidirectional progressive structural optimization model according to the structural optimization variables, wherein the structural optimization variables include bending stiffness, critical buckling load, and failure index;

[0056] Specifically, the above step S10 and the following step S40 can be understood as the optimization of the material by the above outer network, and the step S30 can be understood as the optimization of the (incision) structure under the above inner network (such as Figure 2As shown in S30), in this embodiment, two finite element models (i.e., DCTWS shell unit finite element models) are established, one for nonlinear folding analysis (folding-unfolding analysis) of the incision to determine the corresponding bending stiffness and failure index, and the other for linear buckling analysis of the incision to determine the corresponding critical buckling load; the incision is divided into a number of units, and the sensitivity of each unit in the incision in terms of bending stiffness (compliance (stiffness, strain energy) is linearly related to bending stiffness), failure index and critical buckling load is determined, and then a bidirectional evolutionary structural optimization (BESO) model is used, which is based on the weighted aggregated sensitivity of each unit in terms of bending stiffness, failure index and critical buckling load, and the highest structural performance within a predetermined range is found by gradually adding units.

[0057] Specifically, under the bidirectional progressive structural optimization model, the initial cutout is divided into several units, and the bending stiffness, linear buckling analysis, and failure index sensitivity corresponding to each unit are determined. According to the sensitivity, the initial cutout is optimized by adding or reducing units to iteratively until the bending stiffness, critical buckling load, and failure index of the target cutout corresponding to each structure are maximized under the failure constraint.

[0058] It can be understood that the structural optimization target values ​​(C(X) (compliance) (determination of K b (X)(bending stiffness)) failure index FI(X) and critical buckling load λ1(X), constraint condition FI(X) max <1 and the information at the centroid of each unit. Then the bending stiffness K is calculated by chain derivation, linear buckling finite element equilibrium equation, etc. b (X), failure index FI(X) and critical buckling load λ1(X); specifically, the sensitivity of each unit is determined by linear weighted aggregation based on the sensitivity of each structural optimization variable such as bending stiffness, critical buckling load and failure index. w1, w2, and w3 are the weights of the sensitivity of bending stiffness, critical buckling load, and failure index, respectively; They are the normalized bending stiffness, critical buckling load and sensitivity of failure index respectively, and the influence of each target optimization parameter on the optimization process can be adjusted by weight control to achieve single-objective optimization and multi-objective optimization. Each unit is reordered and classified in the bidirectional progressive optimization model based on dichotomy according to sensitivity, and it is determined whether to be retained or deleted; adjusting the sensitivity weight coefficient can achieve the emphasis on different optimization goals, thereby guiding the optimization direction of the incision.

[0059] In this embodiment, in order to facilitate the description of the operation of integrating the above-mentioned finite element model and the bidirectional progressive structural optimization model, it can be called an incision optimization model, which is arranged in the inner network of the nested two-layer iterative optimization model of the above-mentioned embodiment, that is, based on material optimization, a structure with different materials is generated.

[0060] Specifically, starting from the unit finite element model described by the ABD matrix, the sensitivity expressions of the bending stiffness, critical buckling load and failure index of the woven composite laminate are derived. The linear weighted function method is used to aggregate the sensitivity equations into a single expression, and combined with the bidirectional evolutionary structural optimization method (BESO), a multi-objective optimization model with adjustable weight coefficients is formed to solve the topological optimization problem with structural stability, bending stiffness and maximization of stored strain energy as the objectives and failure index as the constraint, and to determine the optimal incision arrangement scheme under each structure.

[0061] For illustration, the above-mentioned incision optimization model may also refer to the detailed description of the incision optimization method of a composite structure previously applied for by the applicant, or may also refer to other currently disclosed topology optimization methods for woven composite material layer units that consider the folding and unfolding performance, for application in the present embodiment.

[0062] S40: updating the initial structure under the search model, and then performing incision optimization based on the updated structure, iterating until a predetermined optimization target is met, and generating a target structure with a target incision (S40-1).

[0063] Specifically, the predetermined optimization objectives include: for each structure composed of different material optimization variables, after the bidirectional progressive structural optimization model performs incision optimization to determine the target incision, the bending stiffness, critical buckling load and failure index aggregation of the target incision corresponding to each structure under the failure constraint are maximized.

[0064] Specifically, the above predetermined optimization objectives are:

[0065]

[0066] Subject to: X = {x e}; x e =0 or 1; 2, 3, ..., N;

[0067] KU=F&(K-λ1K σ ) m =0;

[0068] FI max (X)<1;

[0069]

[0070] Where T(X) is the optimization target value (total sensitivity); w1, w2, and w3 are the weights of the sensitivity of bending stiffness, critical buckling load, and failure index, respectively; are the normalized bending stiffness, critical buckling load and failure index, respectively. The optimization target value is positively correlated with the bending stiffness and failure index, and negatively correlated with the critical buckling load; x e is any unit in the cutout (area) (divided into several units in the above two-way progressive structural optimization model), x e = 0, the unit is deleted, x e =1, the unit is retained, N is the total number of units; K is the global stiffness matrix (obtained from the ADM matrix corresponding to the structure / initial structural parameters); K σ is the stress-stiffness matrix (obtained from the ADM matrix / initial structural parameters corresponding to the structure); F is the global mechanical load vector (obtained from the ADM matrix / initial structural parameters corresponding to the structure); U is the global displacement vector obtained by the nonlinear folding analysis (of the finite element model); U m The global displacement vector obtained for the linear buckling analysis of the (finite element model); KU = F, (K-λ1K σ ) m =0 are the finite element equilibrium equations for nonlinear folding analysis and linear buckling analysis respectively.

[0071] As an illustration, the above optimization goal: maximizing the aggregation of bending stiffness, critical buckling load and failure index of the target cutouts corresponding to each structure under failure constraints is consistent with the optimization goal (convergence condition) in the bidirectional progressive structural optimization model. That is, for any structure (initial structure / updated structure), its structural parameters (ABD matrix) are determined based on the structure, and nonlinear folding analysis / linear buckling analysis is performed using a finite element model to determine the bending stiffness, critical buckling load and failure index. The sensitivity of each unit in the aggregate cutout (region / design domain) in bending stiffness, critical buckling load and failure index is determined. In the bidirectional progressive structural optimization model, units are added or subtracted according to the sensitivity of each unit to obtain the target cutout corresponding to the structure. Then, in the search model, the structure corresponding to the maximum aggregation of bending stiffness, critical buckling load and failure index under failure constraints is searched based on the target cutouts corresponding to different structures, so as to determine the target structure and obtain the target structure with the target cutout (such as Figure 4 The structure shown as an example).

[0072] It is understandable that if Figure 2 As shown (it is understandable that Figure 2Different colors correspond to processing under different models). The bidirectional progressive structural optimization model is based on the maximization of the bending stiffness, critical buckling load and failure index of the target cutout corresponding to each structure under the failure constraint. After the condition is met, the bidirectional progressive structural optimization model is determined to be converged. Then, under the search model, it is judged again that the bending stiffness, critical buckling load and failure index of the corresponding target cutout are maximized under each structure (material). After the condition is met, the search model converges. After both models converge, the target structure with the target cutout is output, which is in the optimal design in terms of material and cutout design.

[0073] Based on the above, the incision optimization method provided in this embodiment constructs a nested two-layer iterative optimization model, determines different structures through the search model, and dynamically optimizes the braided composite material in combination with the agent model (deep neural network model, DNN), generates ABD matrices corresponding to different structures and imports the incision optimization model (specifically, the (DCTWS macro) finite element model), and combines the bidirectional progressive structural optimization model to maximize the bending stiffness, critical buckling load and failure index aggregation to perform incision iterative optimization, iterate until the predetermined optimization target, the model converges and meets the constraints, thereby forming the synchronous optimization of the braided composite material design and the macroscopic structure incision design.

[0074] Embodiment 2: To achieve the above-mentioned purpose, the present invention further provides a computer device, which may also include multiple computer devices, and the computer device may be a smart phone, tablet computer, laptop computer, desktop computer, etc. that executes a program. The computer device of this embodiment includes at least but is not limited to: a memory, a processor, a network interface, etc. that can be interconnected through a system bus for communication.

[0075] In this embodiment, the memory includes at least one type of computer-readable storage medium, and the readable storage medium includes a flash memory, a hard disk, a multimedia card, a card-type memory, etc. In some embodiments, the memory may be an internal storage unit of a computer device, such as a hard disk or a memory of the computer device. In other embodiments, the memory may also be an external storage device of a computer device, such as a plug-in hard disk equipped on the computer device. Of course, the memory may also include both an internal storage unit of the computer device and an external storage device thereof. In this embodiment, the memory is generally used to store an operating system and various application software installed on the computer device, such as the program code of the incision optimization method of Embodiment 1, etc. In addition, the memory may also be used to temporarily store various types of data that have been output or are to be output.

[0076] The processor may be a central processing unit, a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor is generally used to control the overall operation of a computer device. In this embodiment, the processor is used to run program codes stored in a memory or process data, such as implementing the incision optimization method of embodiment one.

[0077] The network interface may include a wireless network interface or a wired network interface, and the network interface is generally used to establish a communication connection between the computer device and other computer devices. For example, the network interface is used to connect the computer device to an external terminal through a network, establish a data transmission channel and a communication connection between the computer device and the external terminal, etc.

[0078] In this embodiment, the incision optimization method / each model can also be divided into one or more program modules, and the one or more program modules are stored in a memory and executed by one or more processors to complete the present invention.

[0079] Embodiment 3: To achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, which includes multiple storage media, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, etc., on which a computer program is stored, and the program realizes the corresponding function when executed by the processor. The computer-readable storage medium of this embodiment is used to store the processing data of the program / each model, and when executed by the processor, the incision optimization method of embodiment 1 is realized.

[0080] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0081] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method.

[0082] It should be noted that the embodiments of the present invention have better practicability and do not impose any form of limitation on the present invention. Any technician familiar with the field may use the technical content disclosed above to change or modify it into an equivalent effective embodiment. However, any modification or equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for optimizing the cut of a braided composite deployable structure, characterized in that: include: Establishing a search model to determine the material optimization variables of the braided composite deployable structure to obtain an initial structure; Calculating an ABD matrix based on the initial structure using a deep neural network model to generate initial structural parameters; Predetermining an initial cut in the initial structure, establishing a finite element model based on the initial structural parameters to determine structural optimization variables, and iteratively optimizing the initial cut using a bidirectional progressive structural optimization model according to the structural optimization variables, wherein the structural optimization variables include bending stiffness, critical buckling load, and failure index; The initial structure is updated based on the material optimization variables under the search model, and the cutout is optimized with the updated structure, and the process is repeated until a predetermined optimization target is met, thereby generating a target structure with a target cutout.

2. The incision optimization method according to claim 1, characterized in that: The predetermined optimization objectives include: for each structure composed of different material optimization variables, after the bidirectional progressive structural optimization model performs cutout optimization to determine the target cutout, the bending stiffness, critical buckling load and failure index aggregation of the target cutout corresponding to each structure are maximized under failure constraints.

3. The incision optimization method according to claim 1, characterized in that: The third generation non-dominated sorting genetic algorithm is used to establish the search model.

4. The incision optimization method according to claim 1, characterized in that: The material optimization variables include material parameters, geometric parameters and ply parameters; The material parameters include the major and minor axis dimensions of the yarn cross section and the yarn spacing dimensions; The set parameters include the number of warp / weft yarns and the yarn interlacing position; The ply parameters include the number of plies and the single-layer ply laying angle.

5. The incision optimization method according to claim 1 or 4, characterized in that: The initial structure is determined and / or updated by using Latin hypercube sampling based on the material optimization variables.

6. The incision optimization method according to claim 1 or 4, characterized in that: The calculating the ABD matrix based on the initial structure using a deep neural network model to generate initial structural parameters includes: Obtain various material optimization variables corresponding to the initial structure; Based on the various material optimization variables of the initial structure, microscopic RVE homogenization analysis and mesoscopic RVE homogenization analysis are performed to calculate the ABD matrix corresponding to the initial structure to generate initial structure parameters.

7. The incision optimization method according to claim 6, characterized in that: The deep neural network model is pre-built and / or trained offline.

8. The incision optimization method according to claim 1, characterized in that: Under the bidirectional progressive structural optimization model, the initial cutout is divided into several units, and the bending stiffness, linear buckling analysis, and sensitivity of the failure index corresponding to each unit are determined; The initial cutout is optimized by adding or reducing units according to the sensitivity, until the bending stiffness, critical buckling load and failure index of the target cutout corresponding to each structure are maximized under the failure constraint.

9. A computer device, characterized in that: The computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the incision optimization method according to any one of claims 1 to 8 when executing the computer program.

10. A computer storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the incision optimization method according to any one of claims 1 to 8 are implemented.

Images