Virtual testing method of hydrogen storage cylinder of fiber reinforced composite material by introducing random wrinkle defects and probability distribution thereof

Abstract

The virtual testing method of the fiber reinforced composite hydrogen storage cylinder with random wrinkle defects and its probability distribution is introduced, which comprises the following steps: S1, a mesoscopic mechanics model of the wrinkle defects in the fiber reinforced composite material is established; S2, a finite element model of the fiber reinforced composite hydrogen storage cylinder is established, and the elements are divided and numbered; S3, based on the MATLAB- Python program mixed programming, a parameterization and cross-scale implantation method of the random wrinkle defects in the fiber reinforced composite hydrogen storage cylinder is established; S4, the virtual testing of the fiber reinforced composite hydrogen storage cylinder with random wrinkle defects and its probability distribution is carried out under internal pressure loading, and the mechanical response output results are obtained. Based on the parameterization and cross-scale method, the modeling complexity of the wrinkle defects in the macroscopic structure is simplified, and the modeling efficiency is improved; the virtual testing of the fiber reinforced composite hydrogen storage cylinder with random wrinkle defects and its probability distribution is realized.

Description

Technical Field

[0001] The present invention belongs to the field of structural mechanics analysis of fiber reinforced composite materials, and particularly relates to a virtual testing method for predicting the mechanical response of a fiber reinforced composite hydrogen storage cylinder by introducing random defects and their probability distribution. Background Art

[0002] Fiber-reinforced composite structures have been used in automotive high-pressure hydrogen storage cylinders due to their high specific strength and high specific stiffness. However, due to the multi-phase, heterogeneous, and rich microstructural characteristics of fiber-reinforced composites, as well as environmental and processing factors during the manufacturing process, manufacturing defects such as wrinkles, delamination, and fiber debonding are inevitable in the composite reinforcement layer of the cylinder. These manufacturing defects vary in severity and spatial distribution in composite cylinders, resulting in uncertainty in the service performance of composite cylinders and making it difficult to predict their mechanical properties. Currently, domestic and international research on fiber-reinforced composite hydrogen storage cylinders generally establishes deterministic cylinder structural models without considering the impact of initial manufacturing defects. The few statistical analyses of composite cylinders that consider initial uncertainty only use discrete values ​​of fiber strength as discrete input parameters. When considering the impact of specific defects such as wrinkles on the mechanical response of fiber-reinforced composites, researchers typically perform localized and refined structural modeling of individual wrinkle defects in the laminate. If detailed defect modeling is required in fiber-reinforced composite hydrogen storage cylinders or other macrostructures, the complexity of structural modeling will be greatly increased, and further consideration of the need to introduce dispersed defect structural modeling will be difficult to achieve. Therefore, given that the defects in the fiber-reinforced layer of the cylinder composite material vary in severity and have the characteristics of probabilistic distribution, establishing a virtual testing method for fiber-reinforced composite hydrogen storage cylinders that introduces random wrinkle defects and their probabilistic distribution has important theoretical significance and engineering application value. Based on this method, highly feasible and efficient modeling and mechanical response virtual testing of defective composite hydrogen storage cylinders can be carried out, thereby reducing the cost of batch testing of composite hydrogen storage cylinders and helping to improve the credibility of analysis results. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to overcome the deficiencies in the prior art and provide a virtual testing method for fiber-reinforced composite hydrogen storage cylinders that introduces random defects and their probability distribution. To solve the above technical problem, the solution of the present invention is:

[0004] A virtual testing method for fiber-reinforced composite hydrogen storage cylinders that introduces random wrinkle defects and their probability distribution includes the following steps:

[0005] S1. Establish a micromechanical model of wrinkle defects in fiber-reinforced composites;

[0006] S2. Build a finite element model of the fiber-reinforced composite hydrogen storage cylinder, divide the units, and number them;

[0007] S3. Based on a MATLAB-Python hybrid program, a parametric, cross-scale method for implanting random wrinkle defects in fiber-reinforced composite hydrogen storage cylinders was established.

[0008] S4. Conduct virtual testing under internal pressure on a fiber-reinforced composite hydrogen storage cylinder that introduces random wrinkle defects and their probability distribution, and obtain mechanical response output results.

[0009] The process S1 specifically includes the following steps:

[0010] S1.1: Geometric model describing wrinkles in fiber-reinforced composites using cosine functions:

[0011]

[0012] Among them, the subscripts Graded and Uniform represent the gradient and uniform fold forms respectively, A and L are the fold amplitude and wavelength respectively, x is the main direction in the ply layer, z is the ply thickness direction, H w is the thickness of the fiber wrinkled area, and H is the thickness of the composite material plate.

[0013] S1.2: When a representative volume element (RVE) contains a complete wrinkle, a micromechanical model of wrinkle defects in fiber-reinforced composites is developed based on a two-step homogenization method, specifically including:

[0014] 1.2.(a) First, divide the wrinkled representative volume element (RVE) into several narrow bands along the in-plane principal direction x of the global coordinate system. Each ply within each narrow band has the same out-of-plane deflection angle. Perform the first homogenization calculation along the ply thickness within each narrow band. At this point, transform the stiffness matrix [Q] of each individual ply in the principal material coordinate system to the global coordinate system:

[0015]

[0016] Where k represents the position of the single layer in the laminate, n is the total number of layers in the laminate, and θ k For the ply direction, [T θ k ] is the angle between the main direction of the material ply in the plane and the main direction of the global coordinate system, θ k The ply transformation matrix.

[0017] 1.2.(b) The assumptions satisfying the first step of homogenization through the thickness direction are that the interlaminar stress is continuous in the z direction and the deformation is coordinated:

[0018]

[0019] Among them, [C aa ] k ,[C ab ] k and [C bb ] k is the stiffness matrix of a single layer of fiber reinforced composite material, C ij (i, j = 1, 2, 3, 4, 5, 6) is the stiffness coefficient, [C * ] is the equivalent stiffness matrix; {σ *} and {ε *} represents stress and strain, and the superscript * Represents the equivalent or average physical quantity obtained by the first step of homogenization.

[0020] 1.2.(c) In the second step, the change in the stiffness matrix caused by the off-plane deflection of each narrow strip is first calculated:

[0021]

[0022] where l and φ l represent the narrowband number and its off-plane deflection angle, m is the total number of narrowbands, [T φ l ] is the narrowband transformation matrix.

[0023] 1.2.(d) Then, a second homogenization step is performed between the narrow bands along the principal in-plane directions. The stresses of adjacent narrow bands are continuous and the deformations are coordinated at the interface:

[0024]

[0025] Among them, [C * ee ] l ,[C * ef ] l and [C * ff ] l is the equivalent stiffness matrix of the entire laminate obtained by the first step of homogenization [C * ], C * ij (i, j = 1, 2, 3, 4, 5, 6) are the equivalent stiffness coefficients; {σ **} and {ε **} represents stress and strain, and the superscript ** It represents the physical quantity equivalent or averaged in the entire RVE after two-step homogenization calculation, [C ** ] is the equivalent stiffness matrix of the wrinkle defect unit.

[0026] S1.3: Based on the above process, a three-dimensional finite element calculation program "Magic Matrix" for the micromechanical model of composite material wrinkle defects was established in MATLAB. By inputting fiber reinforced composite material parameters, layup parameters and wrinkle parameters, the equivalent stiffness matrix of the wrinkle defect unit was calculated.

[0027] The process S2 specifically includes: establishing models of various components of the composite material structure based on the general finite element software ABAQUS, and assembling the components to obtain the entire structure using the assembly module provided by ABAQUS; performing meshing in the mesh module, and outputting the numbers of n units in the composite material structure after meshing, where the units here correspond to representative volume elements (RVEs) randomly sampled as defects; setting constraints and loading conditions in the load module; and then setting the analysis step (Step) and other general properties.

[0028] The process S3 specifically includes the following steps:

[0029] S3.1: Establish a parametric description of wrinkle defects and their distribution in fiber-reinforced composite hydrogen storage cylinders, specifically including:

[0030] 3.1.(a) Extracting the maximum fiber deflection angle θ max As the characteristic parameter of wrinkle defects, it has an equivalent conversion relationship with the amplitude A and wavelength L parameters that describe the geometric structure of the wrinkle:

[0031] A / L=tan(θ max ) / π (6)

[0032] 3.1.(b) Use normal distribution to describe the severity distribution of wrinkle defects:

[0033]

[0034] Where S represents the maximum fiber deflection angle θ max The mean of x μ Represents standard deviation.

[0035] 3.1.(c) Using the specified area Y i According to a certain probability p i Units (numbers) are randomly selected and assigned characteristic defect parameters to achieve random distribution of defects in space. The defects in the structure are regarded as discrete random variables y (non-defect when y=0) and expressed through a certain relationship. Each randomly selected unit (number) will correspond to a defect characteristic parameter, so the random variable y (discretized defect unit) in the sample space Y (i.e., the sum of the discretized units) can be expressed as:

[0036] g(y)=P(y≠0,y∈Y i )=p i ,(Y=Y1,Y2,...,Y N ) (8)

[0037] S3.2: On the MATLAB platform, establish a program module that generates wrinkle defect characteristic parameters that conform to a certain distribution and implements their random assignment in the unit. In the module, calculate the corresponding equivalent stiffness matrix for the unit assigned with the defect characteristic parameters. The specific implementation method is as follows:

[0038] 3.2.(a) Based on the MATLAB platform, the randsample function is used to randomly generate n numbers in the n unit numbers of the composite material structure. d The defect unit numbers are used as random number inputs for the defect space distribution that meets certain distribution requirements:

[0039] G(n d )=randsample(n,n d ) (9)

[0040] 3.2.(b) Generate n through normrnd function d A wrinkle defect characteristic parameter θ that meets certain mean and expectation max Input parameters as defect structure variables:

[0041] θ max =normrnd(S,x μ ,n d ) (10)

[0042] 3.2.(c) Traverse all units and determine whether they are defective units based on the unit number. If they are defective units, call the three-dimensional finite element calculation program "Magic Matrix" for the microscopic mechanical model of composite material wrinkle defects established according to process S1, introduce material parameters, composite material layup parameters and defect characteristic parameters, calculate the equivalent stiffness matrix of the defective unit and save it; if they are not defective units, calculate the unit stiffness matrix based on the composite material layup parameters and save it.

[0043] S3.3: Based on the secondary development function of ABAQUS-Python, a cross-scale modeling method is established to introduce the mathematical representation of defective units into the composite material structure model. A Python script program submodule is established. By starting the MATLAB engine for Python, the MATLAB random wrinkle defect and its constitutive generation program module is called, and the degenerate stiffness matrix of the wrinkle defect unit or the stiffness matrix of the non-defective unit output by MATLAB is read. Then, the ABAQUS object model data (Mdb) is accessed through the Python script to enter the Python interpreter, and the Model is accessed to realize the material property assignment of the wrinkle defect equivalent material parameters in the ModelDatabase. The stiffness matrices of the non-defective units and defective units in MATLAB are batch-entered into the corresponding units of the composite material structure model under the ABAQUS environment, thereby realizing data interaction with ABAQUS.

[0044] Process S4 specifically involves assembling a global stiffness matrix in ABAQUS / CAE, setting loading pressure parameters, and running a virtual test program for a fiber-reinforced composite hydrogen storage cylinder that introduces random wrinkle defects and their probability distribution. The ABAQUS / Standard solver then calculates the cylindrical coordinate components of the cylinder's surface mechanical response for each distribution parameter. Running this program once is equivalent to conducting a virtual structural test of the fiber-reinforced composite hydrogen storage cylinder. The program is then run a set number of times to obtain the corresponding virtual test mechanical response results.

[0045] The advantages and beneficial effects of the present invention are:

[0046] Manufacturing defects such as wrinkles of varying severity and uneven spatial distribution in the fiber reinforcement layer of composite hydrogen storage cylinders can lead to uncertainty in the cylinder's mechanical properties. Compared with existing structural modeling methods that use local refinement of a single wrinkle defect, the present invention establishes a fiber-reinforced composite hydrogen storage cylinder model based on a parameterized, cross-scale approach that takes into account the severity distribution and spatial random distribution of wrinkle defects. This greatly simplifies the modeling complexity of wrinkle defects in the macrostructure and improves modeling efficiency; it also enables virtual testing of fiber-reinforced composite hydrogen storage cylinders that introduce random wrinkle defects and their probability distribution. Based on this method, it is possible to carry out highly feasible and efficient modeling of composite hydrogen storage cylinders containing dispersed defects and achieve virtual prediction of mechanical responses, which helps to improve the credibility of the analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings described herein are used to provide further understanding of the present invention and constitute a part of this application. The illustrative examples of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0048] Figure 1 Schematic diagram of a two-step uniformization method for wrinkle defects in an embodiment of the present invention;

[0049] Figure 2 A finite element model of a fiber-reinforced composite hydrogen storage cylinder 1 / 2 established in an embodiment of the present invention and a schematic diagram of its constraints and loading;

[0050] Figure 3 This is a flow chart of the analysis procedure for the composite material hydrogen storage cylinder with introduced dispersion defects disclosed in the present invention;

[0051] Figure 4 Schematic diagram of a composite material layer model of a gas cylinder with randomly distributed defects introduced in an embodiment of the present invention;

[0052] Figure 5 is a radial displacement response diagram of the gas cylinder surface in an embodiment of the present invention;

[0053] Figure 6 is a circumferential displacement response diagram of the gas cylinder surface in an embodiment of the present invention;

[0054] Figure 7 This is a graph showing the axial displacement response of the gas cylinder surface in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] First, it should be noted that the present invention is an application of computer technology in the field of structural mechanics analysis of fiber-reinforced composite materials. The implementation of the present invention involves the application of multiple software functional modules. The applicant believes that after carefully reading the application documents and accurately understanding the implementation principles and objectives of the present invention, and in combination with existing publicly known technologies, those skilled in the art can fully utilize their software programming skills to implement the present invention. All matters mentioned in the present application documents fall within this scope, and the applicant will not enumerate them one by one.

[0056] The following describes the effects of the present invention with reference to the accompanying drawings and specific implementation examples based on the method of the present invention:

[0057] S1. Establish a micromechanical model for wrinkle defects in fiber-reinforced composites. This includes the following steps:

[0058] S1.1: In a specific implementation example, the defects in the fiber-reinforced composite material layer of the hydrogen storage cylinder are in the form of uniform wrinkles with the same waveform in the thickness direction.

[0059] z Uniform (x, z)=Acos(2πxL),|x|≤L / 2(1)

[0060] Where A and L are the wrinkle amplitude and wavelength, respectively, x is the main direction within the ply, and z is the ply thickness direction.

[0061] S1.2: By Figure 1 The two-step homogenization method shown in the figure establishes a micromechanical model of wrinkle defects in fiber-reinforced composites, which includes:

[0062] 1.2.(a) First, divide the wrinkled representative volume element (RVE) into several narrow bands along the in-plane principal direction x of the global coordinate system. Each ply within each narrow band has the same out-of-plane deflection angle. Perform the first homogenization calculation along the ply thickness within each narrow band. At this point, transform the stiffness matrix [Q] of each individual ply in the principal material coordinate system to the global coordinate system:

[0063]

[0064] Where k represents the position of the single layer in the laminate, n is the total number of layers in the laminate, and θ k For the ply direction, [T θ k ] is the angle between the main direction of the material ply in the plane and the main direction of the global coordinate system, θ k The ply transformation matrix.

[0065] 1.2.(b) The assumptions satisfying the first step of homogenization through the thickness direction are that the interlaminar stress is continuous in the z direction and the deformation is coordinated:

[0066]

[0067] Among them, [C aa ] k ,[C ab ] k and [C bb ] k is the stiffness matrix of a single layer of fiber reinforced composite material, C ij (i, j = 1, 2, 3, 4, 5, 6) is the stiffness coefficient, [C * ] is the equivalent stiffness matrix; {σ *} and {ε *} represents stress and strain, and the superscript * Represents the equivalent or average physical quantity obtained by the first step of homogenization.

[0068] 1.2.(c) In the second step, the change in the stiffness matrix caused by the off-plane deflection of each narrow strip is first calculated:

[0069]

[0070] where l and φ l represent the narrowband number and its off-plane deflection angle, m is the total number of narrowbands, [T φ l ] is the narrowband transformation matrix.

[0071] 1.2.(d) A second homogenization step is then performed between the strips along the principal directions within the pavement surface, so that stress continuity and deformation coordination are achieved at the interface between adjacent strips:

[0072]

[0073] Among them, [C * ee ] l ,[C * ef ] l and [C * ff ] l is the equivalent stiffness matrix of the entire laminate obtained by the first step of homogenization [C * ], C * ij (i, j = 1, 2, 3, 4, 5, 6) are the equivalent stiffness coefficients; {σ **} and {ε **} represents stress and strain, and the superscript ** It represents the physical quantity equivalent or averaged in the entire RVE after two-step homogenization calculation, [C ** ] is the equivalent stiffness matrix of the wrinkle defect unit.

[0074] S1.3: Based on the above process, a three-dimensional finite element calculation program "Magic Matrix" for the micromechanical model of composite material wrinkle defects was established in MATLAB. By inputting fiber reinforced composite material parameters, layup parameters and wrinkle parameters, the equivalent stiffness matrix of the wrinkle defect unit was calculated.

[0075] S2. Build a finite element model of the fiber-reinforced composite hydrogen storage cylinder, divide the units and number them:

[0076] A specific implementation example uses a 50L, 70MPa nominal working pressure, aluminum liner with a closed end and fiber-reinforced composite hydrogen storage cylinder. The cylinder is 1400mm long and has an inner diameter of 233mm. The cylinder's composite reinforcement layer is 21.76mm thick and is laid alternately in spiral and hoop winding. A 1 / 2 finite element model of the cylinder is established in ABAQUS / CAE. Displacement boundary conditions are applied to the open end face of the cylinder's symmetry surface, and a uniformly distributed internal pressure load is applied to the inner surface. Figure 2 The composite reinforcement layer is divided into 2970 numbered elements, using the three-dimensional solid element C3D8 as the element properties. Other general properties for the analysis step are then set.

[0077] S3. Based on MATLAB-Python program mixing, establish Figure 3The parametric, cross-scale implantation method for random wrinkle defects in fiber-reinforced composite hydrogen storage cylinders is shown. The specific steps include:

[0078] S3.1: Establish a parametric description of wrinkle defects and their distribution in fiber-reinforced composite hydrogen storage cylinders, specifically including:

[0079] 3.1.(a) Extracting the maximum fiber deflection angle θ max As the characteristic parameter of wrinkle defects, it has an equivalent conversion relationship with the amplitude A and wavelength L parameters that describe the geometric structure of the wrinkle:

[0080] A / L=tan(θ max ) / π (6)

[0081] 3.1.(b) In a specific embodiment, the maximum fiber deflection angle θ of the wrinkle is used. max Conforms to X=N(9,2.8 2 ) Distribution of wrinkle defect severity described by normal distribution;

[0082] 3.1.(c) In the specific implementation example, all the elements in the entire area of ​​the fiber reinforced composite material layer of the gas cylinder are measured according to p i =5.05% probability (ie 150 defective units) are randomly assigned characteristic defect parameters to achieve the spatial distribution of defects.

[0083] S3.2: Based on MATLAB programming, establish a program module for generating random wrinkle defects and their constitutive structures. The specific implementation method is as follows:

[0084] 3.2.(a) Based on the MATLAB platform, the randsample function is used to randomly generate n numbers in the n unit numbers of the composite material structure. d The defect unit numbers are used as random number inputs for the defect space distribution that meets certain distribution requirements:

[0085] G(n d )=randsample(n,n d ) (7)

[0086] 3.2.(b) Generate n through normrnd function d A wrinkle defect characteristic parameter θ that meets certain mean and expectation max Input parameters as defect structure variables:

[0087] θ max =normrnd(S,x μ ,n d ) (8)

[0088] 3.2.(c) Traverse all units and determine whether they are defective units based on the unit number. If they are defective units, call the three-dimensional finite element program interface "Magic Matrix" for the microscopic mechanical model of composite material wrinkle defects established according to process S1, introduce the composite material layup parameters and defect characteristic parameters, calculate the degraded stiffness matrix of the defective unit and save it; if they are not defective units, calculate the unit stiffness matrix based on the composite material layup parameters and save it.

[0089] S3.3: Based on the secondary development function of ABAQUS-Python, a cross-scale modeling method is established to introduce the mathematical representation of defect units into the composite material structure model. A Python script program submodule is established, and by starting the MATLAB engine for Python, the random wrinkle defect and its constitutive generation program module of MATLAB is called, and the degraded stiffness matrix of the wrinkle defect unit or the stiffness matrix of the defect-free unit output by MATLAB is read. Then, the ABAQUS object model data (Mdb) is accessed through the Python script to enter the Python interpreter, and the Model is accessed to realize the material property assignment of the wrinkle defect equivalent material parameters in the ModelDatabase, and the stiffness matrices of the non-defective units and defective units in MATLAB are batch-inputted into the corresponding units of the composite material structure model under the ABAQUS environment, thereby realizing data interaction with ABAQUS. The above program is set to execute N times through a for loop, and a virtual test model with N random defect spatial distributions of the fiber reinforced composite material layer of the gas cylinder that meets the same defect distribution is obtained, such as Figure 4 shown.

[0090] S4. Assemble the global stiffness matrix in ABAQUS / CAE. Perform a virtual test under internal pressure loading on a fiber-reinforced composite hydrogen storage cylinder model with random wrinkle defects and their probability distribution in the ABAQUS / Standard solver to obtain the mechanical response output results:

[0091] Set the loading pressure to 70 MPa and run the program developed for the virtual testing method for a fiber-reinforced composite hydrogen storage cylinder with random wrinkle defects and their probability distribution, as described in steps S1 through S3. Calculate the cylindrical coordinate components of the cylinder's surface mechanical response for each distribution parameter. Running this calculation program once is equivalent to performing a virtual structural test of the fiber-reinforced composite hydrogen storage cylinder. Set the program's number of loops to generate the corresponding number of virtual test mechanical response results. Figure 5 、 Figure 6 and Figure 7These are the radial displacement, circumferential displacement, and axial displacement response fields in the cylindrical coordinate system obtained in one of the virtual tests of a fiber-reinforced composite hydrogen storage cylinder with random wrinkle defects.

[0092] Finally, it should be noted that the above examples are merely specific embodiments of the present invention. Obviously, the present invention is not limited to the above examples and is subject to numerous variations. All variations that can be directly derived or conceived by a person of ordinary skill in the art from the disclosure of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A virtual testing method for fiber-reinforced composite hydrogen storage cylinders that introduces random wrinkle defects and their probability distribution is characterized by: The following processes are included: S1. Establish a micromechanical model of wrinkle defects in fiber-reinforced composites; S2. Build a finite element model of the fiber-reinforced composite hydrogen storage cylinder, divide the units, and number them; S3. Based on a MATLAB-Python hybrid program, a parametric, cross-scale method for implanting random wrinkle defects in fiber-reinforced composite hydrogen storage cylinders was established. Specifically include: S3.1: Establish a parametric description of wrinkle defects and their distribution in fiber-reinforced composite hydrogen storage cylinders, specifically including: 3.1.(a) Extracting the maximum fiber deflection angle θ max As the characteristic parameter of wrinkle defects, it has an equivalent conversion relationship with the amplitude A and wavelength L parameters that describe the geometric structure of the wrinkle: A / L=tan(θ max ) / π (6) 3.1.(b) Use normal distribution to describe the severity distribution of wrinkle defects: Where S represents the maximum fiber deflection angle θ max The mean of x μ represents the standard deviation; 3.1.(c) Using the specified area Y i According to a certain probability p i Units are randomly selected and assigned characteristic defect parameters to achieve random distribution of defects in space. The defects in the structure are regarded as discrete random variables y, where y = 0 is non-defect, and are expressed through a certain relationship. Each randomly selected unit will correspond to a defect characteristic parameter, so the discrete random variable y in the sample space Y is expressed as: g(y)=P(y≠0,y∈Y i )=p i ,(Y=Y1,Y2,...,Y N ) (8) S3.2: Based on the MATLAB platform, a program module is established to generate wrinkle defect characteristic parameters that conform to a certain distribution and implement their random assignment in the unit. In the module, the corresponding equivalent stiffness matrix is ​​calculated for the unit assigned with the defect characteristic parameters. The specific implementation method is as follows: 3.2.(a) Based on the MATLAB platform, the randsample function is used to randomly generate n numbers in the n unit numbers of the composite material structure. d The defect unit numbers are used as random number inputs for the defect space distribution that meets certain distribution requirements: G(n d )=randsample(n,n d ) (9) 3.2.(b) Generate n through normrnd function d A wrinkle defect characteristic parameter θ that meets certain mean and expectation max Input parameters as defect structure variables: i max =normrnd(S,x μ ,n d ) (10) 3.2.(c) Loop through all cells and determine whether they are defective based on the cell number. If they are defective, call the "Magic Matrix" 3D finite element calculation program for the composite wrinkle defect micromechanics model established in step S1. Enter the material parameters, composite layup parameters, and defect characteristic parameters, calculate the equivalent stiffness matrix of the defective cell, and save it. If they are not defective, calculate the cell stiffness matrix based on the composite layup parameters and save it. S3.3: Based on the secondary development function of ABAQUS-Python, a cross-scale modeling method is established to introduce the mathematical representation of defective units into the composite material structure model; a Python script program submodule is established to call the MATLAB random wrinkle defect and its constitutive generation program module by starting the MATLAB engine for Python, and read the degenerate stiffness matrix of the wrinkle defect unit or the stiffness matrix of the non-defective unit output by MATLAB; then, the Python script accesses the ABAQUS object model data Mdb to enter the Python interpreter, accesses the Model to realize the material property assignment of the wrinkle defect equivalent material parameters in the Model Database, and batch inputs the stiffness matrices of the non-defective units and defective units in MATLAB into the corresponding units of the composite material structure model in the ABAQUS environment, thereby realizing data interaction with ABAQUS; S4. Conduct virtual testing under internal pressure on a fiber-reinforced composite hydrogen storage cylinder that introduces random wrinkle defects and their probability distribution, and obtain mechanical response output results.

2. The virtual testing method for fiber reinforced composite hydrogen storage cylinders with random wrinkle defects and their probability distribution as claimed in claim 1, characterized in that: Process S1 specifically includes the following steps: S1.1: Geometric model describing wrinkles in fiber-reinforced composites using cosine functions: Among them, the subscripts Graded and Uniform represent the gradient and uniform fold forms respectively, A and L are the fold amplitude and wavelength respectively, x is the main direction in the ply layer, z is the ply thickness direction, H w is the thickness of the fiber wrinkling area, H is the thickness of the composite sheet; S1.2: When a representative volume element (RVE) contains a complete wrinkle, a micromechanical model of wrinkle defects in fiber-reinforced composites is established based on a two-step homogenization method, specifically including: 1.2.(a) First, divide the wrinkle-containing representative volume element RVE into several narrow bands along the in-plane principal direction x of the global coordinate system. Each ply within each narrow band has the same out-of-plane deflection angle. Perform the first homogenization calculation along the ply thickness direction within each narrow band. At this point, transform the stiffness matrix [Q] of each individual ply in the principal material coordinate system to the global coordinate system: Where k (k = 1, 2, ..., n) represents the position of the single layer in the laminate, n is the total number of layers in the laminate, θ k For the ply direction, [T θ k ] is the angle between the main direction of the material ply in the plane and the main direction of the global coordinate system, θ k The ply transformation matrix; 1.2.(b) The assumptions satisfying the first step of homogenization through the thickness direction are that the interlaminar stress is continuous in the z direction and the deformation is coordinated: in, [C aa ] k ,[C ab ] k and [C bb ] k is the stiffness matrix of a single layer of fiber reinforced composite material, C ij (i, j = 1, 2, 3, 4, 5, 6) is the stiffness coefficient, [C * ] is the equivalent stiffness matrix; {σ * } and {ε * } represents stress and strain, and the superscript * represents the equivalent or average physical quantity obtained by the first step of homogenization; 1.2.(c) In the second step, the change in the stiffness matrix caused by the off-plane deflection of each narrow strip is first calculated: where l and φ l represent the narrowband number and its off-plane deflection angle, m is the total number of narrowbands, [Tφ l ] is the narrowband conversion matrix; 1.2.(d) Then, a second homogenization step is performed between the narrow bands along the principal in-plane directions. The stresses of adjacent narrow bands are continuous and the deformations are coordinated at the interface: in, [C * ee ] l ,[C * ef ] l and [C * ff ] l is the equivalent stiffness matrix of the entire laminate obtained by the first step of homogenization [C * ], C * ij (i, j = 1, 2, 3, 4, 5, 6) are the equivalent stiffness coefficients; {σ ** } and {ε ** } represents stress and strain, superscript ** It represents the physical quantity equivalent or averaged in the entire RVE after two-step homogenization calculation, [C ** ] is the equivalent stiffness matrix of the wrinkle defect unit; S1.3: Based on the above process, a three-dimensional finite element calculation program "Magic Matrix" for the micromechanical model of composite material wrinkle defects was established in MATLAB. By inputting fiber reinforced composite material parameters, layup parameters, and wrinkle parameters, the equivalent stiffness matrix of the wrinkle defect unit was calculated.

3. The virtual testing method for fiber reinforced composite hydrogen storage cylinders with random wrinkle defects and their probability distribution as claimed in claim 1, characterized in that: Process S2 specifically includes: establishing models of various components of the composite structure based on the general finite element software ABAQUS, and assembling the components to obtain the entire structure using the Assembly module provided by ABAQUS; performing meshing in the Mesh module, and outputting the numbers of n units in the composite structure after meshing, where the units correspond to the representative volume units RVE randomly selected as defects; setting constraints and loading conditions in the Load module; and then setting the analysis step Step and other general properties.

4. The virtual testing method for fiber reinforced composite hydrogen storage cylinders with random wrinkle defects and their probability distribution as claimed in claim 1, characterized in that: Process S4 includes: assembling the overall stiffness matrix in ABAQUS / CAE, setting the loading pressure parameters, and running the virtual test program of the fiber-reinforced composite material hydrogen storage cylinder with the introduction of random wrinkle defects and their probability distribution, and calculating the components of the cylinder surface mechanical response in the cylindrical coordinate system under various distribution parameters in the ABAQUS / Standard solver; running the above calculation program once is equivalent to performing a structural virtual test of the fiber-reinforced composite material hydrogen storage cylinder; setting the number of program loop runs, and running the corresponding number of virtual test mechanical response results.

Images