Micromechanics-based composite material damage prediction method and related equipment

Abstract

The invention relates to the field of materials, and particularly provides a composite material damage prediction method based on mesomechanics and related equipment, and the method comprises the steps: obtaining macromechanics response data and microstructure feature data of a composite material; constructing a multi-mechanism damage prediction model based on the microstructure feature data through a mesomechanics theory; performing inversion calibration on the model parameters based on macromechanics response data, and verifying the effectiveness of the multi-mechanism damage prediction model through experimental data; and based on the calibrated multi-mechanism damage prediction model, predicting the initial damage strain of the composite material under different loading conditions and microstructures. According to the method, the macromechanics response data and the microstructure feature data of the composite material are obtained, the multi-mechanism damage prediction model is constructed in combination with the micromechanics theory, and model parameters are calibrated and verified through experimental data, so that accurate prediction of the initial damage strain of the composite material under different loading conditions and microstructures is realized.

Description

Technical Field

[0001] The present application relates to the field of material technology, and in particular to a composite material damage prediction method based on micromechanics and related equipment. Background Art

[0002] Composite materials are widely used in aerospace, defense, and other fields due to their excellent mechanical properties and broad application prospects. However, during use, composite materials are inevitably subjected to external loads, which can cause internal damage in the material, thereby affecting its overall performance and service life. Therefore, accurately predicting the damage initiation strain of composite materials is of great significance for assessing their reliability and optimizing their design.

[0003] At present, the research methods for damage prediction of composite materials are mainly divided into three categories: macroscopic experimental methods, microscopic characterization techniques and theoretical model research. Among them, macroscopic experimental methods use experimental means such as uniaxial tension and compression to study the stress-strain behavior of composite materials under different loading conditions, thereby determining the damage threshold strain; microscopic characterization techniques use X-ray computed tomography (CT), scanning electron microscopy (SEM) and other technologies to characterize the microstructure of composite materials and obtain parameters such as particle size, shape and volume fraction; theoretical model research attempts to establish a damage prediction model based on micromechanics theory for prediction.

[0004] However, existing technologies have several drawbacks. For example, macroscopic experimental methods can only provide macroscopic information and cannot fully reveal the impact of microstructure on damage. Microscopic characterization techniques are still limited to describing the microstructure, lacking a direct correlation with macroscopic properties, making them difficult to use for damage prediction. Theoretical models fail to fully consider the complexity of composite material microstructures. Therefore, existing technologies have significant limitations in composite material damage prediction. They cannot fully and comprehensively consider the complexity of microstructure and the effects of multiple factors, resulting in insufficient prediction accuracy and limited applicability.

[0005] The preceding description is intended to provide general background information and does not necessarily constitute prior art. Summary of the Invention

[0006] The embodiments of the present application provide a composite material damage prediction method and related equipment based on micromechanics, which can accurately predict the damage initiation strain of the composite material under different loading conditions and microstructures.

[0007] In a first aspect, the embodiments of the present application provide a composite material damage prediction method based on micromechanics, comprising:

[0008] Obtain macroscopic mechanical response data and microstructural characteristic data of composite materials;

[0009] constructing a multi-mechanism damage prediction model corresponding to the composite material based on the microstructure characteristic data through micromechanics theory;

[0010] Inversely calibrate the model parameters of the multi-mechanism damage prediction model based on the macroscopic mechanical response data, and verify the effectiveness of the multi-mechanism damage prediction model through experimental data;

[0011] Based on the calibrated multi-mechanism damage prediction model, the damage onset strain of the composite material under different loading conditions and microstructures is predicted.

[0012] Optionally, in some embodiments of the present application, the macroscopic mechanical response data includes the stress-strain relationship under different loading rates and sample sizes, and the microstructural characteristic data includes the size distribution, shape parameters, and spatial distribution density of the filling particles. Then, obtaining the macroscopic mechanical response data and microstructural characteristic data of the composite material includes:

[0013] Uniaxial tensile tests were performed on composite material specimens, and multiple sets of stress-strain curves were obtained by changing the loading rate and specimen size.

[0014] The pseudo-strain conversion method is used to eliminate the viscoelastic stress relaxation effect, and the damage initiation strain threshold is determined by analyzing the nonlinear deviation point of the stress-pseudo-strain curve.

[0015] The three-dimensional imaging of composite materials was performed using micro-computed tomography to obtain the spatial distribution and morphological data of the filling particles;

[0016] The particle morphology was parametrically characterized based on the equivalent ellipsoid model, and the principal axis length, slenderness ratio and volume fraction of the particles were calculated.

[0017] Optionally, in some embodiments of the present application, constructing a multi-mechanism damage prediction model corresponding to the composite material based on the microstructure feature data using micromechanics theory includes:

[0018] The homogenized modulus of the composite material is calculated using the equivalent inclusion theory, and the time-varying viscoelastic behavior of the matrix is ​​treated in combination with the Laplace transform.

[0019] A statistical failure criterion including matrix tearing, particle fracture and interface dewetting is established. The failure criterion is based on Weibull distribution and modified strength criterion.

[0020] Optionally, in some embodiments of the present application, the statistical failure criterion includes:

[0021] A matrix tearing criterion based on a modified generalized von Mises yield criterion, wherein the matrix tearing criterion introduces an influence factor of hydrostatic pressure on matrix yield strength;

[0022] A particle fracture criterion based on the Mohr-Coulomb yield criterion, wherein the particle fracture criterion modifies the Weibull distribution parameter in combination with the particle volume effect;

[0023] A dewetting criterion based on the interface normal stress is defined as a randomly distributed threshold value of the interface strength.

[0024] Optionally, in some embodiments of the present application, the inverse calibration of the model parameters of the multi-mechanism damage prediction model based on the macroscopic mechanical response data includes:

[0025] Defining an optimization objective function including interface strength, matrix viscoelastic parameters, and statistical damage parameters, wherein the objective function is based on the relative error between the experimental damage onset strain and the model prediction value;

[0026] A hybrid optimization strategy combining a multi-island genetic algorithm and a pattern search algorithm is adopted to perform global search and local adjustment on the model parameters.

[0027] Optionally, in some embodiments of the present application, verifying the effectiveness of the multi-mechanism damage prediction model through experimental data includes:

[0028] The experimental data are divided into a model parameter optimization dataset and a model validation dataset;

[0029] By comparing the predicted values ​​of the multi-mechanism damage prediction model with the experimental values, the effectiveness and accuracy of the multi-mechanism damage prediction model are verified.

[0030] Optionally, in some embodiments of the present application, the prediction of the damage onset strain of the composite material under different loading conditions and microstructures based on the calibrated multi-mechanism damage prediction model includes:

[0031] By homogenizing the representative volume unit, a quantitative correlation between the specimen size and the damage threshold is established;

[0032] Based on the mapping model between microstructural parameters and macroscopic loading conditions, the damage initiation strain under different particle distributions and loading rates is predicted.

[0033] In a second aspect, an embodiment of the present application provides a composite material damage prediction device based on micromechanics, comprising:

[0034] Data acquisition module, used to obtain macroscopic mechanical response data and microstructural characteristic data of composite materials;

[0035] A model building module, for building a multi-mechanism damage prediction model corresponding to the composite material based on the microstructure characteristic data by using micromechanics theory;

[0036] a model adjustment module, configured to perform inverse calibration on the model parameters of the multi-mechanism damage prediction model based on the macroscopic mechanical response data, and verify the validity of the multi-mechanism damage prediction model through experimental data;

[0037] The damage prediction module is used to predict the damage initiation strain of the composite material under different loading conditions and microstructures based on the calibrated multi-mechanism damage prediction model.

[0038] In a third aspect, an embodiment of the present application further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the composite material damage prediction method based on micromechanics as described in the first aspect are performed.

[0039] In a fourth aspect, an embodiment of the present application further provides a readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the composite material damage prediction method based on micromechanics as described in the first aspect.

[0040] The present application provides a composite material damage prediction method and related equipment based on micromechanics. First, macroscopic mechanical response data and microstructural feature data are obtained to provide basic input for subsequent modeling. Then, based on the micromechanics theory, the microstructural features are linked to the macromechanical behavior to construct a damage prediction model that can comprehensively consider multiple damage mechanisms. Then, the model parameters are inversely calibrated through experimental data to ensure the accuracy of the model parameters. At the same time, the effectiveness of the model is verified through experimental data to ensure the prediction accuracy of the model under different conditions. Finally, the calibrated model is used to predict the damage onset strain of the composite material under different loading conditions and microstructures, providing a scientific basis for the performance optimization and design of the material. It can be seen that the present application achieves accurate prediction of the damage onset strain of the composite material under different loading conditions and microstructures by comprehensively considering the complexity of the microstructure and the multi-factor coupling effect, thereby improving the accuracy and applicability of composite material damage prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 This is an application environment diagram of the composite material damage prediction method based on micromechanics provided in an embodiment of the present application;

[0043] Figure 21 is a flow chart of a composite material damage prediction method based on micromechanics provided in an embodiment of the present application;

[0044] Figure 3 is a schematic diagram of a process for calculating damage initiation strain provided in an embodiment of the present application;

[0045] Figure 4 Schematic diagram of the structure of a composite material damage prediction device based on micromechanics provided in an embodiment of the present application;

[0046] Figure 5 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of systems and methods consistent with aspects of the present application, as detailed in the appended claims.

[0048] It should be noted that, in this document, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive descriptions such as inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.

[0049] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0050] In the subsequent description, the use of suffixes such as "module", "component" or "unit" to represent elements is only for the purpose of facilitating the description of the present application and has no specific meaning. Therefore, "module", "component" or "unit" can be used interchangeably.

[0051] In order to solve the above-mentioned technical problems and overcome the defects of the existing technology, the embodiments of the present application provide a composite material damage prediction method and related equipment based on micromechanics. By comprehensively considering the complexity of the microstructure and the multi-factor coupling effect, it can accurately predict the damage initiation strain of the composite material under different loading conditions and microstructures, thereby improving the accuracy and applicability of composite material damage prediction.

[0052] Figure 1 FIG. 1 is an application environment diagram of a composite material damage prediction method based on micromechanics in an embodiment. Figure 1 , the composite material damage prediction method based on micromechanics is applied to a composite material damage prediction system based on micromechanics. The composite material damage prediction system based on micromechanics includes a terminal 110 and a server 120. The terminal 110 and the server 120 are connected via a network. The terminal 110 can be a desktop terminal or a mobile terminal. The mobile terminal can be at least one of a mobile phone, a tablet computer, a laptop computer, etc. The server 120 can be implemented as an independent server or a server cluster composed of multiple servers. The terminal 110 is used to obtain the macroscopic mechanical response data and microstructural characteristic data of the composite material; based on the microstructural characteristic data through the micromechanics theory, a multi-mechanism damage prediction model corresponding to the composite material is constructed; based on the macroscopic mechanical response data, the model parameters of the multi-mechanism damage prediction model are inversely calibrated, and the effectiveness of the multi-mechanism damage prediction model is verified through experimental data; based on the calibrated multi-mechanism damage prediction model, the damage initiation strain of the composite material under different loading conditions and microstructures is predicted.

[0053] See also Figure 2 , Figure 2 : This is a flow chart of a composite material damage prediction method based on micromechanics provided in one embodiment of the present application. This embodiment mainly uses the composite material damage prediction method based on micromechanics as an example to illustrate the application of the composite material damage prediction method based on micromechanics to computer equipment. The composite material damage prediction method based on micromechanics provided in one embodiment of the present application may specifically include the following steps:

[0054] S1. Obtain macroscopic mechanical response data and microstructural characteristic data of composite materials;

[0055] Specifically, for step S1, first, a uniaxial tensile test is performed to obtain macroscopic mechanical response data of the composite material, such as the stress-strain relationship under different loading rates and sample sizes. In the experiment, a computer servo high and low temperature tensile testing machine is used for testing, and a tensile fixture suitable for bone-shaped specimens is designed. By changing the loading rate (e.g., 0.001s -1 to 1.0s -1) and specimen dimensions (e.g., 8 mm, 11.3 mm, and 25.0 mm diameters) to obtain multiple sets of stress-strain curves. Microstructural data of the composite material is then acquired. For example, three-dimensional imaging of the composite specimen using micro-computed tomography (Micro CT) is used to obtain spatial distribution and morphological data of the filler particles. Image processing techniques are used to reduce noise, segment, and perform three-dimensional reconstruction of the two-dimensional slice images obtained by CT scanning to extract morphological parameters of the filler particles, such as the major axis length and slenderness of the ellipsoid.

[0056] S2. Construct a multi-mechanism damage prediction model for composite materials based on microstructural characteristic data using micromechanics theory;

[0057] Specifically, for step S2, the Mori-Tanaka (MT) method is used to equate the filler particles to rotating ellipsoids embedded in the matrix, and the effective modulus of the composite material is calculated. The time-varying viscoelastic behavior of the matrix is ​​processed in combination with the Laplace transform, and a multi-mechanism damage prediction model is established that includes matrix tearing, particle fracture, and interface dewetting. The model assumes that the matrix is ​​an isotropic linear viscoelastic material and the particles are isotropic linear elastic materials. The Eshelby equivalent inclusion theory is used to consider the effects of particle size distribution, shape parameters, and spatial distribution density on the mechanical behavior of the composite material.

[0058] S3. Inversely calibrate the model parameters of the multi-mechanism damage prediction model based on macroscopic mechanical response data, and verify the effectiveness of the multi-mechanism damage prediction model using experimental data.

[0059] Specifically, for step S3, an optimization objective function is defined that includes interface strength, matrix viscoelastic parameters, and statistical damage parameters, with the relative error between the experimental damage onset strain and the model prediction as the benchmark. A hybrid optimization strategy combining a multi-island genetic algorithm (MIGA) and a pattern search algorithm is used to perform global search and local adjustment of model parameters. Furthermore, the experimental data is divided into a model parameter optimization dataset and a model validation dataset. The validity and accuracy of the model are verified by comparing the model predictions with the experimental values.

[0060] S4. Based on the calibrated multi-mechanism damage prediction model, predict the damage onset strain of composite materials under different loading conditions and microstructures;

[0061] Specifically, for step S4, a quantitative correlation between sample size and damage threshold is established by homogenizing the representative volume element (RVE). Based on a mapping model between microstructural parameters and macroscopic loading conditions, the damage onset strain under different particle distributions and loading rates is predicted. The calibrated multi-mechanism damage prediction model can be used to predict the damage onset strain of composite materials under different loading rates (e.g., low to high speed), different sample sizes (e.g., small to large), and different particle distributions (e.g., uniform to non-uniform).

[0062] It can be seen that this embodiment provides rich input data for subsequent model construction through multi-dimensional macro and micro data acquisition, ensuring the accuracy and reliability of the prediction model; through the introduction of micromechanics theory, the model can more accurately describe the damage mechanism and its evolution law inside the composite material, and the model comprehensively considers the influence of multiple factors such as particle size, shape, distribution density and matrix viscoelasticity, significantly improving the comprehensiveness and accuracy of damage prediction; the model parameters are inverted and calibrated through experimental data to ensure the accuracy of the model parameters, and the effectiveness of the model is verified through experimental data to ensure the prediction accuracy of the model under different conditions; finally, the calibrated model is used to predict the damage onset strain of the composite material under different loading conditions and microstructures, providing a scientific basis for material performance optimization and design.

[0063] Optionally, in some embodiments, the macroscopic mechanical response data may specifically include the stress-strain relationship under different loading rates and sample sizes, and the microstructural characteristic data may specifically include the size distribution, shape parameters, and spatial distribution density of the filling particles. Then, step S1, "obtaining the macroscopic mechanical response data and microstructural characteristic data of the composite material," may specifically include:

[0064] S11. Perform uniaxial tensile tests on the composite material specimens, obtaining multiple sets of stress-strain curves by varying the loading rate and specimen size;

[0065] Specifically, in step S11, a uniaxial tensile test is performed to obtain the stress-strain relationship of the composite material under different loading rates and specimen sizes. In the test, a computer servo high and low temperature tensile testing machine is used for testing, and a tensile fixture suitable for bone-shaped specimens is designed. The loading rate range is 0.001s -1 to 1.0s -1 Specimen sizes included 8mm, 11.3mm, and 25.0mm diameters. The test platform was a computer servo high and low temperature tensile testing machine with a maximum test force of 20kN and a resolution of 1 / 200,000 of the maximum load. The maximum open-loop tensile speed was 5000mm / min. To ensure the validity of the test results, five valid replicates were performed under each test condition. All data are the average of these five test results.

[0066] S12. Use pseudo-strain conversion to eliminate the viscoelastic stress relaxation effect and determine the damage initiation strain threshold by analyzing the nonlinear deviation points of the stress-pseudo-strain curve;

[0067] Specifically, for step S12, by introducing the concept of pseudo-strain, the stress-strain relationship of the viscoelastic material is converted into a constitutive relationship consistent with the form of the elastic material. The specific method is based on the pseudo-variable method of Schapery (nonlinear viscoelastic model), which converts the stress-strain curve into a stress-pseudo-strain curve to eliminate the influence of the viscoelastic stress relaxation effect on the nonlinearity of the curve. Determination of the damage initiation strain threshold: The damage initiation strain threshold is determined by analyzing the nonlinear deviation point of the stress-pseudo-strain curve. When the relative deviation between the stress-pseudo-strain curve and the reference straight line with a slope of 1 reaches 5%, it is considered that the material begins to be damaged.

[0068] S13. Perform three-dimensional imaging of the composite material sample using micro-computed tomography to obtain spatial distribution and morphological data of the filler particles;

[0069] Specifically, in step S13, microcomputed tomography (CT) is used to perform three-dimensional imaging of the composite material specimen to obtain the spatial distribution and morphological data of the filler particles. Scanning parameters include an operating voltage of 75 kV, a current of 133 μA, an image resolution of 3.57 μm, a rotation angle increment of 0.25°, and a scanning angle of 360°. The two-dimensional slice images obtained by the CT scan are subjected to noise reduction, segmentation, and three-dimensional reconstruction to extract morphological parameters of the filler particles, such as the major axis length and slenderness ratio of the rotation ellipsoid.

[0070] S14. Parametric characterization of particle morphology based on an equivalent ellipsoid model, and statistical analysis of particle principal axis length, slenderness ratio, and volume fraction;

[0071] Specifically, in step S14, the filling particles are converted into rotating ellipsoids, and statistical analysis is performed to determine the particle's major axis length, slenderness ratio, and volume fraction. The specific method includes extracting the volume, surface area, and maximum Frett diameter of each particle based on CT scan image processing results; determining the major axis length and slenderness ratio of the equivalent ellipsoid by solving the ellipsoid volume and surface area equations; and performing statistical analysis on all particles to obtain the particle number density distribution within different slenderness ratios and major axis length ranges.

[0072] This embodiment uses the pseudo-strain conversion method to effectively eliminate the influence of viscoelastic stress relaxation effect on the extraction of damage initiation strain, thereby improving the quantification accuracy of the damage threshold. Through micro-CT scanning technology and an equivalent ellipsoid model, detailed microstructural data of the filling particles inside the composite material are obtained, providing key input for model construction.

[0073] Optionally, in some embodiments, step S2 of "constructing a multi-mechanism damage prediction model corresponding to the composite material based on microstructure characteristic data using micromechanics theory" may specifically include:

[0074] S21. Use the equivalent inclusion theory to calculate the homogenized modulus of composite materials and combine it with the Laplace transform to deal with the time-varying viscoelastic behavior of the matrix;

[0075] Specifically, for step S21, the Mori-Tanaka (MT) method is used to equate the filling particles to rotating ellipsoids embedded in the matrix, and the effective modulus of the composite material is calculated. The MT method assumes that the matrix is ​​an isotropic linear viscoelastic material and the particles are isotropic linear elastic materials. The effective modulus of the composite material is calculated by combining the Eshelby equivalent inclusion theory with the Laplace transform to process the time-varying viscoelastic behavior of the matrix. The specific steps include: equating the AP particles to rotating ellipsoids embedded in the matrix, considering the size distribution, shape parameters and spatial distribution density of the particles; based on the MT method, calculating the average strain of the matrix and the average strain of the particles, and then obtaining the effective modulus of the composite material; processing the viscoelastic behavior of the matrix through the Laplace transform in the complex domain to ensure that the model can describe the time-varying characteristics of the material.

[0076] S22. Establish a statistical failure criterion that includes matrix tearing, particle fracture, and interface dewetting. The failure criterion is based on the Weibull distribution and a modified strength criterion.

[0077] Optionally, in some embodiments, the statistical failure criteria include:

[0078] The matrix tearing criterion is based on the modified generalized von Mises yield criterion, which introduces the influence factor of hydrostatic pressure on matrix yield strength.

[0079] Particle fracture criterion based on the Mohr-Coulomb yield criterion, which combines particle volume effect to modify the Weibull distribution parameters;

[0080] The dewetting criterion is based on the interface normal stress, and the dewetting criterion is defined as the randomly distributed threshold value of the interface strength.

[0081] Specifically, for step S22, a modified generalized Von Mises yield criterion is adopted, and the influence factor of hydrostatic pressure on the yield strength of the matrix is ​​introduced. The modified Von Mises criterion can more accurately describe the yield behavior of the matrix under complex stress states. Based on the Mohr-Coulomb yield criterion, the Weibull distribution parameters are modified in combination with the particle volume effect. The Mohr-Coulomb criterion can effectively describe the fracture behavior of particles under the combined action of tensile and shear stresses. It is defined as the random distribution threshold of the interface strength and the dewetting criterion based on the interface normal stress. This criterion can describe the dewetting behavior of the matrix and particle interface under tensile load.

[0082] The Laplace transform is used to convert the time-dependent viscoelastic behavior into an expression in the complex domain, simplifying the computation of the genetic integral. The effective modulus of the matrix is ​​calculated in the complex domain and, combined with Eshelby's equivalent inclusion theory, the effective modulus of the composite material is obtained. The complex domain results are converted back to the time domain using an inverse Laplace transform, yielding the mechanical response of the composite material in the time domain.

[0083] In this embodiment, the model uses the equivalent inclusion theory and modified strength criterion to more accurately describe the mechanical behavior and damage mechanism of composite materials. The model comprehensively considers the influence of multiple factors such as particle size, shape, distribution density, and matrix viscoelasticity, significantly improving the comprehensiveness and accuracy of damage prediction. By processing viscoelastic behavior through Laplace transform, the model can accurately describe the time-varying characteristics of composite materials, improving the applicability of the model.

[0084] Optionally, in some embodiments, the “inverse calibration of model parameters of the multi-mechanism damage prediction model based on macroscopic mechanical response data” in step S3 may specifically include:

[0085] S31. Define an optimization objective function that includes interface strength, matrix viscoelastic parameters, and statistical damage parameters. The objective function is based on the relative error between the experimental damage onset strain and the model prediction value.

[0086] Specifically, in step S31, an objective function is defined to quantify the deviation between the model prediction and the experimental value. The best consistent approximation criterion is used, namely, minimizing the maximum relative deviation between the model prediction and the experimental value. Optimization parameters include interface strength, matrix viscoelastic parameters (such as the Prony series parameter of the relaxation modulus), and statistical damage parameters (such as the Weibull distribution parameter).

[0087] S32. A hybrid optimization strategy combining a multi-island genetic algorithm and a pattern search algorithm is used to perform global search and local adjustment of model parameters;

[0088] Specifically, for step S32, the MIGA (Multi-Island Genetic Algorithm) algorithm evolves in parallel through multiple independent genetic algorithm populations (islands), and individuals within each island undergo selection, crossover, and mutation operations. Individuals migrate regularly between islands to maintain population diversity. The specific parameters are set as follows: the number of islands is 20, the number of individuals per island is 10, and the total number of evolutionary generations is 20. The pattern search algorithm (Hooke-Jeeves) is a direct search method that performs a fine search within a local area through a pattern search strategy. The maximum number of optimization iterations is set to 400. First, MIGA is used to perform a global search to find the global optimal area of ​​the parameters; then, the Hooke-Jeeves algorithm is used to perform local adjustments within this area to further improve the optimization accuracy of the parameters.

[0089] This embodiment uses a precise objective function and a hybrid optimization strategy to ensure that the optimization results of the model parameters can fit the experimental data to the greatest extent, significantly improving the prediction accuracy of the model; the global search capability of the MIGA algorithm and the local optimization capability of the Hooke-Jeeves algorithm are combined to ensure the global optimality and local accuracy of the parameter optimization; the hybrid optimization strategy significantly improves the efficiency of parameter optimization and shortens the optimization time, enabling it to be applied to large-scale parameter optimization problems.

[0090] Optionally, in some embodiments, “verifying the effectiveness of the multi-mechanism damage prediction model through experimental data” in step S3 may specifically include:

[0091] S33. Divide the experimental data into a model parameter optimization dataset and a model validation dataset;

[0092] Specifically, for step S33, the experimental data are divided into two parts, one for model parameter optimization (model parameter optimization data set), and the other for model verification (model verification data set). The parameter optimization data set contains three specimens of different sizes at strain rates of 0.01, 0.02, 0.05, 0.15, and 0.2s -1 The damage threshold strain values ​​measured under loading conditions, a total of 15 sets of data; the validation data set contains specimens of the same size at strain rates of 0.001, 0.005, 0.1, 0.5, and 1.0s -1 The damage initiation strain values ​​under loading conditions also include 15 data sets. Data with different loading rates and specimen sizes were selected to ensure that the data sets cover the applicable range of the model while avoiding overlap between data sets to ensure the independence of the validation data.

[0093] S34. Verify the validity and accuracy of the multi-mechanism damage prediction model by comparing its predicted values ​​with experimental values.

[0094] Specifically, for step S34, the damage onset strain value predicted by the model is compared with the damage onset strain value measured experimentally. The specific method includes: drawing a comparison graph of the predicted value and the experimental value to intuitively show the degree of agreement between the two. Calculate the relative error to evaluate the accuracy of the model prediction, and finally, intuitively display the comparison results of the model prediction value and the experimental value through charts (such as scatter plots, error bar charts). In addition, numerical simulations (such as finite element analysis) and physical experiments can be combined to verify the predictive ability of the model from multiple dimensions. Perform a detailed analysis of the prediction error, identify the main sources of error (such as simplification of model assumptions, randomness of experimental data), and provide a basis for further model improvement.

[0095] This embodiment uses an independent validation data set to ensure the objectivity and reliability of model validation and avoid overfitting. It also verifies the applicability and generalization ability of the model under various loading rates and specimen sizes, ensuring the reliability of the model under different working conditions. Through comparative verification with experimental data, it is confirmed that the model prediction results are highly consistent with the experimental values, thereby enhancing the credibility of the model.

[0096] Optionally, in some embodiments, step S4 of “predicting the damage onset strain of the composite material under different loading conditions and microstructures based on the calibrated multi-mechanism damage prediction model” may specifically include:

[0097] S41. Establish a quantitative correlation between specimen size and damage threshold by performing homogenization analysis on representative volume units;

[0098] Specifically, in step S41, the Mori-Tanaka (MT) method is used to homogenize the RVE and calculate the average stress and strain of the matrix and particles. This homogenization analysis simplifies the complex microstructure into macroscopic mechanical properties. By analyzing the mechanical response of RVEs of varying sizes, a quantitative correlation between specimen size and damage threshold is established. This involves numerically simulating RVEs of varying sizes to obtain their damage onset strains under identical loading conditions; and establishing a mathematical relationship between size parameters and damage threshold through regression analysis.

[0099] S42. Predicting the damage initiation strain under different particle distributions and loading rates based on a mapping model between microstructural parameters and macroscopic loading conditions;

[0100] Specifically, for step S42, a mapping relationship is established between microstructural parameters (such as particle size distribution, shape parameters, interface strength) and macro loading conditions (such as strain rate, sample size). For example, damage initiation strain data under different microstructural parameters and loading conditions are obtained through experiments and numerical simulations. A mapping model is constructed using machine learning methods (such as neural networks, support vector machines) or statistical modeling methods (such as multivariate linear regression). Finally, the mapping model is used to predict the damage initiation strain under different particle distributions and loading rates. For example: predict the damage initiation strain under different particle size distributions, and predict the damage initiation strain under different loading rates.

[0101] This example quantifies the effect of specimen size on the damage threshold through homogenization analysis, providing a theoretical basis for size optimization design. The model that considers size effects and microstructural parameters can more accurately predict damage behavior under different sizes and loading conditions, enhancing the applicability of the model. The mapping model can more accurately predict the damage onset strain under different particle distributions and loading rates, significantly improving the model's prediction accuracy.

[0102] To facilitate understanding of the composite material damage prediction method based on micromechanics provided in this embodiment, the specific implementation process of the composite material damage prediction method will be described in detail below with examples. The specific steps are as follows:

[0103] 1.1 Macro tensile test

[0104] 1.1.1 Experimental Design and Results

[0105] In order to study the strain rate correlation and specimen size correlation of the damage onset strain of the material, three sizes of uniaxial tensile bone-shaped specimens were designed in this embodiment. Since the cross-section of the round rod-shaped tensile specimen is circular, its stress distribution is relatively more uniform when subjected to tension. The defects and unevenness at the edge of the specimen can be better eliminated. This uniform stress distribution helps to better evaluate the overall mechanical properties of the material, especially in strength tests, and can more accurately reflect the true strength of the material. Before the test, the specimen is placed in a drying oven and sealed to prevent the aging effect of air moisture on the propellant. The test platform is a computer servo high and low temperature tensile testing machine with a maximum test force of 20kN and a resolution of 1 / 200000 of the maximum load. The maximum open-loop tensile speed is 5000mm / min. In order to carry out experiments using the above-mentioned propellant bone-shaped specimen and tensile machine, it is necessary to design and process a fixture that can be used in conjunction with the specimen.

[0106] In this example, tensile tests were first conducted on three sizes of propellant uniaxial tensile bone specimens at room temperature (298.15K) at different loading speeds, including 0.001 (4.2 mm / min), 0.005, 0.01, 0.02, 0.05, 0.1, 0.15, 0.2, 0.5, and 1.0 s. -1 Ten different tensile strain rates (4200 mm / min) were used. Due to the random distribution of filler particles, the mechanical properties of the same batch of propellants also vary within a certain range. Furthermore, during the experiment, the propellant samples may have sustained initial damage, leading to deviations and errors in the test results. To ensure the validity of the test results, five valid replicates were performed under each test condition. All data in the following analysis are the average of these five test results.

[0107] 1.1.2 Damage initiation strain analysis

[0108] The damage initiation strain (DIS) refers to the critical strain at which a material or structure begins to develop macroscopic damage (e.g., cracks, holes, etc.) under external load. According to continuum damage mechanics theory, the nonlinear mechanical behavior of a material is primarily attributed to the initiation and evolution of internal damage. This damage evolution leads to a "softening" of the material's macroscopic mechanical properties. Specifically, the material's stiffness decreases as damage progresses. The uniaxial tensile stress-strain curves of this material at various loading rates exhibit significant nonlinear characteristics. During the initial loading phase, the stress-strain relationship is approximately linear. As strain increases, the slope of the tangent to the curve gradually decreases, indicating softening of the material until it eventually fractures. Due to the material's significant viscoelastic properties, the decrease in modulus during tension is primarily due to two factors: stress relaxation due to viscoelasticity and softening due to damage evolution. To accurately extract the DIS from the stress-strain curve, it is necessary to convert the stress-strain curve into a pseudo-strain-stress relationship to eliminate the influence of stress relaxation on the curve's nonlinearity. To this end, a pseudo-variable-based method is proposed to transform the stress-strain relationship of a viscoelastic material into a constitutive relationship consistent with that of an elastic material. By introducing the correspondence principle, for linear viscoelastic materials, pseudo-strain can be constructed based on the linear constitutive model, so that the stress-pseudo-strain relationship of the viscoelastic material is obtained as follows:

[0109] σ=C:ε R (3.1)

[0110]

[0111] In formula (3.1), C represents the stiffness coefficient tensor of the material, and its component expressions are shown in formula (3.3); ε Ris the pseudo strain tensor, and its calculation formula is shown in formula (3.2), where ε is the strain tensor; E(t) is the one-dimensional linear viscoelastic modulus of the propellant; E R The reference elastic modulus can be selected arbitrarily, usually E R = 1. By introducing the reference modulus, the dimension of the pseudo strain is achieved to be 1.

[0112] C ijkl =λδ ij δ kl +μ(δ ik δ jl +δ il δ jk ) (3.3)

[0113] In formula (3.3), λ and μ are Lame constants, which have the following form:

[0114]

[0115] In the above formula, ν is the Poisson's ratio of the material; from formula (3.2), it can be seen that the pseudo-strain is in the form of convolution integral, which includes the genetic integral part of the time correlation of viscoelastic mechanics. Under uniaxial loading conditions, let C = E R , introducing the concept of pseudo-strain into the linear viscoelasticity theory, the one-dimensional stress-pseudo-strain relationship becomes:

[0116] σ=E R ε R (3.6)

[0117] According to equation (3.6), the linear viscoelastic response is linearly related to the pseudo-strain. When the uniaxial tensile strain rate is fixed, the linear viscoelastic modulus E(t) of the propellant is expressed in the form of the Prony series shown in equation (2.11), and the strain and pseudo-strain in the tensile direction have the following relationship:

[0118]

[0119] The relaxation modulus of the HTPB composite solid propellant was obtained through relaxation tests. At room temperature (298.15 K), the Prony series parameters of the relaxation modulus are shown in Table 1, in MPa.

[0120] Table 1 Prony series of material relaxation modulus

[0121]

[0122] According to the relationship between strain and pseudo-strain (3.3), the pseudo-strain values ​​under different stretching rates and strain conditions can be calculated, and then the stress-pseudo-strain curve can be plotted. When the material is in the linear viscoelastic stage, the stress-pseudo-strain curve shows a linear relationship; in the initial loading stage, the stress-pseudo-strain curve of the propellant shows linear viscoelastic characteristics. As the strain increases further, the mechanical behavior of the propellant gradually changes to nonlinear viscoelasticity, and there are differences in the nonlinear initial pseudo-strain values ​​under different strain rates, resulting in the non-linear curves not coinciding. In this embodiment, the horizontal coordinate corresponding to the turning point of the linear segment and the nonlinear segment in the stress-pseudo-strain curve is defined as the damage onset pseudo-strain. In order to facilitate the clarification of the specific value of the damage onset strain, this embodiment adopts the following judgment standard: when the relative deviation between the experimentally measured stress-pseudo-strain curve and the reference line with a slope of 1 reaches 5%, it is considered that the material begins to be damaged. This judgment standard provides a clear basis for the quantification of the damage onset strain.

[0123] Under uniaxial tensile loading at room temperature, both the damage onset pseudostrain and the damage onset strain showed a significant increase with increasing diameter of the bone-shaped specimen. Furthermore, within the strain rate range of 0.001–1.0 s⁻¹, the damage onset pseudostrain exhibited a logarithmic growth trend, while the damage onset strain exhibited a logarithmic decay trend. These results indicate that the damage onset behavior of propellants is not only influenced by specimen geometry but also closely related to the loading rate. These findings provide important experimental evidence for studying the damage evolution mechanism of propellants under different loading conditions.

[0124] 1.2 Analysis of microCT test results

[0125] The components of the solid propellant used in this embodiment include a binder, hydroxy-terminated polybutadiene (HTPB), an oxidizer, ammonium perchlorate (AP), aluminum powder (Al), and functional additives such as a curing agent, a cross-linking agent, and a plasticizer. Before the experiment, the sample was sealed in a drying oven to prevent the aging effect of air moisture on the propellant. The sample was cut from the center of the bone-shaped propellant sample, and the sample size for micro-CT scanning analysis was 4mm×4mm×4mm. The statistical value of the microstructural parameters is the average value of the scanning result analysis of the three scanned samples. This experiment was carried out based on the Sky scan 1172 micro-CT, with a spatial resolution of up to 0.5μm, which meets the experimental requirements.

[0126] CT slice images of HTPB propellant samples obtained by scanning are a series of grayscale images. MicroCT reconstructed images often contain some noise, which, if not filtered, can affect image segmentation, 3D reconstruction, and quantitative analysis. Therefore, after obtaining 2D slice reconstructed images, they must first be filtered. Neighborhood averaging and median filtering are commonly used. Based on the CT reconstruction results, this paper uses median filtering to address noise in the reconstructed images. By replacing the original pixel value with the median grayscale value within a pixel neighborhood, this method effectively removes random noise while preserving edge details. The microstructure of propellants is highly complex due to varying microstructures and material properties. To study specific microstructures, image segmentation methods such as edge detection-based segmentation, threshold-based segmentation, and tracking-based segmentation are essential to segment the structures of interest. Threshold-based segmentation methods use the distribution characteristics of the grayscale values ​​of each element in the grayscale image to determine the corresponding element by determining a threshold. Preliminary processing and analysis of the reconstructed image allows the grayscale range of each microstructure to be determined, enabling accurate segmentation.

[0127] To further enhance image readability and intuitiveness, the segmented image is rendered in 3D, with the filler particles rendered in different colors based on their size. All image processing is performed on the 3D image. To more accurately capture the microstructure of the solid propellant filler particles, a more detailed segmentation of the 3D image is required. This is accomplished using the built-in "Separate Objects" function based on an expansion algorithm. After identifying the solid particles, the eroded solid phase is segmented, similar to the adaptive watershed segmentation technique. Following watershed segmentation, each filler particle is extracted from the reconstructed image to determine its morphological parameters.

[0128] 1.2.1 Characterization and analysis of solid particles

[0129] In order to quantitatively characterize the microstructure of HTPB propellant, Li Shiqi et al. characterized the size of AP particles as the equivalent diameter d and the shape of AP particles as the sphericity s, which are defined as follows:

[0130]

[0131] Where V is the volume of the mesostructure.

[0132]

[0133] A is the surface area of ​​the mesostructure. The closer the value of s is to 1, the closer the shape of the mesostructure is to a sphere. When s = 1, the shape of the mesostructure becomes spherical. However, this paper argues that this correspondence is not accurate enough, primarily because when the mesostructure is spherical, the sphericity s is 1, and when the sphericity s is 1, the mesostructure is not necessarily spherical. In other words, the sphericity s and the shape characteristics of the filler particles do not have a one-to-one correspondence. This is because the surface of AP particles is uneven and not completely convex. Furthermore, using equivalent diameter d and sphericity s to characterize the size and shape of propellant filler particles makes it difficult to apply statistical results to mesoscopic mathematical model calculations and analysis. For example, the shape of a particle cannot be determined based on the value of the equivalent diameter d, and whether the particle is elongated or flat cannot be determined based on the value of the sphericity s. Studies have shown that the morphology of HTPB propellant filler particles is mostly irregular polyhedrons, but there are also particles that are closer to ellipsoids or spheres. Taking into account the irregularity of the shape of the filling particles, the rotation tensor analysis method is introduced to calculate the eigenvectors of the HTPB propellant filling particles and construct a characteristic ellipsoid, whose semi-axis lengths are a, b, and c (a>b>c), respectively. The calculation formulas for the elongation index EI and the flattening index FI are defined as follows.

[0134] EI=b 2 / a 2 (2.3)

[0135] FI=c 2 / b 2 (2.4)

[0136] The smaller the EI, the more slender the particle shape; the smaller the FI, the flatter the particle. This method can effectively quantitatively describe the three-dimensional structure of the filling particles, but it is not convenient for applying micromechanics methods to achieve homogenized inclusion analysis. In order to facilitate statistical analysis of particle morphology and subsequent micromechanics analysis, this example defines the size and shape parameters of the propellant as a rotating ellipsoid inclusion, with the shape parameters being the rotation axis radius a1, a2=a3=a and the slenderness ratio ρ=a1 / a. In order to obtain the semi-axis length and slenderness ratio of the filling particles, this paper performs calculations and analysis according to the following steps:

[0137] (1) First, the volume V, surface area A, and maximum Fret diameter Dmax (i.e., the maximum distance between any two points on the particle surface) of each particle are obtained based on scanned and reconstructed image processing and analysis. The maximum Fret diameter can be calculated through image preprocessing, image binarization, regional attribute calculation, particle edge segmentation, parallel direction selection, and maximum diameter calculation. In this embodiment, 72 parallel planes at different angles are selected for each particle to obtain the maximum Fret diameter.

[0138] (2) For rotating ellipsoid particles, the calculation formulas for their volume and surface area are Equation (1.13) and Equation (1.14), respectively. Solve the equations based on these two formulas and calculate the solutions to the equations. The number of solutions may not be unique.

[0139]

[0140] (3) If the solution ρ≥1, then a max =ρ·a; if ρ<1, then a max =a;

[0141] (4) Determine which set of solutions has the closest value of amax to Dmax, and take the closest set of solutions as the shape parameter values ​​of the equivalent rotating ellipsoid.

[0142] After obtaining the equivalent ellipsoidal parameters for all AP and Al particles within the HTPB propellant, a statistical analysis was performed based on the length of the principal axis radius a1 and the slenderness ratio ρ. Overall, the principal axis radius length distribution histogram of the AP particles exhibits a single peak, while the slenderness ratio distribution histogram exhibits a bimodal characteristic, with one peak located between 0 and 1 and the other peak with a horizontal coordinate greater than 1. This indicates that AP particles have two typical shapes: a flattened ellipsoid of rotation and a rugby-shaped elongated ellipsoid of rotation. The total volume fraction of AP particles is 52.17%, and the volume fraction of AP particles with an aspect ratio ρ between 0 and 1 is 34.18%, accounting for 59.52% of the total volume. This indicates that flattened ellipsoidal AP particles account for a larger proportion, while ellipsoidal AP particles with larger principal axis radius lengths are mostly rugby-shaped elongated ellipsoids of rotation.

[0143] The damage initiation strain prediction model of composite solid propellant is established. The microscopic parameters required are the volume fraction c of filling particles with different slenderness ratios ρ. ρ and specific surface area A ρ , its calculation formula is shown in formula (1.15) and formula (1.16), c ρ and A ρ The trend of DE with the change of ρ is similar to the distribution histogram of the number density of AP particles with the aspect ratio, and also has distribution characteristics.

[0144]

[0145] The distribution histogram of AP particles is similar to that of Al particles. The distribution histogram of the principal axis radius length of Al particles also has a single peak characteristic, while the distribution histogram of its aspect ratio has a bimodal characteristic.

[0146] 1.3 Damage initiation strain model for composite solid propellants

[0147] Current research on the damage initiation strain of solid propellants mainly focuses on the macroscale. Whether based on the characteristic parameter method or the cumulative damage theory, the microscopic component structure of the propellant and its influence on the damage behavior are not fully considered. Specifically, the existing research methods fail to effectively evaluate the combined effects of the propellant's internal microstructure (such as filler distribution, matrix-filler interface characteristics, etc.) and the specimen geometry (volume effect) and loading rate (strain rate effect) on damage. This macroscale limitation leads to an incomplete understanding of the damage initiation mechanism. To make up for this research deficiency, this embodiment will propose a damage initiation strain analysis method based on the micromechanics of composite materials. By introducing the micromechanics theory, this method comprehensively considers the heterogeneity of the propellant's internal microstructure and its relationship with the macroscopic mechanical response, aiming to more accurately describe the initial conditions of damage initiation and its evolution law.

[0148] 1.3.1 Description of some operation rules in the model

[0149] To facilitate understanding and ensure the conciseness and logic of the paper, this section will briefly describe the calculation rules used:

[0150] Let f(s) be the Laplace transform of f(t), where s is a complex variable.

[0151] f(s)=L[f(t)] (3.8)

[0152] If P is the transversely isotropic tensor, it can be expressed as a 6×6 matrix:

[0153]

[0154] Among them 2P 2323 =P 2222 -P 2233 If the tensor P is simply written as P = (c, g, h, d, e, f), then c = P 2222 +P 2233 g=h=P 1122 , d=P 1111 , e=2P 2323 , f=2P 1212 , the tensor P corresponds to a 6×6 matrix as shown in formula (3.10):

[0155]

[0156] For a tensor H(H ijkl =H jikl =H ijlk ) and H′(H′ ijkl =H′ jikl =H′ ijlk), they can be expressed as H = (c, g, h, d, e, f), H′ = (c′, g′, h′, d′, e′, f′). The operation between H and H′ follows the following rules:

[0157] H±H′=(c±c′,g±g′,h±h′,d±d′,e±e′,f±f′) (3.11)

[0158] H:H′=(cc′+2hg′,gc′±dg′,ch′+hd′,dd′±2h′g,ee′,ff′) (3.12)

[0159]

[0160] Using the above representation method, the fourth-order unit tensor can be expressed as I = (1, 0, 0, 1, 1, 1).

[0161] For a fourth-order isotropic tensor with only two independent components, it can be abbreviated as H = (α, β), and the corresponding subscript form is:

[0162]

[0163] Let H = (α, β) and H′ = (α′, β′). The abbreviation of the isotropic fourth-order tensor defined above complies with the following operation rules:

[0164] H±H′=(α±α′,β±β′) (3.15)

[0165]

[0166] H:H′=(αα′,ββ′) (3.17)

[0167] The fourth-order unit tensor I=(1,1).

[0168] 1.3.2 Micromechanical analysis

[0169] This paper adopts the Mori-Tanaka (MT) method to account for the effects of particle reinforcement. Composite solid propellants consist of a binder matrix (hydroxyl-terminated polybutadiene (HTPB), an oxidizer (ammonium perchlorate), aluminum powder (Al), and functional additives such as curing agents, crosslinkers, and plasticizers. Microscopic scanning studies reveal that under uniaxial tensile loading, smaller particles in composite propellant specimens exhibit less interfacial dewetting than larger particles, and crack propagation after interfacial dewetting primarily occurs along the interfaces of the larger particles. Given that the average size of Al particles is significantly smaller than that of AP particles and their lower volume fraction in the propellant, interfacial dewetting is largely absent during loading. Therefore, the theoretical derivation below simplifies the failure behavior of Al particles and their bonding interface, assuming that the Al particles and their interface with the matrix remain intact during loading, serving only to strengthen the matrix. Based on this assumption, the homogenized mixture of small Al particles and the HTPB binder is considered an equivalent mixed matrix. It is important to note that the "matrix" mentioned below refers to this mixed matrix. This approach not only simplifies the model complexity but also provides a reasonable theoretical basis for subsequent micromechanical analysis.

[0170] The AP particles are equivalent to rotating ellipsoids with different spatial orientations and size distributions embedded in a binder matrix. Since rotating ellipsoids with the same aspect ratio and consistent spatial arrangement direction have the same Eshelby tensor in the macroscopic coordinate system, the AP particles with the same aspect ratio and consistent arrangement direction can be regarded as the same phase material, assuming that the AP particles of each phase are randomly distributed in space. Take a composite solid propellant RVE, assuming that there are N phases of material in the RVE, including the matrix phase and each AP particle phase with a specific aspect ratio and spatial orientation. Use C r Represents the elastic stiffness coefficient tensor of the r-phase material, and r=0 represents the matrix. This section assumes that the matrix is ​​an isotropic linear viscoelastic material, and the AP particles are isotropic linear elastic materials. Based on the MT method, the effective modulus C of the composite material can be calculated by formula (3.18). It should be noted that since this embodiment regards the matrix as a linear viscoelastic material, the Laplace correspondence between linear viscoelasticity and linear elasticity is used in the following derivation process. Specifically, in formulas (3.18) to (3.28), all time-related variables should be replaced with their Laplace transformed forms. At the same time, considering the genetic integral form of linear viscoelasticity, the modulus C0 is related to sC0(s) and Replacement, strain tensor <ε>0 and <ε> r Should be used separately <ε(s)>0 and <ε(s)>r replace.

[0171]

[0172] Where: c r is the volume proportion of the r-phase material, c0 is the volume proportion of the matrix, P r is a fourth-order tensor related to the slenderness ratio of the ellipsoid. For a spheroidal inclusion (rotation axis is x1, a2=a3=a, slenderness ratio ρ=a1 / a), P r The tensor is a transversely isotropic tensor, and its components are expressed as follows. Other unlisted components are 0:

[0173]

[0174] When ρ>1:

[0175]

[0176] When ρ<1:

[0177]

[0178] Where G0 and υ0 are the shear modulus and Poisson's ratio of the matrix, respectively.

[0179] If the uniform strain boundary condition of RVE is , then the average strain of the matrix in RVE is:

[0180]

[0181] Where I is the fourth-order unit tensor,

[0182] Correspondingly, the average strain in the rth type inclusion is:

[0183]

[0184] The fourth-order tensor T r The expression is,

[0185] T r =[I+P r :(C r -C0)] -1 (3.24)

[0186] In order to solve equations (3.18), (3.22) and (3.23), it is necessary to first calculate the volume fraction of different types of AP particles. AP particles are embedded in the HTPB matrix in a randomly distributed form. In order to quantitatively describe the spatial distribution characteristics of AP particles, this paper introduces a distribution density function on a unit radius sphere. where θ and Represent the polar angle and azimuth angle in the spherical coordinate system respectively. The physical meaning of the distribution density function is that the principal axis direction in the unit volume is The number of AP particles with a slenderness ratio of ρ. If the average semi-axis length of the AP particles with a slenderness ratio of ρ is a ρ , whose average volume is V ρ , the volume of RVE is represented by V0, and the total number of particles contained in RVE is NAP:

[0187]

[0188] Along direction The infinitesimal area on the unit sphere can be expressed as The main axis direction is located at the microelement The AP particles with an aspect ratio of ρ are regarded as the rth type inclusions in the MT method. Based on this, the volume fraction of the rth type inclusions can be expressed as:

[0189]

[0190] Then the volume fraction of all AP particles with slenderness ratio ρ in the RVE is:

[0191]

[0192] Substituting equations (3.26) and (3.27) into equation (3.22), the average strain of the propellant matrix can be obtained as:

[0193]

[0194] {·} angle The operator refers to the spatial angle average of the variable, and its definition is:

[0195]

[0196] In the three-dimensional CT microscopic scanning reconstruction, it was found that the orientation of AP particles in space was randomly distributed. Therefore, the AP particle distribution density function and spatial angle In this case, the spatial angle average formula (3.29) can be simplified to:

[0197]

[0198] In order to facilitate the solution of the matrix average strain in Equation (3.28) and the RVE effective modulus in Equation (3.18), simplified notation is used here, and [(C r -C0) -1 +P r ] -1Denoted as W, replace [(C r -C0) -1 +c0P r ] -1 Denoted as R. Since the matrix and ellipsoidal AP particles are both isotropic materials, the fourth-order tensor C r , C0 and P r All satisfy the rotation symmetry of their components, so W and R can be expressed as W = (w1, w2, w3, w4, w5, w6) and R = (r1, r2, r3, r4, r5, r6). Based on equations (3.11) to (3.13), the specific expression of the W component can be derived as:

[0199]

[0200] In formula (3.31) K r and G r are the bulk modulus and shear modulus of the AP particles, respectively. Using the spatial angle average formula (3.30) and referring to formulas (3.14) to (3.17), {W} can be calculated. angle and {R} angle The expressions are:

[0201]

[0202] According to equations (3.22) and (3.28), in order to solve the average strain of the matrix <ε>0 in the propellant RVE model, it is first necessary to determine the uniform strain boundary condition of the RVE: During uniaxial tensile loading, the displacement and strain in the tensile direction are known, while the two strain components perpendicular to the tensile direction are unknown. In order to solve the strain components in these two directions, the Poisson's ratio of the RVE needs to be determined in advance. In this paper, the homogenized bulk modulus of the RVE model is used and shear modulus To calculate Poisson's ratio. Substituting equations (3.26) and (3.27) into equation (3.18), we can obtain the effective modulus tensor of the propellant RVE in the complex domain, which satisfies the following relationship:

[0203]

[0204] Based on formula (3.34), according to the elastic-viscoelastic correspondence principle, the expression of Poisson's ratio of the propellant RVE model in the complex domain can be further deduced as:

[0205]

[0206] Under uniaxial tensile loading, the principal strain direction of RVE is consistent with the tensile direction, and the three shear strain components are all 0. The three principal strains satisfy the following relationship:

[0207] ε 22 (s)=ε 33 (s)=-sν(s)ε 11 (s)(3.36)

[0208] According to equations (3.23) and (3.29), the average strain of all AP particles with a slenderness ratio of ρ and randomly oriented in space can be expressed as:

[0209]

[0210] According to formula (3.24) and the expression of W, we can get T r Has a relationship with W: T r =(C r -C0) -1 :W, also known but:

[0211]

[0212] The stress in different directions at the particle / matrix interface is approximately:

[0213]

[0214] According to the above analysis, the average stress of AP particles with slenderness ratio ρ and random orientation distribution in space is (σ 11 ,σ 22 ,σ 33 ,0,0,0), then the normal stress component of the particle / matrix interface is τ n =cosθ·σ 11 , obviously when cosθ=π or cosθ=0, τ n reaches its maximum value, at which point the direction of maximum normal stress on the interface coincides with the direction of uniaxial tension, and τ n =σ 11 .

[0215] 1.3.3 Statistical failure criteria for each component

[0216] Mesoscopic observations of cross-sections of composite solid propellant specimens subjected to dynamic uniaxial tensile loading over a wide temperature range revealed three typical mesoscopic damage mechanisms within the propellant: matrix tearing, fracture (or crushing) of solid filler particles, and interfacial dewetting of AP particles. Therefore, constructing a propellant damage onset strain model based on mesomechanics theory requires comprehensive consideration of the influence of these three mesoscopic damage mechanisms.

[0217] Based on the propellant microstructure stress analysis results from the previous section and combined with the failure criteria for the adhesive matrix, the failure probability of the composite solid propellant matrix phase can be quantitatively assessed. Given that the adhesive matrix, as a high molecular weight polymer material, has a significant correlation between its yield strength and the hydrostatic pressure component of the stress state, this example uses a modified Von Mises criterion to characterize the random distribution characteristics of the adhesive matrix strength. Furthermore, the matrix tear failure probability is described using the Weibull statistical distribution function, expressed as follows:

[0218]

[0219] Among them, P m (V m ,σ) is the volume V under stress σ m The fracture probability of the matrix, and m m is the material constant, where It represents the minimum equivalent stress required for matrix tearing, F m (σ) is a randomly distributed variable that measures the strength of the adhesive matrix. The order of the principal stress of stress σ is σ1≥σ2≥σ3, and the hydrostatic stress is σ p , then F m (σ) can be expressed as the principal stress:

[0220]

[0221] In the above formula, A is a material constant, and its specific value can be determined by inversion of experimental results.

[0222] Studies have shown that larger AP particles are more likely to break under load. To effectively reflect the effect of size on AP particle breakage, this paper assumes that the probability of AP particle breakage can be described by a Weibull distribution. The probability of failure of a single particle is considered to be:

[0223]

[0224] Referring to formula (3.42), it can be deduced that the total destruction probability of all AP particles in the representative volume unit is:

[0225]

[0226] P in formula (3.43) AP (V AP ,σ) is the volume V under stress σ AP The fracture probability of the AP particle, and m AP is the material constant, where It represents the minimum equivalent stress required for AP particles to break, FAP (σ) is a randomly distributed variable that measures the strength of the adhesive matrix. Considering that AP particles are brittle materials, it can be assumed that the destruction of AP particles is affected by the coupling effect of their maximum shear stress and the normal stress in the corresponding direction. AP The specific form of (σ) refers to the Mohr-Coulomb strength criterion. Assuming that the order of the principal stress magnitude of stress σ is σ1≥σ2≥σ3, then F AP (σ) can be expressed as the principal stress:

[0227] F AP (σ)=(σ1-σ3)+k(σ1+σ3) (3.44)

[0228] Where k is a material constant that can be determined by inversion of experimental results.

[0229] Under tensile loading, the propellant matrix / AP particle bonding interface is prone to dewetting. This paper uses the interface normal stress dewetting criterion to calculate the failure probability of the matrix / AP particle bonding interface. Studies have found that under uniaxial tensile loading, dewetting of the matrix / AP particle bonding interface in HTPB composite solid propellants mostly occurs on the surface of the larger AP particles. To effectively reflect this phenomenon, this paper assumes that the dewetting probability of the matrix / AP particle bonding interface considering a single particle interface is:

[0230]

[0231] Referring to formula (3.45), the total interface failure probability considering all matrix / AP particle bonding interfaces in the RVE can be derived as:

[0232]

[0233] Among them, P f (S f ,σ) is the area S under the action of stress σ f The fracture probability of the interface, m f is the material constant, is the minimum equivalent stress required for dewetting of the matrix / AP particle bonding interface, F f (σ) is a random distribution variable that measures the strength of the matrix / AP particle bonding interface. Let the normal stress on the matrix / AP particle bonding interface be τ n , then F f (σ) can be expressed as:

[0234] F f (σ)=<τ n > (3.47)

[0235] where <> is the Macauley symbol, <τn >=(|τ n |+τ n ) / 2.

[0236] The basic assumption of the Weibull distribution is that a material body of volume V has a statistical distribution of independently interacting microscopic representative units. An object of volume V can be considered to consist of n volume elements of volume V0. Let V0 be the reference volume of the material with a survival rate of P(V0), meaning that when the specimen is loaded to a stress σ, each specimen of volume V0 partially survives. Considering that the volume of the material is V = nV0, the survival rate R(V) of the material of volume V is given by Equation (3.48):

[0237]

[0238] In formula (3.48), R(V0) represents the survival rate of the volume element with volume V0:

[0239] R(V0)=R AP (V AP ,σ)·R m (V m ,σ)·R f (S f ,σ) (3.49)

[0240] In the above formula, R AP (V AP ,σ),R m (V m ,σ) and R f (S f ,σ) are the survival rates of AP particles, matrix and bonding interface in volume element V0, respectively. Substituting equation (3.49) into equation (3.48), we can obtain:

[0241]

[0242] The damage initiation strain of the propellant under uniaxial tensile loading can be obtained by integrating the strain according to formula (3.50), as shown in formula (3.51).

[0243] The overall calculation process of damage initiation strain is as follows: Figure 3As shown in the figure, a uniaxial tensile load was applied to the composite material specimen to obtain macroscopic mechanical response data (such as stress-strain curves). The Mori-Tanaka (MT) method was used to equate the filling particles to rotating ellipsoids embedded in the matrix, and the effective modulus of the composite material was calculated. Based on the MT method, the time-varying viscoelastic behavior of the matrix was processed in combination with Laplace transform to calculate the equivalent Poisson's ratio of the composite material. The average strain of the RVE under uniaxial tensile load was calculated by homogenization analysis. The average strain of the matrix was calculated by the MT method. The average stress of the matrix was calculated from the average strain and equivalent modulus of the matrix. Based on the Eshelby equivalent inclusion theory, the average stress of AP particles with different aspect ratios and their interface normal stress were calculated. The average strain of the particles was calculated from the average stress and mechanical parameters of the particles. Based on the Weibull distribution and the modified strength criterion, the failure probabilities of matrix tearing, particle fracture and interface dewetting were calculated. The total probability of RVE damage was calculated by accumulating the failure probabilities of each microscopic component. The survival rate of the RVE was strain-integrated to determine the damage initiation strain.

[0244] The parameters related to the microstructure, such as the volume fraction of AP particles of each phase in RVE and the specific surface area fraction, are calculated according to the statistical results in the previous section.

[0245]

[0246] 1.4 Model parameter solution, verification and analysis

[0247] 1.4.1 Model parameter solution and model verification

[0248] Bulk modulus K of AP particles r =15056MPa, shear modulus G r =14107MPa. The mechanical properties of the mixed matrix are characterized by a combination of experimental and theoretical methods: First, a stress relaxation test is performed on the matrix film to obtain its time-dependent relaxation modulus; then, based on the Eshelby equivalent inclusion theory, the reinforcement effect of the Al particles is taken into account and the equivalent relaxation modulus of the mixed matrix is ​​calculated by the micromechanical method. The volume fraction and material properties of the Al particles are described in Chapter 2. The relaxation modulus of the mixed matrix is ​​expressed in the form of the Prony series shown in Equation (2.11) using the generalized Maxwell model. At room temperature, the Prony series parameters of the mixed matrix are shown in Table 2, in MPa. The Poisson's ratio is ν0 = 0.499.

[0249] Table 2 Parameters of the viscoelastic constitutive model of the propellant mixture matrix

[0250]

[0251] The solid propellant damage initiation strain prediction model established above contains a total of 11 undetermined parameters, namely: m AP ,k, m m 、A、 and m f . Since it is difficult to directly determine the specific values ​​of these parameters through experiments, and there is a lack of effective theoretical analysis methods. Therefore, this embodiment adopts an inversion analysis method based on experimental data to determine the parameter values. Specifically, by constructing an optimization problem, the macroscopic uniaxial tensile test data is combined with the damage initiation strain prediction model, and a numerical optimization algorithm is used to solve the optimal value of the parameters. The experimental data is divided into two parts: the model parameter optimization data set and the model verification data set. The parameter optimization data set contains three specimens of different sizes at strain rates of 0.01, 0.02, 0.05, 0.15, and 0.2s -1 The damage threshold strain values ​​were measured under loading conditions, totaling 15 sets of data; the model validation data set used specimens of the same size at strain rates of 0.001, 0.005, 0.1, 0.5, and 1.0s -1 The damage initiation strain value under loading conditions also contains 15 sets of data. The deviation between the model prediction value and the experimental value is quantified by constructing an objective function, and the optimal parameters are solved by an iterative optimization method. Considering the model prediction accuracy requirements, this paper uses the best consistent approximation criterion to construct the objective function, that is, to minimize the maximum value of the relative deviation between the model prediction value and the experimental value, as shown in formula (3.52), where and represent the model predicted value and experimental value of the damage initiation strain data of the i-th group, respectively.

[0252]

[0253] In terms of optimization algorithm selection, global optimization algorithms have the ability to search for the global optimum, but their optimization efficiency is low. Direct search methods, while effective in exploring local regions, are prone to falling into local optima and are unsuitable for problems with long computational times. This paper adopts a hybrid optimization strategy to balance global search capability and computational efficiency. The optimization model was established using the multidisciplinary optimization platform Isight integrated with Matlab. The global optimization algorithm used was the Multi-Island Genetic Algorithm (MIGA), while the direct search algorithm used the Hooke-Jeeves algorithm. The MIGA algorithm used 20 islands, 10 individuals per island, and a total of 20 evolutionary generations. All other parameters were set to their default values. The maximum number of optimization iterations for the Hooke-Jeeves algorithm was set to 400, and all other parameters were set to their default values. The initial values ​​of the optimization parameters, the search range, and the final optimization results are detailed in Table 3.

[0254] Table 3 Initial value settings, ranges and results of global optimization

[0255]

[0256] Substituting the parameter optimization results in Table 2 into the established mathematical model, the three different sizes of propellant skeleton specimens were calculated to obtain the following results in the range of 0.001 to 1.0 s: -1 The damage initiation strain of uniaxial tension at ten strain rates. The calculated results show that the maximum relative error is 7.202%.

[0257] Although the strain rate trends of the model predictions and experimental results are highly consistent, indicating that the established mathematical model can well describe the damage initiation behavior of solid propellants, there are still some deviations between the model predictions and the experimental data. The main reasons for this are as follows: First, the established mathematical model does not consider the initial damage within the material, such as microvoid defects in the matrix phase and initial debonding at the AP particle / matrix interface. Second, as a typical particle-reinforced composite material, the mechanical properties of solid propellant have significant randomness, and the parameter values ​​obtained by the optimization algorithm may have some deviations. Finally, because the mechanical properties of each component material of the propellant (such as AP particles and HTPB matrix) show significant strain rate dependence, this example assumes that the model parameters are constants to simplify the parameter determination process. This may lead to a decrease in the model's prediction accuracy under high strain rate conditions.

[0258] 1.4.2 Model parameter impact analysis

[0259] After completing the construction of the composite solid propellant damage initiation strain model, it is necessary to systematically analyze the key parameters in the model to clarify the mechanism of action of each parameter in the model and its contribution to the damage behavior, thereby providing a theoretical basis for the quantitative design and optimization of the mechanical properties of the propellant. This embodiment uses the control variable method to perform parameter sensitivity analysis, and the remaining model parameters are set according to the values ​​in Table 3.4. The analysis objects are selected as cubic specimens with side lengths of 10mm and 100mm, and their corresponding volumes are V = (10mm) 3 and V = (100 mm) 3 , the loading strain rate is selected as R = 1.00s -1 and R = 0.001s -1 Two typical values.

[0260] It can be found that under the condition that the specimen volume and loading rate remain constant, as the parameters With the increase of , the damage initiation strain shows a nonlinear trend of first increasing and then tending to saturation. This phenomenon can be attributed to the parameter The physical meaning of m: It characterizes the average strength of the bonding interface between the filling particles and the matrix, and reflects the intrinsic correlation mechanism between the mechanical properties of the interface and the macroscopic damage resistance of the composite solid propellant. Specifically, when the interface strength is at a low level, enhancing the particle-matrix interface bonding strength can significantly improve the damage threshold of the composite solid propellant; however, when the interface strength reaches a certain critical value, the effect of continuing to improve the interface strength on the damage threshold will gradually weaken and eventually stabilize. This law shows that there is an optimal range for the improvement of the material's damage resistance by interface strength. In addition, for specimens of the same volume, as the loading rate increases, the damage threshold strain shows a downward trend; while at the same loading rate, the increase in specimen volume leads to a decrease in the damage threshold strain. These laws are based on the theoretical prediction results of the current model, and their physical mechanism may be closely related to the stress distribution characteristics and damage evolution process inside the material. Its scope of application and the microscopic mechanism of damage occurrence still need to be further explored and clarified through more experimental verification and detailed data analysis. Under the condition that the specimen volume and loading rate remain constant, the parameter m f and parameters Influencing laws and parameters on damage initiation strain are similar to those of , both showing the characteristics of rising first and then tending to saturation. However, the parameter m f The effect on damage initiation strain is more significant, suggesting that it may play a more important role in controlling propulsion damage behavior. This finding provides important theoretical guidance for the optimization design of subsequent propellant formulations.

[0261] Under conditions where the specimen volume and loading rate remain constant, the three parameters exhibit similar influence trends: as the parameter values ​​increase, the damage initiation strain first shows a clear upward trend, then gradually stabilizes, and eventually reaches a saturation value. This nonlinear response indicates that the mechanical properties of the binder matrix have a significant impact on the damage behavior of the composite solid propellant, but this influence has a critical threshold. Beyond this threshold, further improving the matrix properties will significantly reduce the effect of improving the material's damage resistance.

[0262] The effect of parameter k on the damage initiation strain shows obvious stage characteristics: in the initial stage, as k increases, the damage initiation strain remains basically unchanged; but when k exceeds a certain critical value, the damage initiation strain begins to show a downward trend. It should be pointed out that this law is obtained under uniaxial tension conditions. Whether it is applicable to multiaxial stress states or other complex loading conditions still needs further research on the variation law of damage initiation strain under complex loading conditions to verify. Parameter m AP 、 and The influence of damage initiation strain and adhesive matrix parameter m m 、 and They are similar, both showing the characteristics of rising first and then tending to saturation, so I will not elaborate on them.

[0263] In summary, the composite material damage prediction method based on mesomechanics provided in this embodiment first obtains macroscopic mechanical response data and microstructural characteristic data to provide basic input for subsequent modeling; then, based on the mesomechanics theory, the microstructural characteristics are linked to the macroscopic mechanical behavior to construct a damage prediction model that can comprehensively consider multiple damage mechanisms; then, the model parameters are inversely calibrated through experimental data to ensure the accuracy of the model parameters; at the same time, the effectiveness of the model is verified through experimental data to ensure the prediction accuracy of the model under different conditions; finally, the calibrated model is used to predict the damage onset strain of the composite material under different loading conditions and microstructures, providing a scientific basis for material performance optimization and design.

[0264] It should be understood that although Figure 2 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 2 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0265] To facilitate better implementation of the micromechanics-based composite material damage prediction method of the present application embodiment, the present invention also provides a micromechanics-based composite material damage prediction device based on the aforementioned micromechanics-based composite material damage prediction method. The meanings of the terms herein are the same as those in the aforementioned micromechanics-based composite material damage prediction method. For specific implementation details, please refer to the description in the method embodiment.

[0266] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of a composite material damage prediction device based on micromechanics provided in an embodiment of the present application. The composite material damage prediction device based on micromechanics may specifically include a data acquisition module 201, a model construction module 202, a model adjustment module 203, and a damage prediction module 204. Specifically, it may be as follows:

[0267] Data acquisition module 201, used to acquire macroscopic mechanical response data and microscopic structural characteristic data of the composite material;

[0268] A model building module 202 is configured to build a multi-mechanism damage prediction model corresponding to the composite material based on the microstructure characteristic data using micromechanics theory;

[0269] A model adjustment module 203 is configured to perform inverse calibration on the model parameters of the multi-mechanism damage prediction model based on the macroscopic mechanical response data, and verify the effectiveness of the multi-mechanism damage prediction model through experimental data;

[0270] The damage prediction module 204 is configured to predict the damage initiation strain of the composite material under different loading conditions and microstructures based on the calibrated multi-mechanism damage prediction model.

[0271] The specific definition of the composite material damage prediction device based on micromechanics can be found in the definition of the composite material damage prediction method based on micromechanics above, and will not be repeated here. The various modules in the above-mentioned composite material damage prediction device based on micromechanics can be implemented in whole or in part through software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0272] The composite material damage prediction device based on micromechanics provided in this embodiment provides basic input for subsequent modeling by acquiring macroscopic mechanical response data and microstructural characteristic data; based on the micromechanics theory, it links the microstructural characteristics with the macroscopic mechanical behavior to construct a damage prediction model that can comprehensively consider multiple damage mechanisms; the model parameters are inversely calibrated through experimental data to ensure the accuracy of the model parameters; the effectiveness of the model is verified through experimental data to ensure the prediction accuracy of the model under different conditions; the calibrated model is used to predict the damage onset strain of the composite material under different loading conditions and microstructures, providing a scientific basis for material performance optimization and design.

[0273] In addition, the present invention also provides an electronic device, such as Figure 5 , which shows a schematic structural diagram of an electronic device involved in an embodiment of the present application, specifically:

[0274] The electronic device may include one or more processors 301 of processing cores, one or more computer-readable storage media memories 302, a power supply 303, an input unit 304 and other components. Those skilled in the art will appreciate that Figure 5 The electronic device structure shown in the figure does not constitute a limitation of the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently.

[0275] The processor 301 is the control center of the electronic device. It connects all parts of the electronic device using various interfaces and lines. By running or executing software programs and / or modules stored in the memory 302 and accessing data stored in the memory 302, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. Optionally, the processor 301 may include one or more processing cores; preferably, the processor 301 may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into the processor 301.

[0276] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and the composite material damage prediction method based on micromechanics by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store an operating system, at least one application required for a function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created based on the use of the electronic device, etc. In addition, the memory 302 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0277] The electronic device also includes a power supply 303 for supplying power to various components. Preferably, the power supply 303 can be logically connected to the processor 301 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. The power supply 303 can also include one or more DC or AC power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and other arbitrary components.

[0278] The electronic device may further include an input unit 304, which may be configured to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0279] Although not shown, the electronic device may further include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 301 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 302 according to the following instructions, and the processor 301 will run the application programs stored in the memory 302 to implement various functions as follows:

[0280] Obtain the macroscopic mechanical response data and microstructural characteristic data of the composite material; construct a multi-mechanism damage prediction model based on the microstructural characteristic data through mesomechanics theory; perform inverse calibration of the model parameters based on the macroscopic mechanical response data, and verify the effectiveness of the multi-mechanism damage prediction model through experimental data; based on the calibrated multi-mechanism damage prediction model, predict the damage onset strain of the composite material under different loading conditions and microstructures.

[0281] The specific implementation of the above operations can be found in the previous embodiments and will not be repeated here.

[0282] The embodiments of the present application provide basic input for subsequent modeling by obtaining macroscopic mechanical response data and microstructural feature data; based on the mesomechanical theory, the microstructural features are linked to the macroscopic mechanical behavior to construct a damage prediction model that can comprehensively consider multiple damage mechanisms; the model parameters are inversely calibrated through experimental data to ensure the accuracy of the model parameters; the effectiveness of the model is verified through experimental data to ensure the prediction accuracy of the model under different conditions; the calibrated model is used to predict the damage onset strain of the composite material under different loading conditions and microstructures, providing a scientific basis for material performance optimization and design.

[0283] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be accomplished by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0284] To this end, an embodiment of the present application provides a storage medium storing a plurality of instructions that can be loaded by a processor to execute any of the steps in the micromechanics-based composite material damage prediction provided in the embodiments of the present application. For example, the instructions can execute the following steps:

[0285] Obtain the macroscopic mechanical response data and microstructural characteristic data of the composite material; construct a multi-mechanism damage prediction model based on the microstructural characteristic data through mesomechanics theory; perform inverse calibration of the model parameters based on the macroscopic mechanical response data, and verify the effectiveness of the multi-mechanism damage prediction model through experimental data; based on the calibrated multi-mechanism damage prediction model, predict the damage onset strain of the composite material under different loading conditions and microstructures.

[0286] The specific implementation of the above operations can be found in the previous embodiments and will not be repeated here.

[0287] The storage medium may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0288] Since the instructions stored in the storage medium can execute the steps in any one of the composite material damage prediction methods based on micromechanics provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the composite material damage prediction methods based on micromechanics provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0289] The above is a detailed introduction to a composite material damage prediction method based on micromechanics and related equipment provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A composite material damage prediction method based on micromechanics, characterized in that: The steps include: Obtain macroscopic mechanical response data and microstructural characteristic data of composite materials; constructing a multi-mechanism damage prediction model corresponding to the composite material based on the microstructure characteristic data through micromechanics theory; Inversely calibrate the model parameters of the multi-mechanism damage prediction model based on the macroscopic mechanical response data, and verify the effectiveness of the multi-mechanism damage prediction model through experimental data; Based on the calibrated multi-mechanism damage prediction model, the damage onset strain of the composite material under different loading conditions and microstructures is predicted.

2. The composite material damage prediction method based on micromechanics according to claim 1, characterized in that: The macroscopic mechanical response data includes the stress-strain relationship under different loading rates and sample sizes, and the microstructural characteristic data includes the size distribution, shape parameters, and spatial distribution density of the filling particles. The method of obtaining the macroscopic mechanical response data and microstructural characteristic data of the composite material includes: Uniaxial tensile tests were performed on composite material specimens, and multiple sets of stress-strain curves were obtained by changing the loading rate and specimen size. The pseudo-strain conversion method is used to eliminate the viscoelastic stress relaxation effect, and the damage initiation strain threshold is determined by analyzing the nonlinear deviation point of the stress-pseudo-strain curve. The three-dimensional imaging of composite materials was performed using micro-computed tomography to obtain the spatial distribution and morphological data of the filling particles; The particle morphology was parametrically characterized based on the equivalent ellipsoid model, and the principal axis length, slenderness ratio and volume fraction of the particles were calculated.

3. The composite material damage prediction method based on micromechanics according to claim 1, characterized in that: The multi-mechanism damage prediction model corresponding to the composite material is constructed based on the microstructure characteristic data through the micromechanics theory, including: The homogenized modulus of the composite material is calculated using the equivalent inclusion theory, and the time-varying viscoelastic behavior of the matrix is ​​treated in combination with the Laplace transform. A statistical failure criterion including matrix tearing, particle fracture and interface dewetting is established. The failure criterion is based on Weibull distribution and modified strength criterion.

4. The composite material damage prediction method based on micromechanics according to claim 3, characterized in that: The statistical failure criteria include: A matrix tearing criterion based on a modified generalized von Mises yield criterion, wherein the matrix tearing criterion introduces an influence factor of hydrostatic pressure on matrix yield strength; A particle fracture criterion based on the Mohr-Coulomb yield criterion, wherein the particle fracture criterion modifies the Weibull distribution parameter in combination with the particle volume effect; A dewetting criterion based on the interface normal stress is defined as a randomly distributed threshold value of the interface strength.

5. The composite material damage prediction method based on micromechanics according to claim 1, characterized in that: The inverse calibration of the model parameters of the multi-mechanism damage prediction model based on the macroscopic mechanical response data includes: Defining an optimization objective function including interface strength, matrix viscoelastic parameters, and statistical damage parameters, wherein the objective function is based on the relative error between the experimental damage onset strain and the model prediction value; A hybrid optimization strategy combining a multi-island genetic algorithm and a pattern search algorithm is adopted to perform global search and local adjustment on the model parameters.

6. The composite material damage prediction method based on micromechanics according to claim 1, characterized in that: The validation of the multi-mechanism damage prediction model through experimental data includes: The experimental data are divided into a model parameter optimization dataset and a model validation dataset; By comparing the predicted values ​​of the multi-mechanism damage prediction model with the experimental values, the effectiveness and accuracy of the multi-mechanism damage prediction model are verified.

7. The composite material damage prediction method based on micromechanics according to claim 1, characterized in that: The calibrated multi-mechanism damage prediction model is used to predict the damage initiation strain of the composite material under different loading conditions and microstructures, including: By homogenizing the representative volume unit, a quantitative correlation between the specimen size and the damage threshold is established; Based on the mapping model between microstructural parameters and macroscopic loading conditions, the damage initiation strain under different particle distributions and loading rates is predicted.

8. A composite material damage prediction device based on micromechanics, characterized in that: include: Data acquisition module, used to obtain macroscopic mechanical response data and microstructural characteristic data of composite materials; A model building module, for building a multi-mechanism damage prediction model corresponding to the composite material based on the microstructure characteristic data using micromechanics theory; a model adjustment module, configured to perform inverse calibration on the model parameters of the multi-mechanism damage prediction model based on the macroscopic mechanical response data, and verify the validity of the multi-mechanism damage prediction model through experimental data; The damage prediction module is used to predict the damage initiation strain of the composite material under different loading conditions and microstructures based on the calibrated multi-mechanism damage prediction model.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the composite material damage prediction method based on micromechanics as described in any one of claims 1 to 7 are implemented.

10. A storage medium, characterized in that: The computer program is stored which can be loaded by a processor and executes the composite material damage prediction method based on micromechanics according to any one of claims 1 to 7.

Images