A method, equipment, and product for predicting the fatigue life of carbon fiber composite materials.

Abstract

This application discloses a method, device, and product for predicting the fatigue life of carbon fiber composite materials, relating to the field of material fatigue life prediction. The method includes: determining four sub-models based on the acquired wave velocity data of the S0 and A0 modes and a set layered damage correlation coefficient, respectively, using a fatigue damage evolution model; estimating and updating the parameters of the corresponding sub-models using Bayesian theory and Monte Carlo sampling methods based on the wave velocity data of the S0 and A0 modes; determining the Bayesian evidence for each updated sub-model using a nested sampling algorithm, and selecting the sub-model with the strongest Bayesian evidence as the optimal sub-model; determining the failure threshold for each specimen using a dynamic failure criterion based on the parameters of the optimal sub-model; and predicting the fatigue life of the carbon fiber composite material based on the failure threshold of each specimen and the optimal sub-model. This application can improve the accuracy and stability of fatigue life prediction for carbon fiber composite materials.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of material fatigue life prediction, and in particular to a carbon fiber composite material fatigue life prediction method, device and product. BACKGROUND

[0002] Carbon fiber composite materials have been widely used in the fields of industry, aerospace, automobile and ship due to their high specific strength, high specific stiffness, fatigue resistance and lightweight, and have a broad development prospect. The market demand for carbon fiber composite materials also shows a trend of continuous growth. Therefore, it is of great significance to study the fatigue state evaluation and fatigue life prediction of carbon fiber composite materials to improve their service safety. The phenomenological model based on strength / stiffness is one of the main models for studying the mechanical property degradation and fatigue life prediction of composite materials. In practical applications, the stiffness of the composite structure cannot be directly measured, so various non-destructive testing techniques are often used to extract signal features sensitive to damage to characterize the stiffness degradation, such as amplitude, transit time, wave speed, etc. However, carbon fiber composite materials have higher modulus than other composite materials, and have fatigue resistance, so the stiffness degradation rate of carbon fiber composite materials is relatively small during the fatigue loading process, and the degradation trend of the detected signal features is also weak and easily affected by noise, resulting in larger prediction error of the model. At the same time, due to the complex manufacturing process of carbon fiber composite materials, they have great dispersion, resulting in great randomness in the damage propagation of each specimen during the fatigue loading process. Therefore, the fatigue life prediction demand of carbon fiber composite materials cannot be met by relying on a single mode of observation data and a single model considering only a single fixed damage form.

[0003] The existing technology mainly considers the uncertainty from the specimen, experimental environment, model, etc. and uses the idea of Bayesian model averaging to realize fatigue life prediction for glass fiber composite laminates. However, the life prediction effect of this method depends to some extent on the degradation trend of the observed wave speed degradation data during the fatigue loading process. Therefore, when the observed data is greatly affected by material, environmental and other factors, a single mode of wave speed degradation data and a single model with fixed damage form cannot adapt to and meet the higher demand for fatigue life prediction. For carbon fiber composite materials, their modulus is about 3-4 times higher than that of glass fiber composite materials. During the fatigue loading process, carbon fiber composite materials have stronger fatigue resistance, longer fatigue life cycle and relatively weaker mechanical property degradation, and the observed wave speed degradation data is more affected by noise and the dispersion between specimens is stronger. It can be seen that the above factors make the existing technology unable to meet the demand for fatigue life prediction of carbon fiber composite materials.

[0004] Therefore, there is an urgent need to provide a life prediction method capable of improving the accuracy and stability of life prediction of carbon fiber composites. SUMMARY

[0005] The purpose of the present application is to provide a carbon fiber composite fatigue life prediction method, device and product, which can improve the accuracy and stability of life prediction of carbon fiber composites.

[0006] To achieve the above-mentioned purpose, the present application provides the following solutions:

[0007] In a first aspect, the present application provides a carbon fiber composite fatigue life prediction method, which comprises:

[0008] In the fatigue experiment of the carbon fiber composite specimen, laser ultrasonic detection technology is used to obtain the wave speed data of S0 mode and the wave speed data of A0 mode of Lamb wave; the wave speed data is used to characterize the degradation trend of wave speed with fatigue loading;

[0009] According to the wave speed data of S0 mode and the wave speed data of A0 mode and the set delamination damage correlation coefficient, four sub-models are determined based on the fatigue damage evolution model respectively; the four sub-models include: fatigue damage evolution model considering delamination damage under S0 mode and fatigue damage evolution model not considering delamination damage, and fatigue damage evolution model considering delamination damage under A0 mode and fatigue damage evolution model not considering delamination damage;

[0010] According to the wave speed data of S0 mode and the wave speed data of A0 mode, the parameters of the corresponding sub-models are estimated and updated by using Bayesian theory and Monte Carlo sampling method respectively;

[0011] The Bayesian evidence of the updated sub-models is determined according to the nested sampling algorithm, and the sub-model with the strongest Bayesian evidence is taken as the best sub-model;

[0012] According to the parameters of the best sub-model, the failure threshold of each specimen is determined by using dynamic failure criterion;

[0013] The carbon fiber composite fatigue life prediction is performed according to the failure threshold of each specimen and the best sub-model.

[0014] Optionally, in the fatigue experiment of the carbon fiber composite specimen, laser ultrasonic detection technology is used to obtain the wave speed data of S0 mode and the wave speed data of A0 mode of Lamb wave, which specifically comprises:

[0015] In the fatigue experiment of the carbon fiber composite specimen, the fatigue experiment system setting is used to control the number of fatigue loadings in each cycle;

[0016] When the carbon fiber composite specimen is loaded to the fatigue loading number, a laser is emitted by the laser ultrasonic testing system to excite a specified area on the surface of the carbon fiber composite specimen, generating a guided wave field signal.

[0017] The guided wave field signal will be collected by a sensor attached to the bottom of the carbon fiber composite specimen;

[0018] The guided wave field signal was subjected to wavelet transform at 300kHz and 100kHz respectively to obtain the wave velocity data of the S0 mode and A0 mode of the Lamb wave.

[0019] Optionally, the fatigue damage evolution model specifically includes the following formula:

[0020]

[0021]

[0022] Where N represents the number of fatigue loading cycles, c represents the wave velocity data extracted under a specific number of fatigue loading cycles, ρ represents the material density, and E x E represents the axial stiffness of the laminate after the fatigue loading number corresponding to the wave velocity c, and E0 represents the initial stiffness of the laminate. x1 D represents the axial stiffness after the first cycle of fatigue loading. f D represents the damage factor corresponding to fiber fracture during the first cycle of fatigue loading. mc D represents the damage factor corresponding to the matrix crack. c D is a parameter representing the critical damage state of a carbon fiber composite specimen. dela D represents the damage factor corresponding to layered damage. mc+dela This represents the sum of damage factors corresponding to matrix cracks and delamination damage, where A and m represent material-related parameters, and σ... x D represents the stress level under fatigue loading. total This represents the total damage factor, which includes fiber fracture, matrix cracks, and delamination damage. The superscripts N+1 and N represent the corresponding fatigue loading number, and β represents the correlation coefficient with delamination damage.

[0023] Optionally, the step of determining the Bayesian evidence of the updated sub-models according to the nested sampling algorithm, and selecting the sub-model with the strongest Bayesian evidence as the best sub-model, specifically includes:

[0024] Z = ∫∫L(θ)p(θ)dθ;

[0025] Where Z represents Bayesian evidence, θ represents model parameters, and the model parameters are A, m, β, and D, respectively. c L(θ) represents the likelihood function, and p(θ) represents the prior distribution of the model parameters.

[0026] Optionally, based on the parameters of the optimal sub-model, a dynamic failure criterion is used to determine the failure threshold for each specimen, specifically including:

[0027] D t =1-(1-D) f )(1-γD c );

[0028] Among them, D t D represents the failure threshold corresponding to the damage factor. c The parameter represents the critical damage state of the carbon fiber composite specimen, where γ is an adjustable parameter between 0 and 1.

[0029] Secondly, this application provides a fatigue life prediction device for carbon fiber composite materials, the fatigue life prediction device for carbon fiber composite materials comprising:

[0030] The wave velocity data acquisition module is used to acquire wave velocity data of the S0 mode and A0 mode of Lamb waves in fatigue tests of carbon fiber composite specimens using laser ultrasonic testing technology; the wave velocity data is used to characterize the degradation trend of wave velocity with fatigue loading.

[0031] The sub-model determination module is used to determine four sub-models based on the wave velocity data of the S0 mode and the A0 mode, as well as the set layered damage correlation coefficient, respectively, based on the fatigue damage evolution model. The four sub-models include: a fatigue damage evolution model considering layered damage and a fatigue damage evolution model not considering layered damage under the S0 mode, and a fatigue damage evolution model considering layered damage and a fatigue damage evolution model not considering layered damage under the A0 mode.

[0032] The parameter estimation and update module for the sub-model is used to estimate and update the parameters of the corresponding sub-model based on the wave velocity data of the S0 mode and the wave velocity data of the A0 mode, using Bayesian theory and Monte Carlo sampling method respectively.

[0033] The optimal sub-model determination module is used to determine the Bayesian evidence of the updated sub-models according to the nested sampling algorithm, and to select the sub-model with the strongest Bayesian evidence as the optimal sub-model.

[0034] The failure threshold determination module is used to determine the failure threshold of each specimen based on the parameters of the optimal sub-model and using dynamic failure criteria.

[0035] The fatigue life prediction module is used to predict the fatigue life of carbon fiber composites based on the failure threshold and the optimal sub-model for each specimen.

[0036] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the carbon fiber composite material fatigue life prediction method.

[0037] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the carbon fiber composite material fatigue life prediction method.

[0038] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0039] This application provides a method, device, and product for predicting the fatigue life of carbon fiber composite materials. Based on existing fatigue damage evolution models and combined with Bayesian model selection, a method for predicting the fatigue life of carbon fiber composite materials is proposed. The method considers a fatigue damage evolution model with independent delamination coefficients. Depending on whether the independent delamination coefficient is set to 0, two cases are considered: delamination damage and non-delamination damage, to address the issue of large differences in damage propagation between specimens. The wave velocity data of the two modes are combined with the two delamination damage cases to form four sub-models. Then, leveraging Bayesian evidence to comprehensively evaluate data-model matching and model complexity, model selection is performed, enabling the selection of the sub-model that best matches the current damage state at any point during specimen fatigue loading. Furthermore, in the specific fatigue life prediction module, a dynamic failure criterion is used to determine the failure threshold for each specimen. This further improves the accuracy and stability of fatigue life prediction for carbon fiber composite materials based on the use of multi-modal wave velocity and model selection. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is an application environment diagram of a method for predicting the fatigue life of carbon fiber composite materials according to an embodiment of this application.

[0042] Figure 2 A flowchart illustrating a method for predicting the fatigue life of carbon fiber composite materials according to an embodiment of this application;

[0043] Figure 3This is a schematic diagram of an overall method for predicting the fatigue life of carbon fiber composite materials according to an embodiment of this application;

[0044] Figure 4 This is a visual schematic diagram of a method for predicting the fatigue life of carbon fiber composite materials provided in an embodiment of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] The fatigue life prediction method for carbon fiber composite materials provided in this application can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send wave velocity data of S0 mode and A0 mode to server 104. After receiving the wave velocity data of S0 mode and A0 mode, server 104 determines four sub-models based on the wave velocity data of S0 mode and A0 mode and the set layered damage correlation coefficient, respectively, based on the fatigue damage evolution model. According to the wave velocity data of S0 mode and A0 mode, server 104 estimates and updates the parameters of the corresponding sub-models using Bayesian theory and Monte Carlo sampling method. According to the nested sampling algorithm, the Bayesian evidence of the updated sub-models is determined, and the sub-model with the strongest Bayesian evidence is selected as the optimal sub-model. According to the parameters of the optimal sub-model, the failure threshold of each specimen is determined using dynamic failure criteria. According to the failure threshold of each specimen and the optimal sub-model, the fatigue life of carbon fiber composite material is predicted. Server 104 can feed back the predicted fatigue life of carbon fiber composite material to terminal 102. In addition, in some embodiments, the fatigue life prediction method for carbon fiber composite materials can also be implemented by the server 104 or the terminal 102 separately. For example, the terminal 102 can directly process the wave velocity data of the S0 mode and the wave velocity data of the A0 mode, or the server 104 can obtain the wave velocity data of the S0 mode and the wave velocity data of the A0 mode from the data storage system and process the wave velocity data of the S0 mode and the wave velocity data of the A0 mode.

[0048] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0049] In one exemplary embodiment, such as Figure 2 As shown, a method for predicting the fatigue life of carbon fiber composite materials is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps S201 to S206. Among them:

[0050] S201. In the fatigue test of carbon fiber composite specimens, laser ultrasonic testing technology was used to obtain the wave velocity data of the S0 mode and A0 mode of Lamb wave; the wave velocity data was used to characterize the degradation trend of wave velocity with fatigue loading.

[0051] S202. Based on the wave velocity data of S0 mode and A0 mode, and the set layered damage correlation coefficient, four sub-models are determined based on the fatigue damage evolution model. The four sub-models include: a fatigue damage evolution model considering layered damage and a fatigue damage evolution model not considering layered damage under S0 mode, and a fatigue damage evolution model considering layered damage and a fatigue damage evolution model not considering layered damage under A0 mode.

[0052] S203. Based on the wave velocity data of the S0 mode and the A0 mode, Bayesian theory and Monte Carlo sampling method are used to estimate and update the parameters of the corresponding sub-models respectively.

[0053] S204. The Bayesian evidence of the updated sub-models is determined according to the nested sampling algorithm, and the sub-model with the strongest Bayesian evidence is taken as the best sub-model.

[0054] S205, Based on the parameters of the optimal sub-model, the failure threshold of each specimen is determined using dynamic failure criteria;

[0055] S206, predict the fatigue life of carbon fiber composites based on the failure threshold and optimal sub-model for each specimen.

[0056] This application utilizes the obtained wave velocity data of the S0 mode and A0 mode. The A0 mode has the advantages of stronger signal, larger amplitude, and less susceptibility to noise, especially in the later stages of fatigue loading. Due to the accumulation of fatigue damage, the detected wave field signal becomes weaker and weaker, and the S0 mode signal is not strong, resulting in a much larger wave velocity extraction error compared to the initial state. Based on this, using the A0 mode to define the damage factor can improve the problem of the large wave velocity error of the S0 mode in the later stages of fatigue loading, which cannot reflect the true fatigue damage state. On the other hand, in existing technologies, the A0 mode is mostly used for damage detection, etc. Considering the use of the A0 mode for fatigue damage evolution and life prediction provides a new development potential for the application of the A0 mode. Based on the existing fatigue damage evolution model, combined with the Bayesian model, a method for predicting the fatigue life of carbon fiber composites is proposed. A dynamic failure criterion is used to determine the failure threshold of each specimen, further ensuring the accuracy and stability of the fatigue life prediction of carbon fiber composites. Moreover, this application can predict the fatigue life of carbon fiber composites at any stage of fatigue loading based on the observed wave velocity data. The specific technical solution flowchart is as follows. Figure 3 As shown, Figure 4 forFigure 3 A flowchart combining actual calculation results.

[0057] In an exemplary embodiment, S201 specifically includes:

[0058] S1, In the fatigue test of carbon fiber composite specimens, the fatigue test system settings are used to control the number of fatigue loading cycles in each period.

[0059] S2, When the carbon fiber composite specimen is loaded to the fatigue loading number, the laser ultrasonic testing system emits a laser to excite a specified area on the surface of the carbon fiber composite specimen, generating a guided wave field signal;

[0060] S3, the guided wave field signal will be collected by the sensor attached to the bottom of the carbon fiber composite specimen;

[0061] S4. Perform wavelet transform on the guided wave field signal at 300kHz and 100kHz respectively to obtain the wave velocity data of the S0 mode and A0 mode of the Lamb wave.

[0062] In one exemplary embodiment, the fatigue damage evolution model specifically includes the following formula:

[0063]

[0064] Where N represents the number of fatigue loading cycles, c represents the wave velocity data extracted under a specific number of fatigue loading cycles, ρ represents the material density, and E x E represents the axial stiffness of the laminate after the fatigue loading number corresponding to the wave velocity c, and E0 represents the initial stiffness of the laminate. x1 D represents the axial stiffness after the first cycle of fatigue loading. f D represents the damage factor corresponding to fiber fracture during the first cycle of fatigue loading. mc D represents the damage factor corresponding to the matrix crack. c D is a parameter representing the critical damage state of a carbon fiber composite specimen. dela D represents the damage factor corresponding to layered damage. mc+dela This represents the sum of damage factors corresponding to matrix cracks and delamination damage, where A and m represent material-related parameters, and σ... x D represents the stress level under fatigue loading. total This represents the total damage factor, which includes fiber fracture, matrix cracks, and delamination damage. The superscripts N+1 and N represent the corresponding fatigue loading number, and β represents the correlation coefficient with delamination damage.

[0065] Fiber fracture, matrix cracking, and delamination are the three main forms of fatigue damage in composite materials. As damage accumulates, the stiffness of the laminate decreases. To characterize the stiffness degradation process, the concept of a damage factor is defined in the model. Substituting equation (1) into equations (2) and (3) respectively, the damage factor D can be calculated based on the extracted wave velocity data. f and damage factor D under different fatigue loading cycles mc+dela The value;

[0066] D c It can be calculated through material parameters and layup conditions, representing a critical damage state when all layups other than the 0° layup completely lose their load-bearing capacity and the matrix cracks reach saturation. Due to the large dispersion of specimens, in this application, it is assumed that the critical damage state of each specimen is different.

[0067] By considering the calculation of damage factors using different wave velocities and dividing the model into two cases—considering and not considering layered damage—by setting the layered correlation coefficient β, the above fatigue damage evolution model is developed into four sub-models, such as... Figure 3 As shown.

[0068] In an exemplary embodiment, S203 specifically includes:

[0069] For the sub-model considering layered damage, the model parameters are A, m, and D. c For a model that does not consider layered damage, the model parameters are A, m, and D. c According to Bayesian theory, the product of the prior distribution and the likelihood function can approximate the posterior probability density distribution. The prior distribution is generally given artificially based on prior knowledge of the parameters. Experimental data (wave velocity data) are input into the likelihood function. Based on this, and combined with the Monte Carlo sampling algorithm, the posterior probability density distribution of the model parameters can be obtained by sampling the model parameters. Simultaneously, as wave velocity data is continuously input, the posterior probability density distribution is continuously updated, thereby updating the model parameters. In this step regarding the setting of the prior distribution, D... c It can be calculated from the material parameters, therefore D c The prior distribution is set to a normal distribution with the calculated values ​​as the mean. Meanwhile, for the model parameters A, m, and β, since the prior knowledge is insufficient, they are all set to a uniform distribution.

[0070] Based on the Bayesian parameter update, Bayesian evidence is calculated for the sub-model at different fatigue stages using observed wave velocity data. The formula for the Bayesian evidence is as follows:

[0071] Z=∫∫L(θ)p(θ)dθ (6)

[0072] Where Z represents Bayesian evidence, θ represents model parameters, and the model parameters are A, m, β, and D, respectively. c L(θ) represents the likelihood function, and p(θ) represents the prior distribution of the model parameters.

[0073] The Nested Sampling (NS) algorithm was chosen to compute Bayesian evidence. The core idea of ​​this algorithm is to transform multidimensional integrals into one-dimensional integrals. First, the model parameters are sampled within a given prior distribution range. At the same time, the corresponding likelihood function is calculated based on the sampled model parameters. The Bayesian evidence is then segmented into discrete regions by sorting the likelihood functions. Through continuous sampling iterations, the algorithm continuously approaches the true Bayesian evidence value.

[0074] Based on the above method, Bayesian evidence can be calculated for each sub-model based on the given wave velocity data. The larger the calculated Bayesian evidence value, the better the model fits the observed data and the more it matches the current fatigue damage state. Therefore, the sub-model with the largest Bayesian evidence value is taken as the current model selection result, and finally the model selection is realized at different stages of fatigue loading (reflected by different amounts of given wave velocity data).

[0075] Based on model selection using Bayesian evidence, the failure threshold for each specimen will be determined according to the dynamic failure criterion. The formula for calculating the failure threshold is as follows:

[0076] D t =1-(1-D) f )(1-γD c (7)

[0077] Among them, D t D represents the failure threshold corresponding to the damage factor. c The parameter γ represents the critical damage state of the carbon fiber composite specimen, where γ is an adjustable parameter between 0 and 1. Due to the dispersion among specimens, the dynamic failure criterion assumes that the critical damage of each specimen is different, thus defining D... c As observed data is continuously input, parameter D is used as a model parameter. c The update, D in formula (7) c The value is D. c The mean of the sampled specimens. In addition, γ is an adjustable parameter with a value between 0 and 1. By taking about 3 specimens and performing lifetime predictions at different set γ values, the value with the smallest error between the lifetime prediction result and the experiment is selected. Then, the average value among the specimens is taken as the final value of γ. At the same time, the value of γ can also be adjusted to meet more conservative lifetime prediction requirements, etc.

[0078] The Bayesian model selection addresses the issue of high dispersion among carbon fiber composite specimens. The four sub-models proposed in this application differ primarily in the selected waveguide velocities corresponding to different modes and whether or not delamination damage is considered. These two considerations respectively address the problems of high noise in observed data and high specimen dispersion in fatigue life prediction of carbon fiber composites. The selection of different waveguide velocities mainly affects the model's fit to the observed data; from the perspective of the number of model parameters, whether or not delamination damage is considered affects the model's complexity. Bayesian evidence, being the integral of the likelihood function and prior distribution over the parameter space, provides a comprehensive evaluation of the model for a given model, considering both model-data fit and model complexity. In conclusion, using Bayesian evidence to evaluate the four sub-models established in this application allows for the selection of the sub-model that best matches the fatigue damage state, achieving an accurate assessment of the fatigue damage state.

[0079] Model selection was performed at different fatigue stages using Bayesian evidence, and the updated model results were obtained by combining the Bayesian parameter update results. Based on this, the calculated failure threshold D was... t Input into the selected sub-model, due to the failure threshold D t It is related to the total damage factor D total Correspondingly, given a failure threshold, the damage factor D can be obtained through formula (5). mc+dela The corresponding failure threshold is then substituted into formula (4), and the predicted fatigue life can be obtained through numerical integration. Since the sub-model has already obtained the posterior probability density distribution of the model parameters, the predicted fatigue life also has a probability distribution. The final life prediction result is as follows: Figure 4 As shown.

[0080] The above steps are repeated for each specimen, thereby enabling the prediction of fatigue life of each specimen at different fatigue stages.

[0081] This application has the following effects:

[0082] 1. The use of multimodal wave velocity proposed in this application is based on the use of S0 mode wave velocity to define the damage factor, and provides a new option by using A0 mode wave velocity. First, it can reduce the model error caused by noise. Second, it can improve the problem of inaccurate model prediction caused by the increase of S0 mode extraction error in the later stage of fatigue loading. It can reduce the model's dependence on single mode wave velocity data. Finally, the original scheme of using S0 mode wave velocity is still one of the options, retaining the advantage of S0 mode in effectively characterizing axial stiffness.

[0083] 2. The fatigue damage evolution model with independent delamination coefficient considered in this application will not affect the damage evolution of matrix cracks regardless of whether delamination damage is considered. On the other hand, by using the delamination damage coefficient β as a model parameter and estimating and updating the parameter through Bayesian theory, it can adapt to the situation where the degree of damage propagation is random due to the large dispersion of carbon fiber composite specimens.

[0084] 3. The Bayesian model selection method proposed in this application can select the sub-model that best matches the current fatigue damage state based on the observed data at any stage of fatigue loading, thereby realizing the fatigue damage evolution of carbon fiber composite laminates throughout the entire life cycle. Compared with using only a single fixed model for life prediction, the application of model selection can effectively improve the accuracy and stability of fatigue life prediction, and can achieve effective prediction even in the early stage of fatigue loading.

[0085] Based on the same inventive concept, this application also provides a carbon fiber composite fatigue life prediction device for implementing the aforementioned carbon fiber composite fatigue life prediction method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the carbon fiber composite fatigue life prediction device provided below can be found in the limitations of the carbon fiber composite fatigue life prediction method described above, and will not be repeated here.

[0086] In one exemplary embodiment, a fatigue life prediction device for carbon fiber composite materials is provided, comprising:

[0087] The wave velocity data acquisition module is used to acquire wave velocity data of the S0 mode and A0 mode of Lamb waves in fatigue tests of carbon fiber composite specimens using laser ultrasonic testing technology; the wave velocity data is used to characterize the degradation trend of wave velocity with fatigue loading.

[0088] The sub-model determination module is used to determine four sub-models based on the wave velocity data of the S0 mode and the A0 mode, as well as the set layered damage correlation coefficient, respectively, based on the fatigue damage evolution model. The four sub-models include: a fatigue damage evolution model considering layered damage and a fatigue damage evolution model not considering layered damage under the S0 mode, and a fatigue damage evolution model considering layered damage and a fatigue damage evolution model not considering layered damage under the A0 mode.

[0089] The parameter estimation and update module for the sub-model is used to estimate and update the parameters of the corresponding sub-model based on the wave velocity data of the S0 mode and the wave velocity data of the A0 mode, using Bayesian theory and Monte Carlo sampling method respectively.

[0090] The optimal sub-model determination module is used to determine the Bayesian evidence of the updated sub-models according to the nested sampling algorithm, and to select the sub-model with the strongest Bayesian evidence as the optimal sub-model.

[0091] The failure threshold determination module is used to determine the failure threshold of each specimen based on the parameters of the optimal sub-model and using dynamic failure criteria.

[0092] The fatigue life prediction module is used to predict the fatigue life of carbon fiber composites based on the failure threshold and the optimal sub-model for each specimen.

[0093] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores wave velocity data. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for predicting the fatigue life of carbon fiber composite materials.

[0094] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0095] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0096] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0097] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0098] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0099] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0100] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting fatigue life of a carbon fiber composite material, characterized by, The carbon fiber composite material fatigue life prediction method comprises: In the fatigue experiment of the carbon fiber composite material test piece, laser ultrasonic detection technology is used to obtain wave velocity data of S0 mode and A0 mode of Lamb wave; the wave velocity data is used to characterize the degradation trend of wave velocity with fatigue loading; According to the wave velocity data of S0 mode and A0 mode and the set delamination damage correlation coefficient, four sub-models are determined based on the fatigue damage evolution model respectively; the four sub-models include: fatigue damage evolution models considering delamination damage and not considering delamination damage under S0 mode and fatigue damage evolution models considering delamination damage and not considering delamination damage under A0 mode; According to the wave velocity data of S0 mode and A0 mode, the parameters of the corresponding sub-models are estimated and updated by using Bayesian theory and Monte Carlo sampling method respectively; According to the nested sampling algorithm, the Bayesian evidence of the updated sub-models is determined respectively, and the sub-model with the strongest Bayesian evidence is taken as the best sub-model; According to the parameters of the best sub-model, the failure threshold of each test piece is determined by using the dynamic failure criterion; According to the failure threshold of each test piece and the best sub-model, the fatigue life of the carbon fiber composite material is predicted; In the fatigue experiment of the carbon fiber composite material test piece, laser ultrasonic detection technology is used to obtain wave velocity data of S0 mode and A0 mode of Lamb wave, and specifically comprises: In the fatigue experiment of the carbon fiber composite material test piece, the fatigue experiment system is set to control the number of fatigue loadings in each cycle; When the carbon fiber composite material test piece is loaded to the number of fatigue loadings, the laser ultrasonic detection system emits laser to excite the specified area on the surface of the carbon fiber composite material test piece to generate guided wave field signals; The sensors pasted at the bottom of the carbon fiber composite material test piece are used to collect the guided wave field signals; The guided wave field signals are wavelet transformed at 300 kHz and 100 kHz respectively to obtain wave velocity data of S0 mode and A0 mode of Lamb wave.

2. The carbon fiber composite material fatigue life prediction method according to claim 1, characterized by, The fatigue damage evolution model specifically comprises the following formula: where N denotes the number of fatigue loadings, c denotes the wave speed data extracted at a specific number of fatigue loadings, p denotes the material density, E x denotes the axial stiffness of the laminate at the number of fatigue loadings corresponding to the wave speed c, E0denotes the initial stiffness of the laminate, E x1 denotes the axial stiffness after the first cycle of fatigue loading, D f denotes the damage factor corresponding to fiber breakage at the first cycle of fatigue loading, D mc denotes the damage factor corresponding to matrix cracking, D c denotes the parameter representing the critical damage state of the carbon fiber composite specimen, D dela denotes the damage factor corresponding to delamination damage, D mc+dela denotes the sum of the damage factors corresponding to matrix cracking and delamination damage, A and m denote parameters related to the material, s x denotes the stress level of the fatigue loading, D total denotes the total damage factor including fiber breakage, matrix cracking, and delamination damage, the superscripts N+1and N denote the corresponding number of fatigue loadings, and b denotes a coefficient related to delamination damage.

3. The carbon fiber composite material fatigue life prediction method according to claim 2, characterized by, According to the nested sampling algorithm, the Bayesian evidence of the updated sub-models is determined respectively, and the sub-model with the strongest Bayesian evidence is taken as the best sub-model, specifically comprising: Z=∫∫L(θ)p(θ)dθ; where Z denotes the Bayesian evidence, θ denotes the model parameters, the model parameters are A, m, β, and D, respectively c L(θ) denotes the likelihood function, and p(θ) denotes the prior distribution of the model parameters.

4. The carbon fiber composite material fatigue life prediction method according to claim 3, characterized by, According to the parameters of the best sub-model, the failure threshold of each test piece is determined by using the dynamic failure criterion, specifically comprising: D t = 1 - (1 - D f )(1 - γD c ); where D t represents the failure threshold corresponding to the damage factor, D c represents a parameter of the critical damage state of the carbon fiber composite test piece, and γ represents an adjustable parameter between 0 and 1.

5. A carbon fiber composite material fatigue life prediction device for implementing the carbon fiber composite material fatigue life prediction method according to any one of claims 1 to 4, characterized by, The carbon fiber composite material fatigue life prediction device comprises: A wave velocity data acquisition module is configured to acquire wave velocity data of S0 mode and A0 mode of Lamb wave by using laser ultrasonic detection technology in the fatigue experiment of the carbon fiber composite material test piece; the wave velocity data is used to characterize the degradation trend of wave velocity with fatigue loading; The sub-model determination module is configured to determine four sub-models based on a fatigue damage evolution model according to wave velocity data of the S0 mode and wave velocity data of the A0 mode and a set delamination damage correlation coefficient; the four sub-models include: a fatigue damage evolution model considering delamination damage under the S0 mode, a fatigue damage evolution model not considering delamination damage under the S0 mode, a fatigue damage evolution model considering delamination damage under the A0 mode, and a fatigue damage evolution model not considering delamination damage under the A0 mode; The parameter estimation and updating module of the sub-model is configured to estimate and update parameters of the corresponding sub-model according to the wave velocity data of the S0 mode and the wave velocity data of the A0 mode by using a Bayesian theory and a Monte Carlo sampling method; The optimal sub-model determination module is configured to determine Bayesian evidences of the updated sub-models according to a nested sampling algorithm, and determine the sub-model with the strongest Bayesian evidence as the optimal sub-model; The failure threshold determination module is configured to determine a failure threshold of each test piece by using a dynamic failure criterion according to parameters of the optimal sub-model; The fatigue life prediction module is configured to predict a carbon fiber composite material fatigue life according to the failure threshold of each test piece and the optimal sub-model.

6. A computer device comprising: A memory and a processor to store a computer program stored on the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the carbon fiber composite material fatigue life prediction method of any one of claims 1-4.

7. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the carbon fiber composite material fatigue life prediction method of any one of claims 1-4.

Images