Steel fatigue life prediction method based on physical information neural network of meta learning

Abstract

The application belongs to the technical field of machine learning, and discloses a physical information neural network steel fatigue life prediction method based on meta learning. By introducing a meta model learning sample corresponding to a loading path geometric feature with a GRU module, the weight is updated for different samples through model reasoning, so that the training performance of the physical information neural network is optimized. Unlike static optimization, a group of weights is independently updated and optimized for a specific steel and a sample under a multi-axis loading path. The method introduces the path geometric information with the non-proportional cyclic hardening degree metric through the meta model, so that the prediction result of the physical information neural network has good physical consistency, and has better performance in the multi-axis fatigue life prediction of the steel under a small sample.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of machine learning, and particularly relates to a steel fatigue life prediction method. BACKGROUND

[0002] Fatigue failure is one of the most common failure modes of metal materials under cyclic loading. Compared with uniaxial fatigue life prediction, multiaxial fatigue is more complex due to the involvement of non-proportional additional cyclic hardening effect. In order to ensure the reliability and safety of metal structural components, multiaxial fatigue life prediction has become a research focus. However, although traditional fatigue tests are an important means to study the fatigue properties of materials, they usually consume a large amount of time and cost, and due to the complexity of multiaxial loading path and working conditions, relying solely on experiments to predict the fatigue life of metal materials under complex conditions is obviously limited.

[0003] With the emergence of a large number of empirical methods in recent years, these methods have established various models based on damage parameters such as strain, stress, energy and fracture mechanism. Among them, the critical plane method, as a representative of energy methods, has shown excellent performance in empirical models due to its comprehensive consideration of multiple factors. However, although the prediction accuracy of the critical plane method is relatively high, its applicability is still limited, and the prediction results can only be concentrated within the five times error band.

[0004] The progress of modern computer technology and multiaxial fatigue experimental technology has enabled the continuous accumulation of multiaxial fatigue data for specific metal materials. Based on these data, training artificial neural networks and other models through data-driven algorithms such as machine learning has become a popular research direction. However, due to the "black box" characteristics of machine learning models and their dependence on a large number of samples, these methods have poor prediction results in small sample cases, and lack sufficient explainability and physical background, which severely limits their widespread application.

[0005] Since the physical information neural network was proposed in 2019, as an innovative model that combines physical information and data-driven, it has achieved remarkable results in multiple fields. Developing such a neural network model that combines physical information and data-driven for metal multiaxial fatigue life prediction not only introduces physical background and improves the explainability of the model, but also significantly improves the prediction accuracy under small sample conditions, providing a powerful solution to the limitations of traditional methods and data-driven methods. SUMMARY

[0006] The application aims to provide a physical information neural network steel fatigue life prediction method based on meta-learning, which is based on a critical surface method and damage parameters thereof, constructs a physical information neural network and a physical information loss function, and uses physical information as a meta-model input, thereby introducing physical meaning and background into a data-driven method, increasing the explainability, and ensuring the further improvement of the prediction ability under a small sample.

[0007] In order to achieve the above-mentioned application purposes, the application is implemented by the following technical solutions:

[0008] According to one aspect of the application, a physical information neural network steel fatigue life prediction method based on meta-learning is provided, which comprises the following steps:

[0009] S1, collecting a fatigue experimental data set of a steel to be predicted;

[0010] S2, after selecting a critical surface model according to the steel to be predicted, calculating the parameters required by the critical surface model according to the fatigue experimental data set, and calculating the damage parameters;

[0011] S3, determining the physical information neural network input features and the physical information neural network output features, and performing data preprocessing on the physical information neural network input features and the physical information neural network output features; wherein the physical information neural network input features are the damage parameters obtained in S2, and the physical information neural network output features are fatigue life prediction results;

[0012] S4, constructing a physical information neural network based on an artificial neural network as a basic model, and defining a physical information loss function according to the critical surface model selected in S2, while introducing a physical information term weight;

[0013] S5, using the normal strain and shear strain information in the fatigue experimental data set obtained in S1 as meta-model input features, using the physical information term weight in S4 as meta-model output features, and performing data preprocessing on the meta-model input features;

[0014] S6, constructing a meta-model, and defining the mean square error (MSE) of the physical information neural network as a meta-model loss function;

[0015] S7, randomly dividing the data sets of the meta-model input features, the physical information neural network input features, and the physical information neural network output features into a training set and a test set, and ensuring that the same samples of the meta-model and the physical information neural network are simultaneously divided in the training set or the test set; initializing the meta-model and the physical information neural network, initializing the physical information term weight, and setting the meta-model training period and the physical information neural network training period;

[0016] S8, according to the current physical information item weight, performing physical information neural network training, updating the physical information neural network output feature, calculating the physical information loss function and back propagation to the physical information neural network;

[0017] S9, in the last cycle of physical neural network training, calculating the mean square error MSE of the physical information neural network, and back propagation to the meta-model;

[0018] S10, performing meta-model training, updating the physical information item weight;

[0019] S11, repeating the steps of S8-S10 until the set meta-model training cycle is reached, and obtaining the fatigue life prediction result through the physical information neural network.

[0020] Further, in S2, for the steel material of shear type fatigue fracture, the critical plane model adopts at least two of the FS model, the MGS Yu model, the MGSE Zhu model, the WB model, the Liu model, the CXH model, the Zhu model or the GSE model; for the steel material of tensile type fatigue fracture, the critical plane model adopts at least two of the SWT model, the Liu model, the CXH model, the Zhu model and the GSE model.

[0021] Further, in S2, for carbon steel, the critical plane model adopts the FS model, the SWT model, the MGSE Zhu model, the WB model and the CXH(T) model:

[0022] FS model:

[0023] SWT model:

[0024] MGSE Zhu model:

[0025] WB model:

[0026] CXH(T) model:

[0027] Wherein, σ nmax , τ nmax , Δε nmax , Δγ nmax are, in turn, the maximum normal stress, the maximum shear stress, the maximum normal strain range, the maximum normal stress range corresponding to the critical plane when the plane where the maximum shear strain range is located is taken as the critical plane; σ smax , τ smax , Δε smax , Δγ smaxThey are the maximum normal stress, maximum shear stress, maximum normal strain range, and maximum normal stress range corresponding to the critical surface when the plane with the maximum normal strain range is taken as the critical surface; Δε n is the positive strain stroke; σ' f is the fatigue strength coefficient, τ' f is the shear fatigue strength coefficient, ε' f is the fatigue ductility coefficient, γ' f is the shear fatigue ductility coefficient, b is the fatigue strength index, c is the fatigue ductility index, b0 is the shear fatigue strength index, c0 is the shear fatigue ductility index, k is the FS model constant, k MGSE is the MGSEZhu model constant, S is the WB model constant; E is Young's modulus, G is the shear modulus, σ y is the yield strength, N f is fatigue life; v e is the Poisson's ratio in the elastic stage, which is 0.3; v p is the Poisson's ratio in the plastic stage, which is taken as 0.5;

[0028] Among them, the parameters required for the critical surface model are σ' f , τ' f , ε' f ,γ' f ,b,c,b0,c0,k,k MGSE and S;

[0029] Wherein, the damage parameter is: nmax , τ nmax , Δε nmax , Δγ nmax , σ smax , τ smax , Δε smax , Δγ smax and Δε n .

[0030] Furthermore, in S3, the physical information neural network input feature is σ nmax , τ nmax , Δε nmax , Δγ nmax , σ smax , τ smax , Δε smax , Δγ smax .

[0031] Furthermore, in S3, data preprocessing of the physical information neural network input features and the physical information neural network output features includes:

[0032] Normalize the physical information neural network input features and the logarithmized physical information neural network output features:

[0033]

[0034] Among them, x refers to the input characteristics of each physical information neural network or the logarithmic output characteristics of the physical information neural network, x * represents the normalized value, μ represents the mean of the input features of each physical information neural network or the logarithmic output features of the physical information neural network, and σ represents the standard deviation of the input features of each physical information neural network or the logarithmic output features of the physical information neural network.

[0035] Furthermore, in S4, the physical information loss function is:

[0036] L PINN =MSE+ω1L FS +ω2L SWT +ω3L MGSEZhu +ω4L WB +ω5L CXH(T)

[0037] in:

[0038]

[0039]

[0040] ω1, ω2, ω3, ω4 and ω5 are the weights of physical information items, i is the sample number, n is the number of samples, N exp,i is the experimental life of the i-th sample, N pre,i is the predicted lifespan of the sample numbered i;

[0041] Among them, σ nmax , τ nmax , Δε nmax , Δγ nmax They are the maximum normal stress, maximum shear stress, maximum normal strain range, and maximum normal stress range corresponding to the critical surface when the plane with the maximum shear strain range is taken as the critical surface; σ smax , τ smax , Δε smax , Δγ smax They are the maximum normal stress, maximum shear stress, maximum normal strain range, and maximum normal stress range corresponding to the critical surface when the plane with the maximum normal strain range is taken as the critical surface; Δε n is the positive strain stroke; σ' f is the fatigue strength coefficient, τ' f is the shear fatigue strength coefficient, ε' f is the fatigue ductility coefficient, γ' fis the shear fatigue ductility coefficient, b is the fatigue strength index, c is the fatigue ductility index, b0 is the shear fatigue strength index, c0 is the shear fatigue ductility index, k is the FS model constant, k MGSE is the MGSE Zhu model constant, S is the WB model constant; E is the Young's modulus, G is the shear modulus, and y is the yield strength, N f is the fatigue life; v e is the elastic stage Poisson's ratio, and 0.3 is taken; v p is the plastic stage Poisson's ratio, and 0.5 is taken.

[0042] Further, in S5, the data preprocessing of the meta-model input features includes: normalizing the meta-model input features:

[0043]

[0044] wherein y indicates the meta-model input features, and y' represents the normalized value.

[0045] Further, in S9, the root mean square error MSE of the physical information neural network is:

[0046]

[0047] wherein i is the sample serial number, n is the sample quantity, N exp,i is the experimental life of the i-th sample, N pre,i is the predicted life of the i-th sample.

[0048] According to another aspect of the present application, a computer device is provided, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the computer program to implement the above-mentioned meta-learning-based physical information neural network steel fatigue life prediction method.

[0049] According to another aspect of the present application, a computer readable storage medium is provided, having a computer program stored thereon, the computer program being executed by a processor to implement the above-mentioned meta-learning-based physical information neural network steel fatigue life prediction method.

[0050] The present application has the following beneficial effects:

[0051] (1) The physical information neural network steel multi-axial fatigue life prediction method under meta-learning of the present invention utilizes a meta-learning framework, introduces a meta-model with GRU as the module to learn the geometric features of the loading path corresponding to the sample, and updates the weights for the loading features corresponding to different samples through model reasoning, thereby optimizing the training performance of the physical information neural network. Different from static optimization, for a specific steel and a sample under a multi-axial loading path, a set of weights are independently updated and optimized. This method introduces path geometry information with a non-proportional cyclic hardening degree metric through the meta-model, so that the prediction results of the physical information neural network have better physical consistency, and have better performance in the prediction of multi-axial fatigue life of steel under small samples.

[0052] (2) The meta-learning-based physical information neural network steel multi-axial fatigue life prediction method of the present invention, wherein the meta-learning framework, as a hyperparameter optimization method, can effectively extract feature differences under different conditions by introducing key physical information as meta-model input features. By dynamically optimizing the hyperparameters of the physical information neural network, the prediction performance of the physical information neural network is made closer to the true value. At the same time, the search space is effectively reduced, so that the physical information items in the physical information neural network can fully play their role under different conditions.

[0053] (3) The physical information neural network steel multi-axial fatigue life prediction method under meta-learning of the present invention, in which the physical information neural network loss function has good generalization for fatigue life prediction of different types of steels with different multi-axial loading paths due to the introduction of the weights of the physical information items and the dynamic update of the weights of the physical information items by the meta-learning framework.

[0054] (IV) The method for multi-axial fatigue life prediction of steel using a physical information neural network under meta-learning of the present invention applies the physical information neural network optimized under the meta-learning framework to the field of multi-axial fatigue life prediction of steel, and by introducing the physical information term in the loss function and the loading path geometry information as the meta-model input, the empirical model in the field is used as the physical meaning guidance for model training, thereby obtaining a machine learning life prediction method with both interpretability and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0056] Figure 1 This is a schematic diagram of the framework of the physical information neural network steel fatigue life prediction method based on meta-learning provided by the present invention;

[0057] Figure 2 A schematic diagram of the implementation process of the physical information neural network in the method provided by the present invention;

[0058] Figure 3 The ablation experimental performance of the method provided by the present invention on three metal materials: 316LN, 316L and 304;

[0059] Figure 4 The comparison between the predicted value and experimental value of fatigue life of three metal material data sets of 316LN, 316L and 304 by the method provided by the present invention. DETAILED DESCRIPTION

[0060] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0061] To solve the problems in the background technology, the present invention provides the following solutions to achieve both accuracy and interpretability when predicting the multiaxial fatigue life of steel:

[0062] Traditional empirical models offer good physical meaning, but due to experimental bias and the theoretical nature of these models, their accuracy in predicting the multiaxial fatigue life of steel is limited. Data-driven methods offer good accuracy for predicting the multiaxial fatigue life of steel, but due to the "black box" nature of data-driven models like machine learning, these models lack physical meaning and interpretability. Furthermore, due to the scarcity of fatigue data for complex paths, improving the accuracy of fatigue life predictions using small sample sizes has become a key area of ​​research for data-driven methods.

[0063] Meta-learning refers to a branch of machine learning that aims to design algorithms that can learn how to learn. It accelerates the hyperparameter search process for new tasks by learning optimization strategies on related tasks.

[0064] In order to propose a method that has both physical meaning, interpretability and accuracy, the present invention constructs a loss function based on the critical surface model in the traditional empirical method as the source of physical information, and constructs a physical information neural network on this basis. The weights of each sub-item in the physical information sub-item of the loss function of the physical information neural network are adjusted with the material and loading path to adapt to different multi-axial fatigue characteristics. The weight adjustment is carried out through the training of the meta-learning model. Under the meta-learning framework, the meta-model learns the geometric features of each loading path, which is used to update the weights of the physical information sub-item of the physical information neural network loss function, so as to better predict the multi-axial fatigue life of steel.

[0065] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] Example 1

[0067] like Figure 1 As shown, this embodiment provides a method for predicting multi-axial fatigue life of steel based on a physical neural network based on meta-learning, which specifically includes:

[0068] S1. Collect the fatigue test data set of the steel to be predicted, and divide the data of the fatigue test data set into a training set and a validation set.

[0069] The fatigue experimental data set of the steel to be predicted can be obtained from published literature or fatigue experiments.

[0070] The fatigue test data set specifically includes multi-axial loading hysteresis loops, real-time normal stress, normal strain, shear stress, shear strain signal values, and corresponding life cycles.

[0071] Steel refers to carbon steel and alloy steel, etc.; fatigue test data is obtained based on experiments on steel specimens of certain shape, size and performance made by pressure processing in the form of ingots, billets, etc.

[0072] S2. Select a critical surface model based on the type of steel to be predicted, and then calculate the parameters required for the critical surface model based on the fatigue test data set, and calculate the damage parameters.

[0073] Generally speaking, for steels with shear fatigue fracture type, the selectable critical surface models include FS model, MGS Yu model, MGSE Zhu model, WB model, Liu model, CXH model, Zhu model, and GSE model; for steels with tensile fatigue fracture type, the selectable critical surface models include SWT model, Liu model, CXH model, Zhu model, and GSE model; for each type, the critical surface model can be a combination of any two or more of the above models.

[0074] In this embodiment, for 316LN, 316L, and 304 steels, the critical surface model selected is:

[0075] FS Model:

[0076] SWT Model:

[0077] MGSEZhu model:

[0078] WB Model:

[0079] CXH(T) Model:

[0080] Among them, σ nmax , τ nmax , Δε nmax , Δγ nmax The plane with the maximum shear strain range is taken as the critical surface, the maximum normal stress, maximum shear stress, maximum normal strain range, maximum normal stress range corresponding to the critical surface, σ smax , τ smax , Δε smax , Δγ smax The plane with the maximum normal strain range is taken as the critical surface, the maximum normal stress, maximum shear stress, maximum normal strain range, maximum normal stress range corresponding to the critical surface, Δε n is the positive strain stroke; σ' f is the fatigue strength coefficient, τ' f is the shear fatigue strength coefficient, ε' f is the fatigue ductility coefficient, γ' f is the shear fatigue ductility coefficient, b is the fatigue strength index, c is the fatigue ductility index, b0 is the shear fatigue strength index, c0 is the shear fatigue ductility index, k is the FS model constant, k MGSE is the MGSEZhu model constant, S is the WB model constant; E is Young's modulus, G is the shear modulus, σ y is the yield strength, N f is the fatigue life.

[0081] In addition, the above five critical surface models can also be applied to other carbon steels except 316LN, 316L, and 304 steels.

[0082] The fatigue parameter τ' on the right side of the above critical surface model equation is f ,γ' f , b0 and c0 are obtained by fitting the torsion data; σ' f , ε' f , b and c are obtained by fitting the uniaxial data. Due to the lack of torsion data for 316L, the Mises equivalent stress criterion is used to obtain:

[0083]

[0084] b o =b

[0085] c o =c

[0086] E, G, σ y Please check the data sheet of the corresponding steel material to obtain it.

[0087] Table 1 lists the parameters required for the critical surface model for 316LN, 316L, and 304 stainless steel:

[0088] Table 1 316LN, 316L, and 304 stainless steel material parameters

[0089]

[0090]

[0091] And calculate the corresponding damage parameters according to the following formula:

[0092]

[0093] Where θ is the angle between the calculation plane and the axis of the steel specimen, ε θ is the stress value on the calculation plane, ε x is the axial normal strain value, γ θ is the calculated plane shear strain value, τ xy is the axial shear stress value, σ θ is the calculated plane normal stress value, σ x is the axial normal stress value.

[0094] The normal strain and shear strain ranges are obtained as follows: Δε θ =ε θmax -ε θmin

[0095] Δγ θ =γ θmax -γ θmin

[0096] And normal stress and shear stress range:

[0097] Δσ θ =σ θmax -σ θmin

[0098] Δτ θ =τ θmax -τ θmin

[0099] Based on the critical surface defined by different critical surface methods, the damage function σ is obtained. nmax , τ nmax , Δε nmax , Δγ nmax , σ smax , τ smax , Δε smax , Δγ smax and Δε n .

[0100] S3. Determine the input features and output features of the physical information neural network, and perform data preprocessing on the input features and output features of the physical information neural network.

[0101] Among them, the input characteristics of the physical information neural network adopt the damage parameters obtained by S2, and the output characteristics of the physical information neural network are the fatigue life prediction results.

[0102] The physical information neural network input features selected in this embodiment are: nmax , τ nmax , Δε nmax , Δγ nmax , σ smax , τ smax , Δε smax and Δγ smax .

[0103] Data preprocessing is performed on the input features and output features of the physical information neural network. Specifically, the input features of the physical information neural network and the logarithmically processed output features of the physical information neural network are normalized:

[0104]

[0105] Among them, x refers to the input characteristics of each physical information neural network or the logarithmic output characteristics of the physical information neural network, x * represents the normalized value, μ represents the mean of the input features of each physical information neural network or the logarithmic output features of the physical information neural network, and σ represents the standard deviation of the input features of each physical information neural network or the logarithmic output features of the physical information neural network.

[0106] S4. Construct a physical information neural network based on the artificial neural network model, define the physical information loss function according to the critical surface model selected in S2, and introduce the physical information item weight.

[0107] σ nmax , τ nmax , Δε nmax , Δγ nmax , σ smax , τ smax , Δε smax and Δγ smax As eight input features, an artificial neural network is constructed. An artificial neural network is a mathematical model in the field of machine learning and cognitive science that mimics the structure and function of biological neural networks (the central nervous system of animals, especially the brain).

[0108] The structure of the artificial neural network in this embodiment is 5 hidden layers, and the number of neurons in each layer is 16, 16, 8, 8, and 4. The output is the predicted life span after logarithmic normalization.

[0109] On this basis, the physical information loss function is defined to construct a physical information neural network, which is expressed as:

[0110] L PINN =MSE+ω1L FS +ω2L SWT +ω3L MGSEZhu +ω4L WB +ω5L CXH(T)

[0111] in:

[0112]

[0113]

[0114] ω1, ω2, ω3, ω4 and ω5 are the physical information weights introduced by the above physical information loss function, i is the sample number, n is the number of samples, N exp,i is the experimental life of the i-th sample, N pre,i is the predicted life span of the i-th sample.

[0115] S5. The normal strain and shear strain information in the fatigue test data set in S1 are used as the input features of the meta-model, and the weights of the physical information items in S4 are used as the output features of the meta-model.

[0116] The input features of the meta-model are: ε and γ, namely normal strain and shear strain.

[0117] Normalize the meta-model input features:

[0118]

[0119] Here, y refers to the meta-model input feature, and y' represents the normalized value.

[0120] The output features of the meta-model are: the weights of the physical information items ω1, ω2, ω3, ω4 and ω5 in S4.

[0121] S6. Build a meta-model using the gated recurrent unit as the meta-model module and define the meta-model loss function.

[0122] Using the gated recurrent unit as the module of the meta-model, a total of three identical modules are constructed, with a total of 32 hidden layers. The first gated recurrent unit is pre-connected to the input features, and the last gated recurrent unit (GRU) is followed by a fully connected layer, which outputs the weights required by the physical neural network.

[0123] GRU mainly includes update gate, reset gate, candidate hidden state, and hidden state.

[0124] The update gate is used to determine the proportion of information from the current time step that is passed to the next time step:

[0125] z t =σ(W z ·[h t-1 ,x t ])

[0126] Among them, z t is the output of the update gate, σ is the Sigmoid activation function, W z is the weight matrix, h t-1 is the hidden state of the previous time step, x t is the input for the current time step.

[0127] The reset gate determines the forgetting ratio of the previous time step information:

[0128] r t =σ(W r ·[h t-1 ,x t ])

[0129] Among them, r t is the output of the reset gate, σ is the Sigmoid activation function, W r is the weight matrix, h t-1 is the hidden state of the previous time step, x t is the input for the current time step.

[0130] Under the action of the reset gate, the candidate hidden state of the current time step is calculated, and under the action of the update gate, the weighted sum with the previous hidden state is calculated to obtain the final hidden state.

[0131] The meta-model loss function is defined as:

[0132]

[0133] Among them, i is the sample number of the physical information neural network, n is the number of samples of the physical information neural network, N exp,i is the experimental life of the i-th sample, N pre,i is the predicted life span of the i-th sample.

[0134] S7, randomly divide the dataset of the metamodel input features, the physical information neural network input features, and the physical information neural network output features into a training set and a test set, and ensure that the sample division of the metamodel and the physical information neural network is consistent, that is, the data of the same sample in the metamodel and the physical information neural network are simultaneously divided into the training set or the test set.

[0135] Initialize the metamodel and the physical information neural network, initialize the physical information term weight, and set the metamodel training period and the physical information neural network training period.

[0136] After creating the database for the physical neural network input features and the physical neural network output features in S3 and the metamodel input features and the metamodel output features in S5, a random seed is assigned, and the data corresponding to the samples in the database are randomly divided into a training set and a test set.

[0137] In this embodiment, 80% of the data is used for the training set, and 20% is used for the validation set.

[0138] The metamodel and the physical information network hyperparameters and the physical information term weight are randomly assigned initially. The number of iterations of the metamodel is set to 50, and the number of periods in a single training of the physical neural network is set to 1500.

[0139] S8, according to the current physical information term weight, the physical information neural network is trained, the physical information neural network output features are updated, the physical information loss function is calculated and back-propagated to the physical information neural network.

[0140] According to the physical information term weight, the physical information neural network is trained once every 1500 periods. In each period, the fatigue life is predicted, and the corresponding physical information loss function is calculated. The physical information loss function value is back-propagated to update the hyperparameters for fatigue life prediction in the next period.

[0141] S9, in the last period of a single training of the physical neural network, the MSE is calculated and back-propagated to the metamodel as the loss function value.

[0142] In the 1500th period of a single training of the physical information neural network, the root mean square error MSE of the test set is calculated and back-propagated to the metamodel to update the hyperparameters.

[0143] S10, the metamodel is trained once, and the physical information term weight is updated.

[0144] After the metamodel updates the hyperparameters, it is trained once to update a set of physical information neural network loss function physical information term required weights.

[0145] S11, repeating the steps S8-S10 until a set meta-model training period, obtaining a fatigue life prediction result by the physical information neural network.

[0146] Figure 2 The specific process of meta-learning in the physical information neural network steel fatigue life prediction method is shown.

[0147] Figure 3 The ablation experiment performance of the method proposed in the present application on three metal materials of 316LN, 316L and 304 is shown. Figure 3 It can be seen that the gated recurrent unit-physical information neural network performs better than the artificial neural network and the physical information neural network with unoptimized weights (weight is 1) in the repeated experiment of the multi-axial fatigue data set of the three metal materials.

[0148] Figure 4 The comparison between the predicted value and the experimental value of the multi-axial fatigue life of the method proposed in the present application in the data set of three metal materials of 316LN, 316L and 304 is shown. Figure 4 It can be seen that the predicted value of the gated recurrent unit-physical information neural network on the test set and the training set is within the 2 times error band, and most of them are within the 1.5 times error band.

[0149] Embodiment 2

[0150] A computer device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the computer program to implement the meta-learning-based physical information neural network steel fatigue life prediction method in embodiment 1.

[0151] Embodiment 3

[0152] A computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the meta-learning-based physical information neural network steel fatigue life prediction method in embodiment 1.

[0153] The technical features of the above embodiments can be combined in any way, and to make the description concise, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combinations of the technical features do not exist contradictory, they should be considered as the scope of the present application.

[0154] The principles and implementation modes of the present application are described by specific examples in this paper, and the above description of the examples in the specification is only to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In view of the above, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A physical information neural network steel fatigue life prediction method based on meta-learning, characterized in that: The method includes: S1. Collect fatigue test data set of steel to be predicted; S2. After selecting a critical surface model based on the steel to be predicted, calculate the parameters required for the critical surface model based on the fatigue test data set and calculate the damage parameters; Among them, the parameters required for the critical surface model are , , , , b, c, b0, c0, k, k MGSE and S; Wherein, the damage parameter is: , , , , , , , and ; S3. Determine the input characteristics and output characteristics of the physical information neural network, and perform data preprocessing on the input characteristics and output characteristics of the physical information neural network; wherein the input characteristics of the physical information neural network use the damage parameters obtained in S2, and the output characteristics of the physical information neural network are fatigue life prediction results; S4, constructing a physical information neural network based on the artificial neural network model, and defining the physical information loss function according to the critical surface model selected in S2, while introducing the physical information item weight; The physical information loss function is: ; in: ; ; ; ; ; ; , , , and is the weight of the physical information item, i is the sample number, n is the number of samples, is the experimental life of the i-th sample, is the predicted lifespan of the sample numbered i; in, , , , The order is the maximum normal stress, maximum shear stress, maximum normal strain range, and maximum shear strain range corresponding to the critical surface when the plane where the maximum normal strain range is located is taken as the critical surface; , , , The order is the maximum normal stress, maximum shear stress, maximum normal strain range, and maximum shear strain range corresponding to the critical surface when the plane where the maximum shear strain range is located is taken as the critical surface; is the positive strain stroke; is the fatigue strength coefficient, is the shear fatigue strength coefficient, is the fatigue ductility coefficient, is the shear fatigue ductility coefficient, b is the fatigue strength index, c is the fatigue ductility index, b0 is the shear fatigue strength index, c0 is the shear fatigue ductility index, k is the FS model constant, k MGSE is the MGSEZhu model constant, S is the WB model constant; E is Young's modulus, G is the shear modulus, is the yield strength, is fatigue life; v e is the Poisson's ratio in the elastic stage; v p is the Poisson's ratio in the plastic stage; S5, using the normal strain and shear strain information in the fatigue test data set obtained in S1 as the meta-model input features, using the physical information item weights in S4 as the meta-model output features, and performing data preprocessing on the meta-model input features; S6. Construct a meta-model and define the mean square error (MSE) of the physical information neural network as the meta-model loss function; S7. Randomly divide the data sets of metamodel input features, physical information neural network input features, and physical information neural network output features into a training set and a test set, and ensure that the data of the same samples of the metamodel and the physical information neural network are divided into the training set or the test set at the same time; initialize the metamodel and the physical information neural network, initialize the weights of the physical information items, and set the metamodel training cycle and the physical information neural network training cycle; S8. Perform physical information neural network training based on the current physical information item weights, update the output features of the physical information neural network, calculate the physical information loss function, and back-propagate it to the physical information neural network; S9. In the last cycle of physical neural network training, the mean square error (MSE) of the physical information neural network is calculated and back-propagated to the meta-model. S10, perform meta-model training and update the weights of physical information items; S11. Repeat steps S8-S10 until the set meta-model training cycle is reached, and the fatigue life prediction result is obtained through the physical information neural network.

2. The method for predicting steel fatigue life based on a physical information neural network using meta-learning according to claim 1, characterized in that: In S2, for steel with shear fatigue fracture type, the critical surface model adopts at least two of the FS model, MGS Yu model, MGSE Zhu model, WB model, Liu model, CXH model, Zhu model or GSE model; for steel with tensile fatigue fracture type, the critical surface model adopts at least two of the SWT model, Liu model, CXH model, Zhu model and GSE model.

3. The method for predicting steel fatigue life based on a physical information neural network using meta-learning according to claim 2, characterized in that: In S2, for carbon steel, the critical surface model adopts FS model, SWT model, MGSEZhu model, WB model and CXH(T) model: FS Model: ; SWT Model: ; MGSEZhu model: ; WB Model: ; CXH(T) Model: ; in, , , , The order is the maximum normal stress, maximum shear stress, maximum normal strain range, and maximum shear strain range corresponding to the critical surface when the plane where the maximum normal strain range is located is taken as the critical surface; , , , The order is the maximum normal stress, maximum shear stress, maximum normal strain range, and maximum shear strain range corresponding to the critical surface when the plane where the maximum shear strain range is located is taken as the critical surface; is the positive strain stroke; is the fatigue strength coefficient, is the shear fatigue strength coefficient, is the fatigue ductility coefficient, is the shear fatigue ductility coefficient, b is the fatigue strength index, c is the fatigue ductility index, b0 is the shear fatigue strength index, c0 is the shear fatigue ductility index, k is the FS model constant, k MGSE is the MGSEZhu model constant, S is the WB model constant; E is Young's modulus, G is the shear modulus, is the yield strength, is fatigue life; v e is the Poisson's ratio in the elastic stage; v p is the Poisson's ratio in the plastic stage.

4. The method for predicting steel fatigue life based on a physical information neural network using meta-learning according to claim 3, characterized in that: In S3, the physical information neural network input feature is , , , , , , , .

5. The method for predicting steel fatigue life based on a physical information neural network using meta-learning according to claim 1, characterized in that: In S3, data preprocessing of the physical information neural network input features and the physical information neural network output features includes: Normalize the physical information neural network input features and the logarithmized physical information neural network output features: ; Among them, x refers to the input characteristics of each physical information neural network or the logarithmic output characteristics of the physical information neural network, represents the normalized value, Represents the mean of each physical information neural network input feature or the logarithmic physical information neural network output feature, Represents the standard deviation of each physical information neural network input feature or the logarithmic physical information neural network output feature.

6. The method for predicting steel fatigue life based on a physical information neural network using meta-learning according to claim 1, characterized in that: In S4, v e Take 0.3; v p Take 0.

5.

7. The method for predicting steel fatigue life based on a physical information neural network using meta-learning according to claim 1, characterized in that: In S5, data preprocessing of the metamodel input features includes normalizing the metamodel input features: ; in, Refers to the meta-model input features, Indicates the normalized value.

8. The method for predicting steel fatigue life based on a physical information neural network using meta-learning according to claim 1, characterized in that: In S9, the root mean square error (MSE) of the physical information neural network is: ; Among them, i is the sample number, n is the number of samples, is the experimental life of the i-th sample, is the predicted life span of the i-th sample.

9. A computer device comprising: 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 meta-learning-based physical information neural network steel fatigue life prediction method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting steel fatigue life based on a physical information neural network using meta-learning as described in any one of claims 1 to 7 is implemented.

Images