Composite material microscale burn-off erosion morphology prediction method, device, equipment and storage medium

Abstract

The application provides a composite material mesoscopic burning-ablation morphology prediction method, device, equipment and storage medium, and is used for the technical field of composite material mesoscopic morphology evolution prediction. The method comprises the following steps: acquiring a data set; determining initial conditions, boundary conditions and ablation criteria of a composite material mesoscopic burning-ablation model; constructing a physical information neural network model based on the evolution of the mesoscopic burning-ablation surface morphology; inputting the data set into the physical information neural network model for iterative training to obtain a trained physical information neural network model; and inputting a composite material to be detected into the trained physical information neural network model to obtain a prediction result. The method can effectively approximate the mesoscopic morphology evolution in the composite material burning-ablation process, greatly improves the speed and accuracy of solving the composite material mesoscopic burning-ablation morphology evolution, and has great application prospect.

Description

Technical Field

[0001] This application relates to the field of microstructure evolution prediction technology for composite materials, and in particular to a method, apparatus, equipment, and storage medium for rapid prediction of microstructure burn-erosion morphology evolution of composite materials based on physical information neural networks. Background Technology

[0002] Carbon-carbon composites are widely used in aerospace, especially in thermal protection systems operating under high-temperature, high-speed airflow conditions. However, in extreme environments, C / C composites undergo ablation, leading to surface degradation, crack propagation, and accelerated ablation rates, directly impacting material performance and lifespan. Traditional methods for solving ablation morphology evolution, such as numerical simulations, typically rely on detailed calculations based on physical equations (e.g., heat conduction, fluid dynamics, chemical reactions), resulting in complex, time-consuming calculations with high computational resource requirements, failing to meet the real-time and efficiency demands of engineering practice. Furthermore, the evolution of ablation morphology at the microscale is influenced by various factors, including material properties, flow field dynamics, and chemical reactions, often rendering traditional methods incapable of providing accurate and efficient solutions at the microscale.

[0003] Therefore, there is an urgent need for a new method for rapid prediction of the microscopic burn-erosion morphology evolution of composite materials to solve the problems encountered in the existing technology. Summary of the Invention

[0004] This application provides a method, apparatus, device, and storage medium for rapid prediction of the microstructure evolution of composite materials during the burn-erosion process based on a physical information neural network, which can be used to quickly and efficiently predict the microstructure evolution of composite materials during the burn-erosion process.

[0005] According to the first aspect disclosed in this application, this application provides a method for predicting the microstructure of scorch-erosion morphology of composite materials, including:

[0006] Obtain the dataset; wherein the dataset includes the micro-ablation morphology of the carbon-carbon composite surface obtained by processing the micro-ablation model of the constructed carbon-carbon composite material based on the fixed volume fraction method.

[0007] Determine the initial conditions, boundary conditions, and erosion criteria for the microscopic burn-erosion model of composite materials;

[0008] Construct a physical information neural network model based on the evolution of microscopic sintered-eroded surface morphology of composite materials;

[0009] The dataset is input into the physical information neural network model for iterative training to obtain the trained physical information neural network model.

[0010] The composite material to be tested is input into the trained physical information neural network model to obtain the prediction results of the microscopic burn-erosion morphology evolution of the composite material.

[0011] The beneficial effects of adopting the above technical solution are as follows: This application mainly combines deep learning technology and physical information neural network technology to solve the thermochemical ablation equation. Under the premise of ensuring physical consistency, the neural network can be trained to effectively approximate the microscopic morphology evolution of composite materials during the ablation-exfoliation process, which greatly improves the speed and accuracy of solving the microscopic ablation-exfoliation morphology evolution of composite materials. The technology of this application can be widely used in the aerospace field, especially in applications that require accurate prediction of ablation and exfoliation processes and their impact on structural performance, such as aircraft outer surfaces, engine nozzles, and thermal protection systems.

[0012] In one feasible implementation, obtaining the dataset includes:

[0013] Construct a microscopic geometric model of the composite material;

[0014] Based on the aforementioned microscopic geometric model, a composite material ablation model is constructed;

[0015] Based on the solid volume fraction method, the ablation retreat calculation is performed on each unit in the geometric model according to the microscopic ablation model of the composite material to obtain the solid volume fraction dataset of each unit, so as to constitute the microscopic ablation-exfoliation morphology of the composite material surface.

[0016] In one feasible implementation, the microscopic geometric model for constructing the composite material includes:

[0017] A microscopic geometric model of carbon-carbon composite materials, including carbon fibers and a carbon matrix, was established based on TexGen, and the microscopic geometric model was transformed into a discrete model of cubic elements.

[0018] The carbon fibers and carbon matrix are endowed with material properties including fiber volume fraction, density, thermal conductivity, porosity, and specific heat capacity.

[0019] In one feasible implementation, constructing the composite material ablation model based on the microscopic geometric model includes:

[0020] Construct the spatial surface function of the composite material in three-dimensional coordinates; when the composite material is ablated and retreats, each point on the surface retreats a certain distance along the normal vector direction within a certain time interval to form a new surface function after ablation; and the gas diffusion equation of the carbon matrix of the composite material during ablation and retreat.

[0021] The spatial surface function of the composite material in three-dimensional coordinates is:

[0022] As the composite material ablates and recedes, each point on the surface recedes a certain distance along the normal vector direction within a certain time interval, forming a new surface function after ablation.

[0023]

[0024]

[0025] in, It is a material surface function, present at the gas-solid or liquid-solid interface. ; These are spatial coordinates; It is time; It is the reaction concentration; It refers to the speed at which the interface goes back; k It is a heterogeneous reaction rate; It is the molar volume of the solid phase; It is the outward normal of the solid phase interface; It is a gradient operator; It is the magnitude of the vector;

[0026] The gas diffusion equation for the carbon matrix of the composite material during ablation retreat is as follows:

[0027]

[0028] in, It is the reaction concentration; It is time; It is the diffusion rate; It is the gradient operator.

[0029] In one feasible implementation, the solid volume fraction-based method involves performing ablation regression calculations on each element of the geometric model according to a microscopic ablation model of the composite material to obtain a dataset of the solid volume fraction for each element, including:

[0030] Define the solid volume fraction of each unit as:

[0031]

[0032] The oxygen diffusion equation and the new surface function after ablation, rewritten using the solid volume fraction, are as follows:

[0033]

[0034]

[0035] in, It is the volume of the solid in the unit cell; It is the unit volume; Indicates a gas phase unit. Represents a solid-state unit; 0 < <1 represents a hybrid unit; It is the reaction concentration; It is time; It is the diffusion coefficient; It is the outward normal of the solid phase interface; k It is a heterogeneous reaction rate; It is a gradient operator; It is the absolute value of the screen back speed;

[0036] The finite volume method was used to solve the oxygen diffusion equation and the new surface function after ablation, and a dataset of solid volume fractions was obtained.

[0037] In one feasible implementation, determining the initial conditions, boundary conditions, and erosion criteria for the composite material microstructure burn-erosion model includes:

[0038] The initial conditions for defining the computational domain include the temperature field, gas concentration, and material morphology; the temperature field is defined as the initial temperature of the composite material and the environment. The gas concentration is set to the ambient gas concentration. The material morphology is defined as the initial surface shape and microstructure distribution of the material. ;

[0039] The boundary conditions for the computational domain include a top boundary and lateral boundaries; the top boundary uses Dirichlet conditions. (in (This represents the gas concentration in the initial state); the lateral boundary is symmetrical, meaning the normal gradient between temperature and gas concentration is 0.

[0040] The erosion criterion is a mechanical erosion criterion, specifically:

[0041]

[0042]

[0043]

[0044]

[0045] in, It is the first in the composite material model The height of the floor, ; It is the maximum height of the surface yarn in the composite material model; It is the average height of the surface matrix in the composite material model; It is the mechanical erosion coefficient related to materials and the environment; It is the width of the weft yarn in the composite material model; It refers to the number of warp yarns covering the weft yarns; The length of the weft yarn; It is the first The height of the weft yarn; warp is the warp yarn in the composite material model; welf is the weft yarn in the composite material model; if n is warp, then s is welf; if n is welf, then s is warp;

[0046] Its mechanical erosion at the microscale is as follows:

[0047]

[0048] in, It is the average height of the surface substrate; It is the height of the surface yarn in the composite material model; It is the amount of mechanical erosion.

[0049] In one feasible implementation, the construction of a physical information neural network model based on the evolution of the microscopic burn-eradication surface morphology of composite materials includes:

[0050] Based on the determined initial conditions, boundary conditions, and erosion criteria of the composite material microstructure burn-erosion model, residual points, boundary constraint points, and initial constraint points for training are selected within the computational domain.

[0051] Construct a custom multilayer perceptron neural network structure;

[0052] A neural network is constructed based on a custom improved multilayer perceptron structure, and training points are input for training to establish a physical information neural network model.

[0053] The loss function is obtained from the physical information neural network model, and the model is trained to obtain the final physical information neural network model based on the evolution of the microscopic burn-eradication surface morphology of composite materials.

[0054] In one feasible implementation, selecting residual points, boundary constraint points, and initial constraint points for training within the computational domain includes:

[0055] Latin hypercube random sampling is used to select residual points in the computational domain of composite materials;

[0056] Set boundary constraint points at the top boundary of the composite material, and set boundary constraint points at the lateral boundary using symmetry conditions;

[0057] Initial constraint points for temperature and gas concentration are set on the surface and inside the composite material.

[0058] In one feasible implementation, the construction of a custom multilayer perceptron neural network structure includes: an input layer, an encoder layer, a fusion layer, a hidden layer, an automatic micro-layer, and an output layer.

[0059] The encoder layer includes an ablation encoder and an ablation encoder; the ablation encoder is used to perform feature encoding processing on the input data of the composite material ablation process and output an ablation process feature vector; the ablation encoder is used to perform feature encoding processing on the input data of the composite material ablation process and output an ablation process feature vector.

[0060] The input data for the composite material ablation process includes the surface temperature and gas concentration during the composite material ablation process;

[0061] The input data for the composite material ablation process includes airflow velocity and surface morphology during the ablation process.

[0062] In one feasible implementation, obtaining the loss function based on the physical information neural network model includes:

[0063] Physical equation constraints are introduced into the physical information neural network model. The physical constraint loss, boundary condition loss, initial condition loss and mechanical erosion condition loss are calculated using automatic differentiation technology. The final loss function is obtained by weighted summation.

[0064] In one feasible implementation, the physical constraint loss, boundary condition loss, initial condition loss, and mechanical erosion condition loss take the following form:

[0065] The physical constraint loss, i.e., the loss function of the gas diffusion equation, is:

[0066]

[0067] in, It is the reaction concentration; It is time; It is the diffusion coefficient; Indicates a gas phase unit. Represents a solid-state unit; 0 < <1 represents a hybrid unit; It is a gradient operator; It is the physical constraint loss function;

[0068] The loss function for the boundary conditions and initial conditions is:

[0069]

[0070]

[0071] in, It is the reaction concentration; It represents the gas concentration in the initial state. It refers to the concentration of ambient gases; These are weighting coefficients used to balance the relative importance of different losses; It is the temperature of the composite material; It is the initial temperature of the environment; It is the boundary condition loss function; It is the initial condition loss function;

[0072] The loss function for the mechanical erosion condition is:

[0073]

[0074] in, It is the average height of the surface substrate; It is the height of the surface yarn in the composite material model; It is the amount of mechanical erosion. It is the mechanical erosion constraint loss function;

[0075] The final loss function is a weighted sum of the physical constraint loss, boundary condition loss, initial condition loss, and mechanical erosion constraint loss:

[0076] .

[0077] in, These are weighting coefficients used to balance the relative importance of different sub-loss functions; This is the final loss function; according to the second aspect disclosed in this application, this application provides a rapid prediction device for the evolution of microscopic burn-eradication morphology of composite materials based on a physical information neural network, comprising:

[0078] The data acquisition module is used to acquire a dataset; wherein, the dataset includes the micro-ablation morphology of the carbon-carbon composite surface obtained by processing the micro-ablation model of the constructed carbon-carbon composite material based on the fixed volume fraction method.

[0079] The model condition verification module is used to determine the initial conditions, boundary conditions, and erosion criteria of the composite material microstructure burn-erosion model.

[0080] The model building module is used to construct a physical information neural network model based on the evolution of the microscopic sintered-eroded surface morphology of composite materials.

[0081] The model training module is used to input the dataset into the physical information neural network model for iterative training to obtain the trained physical information neural network model.

[0082] The rapid prediction module for the evolution of microscopic sintering-erosion morphology is used to input the composite material to be tested into the trained physical information neural network model to obtain the prediction results of the microscopic sintering-erosion morphology evolution of the composite material.

[0083] According to a third aspect disclosed in this application, an electronic device is provided, including a processor and a memory communicatively connected to the processor;

[0084] The memory stores computer-executed instructions;

[0085] The processor executes computer execution instructions stored in the memory to implement the method described in any one of the first aspects.

[0086] According to a fourth aspect disclosed in this application, a computer-readable storage medium is provided, wherein computer-executable instructions are stored therein, which, when executed by a processor, are used to implement the method described in any one of the first aspects.

[0087] According to the fifth aspect disclosed in this application, a computer program product is provided, comprising a computer program that, when executed by a processor, is used to implement the method described in any one of the first aspects.

[0088] Compared with the prior art, this application has the following beneficial effects:

[0089] (1) The fast prediction method for the microscopic burn-eradication morphology evolution of composite materials based on physical information neural network provided in this application solves the problems of traditional finite volume method in mesh generation, dimensionality curse of difference scheme, multi-scale processing, etc. It has significant advantages in computational efficiency and flexibility. Especially in the fields of complex boundary conditions, high-dimensional problems and data and physical coupling solutions, PINN (physical information neural network) provides a brand-new solution for the fast simulation and prediction of burn-eradication problems.

[0090] (2) Compared with the fully connected neural network structure of the baseline PINN model, the custom multilayer perceptron structure proposed in this application has significant advantages in terms of expressive power, physical constraint handling ability and computational efficiency through specific optimization for ablation and erosion problems. It can also better handle complex boundaries, multi-scale problems and generalization ability, thereby improving the accuracy and training stability of the model.

[0091] (3) The PINN method proposed in this application can directly predict the morphological evolution of the entire composite material micro-burning-erosion process from the physical equation through the neural network. It does not require layer-by-layer calculation and surface updating. The network can solve the entire process at once and directly give the overall burning-erosion morphology under time evolution, which greatly improves the computational efficiency and reduces the time cost of layer-by-layer calculation. Attached Figure Description

[0092] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0093] Figure 1 A flowchart illustrating the rapid prediction method for the microscopic burn-eradication morphology evolution of composite materials provided in this application embodiment;

[0094] Figure 2 A schematic diagram of the micro-geometric model of the composite material provided in the embodiments of this application;

[0095] Figure 3 A schematic diagram illustrating the application of boundary conditions to the composite material geometry model provided in this embodiment;

[0096] Figure 4 A schematic diagram of the structure of a custom improved multilayer perceptron provided in the embodiments of this application.

[0097] Figure 5 A schematic diagram of the structure of the rapid prediction device for the microscopic burn-eradication morphology evolution of composite materials provided in this application embodiment;

[0098] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0099] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0100] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0101] To improve the speed and accuracy of solving the scorching-erosion morphology evolution of composite materials, this application proposes a fast prediction method for the microscopic scorching-erosion morphology evolution based on Physical Information Neural Network (PINN). This application combines deep learning technology and physical law constraints, and trains the neural network to effectively approximate the microscopic morphology evolution in the complex scorching-erosion process of composite materials, providing an efficient and accurate solution prediction method.

[0102] The technical solution of the rapid prediction method for the microscopic burn-eradication morphology evolution of composite materials based on physical information neural networks provided in this application will be described in detail below through specific embodiments. It should be noted that the following embodiments may exist alone or in combination with each other, and the same or similar content may not be described again in different embodiments.

[0103] Figure 1 A flowchart illustrating a rapid prediction method for the microscopic burn-eradication morphology evolution of composite materials based on a physical information neural network, as provided in this application embodiment, is shown below. Figure 1 In some embodiments, the rapid prediction method for the microscopic burn-eradication morphology evolution of the composite material includes the following steps:

[0104] S101, Obtain the dataset; wherein, the dataset includes the micro-ablation morphology of the carbon-carbon composite surface obtained by processing the micro-ablation model of the constructed carbon-carbon composite material based on the fixed volume fraction method.

[0105] S102, determine the initial conditions, boundary conditions and erosion criteria of the composite material microstructure burn-erosion model;

[0106] S103, Construct a physical information neural network model based on the evolution of the microscopic sintered-eroded surface morphology of composite materials;

[0107] S104, Input the dataset into the physical information neural network model for iterative training to obtain the trained physical information neural network model;

[0108] S105, input the composite material to be tested into the trained physical information neural network model to obtain the prediction results of the microscopic burn-erosion morphology evolution of the composite material.

[0109] In step S101, the specific steps for obtaining the dataset are as follows:

[0110] A mesoscopic geometric model of the composite material is constructed. In this embodiment, a mesoscopic geometric model of a C / C composite material is constructed, comprising two parts: carbon fibers and a carbon matrix. First, a mesoscopic geometric model is established using TexGen, and then the mesoscopic geometric model is transformed into a discrete model of cubic elements, such as... Figure 2As shown, the carbon fibers and the matrix are then given material properties such as fiber volume fraction, density, thermal conductivity, porosity, and specific heat capacity.

[0111] Based on the above microscopic geometric model, a composite material ablation model is constructed. In this embodiment, a spatial surface function of the carbon-carbon composite material in three-dimensional coordinates is constructed. When the composite material ablates and retreats, each point on the surface retreats a certain distance along the normal vector direction within a certain time interval to form a new surface function after ablation. The gas diffusion equation of the carbon matrix of the composite material during ablation and retreat is also provided. Specifically:

[0112] Thermochemical ablation of carbonized composite materials mainly includes carbon oxidation, nitridation, and sublimation reactions. Thermochemical ablation primarily occurs in carbonized composite materials due to oxidation. In this embodiment, only oxygen diffusion and thermochemical reactions at the ablated surface of the C / C composite material are considered to establish a microscopic ablation model. During thermochemical ablation, heat and mass transfer from carbon oxidation leads to the decay of the gas-solid interface of the C / C composite material. This decay alters the distribution of oxygen concentration in the boundary layer, which in turn affects the carbon reaction rate and the gas-solid interface decay rate. Therefore, the carbon reaction rate and the boundary layer oxygen concentration distribution are crucial parameters for establishing an oxidation kinetic model characterizing the surface decay of the C / C composite material. First, a spatial surface function of the C / C composite material in three-dimensional coordinates is constructed to characterize the gas / solid interface. When the composite material regresses during ablation, each point on the surface regresses a certain distance along the normal direction within a time interval, forming a new surface function after ablation.

[0113] The spatial surface function of the composite material in three-dimensional coordinates is:

[0114]

[0115] As the composite material ablates and recedes, each point on the surface recedes a certain distance along the normal vector direction within a certain time interval, forming a new surface function after ablation.

[0116]

[0117]

[0118]

[0119] in, It is a material surface function, present at the gas-solid or liquid-solid interface. ; These are spatial coordinates; It is time; It is the reaction concentration; It refers to the speed at which the interface goes back;k It is a heterogeneous reaction rate; It is the molar volume of the solid phase; It is the outward normal of the solid phase interface; It is a gradient operator; It is the magnitude of the vector;

[0120] When the carbon matrix of the C / C composite material ablates and retreats, according to the law of conservation of mass, the gas diffusion equation is:

[0121]

[0122] in, It is the reaction concentration; It is time; It is the diffusion rate; It is a gradient operator;

[0123] Based on the solid volume fraction method, ablation regression calculations are performed on each element in the discrete model according to the microscopic ablation model of the composite material to obtain the solid volume fraction dataset for each element, thus constructing the microscopic ablation-exfoliation morphology of the composite material surface; specifically including:

[0124] Based on the solid volume fraction method, the solid volume fraction of each unit in the discrete model is defined as:

[0125]

[0126] in, It is the volume of the solid in the unit cell. It is the unit volume;

[0127] Introducing the solid volume fraction, the above oxygen diffusion equation and the new surface function after ablation are rewritten as follows:

[0128]

[0129]

[0130] in, It is the volume of the solid in the unit cell; It is the unit volume; Indicates a gas phase unit. Represents a solid-state unit; 0 < <1 represents a hybrid unit; It is the reaction concentration; It is time; It is the diffusion coefficient; It is the outward normal of the solid phase interface; k It is a heterogeneous reaction rate; It is a gradient operator; It is the absolute value of the screen back speed;

[0131] The finite volume method was used to solve the oxygen diffusion equation and the new surface function after ablation, and a dataset of solid volume fractions was obtained.

[0132] In step S102, the initial conditions, boundary conditions, and erosion criteria of the composite material microstructure burn-erosion model are determined. The specific steps are as follows:

[0133] The initial conditions for defining the computational domain include the temperature field, gas concentration, and material morphology; the temperature field is defined as the initial temperature of the composite material and the environment. The gas concentration is set to the ambient gas concentration. The material morphology is defined as the initial surface shape and microstructure distribution of the material. ;

[0134] Establish a unit cell model, including warp yarns, yarns, and the matrix, and set the boundary conditions of the computational domain, including the top boundary and lateral boundaries, such as... Figure 2 As shown; the top boundary adopts the Dirichlet condition. (in (This represents the gas concentration in the initial state); the lateral boundary is symmetrical, meaning the normal gradient between temperature and gas concentration is 0.

[0135] Specifically, to accurately simulate the ablation morphology of C / C composite materials, the initial conditions are crucial. These conditions mainly include the temperature field, gas concentration, and material morphology. These conditions determine the initial state of the simulation, thus affecting the predicted ablation behavior. The temperature field refers to the initial temperature distribution inside and on the surface of the material, usually set to the ambient temperature. Material morphology refers to the surface shape and microstructural characteristics of a material, i.e. Gas concentration refers to the initial concentration of ablation reaction products, usually set as the ambient concentration, i.e. ;in Indicates the initial temperature of the material and the environment. Indicates the concentration of ambient gases. This represents the initial surface height or topography distribution; the boundary conditions for the model domain mainly include the top boundary and the lateral boundaries. The top boundary uses the Dirichlet condition, i.e. The lateral boundaries are assumed to be symmetrical, i.e., the normal gradients of temperature and gas concentration are zero. During thermochemical ablation, when most of the fiber height exceeds the carbon matrix surface, the constraint force of the carbon matrix on the exposed fibers gradually decreases. This is considered by taking into account the mass loss and morphological changes caused by non-thermochemical ablation, such as mechanical ablation caused by airflow shear and pneumatic blowing. Therefore, in a high-speed airflow environment, the airflow shear force on the ablated parallel fibers far exceeds the critical force required for mechanical ablation. After mechanical ablation of the parallel yarns, the remaining material will continue thermochemical ablation until the matrix surface is again lower than the yarn surface height. During this process, thermochemical ablation and mechanical ablation will alternate until all C / C composite materials are completely consumed. In this cycle, mechanical ablation will change the conditions of subsequent thermochemical ablation, such as the position of the gas / solid reaction interface and the distribution of gas concentration. At the same time, mechanical ablation will also affect the ablation performance of C / C composite materials, such as surface degradation and changes in surface roughness. Based on the mechanical ablation behavior of parallel yarns, the ablation criteria for C / C composite materials are as follows:

[0136]

[0137]

[0138]

[0139]

[0140] in, It is the first in the composite material model The height of the floor, ; It is the maximum height of the surface yarn in the composite material model; It is the average height of the surface matrix in the composite material model; It is the mechanical erosion coefficient related to materials and the environment; It is the width of the weft yarn in the composite material model; It refers to the number of warp yarns covering the weft yarns; The length of the weft yarn; It is the first The height of the weft yarn; warp is the warp yarn in the composite material model; welf is the weft yarn in the composite material model; if n is warp, then s is welf; if n is welf, then s is warp;

[0141] When the mechanical erosion criterion in the above formula is met, the height of the exposed yarn is greater than... Parts will be peeled off; therefore, the amount of mechanical erosion at the microscale is:

[0142]

[0143] in, It is the average height of the surface substrate; It is the height of the surface yarn in the composite material model; It is the amount of mechanical erosion.

[0144] In step S103, a physical information neural network model based on the evolution of the microscopic sintering-erosion surface morphology of composite materials is constructed. The specific steps are as follows:

[0145] Based on the defined initial conditions, boundary conditions, and erosion criteria of the composite material microstructure burn-erosion model, residual points, boundary constraint points, and initial constraint points for training are selected within the computational domain; specifically:

[0146] To construct a physical information neural network that conforms to the evolution of ablation-ablation surface morphology, it is necessary to rationally select residual points, boundary constraint points, and initial constraint points within the computational domain. These points are crucial for training the neural network, ensuring that the model can effectively satisfy physical constraints and provide accurate predictions when solving the thermochemical ablation process. Residual points are used to minimize the residuals of the physical equations during neural network optimization. For the thermochemical ablation simulation of C / C composite materials, in this embodiment, residual points are distributed in the internal region and near the surface of the material, especially in the area where the airflow contacts the material surface, and a Latin hyperparameter is used. Square sampling involves random sampling within the computational domain to ensure better coverage. Boundary constraint points are used to apply physical boundary conditions, ensuring that the solution of the neural network is consistent with the boundary conditions of the problem. For the thermochemical ablation simulation of C / C composite materials, this embodiment sets boundary constraint points on the material surface (top boundary), specifies the surface temperature as the ambient temperature, and applies symmetry conditions on the lateral boundaries. Initial constraint points are used to specify the initial conditions of the simulation, ensuring that the neural network can correctly reflect the initial state of the material when training begins. For the thermochemical ablation simulation of C / C composite materials, this embodiment sets initial conditions for temperature and gas concentration on the surface and inside.

[0147] Construct a custom multilayer perceptron neural network structure, and a custom improved multilayer perceptron neural network structure, such as... Figure 4 As shown; a neural network is constructed based on a custom-designed improved multilayer perceptron structure, and trained using input training points to establish a physical information neural network model; specifically:

[0148] To construct a physical information neural network for the evolution of ablation-exfoliation surface morphology, a basic multilayer perceptron is first defined. This network, primarily through weighted connections and nonlinear activations across multiple layers, effectively learns complex patterns and relationships. Its main advantages are its simple structure, ease of implementation, and strong expressive power. Based on this framework, to enable the neural network to simultaneously learn and process physical information and features related to both ablation and exfoliation processes, this embodiment adds an ablation encoder and an exfoliation encoder to the standard fully connected neural network structure. These encoders extract and transform important features related to ablation and exfoliation by specifically encoding or mapping the original input. This custom neural network structure can handle complex nonlinear relationships and adapt to high-dimensional, dynamically changing ablation and exfoliation processes. First, the input layer includes information such as time and spatial location (coordinates). Second, an ablation encoder and an exfoliation encoder are added. The purpose of the ablation encoder is to encode ablation-related input data, such as gas concentration and surface temperature, as features. The output will be a high-dimensional feature vector containing important physical information during the ablation process, such as linear ablation decay rate and mass ablation rate. This high-dimensional feature vector is denoted as... The purpose of the ablation encoder is to extract input data related to the ablation process on the material surface, such as airflow velocity, for feature encoding. The output is a feature vector containing information about the ablation process, i.e., the mechanical erosion amount. This feature vector is denoted as... The outputs of the two encoders mentioned above and Feature fusion is performed through a feature fusion layer, and the fused feature vector is... The fused feature vector is then passed as input to subsequent neural network layers for further nonlinear transformation; the fused feature vector is then passed to multiple hidden layers, where fully connected layers process the fused features.

[0149]

[0150] in, It is an activation function (such as ReLU, tanh, etc.). It is the weight matrix of the first layer in the neural network. It is a bias vector. This represents the fused feature vector; It represents the hidden state of the l-th layer in the neural network;

[0151] The loss function is obtained from the physical information neural network model, and the model is trained to obtain the final physical information neural network model based on the evolution of the microscopic sintering-erosion surface morphology of composite materials; specifically:

[0152] Physical equation constraints are introduced into the physical information neural network model. Automatic differentiation techniques are used to calculate the physical constraint loss, boundary condition loss, initial condition loss, and mechanical erosion condition loss. A weighted sum is then used to obtain the final loss function. The specific process is as follows:

[0153] Based on the aforementioned description of the physical information of ablation and erosion, the loss function is obtained from the neural network. This embodiment introduces constraints from physical equations during the neural network training process. Specifically, automatic differentiation is added between the last layer of the fully connected layer and the output layer. The purpose is to enable the network to calculate the gradients of physical equations, such as the gas diffusion equation and the new surface function after ablation, and to optimize the model parameters based on these gradients. Automatic differentiation automatically calculates the gradient of the loss function with respect to the network weights and inputs by tracking every operation in the network computation process. The physical constraint loss, boundary condition loss, initial condition loss, and mechanical ablation condition loss take the following forms:

[0154] The physical constraint loss, i.e., the loss function of the gas diffusion equation, is:

[0155]

[0156] in, It is the reaction concentration; It is time; It is the diffusion coefficient; Indicates a gas phase unit. Represents a solid-state unit; 0 < <1 represents a hybrid unit; It is a gradient operator; It is the physical constraint loss function;

[0157] The loss function for the boundary conditions and initial conditions is:

[0158]

[0159]

[0160] in, It is the reaction concentration; It represents the gas concentration in the initial state. It refers to the concentration of ambient gases; These are weighting coefficients used to balance the relative importance of different losses; It is the temperature of the composite material; It is the initial temperature of the environment; It is the boundary condition loss function; It is the initial condition loss function;

[0161] The loss function for the mechanical erosion condition is:

[0162]

[0163] in, It is the average height of the surface substrate; It is the height of the surface yarn in the composite material model; It is the amount of mechanical erosion; It is the mechanical erosion constraint loss function;

[0164] This embodiment further combines the above loss functions in a weighted sum form by setting weighting factors. The weighting factors determine the relative importance of the three losses in the final loss function, which can be expressed as follows:

[0165]

[0166] in, These are weighting coefficients used to balance the relative importance of different sub-loss functions; It is the final loss function;

[0167] This application constructs an ablation-exfoliation model based on a Physical Information Neural Network (PINN) through the above steps, and calculates the physical constraint loss by automatic differentiation, combining it with the losses of boundary conditions and initial conditions for training. During training, an optimizer (such as Adam) can be used to update the network parameters, and weight factors are set to balance the physical loss and data loss. After training, the model's performance is verified by evaluation metrics and visualization methods.

[0168] Finally, based on the trained physical information neural network, this application can solve the surface morphology evolution of composite material micro-ablation-exfoliation model under different working conditions. The specific steps are: (1) Loading the trained model: After training, the trained model is saved and loaded so as to predict the surface morphology evolution under different working conditions; (2) Inputting different working condition data: The working condition data can be different physical conditions such as gas diffusion concentration, pressure field, time, etc. By inputting these data into the trained model, the physical information neural network will predict the corresponding surface morphology changes according to the physical laws it has learned; (3) Predicting the micro-surface morphology: By inputting the inputs under different working conditions, such as different spatial locations and time points, into the network, the corresponding surface morphology, such as ablation depth and mechanical exfoliation amount, can be obtained.

[0169] The PINN method proposed in this application can quickly predict the morphological evolution of the entire ablation process by directly starting from the physical equations through a neural network. Without requiring layer-by-layer calculations and surface updates, the network can solve the entire process in one go and directly provide the overall ablation morphology over time, greatly improving computational efficiency and showing great promise for application.

[0170] Figure 5 This is a schematic diagram of a rapid prediction device for the microscopic burn-erosion morphology evolution of composite materials provided in this application embodiment. (See attached diagram.) Figure 5 The device for rapid prediction of microscopic burn-ebulation morphology evolution includes various functional modules for implementing the aforementioned rapid prediction method for microscopic burn-ebulation morphology evolution. Any functional module can be implemented by software and / or hardware.

[0171] In some embodiments, the rapid prediction device 1000 for the evolution of microscopic sintering-erosion morphology of composite materials based on physical information neural networks includes a data acquisition module 1001, a model condition verification module 1002, a model construction module 1003, a model training module 1004, and a rapid prediction module 1005 for the evolution of microscopic sintering-erosion morphology; wherein:

[0172] The data acquisition module 1001 is used to acquire a dataset, wherein the dataset includes the micro-ablation morphology of the carbon-carbon composite surface obtained by processing the micro-ablation model of the constructed carbon-carbon composite material based on the fixed volume fraction method.

[0173] The model condition verification module 1002 is used to determine the initial conditions, boundary conditions, and erosion criteria of the composite material microstructure burn-erosion model;

[0174] Model building module 1003 is used to build a physical information neural network model based on the evolution of the microscopic sintered-eroded surface morphology of composite materials;

[0175] The model training module 1004 is used to input the dataset into the physical information neural network model for iterative training to obtain the trained physical information neural network model;

[0176] The rapid prediction module 1005 for the evolution of micro-sintering-erosion morphology is used to input the composite material to be tested into the trained physical information neural network model to obtain the prediction results of the evolution of micro-sintering-erosion morphology of the composite material.

[0177] The composite material microstructure sintering-erosion morphology evolution rapid prediction device 1000 provided in this application embodiment is used to execute the technical solution provided in the aforementioned composite material microstructure sintering-erosion morphology evolution rapid prediction method embodiment. Its implementation principle and technical effect are similar to those in the aforementioned method embodiment, and will not be repeated here.

[0178] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing elements, entirely in hardware, or partially in software via processing elements and partially in hardware. For example, the data acquisition module 1001 can be a separate processing element, or it can be integrated into a chip in the above device. Alternatively, it can be stored as program code in the memory of the above device, and its functions can be called and executed by a processing element. The implementation of other modules is similar. Moreover, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through integrated logic circuits in the hardware of the processor element or through software instructions.

[0179] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. (See attached diagram.) Figure 6 The electronic device 1100 includes: a processor 1101, and a memory 1102 communicatively connected to the processor 1101;

[0180] Memory 1102 stores computer-executed instructions;

[0181] The processor 1101 executes the computer execution instructions stored in the memory 1102 to realize the technical solution of the aforementioned method for rapid prediction of the microscopic burn-erosion morphology evolution of composite materials.

[0182] In the aforementioned electronic device 1100, the memory 1102 and the processor 1101 are electrically connected directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines, such as bus connections. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be classified as address buses, data buses, control buses, etc., but this does not mean that there is only one bus or one type of bus. The memory 1102 stores computer execution instructions for implementing the aforementioned method for predicting the surface roughness of thermal protection materials, including at least one software functional module that can be stored in the memory 1102 in the form of software or firmware. The processor 1101 executes various functional applications and data processing by running the software programs and modules stored in the memory 1102.

[0183] The memory 1102 includes at least one type of readable storage medium, not limited to Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory 1102 stores programs, and the processor 1101 executes the programs after receiving execution instructions. Furthermore, the software programs and modules within the memory 1102 may also include an operating system, which may include various software components and / or drivers for managing system tasks (e.g., memory management, storage device control, power management, etc.) and can communicate with various hardware or software components to provide an operating environment for other software components.

[0184] Processor 1101 can be an integrated circuit chip with signal processing capabilities. The aforementioned processor 1101 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor, or processor 1101 can be any conventional processor.

[0185] The electronic device 1100 is used to execute the technical solution provided in the aforementioned embodiment of the rapid prediction method for the microscopic burn-erosion morphology evolution of composite materials. Its implementation principle and technical effect are similar to those in the aforementioned method embodiment, and will not be repeated here.

[0186] This application also provides a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the technical solution of the aforementioned method for predicting the surface roughness of thermal protection materials.

[0187] The aforementioned computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The computer-readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0188] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Alternatively, the readable storage medium can be an integral part of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components within the control device of a thermal protection material surface roughness prediction device.

[0189] This application also provides a computer program product, including a computer program that, when executed by a processor, is used to implement the aforementioned technical solution for predicting the surface roughness of thermal protection materials.

[0190] In the above embodiments, those skilled in the art will understand that the above method embodiments can be implemented entirely or partially by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented entirely or partially in the form of a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless network, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0191] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for predicting the microscopic burn-erosion morphology of composite materials, characterized in that, include: Obtain the dataset; wherein the dataset includes the micro-ablation morphology of the carbon-carbon composite surface obtained by processing the micro-ablation model of the constructed carbon-carbon composite material based on the fixed volume fraction method. Determine the initial conditions, boundary conditions, and erosion criteria for the microscopic burn-erosion model of composite materials; Construct a physical information neural network model based on the evolution of microscopic sintered-eroded surface morphology of composite materials; The dataset is input into the physical information neural network model for iterative training to obtain the trained physical information neural network model. The composite material to be tested is input into the trained physical information neural network model to obtain the prediction results of the microscopic burn-erosion morphology evolution of the composite material. The construction of the physical information neural network model based on the evolution of the microscopic burn-eradication surface morphology of composite materials includes: Based on the determined initial conditions, boundary conditions, and erosion criteria of the composite material microstructure burn-erosion model, residual points, boundary constraint points, and initial constraint points for training are selected within the computational domain. Construct a custom multilayer perceptron neural network structure; A neural network is constructed based on a custom improved multilayer perceptron structure, and training points are input for training to establish a physical information neural network model. The loss function is obtained from the physical information neural network model, and the model is trained to obtain the final physical information neural network model based on the evolution of the microscopic sintering-erosion surface morphology of composite materials. The selection of residual points, boundary constraint points, and initial constraint points for training within the computational domain includes: Latin hypercube random sampling is used to select residual points in the computational domain of composite materials; Set boundary constraint points at the top boundary of the composite material, and set boundary constraint points at the lateral boundary using symmetry conditions; Initial constraint points for temperature and gas concentration are set on the surface and inside the composite material.

2. The method according to claim 1, characterized in that, The acquisition of the dataset includes: Construct a microscopic geometric model of the composite material; Based on the aforementioned microscopic geometric model, a composite material ablation model is constructed; Based on the solid volume fraction method, the ablation retreat calculation is performed on each unit in the geometric model according to the microscopic ablation model of the composite material to obtain the solid volume fraction dataset of each unit, so as to constitute the microscopic ablation-exfoliation morphology of the composite material surface.

3. The method according to claim 2, characterized in that, The microscopic geometric model for constructing the composite material includes: A microscopic geometric model of carbon-carbon composite materials, including carbon fibers and a carbon matrix, was established based on TexGen, and the microscopic geometric model was transformed into a discrete model of cubic elements. The carbon fibers and carbon matrix are endowed with material properties including fiber volume fraction, density, thermal conductivity, porosity, and specific heat capacity.

4. The method according to claim 3, characterized in that, The construction of the composite material ablation model based on the microscopic geometric model includes: Construct the spatial surface function of the composite material in three-dimensional coordinates; when the composite material is ablated and retreats, each point on the surface retreats a certain distance along the normal vector direction within a certain time interval to form a new surface function after ablation; and the gas diffusion equation of the carbon matrix of the composite material during ablation and retreat. The spatial surface function of the composite material in three-dimensional coordinates is: As the composite material ablates and recedes, each point on the surface recedes a certain distance along the normal vector direction within a certain time interval, forming a new surface function after ablation. in, It is a material surface function, present at the gas-solid or liquid-solid interface. ; These are spatial coordinates; It is time; It is the reaction concentration; It refers to the speed at which the interface goes back; k It is a heterogeneous reaction rate; It is the molar volume of the solid phase; It is the outward normal of the solid phase interface; It is a gradient operator; It is the magnitude of the vector; The gas diffusion equation for the carbon matrix of the composite material during ablation retreat is as follows: in, It is the reaction concentration; It is time; It is the diffusion rate; It is the gradient operator.

5. The method according to claim 4, characterized in that, The solid volume fraction method involves performing ablation regression calculations on each element of the geometric model based on the microscopic ablation model of the composite material to obtain a dataset of the solid volume fraction for each element, including: Define the solid volume fraction of each unit as: in, It is the volume of the solid in the unit cell. It is the unit volume; The oxygen diffusion equation and the new surface function after ablation, rewritten using the solid volume fraction, are as follows: in, It is the volume of the solid in the unit cell; It is the unit volume; Indicates a gas phase unit. Represents a solid-state unit; 0 < <1 represents a hybrid unit; It is the reaction concentration; It is time; It is the diffusion coefficient; It is the outward normal of the solid phase interface; k It is a heterogeneous reaction rate; It is a gradient operator; It is the absolute value of the screen back speed; The finite volume method was used to solve the oxygen diffusion equation and the new surface function after ablation, and a dataset of solid volume fractions was obtained.

6. The method according to claim 1, characterized in that, The initial conditions, boundary conditions, and erosion criteria for determining the microscopic burn-erosion model of composite materials include: The initial conditions for defining the computational domain include the temperature field, gas concentration, and material morphology; the temperature field is defined as the initial temperature of the composite material and the environment. The gas concentration is set to the ambient gas concentration. The material morphology is defined as the initial surface shape and microstructure distribution of the material. ; The boundary conditions for the computational domain include a top boundary and lateral boundaries; the top boundary uses Dirichlet conditions. ,in This represents the gas concentration in the initial state; the lateral boundary is symmetrical, meaning the normal gradient between temperature and gas concentration is 0. The erosion criterion is a mechanical erosion criterion, specifically: in, It is the first in the composite material model The height of the floor, ; It is the maximum height of the surface yarn in the composite material model; It is the average height of the surface matrix in the composite material model; It is the mechanical erosion coefficient related to materials and the environment; It is the width of the weft yarn in the composite material model; It refers to the number of warp yarns covering the weft yarns; The length of the weft yarn; It is the first Height of the weft yarns; Its mechanical erosion at the microscale is as follows: in, It is the average height of the surface substrate; It is the height of the surface yarn in the composite material model; It is the amount of mechanical erosion.

7. The method according to claim 1, characterized in that, The custom multilayer perceptron neural network structure is constructed, which includes: an input layer, an encoder layer, a fusion layer, a hidden layer, an automatic micro-layer, and an output layer. The encoder layer includes an ablation encoder and an ablation encoder; the ablation encoder is used to perform feature encoding processing on the input data of the composite material ablation process and output an ablation process feature vector; the ablation encoder is used to perform feature encoding processing on the input data of the composite material ablation process and output an ablation process feature vector. The input data for the composite material ablation process includes the surface temperature and gas concentration during the composite material ablation process; The input data for the composite material ablation process includes airflow velocity and surface morphology during the ablation process.

8. The method according to claim 7, characterized in that, The loss function obtained based on the physical information neural network model includes: Physical equation constraints are introduced into the physical information neural network model. The physical constraint loss, boundary condition loss, initial condition loss and mechanical erosion condition loss are calculated using automatic differentiation technology. The final loss function is obtained by weighted summation.

9. The method according to claim 8, characterized in that: The physical constraint loss, boundary condition loss, initial condition loss, and mechanical erosion condition loss take the following forms: The physical constraint loss, i.e., the loss function of the gas diffusion equation, is: in, It is the reaction concentration; It is time; It is the diffusion coefficient; Indicates a gas phase unit. Represents a solid-state unit; 0 < <1 represents a hybrid unit; It is a gradient operator; It is the physical constraint loss function; The loss function for the boundary conditions and initial conditions is: in, It is the reaction concentration; It represents the gas concentration in the initial state. It refers to the concentration of ambient gases; These are weighting coefficients used to balance the relative importance of different losses; It is the temperature of the composite material; It is the initial temperature of the environment; It is the boundary condition loss function; It is the initial condition loss function; The loss function for the mechanical erosion condition is: in, It is the average height of the surface substrate; It is the height of the surface yarn in the composite material model; It is the amount of mechanical erosion; It is the mechanical erosion constraint loss function; The final loss function is a weighted sum of the physical constraint loss, boundary condition loss, initial condition loss, and mechanical erosion constraint loss: ； in, These are weighting coefficients used to balance the relative importance of different sub-loss functions; It is the final loss function.

10. A composite material microstructure prediction device for performing the method according to any one of claims 1-9, characterized in that, include: The data acquisition module is used to acquire a dataset; wherein, the dataset includes the micro-ablation morphology of the carbon-carbon composite surface obtained by processing the micro-ablation model of the constructed carbon-carbon composite material based on the fixed volume fraction method. The model condition verification module is used to determine the initial conditions, boundary conditions, and erosion criteria of the composite material microstructure burn-erosion model. The model building module is used to construct a physical information neural network model based on the evolution of the microscopic sintered-eroded surface morphology of composite materials. The model training module is used to input the dataset into the physical information neural network model for iterative training to obtain the trained physical information neural network model. The rapid prediction module for the evolution of microscopic sintering-erosion morphology is used to input the composite material to be tested into the trained physical information neural network model to obtain the prediction results of the microscopic sintering-erosion morphology evolution of the composite material.

11. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 9.

Images