SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method

Abstract

The present invention provides a SiC / SiC pyrolytic carbon interface optimization method based on a neural network and a phase field fracture method. The method comprises: first constructing a representative volume unit model of a SiC fiber-reinforced composite material; then using the phase field fracture method to simulate the transverse tensile strength of the representative volume unit model of the SiC fiber-reinforced composite material; finally, using the transverse tensile strength as a criterion for screening the optimal interface thickness and training an inverse neural network to predict and optimize the optimal interface thickness of SiC fibers with different material parameters. The present invention uses an accurate and efficient representative volume unit model and combines the phase field fracture method with an inverse design neural network. By simulating crack behavior and predicting the optimal interface thickness of fibers with different radii, elastic moduli, critical energy release rates, and strengths, the PyC interface parameters are systematically optimized, thereby improving the mechanical properties and reliability of the SiC / SiC composite material.

Description

Technical Field

[0001] The present invention belongs to the technical field of silicon carbide composite materials, relates to the optimization design of silicon carbide composite materials, and specifically relates to a SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method. Background Art

[0002] SiC / SiC composites are continuous silicon carbide fiber-reinforced silicon carbide ceramic-based composites. Due to their low density, high strength, high-temperature resistance, and excellent oxidation resistance, they hold great promise for application in advanced nuclear energy, aerospace, and other fields. In SiC / SiC composites, the pyrolytic carbon (PyC) interface layer is a key structure connecting the fibers to the matrix. It is deposited on the fiber surface primarily through processes such as chemical vapor infiltration and precursor impregnation and pyrolysis. The PyC interface layer not only protects the fibers from damage during the fabrication process but also significantly influences the mechanical properties and fracture behavior of the composite by regulating the bond strength between the fibers and the matrix. Studies have shown that appropriate PyC interface thickness can improve the fracture toughness of the material and prevent brittle failure. Optimizing interface parameters (such as thickness and microstructure) is a key technical challenge in enhancing the overall performance of SiC / SiC composites. However, traditional optimization methods often rely on trial-and-error experiments, which are limited by low efficiency and high cost.

[0003] In recent years, with the development of computational materials science, numerical simulation technology has provided a new path for interface optimization. Phase-field fracture analysis is an advanced simulation method that can effectively describe the initiation, propagation, and coalescence of cracks in materials, making it particularly suitable for studying the complex fracture mechanisms of composite materials. Furthermore, neural networks, as powerful machine learning algorithms, can rapidly establish mappings between process parameters and material properties by learning from experimental or simulated data, enabling efficient prediction and design optimization.

[0004] In the field of SiC / SiC composites, there is limited research on numerical simulation of cracks in mesoscopic models of SiC / SiC composites. Currently, crack simulation methods fall into two main categories: one is geometric description methods, such as the element deletion method, the interface element method, and the cohesive force model method; the other is non-geometric description methods, such as the extended finite element method, in which cracks can break away from the mesh and propagate within the mesh. For example, in the paper "Numerical Simulation of the Tensile Mechanical Behavior of Defective C / SiC Plain Woven Composites," the authors, based on a mesoscopic unit cell model, employed the Linde failure criterion and the Weibull distribution to simulate the random distribution of pore defects. They successfully predicted the tensile stress-strain behavior of the plain woven composite, which was in good agreement with experimental data. However, this study used a progressive damage model and did not consider the dynamic degradation of interfacial phases (such as the pyrolytic carbon layer) and its influence on the crack path. For example, in the paper "Molecular Dynamics Study on the Tensile Behavior of SiC Nanofiber / C / SiC Composites", the authors used molecular dynamics simulation methods to numerically simulate how the thickness of the amorphous carbon (aC) coating on SiC nanofibers significantly affects the fracture mode and mechanical properties of the composite material; however, due to the small simulation scale of the molecular dynamics method, it is difficult to guide the optimization of macroscopic preparation process parameters.

[0005] In summary, there is little research on the numerical simulation of mesoscopic cracks in SiC / SiC composites at home and abroad, and there are limitations in the characterization of composite material defects and simulation of interface fracture behavior, resulting in defects in interface optimization design based on numerical simulation. Summary of the Invention

[0006] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method, so as to solve the technical problem that the existing technology has limitations in the characterization of composite material defects and simulation of interface fracture behavior, resulting in defects in interface optimization design based on numerical simulation.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0008] A SiC / SiC pyrolytic carbon interface optimization method based on a neural network and a phase field fracture method comprises: firstly constructing a representative volume unit model of a SiC fiber reinforced composite material; then using the phase field fracture method to perform a transverse tensile simulation on the representative volume unit model of the SiC fiber reinforced composite material to obtain the transverse tensile strength; finally, using the transverse tensile strength as a criterion for screening the optimal interface thickness and training an inverse neural network to achieve the prediction and optimization of the optimal interface thickness of SiC fibers with different material parameters.

[0009] The present invention also has the following technical features:

[0010] The method specifically comprises the following steps:

[0011] Step 1: Construct a representative volume element model of SiC fiber reinforced composite materials:

[0012] Step 1.1, set the cross-sectional size parameters of SiC / SiC composite material:

[0013] Set the cross-sectional dimensions of the SiC fiber-reinforced composite material. The cross-sectional dimensions of the SiC fiber-reinforced composite material include: SiC matrix length, SiC matrix width, SiC fiber radius r, and PyC boundary layer thickness h. Establish a SiC / SiC composite material model based on these parameters. Specifically, in step 1.1, the SiC / SiC composite material model is a transverse tensile model containing a single fiber. The SiC matrix length and width are 25 μm, the SiC fiber radius r is 4-8 μm, and the PyC boundary layer thickness h is 50-550 nm.

[0014] Step 1.2, mesh division:

[0015] Mesh the SiC / SiC composite model constructed in step 1.1 using quadrilateral elements and refine the mesh in areas where cracks are likely to grow to improve the accuracy of the calculation results. This completes the construction of a representative volume element model of the SiC fiber-reinforced composite material. Specifically, in step 1.2, when meshing, set the mesh size to 1 / 100 of the SiC matrix width in step 1.1; when refining the mesh, set the mesh size to 1 / 500 to 1 / 400 of the SiC matrix width in step 1.1.

[0016] Step 2: Phase field fracture simulation:

[0017] Step 2.1, set material property parameters:

[0018] The material property parameters include: Young's modulus E1 of the SiC fiber interface, 150 GPa≤E1≤430 GPa; Young's modulus E2 of the SiC matrix interface, 150 GPa≤E2≤430 GPa; Young's modulus E3 of the PyC interface, 10 GPa≤E3≤30 GPa; strength Ft1 of the SiC fiber interface, 300 MPa≤Ft1≤600 MPa; strength Ft2 of the SiC fiber interface, 300 MPa≤Ft2≤600 MPa; strength Ft3 of the PyC interface, 20 MPa≤Ft3≤60 MPa; critical energy release rate Gc1 of the SiC fiber interface, 6 J / m 2 ≤Gc1≤10 J / m 2; Critical energy release rate of SiC matrix interface Gc2,6 J / m 2 ≤Gc2≤10 J / m 2 ; Critical energy release rate of PyC interface Gc3, 0.4 J / m 2 ≤Gc3≤0.6 J / m 2 ; The above parameters are randomly selected within the value range.

[0019] The material property parameters also include: the Poisson's ratio v1 of the SiC fiber interface, v1 is 0.185; the Poisson's ratio v2 of the SiC matrix interface, v2 is 0.185; and the Poisson's ratio v3 of the PyC interface, v3 is 0.18. After the values of each material property parameter are determined, the maximum and minimum normalization processing is performed within the respective value range using Formula I; Formula I is as follows:

[0020] Formula I.

[0021] Where:

[0022] X represents the normalized value.

[0023] A indicates the current value.

[0024] A max Indicates the maximum value among all values.

[0025] A min Indicates the minimum value among all values.

[0026] Step 2.2, set boundary conditions:

[0027] Set displacement constraints on the left and right sides of the model created in step 1.1.

[0028] Step 2.3, phase-field fracture simulation:

[0029] The phase field cohesion control equation and related parameters are set, and a partial differential equation solver is set to solve the phase field cohesion control equation, and related parametric equations and related material parameters are set; the phase field cohesion control equation is as follows:

[0030] Formula II.

[0031] Where:

[0032] represents the governing equation for phase field cohesion.

[0033] Represents the phase field value.

[0034] represents the first derivative of the degradation function.

[0035] The strain energy density at a point is The historical maximum value in the time period.

[0036] represents the critical energy release rate.

[0037] Represents the first derivative of the crack geometry function.

[0038] represents the proportionality constant.

[0039] Represents the characteristic width of the phase field model.

[0040] Step 2.4, output of transverse tensile strength:

[0041] After the phase field fracture simulation, the model stress-strain curve was extracted and the transverse tensile strength was obtained from the stress-strain curve.

[0042] Step 3: Construction and training of neural network:

[0043] Step 3.1, dataset collection:

[0044] Repeat step 2 N times to obtain N transverse tensile strengths to form a forward neural network training set; N is an integer greater than or equal to 200 and less than or equal to 2000.

[0045] Step 3.2, forward neural network training:

[0046] For the forward neural network, its input is the thickness h of the PyC boundary layer, the radius r of the SiC fiber, the Young's modulus E1 of the SiC fiber interface, the Young's modulus E2 of the SiC matrix interface, the Young's modulus E3 of the PyC interface, the strength Ft1 of the SiC fiber interface, the strength Ft2 of the SiC fiber interface, the strength Ft3 of the PyC interface, the critical energy release rate Gc1 of the SiC fiber interface, the critical energy release rate Gc2 of the SiC matrix interface, and the critical energy release rate Gc3 of the PyC interface; the output is the transverse tensile strength.

[0047] Specifically, in step 3.2, the input layer of the forward neural network has 11 parameters, the hidden layer has 5 layers, and each hidden layer contains 256 neurons; the output layer of the forward neural network has 1 parameter, and the ReLU activation function is used between layers to maintain the nonlinearity of the model. The Adam algorithm is used for model optimization. The initial learning rate of the Adam optimization algorithm is 1e-3, and the weight decay coefficient is set to 1e-5. The number of training times is 1000.

[0048] Step 3.3, obtain the optimal solution:

[0049] The forward neural network is used to traverse the thickness h of the PyC boundary layer under different SiC fiber radii r and different material property parameters. The transverse tensile strength is used as the criterion to obtain the optimal PyC thickness, and the results are used to collect data sets for the reverse neural network.

[0050] Step 3.4, reverse neural network training:

[0051] For the inverse neural network, its input is the SiC fiber radius r, Young's modulus E1 of the SiC fiber interface, Young's modulus E2 of the SiC matrix interface, Young's modulus E3 of the PyC interface, strength Ft1 of the SiC fiber interface, strength Ft2 of the SiC fiber interface, strength Ft3 of the PyC interface, critical energy release rate Gc1 of the SiC fiber interface, critical energy release rate Gc2 of the SiC matrix interface and critical energy release rate Gc3 of the PyC interface; the output is the optimal PyC thickness; after the training is completed, the optimization model of the pyrolytic carbon interface in SiC / SiC composite materials is obtained.

[0052] Specifically, in step 3.4, the input layer of the inverse neural network is 10 parameters, the hidden layer is 5 layers, each hidden layer contains 256 neurons, and the output layer is 1 parameter. The Sigmoid function is used as the activation function between layers and the nonlinearity of the model is maintained. The Adam algorithm is used for model optimization, and the initial learning rate is set to 5e-3, and the number of training times is 1000.

[0053] The beneficial technical effects of the present invention compared with the prior art are as follows:

[0054] (I) The present invention uses an accurate and efficient representative volume element model and combines the phase-field fracture method with an inverse design neural network. By simulating crack behavior and predicting the optimal interface thickness of fibers with different radii, elastic moduli, critical energy release rates, and strengths, the PyC interface parameters are systematically optimized, thereby improving the mechanical properties and reliability of SiC / SiC composites.

[0055] (II) The phase-field fracture method used in the present invention is a method that can simulate fracture in a continuous medium. It can accurately capture the entire process of material damage from crack nucleation to crack propagation and final fracture, and can solve the problems of complex fracture modes and multi-physical field coupling.

[0056] (III) The present invention is entirely carried out on a computer, which reduces unnecessary experimental trial and error and experimental dependence, and improves the efficiency of interface optimization design of SiC / SiC composite materials.

[0057] (IV) The inverse optimization design neural network used in the present invention is composed of a forward neural network and an inverse neural network connected in series. It can be applied not only to the interface optimization design of various SiC / SiC composite materials containing bundled SiC fibers, PyC interfaces and SiC matrices, but can also be extended to other brittle composite materials containing interfaces, that is, it has wide applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Flowchart of the SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method.

[0059] Figure 2 Schematic diagram of the neural network structure obtained in an embodiment of the present invention.

[0060] Figure 3 This is the prediction result of the interface optimization model obtained in the embodiment of the present invention for the optimal interface thickness of certain fibers.

[0061] The technical solution of the present invention is further described below in conjunction with embodiments. DETAILED DESCRIPTION

[0062] In the present invention, SiC refers to silicon carbide, the two SiCs in "SiC / SiC" refer to SiC fiber and SiC matrix, respectively, and PyC refers to pyrolytic carbon.

[0063] It should be noted that all software, algorithms, and functions used in the present invention, unless otherwise specified, are software, algorithms, and functions known in the art. For example:

[0064] The modeling and phase-field fracture simulation software adopts the commonly used Comsol 6.0 software known in the prior art.

[0065] The Sigmoid function, also known as the S-shaped growth curve, is a commonly used activation function. It generally exists between the input layer and the output layer of a neural network. Its function is to add some nonlinear factors to the neural network.

[0066] The Adam (Adaptive Moment Estimation) algorithm is an optimization algorithm that combines the momentum method and adaptive learning rate. It was proposed by DPKingma and J.Ba in 2014.

[0067] Represents the degradation function, which is generally chosen as: ;in is the phase field value to be solved.

[0068] In accordance with the above technical solution, specific embodiments of the present invention are given below. It should be noted that the present invention is not limited to the following specific embodiments, and all equivalent changes made on the basis of the technical solution of this application fall within the protection scope of the present invention.

[0069] Example:

[0070] This embodiment provides a SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method; Figure 1 As shown, the method specifically includes the following steps:

[0071] Step 1: Construct a representative volume element model of SiC fiber reinforced composite materials:

[0072] Step 1.1, set the cross-sectional size parameters of SiC / SiC composite material:

[0073] To simulate the transverse tensile strength of SiC fiber-reinforced composites, it is necessary to set the dimensions of the SiC fiber-reinforced composite cross-section. The cross-sectional dimensional parameters of the SiC fiber-reinforced composite include: a SiC matrix length of 25 μm, a SiC matrix width of 25 μm, a SiC fiber radius r of 4 to 8 μm, and a PyC boundary layer thickness h of 50 to 550 nm. A SiC / SiC composite model is established based on the set dimensional parameters (this SiC / SiC composite model is a transverse tensile model containing a single SiC fiber).

[0074] Step 1.2, mesh division:

[0075] Use quadrilateral elements to mesh the SiC / SiC composite material model built in step 1.1. Set the mesh size to 1 / 100 of the SiC matrix width in step 1.1. Also, mesh the area where cracks may grow with a mesh size of 1 / 500 of the SiC matrix width in step 1.1 to make the calculation results more accurate. This completes the construction of a representative volume element model of SiC fiber-reinforced composite materials.

[0076] Step 2: Phase field fracture simulation:

[0077] Step 2.1, set material property parameters:

[0078] The values of the material property parameters are as follows: Young's modulus E1 of the SiC fiber interface, 150 GPa≤E1≤430 GPa; Young's modulus E2 of the SiC matrix interface, 150 GPa≤E2≤430 GPa; Young's modulus E3 of the PyC interface, 10 GPa≤E3≤30 GPa; strength Ft1 of the SiC fiber interface, 300 MPa≤Ft1≤600 MPa; strength Ft2 of the SiC fiber interface, 300 MPa≤Ft2≤600 MPa; strength Ft3 of the PyC interface, 20 MPa≤Ft3≤60 MPa; critical energy release rate Gc1 of the SiC fiber interface, 6 J / m 2 ≤Gc1≤10 J / m 2 ; Critical energy release rate of SiC matrix interface Gc2,6 J / m 2 ≤Gc2≤10 J / m 2 ; Critical energy release rate of PyC interface Gc3, 0.4 J / m 2 ≤Gc3≤0.6 J / m 2 Randomly select values for each material property parameter within the above range. Furthermore, the Poisson's ratios v1 and v1 at the SiC fiber interface were set to 0.185; the Poisson's ratios v2 and v2 at the SiC matrix interface were set to 0.185; and the Poisson's ratios v3 and v3 at the PyC interface were set to 0.18.

[0079] The material property parameters need to be normalized individually using Formula I, which is as follows:

[0080] Formula I.

[0081] Where:

[0082] X represents the normalized value.

[0083] A indicates the current value.

[0084] A max Indicates the maximum value among all values.

[0085] A min Indicates the minimum value among all values.

[0086] Step 2.2, set boundary conditions: In order to simulate the transverse tensile strength of SiC / SiC composite materials, set displacement constraints on the left and right sides of the model established in step 1.1.

[0087] Step 2.3, phase field fracture simulation: Set the phase field cohesion control equation and related parameters, set the partial differential equation solver to solve the phase field cohesion control equation, set the related parametric equations and related material parameters; the phase field cohesion control equation is as follows:

[0088] Formula II.

[0089] Where:

[0090] represents the governing equation for phase field cohesion.

[0091] Represents the phase field value, to be solved, used to describe the fracture situation; 0 represents completely intact, and 1 represents completely broken.

[0092] It represents the first derivative of the degradation function, which is used to reflect the reduction of strain energy due to fracture damage.

[0093] The strain energy density at a point is The historical maximum value in the time period.

[0094] represents the critical energy release rate, which is related to the material properties and is set in step 2.

[0095] Represents the first derivative of the crack geometry function.

[0096] Represents the proportional constant; generally takes the value .

[0097] Represents the characteristic width of the phase field model, which is used to control the degree of crack diffusion; it is generally taken as 1 / 4 of the grid size.

[0098] Step 2.4, output of transverse tensile strength: After the phase field fracture simulation is completed, extract the model stress-strain curve, where the horizontal axis is strain and the vertical axis is stress. The value of the highest point (maximum value) obtained from the stress-strain curve is the transverse tensile strength.

[0099] Step 3: Construction and training of neural network:

[0100] Step 3.1, data set collection: Repeat step 2 1000 times to obtain 1000 transverse tensile strengths to form the forward neural network training set.

[0101] Step 3.2, forward neural network training: For the forward neural network, its input is the thickness h of the PyC boundary layer, the radius r of the SiC fiber, the Young's modulus E1 of the SiC fiber interface, the Young's modulus E2 of the SiC matrix interface, the Young's modulus E3 of the PyC interface, the strength Ft1 of the SiC fiber interface, the strength Ft2 of the SiC fiber interface, the strength Ft3 of the PyC interface, the critical energy release rate Gc1 of the SiC fiber interface, the critical energy release rate Gc2 of the SiC matrix interface, and the critical energy release rate Gc3 of the PyC interface; the output is the transverse tensile strength.

[0102] As a specific solution of this embodiment, the forward neural network input layer has 11 parameters, the hidden layer has 5 layers, and each hidden layer contains 256 neurons; the forward neural network output layer has 1 parameter, and the ReLU activation function is used between layers to maintain the nonlinearity of the model. The Adam algorithm is used for model optimization, the initial learning rate of the Adam optimization algorithm is 1e-3, and the weight decay coefficient is set to 1e-5, and the number of training times is 1000.

[0103] Step 3.3, obtain the optimal solution: the forward neural network is traversed for the thickness h of the PyC boundary layer at different SiC fiber radii r and different material property parameters (including Young's modulus E1 of the SiC fiber interface, Young's modulus E2 of the SiC matrix interface, Young's modulus E3 of the PyC interface, strength Ft1 of the SiC fiber interface, strength Ft2 of the SiC fiber interface, strength Ft3 of the PyC interface, critical energy release rate Gc1 of the SiC fiber interface, critical energy release rate Gc2 of the SiC matrix interface, and critical energy release rate Gc3 of the PyC interface), and the transverse tensile strength is used as the criterion to obtain the optimal PyC thickness. The results are used to collect data sets for the reverse neural network.

[0104] Step 3.4, reverse neural network training: For the reverse neural network, its input is the SiC fiber radius r, the elastic modulus E1 of the SiC fiber interface, the Young's modulus E2 of the SiC matrix interface, the Young's modulus E3 of the PyC interface, the strength Ft1 of the SiC fiber interface, the strength Ft2 of the SiC fiber interface, the strength Ft3 of the PyC interface, the critical energy release rate Gc1 of the SiC fiber interface, the critical energy release rate Gc2 of the SiC matrix interface, and the critical energy release rate Gc3 of the PyC interface; the output is the optimal PyC thickness. After the training is completed, the optimization model of the pyrolytic carbon interface in the SiC / SiC composite material is obtained (such as Figure 3 The obtained neural network structure is shown as Figure 2 shown.

[0105] As a specific solution of this embodiment, the inverse neural network input layer is 10 parameters, the hidden layer is 5 layers, each hidden layer contains 256 neurons, and the output layer is 1 parameter. The Sigmoid function is used as the activation function between layers and the nonlinearity of the model is maintained. The Adam algorithm is used for model optimization, and the initial learning rate is set to 5e-3, and the number of training times is 1000.

[0106] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method, characterized in that: The method includes: first constructing a representative volume unit model of SiC fiber reinforced composite materials; then using the phase field fracture method to simulate the transverse tensile strength of the representative volume unit model of SiC fiber reinforced composite materials to obtain the transverse tensile strength; finally, using the transverse tensile strength as a criterion to screen the optimal interface thickness and train an inverse neural network to predict and optimize the optimal interface thickness of SiC fibers with different material parameters; The method specifically comprises the following steps: Step 1: Construct a representative volume element model of SiC fiber reinforced composite materials: Step 1.1, set the cross-sectional size parameters of SiC / SiC composite material: Set the size of the cross section of the SiC fiber reinforced composite material. The cross section size parameters of the SiC fiber reinforced composite material include: SiC matrix length, SiC matrix width, SiC fiber radius r and the thickness of the PyC boundary layer h ;Establish a SiC / SiC composite material model based on the set dimensional parameters; In step 1.1, the length of the SiC matrix is 25 μm, the width of the SiC matrix is 25 μm, and the radius of the SiC fiber is r The thickness of the PyC boundary layer is 4 to 8 μm. h 50~550nm; Step 1.2, mesh division: Use quadrilateral elements to mesh the SiC / SiC composite material model built in step 1.1, and refine the mesh in the area where cracks may grow to make the calculation results more accurate. This completes the construction of the representative volume element model of the SiC fiber-reinforced composite material. Step 2: Phase field fracture simulation: Step 2.1, set material property parameters: The material property parameters include: Young's modulus of SiC fiber interface E 1. Young's modulus of SiC matrix interface E 2. Young's modulus of the PyC interface E 3. Strength of SiC fiber interface Ft 1. Strength of SiC fiber interface Ft 2. Strength of the PyC interface Ft 3. Critical energy release rate of SiC fiber interface Gc 1. Critical energy release rate at the SiC matrix interface Gc Critical energy release rate at the interface between 2 and PyC Gc 3. The above parameters are randomly selected within the value range; The material property parameters also include: Poisson's ratio of SiC fiber interface v 1 、 Poisson's ratio of SiC substrate interface v 2 and Poisson's ratio of the PyC interface v 3. Poisson's ratio is directly taken; After taking the values of each material property parameter, the maximum and minimum normalization processing is performed within the respective value range; Step 2.2, set boundary conditions: Set displacement constraints on the left and right sides of the model created in step 1.1; Step 2.3, phase-field fracture simulation: Set the phase field cohesion control equation and related parameters, set the partial differential equation solver to solve the phase field cohesion control equation, and set the related parametric equations and related material parameters; Step 2.4, output of transverse tensile strength: After the phase field fracture simulation, the model stress-strain curve was extracted and the transverse tensile strength was obtained from the stress-strain curve.

2. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 1, characterized in that: The method further comprises the steps of: Step 3: Construction and training of neural network: Step 3.1, dataset collection: Repeat step 2 N times to obtain N transverse tensile strengths, which constitute a forward neural network training set; N is an integer greater than or equal to 200 and less than or equal to 2000; Step 3.2, forward neural network training: For the forward neural network, the input is the thickness of the PyC boundary layer h、 SiC fiber radius r、 Young's modulus of SiC fiber interface E 1 、 Young's modulus of the SiC matrix interface E 2 、 Young's modulus of the PyC interface E 3 、 SiC fiber interface strength Ft 1 、 SiC fiber interface strength Ft 2. Strength of the PyC interface Ft 3 、 Critical energy release rate at SiC fiber interface Gc 1 、 Critical energy release rate of SiC matrix interface Gc 2 and the critical energy release rate of the PyC interface Gc 3; Output is transverse tensile strength; Step 3.3, obtain the optimal solution: PyC boundary layer thickness for forward neural networks h At different SiC fiber radii r The traversal of different material property parameters is carried out, and the optimal PyC thickness is obtained with the transverse tensile strength as the criterion. The results are used to collect the data set of the inverse neural network; Step 3.4, reverse neural network training: For the inverse neural network, the input is the SiC fiber radius r、 Young's modulus of SiC fiber interface E 1 、 Young's modulus of the SiC matrix interface E 2 、 Young's modulus of the PyC interface E 3 、 SiC fiber interface strength Ft 1 、 SiC fiber interface strength Ft 2. Strength of the PyC interface Ft 3 、 Critical energy release rate at SiC fiber interface Gc 1 、 Critical energy release rate of SiC matrix interface Gc 2 and the critical energy release rate of the PyC interface Gc 3. The output is the optimal PyC thickness. After training, the optimization model of the pyrolytic carbon interface in SiC / SiC composite materials is obtained.

3. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 1, characterized in that: In step 1.1, the SiC / SiC composite material model is a transverse tensile model containing a single fiber.

4. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 1, characterized in that: In step 1.2, when performing mesh division, the mesh size is set to 1 / 100 of the SiC substrate width in step 1.1; when performing mesh encryption, the mesh size is set to 1 / 500 to 1 / 400 of the SiC substrate width in step 1.

1.

5. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 1, characterized in that: In step 2.1, the range and value of the material property parameters are: Young's modulus of SiC fiber interface E 1,150 GPa≤ E 1≤430 GPa; Young's modulus of the SiC matrix interface E 2,150 GPa≤ E 2≤430 GPa; Young's modulus of the PyC interface E 3, 10 GPa≤ E 3≤30 GPa; SiC fiber interface strength Ft 1,300 MPa≤ Ft 1≤600 MPa; SiC fiber interface strength Ft 2,300 MPa≤ Ft 2≤600 MPa; Strength of the PyC interface Ft 3, 20 MPa≤ Ft 3≤60 MPa; Critical energy release rate at SiC fiber interface Gc 1,6 J / m 2 ≤ Gc 1≤10 J / m 2 ; Critical energy release rate of SiC matrix interface Gc 2,6 J / m 2 ≤ Gc 2≤10 J / m 2 ; Critical energy release rate at the PyC interface Gc 3,0.4 J / m 2 ≤ Gc 3≤0.6 J / m 2 ; Poisson's ratio of SiC fiber interface v 1, v 1 takes the value of 0.185; Poisson's ratio of SiC substrate interface v 2, v 2 takes the value of 0.185; Poisson's ratio of the PyC interface v 3. v 3The value is 0.

18.

6. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 1, characterized in that: In step 2.1, the material property parameters are normalized using Formula I, which is as follows: Formula I; Where: X Indicates the normalized value; A Indicates the current value; A max Indicates the maximum value among all values; A min Indicates the minimum value among all values.

7. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 1, characterized in that: In step 2.3, the phase field cohesion governing equation is as follows: Formula II; Where: represents the governing equation of phase field cohesion; represents the phase field value; represents the first derivative of the degradation function; The strain energy density at a point is The historical maximum value within the time period; represents the critical energy release rate; represents the first derivative of the crack geometry function; represents the proportionality constant; Represents the characteristic width of the phase field model.

8. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 2, characterized in that: In step 3.2, the input layer of the forward neural network has 11 parameters, the hidden layer has 5 layers, and each hidden layer contains 256 neurons; the output layer of the forward neural network has 1 parameter, and the ReLU activation function is used between layers to maintain the nonlinearity of the model. The Adam algorithm is used for model optimization. The initial learning rate of the Adam optimization algorithm is 1e-3, and the weight decay coefficient is set to 1e-5. The number of training times is 1000.

9. The SiC / SiC pyrolytic carbon interface optimization method based on neural network and phase field fracture method according to claim 2, characterized in that: In step 3.4, the input layer of the inverse neural network is 10 parameters, the hidden layer is 5 layers, each hidden layer contains 256 neurons, and the output layer is 1 parameter. The Sigmoid function is used as the activation function between layers to maintain the nonlinearity of the model. The Adam algorithm is used for model optimization, and the initial learning rate is set to 5e-3, and the number of training times is 1000.

Images