Bionic layered ceramic matrix composite bending strength prediction method and device and storage medium
By constructing a bending strength prediction model based on supervised learning algorithm, the problem of difficult to predict the bending strength of bionic layered ceramic matrix composites in the existing technology is solved, and accurate prediction and efficient design of material strength are achieved, reducing the cycle and cost of new materials research and development.
Patent Information
- Application Number
- CN202411884859.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2025-04-11
AI Technical Summary
The prior art is difficult to effectively predict the bending strength of bionic layered ceramic matrix composite materials, resulting in the problems of long research and development cycles and high costs of new materials.
By obtaining the initial sample data set, including bending strength parameters, structural parameters and molding parameters, a supervised learning algorithm is used to build a bending strength prediction model, and combining boxing operation and hyperparameter tuning to achieve accurate prediction of bending strength.
The bending strength of the material corresponding to the coupling of multiple factors can be obtained without design and R&D, which significantly accelerates the design of high-strength bionic layered ceramic matrix composite materials, and solves the problems of long research and development cycle and high cost of new materials.
Smart Images
Figure CN119808566A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method, a device and a storage medium for predicting the bending strength of a bionic layered ceramic-based composite material, and belongs to the technical field of design and development of ceramic-based composite materials. Background Art
[0002] Bionic ceramic matrix composites are important alternative materials in the fields of protective armor, medical equipment, aerospace, etc. due to their excellent characteristics such as light weight, high strength, high hardness and high impact resistance. Due to the diversity of its material composition, preparation method and structural characteristics, its mechanical properties are more designable than traditional bulk ceramics, and can meet design and application requirements such as super strength, super hardness or both strength and toughness.
[0003] In the past, scholars have conducted a lot of research on component optimization and preparation methods. In terms of component selection, there are many choices of ceramics for the substrate layer, including oxides, borides, nitrides and carbides. In order to achieve close interface bonding and appropriate residual stress, the interface layer material can be ceramic, or metal and polymer materials. According to different material compositions, people have successively proposed methods such as tape casting, freeze casting, thermal spraying, EPD electrophoretic deposition, LbL layer-by-layer self-assembly, and co-extrusion molding to obtain controllable or semi-controllable layered structures. Due to the differences in intra-layer density, interface bonding force and residual stress brought about by different preparation methods, the mechanical properties of the prepared layered materials, such as strength, hardness and toughness, show great differences.
[0004] Although there are many related theoretical and experimental studies, these studies are often based on extremely limited structural characteristics and have limited guiding role in the design of layered materials. In view of the complexity of the mechanical properties of layered materials caused by the coupling of multiple factors, the design and development of such materials still mainly rely on the "trial and error method", which is difficult to solve the problems of long development cycle and high cost of new materials, and limits the further development of ceramic-based layered composite materials. Summary of the invention
[0005] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a method, device and storage medium for predicting the flexural strength of a bionic layered ceramic-based composite material, which can obtain the flexural strength of the material corresponding to the coupling of multiple factors without the need for design and research and development, and can solve the problems of long research and development cycle and high cost of new materials in the prior art.
[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:
[0007] In a first aspect, the present invention provides a method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material, comprising the following steps:
[0008] Acquire an initial sample data set of the bionic layered ceramic matrix composite material, wherein each sample data in the initial sample data set includes a bending strength parameter, a structural parameter and a forming parameter of the bionic layered ceramic matrix composite material;
[0009] Input characteristic data is calculated based on structural parameters and molding parameters;
[0010] Based on the input feature data and the corresponding flexural strength parameters, flexural strength prediction models under different supervised learning algorithms are constructed. In the model training process, the flexural strength parameters are converted into flexural strength grades by using binning operation.
[0011] Controlling each flexural strength prediction model to output scores of multiple evaluation indicators according to the prediction situation, respectively calculating the comprehensive index scores of each flexural strength prediction model according to the scores of the multiple evaluation indicators, and taking the flexural strength prediction model corresponding to the highest comprehensive index score as the optimal flexural strength prediction model;
[0012] The real-time input characteristic data is calculated according to the real-time structural parameters and forming parameters, and then the real-time input characteristic data is input into the optimal bending strength prediction model to obtain the predicted bending strength grade.
[0013] The acquisition of the initial sample data set includes: from the CNKI document library and the SCI document library, by using bionic layered ceramic matrix composite materials and mechanical properties as keywords, obtaining material data of bionic layered ceramic matrix composite materials samples to form an initial sample data set; the initial sample data set at least includes bionic layered ceramic matrix composite materials of different systems such as ceramic-ceramic and ceramic-metal.
[0014] The structural parameters include substrate layer thickness, interface layer thickness, substrate-interface layer thickness ratio and relative number of layers; the molding parameters include elements and contents, molding method, molding temperature, molding pressure and molding time.
[0015] Supervised learning algorithms include NB, RF, DT, XGBoost, LightGBM, KNN, and ANN algorithms, which are called from the scikit-learn library in Python.
[0016] During the training process of the flexural strength prediction model, 70% of the input feature data and the corresponding flexural strength parameters were used as the training set, and the remaining 30% were used as the test set.
[0017] Before the bending strength prediction model outputs the scores of multiple evaluation indicators, hyperparameter tuning is performed as follows: the Scikit-learn library in Python is used to call GridSearch CV to tune the hyperparameters of the algorithm, and the hyperparameter search space is selected with the default parameters as the center.
[0018] The evaluation indicators include overall classification accuracy, recall rate, precision, F1 score and area under the receiver operating characteristic curve. In the comprehensive indicator score, the overall classification accuracy score of the test set has a weight of 30%, the overall classification accuracy score of the validation set has a weight of 30%, the recall score has a weight of 10%, the precision score has a weight of 10%, the F1 score has a weight of 10%, and the area under the receiver operating characteristic curve score has a weight of 10%.
[0019] Overall classification accuracy The calculation formula is: (1),
[0020] The calculation formula for recall is: (2),
[0021] The calculation formula of precision is: (3),
[0022] The F1 score calculation formula is: (4),
[0023] Among them, TP, TN, FP and FN are confusion matrix parameters, representing true positive, true negative, false positive and false negative respectively.
[0024] In a second aspect, the present invention provides a device for predicting the flexural strength of a biomimetic layered ceramic matrix composite material, comprising:
[0025] An initial sample data set acquisition module is used to acquire an initial sample data set of a bionic layered ceramic matrix composite material, wherein each sample data in the initial sample data set includes a bending strength performance parameter, a structural parameter and a forming parameter of the bionic layered ceramic matrix composite material;
[0026] An input feature conversion module is used to calculate input feature data based on structural parameters and molding parameters;
[0027] The flexural strength prediction model training module is used to construct a flexural strength prediction model under different supervised learning algorithms based on input feature data and corresponding flexural strength parameters, wherein a binning operation is used in the model training process to convert the flexural strength parameters into flexural strength grades;
[0028] The optimal bending strength prediction model determination module is used to control each bending strength prediction model to output scores of multiple evaluation indicators according to the prediction situation, calculate the comprehensive index scores of each bending strength prediction model according to the scores of multiple evaluation indicators, and take the bending strength prediction model corresponding to the highest comprehensive index score as the optimal bending strength prediction model;
[0029] The module for calculating the predicted bending strength grade is used to calculate the real-time input characteristic data according to the real-time structural parameters and forming parameters, and then input the real-time input characteristic data into the optimal bending strength prediction model to obtain the predicted bending strength grade.
[0030] In a third aspect, the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the method for predicting the flexural strength of a bionic layered ceramic-based composite material is implemented.
[0031] The beneficial effects of the present invention are as follows: The present invention provides a method, device and storage medium for predicting the flexural strength of a bionic layered ceramic-based composite material. Based on input feature data and corresponding flexural strength parameters, a flexural strength prediction model under different supervised learning algorithms is constructed. The influence of structural parameters and molding parameters on the strength of bionic layered ceramic-based composite materials is comprehensively considered. By introducing a machine learning algorithm, the flexural strength of bionic layered ceramic-based composite materials can be accurately predicted, thereby accelerating the design of high-strength bionic layered ceramic-based composite materials. The present invention can obtain the material flexural strength corresponding to multi-factor coupling without the need for design and research and development, and can solve the problems of long research and development cycle and high cost of new materials in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is a schematic flow chart of a method for predicting the flexural strength of a bionic layered ceramic-based composite material according to the present invention;
[0033] Figure 2 It is the ACC score diagram of each model in the present invention. DETAILED DESCRIPTION
[0034] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the protection scope of the present invention.
[0035] The present invention discloses a method for predicting the flexural strength of a bionic layered ceramic matrix composite material. The method aims to improve the strength of the layered ceramic matrix composite material, comprehensively considers the layered composite material based on oxide, boride, nitride and carbide ceramics, and uses a machine learning method to study the importance and influence of the structural characteristics of the layered composite material on the material strength, in order to provide guidance for the structural design of new layered ceramic composite materials, such as Figure 1 As shown, it mainly includes the following steps:
[0036] Step 1: obtaining an initial sample data set of the bionic layered ceramic matrix composite material, wherein each sample data in the initial sample data set includes bending strength parameters, structural parameters and forming parameters of the bionic layered ceramic matrix composite material.
[0037] Step 2, calculating input feature data based on structural parameters and molding parameters;
[0038] Step three, based on the input feature data and the corresponding flexural strength parameters, a flexural strength prediction model under different supervised learning algorithms is constructed, wherein the binning operation is used in the model training process to convert the flexural strength parameters into flexural strength grades.
[0039] Step 4: Control each flexural strength prediction model to output scores of multiple evaluation indicators according to the prediction situation, calculate the comprehensive index scores of each flexural strength prediction model according to the scores of multiple evaluation indicators, and take the flexural strength prediction model corresponding to the highest comprehensive index score as the optimal flexural strength prediction model.
[0040] Step 5: Calculate the real-time input characteristic data according to the real-time structural parameters and forming parameters, and then input the real-time input characteristic data into the optimal bending strength prediction model to obtain the predicted bending strength grade.
[0041] The present invention comprehensively considers the influence of structural parameters and molding parameters on the strength of bionic layered ceramic-based composite materials, and realizes accurate prediction of the flexural strength of bionic layered ceramic-based composite materials by introducing a machine learning algorithm, thereby accelerating the design of high-strength bionic layered ceramic-based composite materials. The present invention can obtain the material flexural strength corresponding to multi-factor coupling without the need for design and research and development, and can solve the problems of long research and development cycle and high cost of new materials in the prior art.
[0042] Example 2
[0043] like Figure 1 As shown, the present invention discloses a method for predicting the flexural strength of a bionic layered ceramic matrix composite material, comprising the following steps:
[0044] Step 1, obtaining an initial sample data set of bionic layered ceramic matrix composite materials, each sample data in the initial sample data set includes the bending strength performance parameters, structural parameters and molding parameters of the bionic layered ceramic matrix composite materials. The initial sample data set acquisition includes: from the HowNet document library and the SCI document library, by using bionic layered ceramic matrix composite materials and mechanical properties as keywords, obtaining the material data of the bionic layered ceramic matrix composite material samples to form the initial sample data set; the initial sample data set contains at least bionic layered ceramic matrix composite materials of different systems of ceramic-ceramic and ceramic-metal.
[0045] The initial sample data set directly affects the performance and generalization of the constructed model. To ensure the reliability of the data, the present invention uses "bionic layered ceramic-based composite materials" and "mechanical properties" as keywords to obtain data from published literature. It should be noted that the sample data of the obtained bionic layered ceramic-based composite materials include at least bionic layered ceramic-based composite materials of different systems such as ceramic-ceramic and ceramic-metal. The parameters closely related to the strength performance of bionic layered ceramic-based composite materials mainly include flexural strength, yield strength, tensile strength, etc. Considering the number of relevant samples, the present invention selects flexural strength as a parameter indicator to measure strength performance, and a total of 105 flexural strength data are collected.
[0046] The structural parameters of bionic layered ceramic matrix composites include the thickness of the matrix layer (h1), the thickness of the interface layer (h2), the thickness ratio of the matrix to the interface (Ratio=h1 / h2) and the relative number of layers (the ratio of the sample thickness h to the total number of matrix and interface layers l, i.e. h / l). The molding parameters include elements and contents (E&C), molding method (Tech), molding temperature (Temp), molding pressure (P) and molding time (t).
[0047] The structural parameters include the thickness of the substrate layer, the thickness of the interface layer, the layer thickness ratio of the substrate to the interface, and the relative number of layers; the molding parameters include elements and contents, molding method, molding temperature, molding pressure, and molding time.
[0048] Step 2: Calculate input feature data based on structural parameters and molding parameters.
[0049] Step three, based on the input feature data and the corresponding flexural strength parameters, a flexural strength prediction model under different supervised learning algorithms is constructed, wherein the binning operation is used in the model training process to transform the regression problem into a classification problem, and the flexural strength parameters are converted into flexural strength grades. Machine learning is good at capturing the connection directly from input parameters and performance parameters, and its unique training method can implicitly capture the mapping relationship between given input and output to achieve accurate prediction. Through machine learning, material data can be modeled to make up for the limitations of the research and development of high-strength bionic layered ceramic-based composite materials that rely on traditional experimental trial and error methods and complex domain knowledge.
[0050] The supervised learning algorithms in the present invention include NB (Naive Bayes), RF (Random Forest), DT (Decision Tree), XGBoost, LightGBM, KNN (K Nearest Neighbor Classification) and ANN (Artificial Neural Network) algorithms, which are called from the scikit-learn library in python. During the training of the flexural strength prediction model, 70% of the input feature data and the corresponding flexural strength parameters are used as a training set, and the remaining 30% are used as a test set.
[0051] Step 4: Control each flexural strength prediction model to output scores of multiple evaluation indicators according to the prediction situation, calculate the comprehensive index scores of each flexural strength prediction model according to the scores of multiple evaluation indicators, and take the flexural strength prediction model corresponding to the highest comprehensive index score as the optimal flexural strength prediction model.
[0052] Before the bending strength prediction model outputs the scores of multiple evaluation indicators, hyperparameter tuning is performed, as follows: the Scikit-learn library in Python is used to call GridSearch CV to perform hyperparameter tuning of the algorithm, and the hyperparameter search space is selected with the default parameters as the center. In this embodiment, the ANN algorithm hyperparameters used to construct the strength prediction model include Dense, Units1, Units2, Units3, Units4, Units5, and optimizer, and the search space is shown in Table 1:
[0053] Table 1 ANN algorithm hyperparameter search space
[0054]
[0055] The evaluation indicators include overall classification accuracy, recall rate, precision, F1 score and area under the receiver operating characteristic curve. In the comprehensive indicator score, the overall classification accuracy score of the test set has a weight of 30%, the overall classification accuracy score of the validation set has a weight of 30%, the recall score has a weight of 10%, the precision score has a weight of 10%, the F1 score has a weight of 10%, and the area under the receiver operating characteristic curve score has a weight of 10%.
[0056] Overall classification accuracy The calculation formula is: (1),
[0057] The calculation formula for recall is: (2),
[0058] The calculation formula of precision is: (3),
[0059] The F1 score calculation formula is: (4),
[0060] Among them, TP, TN, FP and FN are confusion matrix parameters, representing true positive, true negative, false positive and false negative respectively.
[0061] In the present invention, the ACC scores of each model are as follows: Figure 1 The scores of the remaining four indicators are shown in Table 2.
[0062] Table 2 The scores of the other four indicators of each model of the present invention
[0063]
[0064] Through the above weight calculation, in the present invention, the scores of each model are: ANN: 0.8737, KNN: 0.8677, LightGBM: 0.8363, XGBoost: 0.8395, DT: 0.8481, RF: 0.8297 and NB: 0.7684. It can be seen that the ANN model can perform the best prediction and is therefore selected as the optimal bending strength prediction model.
[0065] Step 6: Calculate the real-time input feature data based on the real-time structural parameters and forming parameters, and then input the real-time input feature data into the optimal bending strength prediction model to obtain the predicted bending strength grade. In addition, the optimal bending strength prediction model is hyperparameter tuned before use, as follows: Use the Scikit-learn library in Python to call GridSearch CV to perform hyperparameter tuning of the algorithm, and the hyperparameter search space is selected with the default parameters as the center.
[0066] The present invention is based on SHAP interpretable machine learning to analyze the importance of each feature under the constructed model and its influence law, which is used as the design result of high-strength bionic layered ceramic matrix composite materials to guide the subsequent design of high-strength bionic layered ceramic matrix composite materials.
[0067] Example 3
[0068] This embodiment discloses a device for predicting the bending strength of a biomimetic layered ceramic matrix composite material, comprising:
[0069] An initial sample data set acquisition module is used to acquire an initial sample data set of a bionic layered ceramic matrix composite material, wherein each sample data in the initial sample data set includes a bending strength performance parameter, a structural parameter and a forming parameter of the bionic layered ceramic matrix composite material;
[0070] An input feature conversion module is used to calculate input feature data based on structural parameters and molding parameters;
[0071] The flexural strength prediction model training module is used to construct a flexural strength prediction model under different supervised learning algorithms based on input feature data and corresponding flexural strength parameters, wherein a binning operation is used in the model training process to convert the flexural strength parameters into flexural strength grades;
[0072] The optimal bending strength prediction model determination module is used to control each bending strength prediction model to output scores of multiple evaluation indicators according to the prediction situation, calculate the comprehensive index scores of each bending strength prediction model according to the scores of multiple evaluation indicators, and take the bending strength prediction model corresponding to the highest comprehensive index score as the optimal bending strength prediction model;
[0073] The predicted bending strength grade calculation module is used to calculate the real-time input characteristic data according to the real-time structural parameters and forming parameters, and then input the real-time input characteristic data into the optimal bending strength prediction model to obtain the predicted bending strength grade.
[0074] Example 4
[0075] This embodiment discloses a computer-readable storage medium on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the method for predicting the flexural strength of a bionic layered ceramic-based composite material is implemented.
[0076] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0077] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0079] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material, characterized by: The following steps are involved: Acquire an initial sample data set of the bionic layered ceramic matrix composite material, wherein each sample data in the initial sample data set includes a bending strength parameter, a structural parameter and a forming parameter of the bionic layered ceramic matrix composite material; Input characteristic data is calculated based on structural parameters and molding parameters; Based on the input feature data and the corresponding flexural strength parameters, flexural strength prediction models under different supervised learning algorithms are constructed. In the model training process, the flexural strength parameters are converted into flexural strength grades by using binning operation. Controlling each flexural strength prediction model to output scores of multiple evaluation indicators according to the prediction situation, respectively calculating the comprehensive index scores of each flexural strength prediction model according to the scores of the multiple evaluation indicators, and taking the flexural strength prediction model corresponding to the highest comprehensive index score as the optimal flexural strength prediction model; The real-time input characteristic data is calculated according to the real-time structural parameters and forming parameters, and then the real-time input characteristic data is input into the optimal bending strength prediction model to obtain the predicted bending strength grade.
2. The method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material according to claim 1, characterized in that: The acquisition of the initial sample data set includes: from the CNKI document library and the SCI document library, by using bionic layered ceramic matrix composite materials and mechanical properties as keywords, obtaining material data of bionic layered ceramic matrix composite materials samples to form an initial sample data set; the initial sample data set at least includes bionic layered ceramic matrix composite materials of different systems such as ceramic-ceramic and ceramic-metal.
3. The method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material according to claim 1, characterized in that: The structural parameters include substrate layer thickness, interface layer thickness, substrate-interface layer thickness ratio and relative number of layers; the molding parameters include elements and contents, molding method, molding temperature, molding pressure and molding time.
4. The method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material according to claim 1, characterized in that: Supervised learning algorithms include NB, RF, DT, XGBoost, LightGBM, KNN, and ANN algorithms, which are called from the scikit-learn library in Python.
5. The method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material according to claim 1, characterized in that: During the training process of the flexural strength prediction model, 70% of the input feature data and the corresponding flexural strength parameters were used as the training set, and the remaining 30% were used as the test set.
6. The method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material according to claim 1, characterized in that: Before the bending strength prediction model outputs the scores of multiple evaluation indicators, hyperparameter tuning is performed as follows: the Scikit-learn library in Python is used to call GridSearch CV to tune the hyperparameters of the algorithm, and the hyperparameter search space is selected with the default parameters as the center.
7. The method for predicting the flexural strength of a biomimetic layered ceramic matrix composite material according to claim 1, characterized in that: The evaluation indicators include overall classification accuracy, recall rate, precision, F1 score and area under the receiver operating characteristic curve. In the comprehensive indicator score, the overall classification accuracy score of the test set has a weight of 30%, the overall classification accuracy score of the validation set has a weight of 30%, the recall score has a weight of 10%, the precision score has a weight of 10%, the F1 score has a weight of 10%, and the area under the receiver operating characteristic curve score has a weight of 10%.
8. The method for predicting the flexural strength of biomimetic layered ceramic matrix composite materials according to claim 7, characterized in that: Overall classification accuracy The calculation formula is: (1), The calculation formula for recall is: (2), The calculation formula of precision is: (3), The F1 score calculation formula is: (4), Among them, TP, TN, FP and FN are confusion matrix parameters, representing true positive, true negative, false positive and false negative respectively.
9. A device for predicting the flexural strength of a biomimetic layered ceramic matrix composite material, characterized by: include: An initial sample data set acquisition module is used to acquire an initial sample data set of a bionic layered ceramic matrix composite material, wherein each sample data in the initial sample data set includes a bending strength performance parameter, a structural parameter and a forming parameter of the bionic layered ceramic matrix composite material; An input feature conversion module is used to calculate input feature data based on structural parameters and molding parameters; The flexural strength prediction model training module is used to construct a flexural strength prediction model under different supervised learning algorithms based on input feature data and corresponding flexural strength parameters, wherein a binning operation is used in the model training process to convert the flexural strength parameters into flexural strength grades; The optimal bending strength prediction model determination module is used to control each bending strength prediction model to output scores of multiple evaluation indicators according to the prediction situation, calculate the comprehensive index scores of each bending strength prediction model according to the scores of multiple evaluation indicators, and take the bending strength prediction model corresponding to the highest comprehensive index score as the optimal bending strength prediction model; The predicted bending strength grade calculation module is used to calculate the real-time input characteristic data according to the real-time structural parameters and forming parameters, and then input the real-time input characteristic data into the optimal bending strength prediction model to obtain the predicted bending strength grade.
10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the method for predicting the flexural strength of a bionic layered ceramic matrix composite material as described in any one of claims 1 to 8 is implemented.