Component design method and device for titanium-based composite material, electronic equipment and storage medium

Abstract

The invention discloses a component design method and device of a titanium-based composite material, electronic equipment and a storage medium, the method and device are applied to the electronic equipment, specifically, multiple reinforcement elements are screened based on a pre-constructed machine learning model, and an optimal reinforcement with harmony meeting user requirements is obtained; first principle simulation is conducted on the optimal reinforcement and the titanium matrix based on a material performance simulation tool, and a simulation result is obtained and comprises material component data; and sample processing preparation is conducted based on the material component data, and the titanium-based composite material sample is obtained. According to the method, the reinforcement material having the most correlation for improving the performance of the titanium-based composite material is found out by utilizing the machine learning model, and then the screened reinforcement element is simulated and analyzed by utilizing the first principle so as to judge the interface bonding condition between the reinforcement element and the titanium matrix; the situation that the comprehensive mechanical property of the material is reduced due to large defects generated in the actual preparation process is avoided, and therefore the current complex and changeable performance requirements for the titanium-based composite material are met.

Description

Technical Field

[0001] This application relates to the field of titanium alloy technology, and more specifically, to a method, apparatus, electronic device, and storage medium for designing the composition of titanium-based composite materials. Background Technology

[0002] Titanium-based composite materials, with their unique and excellent properties such as high strength, low density, good corrosion resistance and high temperature stability, have broad application prospects in high-end manufacturing industries such as aerospace, defense and automobile manufacturing. They can be used to manufacture key components such as aircraft engine parts, aircraft structural parts and automobile engine parts to achieve lightweight and high performance of equipment, and improve energy utilization efficiency and overall performance.

[0003] However, the development of traditional titanium-based composite materials faces numerous challenges. On one hand, the performance of materials is influenced by a combination of factors, including the alloy composition of the titanium matrix, the type of reinforcing phase, preparation process parameters (such as temperature, pressure, and time), and microstructural characteristics (such as grain size and phase distribution). These factors exhibit complex nonlinear relationships, and traditional development methods often require extensive experimental trial and error, consuming significant time, manpower, material resources, and financial investment. Furthermore, they struggle to fully explore the material's performance potential, easily missing out on potential high-performance material systems. On the other hand, with continuous technological advancements, the performance requirements for titanium-based composite materials are becoming increasingly stringent. They not only require excellent mechanical properties but also meet specific physical and chemical performance requirements, such as high thermal conductivity, low coefficient of thermal expansion, and good biocompatibility, to adapt to the needs of specific application scenarios. Traditional development models struggle to quickly respond to these complex and ever-changing performance demands, severely hindering the large-scale application and performance improvement of titanium-based composite materials. Summary of the Invention

[0004] In view of this, this application provides a method, apparatus, electronic device and storage medium for designing titanium-based composite materials based on machine learning methods to meet the current complex and varied performance requirements of titanium-based composite materials.

[0005] To achieve the above objectives, the following solution is proposed:

[0006] A method for designing the composition of a titanium-based composite material, applied to electronic devices, the method comprising the following steps:

[0007] Based on a pre-built machine learning model, various enhancer elements are screened to obtain the best enhancer that meets the user's needs in terms of harmony;

[0008] First-principles simulations were performed on the optimal reinforcement and titanium matrix using a material property simulation tool to obtain simulation results, which include material composition data.

[0009] Based on the material composition data, a sample was processed and prepared to obtain a titanium-based composite material sample.

[0010] Optionally, the step of filtering various enhancer elements based on a pre-built machine learning model to obtain the optimal enhancer that satisfies user needs includes the following steps:

[0011] Based on the machine learning model, common enhancer elements are screened to obtain multiple candidate enhancer elements;

[0012] The plurality of candidate enhancer elements are sorted according to their importance;

[0013] One or more of the candidate enhancer elements that are ranked first are selected as the best enhancer.

[0014] Optionally, the optimal reinforcement and titanium matrix are subjected to first-principles simulations using a material property simulation tool to obtain simulation results, which include material composition data, and include the following steps:

[0015] The CASTEP module in Materials Studio, a first-principles quantum mechanics software, was used to simulate the ratio between the optimal reinforcing agent and the titanium matrix, and the simulation results were obtained.

[0016] Optionally, in the process of using the CASTEP module in Materials Studio, a first-principles quantum mechanics software, to sequentially simulate the ratio between the optimal reinforcement and the titanium matrix and obtain the simulation results:

[0017] The interaction between ions and valence electrons is replaced by an ultrasoft pseudopotential.

[0018] The generalized gradient approximation of the Perdew-Burke-Ernzerhof method is used to correct for electron-electron interactions and associated potentials;

[0019] Set the cutoff value to 280eV;

[0020] The self-consistent field solution was obtained using the Kohn-Sham equation and energy functionals until the total energy of the SCF converged to 4.5 × 10⁻⁶. - 6 eV / atom;

[0021] The material composition data is determined by simulating the bonding energy and coefficient of thermal expansion between each of the optimal reinforcements and the titanium matrix.

[0022] Optionally, the step of processing and preparing the sample based on the material composition data to obtain the titanium-based composite material sample includes the following steps:

[0023] Based on the material composition data, the optimal reinforcement is mixed with the titanium matrix in sequence, and then laser selective melting is performed to obtain the titanium-based composite material sample.

[0024] Optional steps may also be included:

[0025] A structured database is established by integrating current literature and existing experimental data. The structured database includes a data sample set, and the constructed deep neural network is trained based on the data sample set to obtain the machine learning model.

[0026] A composition design device for titanium-based composite materials, applied to electronic devices, the composition design device comprising:

[0027] The element filtering module is configured to filter multiple enhancer elements based on a pre-built machine learning model to obtain the best enhancer that meets the user's needs in terms of compatibility.

[0028] The composition simulation module is configured to perform first-principles simulations on the optimal reinforcement and titanium matrix based on a material property simulation tool to obtain simulation results, which include material composition data.

[0029] The preparation control module is configured to process and prepare samples based on the material composition data to obtain titanium-based composite material samples.

[0030] Optional, also includes:

[0031] The model building module is configured to establish a structured database by integrating current literature and existing experimental data. The structured database includes a data sample set, and the constructed deep neural network is trained based on the data sample set to obtain the machine learning model.

[0032] An electronic device includes at least one processor and a memory connected to the processor, wherein:

[0033] The memory is used to store computer programs or instructions;

[0034] The processor is used to execute the computer program or instructions to enable the electronic device to implement the component design method as described above.

[0035] A computer-readable storage medium is applied to an electronic device, the storage medium carrying one or more computer programs that can be executed by the electronic device, thereby enabling the electronic device to implement the composition design method as described above.

[0036] As can be seen from the above technical solution, this application discloses a method, apparatus, electronic device, and storage medium for designing the composition of titanium-based composite materials. This method and apparatus are applied to electronic devices, specifically involving the screening of various reinforcing elements based on a pre-built machine learning model to obtain the optimal reinforcing element that meets user requirements; performing first-principles simulations on the optimal reinforcing element and the titanium matrix using material performance simulation tools to obtain simulation results, including material composition data; and processing and preparing samples based on the material composition data to obtain titanium-based composite material samples. This application utilizes a machine learning model to identify the reinforcing material most relevant to improving the performance of titanium-based composite materials, and then uses first-principles simulation analysis to evaluate the interfacial bonding between the selected reinforcing elements and the titanium matrix. This avoids significant defects during actual preparation that could reduce the overall mechanical properties of the material, thereby meeting the current complex and varied performance requirements of titanium-based composite materials. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart illustrating a method for designing the composition of a titanium-based composite material according to an embodiment of this application.

[0039] Figure 2 This is a schematic diagram showing the binding energy and coefficient of thermal expansion of an embodiment of this application;

[0040] Figure 3 This is a schematic diagram of the LPBF laser forming process according to an embodiment of this application;

[0041] Figure 4 The specific morphology of titanium boride-reinforced titanium matrix composites prepared by the LPBF process is shown;

[0042] Figure 5 The specific morphology of the titanium carbide-reinforced titanium matrix composite material prepared by the LPBF process is shown;

[0043] Figure 6The specific morphology of the graphene-reinforced titanium-based composite material prepared by the LPBF process is shown.

[0044] Figure 7 The specific morphology of the graphene-titanium boride reinforced titanium matrix composite material prepared by the LPBF process is shown.

[0045] Figure 8 DOS diagram of titanium matrix reinforced by three reinforcing elements;

[0046] Figure 9 The graph shows the test results of the mechanical properties of titanium-based composite materials.

[0047] Figure 10 This is a schematic diagram illustrating the principle of a deep neural network.

[0048] Figure 11 This is a block diagram of a composition design device for a titanium-based composite material according to an embodiment of this application;

[0049] Figure 12 A block diagram of a component design apparatus for another titanium-based composite material according to an embodiment of this application;

[0050] Figure 13 This is a block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0051] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0052] The inventors of this application have discovered in practice that machine learning and first-principles simulation techniques have brought new opportunities for the research and development of titanium-based composite materials. Specifically, machine learning technology can establish quantitative models of the relationship between material composition, structure, and properties through learning and analysis of massive amounts of experimental and material performance data. This allows for rapid prediction of material performance under different conditions and the screening of material combinations with potential high performance. Compared to traditional experimental trial and error, machine learning offers a significant cost advantage in early-stage research. It can rapidly perform a large number of calculations and simulations in a virtual space, avoiding the costs associated with expensive experimental equipment purchases, material preparation, and complex testing processes. It can also perform preliminary screening of a large number of material systems in a short time, greatly narrowing down the scope requiring actual experimental verification, thereby effectively reducing research and development costs.

[0053] First-principles simulations, starting from the quantum mechanical level, accurately calculate various physicochemical properties of materials based on their atomic and electronic structures, providing a theoretical foundation for predicting material properties. They can deeply reveal the microscopic mechanisms behind material properties, providing more accurate physicochemical information for building machine learning models and improving the prediction accuracy of the models. By combining first-principles simulations with machine learning, the advantages of both can be fully utilized to achieve synergy. First-principles simulations provide high-precision theoretical data support for machine learning, while machine learning can quickly process and analyze large-scale simulation data, uncovering patterns and trends, further optimizing the calculation process and parameter selection of first-principles simulations, and improving simulation efficiency. Based on the above basic content, this application proposes the following embodiments.

[0054] Figure 1 This is a flowchart illustrating a method for designing the composition of a titanium-based composite material according to an embodiment of this application.

[0055] like Figure 1 As shown, the composition design method provided in this application is applied to electronic devices to design the components and content of each component of a titanium-based composite material to obtain a titanium-based composite material with satisfactory performance. This electronic device can be understood as a computer, server, or cloud platform with data computing and information processing capabilities. The composition design method specifically includes the following steps:

[0056] S1. Based on the machine learning module, the best enhancer is selected from a variety of enhancer elements.

[0057] Based on a pre-built machine learning model, various enhancer elements are screened to obtain the optimal enhancer that best meets user needs. The specific process is as follows:

[0058] First, a machine learning model is used to simulate and screen common reinforcing elements to select those with better compatibility in titanium-based composites. Then, the importance of each reinforcing element is ranked. Finally, the reinforcing elements with the highest ranking are determined as the best reinforcing elements.

[0059] In this invention, the machine learning model specifically employs a deep neural network (DNN) model. The DNN includes 3-5 hidden layers, with 34-256 neurons per layer. The activation function is ReLU or Sigmoid. The database setup typically involves the following steps: first, data cleaning to remove outliers and fill in missing values ​​using K-nearest neighbor imputation; then, feature standardization, with numerical features such as the atomic weight of the enhancer normalized using Z-score normalization; finally, feature parameters with a correlation greater than 0.3 with the target performance are selected using Pearson correlation coefficient and recursive feature elimination.

[0060] Through practical screening, this application identifies the most common reinforcements that are most relevant to performance improvement, namely TiC, TiB, Gr, SiC, B4C, Ti5Si3, and TiN.

[0061] S2. First-principles simulations were performed on the optimal reinforcement and titanium matrix.

[0062] That is, first-principles simulations are performed on the optimal reinforcement and titanium matrix using material property simulation tools to obtain the corresponding simulation results, which include at least material composition data.

[0063] The optimal reinforcer, selected through machine learning, will be verified through first-principles simulation. In practice, the CASTEP module of Materials Studio, a quantum mechanics software, will be used to simulate the optimal reinforcer with a titanium matrix, analyzing the interaction mechanism between the reinforcer and the titanium matrix, as well as the in-situ transformation retention, laying the groundwork for subsequent comprehensive mechanical property simulation.

[0064] In the simulation, an ultrasoft pseudopotential was used to replace the interaction between ions and valence electrons; the generalized gradient approximation (GGA) of the Perdew-Burke-Ernzerhof (PBE) method was used to correct for electron-electron interactions and associated potentials; the cutoff value was set to 280 eV; and a self-consistent field (SCF) solution was obtained using the Kohn-Sham (KS) equation and energy functionals until the total SCF energy converged to 4.5 × 10⁻⁶. -6 eV / atom. Finally, by simulating the binding energy and thermal expansion coefficient between multiple elements and the titanium matrix, such as... Figure 2 As shown, the binding energy priority order is Gr>TiC>TiB>TiB2>SiC>B4C>Ti5Si3, and the thermal expansion coefficient priority order is TiB>Gr>TiC>TiB2>Ti5Si3>B4C>SiC. This further narrows down the range to three reinforcing elements: TiB, Gr, and TiC. Therefore, this application prepares titanium-based composite materials reinforced by these three reinforcing elements and attempts to achieve synergistic reinforcement between them.

[0065] S3. Based on the material composition data, sample processing and preparation are carried out to obtain titanium-based composite material samples.

[0066] Selective laser melting and forming of titanium composite powder using LPBF technology involves selectively melting and forming titanium composite powder layer by layer using lasers to obtain high-performance titanium-based composite materials. The forming process is as follows: Figure 3 As shown.

[0067] In some embodiments of the present invention, the process parameters for selective laser melting forming include: powder spreading speed of 0.04 mm / s to 0.08 mm / s, laser power of 240 W to 300 W, scanning speed of 1800 mm / s to 2000 mm / s, and scanning spacing of 0.10 mm to 0.14 mm.

[0068] For example, the powder spreading speed can be one of 0.04 mm / s, 0.05 mm / s, 0.06 mm / s, 0.07 mm / s, or 0.08 mm / s, or any value within the above range. The laser power can be one of 240 W, 250 W, 260 W, 270 W, 280 W, 290 W, or 300 W, or any value within the above range. The scanning speed can be one of 1800 mm / s, 1900 mm / s, or 2000 mm / s, or any value within the above range. The scanning spacing can be one of 0.10 mm, 0.11 mm, 0.12 mm, 0.13 mm, or 0.14 mm, or any value within the above range. In some embodiments of the present invention, titanium-based composite materials are formed using LPBF technology, with the following process parameters: powder spreading speed of 0.06 mm / s, laser power of 280 W, scanning speed of 2000 mm / s, and scanning spacing of 0.12 mm. The following describes the preparation of titanium-based composite materials using TiB, Gr, and TiC as reinforcing elements and their combinations:

[0069] 1. Titanium diboride was used as a reinforcing phase source, reacting with the titanium matrix to generate a reinforcing phase. The titanium composite powder was then subjected to selective laser melting (LBL). The powder mixing was performed using high-energy ball milling at a speed of 200 r / min, a ball-to-powder ratio of 4:1, and a milling time of 6 h. To prevent oxidation of the powder during high-energy ball milling, an argon-sealed environment was used. The titanium diboride addition amounts were 0.25 wt%, 0.5 wt%, and 0.75 wt%. The specific LBL process parameters were: powder spreading speed of 0.06 mm / s, laser power of 280 W, scanning speed of 2000 mm / s, and scanning spacing of 0.12 mm.

[0070] Figure 4 The specific morphology of the titanium boride-reinforced titanium matrix composite prepared by the LPBF process is shown. At a reinforcement ratio of 0.5 wt%, the titanium boride-reinforced titanium matrix composite achieved a density of over 99%, a tensile strength of 1211 MPa, an elongation of 4.11%, and a Young's modulus of 127 GPa. Due to the low solid solubility of boron in titanium, the boron is distributed in a needle-like pattern near the grain boundaries, resulting in fine grains and no cracks.

[0071] 2. Graphite powder is used as a carbon source to react with a titanium matrix to generate a titanium carbide reinforcing phase. The graphite-titanium composite powder is then formed by selective laser melting. The powder mixing is performed using high-energy ball milling at a speed of 200 r / min, a ball-to-powder ratio of 4:1, and a milling time of 6 hours. To prevent oxidation of the powder during high-energy ball milling, an argon-sealed environment is used. The graphite powder addition amounts are 0.25 wt%, 0.5 wt%, 0.75 wt%, and 1.5 wt%. The specific process parameters are: powder spreading speed of 0.06 mm / s, laser power of 280 W, scanning speed of 2000 mm / s, and scanning spacing of 0.12 mm.

[0072] Figure 5 The specific morphology of the titanium carbide-reinforced titanium matrix composite prepared by the LPBF process is shown. With a reinforcement ratio of 0.25 wt%, the titanium boride-reinforced titanium matrix composite achieved a density of over 99%, a tensile strength of 1307 MPa, an elongation of 3.15%, and a Young's modulus of 131 GPa. Compared to titanium boride, titanium carbide exhibits a granular distribution within the titanium matrix, resulting in a higher overall tensile strength improvement in the titanium matrix composite, but with a significant reduction in plasticity.

[0073] 3. Using graphene as a reinforcing agent, graphene-titanium composite powder was selectively melted and shaped using laser. To reduce damage to the graphene lattice structure, low-energy ball milling was employed for powder mixing. The milling speed was 96 r / min, the ball-to-powder ratio was 4:1, and the milling time was 6 h. To avoid oxidation of the powder during high-energy ball milling, an argon-sealed environment was used. The graphene addition amounts were 0.25 wt%, 0.5 wt%, 0.75 wt%, and 1.5 wt%, respectively. The specific process parameters were: powder spreading speed of 0.06 mm / s, laser power of 280 W, scanning speed of 2000 mm / s, and scanning spacing of 0.12 mm.

[0074] Figure 6 The morphology of the graphene-reinforced titanium matrix composite prepared by the LPBF process is shown. With a reinforcement ratio of 1.0 wt%, the graphene-reinforced titanium matrix composite achieved a density of over 99%, a tensile strength of 1344 MPa, an elongation of 3.24%, and a Young's modulus of 141 GPa. Compared to titanium boride and titanium carbide, the graphene-reinforced titanium matrix composite exhibited the highest overall mechanical properties. Graphene was uniformly distributed within the titanium matrix, and some titanium carbide was formed at the boundaries. It is speculated that the superior performance of this composite material is due not only to the excellent properties of graphene itself but also to the synergistic effect of the graphene and titanium carbide.

[0075] 4. Using graphene and titanium diboride as reinforcements, titanium diboride-graphene-titanium composite powder was formed by selective laser melting (LFBF). To reduce lattice damage to graphene, a composite ball milling process was used for powder mixing. First, titanium diboride and titanium powder were mixed by high-energy ball milling, with process parameters as described in Example 1. Then, graphene was added to the composite powder. The ball milling speed was 96 r / min, the ball-to-powder ratio was 4:1, and the milling time was 6 h. To avoid oxidation of the powder during high-energy ball milling, an argon-sealed environment was used. The graphene addition amounts were 0.25 wt%, 0.5 wt%, 0.75 wt%, and 1.5 wt%, respectively, and the titanium diboride addition amount was the same as that of graphene. The specific LFBF process parameters were: powder spreading speed of 0.06 mm / s, laser power of 280 W, scanning speed of 2000 mm / s, and scanning spacing of 0.12 mm.

[0076] Figure 7 The morphology of the graphene-titanium boride reinforced titanium matrix composite prepared by the LPBF process is shown. With a reinforcement ratio of 1.0 wt%, the graphene-reinforced titanium matrix composite achieved a density of over 99%, a tensile strength of 1377 MPa, an elongation of 4.1%, and a Young's modulus of 143 GPa. Compared to titanium boride, titanium carbide, and graphene, the titanium boride-graphene composite reinforced titanium matrix composite exhibited the highest overall mechanical properties. Graphene was uniformly distributed within the titanium matrix, with some titanium carbide forming at the boundaries, while titanium boride was distributed in the matrix in a fine needle-like morphology. The two reinforced each other synergistically, improving the overall mechanical properties of the composite on one hand, and on the other hand, the addition of titanium boride inhibited crack initiation, passively inhibited crack propagation, and improved the plasticity of the composite.

[0077] Figure 8 The DOS diagram shows the titanium matrix reinforced with three reinforcing elements. Mechanical testing has confirmed that its performance meets user requirements; specific mechanical test results are as follows. Figure 9 As shown.

[0078] As can be seen from the above technical solution, this embodiment provides a method for designing the composition of titanium-based composite materials. This method is applied to electronic devices. Specifically, it involves screening various reinforcing elements based on a pre-built machine learning model to obtain the optimal reinforcing element that meets user needs; performing first-principles simulations on the optimal reinforcing element and the titanium matrix using material performance simulation tools to obtain simulation results, including material composition data; and processing and preparing samples based on the material composition data to obtain titanium-based composite material samples. This application utilizes a machine learning model to identify the reinforcing material most relevant to improving the performance of titanium-based composite materials, and then uses first-principles simulation analysis to evaluate the interfacial bonding between the selected reinforcing elements and the titanium matrix. This avoids significant defects in the actual preparation process that reduce the overall mechanical properties of the material, thereby meeting the current complex and varied performance requirements of titanium-based composite materials.

[0079] In summary, applying machine learning and first-principles simulation techniques to predictive research and development of high-performance titanium-based composite materials can not only effectively reduce trial-and-error costs in early-stage research, but also achieve closed-loop optimization of the model through feedback from actual samples. This has significant scientific and practical value and is expected to become a key technology for the future research and development of titanium-based composite materials, providing strong material support for the development of high-end manufacturing.

[0080] In one specific embodiment of this application, the machine learning model is constructed through the following operations.

[0081] In this application, to achieve efficient prediction of the interaction between the titanium-based composite reinforcement and the matrix, a deep neural network (DNN) is used as the core architecture of the machine learning model, such as... Figure 10 As shown. First, various structural parameters of the reinforcements in current titanium-based composite materials are systematically collected and organized into a database. These parameters include ionic radius, covalent radius, ionization energy, and electronic configuration, which are used as inputs to the model, while element diffusion and migration energies are used as outputs. The deep neural network consists of 3–5 hidden layers, with 34–256 neurons per layer, and uses ReLU or Sigmoid activation functions. The database is typically built using the following steps: first, data cleaning to remove outliers, and then filling in missing values ​​using K-nearest neighbor interpolation.

[0082] To ensure effective training and validation of the model, the collected dataset was allocated in a ratio of 8:2, 7:3, and 6:4, with a larger proportion of the data used for training the machine learning model and the remaining data used to validate its accuracy. During model training, Bayesian optimization was employed to adjust the model's hyperparameters, such as learning rate, batch size, and number of hidden layers. Furthermore, mean squared error was chosen as the loss function for the regression task, combined with L2 regularization. The dropout rate was also analyzed, with a value ranging from 0 to 0.5 to effectively prevent overfitting.

[0083] Meanwhile, to further verify the reliability of the model's predictions, this application utilized the CASTEP module in Materials Studio, a first-principles quantum mechanics software based on density functional theory, to simulate the interaction between the reinforcement and the material. During the simulation, detailed settings were implemented for various fundamental parameters: an ultrasoft pseudopotential was used to replace the interaction between ions and valence electrons; the generalized gradient approximation (GGA) of the Perdew-Burke-Ernzerhof (PBE) method was used to correct for electron-electron interactions and associated potentials; the cutoff value was set to 280–310 eV; and a self-consistent field (SCF) solution was obtained using the Kohn-Sham (KS) equation and energy functionals until the total SCF energy converged to 4.5–6.0 × 10⁻⁶ eV. -6 eV / atom. In addition, other factors that may affect model performance were also considered, such as data preprocessing methods (e.g., standardization or normalization) and the model's generalization ability.

[0084] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0085] Although the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous.

[0086] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0087] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer.

[0088] Figure 11 This is a block diagram of a component design device for a titanium-based composite material according to an embodiment of this application.

[0089] like Figure 11 As shown, the composition design device provided in this application is applied to an electronic device for designing the components and content of each component of a titanium-based composite material to obtain a titanium-based composite material with satisfactory performance. This electronic device can be understood as a computer, server, or cloud platform with data computing and information processing capabilities. Specifically, the composition design device includes an element screening module 10, a composition simulation module 20, and a preparation control module 30.

[0090] The element filtering module is used to select the best enhancer from a variety of enhancer elements based on the machine learning module.

[0091] Based on a pre-built machine learning model, various enhancer elements are screened to obtain the optimal enhancer that best meets user needs. The specific process is as follows:

[0092] First, a machine learning model is used to simulate and screen common reinforcing elements to select those with better compatibility in titanium-based composites. Then, the importance of each reinforcing element is ranked. Finally, the reinforcing elements with the highest ranking are determined as the best reinforcing elements.

[0093] In this invention, the machine learning model specifically employs a deep neural network (DNN) model. The DNN includes 3-5 hidden layers, with 34-256 neurons per layer. The activation function is ReLU or Sigmoid. The database setup typically involves the following steps: first, data cleaning to remove outliers and fill in missing values ​​using K-nearest neighbor imputation; then, feature standardization, with numerical features such as the atomic weight of the enhancer normalized using Z-score normalization; finally, feature parameters with a correlation greater than 0.3 with the target performance are selected using Pearson correlation coefficient and recursive feature elimination.

[0094] Through practical screening, this application identifies the most common reinforcements that are most relevant to performance improvement, namely TiC, TiB, Gr, SiC, B4C, Ti5Si3, and TiN.

[0095] The composition simulation module is used to perform first-principles simulations of the optimal reinforcement and titanium matrix.

[0096] That is, first-principles simulations are performed on the optimal reinforcement and titanium matrix using material property simulation tools to obtain the corresponding simulation results, which include at least material composition data.

[0097] The optimal reinforcer, selected through machine learning, will be verified through first-principles simulation. In practice, the CASTEP module of Materials Studio, a quantum mechanics software, will be used to simulate the optimal reinforcer with a titanium matrix, analyzing the interaction mechanism between the reinforcer and the titanium matrix, as well as the in-situ transformation retention, laying the groundwork for subsequent comprehensive mechanical property simulation.

[0098] In the simulation, an ultrasoft pseudopotential was used to replace the interaction between ions and valence electrons; the generalized gradient approximation (GGA) of the Perdew-Burke-Ernzerhof (PBE) method was used to correct for electron-electron interactions and associated potentials; the cutoff value was set to 280 eV; and a self-consistent field (SCF) solution was obtained using the Kohn-Sham (KS) equation and energy functionals until the total SCF energy converged to 4.5 × 10⁻⁶. -6 eV / atom. Finally, by simulating the binding energy and thermal expansion coefficient between multiple elements and the titanium matrix, such as... Figure 2As shown, the binding energy priority order is Gr>TiC>TiB>TiB2>SiC>B4C>Ti5Si3, and the thermal expansion coefficient priority order is TiB>Gr>TiC>TiB2>Ti5Si3>B4C>SiC. This further narrows down the range to three reinforcing elements: TiB, Gr, and TiC. Therefore, this application prepares titanium-based composite materials reinforced by these three reinforcing elements and attempts to achieve synergistic reinforcement between them.

[0099] The preparation control module is used to process and prepare samples based on material composition data to obtain titanium-based composite material samples.

[0100] Selective laser melting and forming of titanium composite powder using LPBF technology involves selectively melting and forming titanium composite powder layer by layer using lasers to obtain high-performance titanium-based composite materials. The forming process is as follows: Figure 3 As shown.

[0101] In some embodiments of the present invention, the process parameters for selective laser melting forming include: powder spreading speed of 0.04 mm / s to 0.08 mm / s, laser power of 240 W to 300 W, scanning speed of 1800 mm / s to 2000 mm / s, and scanning spacing of 0.10 mm to 0.14 mm.

[0102] For example, the powder spreading speed can be one of 0.04 mm / s, 0.05 mm / s, 0.06 mm / s, 0.07 mm / s, or 0.08 mm / s, or any value within the above range. The laser power can be one of 240 W, 250 W, 260 W, 270 W, 280 W, 290 W, or 300 W, or any value within the above range. The scanning speed can be one of 1800 mm / s, 1900 mm / s, or 2000 mm / s, or any value within the above range. The scanning spacing can be one of 0.10 mm, 0.11 mm, 0.12 mm, 0.13 mm, or 0.14 mm, or any value within the above range. In some embodiments of the present invention, titanium-based composite materials are formed using LPBF technology, with the following process parameters: powder spreading speed of 0.06 mm / s, laser power of 280 W, scanning speed of 2000 mm / s, and scanning spacing of 0.12 mm.

[0103] As can be seen from the above technical solution, this embodiment provides a composition design device for titanium-based composite materials. This device is applied to electronic devices. Specifically, it screens various reinforcing elements based on a pre-built machine learning model to obtain the optimal reinforcing element that meets user needs; it then performs first-principles simulations on the optimal reinforcing element and the titanium matrix using a material performance simulation tool to obtain simulation results, including material composition data; finally, it processes and prepares samples based on the material composition data to obtain titanium-based composite material samples. This application utilizes a machine learning model to identify the reinforcing material most relevant to improving the performance of titanium-based composite materials, and then uses first-principles simulation analysis to evaluate the interfacial bonding between the selected reinforcing elements and the titanium matrix. This avoids significant defects in the actual preparation process that reduce the overall mechanical properties of the material, thereby meeting the current complex and varied performance requirements of titanium-based composite materials.

[0104] In summary, applying machine learning and first-principles simulation techniques to predictive research and development of high-performance titanium-based composite materials can not only effectively reduce trial-and-error costs in early-stage research, but also achieve closed-loop optimization of the model through feedback from actual samples. This has significant scientific and practical value and is expected to become a key technology for the future research and development of titanium-based composite materials, providing strong material support for the development of high-end manufacturing.

[0105] In one specific embodiment of this application, a model building module 40 is also included, such as... Figure 12 As shown, this module is used to build the machine learning model through the following operations.

[0106] In this application, to achieve efficient prediction of the interaction between the titanium-based composite reinforcement and the matrix, a deep neural network (DNN) is used as the core architecture of the machine learning model, such as... Figure 10 As shown. First, various structural parameters of the reinforcements in current titanium-based composite materials are systematically collected and organized into a database. These parameters include ionic radius, covalent radius, ionization energy, and electronic configuration, which are used as inputs to the model, while element diffusion and migration energies are used as outputs. The deep neural network consists of 3–5 hidden layers, with 34–256 neurons per layer, and uses ReLU or Sigmoid activation functions. The database is typically built using the following steps: first, data cleaning to remove outliers, and then filling in missing values ​​using K-nearest neighbor interpolation.

[0107] To ensure effective training and validation of the model, the collected dataset was allocated in a ratio of 8:2, 7:3, and 6:4, with a larger proportion of the data used for training the machine learning model and the remaining data used to validate its accuracy. During model training, Bayesian optimization was employed to adjust the model's hyperparameters, such as learning rate, batch size, and number of hidden layers. Furthermore, mean squared error was chosen as the loss function for the regression task, combined with L2 regularization. The dropout rate was also analyzed, with a value ranging from 0 to 0.5 to effectively prevent overfitting.

[0108] Meanwhile, to further verify the reliability of the model's predictions, this application utilized the CASTEP module in Materials Studio, a first-principles quantum mechanics software based on density functional theory, to simulate the interaction between the reinforcement and the material. During the simulation, detailed settings were implemented for various fundamental parameters: an ultrasoft pseudopotential was used to replace the interaction between ions and valence electrons; the generalized gradient approximation (GGA) of the Perdew-Burke-Ernzerhof (PBE) method was used to correct for electron-electron interactions and associated potentials; the cutoff value was set to 280–310 eV; and a self-consistent field (SCF) solution was obtained using the Kohn-Sham (KS) equation and energy functionals until the total SCF energy converged to 4.5–6.0 × 10⁻⁶ eV. -6 eV / atom. In addition, other factors that may affect model performance were also considered, such as data preprocessing methods (e.g., standardization or normalization) and the model's generalization ability.

[0109] The units described in the embodiments of this disclosure can be implemented in software or in hardware. The name of a unit does not necessarily limit the unit itself; for example, the first acquisition unit can also be described as "a unit that acquires at least two Internet Protocol addresses".

[0110] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0111] Figure 13 This is a block diagram of an electronic device according to an embodiment of this application.

[0112] The following is for reference. Figure 13This document illustrates a structural diagram suitable for implementing the electronic device in the embodiments of this disclosure. The terminal device in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. This electronic device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this disclosure.

[0113] The electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from an input device 606 into a random access memory (RAM) 603. The RAM also stores various programs and data required for the operation of the electronic device. The processing unit, ROM, and RAM are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0114] Typically, the following devices can be connected to the I / O interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows the electronic device to communicate wirelessly or wiredly with other devices to exchange data. Although electronic devices with various devices are shown in the figures, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0115] This application also provides an embodiment of a computer-readable storage medium.

[0116] The aforementioned computer-readable storage medium is used in an electronic device and carries one or more computer programs. When these programs are executed by the electronic device, the device screens various reinforcing elements based on a pre-built machine learning model to obtain the optimal reinforcing element that satisfies user requirements. First-principles simulations are then performed on the optimal reinforcing element and the titanium matrix using a material performance simulation tool to obtain simulation results, including material composition data. Based on the material composition data, sample processing and preparation are then carried out to obtain a titanium-based composite material sample. This application utilizes a machine learning model to identify the reinforcing material most relevant to improving the performance of titanium-based composite materials. Then, first-principles simulation analysis is used to evaluate the interfacial bonding between the selected reinforcing elements and the titanium matrix, avoiding significant defects in the actual preparation process that could reduce the overall mechanical properties of the material. This meets the current complex and varied performance requirements for titanium-based composite materials.

[0117] It should be noted that the computer-readable medium described above in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0118] In this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0119] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0120] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0121] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0122] The technical solution provided by the present invention has been described in detail above. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for designing the composition of a titanium-based composite material, applied to electronic devices, characterized in that, The component design method includes the following steps: Based on a pre-built machine learning model, various enhancer elements are screened to obtain the best enhancer that meets the user's needs in terms of harmony; First-principles simulations were performed on the optimal reinforcement and titanium matrix using a material property simulation tool to obtain simulation results, which include material composition data. Based on the material composition data, a sample was processed and prepared to obtain a titanium-based composite material sample.

2. The component design method as described in claim 1, characterized in that, The process of filtering various enhancer elements based on a pre-built machine learning model to obtain the optimal enhancer that satisfies user needs includes the following steps: Based on the machine learning model, common enhancer elements are screened to obtain multiple candidate enhancer elements; The plurality of candidate enhancer elements are sorted according to their importance; One or more of the candidate enhancer elements that are ranked first are selected as the best enhancer.

3. The component design method as described in claim 1, characterized in that, The optimal reinforcement and titanium matrix are subjected to first-principles simulations using a material property simulation tool to obtain simulation results, which include material composition data and include the following steps: The CASTEP module in Materials Studio, a first-principles quantum mechanics software, was used to simulate the ratio between the optimal reinforcing agent and the titanium matrix, and the simulation results were obtained.

4. The component design method as described in claim 1, characterized in that, In the process of simulating the optimal ratio between the reinforcement and the titanium matrix using the CASTEP module in Materials Studio, a first-principles quantum mechanics software, the following simulation results were obtained: The interaction between ions and valence electrons is replaced by an ultrasoft pseudopotential. The generalized gradient approximation of the Perdew-Burke-Ernzerhof method is used to correct for electron-electron interactions and associated potentials; Set the cutoff value to 280eV; The self-consistent field solution was obtained using the Kohn-Sham equation and energy functionals until the total energy of the SCF converged to 4.5 × 10⁻⁶. -6 eV / atom; The material composition data is determined by simulating the bonding energy and coefficient of thermal expansion between each of the optimal reinforcements and the titanium matrix.

5. The component design method as described in claim 1, characterized in that, The process of preparing a titanium-based composite material sample based on the material composition data includes the following steps: Based on the material composition data, the optimal reinforcement is mixed with the titanium matrix in sequence, and then laser selective melting is performed to obtain the titanium-based composite material sample.

6. The component design method according to any one of claims 1 to 5, characterized in that, It also includes the following steps: A structured database is established by integrating current literature and existing experimental data. The structured database includes a data sample set, and the constructed deep neural network is trained based on the data sample set to obtain the machine learning model.

7. A composition design device for titanium-based composite materials, applied to electronic devices, characterized in that, The component design device includes: The element filtering module is configured to filter multiple enhancer elements based on a pre-built machine learning model to obtain the best enhancer that meets the user's needs in terms of compatibility. The composition simulation module is configured to perform first-principles simulations on the optimal reinforcement and titanium matrix based on a material property simulation tool to obtain simulation results, which include material composition data. The preparation control module is configured to process and prepare samples based on the material composition data to obtain titanium-based composite material samples.

8. The component design apparatus as described in claim 7, characterized in that, Also includes: The model building module is configured to establish a structured database by integrating current literature and existing experimental data. The structured database includes a data sample set, and the constructed deep neural network is trained based on the data sample set to obtain the machine learning model.

9. An electronic device, characterized in that, The electronic device includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs or instructions; The processor is used to execute the computer program or instructions to enable the electronic device to implement the component design method as described in any one of claims 1 to 6.

10. A computer-readable storage medium applied to an electronic device, the storage medium carrying one or more computer programs that can be executed by the electronic device to enable the electronic device to implement the composition design method as described in any one of claims 1 to 6.

Images