Graphene aluminum alloy performance optimization simulation method and system based on digital twinning

Abstract

The invention provides a graphene aluminum alloy performance optimization simulation method and system based on digital twinning, and relates to the technical field of digital twinning, the method comprises the following steps: collecting electrical parameters, thermal parameters and mechanical parameters of aluminum alloy samples under different graphene contents; performing performance feature extraction on the electrical parameters, the thermal parameters and the mechanical parameters to obtain electrical performance features, thermal performance features and mechanical performance features; constructing a digital twinborn model, performing integrated solution according to the electrical performance characteristics, the thermal performance characteristics and the mechanical performance characteristics, and outputting simulation performance data; and analyzing the simulation performance data, and proposing a structure optimization suggestion according to the generated performance evaluation. According to the invention, the technical problem of poor digital simulation efficiency of the graphene aluminum alloy in the prior art can be solved, and the technical effect of improving the digital simulation efficiency is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital twinning, in particular to a graphene aluminum alloy performance optimization simulation method and system based on digital twinning. BACKGROUND

[0002] With the development of new material technology, graphene reinforced aluminum alloy gradually becomes a key material in the fields of aerospace, electronic devices and high-end manufacturing due to its excellent electrical conductivity, thermal conductivity and mechanical properties.

[0003] At present, the existing aluminum alloy performance optimization method mainly relies on empirical design or single numerical simulation means, which can only predict and analyze the single performance index of the material, and lacks a comprehensive understanding of the multi-physical field linkage relationship of electricity, heat and mechanics. In addition, the existing method has limitations in processing technology scheme generation, simulation verification and deviation compensation, and cannot realize closed-loop optimization from microstructure design to actual process control, and it is also difficult to dynamically reflect the comprehensive influence of different graphene content, grain size distribution and interface bonding state on material performance.

[0004] In summary, the existing technology lacks the ability of systematic modeling, digital simulation and closed-loop process optimization of the multi-physical field performance of graphene aluminum alloy, which leads to the technical problem that the material performance cannot meet the multi-objective requirements of electrical conductivity, thermal conductivity and mechanical strength. SUMMARY

[0005] The purpose of the present application is to provide a graphene aluminum alloy performance optimization simulation method and system based on digital twinning, to solve the technical problem that the material performance cannot meet the multi-objective requirements of electrical conductivity, thermal conductivity and mechanical strength due to the lack of systematic modeling, digital simulation and closed-loop process optimization ability of the multi-physical field performance of graphene aluminum alloy in the prior art.

[0006] In view of the above problems, the present application provides a graphene aluminum alloy performance optimization simulation method and system based on digital twinning.

[0007] In the first aspect, the present application provides a graphene aluminum alloy performance optimization simulation method based on digital twinning, which is realized by a graphene aluminum alloy performance optimization simulation system based on digital twinning, comprising: collecting electrical, thermal and mechanical parameters of aluminum alloy samples under different graphene contents; extracting performance characteristics of the electrical, thermal and mechanical parameters to obtain electrical, thermal and mechanical performance characteristics; constructing a digital twinning model, integrating solving according to the electrical, thermal and mechanical performance characteristics, and outputting simulation performance data; analyzing the simulation performance data and proposing structure optimization suggestions according to the generated performance evaluation.

[0008] Preferably, the graphene-aluminum alloy performance optimization simulation method based on digital twinning further comprises: data preprocessing of the electrical parameters, the thermal parameters and the mechanical parameters, wherein the data preprocessing comprises time synchronization, denoising, missing value compensation and data consistency verification; fitting and characterization of the preprocessed electrical parameters to obtain temperature dependence and strain dependence characterization of electrical performance, and generation of the electrical performance characteristics through dimension reduction; transient thermal response analysis and feature extraction of the preprocessed thermal parameters to determine temperature rise, thermal diffusion and thermal response time sequence characteristics under thermal excitation, and generation of the thermal performance characteristics through dimension reduction; decomposition and characterization of the preprocessed mechanical parameters to extract elastic response, yield behavior and nonlinear plastic deformation characteristics, and generation of the mechanical performance characteristics through dimension reduction.

[0009] Preferably, the graphene-aluminum alloy performance optimization simulation method based on digital twinning further comprises: obtaining three-dimensional microstructure information according to microstructure characterization data of the aluminum alloy matrix; embedding spatial distribution and interface properties of graphene particles according to the three-dimensional microstructure information to generate a three-dimensional virtual structure model, wherein the spatial distribution and interface properties include different grain sizes, different porosities and different interface bonding states of the same graphene content; correlating the electrical performance characteristics, the thermal performance characteristics and the mechanical performance characteristics in the three-dimensional virtual structure model, and performing grid division and discretization on the three-dimensional virtual structure model to specify material properties and interface parameters based on multi-dimensional feature mapping for discrete elements to obtain a discretized model; introducing boundary conditions and initial field distribution in the discretized model to obtain the digital twinning model, wherein the boundary conditions and the initial field distribution are generated based on corresponding experimental conditions.

[0010] Preferably, the graphene-aluminum alloy performance optimization simulation method based on digital twinning further comprises: calculating a thermal input boundary based on the electrical performance characteristics according to the three-dimensional virtual structure model; taking the thermal input boundary as a thermal field boundary condition, calculating thermal stress and thermal deformation based on the calculated temperature change to obtain mechanical field response; correlating the mechanical field response with the electrical parameters to generate a multi-field linkage feature matrix.

[0011] Preferably, the graphene-aluminum alloy performance optimization simulation method based on digital twinning further comprises: dimensionality reduction processing of the multi-field linkage feature matrix to extract dominant influence factors; constructing a performance response surface model based on the dominant influence factors; performing sensitivity analysis on the performance response surface model to determine key design parameters; establishing a multi-objective optimization function based on the key design parameters, taking electrical conductivity performance, thermal conductivity performance and strength indicators as optimization objectives; dynamically updating the key design parameters using an iterative optimization algorithm to output an optimized parameter set.

[0012] Preferably, the graphene aluminum alloy performance optimization simulation method based on digital twin further includes: inputting the optimization parameter set into the digital twin model and re-solving the multiphysics field; obtaining the updated electric field, thermal field, and mechanical field results, and generating an updated multi-field linkage feature matrix; comparing the differences between the old and new feature matrices of the multi-field linkage feature matrix and the updated multi-field linkage feature matrix, and calculating the performance improvement rate; evaluating whether the optimization effect meets the preset convergence condition based on the performance improvement rate; if not, adjusting the iteration step size and weight coefficients, entering the next round of optimization loop, until the condition is met, and outputting the simulated performance data.

[0013] Preferably, the graphene aluminum alloy performance optimization simulation method based on digital twin further includes: recording the parameter change trajectory of each iteration during the optimization cycle, and constructing a time-series correlation model of parameter evolution sequence and performance response sequence; identifying key change nodes based on the time-series correlation model and judging the performance improvement trend; freezing the parameter set and generating an optimized digital twin model when the performance improvement tends to stabilize; and comparing the predicted performance of the optimized digital twin model with the experimental results to obtain the verification comparison results.

[0014] Preferably, the simulation method for optimizing the performance of graphene aluminum alloy based on digital twins further includes: correcting the parameters of the twin model based on the verification comparison results; adjusting the weights of the multi-field coupling parameters in the optimized digital twin model using a backpropagation mechanism, and re-executing the integrated solution to obtain the corrected simulation performance data; inputting the corrected simulation performance data into the performance evaluation module to calculate the comprehensive performance score; and automatically generating the structural optimization suggestions based on the comprehensive performance score.

[0015] Preferably, the graphene aluminum alloy performance optimization simulation method based on digital twin further includes: generating a process adjustment scheme based on the structural optimization suggestions and converting it into executable process control instructions; simulating the execution of the process control instructions in a virtual machining environment and identifying potential machining deviations based on the simulation results; compensating the control instructions for parameters based on the deviation analysis results, outputting the corrected process control instructions, and updating the control module of the digital twin model.

[0016] Secondly, this application also provides a graphene-aluminum alloy performance optimization simulation system based on digital twins, used to execute the graphene-aluminum alloy performance optimization simulation method based on digital twins as described in the first aspect, including: a parameter acquisition module for acquiring electrical, thermal, and mechanical parameters of aluminum alloy samples with different graphene contents; a feature extraction module for extracting performance features from the electrical, thermal, and mechanical parameters to obtain electrical performance features, thermal performance features, and mechanical performance features; a data output module for constructing a digital twin model, performing integrated solution based on the electrical, thermal, and mechanical performance features, and outputting simulated performance data; and a suggestion proposal module for analyzing the simulated performance data and proposing structural optimization suggestions based on the generated performance evaluation.

[0017] The technical solution provided in this application has at least the following technical effects or advantages: by achieving the technical goal of integrated optimization and closed-loop simulation control of multi-physics field performance of graphene aluminum alloy based on digital twin, it can quickly predict the comprehensive performance of materials, dynamically adjust key design parameters and generate optimized process schemes under different graphene contents, grain sizes and interface bonding states, thereby significantly improving the electrical conductivity, thermal conductivity and mechanical strength of materials.

[0018] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in 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 merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the simulation method for performance optimization of graphene aluminum alloys based on digital twins, as described in this application.

[0021] Figure 2 This is a schematic diagram of the structure of the graphene aluminum alloy performance optimization simulation system based on digital twins in this application.

[0022] Explanation of reference numerals in the attached diagram: Parameter acquisition module 1, Feature acquisition module 2, Data output module 3, Suggestion proposal module 4. Detailed Implementation

[0023] This application provides a digital twin-based method and system for optimizing the performance of graphene-aluminum alloys. It addresses the technical problem in existing technologies where the lack of systematic modeling, digital simulation, and closed-loop process optimization capabilities for the multi-physics properties of graphene-aluminum alloys makes it difficult to simultaneously meet multiple performance targets, including electrical conductivity, thermal conductivity, and mechanical strength. The application achieves the technical goal of integrated optimization and closed-loop simulation control of the multi-physics properties of graphene-aluminum alloys based on digital twins. This enables rapid prediction of the material's comprehensive performance under different graphene contents, grain sizes, and interface bonding states, dynamic adjustment of key design parameters, and generation of optimized process schemes, thereby significantly improving the material's electrical conductivity, thermal conductivity, and mechanical strength.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a simulation method for performance optimization of graphene aluminum alloys based on digital twins, which is applied to a simulation system for performance optimization of graphene aluminum alloys based on digital twins. The method specifically includes the following steps: S1: Collect electrical, thermal, and mechanical parameters of aluminum alloy samples with different graphene contents.

[0026] Specifically, collecting electrical parameters of aluminum alloy samples with different graphene contents refers to measuring the conductivity characteristics, such as conductivity, resistivity, and their response to temperature and strain, of alloy samples prepared by mixing graphene with an aluminum matrix at different doping ratios using electrical measuring instruments. Electrical parameters reflect the ability of electrons to migrate within the material and are important criteria for judging the electrical stability and conductivity efficiency of composite materials.

[0027] Meanwhile, collecting thermal parameters refers to measuring the thermal conductivity, specific heat capacity, and thermal diffusion characteristics of a material under heated conditions using thermal analysis methods such as laser scintillation, steady-state methods, or differential scanning calorimetry. Thermal parameters characterize a material's ability to transfer and absorb heat, and are crucial for evaluating the impact of graphene doping on the thermal stability and heat dissipation performance of aluminum alloys.

[0028] Furthermore, collecting mechanical parameters refers to obtaining data such as the stress-strain relationship, elastic modulus, yield strength, and elongation at break of a material under external force through tensile, compression, bending, or fatigue tests. These mechanical parameters reveal the influence of graphene on the microstructure reinforcement and interfacial bonding of aluminum-based composites, reflecting the overall mechanical load-bearing capacity and deformation characteristics.

[0029] S2: Extract performance characteristics from the electrical, thermal, and mechanical parameters to obtain electrical, thermal, and mechanical performance characteristics.

[0030] Furthermore, this application also includes: preprocessing the electrical parameters, thermal parameters, and mechanical parameters, wherein the data preprocessing includes time synchronization, noise reduction, missing value compensation, and data consistency verification; fitting and characterizing the preprocessed electrical parameters to obtain temperature dependence and strain dependence characteristics of electrical performance, and generating the electrical performance features through dimensional reduction; performing transient thermal response analysis and feature extraction on the preprocessed thermal parameters to determine the time series characteristics of temperature rise, thermal diffusion, and thermal response under thermal excitation, and generating the thermal performance features through dimensional reduction; decomposing and characterizing the preprocessed mechanical parameters to extract elastic response, yield behavior, and nonlinear plastic deformation characteristics, and generating the mechanical performance features through dimensional reduction.

[0031] Specifically, data preprocessing for electrical, thermal, and mechanical parameters involves performing a series of standardized operations on the raw experimental data after acquisition to eliminate noise, missing values, and time asynchrony, enabling comparison of different types of data under a unified time reference and unit system. Time synchronization refers to aligning data from different sensors or experimental devices to the same time scale to ensure the comparability of electrical, thermal, and mechanical data under the same physical conditions. Noise reduction removes random interference or systematic errors through filtering or signal decomposition algorithms to preserve the main trend signals. Missing value compensation uses methods such as interpolation, regression, or neighborhood averaging to fill data gaps and ensure data continuity. Data consistency verification is used to check for abnormal deviations in data from different experimental channels or devices.

[0032] Fitting and characterizing preprocessed electrical parameters refers to using mathematical models to describe their relationship with temperature and strain based on processed experimental data such as conductivity and resistivity. Temperature dependence characterization represents the law of electrical performance changing with temperature; for example, as temperature increases, the conductivity of a metal may decrease due to increased carrier scattering. Strain dependence characterization represents the characteristics of electrical performance affected by mechanical deformation, such as the change in resistance caused by changes in the crystal structure during stretching. After obtaining the laws through polynomial fitting or exponential models, dimensionality reduction algorithms, such as principal component analysis or linear discriminant analysis, can be used to extract the most representative feature vectors, thereby generating electrical performance characteristics.

[0033] Transient thermal response analysis and feature extraction of pre-processed thermal parameters refers to analyzing the temperature changes and heat transfer behavior of materials under dynamic thermal loading conditions. Transient thermal response describes the heat conduction process of a material over a short period after heating; time-series analysis can yield the rate of temperature rise, thermal diffusivity, and temperature recovery characteristics. The temperature rise under thermal excitation reflects the material's heat capacity, thermal diffusivity characterizes the efficiency of heat transfer within the material, and the time-series characteristics of the thermal response reveal the system's stability and delay effects over time.

[0034] The decomposition and characterization of pretreated mechanical parameters refers to the regional division and characteristic analysis of the stress-strain curves obtained from mechanical experiments. The elastic response describes the behavior of the material in the reversible deformation stage, and its slope corresponds to the elastic modulus; the yield behavior reflects the critical point of the material's transition from the elastic stage to the plastic stage and is a key indicator for measuring structural stability; the nonlinear plastic deformation characteristics describe the irreversible deformation law of the material under large strain, reflecting the micro-interface bonding and graphene reinforcement.

[0035] S3: Construct a digital twin model, solve it in an integrated manner based on the electrical performance characteristics, thermal performance characteristics and mechanical performance characteristics, and output simulated performance data.

[0036] Furthermore, this application also includes: obtaining three-dimensional microscopic information based on the microscopic characterization data of the aluminum alloy matrix; embedding the spatial distribution and interface properties of graphene particles into the three-dimensional microscopic information to generate a three-dimensional virtual structure model, wherein the spatial distribution and interface properties include different grain sizes, different porosities, and different interface bonding states with the same graphene content; associating the electrical performance characteristics, the thermal performance characteristics, and the mechanical performance characteristics in the three-dimensional virtual structure model, and performing meshing and discretization on the three-dimensional virtual structure model, specifying material properties and interface parameters based on multidimensional feature mapping for the discrete elements to obtain a discretized model; introducing boundary conditions and initial field distribution into the discretized model to obtain the digital twin model, wherein the boundary conditions and initial field distribution are generated based on corresponding experimental conditions.

[0037] Furthermore, this application also includes: calculating the thermal input boundary based on the three-dimensional virtual structural model and the electrical performance characteristics; using the thermal input boundary as the thermal field boundary condition, calculating the thermal stress and thermal deformation from the solved temperature changes to obtain the mechanical field response; and performing correlation analysis between the mechanical field response and the electrical parameters to generate a multi-field linkage feature matrix.

[0038] Furthermore, this application also includes: performing dimensionality reduction processing on the multi-field linkage feature matrix to extract the dominant influencing factors; constructing a performance response surface model based on the dominant influencing factors; performing sensitivity analysis on the performance response surface model to determine key design parameters; establishing a multi-objective optimization function based on the key design parameters, with electrical conductivity, thermal conductivity, and strength indicators as optimization objectives; and dynamically updating the key design parameters using an iterative optimization algorithm to output an optimized parameter set.

[0039] Furthermore, this application also includes: inputting the optimized parameter set into the digital twin model and re-solving the multiphysics field; obtaining the updated electric field, thermal field, and mechanical field results, and generating an updated multiphysics linkage feature matrix; comparing the differences between the old and new feature matrices of the multiphysics linkage feature matrix and the updated multiphysics linkage feature matrix, and calculating the performance improvement rate; evaluating whether the optimization effect meets the preset convergence condition based on the performance improvement rate; if not, adjusting the iteration step size and weight coefficients, entering the next round of optimization loop, until the condition is met, and outputting the simulated performance data.

[0040] Furthermore, this application also includes: recording the parameter change trajectory of each iteration during the optimization loop, and constructing a time-series correlation model of parameter evolution sequence and performance response sequence; identifying key change nodes based on the time-series correlation model and judging the performance improvement trend; freezing the parameter set and generating an optimized digital twin model when the performance improvement tends to stabilize; and comparing the prediction performance of the optimized digital twin model with the experimental results to obtain the verification comparison results.

[0041] Specifically, three-dimensional microscopic information is obtained from the microscopic characterization data of the aluminum alloy matrix. This involves using techniques such as microscopy, X-ray tomography, or electron backscatter diffraction to obtain detailed information about the geometric morphology, size distribution, and interconnections of the grains within the material. Three-dimensional microscopic information refers to multidimensional data describing the spatial morphology of the material's internal microstructure. It includes not only the material's spatial geometric distribution but also reflects microscopic features such as grain boundaries, phase boundaries, pores, and inclusions. By acquiring three-dimensional microscopic information, a digital foundation reflecting the true microstructure can be constructed, providing a basis for subsequent model building and performance mapping.

[0042] Secondly, a three-dimensional virtual structural model is generated by embedding the spatial distribution and interface properties of graphene particles into the three-dimensional microstructure information. Graphene particles are embedded into the known microstructure of aluminum alloys according to certain rules and probability distributions, so that the model retains the basic geometric structure of the aluminum matrix while introducing the distribution characteristics of the graphene reinforcing phase. Spatial distribution refers to the arrangement and density pattern of graphene particles in three-dimensional space, while interface properties describe the bonding characteristics between graphene and the aluminum matrix, such as interfacial bonding strength, thermal conductivity, or charge migration ability. With the same graphene content, multiple different virtual structural instances can be generated by adjusting the grain size, porosity, and interfacial bonding state to study the influence of different microscopic parameters on macroscopic performance. For example, when the porosity increases from 2% to 5%, even if the graphene content remains unchanged, the overall thermal conductivity will significantly decrease.

[0043] Next, the electrical, thermal, and mechanical properties are correlated and expressed in the 3D virtual structural model. This involves assigning corresponding material properties to each spatial unit or node of the virtual structure, enabling the model to simultaneously reflect the coupled behavior of multiple electrical, thermal, and mechanical fields. The extracted feature parameters are mapped to the model's spatial units, thus reflecting the differences in characteristics across different regions during calculation. Then, the 3D virtual structural model is meshed and discretized, decomposing the continuous 3D model into a large number of small finite elements for solution using methods such as finite element analysis. The discrete elements are assigned material properties and interface parameters under multi-dimensional feature mappings. Each element contains various information such as electrical conductivity, thermal diffusivity, and elastic modulus, enabling it to play an independent role in coupled field analysis.

[0044] Finally, boundary conditions and initial field distributions are introduced into the discretized model to obtain a digital twin model. Specific constraints and initial states, such as temperature distribution, load magnitude, external electric field, or heat flux boundaries, are applied to the discretized virtual structure. Boundary conditions determine the response under the influence of the external environment, while the initial field distribution describes the model's state at the start of the simulation, such as the initial temperature field or stress field. Boundary conditions and initial field distributions are generated based on actual experimental conditions to ensure physical consistency of the model's calculation results. For example, when a stable heat flux boundary condition is applied in the experiment, the digital twin model should also apply the same heat flux boundary condition, allowing the virtual simulation results to be compared and verified with real experimental results.

[0045] Furthermore, based on the three-dimensional virtual structural model, the heat input boundary is calculated according to the electrical performance characteristics, that is, the thermal effect generated under the action of current or electric field is derived using the electrical characteristics of the material. The three-dimensional virtual structural model refers to a digital model constructed in a computer that includes the actual spatial distribution of the material, interface properties, and multi-physics parameters. The electrical performance characteristics include information such as conductivity, resistivity, and current density distribution, which are used to reflect the ability of current to conduct through the material. When current flows through the material, Joule heating is generated due to the resistance effect, and the spatial distribution of Joule heating determines the heat input boundary, that is, the heat flow input at various locations in the model.

[0046] Next, the heat input boundary is used as the thermal field boundary condition. The obtained temperature changes are then used to calculate thermal stress and thermal deformation, thus obtaining the mechanical field response. The thermal field boundary condition refers to the constraint conditions on temperature distribution or heat flux in the model, which determines the temperature gradient change of the material during heating. When the heat input is applied as the thermal boundary to the model, the temperature variation with time and space can be obtained by solving the heat conduction equation. Temperature changes lead to thermal stress and thermal deformation within the material. Thermal stress is the internal stress formed due to uneven thermal expansion in different regions, while thermal deformation is the change in the overall or local dimensions of the material. Through numerical calculation, the distribution law of stress and deformation can be obtained, thus yielding the complete mechanical field response.

[0047] Finally, the mechanical field response is correlated with the electrical parameters to generate a multi-field linkage feature matrix. This signifies the establishment of a multi-dimensional data interaction mapping relationship based on the three physical fields of electricity, heat, and force. The mechanical field response includes information such as stress distribution, strain field, and deformation, while the electrical parameters include current density, potential difference, and conduction path characteristics. Correlation analysis refers to analyzing the coupling laws between different fields through numerical calculations or data fitting methods, such as how changes in the current path affect the thermal stress distribution, or how thermal strain leads to changes in resistivity. The multi-field linkage feature matrix is ​​a systematic expression of the coupling relationship, recording the correlation and interaction strength between various physical quantities in matrix form.

[0048] Furthermore, the multi-field linkage feature matrix is ​​subjected to dimensionality reduction to extract dominant influencing factors. This represents the compression of the complex coupling relationships between multiple physical fields such as electricity, heat, and force into a lower-dimensional space using mathematical methods, in order to identify the variables that have the greatest impact on system performance. The multi-field linkage feature matrix is ​​a high-dimensional matrix describing the interaction between different physical quantities, containing various features such as current density, temperature gradient, thermal stress, and strain energy. Dimensionality reduction is achieved through methods such as principal component analysis, independent component analysis, or autoencoder networks, thereby removing redundant information and retaining the main directions of change. Dominant influencing factors refer to the key variables that can still explain the changes in system performance after dimensionality reduction. For example, in the multi-field coupled system of graphene-aluminum alloy, changes in electrical conductivity and fluctuations in thermal conductivity may become dominant factors, while the impact of microporosity on performance is relatively small.

[0049] Secondly, constructing a performance response surface model based on dominant influencing factors refers to using the functional relationship between extracted key variables and target performance to form a mathematical approximation model that can be used to quickly predict performance changes. The performance response surface model is a multivariate regression model capable of predicting response results under different combinations of variables using a limited number of sample data. This model is used in complex systems to replace computationally intensive numerical simulations for efficient optimization. The construction of response surface models typically employs methods such as polynomial fitting or Gaussian process regression to describe the nonlinear relationship between electrical conductivity, thermal conductivity, and mechanical strength. For example, as the graphene content gradually increases from low to high, the model can predict that while electrical conductivity improves, the yield strength of the material may exhibit nonlinear fluctuations.

[0050] Next, sensitivity analysis is performed on the performance response surface model to determine key design parameters. This involves quantitatively calculating the impact of each input variable on the output performance to identify the design factors that most significantly influence system performance changes. Sensitivity analysis reveals the dependence of performance changes on specific parameters, such as the sensitivity of electrical conductivity to graphene distribution density and the sensitivity of thermal conduction to interfacial bonding strength. This analysis allows for the elimination of minor factors with minimal performance contribution, concentrating optimization efforts on the most significant parameters. For example, if the analysis shows that thermal conductivity is primarily affected by porosity changes, while electrical conductivity is more dependent on the orientation angle of the graphene sheets, then porosity and orientation angle become the key design parameters.

[0051] Then, a multi-objective optimization function is established based on key design parameters, with electrical conductivity, thermal conductivity, and strength as optimization objectives. This involves constructing a comprehensive optimization model that simultaneously considers multiple performance indicators after determining the main influencing variables. The multi-objective optimization function is used to balance the trade-offs between different properties. For example, improving electrical conductivity may lead to a decrease in material strength; therefore, the coordination of all three must be considered simultaneously. The optimization objectives include electrical conductivity, thermal conductivity, and strength. By rationally designing the optimization function, a dynamic balance can be established among the properties. For example, weighting coefficients can be introduced into the objective function to coordinate the increases in electrical conductivity and thermal conductivity while controlling the decrease in strength within an acceptable range.

[0052] Finally, an iterative optimization algorithm is used to dynamically update key design parameters and output an optimized parameter set. This refers to the process of finding the optimal solution through continuous iterative calculations, gradually approaching the optimal performance state. Iterative optimization algorithms can employ genetic algorithms, particle swarm optimization, or gradient descent algorithms, etc., to achieve dynamic adjustment of parameters by repeatedly calculating performance responses and error feedback. The optimized parameter set refers to the final set of optimal design variables that, while satisfying multiple objective constraints, simultaneously achieves an ideal balance between the material's electrical conductivity, thermal conductivity, and mechanical properties.

[0053] Furthermore, the optimized parameter set is input into the digital twin model, and a new integrated multiphysics solution is performed. This means that the optimal or near-optimal design parameters obtained during the optimization process, such as graphene content distribution, grain size, and interfacial bonding strength, are re-inputted into the digital twin model, serving as input variables in the new numerical calculations. The digital twin model is a high-precision model that integrates virtual simulation and real data, capable of simultaneously simulating the interactions between electric, thermal, and mechanical fields. Integrated solution refers to considering the coupling effects of multiple physics fields within a unified numerical calculation framework, rather than solving them independently. For example, during the solution process, Joule heating generated by the electric field affects the thermal field distribution, and thermal expansion reacts to the stress distribution of the mechanical field, thus forming dynamic feedback. By substituting the optimized parameter set into the model, the comprehensive performance of the material under improved conditions can be calculated more accurately.

[0054] Next, the updated electric, thermal, and mechanical field results are obtained, and an updated multi-field linkage feature matrix is ​​generated. This matrix represents the extraction of distribution data for the electric, thermal, and mechanical fields from the calculation results after integrated solution, and their transformation into structured feature representations. The electric field results include potential distribution, current density, and local conductivity variations; the thermal field results include temperature gradient, heat flux, and thermal diffusion rate; and the mechanical field results include stress, strain, and plastic deformation characteristics. After standardization and data fusion, an updated multi-field linkage feature matrix is ​​formed to characterize the synergistic relationships between the various physical fields. Compared to the initial matrix, the updated matrix reflects the performance trends of the optimized material, such as a more uniform thermal conductivity distribution, a smaller current density concentration area, or a weakened stress concentration, indicating improved material performance.

[0055] Then, the differences between the old and new multi-physics interconnected feature matrices are compared, and the performance improvement rate is calculated. This refers to the quantitative comparison of the model results before and after optimization using mathematical metrics to determine the extent of performance improvement. Difference comparison can be performed using methods such as Euclidean distance, correlation coefficient, or principal component projection difference to calculate the difference between matrices. The performance improvement rate represents the percentage improvement in overall performance of the multi-physics features before and after optimization, such as the percentage increase in conductive path continuity, thermal diffusion efficiency, or structural stress concentration. A high performance improvement rate indicates a significant positive contribution of the optimized parameters to the model performance; conversely, a low rate indicates that further parameter adjustments are needed.

[0056] Next, the optimization effect is evaluated based on the performance improvement rate to determine whether it meets the preset convergence criteria. This means comparing the performance improvement result with a pre-set target threshold to determine whether the optimization process can stop. Convergence criteria typically refer to standards such as the performance index change rate being less than a set threshold, the error being below the tolerance range, or the number of iterations reaching the upper limit. If the performance improvement rate has stabilized, meaning the model output no longer changes significantly, it indicates that convergence has been achieved and further optimization is unnecessary. For example, convergence can be determined when the performance improvement rate difference between two consecutive iterations is less than 1%. The evaluation process is equivalent to detecting whether the algorithm has found a stable solution in the multi-objective space, thereby avoiding overcomputation or getting trapped in local optima.

[0057] Finally, if the conditions are not met, the iteration step size and weight coefficients are adjusted, and the next optimization loop begins until the conditions are met. Simulated performance data is then output, indicating that when the optimization has not yet reached convergence, the optimization strategy is adaptively adjusted, and iteration continues. The iteration step size refers to the magnitude of parameter updates in each optimization. Too large a step size may cause oscillations, while too small a step size will result in slow convergence. The weight coefficients determine the importance distribution of different objectives in the optimization function, such as the proportions of electrical conductivity, thermal conductivity, and mechanical properties in the overall objective. By dynamically adjusting variables, the algorithm can be guided towards a better solution. As the number of iterations increases, the optimization path gradually converges to the optimal performance region. When the performance index reaches the target threshold, the final simulated performance data is output for subsequent verification and analysis.

[0058] Furthermore, during the optimization loop, the parameter change trajectory of each iteration is recorded, and a time-series correlation model of parameter evolution sequence and performance response sequence is constructed. This model systematically records the changes in design parameters such as graphene content distribution, grain size, and interface bonding state in each optimization iteration, while simultaneously recording the corresponding performance responses such as electrical conductivity, thermal conductivity, and mechanical properties. The parameter evolution sequence refers to the sequence of parameter values ​​changing with time or iteration steps as the iteration rounds progress, while the performance response sequence refers to the sequence of performance indicators obtained in each iteration changing with time or iteration steps. The time-series correlation model uses mathematical or data analysis methods, such as time series analysis, principal component analysis, or regression models, to establish a dynamic correlation between parameter changes and performance changes, revealing the contribution relationship of each parameter to performance improvement.

[0059] Next, key change nodes are identified based on a time-series correlation model to determine performance improvement trends. This means that the constructed time-series model is used to analyze the changes in performance indicators with iterations, identifying key points where performance changes are significant or trends reverse. Key change nodes refer to the positions in the performance indicator or parameter evolution curve where the rate of change is greatest, the trend reverses, or fluctuations are significant. By identifying these nodes, it can be determined whether the performance improvement during the optimization process is continuously increasing, stabilizing, or oscillating. For example, if the thermal conductivity reaches its peak in the seventh iteration, while the mechanical performance does not change much in the eighth iteration, then subsequent iterations can be considered key change nodes to assess whether the optimization strategy needs to be adjusted.

[0060] Then, when the performance improvement stabilizes, the parameter set is frozen and an optimized digital twin model is generated. This means that during the iteration process, when the change in each performance index is below a preset threshold or the change is close to zero for several consecutive rounds, the optimization process is considered to have converged. Therefore, the current parameter values ​​are fixed and no longer updated, forming the final optimized parameter set. The frozen parameter set is considered the optimal or near-optimal solution and is used to construct the optimized digital twin model. The optimized digital twin model is a virtual model constructed using the final parameters that accurately reflects the material's performance under multiphysics conditions, providing a foundation for subsequent simulation predictions and experimental verification.

[0061] Finally, the predicted performance of the optimized digital twin model is compared with the experimental results to obtain validation comparison results. This means that by comparing and analyzing the performance indicators predicted by the model, such as electrical conductivity, thermal diffusivity, and mechanical strength, with actual experimental measurements, errors, biases, or correlations are calculated to evaluate the accuracy of the digital twin model. Validation comparison results can reveal the predictive reliability of the model on different performance indicators. For example, an error of 2% for electrical performance, 5% for thermal performance, and 3% for mechanical performance indicates that the model's overall predictive performance is high. At the same time, it can also identify shortcomings in the model, facilitating subsequent parameter correction or optimization.

[0062] S4: Analyze the simulated performance data and propose structural optimization suggestions based on the generated performance evaluation.

[0063] Furthermore, this application also includes: correcting the parameters of the twin model based on the verification comparison results; adjusting the weights of the multi-field coupling parameters in the optimized digital twin model using a backpropagation mechanism, and re-executing the integrated solution to obtain the corrected simulation performance data; inputting the corrected simulation performance data into the performance evaluation module to calculate the comprehensive performance score; and automatically generating the structural optimization suggestions based on the comprehensive performance score.

[0064] Furthermore, this application also includes: generating a process adjustment scheme based on the structural optimization suggestions and converting it into executable process control instructions; simulating the execution of the process control instructions in a virtual machining environment and identifying potential machining deviations based on the simulation results; compensating the control instructions for parameters based on the deviation analysis results, outputting the corrected process control instructions, and updating the control module of the digital twin model.

[0065] Specifically, adjusting the parameters of the twin model based on the verification and comparison results means comparing the predictive performance of the optimized digital twin model with actual experimental results to identify biases or errors in the model and then adjusting the model parameters. Twin model parameters include material properties, interface characteristics, and multi-field coupling coefficients, which describe the model's response to electric, thermal, and mechanical fields. By adjusting the parameters, the model's simulation results can be made closer to actual experimental data, thereby improving prediction accuracy.

[0066] Next, the weights of the multi-field coupling parameters in the optimized digital twin model are adjusted using a backpropagation mechanism, and the integrated solution is re-executed to obtain corrected simulation performance data. This demonstrates that the backpropagation method is used to propagate simulation errors along the model's computational path to update the weights of the multi-field coupling parameters, making the interaction responses between the various physics fields more accurate. Multi-field coupling parameters refer to the coupling coefficients between electric, thermal, and mechanical fields, such as the relationship between thermal stress and electrical conductivity. By re-executing the integrated solution, updated simulation performance data, including electrical, thermal, and mechanical performance indicators, can be generated while considering multi-physics interactions.

[0067] Then, the corrected simulated performance data is input into the performance evaluation module to calculate the comprehensive performance score. This means that the various performance indicators obtained from the simulation are used as input, and the overall performance of the material is quantitatively evaluated through the performance evaluation algorithm. The performance evaluation module can include a weighted scoring method, combining indicators such as electrical conductivity, thermal diffusivity, and mechanical strength into a comprehensive score, which facilitates the judgment of the overall performance of the material.

[0068] Finally, structural optimization suggestions are automatically generated based on the comprehensive performance score. This means that, based on the calculated comprehensive performance score, structural design or process adjustment schemes can be proposed, such as increasing the graphene content to improve conductivity, optimizing grain size distribution to enhance mechanical strength, or improving interface bonding to enhance thermal conductivity. Automatic generation means that actionable optimization suggestions can be output without manual intervention, facilitating subsequent process implementation and model updates.

[0069] Furthermore, based on the structural optimization suggestions, a process adjustment plan is generated and converted into executable process control instructions. This means that, based on the calculated comprehensive performance score and optimization suggestions, the plan for adjusting the material structure or processing technology is transformed into specific operation instructions that can be directly used for processing equipment control. The process adjustment plan includes parameters such as graphene filling amount, grain size distribution, heat treatment temperature, and cooling rate.

[0070] Next, process control commands are simulated and executed in a virtual machining environment. Based on the simulation results, potential machining deviations are identified. This means that in a digital machining simulation environment, process control commands are first executed, and the temperature distribution, stress field, and deformation of the material during processing are calculated through simulation to identify potential deviations and machining risks. The virtual machining environment is a simulation platform built from a digital twin system, capable of predicting machining results without actual production. For example, the simulation might reveal that due to localized heat concentration, the grain size in certain areas is too large, or that excessive stress leads to the formation of microcracks.

[0071] Then, based on the deviation analysis results, parameter compensation is performed on the control commands, outputting corrected process control commands and updating the control module of the digital twin model. This means that adjustments are made to the original process control commands based on the deviations identified in the simulation, such as modifying heating time, heating rate, or cooling method, to ensure that the actual processing results conform as closely as possible to the optimization target. Simultaneously, the control module of the digital twin model also updates and corrects the parameters to ensure consistency between subsequent simulations and actual processing. Parameter compensation includes adjusting temperature, adjusting applied force, or changing the loading sequence.

[0072] In summary, the graphene aluminum alloy performance optimization simulation method based on digital twins provided in this application has the following technical effects: by achieving the technical goal of integrated optimization and closed-loop simulation control of multi-physics field performance of graphene aluminum alloys based on digital twins, it can quickly predict the comprehensive performance of materials, dynamically adjust key design parameters and generate optimized process schemes under different graphene contents, grain sizes and interface bonding states, thereby significantly improving the electrical conductivity, thermal conductivity and mechanical strength of the materials.

[0073] Example 2: Based on the same inventive concept as the graphene aluminum alloy performance optimization simulation method based on digital twins in the foregoing examples, this application also provides a graphene aluminum alloy performance optimization simulation system based on digital twins. Please refer to the appendix. Figure 2 The system includes: a parameter acquisition module 1, used to acquire electrical, thermal, and mechanical parameters of aluminum alloy samples with different graphene contents; a feature extraction module 2, used to extract performance features from the electrical, thermal, and mechanical parameters to obtain electrical, thermal, and mechanical performance features; a data output module 3, used to construct a digital twin model, perform integrated solution based on the electrical, thermal, and mechanical performance features, and output simulated performance data; and a suggestion proposal module 4, used to analyze the simulated performance data and propose structural optimization suggestions based on the generated performance evaluation.

[0074] Furthermore, the graphene-based aluminum alloy performance optimization simulation system based on digital twins is also used for: preprocessing the electrical parameters, thermal parameters, and mechanical parameters, wherein the data preprocessing includes time synchronization, noise reduction, missing value compensation, and data consistency verification; fitting and characterizing the preprocessed electrical parameters to obtain the temperature dependence and strain dependence of electrical properties, and generating the electrical performance features through dimensional reduction; performing transient thermal response analysis and feature extraction on the preprocessed thermal parameters to determine the time series characteristics of temperature rise, thermal diffusion, and thermal response under thermal excitation, and generating the thermal performance features through dimensional reduction; and decomposing and characterizing the preprocessed mechanical parameters to extract elastic response, yield behavior, and nonlinear plastic deformation features, and generating the mechanical performance features through dimensional reduction.

[0075] Furthermore, the graphene-aluminum alloy performance optimization simulation system based on digital twins is also used for: obtaining three-dimensional microscopic information based on the microscopic characterization data of the aluminum alloy matrix; embedding the spatial distribution and interface properties of graphene particles into the three-dimensional microscopic information to generate a three-dimensional virtual structure model, wherein the spatial distribution and interface properties include different grain sizes, different porosities, and different interface bonding states with the same graphene content; associating and expressing the electrical performance characteristics, thermal performance characteristics, and mechanical performance characteristics in the three-dimensional virtual structure model, and performing meshing and discretization on the three-dimensional virtual structure model, specifying material properties and interface parameters based on multidimensional feature mapping for the discrete elements to obtain a discretized model; introducing boundary conditions and initial field distribution into the discretized model to obtain the digital twin model, wherein the boundary conditions and initial field distribution are generated based on corresponding experimental conditions.

[0076] Furthermore, the graphene aluminum alloy performance optimization simulation system based on digital twins is also used to: calculate the thermal input boundary based on the three-dimensional virtual structural model and the electrical performance characteristics; use the thermal input boundary as the thermal field boundary condition, calculate the thermal stress and thermal deformation from the solved temperature changes, and obtain the mechanical field response; and perform correlation analysis between the mechanical field response and the electrical parameters to generate a multi-field linkage feature matrix.

[0077] Furthermore, the graphene aluminum alloy performance optimization simulation system based on digital twins is also used for: dimensionality reduction of the multi-field linkage feature matrix to extract dominant influencing factors; constructing a performance response surface model based on the dominant influencing factors; performing sensitivity analysis on the performance response surface model to determine key design parameters; establishing a multi-objective optimization function based on the key design parameters, with electrical conductivity, thermal conductivity, and strength as optimization objectives; and dynamically updating the key design parameters using an iterative optimization algorithm to output an optimized parameter set.

[0078] Furthermore, the graphene aluminum alloy performance optimization simulation system based on digital twins is also used for: inputting the optimization parameter set into the digital twin model and re-solving the multiphysics field in an integrated manner; obtaining the updated electric field, thermal field, and mechanical field results, and generating an updated multi-field linkage feature matrix; comparing the differences between the old and new feature matrices of the multi-field linkage feature matrix and the updated multi-field linkage feature matrix, and calculating the performance improvement rate; evaluating whether the optimization effect meets the preset convergence condition based on the performance improvement rate; if not, adjusting the iteration step size and weight coefficients, entering the next round of optimization loop, until the condition is met, and outputting the simulated performance data.

[0079] Furthermore, the graphene aluminum alloy performance optimization simulation system based on digital twins is also used for: recording the parameter change trajectory of each iteration during the optimization cycle, and constructing a time-series correlation model between the parameter evolution sequence and the performance response sequence; identifying key change nodes based on the time-series correlation model and judging the performance improvement trend; freezing the parameter set and generating the optimized digital twin model when the performance improvement tends to stabilize; and comparing the predicted performance of the optimized digital twin model with the experimental results to obtain the verification comparison results.

[0080] Furthermore, the graphene aluminum alloy performance optimization simulation system based on digital twins is also used to: correct the parameters of the twin model based on the verification comparison results; adjust the weights of the multi-field coupling parameters in the optimized digital twin model using a backpropagation mechanism, and re-execute the integrated solution to obtain the corrected simulation performance data; input the corrected simulation performance data into the performance evaluation module to calculate the comprehensive performance score; and automatically generate the structural optimization suggestions based on the comprehensive performance score.

[0081] Furthermore, the graphene aluminum alloy performance optimization simulation system based on digital twin is also used to: generate process adjustment schemes based on the structural optimization suggestions and convert them into executable process control instructions; simulate and execute the process control instructions in a virtual machining environment, identify potential machining deviations based on the simulation results; perform parameter compensation on the control instructions based on the deviation analysis results, output the corrected process control instructions, and update the control module of the digital twin model.

[0082] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The graphene aluminum alloy performance optimization simulation method and specific examples based on digital twins in the aforementioned embodiment one are also applicable to the graphene aluminum alloy performance optimization simulation system based on digital twins in this embodiment. Through the foregoing detailed description of the graphene aluminum alloy performance optimization simulation method based on digital twins, those skilled in the art can clearly understand the graphene aluminum alloy performance optimization simulation system based on digital twins in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0083] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0084] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A simulation method for performance optimization of graphene aluminum alloys based on digital twins, characterized in that, include: Electrical, thermal, and mechanical parameters of aluminum alloy samples with different graphene contents were collected. The electrical, thermal, and mechanical parameters are subjected to performance feature extraction to obtain electrical performance features, thermal performance features, and mechanical performance features. A digital twin model is constructed, and the electrical, thermal, and mechanical performance characteristics are solved in an integrated manner to output simulated performance data. Analyze the simulated performance data and propose structural optimization suggestions based on the generated performance evaluation.

2. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 1, characterized in that, The electrical, thermal, and mechanical parameters are subjected to performance feature extraction to obtain electrical performance characteristics, thermal performance characteristics, and mechanical performance characteristics, including: The electrical parameters, thermal parameters, and mechanical parameters are subjected to data preprocessing, wherein the data preprocessing includes time synchronization, noise reduction, missing value compensation, and data consistency verification. The preprocessed electrical parameters are fitted and characterized to obtain the temperature dependence and strain dependence of the electrical performance. The electrical performance features are generated through dimensionality reduction. Transient thermal response analysis and feature extraction are performed on the preprocessed thermal parameters to determine the time-series characteristics of temperature rise, thermal diffusion and thermal response under thermal excitation, and the thermal performance characteristics are generated by dimensionality reduction. The preprocessed mechanical parameters are decomposed and characterized to extract elastic response, yield behavior and nonlinear plastic deformation features, and the mechanical performance features are generated through dimensional reduction.

3. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 1, characterized in that, Building a digital twin model includes: Three-dimensional microscopic information is obtained from the microscopic characterization data of the aluminum alloy matrix; A three-dimensional virtual structure model is generated by embedding the spatial distribution and interface properties of graphene particles into the three-dimensional microscopic information. The spatial distribution and interface properties include different grain sizes, different porosities, and different interface bonding states for the same graphene content. The electrical, thermal, and mechanical properties are associated and expressed in the three-dimensional virtual structure model. The three-dimensional virtual structure model is then meshed and discretized. Material properties and interface parameters based on multidimensional feature mapping are assigned to the discrete elements to obtain the discretized model. Boundary conditions and initial field distributions are introduced into the discretized model to obtain the digital twin model, wherein the boundary conditions and initial field distributions are generated based on the corresponding experimental conditions.

4. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 3, characterized in that, Before associating and expressing the electrical performance characteristics, thermal performance characteristics, and mechanical performance characteristics in the three-dimensional virtual structural model, the following steps are included: Based on the three-dimensional virtual structure model, the heat input boundary is calculated according to the electrical performance characteristics; The thermal input boundary is used as the thermal field boundary condition. The obtained temperature change is used to calculate the thermal stress and thermal deformation to obtain the mechanical field response. The mechanical field response and the electrical parameters are correlated and analyzed to generate a multi-field linkage feature matrix.

5. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 4, characterized in that, The electrical performance characteristics, thermal performance characteristics, and mechanical performance characteristics are expressed in association within the three-dimensional virtual structural model, including: The multi-field linkage feature matrix is ​​subjected to dimensionality reduction processing to extract the dominant influencing factors; A performance response surface model was constructed based on the aforementioned dominant influencing factors; Sensitivity analysis was performed on the performance response surface model to determine key design parameters; A multi-objective optimization function is established based on the key design parameters, with electrical conductivity, thermal conductivity and strength as the optimization objectives. An iterative optimization algorithm is used to dynamically update the key design parameters and output an optimized parameter set.

6. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 5, characterized in that, An integrated solution is performed based on the aforementioned electrical, thermal, and mechanical performance characteristics to output simulated performance data, including: The optimized parameter set is input into the digital twin model, and the multiphysics field solution is performed again. Obtain the updated electric field, thermal field, and mechanical field results, and generate the updated multi-field linkage feature matrix; The performance improvement rate is calculated by comparing the differences between the old and new feature matrices of the multi-field linkage feature matrix and the updated multi-field linkage feature matrix. The optimization effect is evaluated based on the performance improvement rate to determine whether it meets the preset convergence condition. If the conditions are not met, the iteration step size and weight coefficients are adjusted, and the next round of optimization loop is entered until the conditions are met, at which point the simulated performance data is output.

7. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 6, characterized in that, After performing an integrated solution based on the aforementioned electrical, thermal, and mechanical performance characteristics, and outputting simulated performance data, the following is included: During the optimization loop, the parameter change trajectory of each iteration is recorded, and a time-series correlation model between the parameter evolution sequence and the performance response sequence is constructed. Based on the aforementioned temporal correlation model, key change nodes are identified, and performance improvement trends are determined. When the performance improvement stabilizes, the parameter set is frozen and an optimized digital twin model is generated; The prediction performance of the optimized digital twin model was compared with the experimental results to obtain the verification comparison results.

8. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 7, characterized in that, Analyze the simulated performance data and propose structural optimization suggestions based on the generated performance evaluation, including: The parameters of the twin model are corrected based on the verification and comparison results. The weights of the multi-field coupling parameters in the optimized digital twin model are adjusted using the backpropagation mechanism, and the integrated solution is re-executed to obtain the corrected simulation performance data. Input the corrected simulated performance data into the performance evaluation module to calculate the overall performance score; The structural optimization suggestions are automatically generated based on the overall performance score.

9. The method for performance optimization simulation of graphene aluminum alloy based on digital twin as described in claim 8, characterized in that, After analyzing the simulated performance data and proposing structural optimization suggestions based on the generated performance evaluation, the following steps are included: Based on the structural optimization suggestions, a process adjustment plan is generated and converted into executable process control instructions; The process control commands are simulated and executed in a virtual machining environment, and potential machining deviations are identified based on the simulation results. Based on the deviation analysis results, parameter compensation is performed on the control commands, the corrected process control commands are output, and the control module of the digital twin model is updated.

10. A graphene-based aluminum alloy performance optimization simulation system based on digital twins, characterized in that, The steps for implementing the digital twin-based graphene aluminum alloy performance optimization simulation method according to any one of claims 1 to 9 include: The parameter acquisition module is used to collect the electrical, thermal, and mechanical parameters of aluminum alloy samples with different graphene contents. The feature extraction module is used to extract performance features from the electrical parameters, thermal parameters, and mechanical parameters to obtain electrical performance features, thermal performance features, and mechanical performance features. The data output module is used to construct a digital twin model, perform integrated solutions based on the electrical performance characteristics, thermal performance characteristics, and mechanical performance characteristics, and output simulated performance data. It is suggested that a module be proposed to analyze the simulated performance data and propose structural optimization suggestions based on the generated performance evaluation.

Images