Capacitor structure design optimization method and system and storage medium

Abstract

The invention relates to the technical field of electrical design and intelligent optimization calculation, and discloses a capacitor structure design optimization method and system and a storage medium. The method comprises the following steps: carrying out modeling processing on capacitor structure parameters, and dividing a parameter space to obtain an initial model set; obtaining a multi-physical field simulation model according to the geometric model and the material attributes; obtaining performance indexes such as capacitance value, heat distribution and stress based on the simulation model; setting a target function and constraint conditions, and operating an optimization algorithm to obtain an optimal solution set; performing feedback control processing on an optimization result, and analyzing a convergence path to obtain a structure parameter; according to the invention, the execution efficiency of capacitor structure design optimization is improved, and the consistency and stability of the optimization process and the performance prediction result are improved.

Description

Technical Field

[0001] The present application relates to the field of electrical design and intelligent optimization computing technology, and in particular to a capacitor structure design optimization method, system and storage medium. Background Art

[0002] As core energy storage and filtering components in power electronic devices, capacitors' structural design directly impacts their electrical stability, thermal management capabilities, and mechanical reliability. With the rapid development of high-frequency, high-voltage, and highly integrated electronic systems, capacitors face higher demands in terms of electrical performance and structural adaptability, prompting their structural design to evolve from traditional two-dimensional parameter control to multi-dimensional coupled optimization.

[0003] Existing capacitor design schemes mainly rely on engineering experience and finite parameter trial calculations to select structural configurations, and usually perform performance verification through static modeling and single physical field simulation. This approach has drawbacks such as low design efficiency, insufficient response accuracy, and unclear structure-performance mapping relationships. In particular, it is difficult to obtain the globally optimal structural configuration results when faced with complex material parameters, nonlinear boundary conditions, and coupled physical responses. Some studies have introduced computer-aided simulation methods to improve modeling and analysis capabilities, but they still focus on fixed simulation paths and single-objective optimization. There is a lack of integrated optimization methods based on collaborative control of multiple performance indicators, convergence path feedback, and high-dimensional design space search mechanisms. As a result, the optimization process is time-consuming and prone to falling into local optimality, which seriously restricts the rapid development and deployment of capacitors in high-performance application scenarios.

[0004] Therefore, there is an urgent need for an integrated design optimization method with structural modeling, multi-physics field simulation, performance extraction, global optimization and error feedback capabilities to achieve efficiency improvement and performance enhancement in capacitor structure design. Summary of the Invention

[0005] The present application provides a capacitor structure design optimization method, system and storage medium for improving the structural configuration efficiency and multi-performance response accuracy of the capacitor under complex working conditions, achieving comprehensive optimization goals for electrical performance, thermal stability and mechanical strength, and solving the problems of low structural modeling efficiency, limited simulation accuracy and slow optimization convergence in existing capacitor design.

[0006] In a first aspect, the present application provides a capacitor structure design optimization method, the capacitor structure design optimization method comprising: Modeling capacitor structural parameters and discretely dividing and combining the capacitor structural parameters in the model space to obtain a structural configuration model data set, wherein the capacitor structural parameters include geometric parameters and material properties; Setting input conditions according to the geometric parameters and material properties, and obtaining a calculation data set through a preset multi-physics field simulation model; Based on the extraction of the calculated data set, performance index data is obtained, wherein the performance index data includes capacitance value, electric field strength, thermal distribution and mechanical stress value; Setting an objective function and constraints based on the performance indicator data, and then executing an optimization algorithm iteration process to obtain a global optimal solution parameter combination, wherein the optimization algorithm iteration process comprises: setting an algorithm type and initializing parameters, introducing a proxy model and a Bayesian-assisted algorithm to iterate until an iteration termination condition is met and the global optimal solution parameter combination is obtained; The global optimal solution parameter combination and the performance prediction error are analyzed and processed, and the convergence termination judgment criteria are compared to obtain the structural parameter configuration.

[0007] In a second aspect, the present application provides a capacitor structure design optimization system, the capacitor structure design optimization system comprising: a structural modeling module, configured to model capacitor structural parameters and discretely divide and combine the capacitor structural parameters in a model space to obtain a structural configuration model data set, wherein the capacitor structural parameters include geometric parameters and material properties; A simulation analysis module, configured to set input conditions according to the geometric parameters and material properties and obtain a calculation data set through a preset multi-physics field simulation model; A performance extraction module, configured to extract performance index data based on the calculation data set, wherein the performance index data includes capacitance value, electric field strength, thermal distribution and mechanical stress value; An optimization solution module is used to set the objective function and constraints based on the performance indicator data, and then execute the optimization algorithm iteration process to obtain the global optimal solution parameter combination, wherein the optimization algorithm iteration process includes: setting the algorithm type and initializing the parameters, introducing the surrogate model and the Bayesian-assisted algorithm iteration until the iteration termination condition is met to obtain the global optimal solution parameter combination; The result evaluation module is used to analyze and process the global optimal solution parameter combination and the performance prediction error, and compare the convergence termination judgment criteria to obtain the structural parameter configuration.

[0008] In a third aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on the computer, the computer executes the above-mentioned capacitor structure design optimization method.

[0009] In the technical solution provided by the present application, by modeling the parameter characteristics of the capacitor structure and constructing a discrete combination of structural variable space, the systematization and standardization of the structural sample generation process are achieved, which solves the problems of insufficient structural configuration schemes and incomplete variable coverage in traditional designs. By introducing material properties and geometric information to establish a coupled simulation model of electric field, thermal field and mechanical field, the simulation distortion problem caused by the separation modeling of multiple physical responses in the existing simulation method is solved. In the simulation solution stage, multiple performance indicators including capacitance value, electric field strength, thermal distribution and mechanical stress are extracted, which enhances the comprehensiveness and accuracy of the structural performance evaluation. On this basis, the extracted performance data are used for objective function construction and constraint setting, and the iterative search of the structural parameter space is realized with the help of the global optimization algorithm. At the same time, a neural network proxy model is introduced to replace some high-cost simulation computers. The algorithm is used to reduce computing resource consumption and improve the convergence speed of the optimization process. The Bayesian optimization mechanism is combined to realize the search path guidance based on uncertainty prediction to enhance the global search capability. During the optimization process, the optimization path and fitness change trend are recorded to construct a learning scheduling strategy to dynamically adjust the search range and step size to solve the problems of high-dimensional optimization that are easy to fall into local optimality and unstable algorithm execution efficiency. In the optimization result output stage, the error analysis and feedback mechanism is introduced to compare the predicted performance with the actual simulation results, and secondary judgment and iterative feedback are performed based on the error threshold to ensure the accuracy and feasibility of the structural parameters. In the overall design process, the core technical shortcomings of traditional structural optimization methods in modeling efficiency, simulation accuracy, convergence stability and result verification are effectively overcome, and the efficient optimization design of capacitor structures and the coordinated control of multiple performance objectives are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A schematic diagram of an embodiment of a capacitor structure design optimization method in an embodiment of the present application; Figure 2 This is a diagram of the evolution of Bayesian optimization sampling density in an embodiment of this application; Figure 3 This is a comparison chart of the convergence performance of the optimization algorithm in the embodiment of this application; Figure 4 A schematic diagram of an embodiment of a capacitor structure design optimization method system in an embodiment of the present application; Figure 5 It is a schematic block diagram of the structure of a capacitor structure design optimization device in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The embodiments of the present application provide a capacitor structure design optimization method and system. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products or devices.

[0013] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In one embodiment of the present application, a capacitor structure design optimization method includes: Step S1 : Modeling capacitor structural parameters, and discretely dividing and combining the capacitor structural parameters in the model space to obtain a structural configuration model data set, wherein the capacitor structural parameters include geometric parameters and material properties.

[0014] Specifically, a set of variables, such as the electrode arrangement, insulation distance, and material properties of the structural body, is defined. These structural parameters are discretized into several value levels using numerical intervals. Multiple sets of structural parameter combinations covering the design space are generated using full combination, Latin hypercube, or orthogonal sampling methods. The corresponding 3D model is constructed based on the parameter grouping results. The model dataset, indexed by structural parameters, records spatial layout and material distribution information. This not only reflects the diversity of structural samples, but also facilitates subsequent simulation calls and enables data flow from parameter space mapping to structural models.

[0015] Step S2: setting input conditions according to the geometric parameters and material properties, and obtaining a calculation data set through a preset multi-physics field simulation model.

[0016] Specifically, when extracting geometric parameters and material properties, based on the pre-set geometric variables and material property tables of media, electrodes, etc., dimensional parameters, layer thickness information, and material thermal conductivity, electrical conductivity, elastic modulus and other data are extracted from the structural model, and then uniformly organized according to the field requirements of the input template to generate a structural input configuration table and material property table.

[0017] When setting the simulation model input conditions based on the extracted parameters, electrode excitation, heat source terms, and mechanical constraints are set according to the boundary requirements corresponding to the electric field, thermal field, and force field. A simulation domain is established within the spatial range, and the mesh discretization of this area is performed to generate a high-density or locally encrypted simulation mesh file. After defining the control equations for the electric field distribution equation, heat conduction equation, and structural mechanics equilibrium equation, a multi-physics field coupling relationship is established and interactive boundaries and variable mappings are set to construct a complete coupled simulation model. When numerically solving this coupled model, the solver iterative algorithm is used to obtain the electric potential distribution, temperature field distribution, and stress field response results in sequence, and the response data at the spatial nodes in each physical domain are extracted to generate a calculation data set.

[0018] Step S3: Based on the extraction of the calculation data set, performance index data is obtained, and the performance index data includes capacitance value, electric field strength, thermal distribution and mechanical stress value.

[0019] Specifically, when extracting the calculation data set, the electric field, thermal field and mechanical field response data are respectively classified into independent data channels according to the physical type label of the simulation output, and a data index table is established through the structural parameter mapping relationship for the rapid extraction and group analysis of subsequent performance indicators. In the process of analyzing the electric field distribution data, the potential difference is calculated by extracting the potential values ​​of the anode and cathode regions. At the same time, the charge density data is extracted and integrated at the electrode surface nodes to obtain the total charge of the electrode. The capacitance value under the current structural configuration is calculated by calculating the ratio of the charge to the potential difference. The electric field strength data is then gradient-solved to obtain the spatially distributed intensity vector field, and the intensity extreme value area is identified and marked in the distribution diagram for analyzing the electric field concentration trend. In the temperature field analysis, the temperature rise rate and the spatial heat distribution state are obtained by statistically calculating the temperature changes of each node in the time series, and the heat accumulation area and the average heat diffusion path parameters are extracted accordingly. Finally, the structural displacement and stress field results are jointly solved, and the structural stress response value is obtained based on the nodal strain recovery method. The maximum stress point is identified in combination with the regional coordinates, and the mechanical stress value is extracted to form a complete performance index data set for optimization control and reliability analysis.

[0020] Step S4: setting the objective function and constraints according to the performance indicator data, and then executing the optimization algorithm iteration process to obtain the global optimal solution parameter combination, wherein the optimization algorithm iteration process is: setting the algorithm type and initializing the parameters, introducing the agent model and Bayesian assisted algorithm iteration until the iteration termination condition is met to obtain the global optimal solution parameter combination.

[0021] Specifically, when setting the objective function and constraint conditions, the target parameters are first selected based on the key parameters in the performance indicators, such as capacitance, electric field strength, temperature rise level, and mechanical stress value, and the weight distribution ratio is set according to the optimization priority or performance requirements to construct an objective function model with a multi-objective expression. At the same time, combined with information such as the physical value range of the structural parameters, design tolerances, material limitations, and engineering application requirements, a set of constraint functions containing inequalities and equalities is constructed to limit the solution space boundary and feasibility domain. The above objective function expression and constraint function set are input into the optimization control module together, and the initial search point or random population is configured according to the type of optimization algorithm to form the optimization starting state data set.

[0022] During the execution of the optimization algorithm, multi-dimensional convergence control conditions, including error thresholds, objective function change rates, and maximum number of iterations, are set. Based on these conditions, the current solution set is judged in real time to determine whether it meets the termination criteria. Subsequently, the objective function is evaluated for the structural parameter combinations based on the selected optimization algorithm to obtain a fitness score dataset. A surrogate model is introduced to pre-estimate and screen high-cost simulation steps, yielding a candidate evaluation solution set to improve iteration efficiency. During the convergence process, a refined analysis of the current optimal region is performed. A perturbation mechanism is used to perform perturbation sampling on the local parameter interval. A Bayesian update strategy is then used to reconstruct the sampling distribution based on prediction uncertainty to obtain a refined set of optimized candidate solutions. During the iterative execution, the optimization path and intermediate solution states of each round are continuously recorded. The fitness evolution trajectory and historical search strategy behavior are analyzed. A learning scheduling model is established to dynamically adjust the search parameters. Finally, within the solution set that meets the convergence conditions, a structural parameter combination with the optimal objective function, satisfied constraints, and a controllable error range is selected as the global optimal solution output, providing a reliable basis for structural configuration and performance prediction.

[0023] Step S5: Analyze and process the global optimal solution parameter combination and the performance prediction error, and compare the convergence termination judgment criteria to obtain the structural parameter configuration.

[0024] Specifically, based on the multi-physics field simulation results, the simulation data of each group of structural models is processed, the potential distribution and charge density data of the electrode area are extracted, the capacitance value is calculated, the extreme value distribution of the electric field intensity is obtained, the statistical parameters of the temperature field and the uniformity of the thermal distribution are analyzed, and the mechanical stress concentration area and its maximum value are identified. All performance indicators are archived according to the structural parameter index to form a structure-performance mapping database. Subsequently, multi-objective or constrained optimization calculations are performed on the database in the form of objective functions and constraints. Parameter search is performed based on the global optimization algorithm and adaptive sampling mechanism. The newly generated parameter group is fed back to the simulation process, and it is iterated until the performance indicators meet the preset goals. The optimization output results are compared with the simulation performance error, and the validity of the solution is judged according to the set accuracy and convergence criteria. The closed-loop control of parameter optimization, performance simulation and convergence feedback is realized to obtain a capacitor structure configuration scheme with optimal performance, reasonable structure and practical feasibility.

[0025] In the embodiment of the present application, by modeling the capacitor structural parameters and constructing a discretized structural variable space, a systematic structural sample set is formed to solve the problems of insufficient structural solutions and limited variable coverage in traditional designs. Geometric parameters and material properties are introduced to construct a coupled simulation model of electric field, thermal field and mechanical field to avoid the response distortion caused by modeling the physical field separately. Based on this model, capacitance value, electric field strength, thermal distribution and mechanical stress indicators are extracted to improve the integrity of performance evaluation. Performance data is used to construct the objective function and set constraints, and the iterative solution of structural parameters is achieved through a global optimization algorithm. A proxy model is introduced to replace some high-overhead simulations to reduce computational costs, and a Bayesian optimization strategy is used to implement uncertainty-based search path guidance and improve global search efficiency. The optimization path and fitness changes are recorded, a learning scheduling mechanism is constructed, and the search strategy is dynamically adjusted to improve convergence performance. In the output stage, based on the comparison of prediction error and simulation results, a convergence judgment and feedback mechanism is established to ensure that the final structural parameters meet the feasibility and accuracy requirements. This solution improves modeling efficiency and simulation accuracy, enhances the stability and reliability of the optimization algorithm, and achieves multi-performance collaborative optimization of the capacitor structure.

[0026] In a specific embodiment, the process of executing step S1 may specifically include the following steps: The basic configuration of the capacitor structure is analyzed, and a set of adjustable structural parameters is obtained based on geometric parameter feature extraction. Range setting and classification of the parameter set are performed, and a structural variable space is obtained by performing standardization processing based on constraint boundary setting results, wherein the constraint boundary setting results include structural parameter ranges and performance index limits; Discretization is performed on the structural variable space, and multiple groups of parameter combination samples are generated based on the combination of the variable dimensions; Performing modeling based on the parameter combination samples and executing a three-dimensional structure generation process in batches to obtain an initial structure model set; The initial structure model set is converted into a unified format, and the structure configuration model data set is obtained by sorting the readable model data set.

[0027] Specifically, the capacitor structure is decomposed into core components such as electrodes, dielectric layers, and end faces. Geometric parameters such as length, width, thickness, gap, and winding radius, as well as material properties, are extracted for these components to form a set of adjustable structural parameters. A reasonable value range is set for each parameter based on the device's operating environment and actual engineering requirements. Parameters are categorized by function, physical action, or process feasibility, ensuring both differentiation and independence. Boundary conditions are used for standardized expression, constructing a complete structural variable space and achieving standardized and unified parameter definitions.

[0028] By discretizing each parameter within the structural variable space according to intervals or step sizes, and employing multivariate combinations such as full combinations, orthogonal experiments, or Latin hypercube sampling, multiple sets of parameter combination samples covering the design space are generated, each corresponding to a specific structure. Using these parameter combination samples, 3D modeling tools are invoked to directly map structural parameters to 3D structural features, enabling automated batch model generation. All structural models are automatically archived using parameter groups as indexes, and material properties and related boundary information are synchronously annotated in the model data, ensuring that each model has complete physical and geometric information.

[0029] The batch-generated structural models are converted into a unified format and organized into a universally readable structural configuration model dataset through geometric model simplification, boundary consistency processing, and attribute normalization. This dataset supports the migration, reuse, and data flow of structural models between different simulation platforms.

[0030] Taking the design of metallized polypropylene film capacitors as an example, a detailed analysis of their winding structure, electrode layer, and polypropylene film components was conducted. Key structural parameters such as metallization layer thickness, polypropylene film thickness, plate length, plate spacing, and winding radius were extracted, and the material properties of the metal and polypropylene used were collected. Based on the actual electrical performance and process requirements, a reasonable value range was set for each parameter. The parameters were classified and standardized based on material properties and structural functions to form a complete structural variable space. Interval discretization and orthogonal experimental design were used to combine parameters in multiple dimensions, automatically generating a large number of structural parameter groups with different configurations. Corresponding 3D structural models were then generated in batches using 3D modeling tools. Key information such as parameter indexes, material properties, and end face structures were simultaneously archived for unified model management. All structural models were formatted and data standardized to form a structural configuration model dataset that is easily accessible for multi-physics field simulation analysis. This provides efficient and automated data support for the multi-performance simulation, optimization, and engineering design of metallized polypropylene film capacitors.

[0031] In a specific embodiment, the process of executing step S2 may specifically include the following steps: Extracting parameters of the geometric parameters and material properties, and arranging the parameters according to preset input format requirements; Setting the input conditions based on the extracted input parameters and boundary conditions, and initializing the simulation area grid; Modeling the electric field distribution equation, the heat conduction equation, and the mechanical equilibrium equation respectively, and coupling them to obtain the multi-physics field simulation model; Numerically solving the multi-physics field simulation model and extracting response data of each physical field to obtain simulation results; The simulation result data is formatted, and the results of various performance indicators are marked and summarized to obtain the calculation data set.

[0032] Specifically, when extracting parameters, various parameters such as electrode size, dielectric thickness, material conductivity, dielectric constant, thermal conductivity, etc. are classified according to simulation requirements. Based on the preset data input template, the parameter names, units and numerical ranges are unified, and a list of structural and material parameters is established. Non-standard or missing data is supplemented and corrected, and parameters are normalized to ensure the consistency and standardization of subsequent data flow and automated calls.

[0033] The sorted input parameters are mapped to the boundary conditions, and typical boundary conditions such as electrode voltage, end face grounding, heat source distribution, and fixed support are assigned to the corresponding structural surfaces or nodes. At the same time, an adaptive simulation area grid is generated according to the geometric characteristics and accuracy requirements of the structural model. The automatic meshing tool is used to subdivide the structural domain to achieve spatial discretization of the physical field distribution, so that a one-to-one index relationship is established between parameters, boundaries, and grid information in the data structure, providing a high-quality spatial foundation for solving physical equations.

[0034] Based on the aforementioned parameters and mesh information, a set of physical equations for electric field distribution, heat conduction, and mechanical equilibrium is established. Structural and material characteristics are mapped into equation coefficients and initial conditions, and a coupled multi-physics simulation model is constructed. Numerical methods such as finite element methods are used to automatically solve the model and output the response data for each physical field. Simulation results such as electric field intensity, temperature distribution, and stress response are archived according to parameter indexes. The simulation output is formatted in a standardized format, and performance indicators such as capacitance, extreme value regions, maximum temperature rise, and maximum stress are extracted and labeled. This enables automatic linkage and archiving of structural parameters, simulation models, and performance data, resulting in a complete computational dataset that provides data support and a basis for subsequent performance analysis and optimization decisions.

[0035] Taking the design of IGBT absorption capacitors as an example, the capacitor structure design optimization method provided in this application can carry out multi-dimensional design based on the high requirements for capacitance stability, thermal management capabilities and mechanical structural strength under high-frequency and high-voltage working conditions. By modeling its geometric parameters and constructing a sample space of structural variable combinations to provide a set of structural candidates for the optimization process, a multi-physics field simulation model is used to couple the electric field distribution, heat conduction path and mechanical stress concentration area to simulate and extract key performance indicators; by constructing an objective function including capacitance value, electric field strength, temperature rise distribution and stress response, and introducing an agent model and Bayesian optimization strategy for multiple rounds of iterative search, the optimization convergence efficiency is improved and the simulation cost is reduced; the search strategy is dynamically adjusted through a learning scheduling mechanism to adapt to the reliability requirements of the IGBT absorption capacitor under different working conditions; through error feedback control, the optimal structural parameter combination with the smallest simulation prediction deviation is screened, so that the absorption capacitor can maintain stable electrical performance and safe structural strength while improving energy absorption efficiency, which is suitable for high-reliability operation requirements in power modules, power conversion devices and other occasions.

[0036] In a specific embodiment, the execution process may specifically include the following steps: Establishing the electric field distribution equation and setting electrode boundary conditions; Establishing the heat conduction equation and setting the material thermal conductivity and heat source terms; Establishing the mechanical equilibrium equation and setting stress boundaries and displacement constraints; Carry out variable association processing for the three types of control equations: electricity, heat and force, and uniformly process the coupled boundary interaction terms; The coupled control equations are packaged, and a multi-physics field simulation solution interface is constructed to obtain the multi-physics field simulation model.

[0037] Specifically, when constructing the electric field distribution equation, the spatial position parameters of the electrode area are extracted according to the capacitor geometric model, and a grid node system is generated. Through the boundary identification rules, a fixed potential value is applied to the anode and cathode, and the boundary is set to insulation or conduction type to form a closed electric field area. The electric potential distribution relationship is constructed through the Poisson equation, and the potential gradient is calculated based on the material conductivity and charge density parameters for subsequent capacitance value and electric field strength analysis. Voltage excitation conditions are set at the electrode boundaries to ensure that the simulation field has an electric field driving source, ensuring the integrity of the field distribution calculation and boundary consistency.

[0038] During the modeling of the heat conduction governing equations, the thermal conductivity parameters of the materials corresponding to the geometry are extracted, and heat source terms are applied to any existing built-in heat sources. Steady-state or transient heat conduction equations are used to describe the diffusion of heat between the medium and the conductor. Thermal boundary conditions are used to set natural convection or constant temperature boundaries to simulate the effects of air cooling environments. Enhanced meshing is applied to areas of high heat flux density to improve simulation accuracy, ensuring that the heat distribution results accurately reflect the concentrated trend of temperature rise, providing data support for subsequent thermal stability assessments and structural heat resistance optimization.

[0039] During the mechanical equilibrium equation establishment and coupling processing phase, external constraints are imposed on the capacitor's stressed areas, especially the electrode interfaces and fixed boundary areas, setting the structural displacement zero point and defining free boundaries to simulate the actual assembly state. The stress-strain field is established based on the basic equations of elasticity. The structural stiffness matrix is ​​constructed using the material elastic modulus and Poisson's ratio parameters, and variables are associated with the temperature load of the thermal field and the electromotive force of the electric field. Coupling terms are introduced for intersecting boundaries to achieve a unified fusion of electrical, thermal, and mechanical boundary data. By encapsulating the control equation group and constructing a numerical solution interface, a complete multi-physics coupling model is formed to achieve a precise prediction of the structural response behavior under real working conditions.

[0040] Taking the design of DC-Link capacitors for low-profile PCBs as an example, these capacitors are typically used in compact power modules or inverter circuits, placing stringent requirements on structural compactness, electric field uniformity, thermal conductivity efficiency, and stress distribution control. During the design process, an electric field distribution control equation was established to evaluate the potential gradient variation under different electrode structures. The electrode spacing, electrode plate area, and conductive path were carefully modeled to accurately predict the capacitance value and local electric field strength. Furthermore, electrode boundary conditions were set to simulate actual operating voltages to ensure simulation field consistency. During the thermal conduction modeling process, the heat exchange efficiency between the upper and lower surfaces of the capacitor and the PCB was considered, thermal conductivity parameters between different material layers were introduced, and the mounting interface was set as a convection boundary to evaluate the impact of heat accumulation on capacitor performance. Transient thermal field simulation was used to simulate the temperature rise distribution after long-term power-on, guiding material selection and package structure optimization. In structural mechanics modeling, stress boundaries and displacement constraints are defined based on the solder joint connection constraints and the capacitor's internal support structure. This is used to identify regions of potential deformation during low-profile press assembly and to collaboratively analyze the structural response induced by thermal expansion. By uniformly coupling the aforementioned electric, thermal, and stress fields and introducing boundary interaction terms, a complete set of governing equations is constructed. This model accurately predicts the capacitance stability, thermal safety, and mechanical reliability of DC-Link capacitors under varying package thicknesses and contact configurations, providing an optimization basis and structural support for their safe application in high-speed, high-density power electronics systems.

[0041] In a specific embodiment, the process of executing step S3 may specifically include the following steps: Classify multi-physics simulation results and extract various types of physical response data; Analyze the electric field distribution data and obtain the capacitance value based on the potential difference and charge density data between electrodes; The gradient of the electric field intensity data is calculated, and then the extreme value areas are marked to obtain the electric field intensity distribution map; Conduct statistical analysis on the temperature field simulation results and obtain thermal distribution parameters based on the temperature rise trend data; The displacement and stress field data are solved, and after identifying the maximum stress area, the mechanical stress value is extracted.

[0042] Specifically, when classifying the results of multi-physics field simulations, the electric field, thermal field, and mechanical response data are extracted into independent data channels, and the simulation outputs are automatically grouped according to physical property labels. The electric potential distribution, electric field intensity, temperature field data, and stress and strain data are respectively assigned to the corresponding structural variable indexes to facilitate the rapid extraction of subsequent performance indicators. During the classification process, structured operations on multi-field simulation data are achieved through data field matching and simulation area coordinate screening, and a unified response data interface is established to ensure the accuracy and consistency of the performance extraction process, providing a basic data source for performance evaluation and optimization.

[0043] When analyzing and processing the electric field distribution data, the potential values ​​at the electrode boundaries are extracted and the anode and cathode regions are determined based on position mapping. After numerically differentiating the potential difference between the two, the charge density data at the electrode surface nodes is extracted. The total electrode charge is calculated by integrating the node charge. The capacitance corresponding to the target structure is then derived from the ratio of charge to potential difference. This capacitance value is then entered into a performance index table, used to establish a correspondence between structural parameters and electrical response performance, and provides a computational basis for optimizing the objective function.

[0044] When performing gradient field analysis on electric field intensity data, the spatial gradient of the electric potential field is solved within the entire simulation area to obtain the electric field intensity distribution vector field. By performing extreme value analysis on the intensity gradient, areas of electric field concentration are identified and high-intensity distribution bands are marked. The corresponding areas are located in the geometric structure to identify electric field focusing risks. Using thermal field simulation data, the temperature distribution trend in the simulation time series is analyzed, and the heat accumulation and conduction efficiency are evaluated in combination with the temperature rise rate of each node. A coupled analysis of the displacement and stress simulation results is performed to extract the location of the maximum stress value and evaluate the risk of structural failure caused by stress concentration. This forms a complete set of multi-physical performance response indicators for subsequent structural optimization and safety design evaluation.

[0045] Take the design of anti-interference capacitors as an example. These capacitors are often used to suppress high-frequency noise interference in power supply lines. They place high demands on capacitance stability, electric field uniformity, thermal stability, and structural reliability. During the structural design process, precise control of electrode layout and dielectric thickness is required to avoid breakdown failure due to electric field concentration. By establishing the governing equations for electric field distribution and setting electrode boundary conditions, simulation analysis of the potential distribution and electric field strength under different structural configurations is performed to evaluate the stability of the electrical response under high-frequency interference conditions. Furthermore, thermal conduction modeling considers the thermal resistance relationship between the package case, leads, and potting material, and a heat source term is set at the actual operating current to simulate heat accumulation. Temperature field analysis identifies hotspots and optimizes heat conduction paths to minimize the impact of localized temperature rise on electrical performance. During the mechanical modeling phase, stress boundaries that the device may experience during welding, transportation, and vibration are considered. Stress concentration analysis is performed at the electrode-dielectric interface to identify risk points that could lead to microcracks or interlayer separation. After coupling the electric, thermal and mechanical field data, a unified control equation is constructed to evaluate the comprehensive electro-thermal and mechanical response behavior of the anti-interference capacitor under interference conditions, providing a structural optimization basis and simulation verification means to improve its reliability and service life in electromagnetic compatibility applications.

[0046] In a specific embodiment, the process of executing step S4 may specifically include the following steps: Select and assign weights to target parameters corresponding to performance indicators, and establish an objective function expression; Construct a set of constraint functions based on the structural parameter range, performance requirements and engineering constraints; Input the objective function expression and the constraint function set into the optimization control module, and set the initial population or search point; According to the preset iteration upper limit and convergence criterion, the optimization algorithm is driven to iteratively search in the solution space until the termination condition is met; The solution set that meets the convergence condition in the iterative results is screened to obtain the global optimal solution parameter combination.

[0047] Specifically, based on the multiple sets of structural response data obtained in the performance extraction stage, target parameters with design-dominant significance are selected from the performance index data set, such as capacitance value, electric field uniformity, maximum temperature rise or stress distribution range, etc., and multiple target parameters are weighted according to the design intention to construct a weighted linear or nonlinear objective function expression. The objective function is input into the optimization module in standard mathematical form to drive the search direction. At the same time, the tendency of the optimization target is controlled in combination with the application scenario to ensure the balance between multiple objectives and the adjustability of the design constraints.

[0048] By combining the domain boundaries of structural parameters, the physical limits of materials, process manufacturing capabilities, and thermal, electrical, and mechanical safety specifications, all possible ranges of structural variables and performance indicators are organized and consolidated to construct a multidimensional set of constraint functions for engineering feasibility. These functions are used to restrict the search paths for parameter combinations and performance responses during the optimization process, preventing non-physical structures or unfeasible designs from entering the calculation process. The objective function and constraint functions are input into the optimization control module, and population initialization parameters or search starting points are set, including control variables such as parameter dimensions, number of individuals, and initial range, to establish the initial solution space distribution for subsequent iterative calculations.

[0049] During the optimization algorithm operation phase, the iterative process is executed according to the set maximum number of iterations, convergence error threshold and fitness change rules, and multiple rounds of structural parameter space search are carried out. The simulation module is called to perform performance calculation and objective function evaluation on the parameter group generated in each round. The iteration history is subjected to path tracking and convergence analysis. High-quality solution groups that meet the convergence criteria are screened out from the entire solution set, and solution points that do not meet the constraints or have large fluctuations are removed. The structural parameter combination with stable convergence and optimal performance is output, and this solution group is used as a recommended solution for performance verification and engineering design.

[0050] In a specific embodiment, executing the preset iteration upper limit and convergence criterion to drive the optimization algorithm to iteratively search in the solution space until the termination condition is met may specifically include the following steps: Set the optimization algorithm type and parameter configuration, initialize the algorithm's initial search space, and obtain the optimization starting state data set; Judging a preset convergence control condition to obtain a set of optimization process termination judgment criteria, wherein the preset convergence control condition includes an error threshold and the iteration upper limit; Performing a search process based on the optimization algorithm and performing an objective function evaluation on the structural parameter combination to obtain a fitness score data set, while introducing a proxy model for auxiliary prediction to obtain a candidate evaluation result; Analyze the current convergence region and perform local perturbation sampling on the solution space, and then use the Bayesian update mechanism to generate a finely optimized set of candidate solutions; The iterative path and intermediate solutions of each round are recorded, and the optimization convergence trajectory and learning scheduling history are analyzed to obtain the optimized execution path dataset.

[0051] Specifically, the execution process is completed by combining an improved multi-objective particle swarm optimization algorithm with a proxy model auxiliary mechanism. The specific implementation goal is to minimize the multi-objective function: ,in , represents the capacitor structure design parameter vector, such as metallization layer thickness, dielectric layer thickness, plate spacing, etc., Indicates the Each objective function is defined based on the response data obtained from multi-physics simulation, and the physical dimensions remain consistent. For example, the unit of capacitance is , the electric field strength is V / m, the temperature is K, and the mechanical stress is Pa.

[0052] For global optimization, set the algorithm population size N, search dimension m, and initialize the speed and location ,in , the search space is based on the upper and lower limits of the design parameters set up.

[0053] Set convergence criteria including maximum number of iterations The rate of change of the objective function is less than the threshold , For the The objective function is related to the design parameters at the iteration The value of Indicates the The objective function is about The value of , that is, satisfies: .

[0054] The MOPSO iterative formula is used to update the velocity and position of each particle: , in, is the inertia factor, For particles In the iteration The velocity vector at step time, For particles In the iteration The updated velocity vector at step time, and is the learning factor, for Random numbers in the interval, is the historical optimal position of the particle, is the current global optimal position. For particles In the iteration The position vector at the time of the step directly corresponds to the capacitor design parameters. After each round of iteration, the new structural parameters Input simulation module and calculate its corresponding , construct a fitness score set.

[0055] A surrogate model is further introduced to reduce simulation overhead, and a radial basis function neural network (RBF-NN) is used to fit the objective function response surface: ,in, is the Gaussian kernel function, As the center point, is the weight, Indicates the Proxy estimates of performance indicators.

[0056] Select the optimal solution set in the current convergence region and resample the solution space in combination with the Upper Confidence Bound (UCB) criterion in Bayesian optimization: ,in, and are the predicted mean and standard deviation of the surrogate model for the current design point, is a parameter that controls the degree of exploration. Figure 2 As shown in the figure, the Bayesian optimization sampling process shows an evolutionary trend from extensive exploration to hotspot focusing, and the sampling distribution gradually concentrates on the potential optimal area in the parameter space, enhancing the intelligent exploration capability of the understanding space.

[0057] Record the path history of each iteration , and build a scheduling strategy function based on the fitness trend: , through analysis Rate of change adjustment Parameters are used to achieve dynamic adjustment and convergence control of algorithm behavior. Figure 3 As shown in the figure, the improved optimization algorithm is superior to the original algorithm in terms of objective function convergence speed and stability, which verifies the global search capability and iteration efficiency improvement effect after the collaborative introduction of the scheduling mechanism and the agent model.

[0058] In a specific embodiment, the process of executing step S5 may specifically include the following steps: Extracting structural parameters and performance indicators from the global optimal solution parameter combination to form an optimal solution candidate set; Calculating a relative error value based on the prediction performance of the optimal solution candidate set and the simulation verification results; Compare the relative error value with the set accuracy threshold to determine whether the prediction accuracy meets the convergence requirement, and return the solution set that does not meet the accuracy requirement to the optimization module to re-execute the iterative process; The optimal parameter configuration that meets the error range and the performance requirements is used as the structural parameter configuration.

[0059] Specifically, the optimal structural parameter combination output by the optimization algorithm is decoded and restored, and its physical field response results are re-evaluated in combination with the simulation model, and key performance indicators such as capacitance value, electric field strength, thermal distribution and stress value are simultaneously extracted. The predicted output of the proxy model is compared, and the relative error percentage of each performance dimension is calculated item by item to generate an error vector data set. The error value is then compared dimension by dimension with the pre-set accuracy threshold. The candidate solutions with the predicted performance credibility that meets the standard are screened out through Boolean judgment, and the samples with a high degree of consistency between the predicted results and the actual simulation response are marked, and a validity judgment mechanism for the optimization results is established.

[0060] The structural parameter combinations that fail the error tolerance judgment are returned to the input port of the optimization module, inheriting the path information and convergence trajectory of the previous round. The algorithm iteration process is restarted under the same objective function and constraints, maintaining search diversity and population activity to avoid falling into local optimality, and dynamically adjusting the sampling samples and training frequency of the proxy model to strengthen the learning density of the error sample area, drive the algorithm to focus on re-optimizing the error-sensitive areas, and improve the matching stability between the prediction model and the physical simulation.

[0061] For structural parameter groups that pass the convergence accuracy assessment and meet the performance target constraints, the results are calibrated and recorded, and automatically archived to the structural optimization results set. A standardized design configuration document and an input interface file that can be directly called by the modeling platform are simultaneously generated. This file associates the objective function value, performance indicator details, error vectors, and optimization path identifiers, supporting visualization, simulation reproduction, and result traceability. Ultimately, the final version of the structural parameters is generated at the structural configuration output end, balancing functionality, accuracy, and feasibility.

[0062] Taking the design of axial IGBT absorption capacitors as an example, the convergence verification and output configuration of structural optimization parameters are achieved through the above-mentioned S5 steps, which has a clear implementation path in engineering. For this type of capacitor, its working environment involves high-frequency and high-voltage pulse conditions, and capacitance value, electric field uniformity, thermal stability and mechanical reliability are key performance indicators. After the optimization is completed, the optimal solution extracted includes multiple parameter combinations such as electrode spacing, metallization layer thickness, dielectric film stacking structure, etc. By comparing with the multi-physics field simulation results, the actual value and proxy prediction value corresponding to the objective function are obtained respectively, and the relative error percentage is calculated for each target dimension. For example, the capacitance value error is controlled within 2%, the maximum electric field strength point error does not exceed 1.5%, and the temperature rise prediction deviation is maintained within 3%.

[0063] When the error of some solution groups exceeds the preset convergence threshold, the system sends the group of parameters back to the optimization algorithm module for re-evaluation, and calls new samples in the approximate area to supplement the training of the proxy model to improve the proxy accuracy to avoid "biased convergence". At the same time, the local perturbation mechanism is enabled to generate a new candidate solution set in a small range around the current convergence area for simulation evaluation to ensure that the optimization process can continue to converge to a more accurate solution, and continuously collect information from the error hotspot area to feed back to the model training layer.

[0064] For optimal structural combinations that meet error requirements, such as a solution with a metallized electrode thickness of 0.75μm, a dielectric film thickness of 4.2μm, and a plate spacing of 1.1mm, simulation verification demonstrates high-frequency dielectric stability, electric field distribution uniformity, and stress release capabilities that meet IGBT application requirements. The system then marks this configuration as the final optimization result output and associates all corresponding simulation data, optimization path information, and error records, archiving it as a reproducible design input that can be directly exported to CAD modeling or process development platforms, achieving an integrated closed-loop design-verification-mass production process. This process improves design efficiency while ensuring the physical feasibility and high reliability of the design solution.

[0065] The above describes the capacitor structure design optimization method in the embodiment of the present application. The following describes the capacitor structure design optimization method system in the embodiment of the present application. Figure 4 In one embodiment of the capacitor structure design optimization method system of the present application, the following embodiments are included: The structural modeling module 201 is used to model capacitor structural parameters and discretely divide and combine the capacitor structural parameters in the model space to obtain a structural configuration model data set. The capacitor structural parameters include geometric parameters and material properties.

[0066] The simulation analysis module 202 is used to set input conditions according to the geometric parameters and material properties, and obtain a calculation data set through a preset multi-physics field simulation model.

[0067] The performance extraction module 203 is configured to extract the calculated data set to obtain performance index data, wherein the performance index data includes capacitance value, electric field strength, thermal distribution and mechanical stress value.

[0068] The optimization solution module 204 is used to set the objective function and constraint conditions according to the performance indicator data, and then execute the optimization algorithm iteration process to obtain the global optimal solution parameter combination, wherein the optimization algorithm iteration process is: setting the algorithm type and initializing the parameters, introducing the agent model and Bayesian auxiliary algorithm iteration until the iteration termination condition is met to obtain the global optimal solution parameter combination.

[0069] The result evaluation module 205 is used to analyze and process the global optimal solution parameter combination and the performance prediction error, and compare the convergence termination judgment criteria to obtain the structural parameter configuration.

[0070] Through the collaborative cooperation of various components, the system can achieve closed-loop optimization of the entire process of capacitor structure design, including key links such as structural modeling, physical simulation, performance extraction, parameter optimization and result evaluation, effectively improving the degree of automation and performance accuracy of the design. The structural modeling module 201 provides a parametric configuration model as the input basis of the system; the simulation analysis module 202 performs multi-physics field solutions based on the modeling results to obtain multi-dimensional performance data such as electricity, heat, and force; the performance extraction module 203 performs quantitative analysis on the simulation results and extracts key indicators for subsequent optimization; the optimization solution module 204 performs algorithm iteration based on the objective function and constraints to find the optimal structural solution; the result evaluation module 205 performs error verification and convergence judgment on the optimization results to ensure the effectiveness and stability of the structural configuration. Through the integrated operation of the above modules, the system can accurately and efficiently complete the capacitor structure design optimization task, and is suitable for the collaborative design of structural performance under complex constraints.

[0071] above Figure 4 The capacitor structure design optimization method system in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The capacitor structure design optimization device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0072] Reference Figure 5 In an embodiment of the present invention, a capacitor structure design optimization device is also provided. The capacitor structure design optimization device can be a server, and its internal structure can be as follows Figure 5 As shown. The capacitor structure design optimization device includes a processor, a memory, a display screen, an input device, a network interface and a database connected via a system bus. Among them, the computer-designed processor is used to provide computing and control capabilities. The memory of the capacitor structure design optimization device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the capacitor structure design optimization device is used to store the corresponding data in this embodiment. The network interface of the capacitor structure design optimization device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0073] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the capacitor structure design optimization device to which the solution of the present invention is applied.

[0074] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, and when the instructions are run on a computer, the computer executes the steps of the capacitor structure design optimization method.

[0075] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0076] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a capacitor structure design optimization device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A capacitor structure design optimization method, characterized in that: The capacitor structure design optimization method comprises: S1. Modeling capacitor structural parameters and discretely dividing and combining the capacitor structural parameters in a model space to obtain a structural configuration model data set, wherein the capacitor structural parameters include geometric parameters and material properties; S2. Setting input conditions according to the geometric parameters and material properties, and obtaining a calculation data set through a preset multi-physics field simulation model; S3. Obtaining performance index data based on the extraction of the calculated data set, wherein the performance index data includes capacitance value, electric field strength, thermal distribution and mechanical stress value; S4. Setting an objective function and constraints based on the performance indicator data, and then executing an optimization algorithm iteration process to obtain a global optimal solution parameter combination, wherein the optimization algorithm iteration process comprises: setting an algorithm type and initializing parameters, introducing a proxy model and a Bayesian-assisted algorithm, and iterating until an iteration termination condition is satisfied to obtain the global optimal solution parameter combination; S5. Analyze and process the global optimal solution parameter combination and the performance prediction error, and compare the convergence termination judgment criteria to obtain the structural parameter configuration.

2. The capacitor structure design optimization method according to claim 1, characterized in that: Said S1 comprises: The basic configuration of the capacitor structure is analyzed, and a set of adjustable structural parameters is obtained based on geometric parameter feature extraction. Range setting and classification of the parameter set are performed, and a structural variable space is obtained by performing standardization processing based on constraint boundary setting results, wherein the constraint boundary setting results include structural parameter ranges and performance index limits; Discretization is performed on the structural variable space, and multiple groups of parameter combination samples are generated based on the combination of the variable dimensions; Performing modeling based on the parameter combination samples and executing a three-dimensional structure generation process in batches to obtain an initial structure model set; The initial structure model set is converted into a unified format, and the structure configuration model data set is obtained by sorting the readable model data set.

3. The capacitor structure design optimization method according to claim 1, characterized in that: The S2 includes: Extracting parameters of the geometric parameters and material properties, and arranging the parameters according to preset input format requirements; Setting the input conditions based on the extracted input parameters and boundary conditions, and initializing the simulation area grid; Modeling the electric field distribution equation, the heat conduction equation, and the mechanical equilibrium equation respectively, and coupling them to obtain the multi-physics field simulation model; Numerically solving the multi-physics field simulation model and extracting response data of each physical field to obtain simulation results; The simulation result data is formatted, and the results of various performance indicators are marked and summarized to obtain the calculation data set.

4. The capacitor structure design optimization method according to claim 3, characterized in that: The electric field distribution equation, the heat conduction equation and the mechanical equilibrium equation are modeled separately, and coupled to obtain the multi-physics field simulation model, including: Establishing the electric field distribution equation and setting electrode boundary conditions; Establishing the heat conduction equation and setting the material thermal conductivity and heat source terms; Establishing the mechanical equilibrium equation and setting stress boundaries and displacement constraints; Carry out variable association processing for the three types of control equations: electricity, heat and force, and uniformly process the coupled boundary interaction terms; The coupled control equations are packaged, and a multi-physics field simulation solution interface is constructed to obtain the multi-physics field simulation model.

5. The capacitor structure design optimization method according to claim 1, characterized in that: The S3 includes: Classify multi-physics simulation results and extract various types of physical response data; Analyze the electric field distribution data and obtain the capacitance value based on the potential difference and charge density data between electrodes; The gradient of the electric field intensity data is calculated, and then the extreme value areas are marked to obtain the electric field intensity distribution map; Conduct statistical analysis on the temperature field simulation results and obtain thermal distribution parameters based on the temperature rise trend data; The displacement and stress field data are solved, and after identifying the maximum stress area, the mechanical stress value is extracted.

6. The capacitor structure design optimization method according to claim 1, characterized in that: The S4 includes: Select and assign weights to target parameters corresponding to performance indicators, and establish an objective function expression; Construct a set of constraint functions based on the structural parameter range, performance requirements and engineering constraints; Input the objective function expression and the constraint function set into the optimization control module, and set the initial population or search point; According to the preset iteration upper limit and convergence criterion, the optimization algorithm is driven to iteratively search in the solution space until the termination condition is met; The solution set that meets the convergence condition in the iterative results is screened to obtain the global optimal solution parameter combination.

7. The capacitor structure design optimization method according to claim 6, characterized in that: The optimization algorithm is driven to iteratively search in the solution space according to a preset iteration upper limit and convergence criterion until a termination condition is met, including: Set the optimization algorithm type and parameter configuration, initialize the algorithm's initial search space, and obtain the optimization starting state data set; Judging a preset convergence control condition to obtain a set of optimization process termination judgment criteria, wherein the preset convergence control condition includes an error threshold and the iteration upper limit; Performing a search process based on the optimization algorithm and performing an objective function evaluation on the structural parameter combination to obtain a fitness score data set, while introducing a proxy model for auxiliary prediction to obtain a candidate evaluation result; Analyze the current convergence region and perform local perturbation sampling on the solution space, and then use the Bayesian update mechanism to generate a finely optimized set of candidate solutions; The iterative path and intermediate solutions of each round are recorded, and the optimization convergence trajectory and learning scheduling history are analyzed to obtain the optimized execution path dataset.

8. The capacitor structure design optimization method according to claim 1, characterized in that: The S5 includes: Extracting structural parameters and performance indicators from the global optimal solution parameter combination to form an optimal solution candidate set; Calculating a relative error value based on the prediction performance of the optimal solution candidate set and the simulation verification results; Compare the relative error value with the set accuracy threshold to determine whether the prediction accuracy meets the convergence requirement, and return the solution set that does not meet the accuracy requirement to the optimization module to re-execute the iterative process; The optimal parameter configuration that meets the error range and the performance requirements is used as the structural parameter configuration.

9. A capacitor structure design optimization system, used to implement the capacitor structure design optimization method according to any one of claims 1 to 8, characterized in that: The capacitor structure design optimization system includes: a structural modeling module, configured to model capacitor structural parameters and discretely divide and combine the capacitor structural parameters in a model space to obtain a structural configuration model data set, wherein the capacitor structural parameters include geometric parameters and material properties; A simulation analysis module, configured to set input conditions according to the geometric parameters and material properties and obtain a calculation data set through a preset multi-physics field simulation model; A performance extraction module, configured to extract performance index data based on the calculation data set, wherein the performance index data includes capacitance value, electric field strength, thermal distribution and mechanical stress value; An optimization solution module is used to set the objective function and constraints based on the performance indicator data, and then execute the optimization algorithm iteration process to obtain the global optimal solution parameter combination, wherein the optimization algorithm iteration process includes: setting the algorithm type and initializing the parameters, introducing the surrogate model and the Bayesian-assisted algorithm iteration until the iteration termination condition is met to obtain the global optimal solution parameter combination; The result evaluation module is used to analyze and process the global optimal solution parameter combination and the performance prediction error, and compare the convergence termination judgment criteria to obtain the structural parameter configuration.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the processor is enabled to execute the capacitor structure design optimization method according to any one of claims 1 to 8.

Images