Nuclear power container mechanical property iterative optimization method and system

Abstract

The invention provides a nuclear power container mechanical property iterative optimization method and system, and the method comprises the steps: training a mechanical property prediction model based on machine learning based on a pre-constructed historical sample data set, and obtaining a trained mechanical property prediction model; based on the mechanical property prediction model, constructing a quantitative relationship between a mechanical property optimization objective function and design index parameters; on the basis of the quantitative relation, optimization of the design index parameters is executed in a parameter space, and the initial optimal point of the design index parameters is obtained; on the basis of the obtained initial optimal point of the design index parameter and the working condition point in the preset neighborhood, combined with the working condition parameters of random and manual supplementation, obtaining supplementation data; and performing mechanical simulation calculation on the basis of the obtained supplementary data, judging whether the initial optimal point of the design index parameter is locally optimal or not on the basis of a simulation calculation result, if so, introducing the supplementary data into a historical sample data set, and iteratively executing the process until the global optimal point of the design index is obtained.

Description

Technical Field

[0001] The present invention belongs to the technical field of mechanical analysis, and in particular relates to a method and system for iterative optimization of mechanical properties of a nuclear capacitor. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Mechanical analysis and assessment of nuclear secondary and tertiary vessels is an important area in nuclear power engineering design. The nuclear power design specifications stipulate that it is necessary to carry out comprehensive stress analysis and calculation of pressure vessels under design conditions, various operating conditions and test conditions, and to carry out corresponding stress assessment according to the stress limits given in the specifications. The loads that need to be considered for nuclear secondary and tertiary vessels usually include: pressure load, deadweight of components and weight of normal storage materials under operation or test conditions (including additional pressure caused by liquid static pressure head and dynamic pressure head), wind load, vibration and seismic load, pipe load and temperature load, etc. For nuclear secondary and tertiary vessels, the current major international nuclear power design specifications (such as: ASME BPVC-III, RCC-M) give corresponding assessment rules and methods according to the main failure modes of the vessels. Specifically, it can be roughly divided into primary stress analysis, primary plus secondary stress analysis, fatigue analysis and buckling analysis.

[0004] When the vessel calculation fails to meet the assessment requirements, it usually means that some design parameters or conditions do not meet the specification requirements or safety standards. Design engineers need to adjust the design parameters to meet the regulations. When modifying the design parameters, they need to consider all relevant factors and ensure that the modified design still meets all specifications and performance requirements. Usually, these modifications need to be verified through iterative calculations and analysis to verify their effectiveness; in addition, any design changes should be reviewed and approved in detail to ensure that no new problems are introduced.

[0005] The inventors found that the existing solutions have the following problems: the existing technology usually takes meeting the design specifications as the analysis goal, generally does not involve optimization issues, and the mechanical properties of the system and the container cannot reach the optimal level; the analysis process of the existing technology includes conventional engineering analysis processes such as "mechanical calculation-result analysis-design modification", and gradually improves the mechanical properties of the container through manual iteration of the above process, which depends on the technical experience of engineering personnel, has high employee training costs and low work efficiency. Summary of the invention

[0006] The embodiments of the present invention provide a method and system for iterative optimization of the mechanical properties of a nuclear capacitor to solve the problem that the traditional optimization process is heavily dependent on the technical experience of engineers and is inefficient, and the problem that the mechanical properties of the nuclear capacitor are easily trapped in the local optimum and cannot reach the global optimum.

[0007] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0008] In a first aspect, the present invention provides a method for iteratively optimizing the mechanical properties of a nuclear capacitor, comprising:

[0009] Step 1: Based on the pre-constructed historical sample data set, the mechanical property prediction model based on machine learning is trained to obtain a trained mechanical property prediction model;

[0010] Step 2: Based on the trained mechanical property prediction model, a quantitative relationship between the mechanical property optimization objective function and the design index parameters is constructed; based on the obtained quantitative relationship, the design index parameters are optimized in the parameter space to obtain the initial optimal point of the design index parameters;

[0011] Step 3: Based on the obtained initial optimal point of the design index parameters and the operating points within the preset neighborhood, combined with the randomly and manually supplemented operating parameters, obtain supplementary data; perform mechanical simulation calculations based on the obtained supplementary data, and determine whether the initial optimal point of the design index parameters is a local optimal point based on the simulation calculation results. If so, introduce the supplementary data into the historical sample data set, and iterate steps 1 to 3 until the global optimal point of the design index is obtained.

[0012] Furthermore, the construction of the historical sample data set is specifically as follows: based on the problem to be optimized, the simulation objective function value of the nuclear capacitor under different working conditions is performed through a mechanical simulation tool, and based on the design index parameters under different working conditions and their corresponding simulation objective function values, the historical sample data set is constructed.

[0013] Furthermore, the acquisition of the supplementary data is specifically as follows: based on the obtained initial optimal point of the design indicator parameter, combined with the operating point within its preset neighborhood, as the first supplementary data; generating a number of operating data by randomly generating, and combining with a number of predetermined operating data, as the second supplementary data; based on the first supplementary data and the second supplementary data, realizing the construction of the final supplementary data.

[0014] Furthermore, before performing iterative optimization of the mechanical properties of the nuclear capacitor, it is necessary to determine the design index parameters, constraints and optimization objective function that affect the mechanical properties of the nuclear capacitor based on a predetermined optimization problem.

[0015] Furthermore, the design index parameters affecting the mechanical properties of the nuclear capacitor include cylinder geometry parameters, head geometry parameters, support geometry parameters, linear support geometry parameters, nozzle geometry parameters, boundary conditions and load parameters.

[0016] Furthermore, the type of the mechanical property prediction model is BP neural network, generalized linear regression model such as ridge regression, elastic grid, random forest or GBDT integrated algorithm.

[0017] In a second aspect, the present invention provides a nuclear capacitor mechanical properties iterative optimization system, comprising:

[0018] A model training unit, which is used to train a mechanical property prediction model based on machine learning based on a pre-built historical sample data set to obtain a trained mechanical property prediction model;

[0019] The parameter optimization unit is used to construct a quantitative relationship between the mechanical performance optimization objective function and the design index parameters based on the trained mechanical performance prediction model; based on the obtained quantitative relationship, the design index parameters are optimized in the parameter space to obtain the initial optimal point of the design index parameters;

[0020] The iterative optimization unit is used to obtain supplementary data based on the obtained initial optimal point of the design index parameters and the operating points within the preset neighborhood, combined with the randomly and manually supplemented operating parameters; perform mechanical simulation calculations based on the obtained supplementary data, and determine whether the initial optimal point of the design index parameters is a local optimal point based on the simulation calculation results. If so, the supplementary data is introduced into the historical sample data set, and the model training, parameter optimization and iterative optimization are iteratively performed until the global optimal point of the design index is obtained.

[0021] In a third aspect, the present invention provides an electronic device comprising a memory, a processor and a computer program stored and running on the memory, wherein the processor implements the iterative optimization method for the mechanical properties of a nuclear capacitor when executing the program.

[0022] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for iterative optimization of the mechanical properties of a nuclear capacitor.

[0023] In a fifth aspect, the present invention provides a computer program product, characterized in that the computer program product includes a computer program, and when the computer program is executed by a processor, it implements the iterative optimization method of the mechanical properties of a nuclear capacitor.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The present invention provides a method and system for iterative optimization of the mechanical properties of a nuclear capacitor. The scheme constructs a quantitative relationship between a mechanical performance optimization objective function and design index parameters by using a nuclear capacitor mechanical performance model based on machine learning trained with a variety of working condition data, and obtains the optimized design index parameters through automated optimization solution, thereby improving the efficiency and accuracy of the optimization solution and greatly reducing the requirements for the technical experience of engineering personnel. At the same time, by adopting an iterative optimization strategy based on supplementary data, the globally optimal design index parameters can be obtained to avoid falling into the local optimum.

[0026] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0028] Figure 1 It is a schematic diagram of a processing flow of an iterative optimization method for mechanical properties of a nuclear capacitor described in an embodiment of the present invention;

[0029] Figure 2 It is a schematic diagram of the overall business logic of a nuclear capacitor mechanical performance iterative optimization system described in an embodiment of the present invention;

[0030] Figure 3 Schematic diagram of the hardware architecture of a nuclear capacitor mechanical property iterative optimization system described in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0032] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0033] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0034] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0035] When the vessel calculation fails to meet the assessment requirements, it usually means that some design parameters or conditions do not meet the specification requirements or safety standards. The design engineer needs to adjust the design to meet the regulations. The following are some common suggested modifications and input parameters that may need to be modified:

[0036] Increase material thickness: Reduce stress levels by increasing the thickness of the material.

[0037] Change material type: Select a material with higher strength or better corrosion resistance to accommodate high stress or corrosive environments.

[0038] Optimize geometric design: adjust the geometry of the cylinder and head, such as changing the type of head or the diameter-to-length ratio of the cylinder to improve stress distribution.

[0039] Strengthen the supporting structure: Strengthen the supporting structure, such as increasing the number of supports or changing the type and layout of supports, to improve the overall stability of the structure.

[0040] Adjust nozzle and accessory design: Consider changing the size, location or orientation of the nozzle.

[0041] Add safety factors: Add safety factors to the design to provide additional safety margins, especially in the face of uncertainty or extreme operating conditions.

[0042] When modifying design parameters, it is necessary to consider all relevant factors and ensure that the modified design still meets all specifications and performance requirements. Typically, these modifications need to be verified through iterative calculations and analysis to verify their effectiveness. In addition, any design changes should be reviewed and approved in detail to ensure that no new problems are introduced.

[0043] However, the existing technology takes meeting the design specifications as the analysis goal, generally does not involve optimization issues, and the mechanical properties of the system and the container are not "excellent"; the analysis process of the existing technology includes conventional engineering analysis processes such as "mechanical calculation-result analysis-design modification", and gradually improves the mechanical properties of the container through manual iteration of the above process, which relies on the technical experience of engineering personnel, has high employee training costs and low work efficiency. At the same time, the various design drawings, calculation and analysis reports and other knowledge-based data generated by the existing technology are difficult to manage centrally, and the data cannot be traced after the employee leaves; repeated analysis and repeated design often occur.

[0044] In order to address the problems existing in the existing technology, such as Figure 1 and Figure 2As shown, in one or more embodiments, a method for iterative optimization of mechanical properties of a nuclear capacitor is provided, which specifically includes the following processing steps:

[0045] Step 1: Based on the pre-constructed historical sample data set, the mechanical property prediction model based on machine learning is trained to obtain a trained mechanical property prediction model;

[0046] Among them, the construction of the historical sample data set is specifically: based on the problem to be optimized, the simulation objective function value of the nuclear capacitor under different working conditions is carried out through a mechanical simulation tool, and the historical sample data set is constructed based on the design index parameters under different working conditions and their corresponding simulation objective function values.

[0047] It should be noted that before performing iterative optimization of the mechanical properties of nuclear capacitors, it is necessary to determine the design index parameters, constraints and optimization objective functions that affect the mechanical properties of nuclear capacitors based on the predetermined optimization problem, including:

[0048] Propose the problem to be optimized in advance (common optimization problems are shown in Table 2), and clarify the variable parameters, optimization objectives and constraints;

[0049] Determine the range of variation of each parameter;

[0050] Generate sample data; wherein the generation of the sample data depends on mechanical simulation software, for example, Ansys series software or Siemens series software can be used to analyze and solve the structural mechanics, fluid mechanics and temperature field of the container. This system needs to be configured with a scheduling interface to drive the mechanical simulation software to automatically perform mechanical calculations and obtain the objective function value under the specified working conditions.

[0051] For multi-parameter and multi-objective optimization problems, since there are many variable parameters, it is not possible to exhaustively list all working conditions. The system should arrange the calculation cases reasonably so that the sample data covers its variation range as evenly as possible.

[0052] In a specific implementation, the design index parameters affecting the mechanical properties of the nuclear capacitor include cylinder geometry parameters, head geometry parameters, support geometry parameters, linear support geometry parameters, nozzle geometry parameters, boundary conditions and load parameters. Specifically, the specific parameters are shown in Table 1.

[0053] Table 1 Common input parameters for container mechanical analysis

[0054]

[0055]

[0056] Table 2 Common analysis objectives of container mechanics

[0057]

[0058]

[0059] Step 2: Based on the trained mechanical property prediction model, a quantitative relationship between the mechanical property optimization objective function and the design index parameters is constructed; based on the obtained quantitative relationship, the design index parameters are optimized in the parameter space to obtain the initial optimal point of the design index parameters;

[0060] Specifically, step 2 specifically includes the following processing steps:

[0061] The machine learning-based nuclear capacitor mechanical properties prediction model obtained through training is used to explore the quantitative relationship between the objective function and the variable parameters; wherein, in the specific implementation, the scheme described in this embodiment provides multiple types of machine learning algorithms (for example: BP neural network, generalized linear regression models such as ridge regression, elastic grid, random forest and GBDT integrated algorithm) and algorithm hyperparameter automatic optimization method (for example: particle method, simulated annealing and genetic evolution) to realize the automatic optimization of model algorithm, and effectively improve the prediction accuracy of the model through the automatic optimization of algorithm hyperparameters.

[0062] By executing the optimization of design index parameters, based on the aforementioned nuclear capacitor mechanical properties prediction model, and by providing multiple types of optimization algorithms and automatic optimization methods for algorithm hyperparameters, automatic optimization of optimization algorithms and automatic optimization of algorithm hyperparameters are achieved, effectively improving the convergence speed of the optimization algorithm.

[0063] Through the above model training and parameter optimization steps, the solution described in this embodiment, with the support of the current sample data, obtains the relationship model between the design index parameters and the objective function, and further obtains the initial optimal point of the design index parameters. On the other hand, the machine learning model and the optimization algorithm have certain analysis errors, and the optimal point obtained by the optimization method may also fall into the local optimum. Therefore, the "optimal point of the design index parameters" obtained in this analysis needs further verification.

[0064] Step 3: Based on the obtained initial optimal point of the design index parameters and the operating points within the preset neighborhood, combined with the randomly and manually supplemented operating parameters, obtain supplementary data; perform mechanical simulation calculations based on the obtained supplementary data, and determine whether the initial optimal point of the design index parameters is a local optimal point based on the simulation calculation results. If so, introduce the supplementary data into the historical sample data set, and iterate steps 1 to 3 until the global optimal point of the design index is obtained.

[0065] In a specific implementation, the acquisition of the supplementary data is specifically as follows: based on the obtained initial optimal point of the design indicator parameter, combined with the operating point within its preset neighborhood, as the first supplementary data; generating a number of operating data by random generation, and combining them with a number of predetermined operating data as the second supplementary data; based on the first supplementary data and the second supplementary data, the final supplementary data is constructed.

[0066] Specifically, step 3 includes the following processing steps:

[0067] In order to verify the aforementioned "optimal design parameters", the system automatically designs new working conditions, drives the simulation software to add new examples, supplements data, and realizes iterative optimization. The specific processing includes the following:

[0068] (a) Optimization point supplementation (i.e., acquisition of the first supplementary data): the aforementioned “optimal point of design parameters” and its nearby operating points are defined as newly added operating conditions. After comparison with the existing data to remove duplicates, the simulation software is driven to perform mechanical calculations.

[0069] (b) Random supplementation and manual supplementation (i.e., acquisition of the second supplementary data). The system randomly adds a number of working conditions. Engineers can also instruct the system to add a number of working conditions based on engineering experience. After comparing with the existing data to remove duplicates, the simulation software is driven to perform mechanical calculations.

[0070] (c) Optimal point verification: After the above calculations are completed, the system can compare the objective function obtained at the "optimal point of design parameters" and its nearby operating points to verify whether the point is a local optimum.

[0071] (d) Iterative optimization: the system can repeat all steps from the second to the fourth step based on the new extended data set, and re-execute the model training and parameter optimization steps. If the model accuracy meets the agreed requirements (such as the determination coefficient is about 0.95 or the mean square error is less than 0.05, etc.), and the two optimization conclusions are the same, it can be considered that the influence of the local optimum is excluded with a high probability, and the optimization conclusion is output as the final recommendation.

[0072] In one or more embodiments, Figure 2 As shown, a nuclear capacitor mechanical properties iterative optimization system is provided corresponding to the above method, including:

[0073] A model training unit, which is used to train a mechanical property prediction model based on machine learning based on a pre-built historical sample data set to obtain a trained mechanical property prediction model;

[0074] The parameter optimization unit is used to construct a quantitative relationship between the mechanical performance optimization objective function and the design index parameters based on the trained mechanical performance prediction model; based on the obtained quantitative relationship, the design index parameters are optimized in the parameter space to obtain the initial optimal point of the design index parameters;

[0075] The iterative optimization unit is used to obtain supplementary data based on the obtained initial optimal point of the design index parameters and the operating points within the preset neighborhood, combined with the randomly and manually supplemented operating parameters; perform mechanical simulation calculations based on the obtained supplementary data, and determine whether the initial optimal point of the design index parameters is a local optimal point based on the simulation calculation results. If so, the supplementary data is introduced into the historical sample data set, and the model training, parameter optimization and iterative optimization are iteratively performed until the global optimal point of the design index is obtained.

[0076] Furthermore, the system described in this embodiment corresponds to the method described in Embodiment 1, and its technical details are described in detail in Embodiment 1, so they will not be repeated here.

[0077] In one or more embodiments, a software and hardware support platform is provided, specifically:

[0078] like Figure 3 The figure shows the schematic diagram of the system hardware architecture, in which the mechanical computing platform deploys professional mechanical software for simulation and calculation of production data; the business backend server is mainly used to organize the functions of each module in series; the algorithm server is mainly used to deploy machine learning model algorithms and optimization algorithms; the data server is mainly used to store key analysis data such as sample data, proxy models and optimization results. The system adopts a distributed microservice deployment method, and each module and server is connected to each other through a service interface. The front end logs in to the business backend server through the API gateway to realize human-computer interaction.

[0079] In further embodiments, there is also provided:

[0080] An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in Embodiment 1 is performed. For the sake of brevity, no further description is given here.

[0081] It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0082] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0083] A computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the method described in embodiment 1 is completed.

[0084] A computer program product, characterized in that the computer program product includes a computer program, and when the computer program is executed by a processor, it implements the iterative optimization method of the mechanical properties of a nuclear capacitor.

[0085] The method in the first embodiment can be directly embodied as being executed by a hardware processor, or by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it is not described in detail here.

[0086] Those skilled in the art will appreciate that the units, i.e., algorithm steps, of the various examples described in the present embodiment can be implemented in electronic hardware or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for iterative optimization of mechanical properties of nuclear capacitors, characterized in that: include: Step 1: Based on the pre-constructed historical sample data set, the mechanical property prediction model based on machine learning is trained to obtain a trained mechanical property prediction model; Step 2: Based on the trained mechanical property prediction model, the quantitative relationship between the mechanical property optimization objective function and the design index parameters is constructed; Based on the obtained quantitative relationship, the optimization of the design index parameters is performed in the parameter space to obtain the initial optimal point of the design index parameters; Step 3: Based on the obtained initial optimal point of the design index parameters and the operating points within the preset neighborhood, combined with the randomly and manually supplemented operating parameters, obtain supplementary data; perform mechanical simulation calculations based on the obtained supplementary data, and determine whether the initial optimal point of the design index parameters is a local optimal point based on the simulation calculation results. If so, introduce the supplementary data into the historical sample data set, and iterate steps 1 to 3 until the global optimal point of the design index is obtained.

2. A method for iterative optimization of mechanical properties of a nuclear capacitor according to claim 1, characterized in that: The construction of the historical sample data set is specifically as follows: based on the problem to be optimized, the simulation objective function value of the nuclear capacitor under different working conditions is performed through a mechanical simulation tool, and the historical sample data set is constructed based on the design index parameters under different working conditions and their corresponding simulation objective function values.

3. A method for iterative optimization of mechanical properties of a nuclear capacitor according to claim 1, characterized in that: The acquisition of the supplementary data is specifically as follows: based on the obtained initial optimal point of the design indicator parameter, combined with the operating point within its preset neighborhood, as the first supplementary data; generating a number of operating data by randomly generating, and combining with a number of predetermined operating data as the second supplementary data; based on the first supplementary data and the second supplementary data, realizing the construction of the final supplementary data.

4. A method for iterative optimization of mechanical properties of a nuclear capacitor according to claim 1, characterized in that: Before performing iterative optimization of the mechanical properties of nuclear capacitors, it is necessary to determine the design index parameters, constraints and optimization objective functions that affect the mechanical properties of nuclear capacitors based on a predetermined optimization problem.

5. A method for iterative optimization of mechanical properties of a nuclear capacitor according to claim 1, characterized in that: The design index parameters that affect the mechanical properties of the nuclear capacitor include cylinder geometry parameters, head geometry parameters, support geometry parameters, linear support geometry parameters, nozzle geometry parameters, boundary conditions and load parameters.

6. A method for iterative optimization of mechanical properties of a nuclear capacitor according to claim 1, characterized in that: The type of the mechanical property prediction model is BP neural network, generalized linear regression model such as ridge regression, elastic grid, random forest or GBDT integrated algorithm.

7. A nuclear capacitor mechanical properties iterative optimization system, characterized in that: include: A model training unit, which is used to train a mechanical property prediction model based on machine learning based on a pre-built historical sample data set to obtain a trained mechanical property prediction model; Parameter optimization unit, which is used to construct the quantitative relationship between the mechanical performance optimization objective function and the design index parameters based on the trained mechanical performance prediction model; Based on the obtained quantitative relationship, the optimization of the design index parameters is performed in the parameter space to obtain the initial optimal point of the design index parameters; The iterative optimization unit is used to obtain supplementary data based on the obtained initial optimal point of the design index parameters and the operating points within the preset neighborhood, combined with the randomly and manually supplemented operating parameters; perform mechanical simulation calculations based on the obtained supplementary data, and determine whether the initial optimal point of the design index parameters is a local optimal point based on the simulation calculation results. If so, the supplementary data is introduced into the historical sample data set, and model training, parameter optimization and iterative optimization are iteratively performed until the global optimal point of the design index is obtained.

8. An electronic device comprising a memory, a processor and a computer program stored and running on the memory, characterized in that: When the processor executes the program, an iterative optimization method for the mechanical properties of a nuclear capacitor as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a method for iteratively optimizing the mechanical properties of a nuclear capacitor as described in any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, characterized in that when the computer program is executed by a processor, it implements a method for iterative optimization of mechanical properties of a nuclear capacitor as described in any one of claims 1 to 6.

Images