A method and system for optimizing the design of a diesel generator set base structure

Abstract

This invention discloses a method and system for optimizing the design of a diesel generator set base structure. The method first establishes a parametric model and a stress simulation model using the position, thickness, width, and number of the base's crossbeams as design parameters and stress minimization as the objective. High-efficiency design parameters are selected through global sensitivity analysis, narrowing the optimization space. Subsequently, orthogonal sampling is performed based on the high-efficiency parameter space, and a radial basis function machine learning model is trained using the stress simulation data of the samples to quickly predict stress. Within the particle swarm optimization framework, candidate samples are generated using a random collaborative strategy, and the optimal sample is selected through the machine learning model for final simulation verification. If the results meet the requirements, the optimal base structure is output. This invention significantly reduces the number of time-consuming simulation calls and improves optimization efficiency through the synergy of parameter selection, surrogate models, and intelligent algorithms.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for optimizing the design of a diesel generator set base structure. Background Technology

[0002] Diesel generator sets, as primary emergency power generation equipment, require extremely high operational stability. However, the generator set base is the foundation for the entire equipment's operation and the main component bearing the weight of the entire device. Therefore, minimizing stress in the design of the diesel generator set while bearing the entire load is one of the key requirements of diesel generator set structural design and an important means to improve the service life of the generator set.

[0003] Currently, stress minimization design for diesel generator set base structures is limited to traditional experimental design methods, resulting in insufficient performance improvement and significantly impacting the service life of diesel generator sets. Furthermore, stress simulation for diesel generator sets is time-consuming, and directly using evolutionary algorithms for optimization design would lead to enormous simulation evaluation costs, making it impossible to obtain a satisfactory diesel generator set base structure within an acceptable design cycle.

[0004] In recent years, machine learning technology has been widely applied in structural optimization scenarios in engineering due to its ability to provide accurate predictions for unknown samples by mining potential patterns based on historical simulation data. Therefore, the accurate predictive power of machine learning can be combined with the global convergence capability of evolutionary algorithms to accelerate the entire optimization process while reducing the number of time-consuming simulation evaluations.

[0005] Considering that the base of a diesel generator set has a large number of structural design parameters, and that the influence effect values ​​of different structural design parameters are significantly different, how to optimize the design of the diesel generator set base structure has become an urgent problem to be solved. Summary of the Invention

[0006] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method and system for optimizing the design of the base structure of a diesel generator set, which aims to solve the above-mentioned problems described in the prior art.

[0007] The first aspect of the present invention is to provide a method for optimizing the design of a diesel generator set base structure, the method comprising:

[0008] Using the position, thickness, width, and number of crossbeams of the diesel generator set base as design parameters, and with stress minimization as the optimization objective, a parametric modeling and stress simulation model of the diesel generator set base are constructed, and a corresponding mathematical optimization problem is output.

[0009] The optimization design space is determined based on the range of values ​​of the design parameters. Within the optimization design space, the sensitivity effect value of each design parameter on stress is calculated using a global sensitivity analysis method. High-efficiency design parameters are selected based on a preset scoring function.

[0010] Within the subspace determined by the high-efficiency design parameters in the optimized design space, a sample dataset of diesel generator set bases is obtained based on orthogonal sampling. Stress simulation is performed on each diesel generator set base in the sample dataset, and the simulation results are used as training data to construct a radial function machine learning model for stress prediction.

[0011] Within the particle swarm optimization framework, a random cooperative strategy is used to generate candidate samples. The stress value of the candidate samples is predicted by the radial function machine learning model, and the best candidate sample with the minimum stress value is searched and determined.

[0012] Stress simulation is performed on the diesel generator set base corresponding to the best candidate sample. If the simulation results meet the preset requirements, the diesel generator set base with the optimal structure is output.

[0013] According to one aspect of the above technical solution, the expression for the mathematical optimization problem is:

[0014] ;

[0015] In the formula, P is the design parameter matrix in the mathematical optimization problem corresponding to the optimal design of the diesel generator set base structure. N The number of beams, and The first and the second N The positional parameters of each beam and The first and the second N The thickness parameters of each beam, and The first and the second N The width parameters of each beam, For stress, This is an expression for calculating stress based on the design parameter matrix P. The upper and lower bounds of the design parameters are represented, Find is the optimal solution to the mathematical optimization problem, Min is the minimum stress on the diesel generator set base, and St is the condition constraint.

[0016] According to one aspect of the above technical solution, the optimization design space is determined based on the value range of the design parameters. Within the optimization design space, a global sensitivity analysis method is used to calculate the sensitivity effect value of each design parameter on stress. The step of selecting high-effect design parameters based on a preset scoring function includes:

[0017] Based on the range of values ​​for the design parameters, the optimal design space for the diesel generator set base is determined; within the optimal design space, stress simulation is performed using the stress simulation model based on each design parameter of the diesel generator set base to obtain the corresponding stress response data.

[0018] Based on the stress response data, the sensitivity effect value of each design parameter on the stress is calculated;

[0019] The sensitivity effect value is normalized according to a preset scoring function to obtain the score value of each design parameter;

[0020] Based on the ranking of the scores, high-efficiency design parameters are selected.

[0021] According to one aspect of the above technical solution, the calculation of the sensitivity effect value is achieved through the Sobol global sensitivity analysis method, including:

[0022] Within the optimized design space, the generation dimension based on uniform design is... m*2d sample matrix T ,in, d=3N+ 1 Indicates the total number of design parameters. N The number of beams, m This represents the number of sample points.

[0023] The sample matrix was calculated using the stress simulation model. A , B and AB i The corresponding stress response values, where the sample matrix A From the sample matrix T The sample matrix is ​​composed of the first d columns. B From the sample matrix T After d Columns constitute a matrix AB i By sample matrix A The i Columns replaced with sample matrix B The i Column generation;

[0024] Based on the stress response value, the total sensitivity effect value of each design parameter is calculated using the variance decomposition method, including:

[0025] Calculate the first i The effect variance of each design parameter V i The expression is:

[0026] ;

[0027] In the formula, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, k Represents the column index of the matrix;

[0028] Calculate the first i Expected effect of each design parameter E i The expression is:

[0029] ;

[0030] And, calculate the first i The total sensitivity effect value of each design parameter S i The expression is:

[0031] ;

[0032] In the formula, Representing the stress response matrix With stress response matrix The variance corresponding to the sum matrix.

[0033] According to one aspect of the above technical solution, the step of normalizing the sensitivity effect value according to a preset scoring function to obtain the score value of each design parameter involves using the scoring function. The total sensitivity effect value was normalized and then sorted by score before screening. α These parameters are used as high-efficiency design parameters;

[0034] In the formula, For the first i The total sensitivity effect value of each design parameter For the first l The total sensitivity effect value of each design parameter.

[0035] According to one aspect of the above technical solution, within the particle swarm optimization framework, the steps of generating candidate samples using a random cooperative strategy, predicting the stress values ​​of the candidate samples using the radial function machine learning model, and searching for and determining the optimal candidate sample with the minimum stress value include:

[0036] Under the particle swarm optimization algorithm architecture, several samples of diesel generator set bases are selected from the sample database to perform a random cooperative evolutionary operation and output the corresponding candidate samples.

[0037] The stress values ​​of all candidate samples are predicted using a radial basis function machine learning model, and the best candidate sample corresponding to the minimum stress value is selected.

[0038] The best candidate sample is used to replace the sample with the largest stress value in the sample database to update the sample database.

[0039] According to one aspect of the above technical solution, under the particle swarm optimization algorithm architecture, the steps of selecting several diesel generator set base samples from the sample database for random cooperative evolutionary operations and outputting the corresponding candidate samples are expressed as follows:

[0040] ;

[0041] In the formula, For the first i 1 candidate sample , and All are uniformly distributed random numbers in the interval [0,1]. , This represents the historical best sample corresponding to the current sample under the particle swarm optimization algorithm architecture. This represents the current best sample under the particle swarm optimization algorithm architecture. Indicates the current sample, Indicates the weighting coefficient. This represents the velocity vector corresponding to the current sample. This indicates a sample randomly selected from the sample database.

[0042] A second aspect of the present invention is to provide a diesel generator set base structure optimization design system, applied to the method described in the above-mentioned technical solution, the system comprising:

[0043] The parametric modeling module is used to construct a parametric model and stress simulation model of the diesel generator set base using the position, thickness, width and number of crossbeams of the crossbeams as design parameters and the stress minimization as the optimization objective, and outputs a mathematical optimization problem accordingly.

[0044] The parameter filtering module is used to determine the optimization design space based on the value range of the design parameters, calculate the sensitivity effect value of each design parameter on stress using a global sensitivity analysis method within the optimization design space, and filter out high-effect design parameters according to a preset scoring function.

[0045] The model building module is used to obtain a sample dataset of diesel generator set bases based on orthogonal sampling in the subspace determined according to the high-efficiency design parameters in the optimization design space, perform stress simulation on each diesel generator set base in the sample dataset, and use the simulation results as training data to build a radial function machine learning model for stress prediction.

[0046] The sample search module is used to generate candidate samples using a random collaborative strategy within the particle swarm optimization framework, predict the stress value of the candidate samples using the radial function machine learning model, and search for and determine the best candidate sample with the minimum stress value.

[0047] The simulation output module is used to perform stress simulation on the diesel generator set base corresponding to the best candidate sample. If the simulation result meets the preset requirements, the optimal diesel generator set base is output.

[0048] A third aspect of the present invention is to provide a readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of the method described in the above-described technical solution.

[0049] A fourth aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method described in the above technical solutions.

[0050] Compared with the prior art, the advantages of the diesel generator set base structure optimization design method and system shown in this invention are as follows:

[0051] The diesel generator set base structure optimization design method provided by this invention performs sensitivity analysis on the design parameters of the diesel generator set base structure to ensure that subsequent optimization design only targets high-effect design parameters. It also designs a particle swarm optimization algorithm based on stochastic cooperation to improve the diversity and collaborative ability of optimization samples during the optimization process. Furthermore, it utilizes the predictive ability of the radial basis function machine learning model to help screen high-quality candidate samples, thereby reducing the number of time-consuming stress simulation calls and improving the optimization efficiency of the entire optimization design method. The optimization design method of this invention provides a systematic solution to the diesel generator set base structure optimization problem. Attached Figure Description

[0052] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0053] Figure 1 A flowchart illustrating the diesel generator set base structure optimization design method provided in an embodiment of the present invention;

[0054] Figure 2 The structural block diagram of the diesel generator set base structure optimization design system provided in the embodiment of the present invention is shown. Detailed Implementation

[0055] To make the objectives, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Several embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of the present invention will be more thorough and complete.

[0056] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0058] Example 1

[0059] Please see Figure 1 The first embodiment of the present invention provides a method for optimizing the design of a diesel generator set base structure, the method comprising steps S10-S50:

[0060] Step S10: Using the position, thickness, width, and number of crossbeams of the diesel generator set base as design parameters, and with stress minimization as the optimization objective, parametric modeling and stress simulation model construction are performed on the diesel generator set base, and the corresponding mathematical optimization problem is output.

[0061] Specifically, in this embodiment, the position, thickness, width, and number of each crossbeam of the diesel generator set base are used as design parameters, and the stress of the diesel generator set base is used as the optimization target. Parametric modeling of the diesel generator set base is performed in Pro / E, and stress simulation model of the diesel generator set base is constructed through Ansys Workbench. Finally, the corresponding mathematical optimization problem is given, and the expression is as follows.

[0062] In this embodiment, the expression for the mathematical optimization problem is:

[0063] ;

[0064] In the formula, P is the design parameter matrix in the mathematical optimization problem corresponding to the optimal design of the diesel generator set base structure. N The number of beams, and The first and the second N The positional parameters of each beam and The first and the second N The thickness parameters of each beam, and The first and the second N The width parameters of each beam, For stress, This is an expression for calculating stress based on the design parameter matrix P. The upper and lower bounds of the design parameters are represented, Find is the optimal solution to the mathematical optimization problem, Min is the minimum stress on the diesel generator set base, and St is the condition constraint.

[0065] Step S20: Determine the optimization design space based on the value range of the design parameters, calculate the sensitivity effect value of each design parameter on stress using a global sensitivity analysis method within the optimization design space, and select high-effect design parameters based on a preset scoring function.

[0066] In this embodiment, the steps of determining the optimized design space based on the value range of the design parameters, calculating the sensitivity effect value of each design parameter on stress using a global sensitivity analysis method within the optimized design space, and selecting high-effect design parameters based on a preset scoring function include:

[0067] Based on the range of values ​​for the design parameters, the optimal design space for the diesel generator set base is determined;

[0068] Within the optimized design space, stress simulation is performed using the stress simulation model based on the various design parameters of the diesel generator set base to obtain the corresponding stress response data.

[0069] Based on the stress response data, the sensitivity effect value of each design parameter on the stress is calculated;

[0070] The sensitivity effect value is normalized according to a preset scoring function to obtain the score value of each design parameter;

[0071] Based on the ranking of the scores, high-efficiency design parameters are selected.

[0072] Specifically, the calculation of the sensitivity effect value is achieved through the Sobol global sensitivity analysis method, including:

[0073] Within the optimized design space, the generation dimension based on uniform design is... m*2d sample matrix T ,in, d=3N+ 1 Indicates the total number of design parameters. N The number of beams, m This represents the number of sample points.

[0074] The sample matrix was calculated using the stress simulation model. A , B and AB i The corresponding stress response values, where the sample matrix A From the sample matrix T The sample matrix is ​​composed of the first d columns. B From the sample matrix T After d Columns constitute a matrix AB i By sample matrix A The i Columns replaced with sample matrix B The i Column generation;

[0075] Based on the stress response value, the total sensitivity effect value of each design parameter is calculated using the variance decomposition method, including:

[0076] Calculate the first i The effect variance of each design parameter V i The expression is:

[0077] ;

[0078] In the formula, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, k Represents the column index of the matrix;

[0079] Calculate the first i Expected effect of each design parameter E i The expression is:

[0080] ;

[0081] And, calculate the first i The total sensitivity effect value of each design parameter S i The expression is:

[0082] ;

[0083] In the formula, Representing the stress response matrix With stress response matrix The variance corresponding to the sum matrix.

[0084] In addition, the step of normalizing the sensitivity effect value according to the preset scoring function to obtain the score value of each design parameter is achieved through the scoring function. The total sensitivity effect value was normalized, and the top α parameters were selected as high-effect design parameters by sorting them by score.

[0085] In the formula, For the first i The total sensitivity effect value of each design parameter For the first l The total sensitivity effect value of each design parameter.

[0086] More specifically and easily understood, step S20 includes:

[0087] The first step is to determine the range of design parameters corresponding to the number, position, thickness, and width of each crossbeam that makes up the base, based on the structural connection and stress requirements of the diesel generator set base, and to establish an optimization design space based on the range of values.

[0088] The second step is to obtain the results based on the uniform design within the optimized design space. m *2 d sample matrix T ,in, d The calculation expression is as follows d =3 N +1, m The number of sample points;

[0089] The third step is to convert the matrix T The former d Columns are used to construct matrices A , matrix T After d Columns are used to construct matrices B ;

[0090] Fourth step, based on the matrixA With matrix B Constructing a matrix ,in, In the matrix A Chinese matrix B The i Column substitution matrix A The i column, and i =1,…, d ;

[0091] The fifth step involves evaluating the matrix using a stress simulation model of the diesel generator set base in Ansys Workbench. A ,matrix B and all matrices The stress response matrix of all corresponding diesel generator set bases, where the matrix... A The corresponding stress response matrix is ,matrix B The corresponding stress response matrix is ,matrix The corresponding stress response matrix is ;

[0092] Step 6: Calculate the variance of the effects of each design parameter, for example, the first... i Formula for the variance of the effect of each design parameter as follows:

[0093] ;

[0094] In the above formula, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, k Represents the column index of the matrix;

[0095] Step 7: Calculate the expected effect of each design parameter, for example, the first... i Expected effect formula for each design parameter as follows:

[0096] ;

[0097] Step 8: Calculate the total sensitivity effect value for each design parameter, for example, the first... i Formula for the total sensitivity effect value of each design parameter as follows:

[0098] ;

[0099] In the above formula, Representing the stress response matrix With stress response matrix The variance corresponding to the sum matrix;

[0100] Step nine: Calculate the score value for each design parameter according to the scoring function. For example, the score for the first parameter... i The score value of each design parameter The calculation formula is as follows:

[0101] ;

[0102] In the above formula, Indicates the first l The total sensitivity effect value of each design parameter;

[0103] Step 10: Sort the scores of all design parameters and filter them. α These design parameters are used as high-efficiency design parameters.

[0104] Step S30: In the subspace determined by the high-efficiency design parameters in the optimized design space, a sample dataset of diesel generator set bases is obtained based on orthogonal sampling. Stress simulation is performed on each diesel generator set base in the sample dataset, and the simulation results are used as training data to construct a radial function machine learning model for stress prediction.

[0105] In this embodiment, a radial basis function machine learning model for stress prediction is constructed. Specifically, a high-efficiency design space is determined based on the value range of the pre-determined high-efficiency design parameters, which is equivalent to a subspace in the optimization design space. Then, orthogonal sampling is performed in this high-efficiency design space to obtain a sample dataset of diesel generator set bases. Each sample in the sample dataset represents a diesel generator set base. Then, stress simulation is performed on each sample in the sample database, i.e., the corresponding diesel generator set base, in Ansys Workbench, and the corresponding stress simulation results are obtained. Finally, the radial basis function machine learning model for stress prediction is constructed using the sample database as input and the corresponding stress simulation results in the sample database as output.

[0106] Step S40: Under the particle swarm optimization algorithm framework, a random cooperative strategy is used to generate candidate samples, the stress value of the candidate samples is predicted by the radial function machine learning model, and the best candidate sample with the minimum stress value is searched and determined.

[0107] In this embodiment, within the particle swarm optimization framework, the steps of generating candidate samples using a random cooperative strategy, predicting the stress values ​​of the candidate samples using the radial function machine learning model, and searching for and determining the optimal candidate sample with the minimum stress value include:

[0108] Under the particle swarm optimization algorithm architecture, several samples of diesel generator set bases are selected from the sample database to perform a random cooperative evolutionary operation and output the corresponding candidate samples.

[0109] The stress values ​​of all candidate samples are predicted using a radial basis function machine learning model, and the best candidate sample corresponding to the minimum stress value is selected.

[0110] The best candidate sample is used to replace the sample with the largest stress value in the sample database to update the sample database.

[0111] More specifically and easily understood, step S40 includes:

[0112] The first step, within the particle swarm optimization algorithm architecture, is to select individuals from the sample database to perform a stochastic cooperative evolutionary operation, generating... The number of candidate samples is given by the following formula:

[0113] ;

[0114] In the formula, For the first i 1 candidate sample , and All are uniformly distributed random numbers in the interval [0,1]. , This represents the historical best sample corresponding to the current sample under the particle swarm optimization algorithm architecture. This represents the current best sample under the particle swarm optimization algorithm architecture. Indicates the current sample, Indicates the weighting coefficient. This represents the velocity vector corresponding to the current sample. This represents a sample randomly selected from the sample database;

[0115] The second step is to use a radial basis function machine learning model to predict the stress values ​​of all candidate samples and select the best candidate sample corresponding to the minimum stress value.

[0116] The third step is to replace the sample with the highest stress value in the sample database with the best candidate sample in order to update the sample database.

[0117] Fourth step, return to the first step, and repeat until the number of iterations reaches the target. Stop at this time, output the best candidate sample, and obtain the optimal diesel generator set base structure corresponding to the best candidate sample.

[0118] Step S50: Perform stress simulation on the diesel generator set base corresponding to the best candidate sample. If the simulation result meets the preset requirements, output the diesel generator set base with the optimal structure.

[0119] In this embodiment, a stress simulation model is used to simulate and analyze the stress of the diesel generator set base structure corresponding to the best candidate sample. Specifically, the stress simulation model is used in Ansys Workbench to simulate and analyze the stress of the diesel generator set base structure. If the simulation results meet the requirements, the diesel generator set base with the optimal structure is output; otherwise, the process jumps to step S30 until the simulation results meet the requirements.

[0120] As a preferred example, a benchmark test function is used to illustrate the optimization performance of the diesel generator set base structure optimization design method provided in this embodiment. The expression of the benchmark test function is as follows:

[0121] ;

[0122]

[0123] In the above formula, This represents the objective function that needs to be minimized. This represents the i-th design parameter. This represents the (i+1)th design parameter. n Indicates the number of design parameters. z This indicates the design parameters, and where indicates the range of values.

[0124] The above benchmark test functions are processed through steps S10 to S50 of the diesel generator set base structure optimization design method provided in this embodiment to obtain experimental results.

[0125] To further illustrate this embodiment, the diesel generator set base structure optimization design method shown in this embodiment is compared with the particle swarm optimization algorithm. The maximum number of simulation evaluations in this embodiment is set to 1000, and the experimental results are shown in Table 1. Under the same number of simulation samples, the method of this embodiment achieves an effect close to the global optimum and is significantly better than the particle swarm optimization algorithm, demonstrating that the method shown in this embodiment has high optimization efficiency. It can be considered that the method of this embodiment performs well in the diesel generator set base structure optimization problem.

[0126] Table 1

[0127]

[0128] In summary, the diesel generator set base structure optimization design method provided in this embodiment performs sensitivity analysis on the design parameters of the diesel generator set base structure to ensure that subsequent optimization design only targets high-efficiency design parameters. It also designs a stochastic cooperative particle swarm optimization algorithm to improve the diversity and collaborative ability of optimization samples during the optimization process. Furthermore, it utilizes the predictive power of the radial basis function machine learning model to help screen high-quality candidate samples, reducing the number of time-consuming stress simulation calls and improving the overall optimization efficiency. This embodiment's optimization design method provides a systematic solution to the diesel generator set base structure optimization problem.

[0129] Example 3

[0130] A second embodiment of the present invention provides a diesel generator set base structure optimization design system, applied to the method described in the above technical solution, the system comprising:

[0131] The parameter modeling module 10 is used to construct a parameter model and stress simulation model for the diesel generator set base using the crossbeam position, crossbeam thickness, crossbeam width and crossbeam number as design parameters, and with stress minimization as the optimization objective, and outputs a mathematical optimization problem accordingly.

[0132] The parameter filtering module 20 is used to determine the optimization design space based on the value range of the design parameters, calculate the sensitivity effect value of each design parameter on stress using a global sensitivity analysis method within the optimization design space, and filter out high-effect design parameters according to a preset scoring function.

[0133] The model building module 30 is used to obtain a sample dataset of diesel generator set bases based on orthogonal sampling in the subspace determined according to the high-efficiency design parameters in the optimization design space, perform stress simulation on each diesel generator set base in the sample dataset, and use the simulation results as training data to build a radial function machine learning model for stress prediction.

[0134] The sample search module 40 is used to generate candidate samples using a random cooperative strategy within the particle swarm optimization framework, predict the stress value of the candidate samples using the radial function machine learning model, and search for and determine the best candidate sample with the minimum stress value.

[0135] The simulation output module 50 is used to perform stress simulation on the diesel generator set base corresponding to the best candidate sample. If the simulation result meets the preset requirements, the optimal diesel generator set base is output.

[0136] In summary, the diesel generator set base structure optimization design system provided in this embodiment performs sensitivity analysis on the design parameters of the diesel generator set base structure to ensure that subsequent optimization design only targets high-effect design parameters. It also designs a particle swarm optimization algorithm based on stochastic cooperation to improve the diversity and collaborative ability of optimization samples during the optimization process. Furthermore, it utilizes the predictive power of the radial basis function machine learning model to help screen high-quality candidate samples, reducing the number of time-consuming stress simulation calls and improving the optimization efficiency of the entire optimization design system. This embodiment's optimization design system provides a systematic solution to the diesel generator set base structure optimization problem.

[0137] Example 4

[0138] A fourth embodiment of the present invention provides a readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of the methods described in the above embodiments.

[0139] Example 5

[0140] A fifth embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the methods described in the above embodiments.

[0141] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0142] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A method for optimizing the design of a base structure of a diesel generator set, characterized in that, The method includes: Using the position, thickness, width, and number of crossbeams of the diesel generator set base as design parameters, and with stress minimization as the optimization objective, a parametric model and stress simulation model of the diesel generator set base are constructed, resulting in a corresponding mathematical optimization problem. The expression of the mathematical optimization problem is as follows: ； In the formula, P is the design parameter matrix in the mathematical optimization problem corresponding to the optimal design of the diesel generator set base structure. N The number of beams, and The first and the second N The positional parameters of each beam and The first and the second N The thickness parameters of each beam, and The first and the second N The width parameters of each beam, For stress, This is an expression for calculating stress based on the design parameter matrix P. The upper and lower bounds of the design parameters are indicated; Find is the optimal solution to the mathematical optimization problem; Min is the minimum stress on the diesel generator set base; and St is the condition constraint. The optimization design space is determined based on the range of values ​​of the design parameters. Within the optimization design space, the sensitivity effect value of each design parameter on stress is calculated using a global sensitivity analysis method. High-efficiency design parameters are selected based on a preset scoring function. Within the subspace determined by the high-efficiency design parameters in the optimized design space, a sample dataset of diesel generator set bases is obtained based on orthogonal sampling. Stress simulation is performed on each diesel generator set base in the sample dataset, and the simulation results are used as training data to construct a radial function machine learning model for stress prediction. Within the particle swarm optimization framework, a random cooperative strategy is used to generate candidate samples. The stress value of the candidate samples is predicted by the radial function machine learning model, and the best candidate sample with the minimum stress value is searched and determined. Stress simulation is performed on the diesel generator set base corresponding to the best candidate sample. If the simulation results meet the preset requirements, the diesel generator set base with the optimal structure is output. The steps include: determining the optimized design space based on the value range of the design parameters; calculating the sensitivity effect value of each design parameter on stress using a global sensitivity analysis method within the optimized design space; and selecting high-effect design parameters based on a preset scoring function. Based on the range of values ​​for the design parameters, the optimal design space for the diesel generator set base is determined; Within the optimized design space, stress simulation is performed using the stress simulation model based on the various design parameters of the diesel generator set base to obtain the corresponding stress response data. Based on the stress response data, the sensitivity effect value of each design parameter on the stress is calculated; The sensitivity effect value is normalized according to a preset scoring function to obtain the score value of each design parameter; Based on the ranking of the scores, high-efficiency design parameters are selected.

2. The diesel generator set base structure optimization design method according to claim 1, characterized in that, The sensitivity effect value is calculated using the Sobol global sensitivity analysis method, including: Within the optimized design space, the generation dimension based on uniform design is... m*2d sample matrix T ,in, d=3N+1 Indicates the total number of design parameters. N The number of beams, m This represents the number of sample points. The sample matrix was calculated using the stress simulation model. A , B and AB i The corresponding stress response values, where the sample matrix A From the sample matrix T The sample matrix is ​​composed of the first d columns. B From the sample matrix T After d Columns constitute a matrix AB i By sample matrix A The i Columns replaced with sample matrix B The i Column generation; Based on the stress response value, the total sensitivity effect value of each design parameter is calculated using the variance decomposition method, including: Calculate the first i The effect variance of each design parameter V i The expression is: ； In the formula, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, The stress response matrix is ​​represented as The k Column values, k Represents the column index of the matrix; Calculate the first i Expected effect of each design parameter E i The expression is: ； And, calculate the first i The total sensitivity effect value of each design parameter The expression is: ； In the formula, Representing the stress response matrix With stress response matrix The variance corresponding to the sum matrix.

3. The diesel generator set base structure optimization design method according to claim 2, characterized in that, The step of normalizing the sensitivity effect value according to the preset scoring function to obtain the score value of each design parameter involves using the scoring function. The total sensitivity effect value was normalized and then sorted by score before screening. α These parameters are used as high-efficiency design parameters; In the formula, For the first i The total sensitivity effect value of each design parameter For the first l The total sensitivity effect value of each design parameter.

4. The diesel generator set base structure optimization design method according to claim 1, characterized in that, Within the particle swarm optimization framework, the steps of generating candidate samples using a random cooperative strategy, predicting the stress values ​​of the candidate samples using the radial function machine learning model, and searching for and determining the optimal candidate sample with the minimum stress value include: Under the particle swarm optimization algorithm architecture, several samples of diesel generator set bases are selected from the sample database to perform a random cooperative evolutionary operation and output the corresponding candidate samples. The stress values ​​of all candidate samples are predicted using a radial function machine learning model, and the best candidate sample corresponding to the minimum stress value is selected. The best candidate sample is used to replace the sample with the largest stress value in the sample database to update the sample database.

5. The diesel generator set base structure optimization design method according to claim 4, characterized in that, Under the particle swarm optimization algorithm architecture, the steps of selecting several diesel generator set base samples from the sample database for random cooperative evolutionary operations and outputting the corresponding candidate samples are expressed as follows: ； In the formula, For the first i 1 candidate sample , and All are uniformly distributed random numbers in the interval [0,1]. , This represents the historical best sample corresponding to the current sample under the particle swarm optimization algorithm architecture. This represents the current best sample under the particle swarm optimization algorithm architecture. Indicates the current sample, Indicates the weighting coefficient. This represents the velocity vector corresponding to the current sample. This indicates a sample randomly selected from the sample database.

6. A diesel generator set base structure optimization design system, characterized in that, The system, applicable to the method of any one of claims 1-5, comprises: The parametric modeling module is used to construct a parametric model and stress simulation model of the diesel generator set base using the position, thickness, width and number of crossbeams of the crossbeams as design parameters and the stress minimization as the optimization objective, and outputs a mathematical optimization problem accordingly. The parameter filtering module is used to determine the optimization design space based on the value range of the design parameters, calculate the sensitivity effect value of each design parameter on stress using a global sensitivity analysis method within the optimization design space, and filter out high-effect design parameters according to a preset scoring function. The model building module is used to obtain a sample dataset of diesel generator set bases based on orthogonal sampling within the subspace determined according to the high-efficiency design parameters in the optimization design space, perform stress simulation on each diesel generator set base in the sample dataset, and use the simulation results as training data to build a radial function machine learning model for stress prediction. The sample search module is used to generate candidate samples using a random collaborative strategy within the particle swarm optimization framework, predict the stress value of the candidate samples using the radial function machine learning model, and search for and determine the best candidate sample with the minimum stress value. The simulation output module is used to perform stress simulation on the diesel generator set base corresponding to the best candidate sample. If the simulation result meets the preset requirements, the optimal diesel generator set base is output.

7. A readable storage medium having computer instructions stored thereon, characterized in that, When executed by the processor, this instruction implements the steps of the method as described in any one of claims 1-5.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method as described in any one of claims 1-5.

Images