Electrostatic ring insulation structure parameter optimization method, system, equipment and medium

Abstract

The invention belongs to the technical field of insulation structure parameter optimization, and discloses an electrostatic ring insulation structure parameter optimization method, system, device and medium, and the optimization method comprises the steps: generating a configuration parameter scheme set, carrying out the coupling calculation of the configuration parameter scheme set, extracting an optimization target output value of the configuration parameter scheme set based on a calculation result, constructing a Kriging agent model by using the optimization target output value; initializing a population of the configuration parameter scheme set by using a genetic algorithm, calculating the fitness of the population of the configuration parameter scheme set, and combining the fitness with a tournament selection technology to obtain an optimal solution set of the configuration parameter scheme; and establishing a forward matrix based on the optimal solution set of the configuration parameter scheme, and calculating and determining the optimal insulation structure parameters of the electrostatic ring through the forward matrix. According to the method, the calculation time is greatly shortened, and the optimization efficiency of the electrostatic ring insulation structure parameters is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of insulation structure parameter optimization, and in particular to an electrostatic ring insulation structure parameter optimization method, system, equipment and medium. Background Art

[0002] The electrostatic ring on the valve side of the converter transformer is an annular structure located at the end of the valve side winding of the converter transformer. It is usually made of conductive materials and is used to improve the electric field distribution and provide insulation support. It is of great significance to ensure the safe and stable operation of the transformer. The electrostatic ring can change the distribution of the electric field lines, making the electric field strength more uniform inside the transformer, thereby avoiding problems such as insulation breakdown caused by excessive electric field strength.

[0003] The insulation structure of the electrostatic ring on the valve side of the converter transformer is one of the key factors to ensure the safe and stable operation of the converter transformer. The insulation structure of the electrostatic ring on the valve side of the converter transformer mainly plays the role of isolating high voltage, preventing current leakage and ensuring uniform electric field distribution. A good insulation structure can improve the insulation performance of the transformer and reduce the risk of insulation failure, thereby ensuring the stable operation of the power system.

[0004] Therefore, it is believed that the insulation design of the electrostatic ring is very important for the safe operation of the converter transformer. The optimization algorithms in the existing technology, such as genetic algorithms and particle swarm methods, require too much time cost when solving nonlinear multi-field coupling and cannot meet the actual engineering requirements. At the same time, the traditional empirical method also faces the problems of large workload and low design efficiency when dealing with the numerous structural configuration parameters of the electrostatic ring on the valve side of the converter transformer, and it is difficult to coordinate the optimization of multiple indicators.

[0005] Therefore, how to provide a method, system, equipment and medium for optimizing the structural parameters of an electrostatic ring insulation is a problem that needs to be solved urgently. Summary of the invention

[0006] The embodiments of the present invention provide a method, system, device and medium for optimizing the parameters of an electrostatic ring insulation structure, so as to solve the problems of heavy workload and low design efficiency in the prior art.

[0007] In order to have a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended to be a general review, nor is it intended to identify key / important components or to delineate the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0008] According to a first aspect of an embodiment of the present invention, a method for optimizing parameters of an electrostatic ring insulation structure is provided.

[0009] In one embodiment, a method for optimizing parameters of an electrostatic ring insulation structure includes:

[0010] The cubic sampling technology and the electrostatic ring insulation structure are used to generate a set of configuration parameter schemes, and the configuration parameter scheme set is coupled calculated. The optimization target output value of the configuration parameter scheme set is extracted based on the calculation results, and the Kriging proxy model is constructed using the optimization target output value and the Lagrangian algorithm.

[0011] The genetic algorithm is used to initialize the population of configuration parameter solutions, and the fitness of the population of configuration parameter solutions is calculated using the Kriging agent model. The fitness is combined with the tournament selection technology to obtain the optimal solution set of configuration parameter solutions.

[0012] Based on the multi-attribute decision analysis method and the optimal solution set of configuration parameter schemes, a forward matrix is ​​established. The comprehensive score of each scheme in the configuration parameter scheme set is calculated through the forward matrix, and the optimal insulation structure parameters of the electrostatic ring are determined according to the scoring results.

[0013] In one embodiment, a cubic sampling technique and an electrostatic ring insulation structure are used to generate a configuration parameter solution set, and a coupling calculation is performed on the configuration parameter solution set. An optimization target output value of the configuration parameter solution set is extracted based on the calculation result. The Kriging proxy model is constructed using the optimization target output value and the Lagrangian algorithm, including:

[0014] Predetermine the configuration parameter range of the electrostatic insulation structure, generate a configuration parameter scheme set of the electrostatic ring insulation structure according to the configuration parameter range and the cubic sampling technology, and establish an electrostatic ring insulation structure model based on the configuration parameter scheme set;

[0015] Implement coupling calculation on the electrostatic ring insulation structure model, and use the calculation results to extract the optimized target output value of the configuration parameter scheme set;

[0016] The expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, and the Kriging equation group is generated based on the expected value. The Kriging model is obtained by fitting the Kriging equation group according to the Lagrangian algorithm.

[0017] The configuration parameter scheme set and the optimization target output value are substituted into the Kriging model for fitting, and the Kriging proxy model for the distribution of the configuration parameter scheme set is obtained based on the fitting results.

[0018] In one embodiment, the expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, a Kriging equation group is generated based on the expected value, and the Kriging model is obtained by fitting the Kriging equation group according to the Lagrangian algorithm, including:

[0019] The expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, the initial equation of the Kriging model is established based on the expected value, and the linear optimal estimation operation is performed on the initial equation according to the optimal properties of the Kriging interpolation function method;

[0020] The Lagrangian algorithm is used to transform the linear optimal estimation result to obtain the Kriging expression equation, and the Kriging expression equation is combined with the initial equation to obtain the Kriging equation system;

[0021] The quadratic matrix is ​​obtained by expressing the Kriging equations in matrix form. The weighted coefficient and Lagrangian coefficient matrix are generated according to the quadratic matrix and the Lagrangian algorithm. The Kriging equations are fitted with the weighted coefficient and Lagrangian coefficient matrix to obtain the Kriging model.

[0022] In one embodiment, a genetic algorithm is used to initialize a population of configuration parameter solutions, and a Kriging agent model is used to calculate the fitness of the population of configuration parameter solutions. Combining the fitness with the tournament selection technique to obtain the optimal solution set of configuration parameter solutions includes:

[0023] Configure the genetic parameters of the genetic algorithm, initialize the population of configuration parameter solutions according to the configuration results, and use the Kriging agent model to introduce a penalty function to calculate the individual fitness value and crowding degree of the population of configuration parameter solutions;

[0024] Based on the calculation results of individual fitness values, the configuration parameter scheme set population is sorted by non-dominated level, and the dominated level of each individual in the configuration parameter scheme set population is analyzed according to the non-dominated level sorting results;

[0025] Arrange the configuration parameter scheme set population in ascending order based on the calculation results of individual dominated level and individual crowding degree, and perform crossover mutation operation on the configuration parameter scheme set population according to the arrangement to obtain the configuration parameter scheme set offspring population;

[0026] Based on the configuration parameter scheme set offspring population, the Kriging proxy model and tournament selection technology are used to perform offspring population screening operations, and the optimal solution set of the configuration parameter scheme is determined based on the selection results.

[0027] In one embodiment, based on the configuration parameter scheme set offspring population, the offspring population screening operation is performed using the Kriging agent model and the tournament selection technique, and the optimal solution set of the configuration parameter scheme is determined based on the selection result, including:

[0028] The fitness and crowding degree of the configuration parameter scheme set offspring population are calculated by using the Kriging proxy model, and the configuration parameter scheme set offspring population is combined with the configuration parameter scheme set population to obtain the set population;

[0029] According to the fitness and crowding degree of the offspring population of the configuration parameter scheme, the aggregate population is non-dominated sorted, and the superior and inferior selection operations are performed on the aggregate population using the sorting results and the tournament selection technology, and a screening aggregate population is generated based on the selection results;

[0030] The number of calculation iterations for fitness and crowding is determined by using the screening set population. The number of calculation iterations is compared with the maximum number of iterations. The task completion is determined based on the difference results. The optimal solution set of the configuration parameter scheme is output according to the task completion results.

[0031] In one embodiment, a forward matrix is ​​established based on a multi-attribute decision analysis method and an optimal solution set of configuration parameter solutions, a comprehensive score of each solution in the configuration parameter solution set is calculated through the forward matrix, and the optimal insulation structure parameters of the electrostatic ring are determined according to the score results, including:

[0032] Determine the number of configuration schemes to be evaluated and the evaluation index based on the optimal solution set of configuration parameter schemes, and establish a forward matrix according to the number of configuration schemes to be evaluated and the evaluation index;

[0033] The forward matrix is ​​standardized to obtain the forward matrix, the optimal solution and the worst solution of the evaluation index are defined based on the forward matrix, and the weight of the evaluation index is set using the subjective weighting technique;

[0034] The weights are used to calculate the distances between the configuration scheme to be evaluated and the optimal solution and the worst solution of the evaluation index, and the comprehensive score of the configuration scheme to be evaluated is obtained based on the distance results;

[0035] Based on the comprehensive scoring results and preset requirements, the optimal insulation structure parameters of the electrostatic ring are determined, and the insulation structure is established using the optimal insulation structure parameters to verify the insulation performance of the electrostatic ring.

[0036] In one embodiment, the calculation formula for the comprehensive score is:

[0037]

[0038] In the formula, S i represents the comprehensive score of the i-th configuration scheme to be evaluated, represents the distance between the i-th configuration scheme to be evaluated and the optimal solution of the evaluation index, Represents the distance between the i-th configuration scheme to be evaluated and the worst solution of the evaluation index.

[0039] According to a second aspect of an embodiment of the present invention, a system for optimizing parameters of an electrostatic ring insulation structure is provided.

[0040] In one embodiment, an electrostatic ring insulation structure parameter optimization system includes:

[0041] A configuration parameter generation unit is used to generate a configuration parameter solution set by using a cubic sampling technique and an electrostatic ring insulation structure, and to perform coupling calculations on the configuration parameter solution set, extract an optimization target output value of the configuration parameter solution set based on the calculation results, and construct a Kriging proxy model by using the optimization target output value and a Lagrangian algorithm;

[0042] A configuration parameter optimization unit is used to initialize the configuration parameter solution set population using a genetic algorithm, calculate the fitness of the configuration parameter solution set population using a Kriging agent model, and combine the fitness with the tournament selection technology to obtain the optimal solution set of the configuration parameter solution;

[0043] The configuration scheme evaluation unit is used to establish a forward matrix based on the multi-attribute decision analysis method and the optimal solution set of the configuration parameter scheme, calculate the comprehensive score of each scheme in the configuration parameter scheme set through the forward matrix, and determine the optimal insulation structure parameters of the electrostatic ring according to the scoring results.

[0044] In one embodiment, the configuration parameter generating unit includes:

[0045] The insulation structure model building module is used to predetermine the configuration parameter range of the electrostatic insulation structure, generate a configuration parameter solution set of the electrostatic ring insulation structure according to the configuration parameter range and the cubic sampling technology, and establish the electrostatic ring insulation structure model based on the configuration parameter solution set;

[0046] A target output value extraction module is used to perform coupling calculations on the electrostatic ring insulation structure model and extract the optimized target output value of the configuration parameter scheme set using the calculation results;

[0047] The Kriging model generation module is used to define the expected value between the actual response value and the predicted difference value using the Kriging interpolation function method, generate the Kriging equation group based on the expected value, and fit the Kriging equation group according to the Lagrangian algorithm to obtain the Kriging model;

[0048] The Kriging proxy model acquisition module is used to substitute the configuration parameter scheme set and the optimization target output value into the Kriging model for fitting operation, and obtain the Kriging proxy model for the distribution of the configuration parameter scheme set based on the fitting results.

[0049] In one embodiment, the expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, a Kriging equation group is generated based on the expected value, and the Kriging model is obtained by fitting the Kriging equation group according to the Lagrangian algorithm, including:

[0050] The expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, the initial equation of the Kriging model is established based on the expected value, and the linear optimal estimation operation is performed on the initial equation according to the optimal properties of the Kriging interpolation function method;

[0051] The Lagrangian algorithm is used to transform the linear optimal estimation result to obtain the Kriging expression equation, and the Kriging expression equation is combined with the initial equation to obtain the Kriging equation system;

[0052] The quadratic matrix is ​​obtained by expressing the Kriging equations in matrix form. The weighted coefficient and Lagrangian coefficient matrix are generated according to the quadratic matrix and the Lagrangian algorithm. The Kriging equations are fitted with the weighted coefficient and Lagrangian coefficient matrix to obtain the Kriging model.

[0053] In one embodiment, the configuration parameter optimization unit includes:

[0054] The fitness and crowding calculation module is used to configure the genetic parameters of the genetic algorithm, initialize the configuration parameter scheme set population according to the configuration results, and use the Kriging agent model to introduce the penalty function to calculate the individual fitness value and crowding of the configuration parameter scheme set population;

[0055] A dominated level analysis module is used to sort the configuration parameter scheme set population by non-dominated level based on the individual fitness value calculation results, and analyze the dominated level of each individual in the configuration parameter scheme set population according to the non-dominated level sorting results;

[0056] A crossover and mutation operation module is used to arrange the configuration parameter scheme set population in ascending order based on the calculation results of the individual dominated level and the individual crowding degree, and to perform a crossover and mutation operation on the configuration parameter scheme set population according to the arrangement to obtain the configuration parameter scheme set offspring population;

[0057] The optimal solution set acquisition module is used to collect offspring populations based on configuration parameter schemes, perform offspring population screening operations using the Kriging agent model and tournament selection technology, and determine the optimal solution set of the configuration parameter scheme based on the selection results.

[0058] In one embodiment, based on the configuration parameter scheme set offspring population, the offspring population screening operation is performed using the Kriging agent model and the tournament selection technique, and the optimal solution set of the configuration parameter scheme is determined based on the selection result, including:

[0059] The fitness and crowding degree of the configuration parameter scheme set offspring population are calculated by using the Kriging proxy model, and the configuration parameter scheme set offspring population is combined with the configuration parameter scheme set population to obtain the set population;

[0060] According to the fitness and crowding degree of the offspring population of the configuration parameter scheme, the aggregate population is non-dominated sorted, and the superior and inferior selection operations are performed on the aggregate population using the sorting results and the tournament selection technology, and a screening aggregate population is generated based on the selection results;

[0061] The number of calculation iterations for fitness and crowding is determined by using the screening set population. The number of calculation iterations is compared with the maximum number of iterations. The task completion is determined based on the difference results. The optimal solution set of the configuration parameter scheme is output according to the task completion results.

[0062] In one embodiment, the configuration scheme evaluation unit includes:

[0063] A forward matrix establishment module is used to determine the number of configuration schemes to be evaluated and the evaluation index based on the optimal solution set of the configuration parameter scheme, and establish a forward matrix according to the number of configuration schemes to be evaluated and the evaluation index;

[0064] The solution definition module is used to perform standardization processing on the forward matrix to obtain the forward matrix, define the optimal solution and the worst solution of the evaluation index based on the forward matrix, and set the weight of the evaluation index using the subjective weighting technology;

[0065] A comprehensive score obtaining module is used to calculate the distance between the configuration scheme to be evaluated and the optimal solution and the worst solution of the evaluation index by using weights, and obtain the comprehensive score of the configuration scheme to be evaluated according to the distance result;

[0066] The insulation performance verification module is used to determine the optimal insulation structure parameters of the electrostatic ring based on the comprehensive scoring results and preset requirements, and use the optimal insulation structure parameters to establish the insulation structure to verify the insulation performance of the electrostatic ring.

[0067] In one embodiment, the calculation formula for the comprehensive score is:

[0068]

[0069] In the formula, S i represents the comprehensive score of the i-th configuration scheme to be evaluated, represents the distance between the i-th configuration scheme to be evaluated and the optimal solution of the evaluation index, Represents the distance between the i-th configuration scheme to be evaluated and the worst solution of the evaluation index.

[0070] According to a third aspect of an embodiment of the present invention, a computer device is provided.

[0071] In one embodiment, the computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0072] According to a fourth aspect of embodiments of the present invention, a computer-readable storage medium is provided.

[0073] In one embodiment, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0074] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0075] 1. The present invention integrates the characteristics of the Kriging interpolation function method and the multi-attribute decision analysis method based on the genetic algorithm, and obtains a fast multi-objective optimization algorithm suitable for the electrostatic ring insulation structure on the converter transformer valve side. The Kriging interpolation function is introduced to predict the performance indicators corresponding to the configuration parameters of different insulation structures, so as to replace the large number of multi-field coupling solution operations required by the genetic algorithm when calculating the fitness, thereby greatly reducing the calculation time and improving the optimization efficiency.

[0076] 2. The present invention adopts the combined optimization algorithm to optimize the insulation structure, which reduces the optimization time and reduces the difference between the predicted value and the actual calculated value, so that the distribution of the electric field and temperature of the optimized insulation structure is significantly improved.

[0077] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0079] Figure 1 is a flow chart of a method for optimizing parameters of an electrostatic ring insulation structure according to an exemplary embodiment;

[0080] Figure 2 is a principle block diagram of an electrostatic ring insulation structure parameter optimization system according to an exemplary embodiment;

[0081] Figure 3 is a specific implementation flow chart of a fast multi-objective optimization method for an electrostatic ring insulation structure on a converter transformer valve side according to an exemplary embodiment;

[0082] Figure 4 is a multi-objective joint optimization flowchart of the Kriging-NSGA-2 algorithm according to an exemplary embodiment;

[0083] Figure 5 The figure is a schematic diagram showing the structure of a computer device according to an exemplary embodiment. DETAILED DESCRIPTION

[0084] The following description and accompanying drawings fully illustrate the specific embodiments of this article so that those skilled in the art can practice them. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. The scope of the embodiments of this article includes the entire scope of the claims, as well as all available equivalents of the claims. Herein, the terms "first", "second", etc. are only used to distinguish one element from another, without requiring or implying any actual relationship or order between these elements. In fact, the first element can also be called the second element, and vice versa. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the structure, device or equipment including a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also include elements inherent to such structure, device or equipment. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the structure, device or equipment including the elements. Each embodiment is described in a progressive manner herein, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the embodiments can be referred to each other.

[0085] The terms "longitudinal", "lateral", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. in this document indicate the orientation or position relationship based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing this document and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In the description of this document, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a mechanical connection or an electrical connection, it can also be the internal communication of two elements, it can be a direct connection, or it can be an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0086] As used herein, the term "plurality" means two or more than two, unless otherwise specified.

[0087] In this document, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0088] In this article, the term "and / or" is a description of the association relationship between objects, indicating that three relationships may exist. For example, A and / or B means: A or B, or, A and B.

[0089] It should be understood that, although the various steps in the flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the figure may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0090] Each module in the device or system of the present application can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above modules.

[0091] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0092] Figure 1 An embodiment of the electrostatic ring insulation structure parameter optimization method of the present invention is shown.

[0093] In this optional embodiment, the electrostatic ring insulation structure parameter optimization method includes:

[0094] Step S101, using cubic sampling technology and electrostatic ring insulation structure to generate a configuration parameter solution set, and performing coupling calculation on the configuration parameter solution set, extracting the optimization target output value of the configuration parameter solution set based on the calculation result, and constructing a Kriging proxy model using the optimization target output value and Lagrangian algorithm;

[0095] Step S103, using a genetic algorithm to initialize the configuration parameter solution set population, and using a Kriging agent model to calculate the fitness of the configuration parameter solution set population, combining the fitness with the tournament selection technology to obtain the optimal solution set of the configuration parameter solution;

[0096] Step S105, establishing a forward matrix based on the multi-attribute decision analysis method and the optimal solution set of the configuration parameter scheme, calculating the comprehensive score of each scheme in the configuration parameter scheme set through the forward matrix, and determining the optimal insulation structure parameters of the electrostatic ring according to the score results.

[0097] In this optional embodiment, when using cubic sampling technology and the electrostatic ring insulation structure to generate a configuration parameter solution set, and coupling calculation is performed on the configuration parameter solution set, the optimization target output value of the configuration parameter solution set is extracted based on the calculation result, and the Kriging proxy model is constructed using the optimization target output value and the Lagrangian algorithm, the configuration parameter range of the electrostatic insulation structure can be determined in advance, and the configuration parameter solution set of the electrostatic ring insulation structure is generated according to the configuration parameter range and the cubic sampling technology, and the electrostatic ring insulation structure model is established based on the configuration parameter solution set; coupling calculation is performed on the electrostatic ring insulation structure model, and the optimization target output value of the configuration parameter solution set is extracted using the calculation result; the expected value between the actual response value and the predicted difference is defined using the Kriging interpolation function method, the Kriging equation group is generated based on the expected value, and the Kriging equation group is fitted according to the Lagrangian algorithm to obtain the Kriging model; the configuration parameter solution set and the optimization target output value are substituted into the Kriging model for fitting operation, and the Kriging proxy model for the distribution of the configuration parameter solution set is obtained based on the fitting result.

[0098] In this optional embodiment, when using the Kriging interpolation function method to define the expected value between the actual response value and the predicted difference, generating a Kriging equation group based on the expected value, and fitting the Kriging equation group according to the Lagrangian algorithm to obtain the Kriging model, the Kriging interpolation function method can be used to define the expected value between the actual response value and the predicted difference, establish the initial equation of the Kriging model based on the expected value, and perform a linear optimal estimation operation on the initial equation according to the optimal properties of the Kriging interpolation function method; the Lagrangian algorithm is used to transform the linear optimal estimation result to obtain the Kriging expression equation, and the Kriging expression equation is combined with the initial equation to obtain the Kriging equation group; the Kriging equation group is expressed in matrix form to obtain a quadratic matrix, weighting coefficients and Lagrangian coefficient matrices are generated according to the quadratic matrix and the Lagrangian algorithm, and the Kriging equation group is fitted using the weighting coefficients and the Lagrangian coefficient matrix to obtain the Kriging model.

[0099] In this optional embodiment, when a genetic algorithm is used to initialize a set population of configuration parameter solutions, and a Kriging agent model is used to calculate the fitness of the set population of configuration parameter solutions, and the fitness is combined with a tournament selection technique to obtain an optimal solution set of configuration parameter solutions, the genetic parameters of the genetic algorithm can be configured, the set population of configuration parameter solutions is initialized according to the configuration results, and a penalty function is introduced using the Kriging agent model to calculate the individual fitness values ​​and crowding of the set population of configuration parameter solutions; the set population of configuration parameter solutions is sorted by non-dominated ranks based on the calculation results of the individual fitness values, and the dominated ranks of each individual in the set population of configuration parameter solutions are analyzed based on the non-dominated rank sorting results; the set population of configuration parameter solutions is sorted in ascending order based on the calculation results of the individual dominated ranks and the individual crowding, and a crossover mutation operation is performed on the set population of configuration parameter solutions according to the arrangement to obtain a set offspring population of configuration parameter solutions; based on the set offspring population of configuration parameter solutions, a offspring population screening operation is performed using the Kriging agent model and the tournament selection technique, and the optimal solution set of configuration parameter solutions is determined based on the selection results.

[0100] In this optional embodiment, based on the offspring population of the configuration parameter solution set, the Kriging proxy model and the tournament selection technique are used to perform the offspring population screening operation, and when the optimal solution set of the configuration parameter solution is determined based on the selection result, the Kriging proxy model can be used to calculate the fitness and crowding of the offspring population of the configuration parameter solution set, and the offspring population of the configuration parameter solution set and the configuration parameter solution set population are spliced ​​and merged to obtain the set population; the set population is non-dominatedly sorted according to the fitness and crowding of the offspring population of the configuration parameter solution set, and the sorting result and the tournament selection technique are used to perform the superior and inferior selection operation on the set population, and a screening set population is generated based on the selection result; the screening set population is used to determine the number of calculation iterations of the fitness and crowding, the difference between the calculated iteration number and the maximum iteration number is compared, the task completion is determined based on the difference result, and the optimal solution set of the configuration parameter solution is output according to the task completion result.

[0101] In this optional embodiment, when a forward matrix is ​​established based on the multi-attribute decision analysis method and the optimal solution set of the configuration parameter scheme, the comprehensive score of each scheme in the configuration parameter scheme set is calculated through the forward matrix, and the optimal insulation structure parameters of the electrostatic ring are determined according to the scoring results, the number of configuration schemes to be evaluated and the evaluation index can be determined based on the optimal solution set of the configuration parameter scheme, and the forward matrix is ​​established according to the number of configuration schemes to be evaluated and the evaluation index; the forward matrix is ​​standardized to obtain a forward matrix, the optimal solution and the worst solution of the evaluation index are defined based on the forward matrix, and the weight of the evaluation index is set using the subjective weighting technology; the distance between the configuration scheme to be evaluated and the optimal solution and the worst solution of the evaluation index is calculated using the weights, and the comprehensive score of the configuration scheme to be evaluated is obtained based on the distance results; the optimal insulation structure parameters of the electrostatic ring are determined based on the comprehensive scoring results and the preset requirements, and the insulation structure is established using the optimal insulation structure parameters to verify the insulation performance of the electrostatic ring.

[0102] In this optional embodiment, when determining the optimal insulation structure parameters of the electrostatic ring based on the comprehensive scoring results and preset requirements, and establishing the insulation structure using the optimal insulation structure parameters to verify the insulation performance of the electrostatic ring, the comprehensive scoring results can be arranged in order of high to low, and a configuration scheme can be selected based on the arrangement results and preset requirements to determine the optimal insulation structure parameters of the electrostatic ring; an insulation structure can be established based on the optimal insulation structure parameters, and the insulation structure can be combined with the electrothermal coupling multi-field distribution calculation technology to verify the insulation performance of the electrostatic ring, and the optimization satisfaction of the optimal insulation structure parameters can be verified based on the insulation performance results.

[0103] Figure 2 An embodiment of an electrostatic ring insulation structure parameter optimization system of the present invention is shown.

[0104] In this optional embodiment, the electrostatic ring insulation structure parameter optimization system includes:

[0105] The configuration parameter generating unit 201 is used to generate a configuration parameter scheme set by using a cubic sampling technique and an electrostatic ring insulation structure, and to perform coupling calculation on the configuration parameter scheme set, extract an optimization target output value of the configuration parameter scheme set based on the calculation result, and construct a Kriging proxy model by using the optimization target output value and a Lagrangian algorithm;

[0106] The configuration parameter optimization unit 203 is used to initialize the configuration parameter solution set population using a genetic algorithm, calculate the fitness of the configuration parameter solution set population using a Kriging agent model, and combine the fitness with the tournament selection technology to obtain the optimal solution set of the configuration parameter solution;

[0107] The configuration scheme evaluation unit 205 is used to establish a forward matrix based on the multi-attribute decision analysis method and the optimal solution set of the configuration parameter scheme, calculate the comprehensive score of each scheme in the configuration parameter scheme set through the forward matrix, and determine the optimal insulation structure parameters of the electrostatic ring according to the scoring results.

[0108] In one embodiment, the configuration parameter generation unit 201 includes: an insulation structure model establishment module (not shown in the figure), a target output value extraction module (not shown in the figure), a Kriging model generation module (not shown in the figure) and a Kriging proxy model acquisition module (not shown in the figure), wherein:

[0109] The insulation structure model building module is used to predetermine the configuration parameter range of the electrostatic insulation structure, generate a configuration parameter solution set of the electrostatic ring insulation structure according to the configuration parameter range and the cubic sampling technology, and establish the electrostatic ring insulation structure model based on the configuration parameter solution set;

[0110] A target output value extraction module is used to perform coupling calculations on the electrostatic ring insulation structure model and extract the optimized target output value of the configuration parameter scheme set using the calculation results;

[0111] The Kriging model generation module is used to define the expected value between the actual response value and the predicted difference value using the Kriging interpolation function method, generate the Kriging equation group based on the expected value, and fit the Kriging equation group according to the Lagrangian algorithm to obtain the Kriging model;

[0112] The Kriging proxy model acquisition module is used to substitute the configuration parameter scheme set and the optimization target output value into the Kriging model for fitting operation, and obtain the Kriging proxy model for the distribution of the configuration parameter scheme set based on the fitting results.

[0113] In this optional embodiment, when using the Kriging interpolation function method to define the expected value between the actual response value and the predicted difference, generating a Kriging equation group based on the expected value, and fitting the Kriging equation group according to the Lagrangian algorithm to obtain the Kriging model, the Kriging interpolation function method can be used to define the expected value between the actual response value and the predicted difference, establish the initial equation of the Kriging model based on the expected value, and perform a linear optimal estimation operation on the initial equation according to the optimal properties of the Kriging interpolation function method; the Lagrangian algorithm is used to transform the linear optimal estimation result to obtain the Kriging expression equation, and the Kriging expression equation is combined with the initial equation to obtain the Kriging equation group; the Kriging equation group is expressed in matrix form to obtain a quadratic matrix, weighting coefficients and Lagrangian coefficient matrices are generated according to the quadratic matrix and the Lagrangian algorithm, and the Kriging equation group is fitted using the weighting coefficients and the Lagrangian coefficient matrix to obtain the Kriging model.

[0114] In this optional embodiment, the configuration parameter optimization unit 203 includes: a fitness crowding degree calculation module (not shown in the figure), a dominated level analysis module (not shown in the figure), a crossover mutation operation module (not shown in the figure) and an optimal solution set acquisition module (not shown in the figure), wherein:

[0115] The fitness and crowding calculation module is used to configure the genetic parameters of the genetic algorithm, initialize the configuration parameter scheme set population according to the configuration results, and use the Kriging agent model to introduce the penalty function to calculate the individual fitness value and crowding of the configuration parameter scheme set population;

[0116] A dominated level analysis module is used to sort the configuration parameter scheme set population by non-dominated level based on the individual fitness value calculation results, and analyze the dominated level of each individual in the configuration parameter scheme set population according to the non-dominated level sorting results;

[0117] A crossover and mutation operation module is used to arrange the configuration parameter scheme set population in ascending order based on the calculation results of the individual dominated level and the individual crowding degree, and to perform a crossover and mutation operation on the configuration parameter scheme set population according to the arrangement to obtain the configuration parameter scheme set offspring population;

[0118] The optimal solution set acquisition module is used to collect offspring populations based on configuration parameter schemes, perform offspring population screening operations using the Kriging agent model and tournament selection technology, and determine the optimal solution set of the configuration parameter scheme based on the selection results.

[0119] In this optional embodiment, based on the offspring population of the configuration parameter solution set, the Kriging proxy model and the tournament selection technique are used to perform the offspring population screening operation, and when the optimal solution set of the configuration parameter solution is determined based on the selection result, the Kriging proxy model can be used to calculate the fitness and crowding of the offspring population of the configuration parameter solution set, and the offspring population of the configuration parameter solution set and the configuration parameter solution set population are spliced ​​and merged to obtain the set population; the set population is non-dominatedly sorted according to the fitness and crowding of the offspring population of the configuration parameter solution set, and the sorting result and the tournament selection technique are used to perform the superior and inferior selection operation on the set population, and a screening set population is generated based on the selection result; the screening set population is used to determine the number of calculation iterations of the fitness and crowding, the difference between the calculated iteration number and the maximum iteration number is compared, the task completion is determined based on the difference result, and the optimal solution set of the configuration parameter solution is output according to the task completion result.

[0120] In this optional embodiment, the configuration scheme evaluation unit 205 includes: a forward matrix establishment module (not shown in the figure), a solution definition module (not shown in the figure), a comprehensive score obtaining module (not shown in the figure) and an insulation performance verification module (not shown in the figure), wherein:

[0121] A forward matrix establishment module is used to determine the number of configuration schemes to be evaluated and the evaluation index based on the optimal solution set of the configuration parameter scheme, and establish a forward matrix according to the number of configuration schemes to be evaluated and the evaluation index;

[0122] The solution definition module is used to perform standardization processing on the forward matrix to obtain the forward matrix, define the optimal solution and the worst solution of the evaluation index based on the forward matrix, and set the weight of the evaluation index using the subjective weighting technology;

[0123] A comprehensive score obtaining module is used to calculate the distance between the configuration scheme to be evaluated and the optimal solution and the worst solution of the evaluation index by using weights, and obtain the comprehensive score of the configuration scheme to be evaluated according to the distance result;

[0124] The insulation performance verification module is used to determine the optimal insulation structure parameters of the electrostatic ring based on the comprehensive scoring results and preset requirements, and use the optimal insulation structure parameters to establish the insulation structure to verify the insulation performance of the electrostatic ring.

[0125] See also Figure 3 to Figure 4 This embodiment proposes a fast multi-objective optimization method for the electrostatic ring insulation structure on the valve side of the converter transformer, which specifically includes three steps:

[0126] Step 1: Kriging-based prediction replaces multi-field calculations: By introducing the Kriging interpolation function method into the insulation structure optimization process, the time-consuming multi-field coupling solution process is replaced by the prediction process of the Kriging proxy model, which greatly improves the computational efficiency of the optimization algorithm.

[0127] Step 2: Combine the fast non-dominated sorting genetic algorithm (NSGA-2) with the genetic algorithm for multi-objective optimization: Since the traditional genetic algorithm cannot directly optimize multiple objectives with different change trends, the fast non-dominated sorting genetic algorithm (NSGA-2) is introduced to improve it, so that the genetic algorithm has the function of completing multi-objective optimization.

[0128] Step 3: Use the TOSIS method to decide on the electrostatic ring insulation structure configuration scheme that best meets the requirements: For the optimal solution set obtained by the joint optimization algorithm, it contains multiple insulation structure parameter combination schemes. The performance indicators corresponding to each configuration scheme have their own advantages and disadvantages. Therefore, it is necessary to select the final scheme according to the design requirements. In order to evaluate and rank each insulation structure configuration scheme, the TOSIS algorithm can be used to calculate the score of each configuration scheme in the Pareto frontier (optimal solution set of configuration parameter schemes) according to different design strategies and the weights of different design indicators, and rank the scores of all schemes in the solution set from high to low, providing an important reference for the selection of the final design scheme.

[0129] In order to facilitate understanding of the above technical solution of this embodiment, the specific implementation process of this embodiment is described in detail below.

[0130] (1) Kriging-based prediction replaces multiple field calculations;

[0131] The Kriging method is also called the Kriging interpolation function method. The core process of introducing the Kriging method to improve the computational efficiency of the traditional optimization algorithm is to replace the time-consuming multi-field coupled finite element calculation process with the prediction process of the Kriging proxy model. The typical characteristics of the Kriging method are unbiasedness and optimality. It can accurately simulate the relationship between input variables and output quantities, thereby accurately interpolating and ultimately outputting the predicted transformer multi-field distribution characteristics under the insulation structure configuration parameters.

[0132] Kriging is a method that predicts n sample values ​​Z(x i ), i = 1, 2, ..., n linear combination to estimate the unknown quantity prediction value Z*(x) at the prediction point x, and the Kriging estimate is:

[0133]

[0134] Among them, w i (x)=[w1(x),w2(x),w3(x),…,w N (x)] T , is the weighted coefficient of the influence of each observation value on the predicted value Z*(x) of the prediction point, N represents the total number of prediction points x, T represents the transposed matrix form of the matrix, and is the operation symbol.

[0135] Kriging is essentially a variant of Gaussian process regression, that is, the regression calculation of random processes. Therefore, for a random process in an arbitrary dimensional space, the response value Z*(x) corresponding to the predicted point x can be expressed as the sum of a linear regression term and a random error term, specifically:

[0136] Z(x)=b T (x)β+ε(x);

[0137] Where b T (x)=[b1(x), b2(x), b3(x),…, b m (x)] T is the basis function, β=[β1, β2, β3,…,β m ] T is the regression coefficient, m represents the quantity, ε(x) represents the data expectation is 0, and the variance is σ 2 The random error term.

[0138] The construction of the Kriging model needs to meet two basic conditions, namely, unbiasedness and optimization. Unbiasedness means that the expectation of the difference between the actual response value and the predicted value should be 0, specifically:

[0139]

[0140] In the formula, E represents the expected value of the difference between the response value and the predicted value, b k Represents the basis function in the response value, and simplifies it to obtain the first equation (initial equation) of the Kriging model, which is:

[0141]

[0142] At the same time, from the optimality of the Kriging method, it can be seen that the variance of the prediction value error is the smallest, and Z*(x) is the linear unbiased optimal estimate, specifically:

[0143]

[0144] In the formula, Var represents variance, Cov represents covariance, and the optimization problem involved in the optimal estimation is transformed into the following form according to the Lagrangian algorithm:

[0145]

[0146] Among them, λ i (i=1,2,3,…,m) are Lagrange multipliers, and for w i (i=1,2,3,…,m) and λ i (i=1, 2, 3, …, m) calculate partial derivatives respectively, m represents the quantity, and obtain the transformed Kriging expression (Kriging equation), which is combined with the first equation of the Kriging model to form the Kriging equation system, which is specifically:

[0147]

[0148] At the same time, the Kriging equations are expressed in matrix form as follows:

[0149]

[0150] Where w = [w1, w2, w3, ..., w N ] T ,λ=[λ1,λ2,λ3,…,λ N ] T , C represents the covariance function matrix between the actual sampling points, B represents the basis function matrix, r(x) represents the covariance matrix between the sampling points and the points to be predicted, b(x) represents the basis function matrix, C, B, r(x) and b(x) are expressed as follows:

[0151]

[0152]

[0153] r(x)=[Cov(Z(x1),Z(x)),Cov(Z(x2),Z(x)),...,Cov(Z(x) N ), Z(x))] T ;

[0154] b(x)=[b1(x), b2(x),..., b m (x)] T ;

[0155] By substituting the input and output data of N sampling points into the covariance function matrix between the actual sampling points, the covariance matrix and the basis function matrix between the sampling points and the points to be predicted can be obtained, thereby completing the fitting of the Kriging equation group and enabling it to have the ability to predict the output value of the specified prediction point x.

[0156] (2) Combined with NSGA-2 algorithm for multi-objective optimization;

[0157] In order to optimize the configuration scheme of the valve-side electrostatic ring insulation structure, this embodiment combines the NSGA-2 algorithm with the Kriging method to optimize multiple optimization objectives, as follows:

[0158] 1. Determine the optimization objectives and constraints of each decision variable, use Latin hypercube sampling method to obtain different insulation structure configuration parameter schemes, model the insulation structure according to different configuration parameter schemes to obtain the insulation structure model, perform multi-field coupling calculations, and extract the corresponding optimization target output value;

[0159] It needs to be explained that the Latin hypercube sampling method is a multidimensional stratified sampling technology. Its essence is to control the position of the sampling points to avoid the overlap of the sampling points in a small neighborhood. The advantage is that the sampling points can be evenly distributed in the sampling space, and the sampling points are random. Therefore, this sampling technology can obtain a smaller estimation variance.

[0160] The specific steps are to use the Latin hypercube sampling method to sample variables in four dimensions: the width of the electrostatic ring, the thickness of the electrostatic ring insulation layer, the arc radius of the upper end of the electrostatic ring, and the arc radius of the upper end of the electrostatic ring insulation layer. Draw multiple groups of electrostatic ring structures according to the different sizes of the structural parameters of the Latin hypercube sampling, establish corresponding insulation structure models for several groups of samples, and carry out multi-field distribution calculations. Perform multi-field coupling calculations on electrostatic ring structures of different sizes, and statistically extract the minimum insulation margin in the oil under AC conditions, the maximum field strength in the oil under DC conditions, and the insulation materials of the electrostatic ring insulation layer under electrostatic ring structures of different sizes.

[0161] 2. Determine the basic parameters of the Kriging function, then substitute the configuration parameter scheme of the insulation structure and the corresponding output (optimization target output value) into the Kriging model for fitting, and obtain the Kriging proxy model that can be used to predict the multi-field distribution corresponding to different insulation structure configuration parameters;

[0162] 3. Configure the basic parameters of the genetic algorithm and initialize the population (configure the parameter scheme set population);

[0163] 4. Call the Kriging model to calculate the fitness value of the individuals in the initial population (the population of the configuration parameter scheme set), and at the same time calculate the crowding degree of each individual. When calculating the fitness of the individuals in the population, add a penalty function to make the individuals meet all the constraints on the one hand, and avoid the configuration scheme that does not conform to the actual situation on the other hand;

[0164] It should be explained that fitness calculation generally adds a penalty function, where the penalty function is:

[0165]

[0166] In the formula, f(x) represents the calculated value of the penalty function, E oil，P Indicates the electric field strength value calculated in oil paper, E max.oc.o represents the maximum electric field strength value, and A represents a constant;

[0167] The fitness fit(x) calculation process is:

[0168] For maxf(x), take fit(x) = f(x) or fit(x) = f(x) - f min , where f min is the lower bound of f(x); for minf(x), take fit(x) = -f(x) or fit(x) = f max -f(x), where f max is an upper bound of f(x).

[0169] The congestion degree is calculated as:

[0170]

[0171] In the formula, CD i represents the crowding degree of the gth individual in the objective function, h(x) represents the objective function value, and x max represents the maximum value of all individuals under the objective function, x min is the minimum value.

[0172] 5. Sort the individuals in the initial population by non-dominated level, and compare them two by two in turn. Sort the dominated level by individual fitness value (convert all individual fitness values ​​into the minimum value type). The initial dominated level of each individual is zero. If an individual is dominated by another individual, its dominated level increases by one.

[0173] 6. Sort the population (configuration parameter scheme set population) from low to high according to the domination level of each individual. If there are two individuals with the same domination level, compare the crowding levels of the two individuals, and the one with the lower crowding level will be ranked higher.

[0174] 7. Perform crossover and mutation operations on the population (configuration parameter scheme set population) to obtain the offspring population (configuration parameter scheme set offspring population), and call the Kriging model on the offspring population (configuration parameter scheme set offspring population) to calculate its fitness value and crowding degree;

[0175] 8. Combine the population (configuration parameter scheme set population) and the offspring population (configuration parameter scheme set offspring population) to obtain the set population, perform non-dominated sorting on the set population again, and then sort the set population by merit through the tournament method to generate the screened population (screened set population);

[0176] 9. Finally, determine whether the calculation has reached the maximum number of iterations. If it has, output the Pareto optimal solution set (optimal solution set for configuration parameter solutions). If it has not, add one to the algebraic number and repeat steps 4 to 9 until the maximum number of iterations is reached and finally output the result (optimal solution set for configuration parameter solutions).

[0177] (3) The TOSIS method (multi-attribute decision analysis method) is used to determine the electrostatic ring insulation structure configuration scheme that best meets the requirements;

[0178] The optimal solution set (optimal solution set of configuration parameter scheme) is obtained by using Kriging and NSGA-2 multi-objective joint optimization, which includes multiple insulation structure parameter combination schemes. The performance indicators corresponding to each configuration scheme have their own advantages and disadvantages. This solution set is the Pareto frontier or Pareto solution set. In order to enable designers to select the final solution according to design requirements, the TOSIS method is used to make decisions to obtain the insulation structure configuration scheme that best meets the requirements. The specific calculation steps of the TOPSIS method are as follows:

[0179] 1. Establish a forward matrix: All optimization target types in the electrostatic ring insulation structure configuration scheme are converted into extremely large targets. If there are d schemes to be evaluated, each scheme has p evaluation indicators, and x represents the prediction point, then the established forward matrix is:

[0180]

[0181] 2. Standardize the forward matrix. The standardized matrix can eliminate the impact of different dimensions of each optimization target on the evaluation calculation. Usually, each element in the forward matrix (e and r represent the number of rows and columns in the matrix, respectively) is divided by the sum of the squares of its column. The standardized calculation method is:

[0182]

[0183] Then the normalized forward matrix (quadratic matrix) is:

[0184]

[0185] 3. Define the optimal solution and the worst solution for each evaluation indicator, that is, determine the maximum and minimum values ​​of each evaluation indicator, specifically:

[0186]

[0187]

[0188] 4. Calculate the distance between each solution and the optimal solution and the worst solution, where u j It represents the weight of each evaluation index. The weight can be either subjective or objective, and is determined by the designer according to his design strategy and needs. represents the distance between the i-th configuration scheme to be evaluated and the optimal solution of the evaluation index, It represents the distance between the i-th configuration scheme to be evaluated and the worst solution of the evaluation index. The specific calculation formula is as follows:

[0189]

[0190]

[0191] Combined with the distance between each solution and the optimal solution and the worst solution, the comprehensive score of each solution is calculated. Finally, the solutions are ranked according to the comprehensive score. The higher the comprehensive score, the better the solution performance.

[0192] The calculation formula for the comprehensive score is:

[0193]

[0194] In the formula, S i represents the comprehensive score of the i-th configuration scheme to be evaluated, represents the distance between the i-th configuration scheme to be evaluated and the optimal solution of the evaluation index, Represents the distance between the i-th configuration scheme to be evaluated and the worst solution of the evaluation index.

[0195] After evaluating and ranking the advantages and disadvantages of each solution in the Pareto front, the transformer designer can then select the configuration solution that meets his design requirements from the recommended solution ranking.

[0196] It also needs to be explained that before optimizing the valve-side electrostatic ring insulation structure, it is necessary to determine the optimization objectives and constraints, and set the value range of the decision variables according to the size limit of the insulation structure. When using the joint optimization algorithm, it is also necessary to set the sampling number for constructing the Kriging model, the population size of the genetic algorithm, the mutation probability, and the number of iterations and other related parameters.

[0197] Taking the electrostatic ring insulation structure as an example: the configuration parameters include the electrostatic ring width, the thickness of the electrostatic ring insulation layer, the arc radius of the upper end of the electrostatic ring, and the arc radius of the upper end insulation layer of the electrostatic ring. The optimal configuration parameter schemes for different structures of the electrostatic ring can be obtained through the above-mentioned fast multi-objective optimization algorithm. The DC electric field distribution and temperature field distribution results before and after optimization are compared. The electric thermal field results calculated by the optimized structure are significantly reduced, which is more conducive to the normal operation of the converter transformer.

[0198] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer 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 computer device is used to store static information and dynamic information data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in the above method embodiment are implemented.

[0199] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0200] In addition, the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiment when executing the computer program.

[0201] In addition, the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0202] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0203] The present invention is not limited to the structures which have been described above and shown in the drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for optimizing the insulation structure parameters of an electrostatic ring, characterized in that: The optimization method includes: The cubic sampling technology and the electrostatic ring insulation structure are used to generate a set of configuration parameter schemes, and the configuration parameter scheme set is coupled calculated. The optimization target output value of the configuration parameter scheme set is extracted based on the calculation results, and the Kriging proxy model is constructed using the optimization target output value and the Lagrangian algorithm. The genetic algorithm is used to initialize the population of configuration parameter solutions, and the fitness of the population of configuration parameter solutions is calculated using the Kriging agent model. The fitness is combined with the tournament selection technology to obtain the optimal solution set of configuration parameter solutions. Based on the multi-attribute decision analysis method and the optimal solution set of configuration parameter schemes, a forward matrix is ​​established. The comprehensive score of each scheme in the configuration parameter scheme set is calculated through the forward matrix, and the optimal insulation structure parameters of the electrostatic ring are determined according to the scoring results.

2. The electrostatic ring insulation structure parameter optimization method according to claim 1 is characterized in that: The method of using cubic sampling technology and an electrostatic ring insulation structure to generate a configuration parameter solution set, performing coupling calculation on the configuration parameter solution set, extracting an optimization target output value of the configuration parameter solution set based on the calculation result, and constructing a Kriging proxy model using the optimization target output value and the Lagrangian algorithm includes: Predetermine the configuration parameter range of the electrostatic insulation structure, generate a configuration parameter scheme set of the electrostatic ring insulation structure according to the configuration parameter range and the cubic sampling technology, and establish an electrostatic ring insulation structure model based on the configuration parameter scheme set; Implement coupling calculation on the electrostatic ring insulation structure model, and use the calculation results to extract the optimized target output value of the configuration parameter scheme set; The expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, and the Kriging equation group is generated based on the expected value. The Kriging model is obtained by fitting the Kriging equation group according to the Lagrangian algorithm. The configuration parameter scheme set and the optimization target output value are substituted into the Kriging model for fitting, and the Kriging proxy model for the distribution of the configuration parameter scheme set is obtained based on the fitting results.

3. The electrostatic ring insulation structure parameter optimization method according to claim 2 is characterized in that: The method of using the Kriging interpolation function to define the expected value between the actual response value and the predicted difference value, generating a Kriging equation group based on the expected value, and fitting the Kriging equation group according to the Lagrangian algorithm to obtain a Kriging model includes: The expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, the initial equation of the Kriging model is established based on the expected value, and the linear optimal estimation operation is performed on the initial equation according to the optimal properties of the Kriging interpolation function method; The Lagrangian algorithm is used to transform the linear optimal estimation result to obtain the Kriging expression equation, and the Kriging expression equation is combined with the initial equation to obtain the Kriging equation system; The quadratic matrix is ​​obtained by expressing the Kriging equations in matrix form. The weighted coefficient and Lagrangian coefficient matrix are generated according to the quadratic matrix and the Lagrangian algorithm. The Kriging model is obtained by fitting the Kriging equations using the weighted coefficient and Lagrangian coefficient matrix.

4. The electrostatic ring insulation structure parameter optimization method according to claim 1 is characterized in that: The method of initializing the configuration parameter scheme set population by using a genetic algorithm, calculating the fitness of the configuration parameter scheme set population by using a Kriging agent model, and combining the fitness with a tournament selection technique to obtain the optimal solution set of the configuration parameter scheme includes: Configure the genetic parameters of the genetic algorithm, initialize the population of configuration parameter solutions according to the configuration results, and use the Kriging agent model to introduce a penalty function to calculate the individual fitness value and crowding degree of the population of configuration parameter solutions; Based on the calculation results of individual fitness values, the configuration parameter scheme set population is sorted by non-dominated level, and the dominated level of each individual in the configuration parameter scheme set population is analyzed according to the non-dominated level sorting results; Arrange the configuration parameter scheme set population in ascending order based on the calculation results of individual dominated level and individual crowding degree, and perform crossover mutation operation on the configuration parameter scheme set population according to the arrangement to obtain the configuration parameter scheme set offspring population; Based on the configuration parameter scheme set offspring population, the Kriging proxy model and tournament selection technology are used to perform offspring population screening operations, and the optimal solution set of the configuration parameter scheme is determined based on the selection results.

5. The electrostatic ring insulation structure parameter optimization method according to claim 4 is characterized in that: The method of performing offspring population screening based on the configuration parameter scheme set and using the Kriging agent model and tournament selection technology, and determining the optimal solution set of the configuration parameter scheme based on the selection result includes: The fitness and crowding degree of the configuration parameter scheme set offspring population are calculated by using the Kriging proxy model, and the configuration parameter scheme set offspring population is combined with the configuration parameter scheme set population to obtain the set population; According to the fitness and crowding degree of the offspring population of the configuration parameter scheme, the aggregate population is non-dominated sorted, and the superior and inferior selection operations are performed on the aggregate population using the sorting results and the tournament selection technology, and a screening aggregate population is generated based on the selection results; The number of calculation iterations for fitness and crowding is determined by using the screening set population. The number of calculation iterations is compared with the maximum number of iterations. The task completion is determined based on the difference results. The optimal solution set of the configuration parameter scheme is output according to the task completion results.

6. The electrostatic ring insulation structure parameter optimization method according to claim 1, characterized in that: The method of establishing a forward matrix based on the multi-attribute decision analysis method and the optimal solution set of the configuration parameter scheme, calculating the comprehensive score of each scheme in the configuration parameter scheme set through the forward matrix, and determining the optimal insulation structure parameters of the electrostatic ring according to the score results includes: Determine the number of configuration schemes to be evaluated and the evaluation index based on the optimal solution set of configuration parameter schemes, and establish a forward matrix according to the number of configuration schemes to be evaluated and the evaluation index; The forward matrix is ​​standardized to obtain the forward matrix, the optimal solution and the worst solution of the evaluation index are defined based on the forward matrix, and the weight of the evaluation index is set using the subjective weighting technique; The weights are used to calculate the distances between the configuration scheme to be evaluated and the optimal solution and the worst solution of the evaluation index, and the comprehensive score of the configuration scheme to be evaluated is obtained based on the distance results; Based on the comprehensive scoring results and preset requirements, the optimal insulation structure parameters of the electrostatic ring are determined, and the insulation structure is established using the optimal insulation structure parameters to verify the insulation performance of the electrostatic ring.

7. The electrostatic ring insulation structure parameter optimization method according to claim 6, characterized in that: The calculation formula of the comprehensive score is: In the formula, S i represents the comprehensive score of the i-th configuration scheme to be evaluated, represents the distance between the i-th configuration scheme to be evaluated and the optimal solution of the evaluation index, Represents the distance between the i-th configuration scheme to be evaluated and the worst solution of the evaluation index.

8. An electrostatic ring insulation structure parameter optimization system, characterized in that: The optimization system includes: A configuration parameter generation unit is used to generate a configuration parameter solution set by using a cubic sampling technique and an electrostatic ring insulation structure, and to perform coupling calculations on the configuration parameter solution set, extract an optimization target output value of the configuration parameter solution set based on the calculation results, and construct a Kriging proxy model by using the optimization target output value and a Lagrangian algorithm; A configuration parameter optimization unit is used to initialize the configuration parameter solution set population using a genetic algorithm, calculate the fitness of the configuration parameter solution set population using a Kriging agent model, and combine the fitness with the tournament selection technology to obtain the optimal solution set of the configuration parameter solution; The configuration scheme evaluation unit is used to establish a forward matrix based on the multi-attribute decision analysis method and the optimal solution set of the configuration parameter scheme, calculate the comprehensive score of each scheme in the configuration parameter scheme set through the forward matrix, and determine the optimal insulation structure parameters of the electrostatic ring according to the scoring results.

9. The electrostatic ring insulation structure parameter optimization system according to claim 8, characterized in that: The configuration parameter generating unit comprises: The insulation structure model building module is used to predetermine the configuration parameter range of the electrostatic insulation structure, generate a configuration parameter solution set of the electrostatic ring insulation structure according to the configuration parameter range and the cubic sampling technology, and establish the electrostatic ring insulation structure model based on the configuration parameter solution set; A target output value extraction module is used to perform coupling calculations on the electrostatic ring insulation structure model and extract the optimized target output value of the configuration parameter scheme set using the calculation results; The Kriging model generation module is used to define the expected value between the actual response value and the predicted difference value using the Kriging interpolation function method, generate the Kriging equation group based on the expected value, and fit the Kriging equation group according to the Lagrangian algorithm to obtain the Kriging model; The Kriging proxy model acquisition module is used to substitute the configuration parameter scheme set and the optimization target output value into the Kriging model for fitting operation, and obtain the Kriging proxy model for the distribution of the configuration parameter scheme set based on the fitting results.

10. The electrostatic ring insulation structure parameter optimization system according to claim 9, characterized in that: The method of using the Kriging interpolation function method to define the expected value between the actual response value and the predicted difference value, generating a Kriging equation group based on the expected value, and fitting the Kriging equation group according to the Lagrangian algorithm to obtain a Kriging model includes: The expected value between the actual response value and the predicted difference is defined by the Kriging interpolation function method, the initial equation of the Kriging model is established based on the expected value, and the linear optimal estimation operation is performed on the initial equation according to the optimal properties of the Kriging interpolation function method; The Lagrangian algorithm is used to transform the linear optimal estimation result to obtain the Kriging expression equation, and the Kriging expression equation is combined with the initial equation to obtain the Kriging equation system; The quadratic matrix is ​​obtained by expressing the Kriging equations in matrix form. The weighted coefficient and Lagrangian coefficient matrix are generated according to the quadratic matrix and the Lagrangian algorithm. The Kriging equations are fitted with the weighted coefficient and Lagrangian coefficient matrix to obtain the Kriging model.

11. The electrostatic ring insulation structure parameter optimization system according to claim 8, characterized in that: The configuration parameter optimization unit comprises: The fitness and crowding calculation module is used to configure the genetic parameters of the genetic algorithm, initialize the configuration parameter scheme set population according to the configuration results, and use the Kriging agent model to introduce the penalty function to calculate the individual fitness value and crowding of the configuration parameter scheme set population; A dominated level analysis module is used to sort the configuration parameter scheme set population by non-dominated level based on the individual fitness value calculation results, and analyze the dominated level of each individual in the configuration parameter scheme set population according to the non-dominated level sorting results; A crossover and mutation operation module is used to arrange the configuration parameter scheme set population in ascending order based on the calculation results of the individual dominated level and the individual crowding degree, and to perform a crossover and mutation operation on the configuration parameter scheme set population according to the arrangement to obtain the configuration parameter scheme set offspring population; The optimal solution set acquisition module is used to collect offspring populations based on the configuration parameter scheme, perform offspring population screening operations using the Kriging agent model and tournament selection technology, and determine the optimal solution set of the configuration parameter scheme based on the selection results.

12. The electrostatic ring insulation structure parameter optimization system according to claim 11, characterized in that: The method of performing offspring population screening based on the configuration parameter scheme set and using the Kriging agent model and tournament selection technology, and determining the optimal solution set of the configuration parameter scheme based on the selection result includes: The fitness and crowding degree of the offspring population of the configuration parameter scheme set are calculated by using the Kriging proxy model, and the offspring population of the configuration parameter scheme set is combined with the configuration parameter scheme set population to obtain the set population; According to the fitness and crowding degree of the offspring population of the configuration parameter scheme, the population is non-dominated sorted, and the superior and inferior selection operations are performed on the population using the sorting results and the tournament selection technology, and a screening population is generated based on the selection results; The number of calculation iterations for fitness and crowding is determined by using the screening set population. The number of calculation iterations is compared with the maximum number of iterations. The task completion is determined based on the difference results. The optimal solution set of the configuration parameter scheme is output according to the task completion results.

13. The electrostatic ring insulation structure parameter optimization system according to claim 8, characterized in that: The configuration scheme evaluation unit comprises: A forward matrix establishment module is used to determine the number of configuration schemes to be evaluated and the evaluation index based on the optimal solution set of the configuration parameter scheme, and establish a forward matrix according to the number of configuration schemes to be evaluated and the evaluation index; The solution definition module is used to perform standardization processing on the forward matrix to obtain the forward matrix, define the optimal solution and the worst solution of the evaluation index based on the forward matrix, and set the weight of the evaluation index using the subjective weighting technology; A comprehensive score obtaining module is used to calculate the distance between the configuration scheme to be evaluated and the optimal solution and the worst solution of the evaluation index by using weights, and obtain the comprehensive score of the configuration scheme to be evaluated according to the distance result; The insulation performance verification module is used to determine the optimal insulation structure parameters of the electrostatic ring based on the comprehensive scoring results and preset requirements, and to establish the insulation structure using the optimal insulation structure parameters to verify the insulation performance of the electrostatic ring.

14. The electrostatic ring insulation structure parameter optimization system according to claim 13, characterized in that: The calculation formula of the comprehensive score is: In the formula, S i represents the comprehensive score of the i-th configuration scheme to be evaluated, represents the distance between the i-th configuration scheme to be evaluated and the optimal solution of the evaluation index, Represents the distance between the i-th configuration scheme to be evaluated and the worst solution of the evaluation index.

15. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images