CN116579040A - Method for optimizing geometry and material parameters of insulator based on finite element simulation - Google Patents

Method for optimizing geometry and material parameters of insulator based on finite element simulation Download PDF

Info

Publication number
CN116579040A
CN116579040A CN202310714205.5A CN202310714205A CN116579040A CN 116579040 A CN116579040 A CN 116579040A CN 202310714205 A CN202310714205 A CN 202310714205A CN 116579040 A CN116579040 A CN 116579040A
Authority
CN
China
Prior art keywords
insulator
parameters
model
simulation
finite element
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310714205.5A
Other languages
Chinese (zh)
Inventor
陈维
王�锋
刘锋
陈振勇
张航
杨瑞
刘康
张广东
吴天存
王翼虎
王海龙
姚永亮
王磊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Gansu Electric Power Co Ltd
Original Assignee
State Grid Gansu Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by State Grid Gansu Electric Power Co Ltd filed Critical State Grid Gansu Electric Power Co Ltd
Priority to CN202310714205.5A priority Critical patent/CN116579040A/en
Publication of CN116579040A publication Critical patent/CN116579040A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
    • G06T17/205—Re-meshing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/04—Power grid distribution networks
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Hardware Design (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Mathematical Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Graphics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Medical Informatics (AREA)
  • Insulators (AREA)

Abstract

The invention belongs to the technical field of insulator structure and material parameter design of power equipment, and discloses a method for optimizing insulator geometry and material parameters based on finite element simulation, which comprises the following steps: s1: explicitly designing indexes and extracting characteristic parameters; s2: determining model parameters; s3: geometric modeling; s4: endowing the model material with properties; s5: meshing; s6: simulating and calculating the distribution of the field intensity on the surface and around the insulator; s7: parameterized scanning simulation calculation is performed, and a simulation data set is established; s8: optimizing and predicting; weighting and normalizing each parameter in a minimum-maximum standardized mode to establish a neural network prediction model; and training the network according to the simulation data set, and inputting the parameter data to be optimized of the insulator to obtain the predicted data after training is completed. The invention can optimally design the geometry and material parameters of the insulator aiming at a specific service environment, and has high calculation efficiency, reliable result and low design period and research and development cost.

Description

Method for optimizing geometry and material parameters of insulator based on finite element simulation
Technical Field
The invention belongs to the technical field of insulator structure and material parameter design of power equipment, and particularly relates to a method for optimizing insulator geometry and material parameters based on finite element simulation.
Background
The insulator transmission line and various electrical equipment have an insulating effect by increasing the creepage distance through a glass or ceramic structure. Insulators are various in variety and shape. Common insulators for transmission main lines comprise porcelain/glass disc type suspension insulators, composite insulators, RTV-coated disc type suspension insulators and the like, and the structure of the common insulators is generally composed of insulator porcelain pieces, steel caps and steel feet. The structure of the insulator is divided into an outer shape (umbrella shape) related to the outer insulation property and an inner structure related to the safety of the inner insulation.
In the actual service of the insulator, the pollution environments such as rainwater, bird droppings and the like cause flashover phenomena to occur frequently, and the safe and stable operation of the power equipment is seriously affected. The existing solutions mostly realize the purposes of large creepage distance and excellent insulating performance by changing parameters of insulator structures such as umbrella skirt radius, shape, inclination angle, materials and the like or changing insulator material properties. However, the conventional design method requires a large amount of engineering experiments to verify, so that a large amount of experiments, high cost and long period become a big pain point for the optimal design of the insulator.
Patent document CN115128407a discloses a method for simulating flashover faults of a V-shaped insulator string based on bird droppings pollution, relates to the technical field of power transmission line maintenance, establishes an environmental model, then simulates the actual running condition of the V-shaped insulator string and the falling state of bird droppings, and sets model parameters, so that the condition that the space electric field changes in the falling process of bird droppings to cause flashover faults is simulated, and provides theoretical basis for fault tracing and optimization design of the V-shaped insulator string; the simulation method of the patent document fully considers the actual structure of the V-shaped insulator string, in the simulation process, not only models and simulates the three-dimensional space around the V-shaped insulator string and the wet pollution accumulated on the surface due to long-term operation are calculated, but also various influencing factors are fully considered, and the calculation accuracy is high. However, the method only aims at fault tracing of the bird droppings flashover problem of the V-shaped insulator string in a wet pollution environment, the multi-environment performance requirement of the insulator string cannot be met in the aspect of model parameter optimization, and meanwhile, the parameterized tracing calculation process of the method adopts a finite element method, so that the calculation speed is very low, and the requirement of rapid iterative design of products cannot be met.
Patent document CN115577558A discloses an umbrella-shaped optimization method of a 220KV hollow porcelain insulator based on COMSOL, which comprises the following steps: s1, setting geometric parameters of a hollow porcelain insulator; s2, importing geometric parameters into COMSOL software to establish a simulation model of the hollow porcelain insulator; s3, performing grid division on the umbrella shape of the hollow porcelain insulator to form a plurality of grid units; s4, setting voltage boundary conditions of electric field distribution, and applying voltage to a simulation model of the hollow porcelain insulator through COMSOL software; s5, calculating an electric field distribution index of each grid unit, and determining the electric field distribution index of the whole hollow porcelain insulator simulation model based on the electric field distribution index of each grid unit; s6, adjusting umbrella skirt parameters of the hollow porcelain insulator, and returning to the step S2; s7, judging whether the geometric parameter adjustment times of the hollow porcelain insulator reach a set value, if so, entering a step S8; s8, selecting geometrical parameters of the hollow porcelain insulator corresponding to the optimal electric field distribution index from multiple simulations as optimal parameters. However, the method of the patent only carries out parameter optimization aiming at umbrella skirt parameters of the insulator, and fails to carry out optimization calculation from the aspect of materials, and meanwhile, the optimization process of the patent also adopts a finite element method, so that the calculation optimization efficiency is low.
Disclosure of Invention
Aiming at the technical problems existing in the prior art, the invention aims to provide a method for optimizing the geometry and material parameters of an insulator based on finite element simulation, which is used for developing an insulator with a novel structure or material according to a specific service environment, intuitively acquiring the insulation performance change caused by the insulator structure optimization through finite element electric field simulation, constructing a neural network prediction model based on neural network deep learning, optimizing the geometry parameters and the material parameters of the insulator, and has the advantages of high calculation efficiency, reliable result and obvious reduction of design period and research and development cost.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a method for insulator geometry and material parameter optimization based on finite element simulation, the method comprising the steps of:
step S1: defining the service design index of the insulator, extracting characteristic parameters which need to be met by the insulation performance of the insulator, and taking the characteristic parameters as excitation conditions of the subsequent simulation step;
step S2: determining model parameters including geometric parameters, material dielectric constant parameters and pollution parameters of the insulator to be optimized;
step S3: geometric modeling;
based on structural design, three-dimensional drawing software is used for establishing a three-dimensional geometric model of the insulator, and the three-dimensional geometric model is imported into finite element analysis software;
step S4: imparting material properties to the model;
step S5: meshing;
step S6: simulation calculation;
adopting a finite element electrostatic field to realize simulation analysis of an insulator parameterized model, and obtaining field intensity distribution on the surface and around the insulator;
step S7: establishing a simulation data set;
according to optimized structure or material parameters in the finite element analysis model of the insulator, parameterized scanning and simulation calculation are carried out to obtain the maximum field intensity value of the fixed-point surface of the insulator, so that a simulation data set is obtained;
step S8: optimizing and predicting;
weighting and normalizing each material parameter and the geometric parameter of the whole model in a minimum-maximum standardized mode, and then establishing a neural network prediction model containing an input layer, an hidden layer and an output layer based on deep learning software of the neural network;
according to the simulation data set obtained in the step S7, training a neural network prediction model; and after the completion, inputting the insulator parameter data to be optimized into the neural network prediction model, and obtaining the prediction data under the corresponding parameters.
Further, in step S1, the extracted feature parameters include at least: under the specified pollution condition, the dry arc distance, the power frequency dry flashover voltage, the power frequency wet flashover voltage and the low frequency breakdown voltage of the insulator are set.
Further, in step S2, the geometric parameters include at least an insulator total height, a foot diameter, a creepage distance, a metal insert depth, and upper, middle, and lower umbrella skirt diameters;
the pollution parameters at least comprise the distribution position and the geometric dimension of the surface pollution of the insulator.
Further, the step S4 specifically includes: the relative dielectric constant and the conductivity of each material in the insulator structure are given according to the physical properties of the insulator physical material.
Further, in step S4, the relative dielectric constant and conductivity of each material in the insulator structure is given, the material including at least insulator umbrella skirt, shielding ring, steel feet, steel cap, dirt, air and rain water.
Further, the step S5 specifically includes:
setting mesh subdivision parameters for the model, and creating a tetrahedral mesh;
the mesh dissection parameters include: maximum cell size, minimum cell size, finesse, cell type, whether near-edge encryption, whether grid is optimized.
Furthermore, the mesh division in the step S5 is performed by default according to professional software, or different structures of the insulator model are mesh-divided according to requirements.
Further, in step S6, the basic equation of the electrostatic field finite element solution is:
in the above formula, epsilon is the dielectric constant of the material, rho is the charge density, and phi is the potential energy.
Further, in step S7, the input variable of the data set is a parameter to be optimized of the insulator, and the output variable is the maximum electric field intensity of the surface of the insulator.
Further, in step S8, training a neural network prediction model according to the simulation data set obtained in step S7, specifically including:
according to the simulation data set obtained in the step S7, training a neural network prediction model, wherein the input set is the input variable in the step S7, the output set is the output variable in the step S7, the network adopts an S-shaped transfer function, and the network weight and the threshold value are continuously adjusted through the inverse error function so as to minimize the inverse error function E.
Still further, the S-shaped transfer function is:
the inverse error function E is:
in the above, t i To desired output, O i Is the computational output of the network.
Compared with the prior art, the invention has the beneficial effects that:
(1) According to the invention, through finite element electric field simulation, insulation performance change caused by insulator structure optimization can be intuitively obtained, meanwhile, a neural network prediction model is constructed based on neural network deep learning, the geometric parameters and material parameters of the insulator are optimally designed, the electric properties such as electric field strength and the like of the insulator can be rapidly obtained, the pain points with high engineering experiment cost and long design period are solved, the design period and cost of the insulator are obviously reduced, and theoretical reference is provided for the optimal design of other devices in an electric power system;
(2) The invention relates to an electric field finite element simulation, which comprises the steps of parameterizing the total height of an insulator, the diameter of an iron foot, the depth of a metal insert, the diameters of upper, middle and lower umbrella skirts, the relative dielectric constant of an insulating material and the like, wherein the optimized parameter is large in volume, and compared with engineering experiment research for controlling a single variable, the finite element simulation model can comprehensively consider the influence of structural parameters and material parameters on the insulating performance of the insulator through parameterized scanning, and solve a plurality of parameter variables of the model at the same time, so that the influence of each parameter on a model result is obtained, and an optimal design scheme is sought;
(3) According to the neural network prediction model, the accuracy of the model can be improved by carrying out weighting and normalization processing on each geometrical parameter and material parameter of the insulator; the neural network deep learning is performed based on the simulation data set, so that the prediction accuracy of the model can be improved; when the optimization design of the insulator parameters is carried out, the prediction data can be obtained only by inputting parameters to be optimized into the model, the calculation efficiency of the whole model is high, the result is reliable, the rapid and batch parameterization optimization calculation can be realized, even the real-time calculation requirement can be met, the optimization design iteration of the insulator parameters can be rapidly and accurately realized, the design period is obviously shortened, the research and development cost is reduced, and meanwhile, more data support is provided for the optimization design of the insulator.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the present invention;
FIG. 2 is a schematic view of a three-dimensional geometric model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a material distribution of a three-dimensional geometric model according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a main grid structure according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a neural network prediction model according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, the embodiment of the invention provides a design method for optimizing geometry and material parameters of an insulator based on finite element simulation. The method comprises the following steps:
step S1: determining a design index;
and (3) defining service indexes of the insulator, and extracting characteristic parameters which are required to be met by the insulating performance of the insulator, wherein the characteristic parameters are used as excitation conditions of the subsequent simulation step.
Wherein the characteristic parameters include: under the specified pollution condition, the dry arc distance, the power frequency dry flashover voltage, the power frequency wet flashover voltage, the low frequency breakdown voltage and the like of the insulator.
Step S2: determining model parameters;
the geometric structure parameters, the material dielectric constant parameters and the pollution parameters of the insulator to be optimized are determined, and the overall height, the diameter of the iron feet, the creepage distance, the depth of the metal insert, the diameters of the upper umbrella skirt, the lower umbrella skirt, the distribution and the size of surface pollution and the like of the insulator can be parameterized.
Step S3: geometric modeling;
based on structural design, three-dimensional drawing software is used for establishing a three-dimensional geometric model of the insulator, and the three-dimensional geometric model is imported into finite element analysis software.
Step S4: imparting material properties to the model;
according to physical properties of the insulator entity material, the insulator structure is endowed with relative dielectric constants and electric conductivities of materials such as insulator umbrella skirt, shielding ring, steel feet, steel caps, pollution, air, rainwater and the like.
Step S5: meshing;
establishing tetrahedral grids, setting the maximum and minimum unit sizes, the fineness and the unit types of the grids before generating the grids, and optimizing the grids.
The step can be performed by default according to professional software, and grid subdivision can be performed on different structures of the insulator according to requirements.
Step S6: simulation calculation;
and adopting a finite element electrostatic field to realize simulation analysis of an insulator parameterized model, and obtaining field intensity distribution on the surface and around the insulator.
The basic equation for solving the finite element of the electrostatic field is as follows:
in the above formula (1), ε is the dielectric constant of the material, ρ is the charge density, and φ is the potential energy.
Step S7: establishing a simulation data set;
and according to the optimized structure or material parameters in the finite element analysis model of the insulator, carrying out parameterized scanning and simulation calculation to obtain the maximum field intensity value of the fixed-point surface of the insulator, thereby obtaining a large number of simulation data sets. The input variable of the data set is the parameter to be optimized of the insulator, and the output variable is the maximum electric field intensity of the surface of the insulator.
Step S8: optimizing and predicting;
weighting and normalizing each material parameter and the geometric parameter of the whole model in a minimum-maximum standardized mode, and then establishing a neural network prediction model containing an input layer, an hidden layer and an output layer based on deep learning software of the neural network;
according to the simulation data set obtained in the step S7, training a neural network prediction model; the input set is the input variable in the step S7, the output set is the output variable in the step S7, the network selects an S-shaped transfer function, and the network weight and the threshold value are continuously adjusted through the inverse error function so as to enable the inverse error function E to be minimum.
Wherein the S-shaped transfer function is:
the inverse error function E is:
in the above formula (3), t i To desired output, O i Is the computational output of the network.
And after the neural network prediction model is trained, inputting the insulator parameter data to be optimized into the neural network prediction model, so as to obtain prediction data.
Example 2
In order to further explain the design method for optimizing the geometry and the material parameters of the insulator based on finite element simulation, the embodiment takes the three-umbrella insulator as the material parameter optimization design example, and details the optimization design process, and specifically comprises the following steps:
step S1: designing indexes;
the design index and the service working condition of the three-umbrella insulator are defined. In the embodiment, the service voltage of the three-umbrella insulator is 55KV, and meanwhile, the requirement that the field intensity of an insulating material is lower than the air breakdown field intensity by 30KV/cm is met.
Step S2: determining model parameters;
in this embodiment, the parameters to be optimized are ceramic material of umbrella skirt in the tri-umbrella insulator, and the change of dielectric constant represents the umbrella skirt of different materials, and the dielectric constants are parameterized here, namely 6, 7, 8 and 9.
Step S3: geometric modeling;
the geometric model is created by the three-dimensional drawing software, and is imported into the finite element analysis software, and the final geometric model is shown in fig. 2.
In this embodiment, only the influence of the material properties on the electrical properties of the insulator is considered, so that only one geometry needs to be constructed.
Step S4: imparting material properties to the model;
the materials of the three-umbrella insulator in the embodiment comprise steel, cement adhesive and ceramic, and the material distribution condition of the three-dimensional geometric model is shown in fig. 3. Meanwhile, the conductivity and relative permittivity of each material are shown in table 1 below:
TABLE 1 conductivity and relative permittivity of the materials of the three umbrella insulators
Material Conductivity (S/m) Relative dielectric constant
Steel and method for producing same 1e 10 1.1e 7
Ceramic material 1e -12 6、7、8、9
Cement adhesive 1e -12 8
Step S5: meshing;
tetrahedral mesh dissection is performed on the tri-umbrella insulator in simulation software, as shown in fig. 4. Selecting integral subdivision, setting the maximum size of the grid to 1000mm, the minimum unit size to 0mm, the fineness to be medium, encrypting near edges, and optimizing the grid. And finally, mesh subdivision of the model is completed, and the total amount of meshes is about 5W.
Step S6: simulation calculation;
the type of electrostatic field analysis was selected and the model score was set to 1. Applying excitation conditions according to design criteria: 55KV potential is applied to the steel foot, and the steel cap is grounded to zero potential. For finite element computation convergence, an air domain needs to be constructed, and the air domain outer surface is set to be an open boundary.
Step S7: establishing a simulation data set;
in this embodiment, the relative dielectric constant of the ceramic material is subjected to parametric scanning, so that the corresponding results need to be simulated separately, and the maximum electric field intensity on the surface of the umbrella skirt is focused. And carrying out parameterized simulation analysis on the three-dimensional electrostatic field by a finite element method to obtain the surface maximum field intensity value of the three-umbrella insulator under different parameters, and forming a simulation data set, as shown in the following table 2.
Table 2 maximum field strength values in the surface of the three umbrella insulators under different parameters
From the concentration data in table 2, it is found that by simulation, the relative dielectric constant of the ceramic needs to be greater than 8 without changing the geometric model, so that no air breakdown occurs. This data has instructive implications for design optimization.
Step S8: optimizing and predicting;
and weighting and normalizing the material parameters and the geometric parameters of the whole model in a minimum-maximum standardized mode, and then establishing a BP neural network model shown in fig. 5, wherein the BP neural network model comprises three layers of networks of an input layer, an hidden layer and an output layer, the number of nodes of the input layer is 4, the number of nodes of the output layer is 1, and the number of nodes of the hidden layer is 6.
The learning algorithm of the network adopts an LM algorithm; s-shaped functions are used as transfer functions from an input layer to an hidden layer, and linear functions are used from the hidden layer to an output layer; and training times are set for 200 times, learning rate is 0.01, and convergence standard is 0.0001.
After training is completed, obtaining a trained neural network prediction model of the tri-umbrella insulator; according to the trained neural network prediction model, parameters to be optimized are input into the neural network prediction model, the performance of the tri-umbrella insulator under other parameter combinations can be rapidly calculated, the prediction of the performance of the tri-umbrella insulator is finally realized, and richer and more reliable data are provided for the optimal design of the tri-umbrella insulator.
The foregoing description is only exemplary of the invention and is not intended to limit the invention. Any modification, equivalent replacement, improvement, etc. made within the scope of the present invention should be included in the protection scope of the present invention.

Claims (11)

1. A method for optimization of insulator geometry and material parameters based on finite element simulation, the method comprising the steps of:
step S1: defining the service design index of the insulator, extracting characteristic parameters which need to be met by the insulation performance of the insulator, and taking the characteristic parameters as excitation conditions of the subsequent simulation step;
step S2: determining model parameters including geometric parameters, material dielectric constant parameters and pollution parameters of the insulator to be optimized;
step S3: geometric modeling;
based on structural design, three-dimensional drawing software is used for establishing a three-dimensional geometric model of the insulator, and the three-dimensional geometric model is imported into finite element analysis software;
step S4: imparting material properties to the model;
step S5: meshing;
step S6: simulation calculation;
adopting a finite element electrostatic field to realize simulation analysis of an insulator parameterized model, and obtaining field intensity distribution on the surface and around the insulator;
step S7: establishing a simulation data set;
according to optimized structure or material parameters in the finite element analysis model of the insulator, parameterized scanning and simulation calculation are carried out to obtain the maximum field intensity value of the fixed-point surface of the insulator, so that a simulation data set is obtained;
step S8: optimizing and predicting;
weighting and normalizing each material parameter and the geometric parameter of the whole model in a minimum-maximum standardized mode, and then establishing a neural network prediction model containing an input layer, an hidden layer and an output layer based on deep learning software of the neural network;
according to the simulation data set obtained in the step S7, training a neural network prediction model; and after the completion, inputting the insulator parameter data to be optimized into the neural network prediction model, and obtaining the prediction data under the corresponding parameters.
2. The method according to claim 1, wherein in step S1, the extracted feature parameters at least include: under the specified pollution condition, the dry arc distance, the power frequency dry flashover voltage, the power frequency wet flashover voltage and the low frequency breakdown voltage of the insulator are set.
3. The method according to claim 1, wherein in step S2, the geometrical parameters include at least total insulator height, foot diameter, creepage distance, metal insert depth and upper, middle and lower shed diameters;
the pollution parameters at least comprise the distribution position and the geometric dimension of the surface pollution of the insulator.
4. The method according to claim 1, wherein the step S4 specifically includes: the relative dielectric constant and the conductivity of each material in the insulator structure are given according to the physical properties of the insulator physical material.
5. The method of claim 4, wherein in step S4, the relative dielectric constant and conductivity of each material in the insulator structure is imparted, the material including at least insulator sheds, shielding rings, steel feet, steel caps, dirt, air and rain.
6. The method according to claim 1, wherein the step S5 specifically includes:
setting mesh subdivision parameters for the model, and creating a tetrahedral mesh;
the mesh dissection parameters include: maximum cell size, minimum cell size, finesse, cell type, whether near-edge encryption, whether grid is optimized.
7. The method according to claim 6, wherein the mesh subdivision in step S5 is performed by default according to a specialized software, or by different structures of the insulator model according to requirements.
8. The method according to claim 1, wherein in step S6, the basic equation for the finite element solution of the electrostatic field is:
in the above formula, epsilon is the dielectric constant of the material, rho is the charge density, and phi is the potential energy.
9. The method according to claim 1, wherein in step S7, the input variable of the data set is a parameter to be optimized of the insulator, and the output variable is a maximum electric field intensity of the surface of the insulator.
10. The method according to claim 9, wherein in step S8, the training of the neural network prediction model is performed according to the simulation data set obtained in step S7, specifically including:
according to the simulation data set obtained in the step S7, training a neural network prediction model, wherein the input set is the input variable in the step S7, the output set is the output variable in the step S7, the network adopts an S-shaped transfer function, and the network weight and the threshold value are continuously adjusted through the inverse error function so as to minimize the inverse error function E.
11. The method of claim 10, wherein the S-shaped transfer function is:
the inverse error function E is:
in the above, t i To desired output, O i Is the computational output of the network.
CN202310714205.5A 2023-06-16 2023-06-16 Method for optimizing geometry and material parameters of insulator based on finite element simulation Pending CN116579040A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310714205.5A CN116579040A (en) 2023-06-16 2023-06-16 Method for optimizing geometry and material parameters of insulator based on finite element simulation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310714205.5A CN116579040A (en) 2023-06-16 2023-06-16 Method for optimizing geometry and material parameters of insulator based on finite element simulation

Publications (1)

Publication Number Publication Date
CN116579040A true CN116579040A (en) 2023-08-11

Family

ID=87545361

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310714205.5A Pending CN116579040A (en) 2023-06-16 2023-06-16 Method for optimizing geometry and material parameters of insulator based on finite element simulation

Country Status (1)

Country Link
CN (1) CN116579040A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116936009A (en) * 2023-09-13 2023-10-24 国网山东省电力公司东营供电公司 Electric field distribution regulation and control method and system for high-voltage insulating dielectric functionally-graded material
CN117556779A (en) * 2023-10-26 2024-02-13 美台高科(上海)微电子有限公司 ESD simulation method
CN117763927A (en) * 2024-02-22 2024-03-26 大连理工大学 automatic updating method for simulation model driven by geometry-grid twinning

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116936009A (en) * 2023-09-13 2023-10-24 国网山东省电力公司东营供电公司 Electric field distribution regulation and control method and system for high-voltage insulating dielectric functionally-graded material
CN116936009B (en) * 2023-09-13 2023-11-28 国网山东省电力公司东营供电公司 Electric field distribution regulation and control method and system for high-voltage insulating dielectric functionally-graded material
CN117556779A (en) * 2023-10-26 2024-02-13 美台高科(上海)微电子有限公司 ESD simulation method
CN117763927A (en) * 2024-02-22 2024-03-26 大连理工大学 automatic updating method for simulation model driven by geometry-grid twinning
CN117763927B (en) * 2024-02-22 2024-05-17 大连理工大学 Automatic updating method for simulation model driven by geometry-grid twinning

Similar Documents

Publication Publication Date Title
CN116579040A (en) Method for optimizing geometry and material parameters of insulator based on finite element simulation
CN110083908B (en) Cable core temperature prediction method based on finite element analysis
CN108717071B (en) Composite insulator umbrella skirt damage diagnosis method
CN110188430B (en) Optimized design method for voltage-sharing structure of sectional type composite insulator of alternating-current 330kV suspension tower
Diaz-Acevedo et al. Optimization of corona ring for 230 kV polymeric insulator based on finite element method and PSO algorithm
CN111859733B (en) Method for optimizing reliability of automobile exhaust system based on ant colony algorithm
CN104899375A (en) Main insulating anti-corona structure optimization calculating method used for high-voltage motor
CN105844029A (en) Research method for high voltage cable terminal joint
CN108154003B (en) Blasting vibration prediction method based on Spark gene expression optimization
CN115128407A (en) V-shaped insulator string flashover fault simulation method based on bird dung pollution
CN104573219B (en) The method that a kind of DC power transmission line electric-field intensity and radio interference are calculated
Fahimi et al. Dynamic modeling of flashover of polymer insulators under polluted conditions based on HGA-PSO algorithm
CN103399980B (en) A kind of optimized Selection collocation method of power transmission and transformation insulator
CN105205202B (en) A kind of carrying current calculation method
CN112765842B (en) Optimization design method for voltage equalizing structure of combined insulator
CN106446418A (en) Cable terminal inflow water defect electric-field distribution analysis method and device
CN110737998B (en) Grading ring optimization design method based on finite element and deep belief network
Gencoglu et al. Computation of AC flashover voltage of polluted HV insulators using a dynamic arc model
CN105912771A (en) Ansys-based analysis method for optimization of influence of insulator detecting and cleaning robot on insulator string distribution voltage
CN116319377A (en) Distributed dynamic state estimation method for power distribution network for resisting network attack
CN113850006A (en) Composite insulator grading ring structure design method based on deep learning
CN111709552A (en) Method for rapidly acquiring critical pollution flashover voltage of composite insulator of power transmission line
Khajavi et al. Optimal design of corona ring in composite insulator considering the effects of tower structure, conductor, hardware and mutual phases
Tayseer et al. Optimum design of spacer material of gas-insulated switchgear by utilizing the technology of functionally graded material using different artificial optimization techniques
CN114113942B (en) Recognition and characteristic correction method and experimental device for oil paper insulation partial discharge

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination