CN114722707A - Preparation method of surface photocureable gradient coating of insulator - Google Patents

Preparation method of surface photocureable gradient coating of insulator Download PDF

Info

Publication number
CN114722707A
CN114722707A CN202210343224.7A CN202210343224A CN114722707A CN 114722707 A CN114722707 A CN 114722707A CN 202210343224 A CN202210343224 A CN 202210343224A CN 114722707 A CN114722707 A CN 114722707A
Authority
CN
China
Prior art keywords
coating
section
insulator
conductivity
gradient
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
CN202210343224.7A
Other languages
Chinese (zh)
Inventor
姚聪伟
孙帅
庞小峰
李兴旺
赵晓凤
王增彬
李盈
杨贤
邰彬
陈祖伟
蔡玲珑
李端姣
洪刚
丘欢
李文栋
邓军波
张冠军
李金殊
孙鹏
王超
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.)
Guangdong Power Grid Co Ltd
Electric Power Research Institute of Guangdong Power Grid Co Ltd
Original Assignee
Guangdong Power Grid Co Ltd
Electric Power Research Institute of Guangdong Power Grid 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 Guangdong Power Grid Co Ltd, Electric Power Research Institute of Guangdong Power Grid Co Ltd filed Critical Guangdong Power Grid Co Ltd
Priority to CN202210343224.7A priority Critical patent/CN114722707A/en
Publication of CN114722707A publication Critical patent/CN114722707A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • C—CHEMISTRY; METALLURGY
    • C09—DYES; PAINTS; POLISHES; NATURAL RESINS; ADHESIVES; COMPOSITIONS NOT OTHERWISE PROVIDED FOR; APPLICATIONS OF MATERIALS NOT OTHERWISE PROVIDED FOR
    • C09D—COATING COMPOSITIONS, e.g. PAINTS, VARNISHES OR LACQUERS; FILLING PASTES; CHEMICAL PAINT OR INK REMOVERS; INKS; CORRECTING FLUIDS; WOODSTAINS; PASTES OR SOLIDS FOR COLOURING OR PRINTING; USE OF MATERIALS THEREFOR
    • C09D7/00—Features of coating compositions, not provided for in group C09D5/00; Processes for incorporating ingredients in coating compositions
    • C09D7/40—Additives
    • C09D7/60—Additives non-macromolecular
    • C09D7/61—Additives non-macromolecular inorganic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/17—Mechanical parametric or variational design
    • 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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01B—CABLES; CONDUCTORS; INSULATORS; SELECTION OF MATERIALS FOR THEIR CONDUCTIVE, INSULATING OR DIELECTRIC PROPERTIES
    • H01B19/00—Apparatus or processes specially adapted for manufacturing insulators or insulating bodies
    • H01B19/04—Treating the surfaces, e.g. applying coatings
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids
    • 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)
  • General Engineering & Computer Science (AREA)
  • Geometry (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Chemical & Material Sciences (AREA)
  • Molecular Biology (AREA)
  • Computer Hardware Design (AREA)
  • Medical Informatics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Inorganic Chemistry (AREA)
  • Materials Engineering (AREA)
  • Wood Science & Technology (AREA)
  • Organic Chemistry (AREA)
  • Application Of Or Painting With Fluid Materials (AREA)

Abstract

The invention discloses a preparation method of a surface photocureable gradient coating of an insulator. The method comprises the steps of aiming at m particle populations in an N-dimensional space of a particle swarm algorithm, representing the position and the speed of each particle population by using the basic conductivity of each section of coating of an insulator to be coated, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating, and initializing the position and the speed of each particle population; calculating the maximum electric field intensity corresponding to each particle population according to the RBF neural network model to obtain the basic conductivity of each section of coating, the nonlinear coefficient of each section of coating and the inner diameter and the outer diameter of each section of coating which are optimal for the insulator to be coated; and carrying out photocuring coating preparation on the insulator to be coated according to the optimal basic conductivity of each section of coating, the nonlinear coefficient of each section of coating and the inner diameter and the outer diameter of each section of coating, so as to obtain the insulator with the photocuring gradient conductivity coating on the surface. The technical scheme of the invention improves the regulation and control precision of the electric field of the insulator coating.

Description

Preparation method of surface photocureable gradient coating of insulator
Technical Field
The invention relates to the technical field of preparation of insulator coatings, in particular to a preparation method of a surface photocureable gradient coating of an insulator.
Background
The solid insulator is an important component in various high-voltage electrical equipment, and plays double roles of structural support and electrical insulation. In the gas-solid composite insulation system, the solid surface insulation level is far lower than the gas gap and the solid insulation level at the same distance, so the surface insulation is a key factor for determining the safe and stable operation of the gas-solid composite insulation system. By changing the dielectric constant/surface conductivity, the distribution of the surface electric field under the action of alternating/direct current voltage can be effectively regulated and controlled, and the surface flashover voltage is improved.
In recent years, the application of a functionally graded permittivity material (. epsilon. -FGM) to insulators has shown a remarkable ability to alleviate electric field distortion at an alternating voltage. Meanwhile, research on a conductivity functionally graded material is being widely conducted.
Currently, researchers have employed a variety of means to regulate surface conductivity. The nonlinear conductive inorganic filler is mixed with an organic material to prepare a nonlinear conductive inorganic/organic composite material, and the nonlinear conductive inorganic/organic composite material is coated on the surface of an insulating material to prepare the nonlinear conductive coating insulator. The surface conductivity gradient insulator can also be constructed by means of plasma fluorination, magnetron sputtering, vapor deposition, laser cladding and the like.
The prior art has the problems of complex coating preparation process, uneven coating surface appearance, untight adhesion between the coating and an insulator and the like. The transition between layers of the gradient coating is not uniform, charge accumulation is easily caused, and electric field distortion is caused. Meanwhile, the design for researching the coating scheme is usually verified through finite element simulation at present, and the optimal coating scheme cannot be obtained by adopting the research idea of firstly designing and then verifying.
Disclosure of Invention
The invention provides a preparation method of a surface photocuring gradient coating of an insulator, which improves the regulation and control precision of an electric field of the insulator coating.
The invention provides a preparation method of an insulator surface photocureable gradient coating, which comprises the following steps:
aiming at m particle populations in the N-dimensional space of the particle swarm algorithm, the basic conductivity of each section of coating of the insulator to be coated, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating are used for representing the position and the speed of each particle population, and the position and the speed of each particle population are initialized; wherein N is 4N, and N is the number of conductivity gradient layers;
calculating the maximum electric field intensity corresponding to each particle population according to the RBF neural network model to obtain the optimal historical position of each particle population and the optimal historical position of each particle population until the maximum electric field intensity of each particle population meets the iteration end condition, and outputting the basic conductivity of each section of the optimal coating of the insulator to be coated, the nonlinear coefficient of each section of the coating, and the inner diameter and the outer diameter of each section of the coating; if the iteration end condition is not met, after the position variable and the speed variable of the particle population are updated, the process of calculating the maximum electric field intensity corresponding to each particle population according to the RBF neural network model is executed again;
and carrying out photocuring coating preparation on the insulator to be coated according to the optimal basic conductivity of each section of coating, the nonlinear coefficient of each section of coating and the inner diameter and the outer diameter of each section of coating, so as to obtain the insulator with the photocuring gradient conductivity coating on the surface.
Further, the establishing process of the RBF neural network model is as follows:
establishing a finite element simulation model according to the number of the conductivity gradient layers of the sample insulator, the nonlinear conductivity of each section of coating, and the inner diameter and the outer diameter of each section of coating;
and establishing the RBF neural network model according to the finite element simulation model.
Further, the nonlinear conductivity of each coating segment is calculated according to the following formula:
Figure BDA0003580115630000021
in the formula, mus0iThe basic conductivity of each section of coating; alpha is alphaiIs the nonlinear coefficient of conductivity; etIs the tangential component of the electric field strength; i is the number of conductivity gradient layers.
Further, obtaining a training sample of the RBF neural network model includes the following steps:
setting the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating of the sample insulator in the finite element simulation model to obtain a sample insulator simulation model;
calculating the distribution of the surface electric field intensity of the sample insulator through a current field interface built in finite element software, and extracting the maximum electric field intensity and the corresponding inner diameter, outer diameter, basic conductivity and nonlinear coefficient of each section of the surface coating in the sample insulator simulation model as training samples; the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating are input items during training, and the maximum electric field intensity is an output item during training.
Further, for m particle populations in an N-dimensional space of a particle swarm algorithm, the position and the speed of each particle population are represented by the basic conductivity of each section of coating of the insulator to be coated, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating, and are initialized, specifically:
for the ith particle population, wherein ViDenotes the velocity, X, of the ith particle populationiIndicates the position, V, of the ith particle populationi＝(Vi1，Vi2,...,ViN)，Xi＝(Xi1,Xi2,...,XiN)，i＝1,2,…,m；
The inner diameter sequence r of each segment of coating1,...,rnCorresponding to position X expressed as a population of particlesiAnd velocity ViThe first n terms of the sequence R of the outer diameters of the coating segments1,...,RnCorresponding to position X expressed as a population of particlesiAnd velocity ViItem n +1 to item 2n, the basic conductivity series mu of each coatings01,...,μs0nCorresponding to position X expressed as a population of particlesiAnd velocity Vi2n +1 to 3n, the non-linear coefficient sequence alpha of each coating0,...,αnCorresponding position X expressed as particle populationiAnd velocity ViThe 3n +1 th to 4n th items of (a); the position and velocity of each particle population is initialized.
Further, the position X of the particle population is updated according to the following formulaiAnd velocity Vi：
Vi(k+1)＝ω·Vi(k)+c1r1·(Pj(k)-Xi(k))+c2r2(Pg(k)-Xi(k))；
Xi(k+1)＝Xi(k)+Vi(k+1)；
Wherein, ω represents the inertial weight and is a non-negative number; c. C1And c2The value is 2 for the learning factor; r is1And r2Is [0,1 ]]A random number within a range; k represents the number of cycles; pjRepresenting the optimal historical position, P, of an individual of a population of particlesgRepresenting the optimal historical location of the population of particles.
Further, according to the optimal basic conductivity of each section of coating, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating, the preparation of the photocureable coating is carried out on the insulator to be coated, and the insulator with the photocureable gradient conductivity coating on the surface is obtained, and the method comprises the following steps:
preparing a coating formula of each section according with the basic conductivity of each section of coating and the nonlinear coefficient requirement of each section of coating by adopting a first inorganic filler and a second inorganic filler, and obtaining a coating mixed solution of each section according to the coating formula of each section; the first inorganic filler is one or more materials with nonlinear conductivity characteristics, and the second inorganic filler is an alumina inorganic filler;
preparing a flexible shielding film ring of each section according to the inner diameter and the outer diameter of each section of coating, and covering the flexible shielding film ring of each section on the surface of the insulator;
spraying the coating mixed solution on the corresponding section from inside to outside aiming at the insulator; only spraying one section of coating each time during spraying, and removing the flexible shielding film ring of the corresponding section; and when spraying of one section of coating is finished, removing the flexible shielding film circular rings of the other sections, carrying out photocuring treatment on the insulator, and covering the flexible shielding film circular rings of all the sections on the surface of the insulator again before spraying of one section of coating.
Further, when the coating mixed solution of each section is obtained according to the coating formula of each section, the solution is uniformly mixed through ultrasonic stirring and vacuum defoaming, and air in the mixed solution is removed.
Further, when the coating mixed solution is sprayed on the corresponding section of the insulator, the mixed solution of the corresponding section is filled into a cavity with a high-speed rotating blade, the solution with higher viscosity is thinned by utilizing the principle of fluid shear thinning, and the solution is sprayed on the surface of the insulator through a high-pressure spray gun.
Furthermore, the flexible shielding film ring is made of silicon rubber and has the thickness of 0.2 mm.
The invention has the following beneficial effects:
the invention provides a preparation method of a surface photocureable gradient coating of an insulator, which is characterized in that the speed and the position of each particle population are represented by the basic conductivity of each section of coating of the insulator, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating; and then, according to the RBF neural network model, iteratively calculating the maximum electric field intensity corresponding to each particle population to obtain the optimal historical position of each particle swarm and the optimal historical position of the particle population, and outputting the basic conductivity of each section of coating of the insulator, the nonlinear coefficient of each section of coating and the inner diameter and the outer diameter of each section of coating.
Drawings
FIG. 1 is a schematic flow chart of a method for preparing a surface photocuring gradient coating on an insulator according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an insulator according to a method for preparing a surface photocuring gradient coating on the surface of the insulator according to an embodiment of the present invention.
FIG. 3 is a graph of the maximum electric field strength along the surface versus the flashover voltage along the surface of an uncoated control group of a surface light-cured gradient coating of an insulator according to an embodiment of the present invention.
Detailed Description
The technical solutions in the present invention will be described clearly and completely with reference to the accompanying drawings, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, a method for preparing an insulator surface photocuring gradient coating according to an embodiment of the present invention includes the following steps:
step S101, aiming at m particle populations in an N-dimensional space of a particle swarm algorithm, representing the position and the speed of each particle population by using the basic conductivity of each section of coating of the insulator to be coated, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating, and initializing the position and the speed of each particle population; wherein N is 4N, and N is the number of conductivity gradient layers.
Step S102, calculating the maximum electric field intensity corresponding to each particle population according to the RBF neural network model to obtain the optimal historical position of each particle population and the optimal historical position of each particle population until the maximum electric field intensity of each particle population meets the iteration end condition, and outputting the basic conductivity of each section of the optimal coating of the insulator to be coated, the nonlinear coefficient of each section of the coating, and the inner diameter and the outer diameter of each section of the coating; and if the iteration end condition is not met, updating the position variable and the speed variable of the particle population, and then re-executing the process of calculating the maximum electric field intensity corresponding to each particle population according to the RBF neural network model.
And S103, carrying out photocureable coating preparation on the insulator to be coated according to the optimal basic conductivity of each section of coating, the nonlinear coefficient of each section of coating and the inner diameter and the outer diameter of each section of coating, so as to obtain the insulator with the photocureable gradient conductivity coating on the surface. The structure of the insulator according to the embodiment of the invention is shown in fig. 2.
As an embodiment, the establishing process of the RBF neural network model is as follows:
establishing a finite element simulation model according to the number of the conductivity gradient layers of the sample insulator, the nonlinear conductivity of each section of coating, and the inner diameter and the outer diameter of each section of coating;
and establishing the RBF neural network model according to the finite element simulation model.
As an example, the nonlinear electrical conductivity of the segments of the coating is calculated according to the following formula:
Figure BDA0003580115630000061
in the formula, mus0iThe basic conductivity of each section of coating; alpha is alphaiThe coefficient is a nonlinear coefficient and reflects the sensitivity of the conductivity to the electric field; etIs the tangential component of the electric field strength; 1,2, …, n; n is the number of conductivity gradient layers.
As one embodiment, the method for acquiring the training sample of the RBF neural network model comprises the following steps:
setting the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating of the sample insulator in the finite element simulation model to obtain a sample insulator simulation model;
calculating the distribution of the surface electric field intensity of the sample insulator through a current field interface built in finite element software, and extracting the maximum electric field intensity and the corresponding inner diameter, outer diameter, basic conductivity and nonlinear coefficient of each section of the surface coating in the sample insulator simulation model as training samples; the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating are input items during training, and the maximum electric field intensity is an output item during training.
As an embodiment, for m particle populations in an N-dimensional space of a particle swarm algorithm, the position and the speed of each particle population are represented by the basic conductivity of each coating of the insulator to be coated, the nonlinear coefficient of each coating, and the inner diameter and the outer diameter of each coating, and are initialized, specifically:
for the ith particle population, wherein ViDenotes the velocity, X, of the ith particle populationiIndicates the position, V, of the ith particle populationi＝(Vi1，Vi2,...,ViN)，Xi＝(Xi1,Xi2,...,XiN) I is 1,2, …, m; the inner diameter sequence r of each segment of coating1,...,rnCorresponding to position X expressed as a population of particlesiAnd velocity ViThe first n terms of the sequence R of the outer diameters of the coating segments1,...,RnCorresponding position X expressed as particle populationiAnd velocity ViN +1 th to 2 ndn terms, base conductivity sequence mu of each coatings01,...,μs0nCorresponding to position X expressed as a population of particlesiAnd velocity Vi2n +1 to 3n, the non-linear coefficient sequence alpha of each coating0,...,αnCorresponding to position X expressed as a population of particlesiAnd velocity ViThe 3n +1 th to 4n th items of (a); the position and velocity of each particle population is initialized.
As one of the embodiments, the position X of the particle population is updated according to the following formulaiAnd velocity Vi：
Vi(k+1)＝ω·Vi(k)+c1r1·(Pj(k)-Xi(k))+c2r2(Pg(k)-Xi(k))；
Xi(k+1)＝Xi(k)+Vi(k+1)；
Wherein, ω represents the inertial weight and is a non-negative number; c. C1And c2The value is 2 for the learning factor; r is1And r2Is [0,1 ]]A random number within a range; k represents the number of cycles; pjRepresenting the optimal historical position, P, of an individual of a population of particlesgRepresenting the optimal historical location of the population of particles.
As an embodiment, according to the optimal basic conductivity of each segment of coating, the nonlinear coefficient of each segment of coating, and the inner diameter and outer diameter of each segment of coating, performing photocuring coating preparation on the insulator to be coated to obtain the insulator with the photocuring gradient conductivity coating on the surface, including the following steps:
preparing a coating formula of each section according with the basic conductivity of each section of coating and the nonlinear coefficient requirement of each section of coating by adopting a first inorganic filler and a second inorganic filler, and obtaining a coating mixed solution of each section according to the coating formula of each section; the first inorganic filler is one or more materials with nonlinear conductivity characteristics, and the second inorganic filler is an alumina inorganic filler;
preparing a flexible shielding film ring of each section according to the inner diameter and the outer diameter of each section of coating, and covering the flexible shielding film ring of each section on the surface of the insulator;
spraying the coating mixed solution on the corresponding section from inside to outside aiming at the insulator; only one section of coating is sprayed each time during spraying, and the flexible shielding film circular ring of the corresponding section is removed; when the spraying of one section of coating is finished, the flexible shielding film circular rings of the other sections are removed, the insulator is subjected to photocuring treatment, and before one section of coating is sprayed, the flexible shielding film circular rings of all the sections are covered on the surface of the insulator again.
As one embodiment, when the coating mixed solution of each section is obtained according to the coating formula of each section, the solution is uniformly mixed by ultrasonic stirring and vacuum defoamation, and air in the mixed solution is removed.
As an embodiment, when spraying the coating mixed solution to the corresponding section of the insulator, the mixed solution of the corresponding section is loaded into a cavity with a high-speed rotating blade, the solution with higher viscosity is thinned by using the principle of fluid shear thinning, and the solution is sprayed on the surface of the insulator by a high-pressure spray gun.
Preferably, the material used for the flexible shielding film ring is silicon rubber, and the thickness of the flexible shielding film ring is 0.2 mm.
As one example of more details therein, step a 01: aiming at m particle populations in the N-dimensional space, representing the position and the speed of each particle population by using the basic conductivity of each section of coating of the insulator to be coated, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating, and initializing the position and the speed of each particle population; wherein N is 4N, and N is the number of conductivity gradient layers. In particular, ViRepresenting the velocity of the ith particle population (i.e., the displacement of the ith particle population in a certain iteration), XiDenotes the position of the ith particle population, where Vi＝(Vi1，Vi2,...,ViN)，Xi＝(Xi1,Xi2,...,XiN) I is 1,2, …, m; the inner diameter sequence r of each segment of coating1,...,rnCorresponding to a position variable X expressed as a population of particlesiAnd velocity variable ViThe first n terms of the sequence R of the outer diameters of the coating segments1,...,RnCorresponding to a position variable X expressed as a population of particlesiAnd velocity variable ViItem n +1 to item 2n, the base conductivity series μ of each coatings01,...,μs0nCorresponding to a position variable X expressed as a population of particlesiAnd velocity variable Vi2n +1 to 3n, the non-linear coefficient sequence alpha of each coating0,...,αnCorresponding to a position variable X expressed as a population of particlesiAnd velocity variable ViThe 3n +1 th item to the 4n th item of (1).
Step A02: calculating the maximum electric field intensity E corresponding to each particle population by using the RBF neural network model obtained by trainingmaxObtaining and storing individual optimal historical positions P of particle populationsjAnd the optimal historical position P of the particle populationg. Wherein, PjAnd PgDefining in the Kth iteration, the individual optimal historical position of the ith particle population is stored in Pj(k) J is 1,2, …, m, and the optimal historical location of all particle populations is stored in pg (k), g is 1, 2.
The RBF neural network model is established in the following process: determining the number n of the conductivity gradient layers of the sample insulator, taking n to be less than or equal to 10 to determine the geometric parameters of the sample insulator model to ensure the operability of the process, wherein ri<Ri，ri＜ri+1，Ri<Ri+1，r1D (d is the minimum radius of the coating region), RnD (D is the maximum radius of the coated area), nonlinear conductivity μ of each coating segmentsiDetermined according to the following equation:
Figure BDA0003580115630000091
wherein, mus0iIs the basic conductivity of each section of coating, the basic conductivity mus0i>1*10-20s/m；αiIs a nonlinear coefficient which represents the sensitivity of the conductivity to the electric field, alphai>1；EtIs the electric field strengthThe tangential component of (a); in the variables, i is 1,2, … and n. And establishing a finite element simulation model according to the number of the conductivity gradient layers of the sample insulator, the nonlinear conductivity of each section of coating, and the inner diameter and the outer diameter of each section of coating.
Acquiring a training sample set of the RBF neural network model: setting the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating of the sample insulator in the finite element simulation model; wherein the inner and outer diameters of each segment are in steps of 10mm, and the basic conductivity of each segment is in steps of 5 x 10-20s/m is the step length, and the nonlinear coefficient of each section takes 0.2 as the step length; and calculating the distribution of the surface electric field intensity of the sample insulator through a current field interface built in finite element software, and extracting the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating of the sample insulator corresponding to the maximum electric field intensity to serve as training samples. And the value ranges of the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating of the sample insulator are consistent with the value ranges of corresponding parameters in the RBF neural network model establishing process.
Establishing the RBF neural network model according to the finite element simulation model, initializing the RBF neural network model, and randomly selecting an activation function center in a sample; determining hidden layer output, and comparing the cosine similarity of the output matrix and the expected matrix; adding the maximum point of the cosine similarity matrix value in the residual points as a new hidden node unit; and finally, training the RBF neural network model, and finishing the training when the error is below an expected value or the number of hidden nodes reaches the maximum value. In the embodiment of the invention, the relation between the input parameters (namely the training sample set) and the output electric field intensity is established through a finite element calculation tool, different training samples are continuously input into the RBF neural network model for calculation, and the relation between the input parameters and the output electric field intensity is simulated through the RBF neural network model, so that the finite element calculation is replaced in the subsequent process, and the purposes of saving the calculation time and improving the calculation efficiency and precision are achieved.
When the RBF neural network model is tested, inputting a test sample set, comparing errors between maximum electric field intensities obtained by finite element software calculation and RBF neural network model calculation under the same test sample set, if the error range is below 1%, testing is qualified, and outputting the RBF neural network model.
Step A03: and determining the maximum iteration times k of the particle swarm algorithm according to the convergence criterion that the output result of each 100 steps is within the range of 100V/m. Judging whether the iteration end condition is met or not according to the maximum electric field intensity of each particle population obtained by calculation, if so, ending the calculation to obtain the optimal basic conductivity mu of each section of coating of the insulator to be coateds0iThe non-linear coefficient alpha of each coatingiInner diameter r of each coatingiAnd outer diameter RiThereby determining the coating scheme. And if not, updating the position and the speed of the particle population, and repeating the steps A02-A03 until an iteration end condition is met.
Updating the position X of the population of particles according to the following formulaiAnd velocity Vi：
Vi(k+1)＝ω·Vi(k)+c1r1·(Pj(k)-Xi(k))+c2r2(Pg(k)-Xi(k))；Xi(k+1)＝Xi(k)+Vi(k+1)；
Wherein, ω represents the inertial weight and is a non-negative number; c. C1And c2The value is 2 for the learning factor; r is a radical of hydrogen1And r2Is [0,1 ]]A random number within a range; k represents the number of cycles; p isjRepresenting the optimal historical position, P, of an individual of a population of particlesgRepresenting the optimal historical location of the population of particles.
Step A04: according to the optimal basic conductivity mu of each section of coating of the insulator to be coateds0iThe non-linear coefficient alpha of each coatingiInner diameter r of each segment of coatingiAnd outer diameter RiAnd performing coating preparation on the insulator to be coated to obtain the insulator with the surface photocuring gradient conductivity coating.
Substep A041: adopting the first inorganic filler and the second inorganic filler to modulate the basic conductivity of each section of coating and the nonlinear coefficient of each section of coatingThe required coating formula of each section is obtained, and the coating mixed solution of each section is obtained according to the coating formula of each section; the first inorganic filler is one or more materials with nonlinear conductivity characteristics, and the second inorganic filler is an alumina inorganic filler; preferably, the first inorganic filler includes, but is not limited to, silicon carbide, zinc oxide, graphene oxide. Specifically, the coating formula of each section (i.e. the optimal basic conductivity μ of each section of coating in step a 03) meeting the basic conductivity of each section of coating and the requirement of the nonlinear coefficient of each section of coating is prepared by adjusting and controlling the proportion, the particle size and the like of the first inorganic filler and the second inorganic filler (i.e. the optimal basic conductivity μ of each section of coating in step a 03)s0iThe non-linear coefficient alpha of each coatingi)。
Preferably, in order to ensure better effect of the light curing process, the volume percentage content of the first inorganic filler and the second inorganic filler is 0 v% to 30 v% in total. Generally, within this range, the change of the content particle size of the filler such as silicon carbide, zinc oxide, graphene oxide, etc. can significantly change the nonlinear electrical conductivity characteristics of the photosensitive resin composite material, and through the test method, the optimal basic electrical conductivity μ of each segment of the coating in the step a03 can be obtaineds0iThe non-linear coefficient alpha of each coatingi。
Preferably, the coating material mixed solution is uniformly mixed by ultrasonic stirring and vacuum defoaming, and air in the coating material mixed solution is removed.
Substep A042: according to the inner diameter and the outer diameter of the coating of each segment (namely, the inner diameter r of the coating of each segment optimized in the step A03)iAnd outer diameter Ri) And preparing the flexible shielding film circular rings of all sections.
Substep A043: and covering the flexible shielding film circular rings of all the sections on the surface of the insulator.
Substep A044: and taking down the flexible shielding film ring of the innermost layer.
Substep a 045: and (4) filling the coating mixed solution which is prepared in the substep A041 and corresponds to the innermost coating into a cavity with a high-speed rotating blade, thinning the solution with higher viscosity by utilizing the principle of fluid shear thinning, and spraying the diluted solution on the surface of the insulator by a high-pressure spray gun. The solution sprayed on the surface of the insulator loses the shearing force effect, and the rheological property is rapidly reduced.
Substep A046: and removing the rest flexible shielding film circular rings, and carrying out ultraviolet light curing treatment on the insulator for 15 min.
Spraying the coating mixed solution of the corresponding section from inside to outside by repeating the operations from substep A043 to substep A046 for the rest sections of the insulator, only spraying one section of coating each time during spraying, and removing the flexible shielding film ring of the corresponding section; when spraying of one section of coating is finished, the flexible shielding film circular rings of the other sections are removed, the insulator is subjected to photocuring treatment, and before spraying of one section of coating, the flexible shielding film circular rings of all the sections are covered on the surface of the insulator again, so that only one section of coating is sprayed before spraying of each layer, and then spraying and curing are carried out. Finally, the insulator with the photocuring gradient conductivity coating on the surface is prepared.
According to the embodiment of the invention, the electric field intensity is calculated through the RBF neural network model, the single calculation time can be shortened to 1/1000, and meanwhile, the error is controlled within 2%, so that the calculation efficiency and the calculation accuracy of the basic conductivity of each section of coating of the insulator to be coated, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating can be effectively improved. By nesting the particle swarm optimization and the artificial neural network model, the coating scheme can be effectively optimized by adopting the particle swarm optimization, as shown in fig. 3, the maximum electric field intensity of the optimized gradient insulator is reduced by 45% relative to that of the homogeneous insulator, and the surface flashover voltage is increased by 22%. Compared with the traditional optimization calculation method with unchanged conductance, the method has the advantages that the electric field regulation effect of the insulator coating is better through a more accurate modeling process, and the actual coating preparation process is guided according to the optimal calculation result by introducing the nonlinear conductance characteristics of the material into the calculation of the RBF neural network model and obtaining the optimal conductance parameters and the space distribution of different areas on the surface of the insulating structure. According to the embodiment of the invention, through combining the nonlinear conductivity with the gradient concept, the electric field mutation caused by the conductivity mutation at the coating junction is reduced, so that the electric field change is more uniform. The insulator coating prepared by the embodiment of the invention has the advantages of quick forming, smooth surface appearance, tight adhesion with the insulator interface, easy realization and industrial application value. The method is applied to the gas insulated switchgear, and the photocuring gradient coating with conductivity can effectively regulate and control the electric field distribution on the surface of the insulator under direct-current voltage, improve the electrical resistance of the edge surface of the insulator and prolong the service life of the gas insulated switchgear.
One of ordinary skill in the art can understand and implement it without inventive effort. While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention. It will be understood by those skilled in the art that all or part of the processes of the above embodiments may be implemented by hardware related to instructions of a computer program, and the computer program may be stored in a computer readable storage medium, and when executed, may include the processes of the above embodiments. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.

Claims (10)

1. A preparation method of an insulator surface photocuring gradient coating is characterized by comprising the following steps:
aiming at m particle populations in an N-dimensional space of a particle swarm algorithm, representing the position and the speed of each particle population by using the basic conductivity of each section of coating of an insulator to be coated, the nonlinear coefficient of each section of coating, and the inner diameter and the outer diameter of each section of coating, and initializing the position and the speed of each particle population; wherein N is 4N, and N is the number of conductivity gradient layers;
calculating the maximum electric field intensity corresponding to each particle population according to the RBF neural network model to obtain the optimal historical position of the particle population and the optimal historical position of the particle population until the maximum electric field intensity of each particle population meets the iteration end condition, and outputting the basic conductivity of each section of optimal coating of the insulator to be coated, the nonlinear coefficient of each section of coating and the inner diameter and the outer diameter of each section of coating; if the iteration end condition is not met, after the position variable and the speed variable of the particle population are updated, the process of calculating the maximum electric field intensity corresponding to each particle population according to the RBF neural network model is executed again;
and carrying out photocuring coating preparation on the insulator to be coated according to the optimal basic conductivity of each section of coating, the nonlinear coefficient of each section of coating and the inner diameter and the outer diameter of each section of coating, so as to obtain the insulator with the photocuring gradient conductivity coating on the surface.
2. The preparation method of the insulator surface photocuring gradient coating according to claim 1, wherein the RBF neural network model is established by the following steps:
establishing a finite element simulation model according to the number of the conductivity gradient layers of the sample insulator, the nonlinear conductivity of each section of coating, and the inner diameter and the outer diameter of each section of coating;
and establishing the RBF neural network model according to the finite element simulation model.
3. The method for preparing the surface photocureable gradient coating of the insulator according to claim 2, wherein the nonlinear conductivity of each segment of the coating is calculated according to the following formula:
Figure FDA0003580115620000011
in the formula, mus0iThe basic conductivity of each section of coating; alpha (alpha) ("alpha")iIs the nonlinear coefficient of conductivity; etIs the tangential component of the electric field strength; i is the number of conductivity gradient layers.
4. The method for preparing the surface photocuring gradient coating of the insulator according to claim 3, wherein the step of obtaining the training sample of the RBF neural network model comprises the following steps:
setting the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating of the sample insulator in the finite element simulation model to obtain a sample insulator simulation model;
calculating the distribution of the surface electric field intensity of the sample insulator through a current field interface built in finite element software, and extracting the maximum electric field intensity and the corresponding inner diameter, outer diameter, basic conductivity and nonlinear coefficient of each section of the surface coating in the sample insulator simulation model as training samples; the inner diameter, the outer diameter, the basic conductivity and the nonlinear coefficient of each section of the surface coating are input items during training, and the maximum electric field intensity is an output item during training.
5. The method for preparing the surface photocureable gradient coating on the insulator according to claim 4, wherein for m particle populations in N-dimensional space of particle swarm optimization, the position and the speed of each particle population are represented by the basic conductivity of each coating of the insulator to be coated, the nonlinear coefficient of each coating, and the inner diameter and the outer diameter of each coating, and are initialized, specifically as follows:
for the ith particle population, wherein ViDenotes the velocity, X, of the ith particle populationiIndicates the position, V, of the ith particle populationi＝(Vi1，Vi2,...,ViN)，Xi＝(Xi1,Xi2,...,XiN)，i＝1,2,…,m；
The inner diameter sequence r of each segment of coating1,...,rnCorresponding to position X expressed as a population of particlesiAnd velocity ViThe first n terms of the sequence R of the outer diameters of the coating segments1,...,RnCorresponding to position X expressed as a population of particlesiAnd velocity ViItem n +1 to item 2n, the basic conductivity series mu of each coatings01,...,μs0nCorresponding to position X expressed as a population of particlesiAnd velocity ViItem 2n +1 to item 3n, coating each segmentIs a non-linear coefficient sequence of0,...,αnCorresponding to position X expressed as a population of particlesiAnd velocity ViThe 3n +1 th to 4n th items of (a); the position and velocity of each particle population is initialized.
6. The method for preparing the surface photocuring gradient coating on the insulator according to claim 5, wherein the position X of the particle population is updated according to the following formulaiAnd velocity Vi：
Vi(k+1)＝ω·Vi(k)+c1r1·(Pj(k)-Xi(k))+c2r2(Pg(k)-Xi(k))；
Xi(k+1)＝Xi(k)+Vi(k+1)；
Wherein, ω represents an inertial weight and is a non-negative number; c. C1And c2The value is 2 for the learning factor; r is a radical of hydrogen1And r2Is [0,1 ]]A random number within a range; k represents the number of cycles; pjRepresenting the optimal historical position, P, of an individual of a population of particlesgRepresenting the optimal historical location of the population of particles.
7. The method for preparing the surface photocureable gradient coating on the insulator according to the claim 6, wherein the preparation of the photocureable coating on the insulator to be coated is carried out according to the optimal basic conductivity of each coating, the nonlinear coefficient of each coating, and the inner diameter and the outer diameter of each coating, so as to obtain the insulator with the surface photocureable gradient conductivity coating, and the method comprises the following steps:
preparing a coating formula of each section according with the basic conductivity of each section of coating and the nonlinear coefficient requirement of each section of coating by adopting a first inorganic filler and a second inorganic filler, and obtaining a coating mixed solution of each section according to the coating formula of each section; the first inorganic filler is one or more materials with nonlinear conductivity characteristics, and the second inorganic filler is an alumina inorganic filler;
preparing a flexible shielding film ring of each section according to the inner diameter and the outer diameter of each section of coating, and covering the flexible shielding film ring of each section on the surface of the insulator;
spraying the coating mixed solution on the corresponding section from inside to outside aiming at the insulator; only one section of coating is sprayed each time during spraying, and the flexible shielding film circular ring of the corresponding section is removed; and when spraying of one section of coating is finished, removing the flexible shielding film circular rings of the other sections, carrying out photocuring treatment on the insulator, and covering the flexible shielding film circular rings of all the sections on the surface of the insulator again before spraying of one section of coating.
8. The method for preparing the surface photocuring gradient coating on the insulator according to claim 7, wherein when the coating mixed solution of each section is obtained according to the coating formula of each section, the solution is uniformly mixed by ultrasonic stirring and vacuum defoaming, and air in the mixed solution is removed.
9. The method for preparing the surface photocuring gradient coating of the insulator according to claim 8, wherein when the coating mixed solution of the corresponding section is sprayed on the insulator, the mixed solution of the corresponding section is filled into a cavity with a high-speed rotating blade, the solution with higher viscosity is thinned by utilizing the principle of fluid shear thinning, and the diluted solution is sprayed on the surface of the insulator by a high-pressure spray gun.
10. The method for preparing the surface photocuring gradient coating on the insulator according to any one of claims 1 to 9, wherein the material for the flexible shielding film ring is silicone rubber, and the thickness of the flexible shielding film ring is 0.2 mm.
CN202210343224.7A 2022-04-02 2022-04-02 Preparation method of surface photocureable gradient coating of insulator Pending CN114722707A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210343224.7A CN114722707A (en) 2022-04-02 2022-04-02 Preparation method of surface photocureable gradient coating of insulator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210343224.7A CN114722707A (en) 2022-04-02 2022-04-02 Preparation method of surface photocureable gradient coating of insulator

Publications (1)

Publication Number Publication Date
CN114722707A true CN114722707A (en) 2022-07-08

Family

ID=82240988

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210343224.7A Pending CN114722707A (en) 2022-04-02 2022-04-02 Preparation method of surface photocureable gradient coating of insulator

Country Status (1)

Country Link
CN (1) CN114722707A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115358131A (en) * 2022-10-19 2022-11-18 广东电网有限责任公司 Insulator design method, device, storage medium and system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115358131A (en) * 2022-10-19 2022-11-18 广东电网有限责任公司 Insulator design method, device, storage medium and system
CN115358131B (en) * 2022-10-19 2023-02-24 广东电网有限责任公司 Insulator design method, device, storage medium and system

Similar Documents

Publication Publication Date Title
Kim et al. An optimal neural network plasma model: a case study
CN114722707A (en) Preparation method of surface photocureable gradient coating of insulator
CN110262233B (en) Optimization method for technological parameters of magnetic control film plating instrument
Khorasani et al. Modeling of TiC-N thin film coating process on drills using particle swarm optimization algorithm
CN106547956B (en) A kind of method and device for the ground total electric field obtaining D.C. high voltage transmission division molded line
Villarroel et al. Particle swarm optimization vs genetic algorithm, application and comparison to determine the moisture diffusion coefficients of pressboard transformer insulation
CN111445965A (en) Design method of carbon fiber reinforced cement-based material based on deep learning
Mohanty et al. Artificial neural networks modelling of breakdown voltage of solid insulating materials in the presence of void
M’Hamdi et al. Multi-objective optimization of 400 kV composite insulator corona ring design
Dadashizadeh Samakosh et al. Experimental‐based models for predicting the flashover voltage of polluted SiR insulators using leakage current characteristics
CN107688862B (en) Insulator equivalent salt deposit density accumulation rate prediction method based on BA-GRNN
CN109816555A (en) A kind of load modeling method based on support vector machines
WO2022263586A1 (en) Method for adjusting parameters of a coating process to manufacture a coated transparent substrate
Belhouchet et al. Artificial neural networks and genetic algorithm modelling and identification of arc parameter in insulators flashover voltage and leakage current
Lin Parameter optimisation of a vacuum plasma spraying process using boron carbide
Bruns Monte Carlo calculations on off-lattice polymer chains. The influence of variation of the excluded volume
CN112765842A (en) Optimization design method for combined insulator voltage-sharing structure
WO2017122404A1 (en) Film forming simulation method, program, and semiconductor processing system
CN107153722A (en) The system and method that the parameter of matching network model is determined using equipment and efficiency
CN108763604A (en) A kind of radial base neural net point collocation solving the response of the statics of composite structure containing interval parameter
CN107491584B (en) Geometric parameter-based shrinkage prediction method for investment casting
Sajjad et al. Application of artificial neural network in predicting flashover behaviour of outdoor insulators under polluted conditions
Ramajo et al. Computational Approach of Dielectric Permitivities in BaTiO3—Epoxy Composites
Xing Prediction of the Optimal Umbrella Shape of Insulators Based on Data Mining Technology
Sun et al. Construction of permittivity graded insulator using topology-optimized lattice structure

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