A design method of GIL dielectric function gradient three-pillar insulator

Abstract

This invention relates to a design method for GIL (Glass-Insulated Linearity) dielectric functional graded three-post insulators. The method first reconstructs the spatial distribution of dielectric constant within the insulator's computational domain using a variable-density topology optimization algorithm, with the joint objectives of homogenizing the tangential electric field along the surface and actively repelling metal particles. Second, it maps the continuous distribution to a three-level gradation model (high, medium, and low), and combines photopolymerization 3D printing with thermosetting integrated casting technology to accurately fabricate the graded insulator. This invention optimizes the axial electric field direction near the posts, subjecting metal particles to an electric field force away from the insulator, thus achieving active particle repellency and significantly improving the surface flashover voltage under particle contamination. Compared to traditional uniform insulators, the method described in this invention… e -FGM insulators maintain a uniform electric field distribution while exhibiting excellent particle suppression performance, effectively preventing insulation failure accidents and significantly improving the operational reliability and stability of high-voltage electrical equipment.

Description

Technical Field

[0001] This invention relates to the field of high-voltage transmission lines in power systems, and more specifically, to a design method for GIL (Glass Inductor Level) dielectric functional gradient three-post insulators. Background Technology

[0002] Gas-insulated metal-enclosed transmission lines (GILs) are key equipment for long-distance, high-capacity power transmission. Their significant advantages, such as low loss, high reliability, and environmental friendliness, make them irreplaceable in power transmission within confined spaces. The three-post insulator, as the core support and insulation component, directly determines the system's operational stability.

[0003] However, traditional homogeneous epoxy resin cast insulators exhibit significant limitations under complex electric fields: First, due to the discontinuity of the dielectric, the tangential electric field distribution at the insulator interface is extremely uneven, easily leading to localized field strength concentrations; second, during equipment assembly and service life, multi-scale metallic particle contaminants inevitably exist within the system due to mechanical wear, electrode vibration, or the accumulation of free particle charges. Driven by an alternating electric field, these particles move towards and adsorb onto the insulator surface, causing severe electric field distortion, which significantly weakens the self-restoring insulation strength of gaseous dielectrics such as SF6, ultimately leading to surface flashover and insulation breakdown.

[0004] Currently, relying solely on passive particle traps is insufficient to eliminate the threat of near-field particles in insulators. Therefore, optimizing the dielectric distribution of insulator materials to achieve synergistic control of interface electric field homogenization and active particle removal has become a key technical requirement for improving the insulation level of high-end power equipment. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, particularly the severe surface electric field distortion and potential metal particle contamination issues associated with AC GIL three-post insulators, this invention proposes a dielectric functional gradient (DCI) method. e The FGM (Follow-Follow-Gram) three-post insulator and its design and fabrication method provide a new approach to improving the insulation performance of three-post insulators and active particle defense.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a design method for a GIL (Glass Inlet Dielectric Function Gradient) three-post insulator, comprising the following steps:

[0007] Step 1: Establish a multi-objective topology optimization model to reconstruct the dielectric constant distribution of the three-post insulator; based on the spatial electric field distribution characteristics of the three-post insulator, establish a topology optimization model with tangential electric field homogenization and axial electric field reconstruction as objective functions to achieve point-by-point continuous gradient design of the dielectric constant inside the insulator in space. Step 2: Discretization of dielectric distribution; The continuous dielectric constant distribution described above is simplified by gradation. The insulator is divided into three characteristic dielectric constant regions: high, medium, and low. Discretization mapping ensures that the simplified distribution model still has the expected electric field control effect. Step 3: Fabrication using a hierarchical integration process e -FGM three-post insulator samples were manufactured using a hybrid method combining 3D printing technology and thermosetting casting. Based on the discretized geometric topology, dielectric materials of different grades were prepared and integrated to ultimately obtain a monolithic structure. e -FGM three-post insulator specimen.

[0008] Step 4: Comprehensive performance characterization and experimental verification; build a metal particle motion trajectory observation platform, verify the effectiveness of FGM insulators in electric field control and active removal of metal particles through comparative experiments, and test its effect on improving surface flashover voltage.

[0009] Furthermore, in step one, in order to quantitatively characterize the spatial distribution of the near-field electric field of the insulator and its influence on the dynamic behavior of metal particles, this invention introduces the electric field deflection coefficient. α def Its definition is as follows:

[0010] In the formula, E x For axial electric field, E z For the radial electric field, this coefficient is used to evaluate the guiding effect of the electric field force on the trajectory of the particle.

[0011] Furthermore, in step one, to synergistically address the dual technical requirements of suppressing surface electric field distortion and guiding metal particles away from the insulator, this invention establishes a multi-objective topology optimization model. The relative permittivity distribution within the computational domain Ω1 of the three-post insulator is reconstructed using the Variable Density Topology Optimization (SIMP) algorithm. The computational model is as follows:

[0012] The objective function of this optimization model contains three core components: 1. Tangential electric field homogenization term f Et :Aims to minimize insulator surface STangential electric field on 1 E t With average field strength E t0 The deviation is reduced to achieve a uniform distribution of the electric field along the surface; 2. Particle removal optimization items f Ex The aim is to optimize the electric field deflection coefficient within a specific region Ω2 by adjusting the direction of the electric field vector to ensure that the particles are subjected to an electric field force away from the surface of the insulator. 3. Numerical stability penalty f grad By introducing a gradient penalty term, numerical instability phenomena such as the "chessboard effect" commonly found in topology optimization are suppressed, ensuring the smoothness of the dielectric constant gradient distribution and manufacturing feasibility.

[0013] Regarding the above optimization model and related formulas, the specific meanings of each parameter are as follows: w a and w b These are the optimization items. f Et and f Ex The weighting factor, C Et and C Ex These are the normalization coefficients for the corresponding optimization terms, used to eliminate the influence of physical dimensions. q It is the penalty weight of the gradient penalty term. h mesh This represents the size of the largest finite element mesh in the computational model. A It is the volume of domain Ω1; in the topology optimization iteration process, the variable density interpolation method is used to determine the relative permittivity of each point in space. e Mapped to artificial material density variables r ,in e max and e min These represent the maximum and minimum relative permittivity values ​​of the materials selected for the insulator, respectively. p This is a penalty factor used to suppress the generation of intermediate density values ​​and ensure that the optimization results are directed towards... e max and e min Polarization occurs, thereby improving manufacturing feasibility.

[0014] Furthermore, in step two, a three-level gradation discretization scheme is adopted. Based on the continuous density distribution calculated by topology optimization, it is mapped into three characteristic dielectric constant regions: high, medium, and low. The relative dielectric constants ε of the three regions are set to 14, 6, and 2, respectively.

[0015] Furthermore, in step three, the specific preparation process is as follows: 1. Preparation of high dielectric constant region: Based on the geometric model of this region, a polytetrafluoroethylene mold is customized; the epoxy resin matrix, curing agent and BaTiO3 high dielectric filler are mixed in a preset ratio, and after vacuum degassing, the mixture is injected into the mold and cross-linked under a preset thermosetting program. After demolding, a high dielectric constant preform is obtained. 2. Preparation of low dielectric constant region: A 3D printing model is established based on the geometry and topology of this region; a low dielectric printing slurry is prepared by mixing photocurable resin, photoinitiator, dispersant and hollow SiO2 microspheres, and then precisely molded using additive manufacturing technology to obtain a low dielectric constant preform; 3. Integral Molding: The high and low dielectric constant preforms are precisely embedded into the three-pillar metal assembly mold according to the optimized spatial positions; then, a dielectric base material composed of epoxy resin, curing agent, and Al2O3 filler is injected into the mold cavity; after overall thermosetting and demolding processes, the desired product is obtained. e -FGM three-post insulator specimen.

[0016] Furthermore, in step four, the metal particle motion observation device used in this invention consists of a high-voltage power supply system and an optical observation system. The power supply system integrates a voltage regulator, a step-up transformer, a current-limiting resistor, and a voltage divider to simulate a high-voltage environment under actual working conditions. The test area is placed inside a controlled glove box with high transparency, and a high-speed camera is used to capture the jumping trajectory of the metal particles. By comparing and analyzing the differences in the particle motion trajectories under different field strengths, the invention verifies... e -FGM three-post insulators have the effect of actively repelling particles and improving flashover voltage.

[0017] Compared with the prior art, the beneficial effects of this invention are: by reconstructing the dielectric constant distribution through multi-objective topology optimization, the tangential electric field on the insulator surface is reduced, and the reconstructed electric field force is used to actively drive away near-field metal particles, thereby blocking the harm of flashover induced by particle adsorption from the root. This provides a practical and feasible technical means to solve the hidden danger of metal particle pollution in AC GIL and enhance the insulation reliability of ultra-high voltage power transmission equipment. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the optimized model of a three-post insulator; Figure 2 Dielectric functional gradient ( e Schematic diagram of the relative permittivity distribution of a three-post insulator (FGM); Figure 3 The electric field distribution along the surface of a three-post insulator; Figure 4 Dielectric functional gradient ( e-FGM) Three-post insulator manufacturing process; Figure 5 The trajectory of metal particles near the three-post insulator; Figure 6 For uniform insulators and e -FGM insulator flashover voltage under particulate pollution and particulate pollution conditions. Detailed Implementation

[0019] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments.

[0020] A method for designing and fabricating a GIL (Glass Inductor) dielectric functional graded three-post insulator includes the following steps: Step 1: Establish a multi-objective topology optimization model and design the dielectric constant distribution of a three-post insulator. Taking a 550kV AC GIL three-post insulator as the application object, a multi-objective topology optimization model is established, and the spatial distribution of the insulator's dielectric constant is reconstructed through calculation. To quantitatively characterize the spatial distribution of the near-field electric field of the insulator and its influence on the dynamic behavior of metal particles, this invention introduces an electric field deflection coefficient. α def This is used to evaluate the guiding effect of electric field force on the trajectory of particles. Its definition is as follows:

[0021] In the formula, E x For axial electric field, E z For the radial electric field, this coefficient is used to evaluate the guiding effect of the electric field force on the trajectory of the particle.

[0022] To address the dual technical requirements of controlling tangential electric field distortion along the surface and guiding metal particles away from the insulator, this invention utilizes a variable density topology optimization algorithm (SIMP) to optimize the computational domain Ω1 of a three-post insulator (e.g., ...). Figure 1 The relative permittivity distribution within (as shown) is reconstructed. The calculation model is as follows:

[0023] The objective function of this optimization model contains three core components: 1. Tangential electric field homogenization term f Et :Aims to minimize such Figure 1 The insulator surface shown S Tangential electric field on 1 E t With average field strength E t0 The deviation is eliminated, thereby eliminating local field concentration and achieving a uniform distribution of the electric field along the surface; 2. Particle removal optimization items f Ex :Aims to optimize such Figure 1 The electric field deflection coefficient is shown within the Ω2 region. The Ω2 region is defined as a hollow cylindrical space adjacent to the inner wall surface of the outer shell, with an axial length of 300 mm and a radial thickness of 30 mm. By adjusting the direction of the electric field vector, it is ensured that the particles experience an electric force away from the insulator surface. 3. Numerical stability penalty f grad By introducing a gradient penalty term, numerical instability phenomena such as the "chessboard effect" commonly found in topology optimization are suppressed, ensuring the smoothness of the dielectric constant gradient distribution and manufacturing feasibility.

[0024] Regarding the above optimization model and related formulas, the specific meanings of each parameter are as follows: w a and w b These are the optimization items. f Et and f Ex The weighting factor, C Et and C Ex These are the normalization coefficients for the corresponding optimization terms, used to eliminate the influence of physical dimensions. q It is the penalty weight of the gradient penalty term. h mesh This represents the size of the largest finite element mesh in the computational model. A It is the volume of domain Ω1; in the topology optimization iteration process, the variable density interpolation method is used to determine the relative permittivity of each point in space. e Mapped to artificial material density variables r ,in e max and e min These represent the maximum and minimum relative permittivity values ​​of the materials selected for the insulator, respectively. p This is a penalty factor used to suppress the generation of intermediate density values ​​and ensure that the optimization results are directed towards... e max and e min Polarization occurs, thereby improving manufacturing feasibility.

[0025] Dielectric functional gradient after topology optimization ( e The relative permittivity distribution of the three-post insulator (-FGM) is as follows: Figure 2As shown, the high dielectric constant region is mainly distributed in the central core of the three support arms and the central main body region; the low dielectric constant region is mainly distributed in the belly of the insulator and the interface region near the metal insert.

[0026] like Figure 3 As shown, a comparison is made between a traditional uniform insulator and the one described in this invention. e The tangential electric field distribution along the surface of the FGM three-post insulator shows that the tangential electric field distribution along the surface of a traditional uniform insulator is extremely uneven. Its high field strength region is mainly concentrated in the central region of the web of the post, with a peak value as high as 3.71 kV / mm. However, the design of this invention... e -FGM insulators effectively diffuse the high-field region through spatial reconstruction of the dielectric constant, and the electric field distribution exhibits significant homogenization characteristics. Its electric field peak value is reduced to 3.29 kV / mm, which is about 11.3% lower than that of a uniform insulator. This significantly suppresses electric field distortion on the insulator surface and improves the electrical reliability of the overall insulation structure.

[0027] Step Two: A three-level gradation discretization scheme is adopted. Based on the continuous material density distribution calculated in Step One, and combined with the modification capability of high-performance insulating materials, the global dielectric constant is discretized and mapped into three characteristic dielectric constant regions: high, medium, and low. The relative dielectric constants of the three characteristic regions are set to 14 (high dielectric region), 6 (dielectric matrix region), and 2 (low dielectric region), respectively. This discretization design, while taking into account manufacturing processes, retains the control effect of the continuous gradient design to the greatest extent.

[0028] Step 3: Fabrication using a hierarchical integration process e -FGM three-post insulator sample, the specific preparation process is as follows: Figure 4 As shown: 1. Fabrication of high dielectric constant regions: such as Figure 4 As shown in ①, firstly, a polytetrafluoroethylene mold is customized based on the geometric model of the high-dielectric region; the epoxy resin matrix, curing agent, and BaTiO3 high-dielectric filler are mixed at a mass ratio of 100:44:410. The mixture is then thoroughly mixed in a mixer, followed by vacuum degassing at 120℃ for 20 minutes to remove air bubbles from the mixture. After degassing, the epoxy composite material is injected into the preheated mold cavity and placed in a forced-air drying oven for a 12-hour thermosetting reaction. After complete curing and cooling to room temperature, the material is demolded to obtain a high-dielectric-constant preform.

[0029] 2. Fabrication of low dielectric constant regions: such as Figure 4As shown in ②, a high-precision 3D model for 3D printing is first established based on the geometric topology of the region. Photocurable resin, photoinitiator, dispersant, and hollow SiO2 microspheres are precisely mixed at a mass ratio of 100:1:0.04:12 and thoroughly stirred to form the printing slurry. A complex topological structure with low dielectric constant is then formed using photocurable 3D printing technology. After printing, a cleaning process is performed to completely remove residual resin from the surface, followed by a secondary UV curing process to enhance its mechanical properties and chemical stability, ultimately yielding a low dielectric constant preform.

[0030] 3. Integrated molding: such as Figure 4 As shown in ③, the high and low dielectric constant preforms prepared above are precisely embedded into the designated positions of the three-pillar metal assembly mold according to the spatial coordinates determined by topology optimization. Subsequently, epoxy resin, curing agent, and Al2O3 filler are mixed at a mass ratio of 100:44:320. After thorough stirring and vacuum degassing for 20 minutes, the dielectric base material is injected into the preheated mold for insert integration. The entire assembly undergoes final thermosetting at a constant temperature of 120℃ for 12 hours. After curing, the mold is opened and demolding and cleaning are performed to finally obtain the integral assembly. e -FGM three-post insulator specimen.

[0031] Step Four: Comprehensive Performance Characterization and Experimental Verification; A metal particle motion trajectory observation platform was constructed, consisting of a high-voltage power supply system and an optical observation system. The power supply system integrates a voltage regulator, step-up transformer, current-limiting resistor, and voltage divider to simulate the high-voltage environment under actual working conditions. The test area was placed inside a highly transparent controlled glove box, and a high-speed camera was used to capture the jumping trajectories of the metal particles. By comparing and analyzing the differences in particle motion trajectories, the performance was verified. e -FGM three-post insulators have the effect of actively repelling particles and improving flashover voltage.

[0032] Figure 5 The trajectory of a spherical metal particle placed 10mm away from the legs of a three-post insulator is shown. Figure 5 As shown in (a), under the electric field generated by a conventional uniform insulator, metal particles are forced to move towards the insulator surface, eventually leading to surface flashover accidents induced by adsorption; in contrast, as Figure 5 As shown in (b), e Under the reconfigured electric field of the FGM insulator, the metal particles exhibit a tendency to move away from the insulator, verifying its active repulsion function.

[0033] Figure 6 It shows uniform insulators and eComparison of flashover voltages of FGM insulators under particulate pollution and particulate pollution conditions. Under particulate pollution-free conditions, the characteristic flashover voltage (cumulative probability 63.2%) of a uniform insulator is 48.5 kV, while... e - The voltage of FGM insulators increased to 54.4 kV, an increase of 12.2%; in the presence of particulate contamination, the flashover voltage of uniform insulators decreased significantly to 44.3 kV, while e The FGM insulator still maintains a high flashover voltage of 54.1kV, almost on par with its particle-free performance, representing a 22.1% improvement compared to the uniform insulator. Experimental results fully demonstrate the capabilities described in this invention. e -FGM insulators offer significant technological advantages in suppressing the impact of particulate contamination and improving the reliability of equipment operation.

Claims

1. A design method for a GIL (Glass-Insulated Linear Dielectric Function Gradient) three-post insulator, characterized in that, Includes the following steps: Step 1: Establish a multi-objective topology optimization model and use the variable density topology optimization algorithm to design the spatial distribution of dielectric constant in the computational domain of the three-post insulator; Step 2: Discretize the continuous dielectric constant distribution and extract the geometric topology model of each region; Step 3: Using a hierarchical integration process, combining 3D printing technology with thermosetting casting technology to prepare... ε -FGM three-post insulator.

2. The method according to claim 1, characterized in that, In step one, a topology optimization model is established with the goals of electric field homogenization and active removal of metal particles, and an electric field deflection coefficient is introduced. α def The influence of the near-field electric field distribution of insulators on the dynamic behavior of metal particles is quantitatively characterized by reconstructing the relative permittivity distribution in the computational domain of a three-post insulator using a variable-density topology optimization algorithm. The electric field deflection coefficient α def Its definition is as follows: In the formula, E x For axial electric field, E z For the radial electric field, this coefficient is used to evaluate the guiding effect of the electric field force on the trajectory of the particle.

3. The method according to claim 2, characterized in that, The relative permittivity distribution within the computational domain Ω1 of the three-post insulator is reconstructed using a variable-density topology optimization algorithm. The computational model is as follows: in, f Et For the tangential electric field homogenization term: This aims to minimize the insulator surface area. S Tangential electric field on 1 E t With average field strength E t0 The deviation is used to achieve a uniform distribution of the electric field along the surface. ； f Ex The particle removal optimization term aims to optimize the electric field deflection coefficient within a specific region Ω2. By adjusting the direction of the electric field vector, it ensures that the particles are subjected to an electric force moving away from the insulator surface. ； f grad Numerical stability penalty: By introducing a gradient penalty term, numerical instability phenomena such as the "chessboard effect" commonly found in topology optimization are suppressed, ensuring the smoothness of the dielectric constant gradient distribution and manufacturing feasibility. ； w a and w b These are the optimization items. f Et and f Ex Weighting factors; C Et and C Ex These are the normalization coefficients for the corresponding optimization terms, used to eliminate the influence of physical dimensions; q It is the penalty weight of the gradient penalty term. h mesh This represents the size of the largest finite element mesh in the computational model. A It is the volume of domain Ω1; in the topology optimization iteration process, the variable density interpolation method is used to determine the relative permittivity of each point in space. ε Mapped to artificial material density variables ρ ,in ε max and ε min These represent the maximum and minimum relative permittivity values ​​of the materials selected for the insulator, respectively. p This is a penalty factor.

4. The method according to claim 1, characterized in that, Step two employs a three-level gradation discretization method, mapping the continuous density distribution calculated by topology optimization into three characteristic dielectric constant regions: high, medium, and low.

5. The method according to claim 1, characterized in that, Step three includes: preparing a high dielectric constant preform using a thermosetting process, preparing a low dielectric constant preform using 3D printing technology, embedding the high and low dielectric constant preforms into a metal mold, and then injecting a dielectric base material and thermosetting the entire assembly to obtain... ε -FGM three-post insulator.

6. The method according to claim 1, characterized in that, The experiment also includes step four: building a metal particle trajectory observation platform to experimentally verify the effectiveness of FGM insulators in electric field control and active removal of metal particles, and testing their effect on improving surface flashover voltage.

7. The method according to claim 6, characterized in that, In step four, the metal particle trajectory observation platform consists of a high-voltage power supply system and an optical observation system. The power supply system integrates a voltage regulator, a step-up transformer, a current-limiting resistor, and a voltage divider to simulate high-voltage operating conditions. The optical observation system captures the jumping trajectory of the metal particles through a high-speed camera set outside a transparent controlled glove box.

8. The method according to claim 7, characterized in that, By comparing and analyzing the differences in the trajectories of metal particles, it was verified that... ε -FGM three-post insulators have the effect of actively repelling particles and improving flashover voltage.

Images