Disc-shaped suspension hybrid insulator and optimized design method thereof

Abstract

The application provides a disc-shaped suspension hybrid insulator and an optimization design method thereof, and relates to the technical field of hybrid insulators.The optimization design method of the disc-shaped suspension hybrid insulator comprises the following steps: a three-dimensional finite element static electric field simulation model is established, and tetrahedral mesh division is performed; a finite element method is used to solve a static electric field control equation, a simulation data set is constructed, normalization processing is performed, a BP neural network prediction model is established and trained; the trained neural network prediction model is used to continuously scan the sheath thickness in the range of 0-4 mm, the optimal coating thickness t* is determined, and an electrical safety design window [t*, 4 mm] is output. By introducing the simulation data set construction, normalization processing, BP neural network prediction model and optimal coating thickness determination algorithm, fast field strength prediction under any working condition is realized, the iron cap field strength reduction rate and the steel foot field strength fluctuation rate are defined as quantitative safety criteria, and an algorithm-driven quantitative evaluation method is provided.

Description

Technical Field

[0001] This invention relates to the field of hybrid insulator technology, and in particular to a disc-shaped suspension hybrid insulator and its optimized design method. Background Technology

[0002] High-altitude regions in my country, such as Tibet, Qinghai, and Xinjiang, are crucial corridors for ultra-high-voltage direct current (UHVDC) transmission projects. These areas, characterized by high altitudes (1000m–5500m), low air density, and severe pollution in some areas, place extremely high demands on the external insulation of transmission lines. Insulators are highly susceptible to forming continuous conductive water films under polluted and wet conditions, leading to flashover accidents and seriously threatening the safe and stable operation of the lines. Currently, two main types of insulators are used in these projects: porcelain disc suspension insulators, which have high mechanical strength and excellent aging resistance, meeting the connection requirements for heavy loads of 300kN and above, but suffer from poor hydrophobicity of the porcelain surface and weak resistance to flashover; and synthetic rubber composite insulators, which exhibit excellent hydrophobicity and hydrophobic migration properties, providing strong resistance to flashover, but lack sufficient mechanical strength under heavy loads and gradually lose their hydrophobicity with age.

[0003] To address these issues, some existing technologies have proposed hybrid insulator solutions, which involve applying an additional HTV silicone rubber coating layer to the outer surface of the ceramic component. However, the question remains whether adding the silicone rubber coating layer will alter the electric field distribution near the insulator fittings, potentially leading to electrical safety hazards such as partial discharge and fitting corrosion. Currently, there is a lack of systematic simulation verification and algorithm-based quantitative design methods, which hinders the engineering promotion and application of hybrid insulators. Summary of the Invention

[0004] In view of this, the present invention proposes a disc-shaped suspension hybrid insulator and its optimized design method to address the technical problem mentioned in the background art: whether adding a silicone rubber coating layer will change the electric field distribution near the insulator fittings, thereby leading to electrical safety hazards such as partial discharge and fitting corrosion. Currently, there is a lack of systematic simulation verification and algorithm-based quantitative design methods, which restricts the engineering promotion and application of hybrid insulators.

[0005] The technical solution of this invention is implemented as follows: In a first aspect, the present invention proposes an optimized design method for a disc-shaped suspension hybrid insulator, comprising: A three-dimensional finite element electrostatic field simulation model was established based on the actual structural dimensions of the disc suspension hybrid insulator, and tetrahedral meshing was performed on the three-dimensional finite element electrostatic field simulation model. The electrostatic field control equations were solved using the finite element method. Parametric simulations were performed under different voltage levels and sheath thicknesses to extract the maximum electric field intensity E of the cap under each condition. cap The maximum electric field strength E of the steel foot pin Construct a simulation dataset; The sheath thickness, applied voltage, electric field strength of the iron cap, and electric field strength of the steel foot in the simulation dataset are normalized. Using the normalized sheath thickness and applied voltage as input variables, and the normalized maximum electric field strength E of the iron cap as the input variable... cap The maximum electric field strength E of the steel foot pin For the output variables, build and train a BP neural network prediction model; Using the trained neural network prediction model, the sheath thickness is continuously scanned within the range of 0 to 4 mm. The values ​​of the reduction rate of the field strength of the iron cap η(t,U) and the fluctuation rate of the field strength of the steel foot δ(t,U) under all verified voltage levels at each scanning point are calculated. The thickness value that meets the preset safety conditions is determined as the optimal sheath thickness t*, and the electrical safety design window [t*, 4 mm] is output.

[0006] In some optional implementations, preferably, the tetrahedral meshing of the three-dimensional finite element electrostatic field simulation model includes: Near-edge dense grid is set at the junction of the lower edge of the iron cap bowl and the porcelain part, and at the base of the steel foot ball head.

[0007] In some alternative implementations, preferably, the setting of the near-edge encrypted mesh includes: the minimum cell size of the near-edge encrypted mesh is not greater than 1 / 5 of the global minimum cell size.

[0008] In some optional implementations, preferably, before continuously scanning the sheath thickness within the range of 0–4 mm using the trained neural network prediction model, the method further includes: The prediction accuracy of the BP neural network prediction model was evaluated using leave-one-out cross-validation, requiring the mean absolute error (MAE) of the field strength prediction for the iron cap and steel foot to be less than 0.2 kV / cm.

[0009] In some optional implementations, preferably, determining the thickness value that satisfies the preset safety conditions as the optimal covering thickness t* includes: This will simultaneously satisfy "η(t,U)≥0 holds true for all verification voltages" and "δ(t,U)≤5% holds true for all verification voltages" and The thickness value that first drops below the threshold ε is determined as the optimal coating thickness t*.

[0010] In some alternative implementations, preferably, the The first drop below the threshold ε includes: The threshold ε ranges from 0.1% / mm to 1.0% / mm.

[0011] In some alternative implementations, preferably, the establishment and training of the BP neural network prediction model includes: The BP neural network prediction model comprises an input layer, hidden layers, and an output layer. The input layer has 2 nodes, the output layer has 2 nodes, and the number of hidden layer nodes is determined using an empirical formula. The network is trained using the LM algorithm. A sigmoid transfer function is used from the input layer to the hidden layer, and a linear transfer function is used from the hidden layer to the output layer. The training convergence criterion is mean squared error (MSE) < 1 × 10⁻⁶. -4 .

[0012] In some optional implementations, preferably, the sheath thickness is continuously scanned within the range of 0–4 mm using a trained neural network prediction model, including: The step size for continuous scanning of the sheath thickness is no greater than 0.1 mm.

[0013] Secondly, the present invention proposes a disc-shaped suspension hybrid insulator, comprising: The ceramic body is made of high-strength electrical porcelain. An iron cap is installed on the upper end of the ceramic body and is bonded and fixed to the ceramic body with adhesive. Steel feet are installed at the lower end of the ceramic body and are bonded and fixed to the ceramic body with adhesive. The sheath, shaped like an outer insulating umbrella skirt, is made of HTV silicone rubber. The sheath covers the entire exposed surface of the ceramic body, with its upper end stopping at the lower edge of the iron cap bowl and its lower end stopping at the root of the steel foot ball head. A cork pad is placed between the iron cap and the ceramic body.

[0014] In some optional embodiments, preferably, the thickness of the sheath is 0.5mm to 4mm; the outer insulating skirt has a double umbrella structure and a nominal creepage distance of not less than 480mm.

[0015] The disc-shaped suspension hybrid insulator and its optimized design method of the present invention have the following advantages over the prior art: (1) By introducing simulation dataset construction, normalization processing, BP neural network prediction model and optimal coating thickness determination algorithm, rapid field strength prediction under any working condition is realized. The field strength reduction rate of iron cap η(t,U) and field strength fluctuation rate of steel foot δ(t,U) are defined as quantitative safety criteria. The optimal coating thickness t* is determined by continuous scanning in the range of 0 to 4 mm using the prediction model. The electrical safety design window [t*, 4 mm] is output. Through the algorithm-driven quantitative evaluation method, the gap of lack of systematic quantitative method for electric field safety evaluation of hybrid insulator fittings in the existing technology is filled, which is conducive to the engineering promotion and application of hybrid insulators. (2) By setting up a near-edge densified grid in the area where the lower edge of the iron cap bowl meets the porcelain piece and at the root of the steel foot ball head, the accuracy of local electric field calculation is ensured; (3) The prediction accuracy of the BP neural network prediction model was evaluated by using leave-one-out cross-validation. The mean absolute error (MAE) of the predicted field strength of the iron cap and steel foot was required to be <0.2 kV / cm to confirm that the prediction model has sufficient accuracy and can replace step-by-step finite element simulation for subsequent rapid analysis. (4) By simultaneously satisfying "η(t,U)≥0 holds true for all verification voltages" and "δ(t,U)≤5% holds true for all verification voltages" and The thickness value that first drops below the threshold ε is determined as the optimal coating thickness t*. The field strength reduction rate of the iron cap η(t,U) and the field strength fluctuation rate of the steel foot δ(t,U) are defined as quantitative safety criteria to improve the credibility and reliability of the optimization method. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the optimized design method of the disc-shaped suspension hybrid insulator in an embodiment of the present invention. Figure 2 This is a schematic diagram of a three-dimensional finite element electrostatic field simulation model in an embodiment of the present invention; Figure 3 This is a schematic diagram comparing the electric field cloud maps of the iron cap region under different coating thicknesses (0 / 1 / 2 / 3 / 4 mm) in an embodiment of the present invention; Figure 4 This is a schematic diagram of the maximum electric field strength of the iron cap as a function of the coating thickness at three voltage levels of 15kV, 20kV, and 25kV in an embodiment of the present invention. Figure 5 This is a schematic diagram of the maximum electric field strength of the steel foot as a function of the coating thickness at three voltage levels of 15kV, 20kV, and 25kV in an embodiment of the present invention. Figure 6 This is a schematic diagram of the structure of the BP neural network prediction model in an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of the disc-shaped suspension hybrid insulator in an embodiment of the present invention; Figure 8 For the present invention Figure 7 A magnified view of part A in the middle.

[0018] Explanation of reference numerals in the attached figures: 1- Porcelain body; 2- Iron cap; 3- Steel foot; 4- Protective sleeve; 5- Cork pad; 6- Locking pin; 7- Adhesive. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] In the description of the embodiments of the present invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in the embodiments of the present invention based on the specific circumstances.

[0021] In the description of the embodiments of the present invention, it should be noted that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of the present invention.

[0022] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0023] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0024] The following disclosure provides numerous different embodiments or examples for implementing various structures of the invention. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the invention. Furthermore, reference numerals and / or letters may be repeated in different examples. Such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. Additionally, examples of various specific processes and materials are provided in this invention; however, those skilled in the art will recognize the applicability of other processes and / or the use of other materials.

[0025] The technical solution will now be explained in detail: Reference Figures 1-6 As shown, a first aspect of the present invention proposes an optimized design method for a disc-shaped suspension hybrid insulator, comprising: Step S1: Establish a three-dimensional finite element electrostatic field simulation model based on the actual structural dimensions of the disc suspension hybrid insulator, and perform tetrahedral meshing on the three-dimensional finite element electrostatic field simulation model; In step S1, the disc-shaped suspension hybrid insulator includes a porcelain body, an iron cap, steel feet, and a sheath. Each component material is assigned a relative permittivity and conductivity. The porcelain body is made of high-strength electrical porcelain material with a relative permittivity ε. r =6.0, conductivity σ=1×10 -12 S / m; the sheath is made of HTV silicone rubber (high-temperature vulcanized silicone rubber), ε r =3.5、σ=1×10 -13 S / m, adhesive εr=4.0, σ=1×10 -12 S / m, both the iron cap and the steel foot are considered ideal conductors; Using COMSOL Multiphysics finite element simulation software, a model was established as follows: Figure 3 The electrical parameters of each material in the three-dimensional axisymmetric electrostatic field simulation model shown are shown in Table 1 below: Table 1 Porcelain body (high-strength electrical porcelain) 6.0 <![CDATA[1×10 -12 ]]> Sheath (HTV silicone rubber) 3.5 <![CDATA[1×10 -13 ]]> Iron cap (ductile iron) —(Ideal conductor) — Steel feet (45Mn2) —(Ideal conductor) — adhesive 4.0 <![CDATA[1×10 -12 ]]> air domain 1.0 0 The tetrahedral meshing of the three-dimensional finite element electrostatic field simulation model includes: setting near-edge denser meshes in the region where the lower edge of the iron cap bowl meets the ceramic part and at the root of the steel foot ball head. These regions are sensitive areas of concentrated electric field, which can ensure the accuracy of local electric field calculations; the total number of global mesh elements is not less than 5 × 10⁻⁶. 4The minimum element size of the near-edge refined mesh is no greater than 1 / 5 of the global minimum element size. Tetrahedral elements are used for mesh generation, with a global maximum element size of 10mm and a minimum element size of 0.1mm. Near-edge refinement (1mm thick, 5 layers) is applied at the junction of the iron cap bowl rim and the ceramic piece, and at the base of the steel foot ball head. The total mesh size is approximately 8 × 10⁻⁶. 4 The governing equation for solving the electrostatic field is Equation (1), and the boundary conditions are set as follows: the steel foot is subjected to a simulation voltage U, the iron cap is grounded (zero potential), and an open boundary is set on the outer surface of the air domain.

[0026] Step S2: Solve the electrostatic field control equations using the finite element method, and perform parametric simulation calculations under different voltage levels and sheath thicknesses to extract the maximum electric field intensity E of the cap under each condition. cap The maximum electric field strength E of the steel foot pin Construct a simulation dataset; In step S2, the governing equation for the electrostatic field is: (1); In equation (1), ε is the dielectric constant of the material (F / m), φ is the potential (V), and ρ is the free charge density (C / m³). For the insulating medium region involved in this invention, ρ=0, the equation degenerates into the Laplace equation; the voltage boundary conditions are: the steel foot is subjected to a simulated voltage U (V), the iron cap is set to zero potential, and the outer surface of the model is set to open boundary conditions; The different voltage levels and sheath thickness conditions include at least two applied voltage levels and at least three sheath thickness conditions. For example, in this embodiment, the sheath thickness t is set to t1=0mm, t2=1mm, t3=2mm, t4=3mm, and t5=4mm, for a total of 5 conditions. Simulations are performed at three applied voltage levels: U1=15kV, U2=20kV, and U3=25kV. A total of 15 sets of results are obtained. The simulation data for each condition is summarized in Table 2 below. Table 2 0mm 5.49 7.32 9.15 15.67 20.90 26.12 1mm 5.11 6.81 8.51 15.59 20.79 25.99 2mm 5.06 6.75 8.44 15.77 21.03 26.29 3mm 4.99 6.66 8.32 15.67 20.89 26.12 4mm 4.97 6.64 8.29 15.76 21.02 26.27 The maximum electric field strength E of the iron hat was extracted from each set of simulation results. cap (kV / cm) and the maximum electric field strength E of the steel foot pin (kV / cm), construct the simulation dataset D={(t ᵢ U ⱼ E cap,ij, E pin,ij ) | i=1,…,5; j=1,2,3}; Step S3: Normalize the sheath thickness, applied voltage, electric field strength of the iron cap, and electric field strength of the steel foot in the simulation dataset; In step S3, to eliminate the influence of the difference in dimensions between the sheath thickness t and the applied voltage U on the neural network training, the min-max normalization method is used to process the simulation dataset. The normalization formula is as follows: (2); In equation (2), x represents each input or output variable, x' is the normalized variable, and x min and x max These are the minimum and maximum values ​​of the variable in the dataset, respectively. After normalization, all variable values ​​fall within the [0,1] interval to improve the training convergence speed and prediction accuracy of the neural network.

[0027] Step S4: Using the normalized sheath thickness and applied voltage as input variables, and the normalized maximum electric field strength E of the iron cap... cap The maximum electric field strength E of the steel foot pin For the output variables, build and train a BP neural network prediction model; In step S4, establish as follows Figure 8 The BP neural network prediction model shown includes an input layer, hidden layers, and an output layer. The input layer has 2 nodes, the output layer has 2 nodes, and the number of hidden layer nodes is determined according to an empirical formula. The empirical formula is: (3); In equation (3), m is the number of hidden layer nodes, and n in n is the number of nodes in the input layer. out Let be the number of nodes in the output layer, and 'a' be an empirical parameter with a value range of [1, 10]. The network is trained using the Levenberg-Marquardt (LM) algorithm. The input layer to the hidden layer uses a sigmoid transfer function, and the hidden layer to the output layer uses a linear transfer function. The sigmoid transfer function is as follows: (4); Using the mean squared error (MSE) as the loss function, the network weights and thresholds are continuously adjusted through backpropagation until the MSE is less than the preset convergence threshold of 1×10⁻⁶. -4 The network is considered to have converged after 1000 training iterations. Using 15 sets of simulation data as the training set, the weight matrices W1 (6×2) and W2 (2×6), and the bias vectors b1 (6×1) and b2 (2×1) are stored after convergence. After training, any combination of (t, U) can be input into the BP neural network prediction model to quickly predict the maximum electric field intensity of the iron cap and steel foot under the corresponding working condition, without needing to rerun the finite element simulation.

[0028] In step S4, the insulator creepage distance, skirt shape parameters, or environmental pollution level can be further incorporated into the input variables of the neural network prediction model to achieve multi-parameter comprehensive optimization design. The method is the same as above and will not be repeated here.

[0029] Step S5: Use leave-one-out cross-validation to evaluate the prediction accuracy of the BP neural network prediction model, requiring the mean absolute error (MAE) of the field strength prediction for the iron cap and steel foot to be <0.2 kV / cm; The leave-one-out method in step S5 specifically involves: each time, selecting one set from 15 sets of data as the validation set, and using the remaining 14 sets for training. After 15 iterations, the mean absolute error (MAE) and maximum error (MaxE) between the predicted and simulated values ​​are calculated. Validation results show that the MAE for the predicted field strength of the iron cap is 0.041 kV / cm, and the maximum error (MaxE) is 0.089 kV / cm; the MAE for the predicted field strength of the steel leg is 0.063 kV / cm, and the maximum error (MaxE) is 0.112 kV / cm, both meeting the accuracy requirement of MAE < 0.2 kV / cm. The accuracy of this prediction model meets the engineering design requirements and can replace step-by-step finite element simulation for subsequent rapid analysis. Step S6: Using the trained neural network prediction model, continuously scan the sheath thickness within the range of 0-4mm, calculate the values ​​of the reduction rate η(t,U) of the iron cap electric field strength and the fluctuation rate δ(t,U) of the steel foot electric field strength at all verified voltage levels at each scan point, determine the thickness value that meets the preset safety conditions as the optimal sheath thickness t*, and output the electrical safety design window [t*, 4mm]. In step S6, determining the thickness value that meets the preset safety conditions as the optimal sheath thickness t* includes: This will simultaneously satisfy "η(t,U)≥0 holds true for all verification voltages" and "δ(t,U)≤5% holds true for all verification voltages" and The thickness value that first drops below the threshold ε is determined as the optimal coating thickness t*. The threshold ε ranges from 0.1% / mm to 1.0% / mm; in this embodiment, ε = 0.5% / mm is used. The step size for continuous scanning of the sheath thickness is no greater than 0.1 mm to ensure the accuracy of determining the optimal thickness t*. The formula for calculating the reduction rate of field strength η(t,U) of the iron cap is as follows: (5); In equation (5), η>0 indicates that the electric field strength of the iron cap decreases after the cladding layer is applied, and η<0 indicates that the electric field strength increases. The electrical safety criterion requires that η(t, U)≥0 holds true for all verification voltage levels, that is, the maximum electric field strength of the iron cap after the cladding layer is applied must not be higher than the value when there is no cladding. The formula for calculating the volatility δ(t,U) of the steel foot field is as follows: (6); In Equation (6), the electrical safety criterion requirement δ(t, U)≤5% holds true for all verification voltage levels, meaning that the disturbance of the cladding layer to the electric field near the steel foot does not exceed 5%, and is considered to have no effect on the electric field of the steel foot. The field strength reduction rate η(t,U) of the iron cap is calculated using formula (5), and the field strength fluctuation rate δ(t,U) of the steel foot is calculated using formula (6). The results are shown in Table 3 below: Table 3 0mm 0% 0% 0% 0% 0% 0% 1mm 6.9% 7.0% 7.0% 0.5% 0.5% 0.5% 2mm 7.8% 7.8% 7.8% 0.6% 0.6% 0.7% 3mm 9.1% 9.0% 9.1% 0% 0% 0% 4mm 9.5% 9.3% 9.4% 0.6% 0.6% 0.6% Using a trained neural network prediction model, continuous scanning was performed within the range of 0–4 mm with a step size of 0.1 mm to calculate the values ​​of η and δ at each scanning point under three voltage levels. The scanning results show that from t=0.5 mm, η is greater than 0 and δ does not exceed 1.0% at all voltage levels, satisfying the basic electrical safety criteria. As t increases, η continues to rise, but the rate of increase gradually slows down. At t=3 mm, η reaches approximately 9% at all three voltage levels, while the marginal increment of η when t continues to increase from 3 mm to 4 mm... The value has been reduced to below 0.3%, and the η-t curve shows a clear inflection point at t=3mm, after which it tends to saturate; the fluctuation of δ does not exceed 1.1% throughout the entire range, and it always meets the safety criterion.

[0030] Based on this, the optimal coating thickness t* = 3mm was determined. Simulation data shows that the reduction rate of the field strength in the iron cap increases significantly as t increases from 0 to 3mm (taking 25kV as an example, η increases from 0 to 9.1%), while from 3mm to 4mm, η only increases by a further 0.3%, a marginal increase ( The η-t curve shows a clear inflection point at t=3mm, with the δ value decreasing to near zero. Simultaneously, the δ value fluctuates by no more than 1.1% across the entire range, consistently meeting the safety criterion. Therefore, t*=3mm is the optimal sheath thickness at the inflection point of the η-t curve's marginal benefit: at this thickness, the improvement in the electric field strength of the cap is close to saturation (η≈9%), and the additional electrical safety benefits gained from further increasing the sheath thickness are extremely limited, making further thickening unnecessary. Overall assessment conclusion: The electrical safety design window for HTV silicone rubber sheaths is [3mm, 4mm], and the recommended optimal sheath thickness design value is t*=3mm.

[0031] Based on the same concept, a second aspect of the present invention, combined with... Figure 7 and Figure 8 As shown, a disc-shaped suspension hybrid insulator is proposed, comprising a porcelain body 1, an iron cap 2, a steel foot 3, a sheath 4, a cork washer 5, and a locking pin 6, wherein: The ceramic body 1 is made of high-strength electrical porcelain; Iron cap 2 is installed on the upper end of the ceramic body 1 and is bonded and fixed to the ceramic body 1 by adhesive 7; Steel feet 3 are installed at the lower end of the ceramic body 1 and are bonded and fixed to the ceramic body 1 by adhesive 7; The sheath 4 is shaped like an outer insulating umbrella skirt and is made of HTV silicone rubber. The sheath 4 covers the entire exposed surface of the ceramic body 1. Its upper end B stops at the lower edge of the bowl of the iron cap 2 and its lower end C stops at the root of the ball head of the steel foot 3. It does not cover any metal fittings. The sheath 4 and the ceramic body are bonded to each other through adhesive 7 to ensure long-term reliable adhesion. A cork pad 5 is disposed between the iron cap 2 and the ceramic body 1; Locking pin 6 is installed inside iron cap 2 and is used to connect adjacent insulator strings.

[0032] In some embodiments, the thickness of the sheath 4 is 0.5mm to 4mm; the outer insulating skirt has a double umbrella structure and a nominal creepage distance of not less than 480mm.

[0033] The technical parameters of each structure of the disc suspension hybrid insulator are shown in Table 4 below: Table 4 Rated mechanical failure load 300 kN Nominal creepage distance 480 mm Structural height 195±6.1 mm Disc diameter 350±14.7 mm Link mark 24R (ball-and-socket type) Porcelain materials High-strength electrical porcelain εr=6.0 Iron Hat 2 Materials QT450-10 Ductile Iron ideal conductor Steel foot material 3 45Mn2 steel ideal conductor HTV Sheath 4 Materials High-temperature vulcanized silicone rubber εr=3.5 Adhesive 7 High-strength cement adhesive 7 εr=4.0 Sheath thickness range 4 0.5mm~4mm Applicable Standards GB / T 1001.1 The disc-shaped suspension hybrid insulator and its optimized design method proposed in this embodiment have the following advantages: (1) The present invention imparts the excellent hydrophobicity and hydrophobic migration properties of silicone rubber to traditional porcelain insulators by covering the surface of high-strength porcelain parts with HTV silicone rubber sheath 4, thereby significantly improving the anti-pollution flashover ability of insulators in high-altitude and heavily polluted environments without changing the mechanical properties of the porcelain parts body 1. (2) Based on traditional finite element simulation, this invention introduces a BP neural network prediction model, which realizes rapid prediction of electric field strength at insulator fittings under any combination of sheath thickness and operating voltage. It eliminates the need for repeated time-consuming finite element simulations, significantly improving parameter optimization efficiency. Taking this embodiment as an example, a single finite element simulation takes about 15 minutes, while the neural network prediction time is negligible. In continuous scanning of the 0-4mm range (step size 0.1mm, a total of 41 scanning points × 3 voltage levels = 123 predictions), the total calculation time is reduced from about 30 hours to a few seconds. (3) This invention proposes two quantitative electrical safety indicators, namely, the field strength reduction rate η of the iron cap and the field strength fluctuation rate δ of the steel foot, which transform the original vague qualitative conclusion of "no adverse effects" into a quantifiable and verifiable mathematical criterion, providing an objective evaluation standard for the design of the coating layer. (4) The present invention outputs the specific optimal design value t* for the thickness of the sheath 4 through the optimal coating thickness determination algorithm (the embodiment verifies t*=3mm), which provides a precise quantitative design basis for engineering applications and avoids the material waste caused by blindly increasing the thickness of the sheath 4; (5) The verification and optimization method of the present invention is universal and can be extended to the electrical safety assessment and sheath thickness optimization design of insulators of other voltage levels, other covering materials and other types. Other parameters other than sheath thickness (such as creepage distance, skirt inclination angle, etc.) can also be included in the neural network input variables for extended optimization.

[0034] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An optimized design method for a disc-shaped suspension hybrid insulator, characterized in that, include: A three-dimensional finite element electrostatic field simulation model was established based on the actual structural dimensions of the disc suspension hybrid insulator, and tetrahedral meshing was performed on the three-dimensional finite element electrostatic field simulation model. The electrostatic field control equations were solved using the finite element method. Parametric simulations were performed under different voltage levels and sheath thicknesses to extract the maximum electric field intensity E of the cap under each condition. cap The maximum electric field strength E of the steel foot pin Construct a simulation dataset; The sheath thickness, applied voltage, electric field strength of the iron cap, and electric field strength of the steel foot in the simulation dataset are normalized. Using the normalized sheath thickness and applied voltage as input variables, and the normalized maximum electric field strength E of the iron cap as the input variable... cap The maximum electric field strength E of the steel foot pin For the output variables, build and train a BP neural network prediction model; Using the trained neural network prediction model, the sheath thickness is continuously scanned within the range of 0 to 4 mm. The values ​​of the reduction rate of the field strength of the iron cap η(t,U) and the fluctuation rate of the field strength of the steel foot δ(t,U) under all verified voltage levels at each scanning point are calculated. The thickness value that meets the preset safety conditions is determined as the optimal sheath thickness t*, and the electrical safety design window [t*, 4 mm] is output.

2. The optimized design method for disc-shaped suspension hybrid insulators as described in claim 1, characterized in that, The tetrahedral meshing of the three-dimensional finite element electrostatic field simulation model includes: Near-edge dense grid is set at the junction of the lower edge of the iron cap bowl and the porcelain part, and at the base of the steel foot ball head.

3. The optimized design method for disc-shaped suspension hybrid insulators as described in claim 2, characterized in that, The setting of the near-edge encrypted mesh includes: the minimum cell size of the near-edge encrypted mesh is not greater than 1 / 5 of the global minimum cell size.

4. The optimized design method for disc-shaped suspension hybrid insulators as described in claim 1, characterized in that, Before continuously scanning the sheath thickness within the range of 0–4 mm using the trained neural network prediction model, the method further includes: The prediction accuracy of the BP neural network prediction model was evaluated using leave-one-out cross-validation, requiring the mean absolute error (MAE) of the field strength prediction for the iron cap and steel foot to be less than 0.2 kV / cm.

5. The optimized design method for disc-shaped suspension hybrid insulators as described in claim 1, characterized in that, The step of determining the thickness value that meets the preset safety conditions as the optimal covering thickness t* includes: This will simultaneously satisfy "η(t,U)≥0 holds true for all verification voltages" and "δ(t,U)≤5% holds true for all verification voltages" and The thickness value that first drops below the threshold ε is determined as the optimal coating thickness t*.

6. The optimized design method for disc-type suspension hybrid insulators as described in claim 5, characterized in that, The The first drop below the threshold ε includes: The threshold ε ranges from 0.1% / mm to 1.0% / mm.

7. The optimized design method for disc-type suspension hybrid insulators as described in claim 1, characterized in that, The establishment and training of the BP neural network prediction model includes: The BP neural network prediction model comprises an input layer, hidden layers, and an output layer. The input layer has 2 nodes, the output layer has 2 nodes, and the number of hidden layer nodes is determined using an empirical formula. The network is trained using the LM algorithm. A sigmoid transfer function is used from the input layer to the hidden layer, and a linear transfer function is used from the hidden layer to the output layer. The training convergence criterion is mean squared error (MSE) < 1 × 10⁻⁶. -4 .

8. The optimized design method for disc-type suspension hybrid insulators as described in claim 1, characterized in that, Using a trained neural network prediction model, the sheath thickness is continuously scanned within the range of 0–4 mm, including: The step size for continuous scanning of the sheath thickness is no greater than 0.1 mm.

9. A disc-shaped suspension hybrid insulator, characterized in that, include: The ceramic body is made of high-strength electrical porcelain. An iron cap is installed on the upper end of the ceramic body and is bonded and fixed to the ceramic body with adhesive. Steel feet are installed at the lower end of the ceramic body and are bonded and fixed to the ceramic body with adhesive. The sheath, shaped like an outer insulating umbrella skirt, is made of HTV silicone rubber. The sheath covers the entire exposed surface of the ceramic body, with its upper end stopping at the lower edge of the iron cap bowl and its lower end stopping at the root of the steel foot ball head. A cork pad is placed between the iron cap and the ceramic body.

10. The disc-shaped suspension hybrid insulator as described in claim 9, characterized in that, The sheath thickness is 0.5mm to 4mm; the outer insulating skirt has a double umbrella structure and a nominal creepage distance of not less than 480mm.

Images