Transformer pressure relief valve installation position optimization method suitable for high-energy arc fault

Abstract

The invention relates to the technical field of transformer pressure relief valve installation position optimization, in particular to a transformer pressure relief valve installation position optimization method suitable for a high-energy arc fault, and the method comprises the steps: building a coordinate system on a transformer top cover, and quantifying the installation position parameters of a pressure relief valve; randomly extracting a mounting position sample point; establishing a transformer high-energy arc fault transient oil pressure simulation calculation model, and calculating oil pressure change corresponding to each sample point during a fault period; fitting a mapping relation between the mounting position of the pressure relief valve and the oil pressure peak value through the proxy model; according to the method, the relationship between the installation position of the pressure relief valve and the oil pressure rise is quantitatively analyzed by constructing a mathematical model, the traditional empirical design is broken through, and a systematic theory and calculation method is provided for selection of the installation position of the pressure relief valve of the transformer.

Description

Technical Field

[0001] This invention relates to the field of transformer pressure relief valve installation position optimization technology, and in particular to a method for optimizing the installation position of transformer pressure relief valves suitable for high-energy arc faults. Background Technology

[0002] In power systems, transformers are critical equipment, and their operational safety and stability directly affect the reliability of the entire power system. Transformers are typically equipped with pressure relief valves. When internal pressure rises, the diaphragm of the pressure relief valve senses the oil pressure surge and releases pressure by discharging the insulating oil inside the transformer, a crucial step in protecting the transformer's safe operation. When an arc fault occurs inside the transformer, the high-energy arc vaporizes and cracks the surrounding insulating oil, generating high-temperature, high-pressure bubbles. The rapid expansion of these bubbles drastically compresses the insulating oil, causing a sudden surge in oil pressure. If the pressure relief valve fails to act effectively and in a timely manner, the oil tank may rupture due to pressure overload, leading to catastrophic consequences such as explosions and fires, seriously threatening the safe and stable operation of the power system. Therefore, improving the explosion-proof capability of transformers has become a critical issue that the power industry urgently needs to address.

[0003] However, the current selection of installation positions for pressure relief valves on transformers is mostly based on experience, failing to fully consider the impact of different installation positions on the pressure relief effect. This results in slow response times when dealing with internal high-energy arc faults, failing to fully utilize the pressure relief capacity, and frequent transformer explosion accidents, thus casting doubt on the existing arrangement of pressure relief valves. Existing research lacks a systematic optimization theory and method for the installation position of pressure relief valves, and no effective design criteria for installation positions have been proposed. Summary of the Invention

[0004] To address the aforementioned problems, the purpose of this invention is to provide an optimization method for the installation position of a transformer pressure relief valve applicable to high-energy arc faults. This method can quantitatively determine the optimal installation position of the pressure relief valve on the transformer top cover, reduce the peak oil pressure inside the transformer during arc faults, and thus reduce the risk of transformer combustion and explosion accidents.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A method for optimizing the installation location of a transformer pressure relief valve for high-energy arc faults includes the following steps: Step 1: Draw a three-dimensional geometric model of the transformer and pressure relief valve based on the actual transformer structure; Step 2: Based on the dimensions of the top cover and the size of the pressure relief valve in the three-dimensional geometric model of the transformer, determine the optimization parameters for the installation position of the pressure relief valve and the variable design space for the optimization parameters. Step 3: Sample points are collected within the variable design space of the parameters to be optimized to obtain a sample set; Step 4: Based on the sample set, perform pressure simulation calculations for the sample points under internal arc fault conditions in the transformer, and obtain the simulation calculation results for all sample points. Step 5: Based on the simulation results of each sample point, construct a surrogate model. The surrogate model is used to fit the functional relationship between the parameters to be optimized at the installation position of the pressure relief valve and the peak value of the internal oil pressure of the transformer. Step 6: Construct a mathematical description of the optimization problem with the objective function of minimizing the peak oil pressure at the internal measuring point of the transformer and the variable design space of the parameters to be optimized for the installation position of the pressure relief valve as the constraint condition; Step 7: Based on the constructed proxy model, use an optimization algorithm to solve the mathematical description of the optimization problem and obtain the installation coordinates of the pressure relief valve when the peak oil pressure inside the transformer is at its minimum under the internal arc fault.

[0006] Furthermore, step one specifically involves: based on the engineering parameters of the target transformer, using engineering drawing software, constructing a full-size three-dimensional geometric model including components such as the tank body, windings, core, riser seat, and pressure relief valve.

[0007] Furthermore, step two specifically involves: determining the range of locations where the pressure relief valve can be installed in a parametric form based on the spatial structure of the top cover and the physical dimensions of the pressure relief valve in the drawn transformer geometric model.

[0008] Furthermore, determining the position parameterization range of the pressure relief valve includes the following steps: (1) Establish a two-dimensional xy coordinate system with the center of the transformer top cover as the origin; (2) Determine the feasible domain of the installation position coordinates of the pressure relief valve, i.e., the variable design space, based on the boundary coordinates of the top cover and the diameter of the pressure relief valve. Furthermore, step three specifically involves: performing spatial sampling within the design space, preferably using Latin hypercube sampling (LHS) to generate an initial sample set.

[0009] Furthermore, the Latin hypercube sampling method includes the following steps: Determine the number of samples and the dimensions of the variables; Hierarchical interval division; The sub-intervals of the variable are randomly arranged, and samples are drawn from the subspace; One-dimensional samples are randomly combined to form a multi-dimensional sample set; Spatial coverage verification.

[0010] Furthermore, step four specifically includes the following steps: (1) Import the three-dimensional geometric model drawn in step one into the Spaceclaim software of the ANSYS platform. According to the coordinate position of the pressure relief valve in each sample point, move and change the position of the pressure relief valve in the model accordingly, and simplify the transformer to improve the simulation calculation efficiency. (2) Based on the simplified solid domain model of the transformer, draw the fluid domain geometric model of the insulating oil part inside the transformer; (3) Import the insulating oil fluid domain model into the ANSYS Fluent module and mesh the fluid domain; (4) Set the simulation solver to a transient solver, set the insulating oil material properties and flow field boundary conditions, and use the RSM turbulence model to simulate the fluid motion state. The RSM is closed by solving the transport equation and dissipation rate equation of Reynolds stress. The transport equation of Reynolds stress is: , In the formula, The average density of the fluid; Represents the Reynolds stress tensor components; This is the turbulent dissipation rate component; For turbulent viscosity; These are spatial coordinate components; For speed; The component is the rotational angular velocity. For the correction of the Reynolds stress by the Coriolis force caused by rotation, For pressure.

[0011] Furthermore, the flow field boundary conditions include wall boundary conditions in contact with the tank wall and windings, pressure relief valve outlet boundary conditions, oil conservator outlet boundary conditions, and boundary motion conditions between the insulating oil and the fault arc bubble. The boundary motion conditions between the insulating oil and the fault arc bubble are updated using a user-defined function, which is described by bubble dynamics. ,Mode middle, Where is the bubble radius; The rate of change of the bubble radius; This refers to the acceleration due to the change in bubble radius. For fluid density; This represents the distance from the bubble to the fluid boundary. It is the energy of the electric arc; Energy dissipated by viscous forces; Energy conversion rate; This represents the pressure at the boundary of the fluid domain.

[0012] (5) Place pressure measuring points in the fluid domain to monitor pressure changes during the fault.

[0013] (6) Set the fault duration and calculation step size to obtain the pressure change results at the measuring point during the internal arc fault of the transformer.

[0014] (7) Repeat the above process (2)-(6) to obtain the simulation results of all sample points.

[0015] Furthermore, step five specifically includes: Based on the design variables and stress simulation results of all sample points, a Kriging surrogate model is preferably constructed.

[0016] Furthermore, the mathematical model of the Kriging proxy model is as follows: , In the formula, To provide a global approximate simulation of the design space, For a Gaussian steady-state process, To optimize the target fitting function.

[0017] Will Using constants If we express this as an expression, then the mathematical model can be represented as: , Its covariance matrix can be expressed as: , In the formula, R is the relation matrix, and R' represents any two points. and Related functions.

[0018] Using a Gaussian correlation function, the expression is as follows: , In the formula, and They represent and The One element, This represents the relevant parameter vector.

[0019] The response estimates for points outside the sample points are: , In the formula, The correlation vector between the initial sample data is as follows: , The regression coefficient β and random variance can be obtained using the generalized least squares regression method. 2 The estimated value: , , After obtaining the relevant parameters, the surrogate model is used to predict the response of unknown sample points in the design space.

[0020] Furthermore, step six specifically involves: taking minimizing the peak oil pressure at the internal measuring points of the transformer as the objective, and using the coordinate range of the possible installation positions of the pressure relief valve on the top cover as constraints, constructing the mathematical description of the transformer pressure relief valve installation position optimization problem as follows: , In the formula, The coordinates of the center of the pressure relief valve; This refers to the number of pressure relief valves; and These are the start and end times of the fault, respectively. For time; This refers to the feasible domain for the installation location.

[0021] Furthermore, step seven specifically involves: based on the constructed proxy model, preferably, using an improved simulated annealing algorithm to find the optimal solution to the optimization problem, thereby obtaining the optimal installation position of each pressure relief valve.

[0022] Furthermore, the improved simulated annealing algorithm includes the following steps: Initialize basic parameters; Construct the current solution; Perform a neighborhood search on the current solution to obtain the neighborhood solution; Update the current solution; If the termination condition is met, the algorithm ends and outputs the optimal solution.

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention discloses a method for optimizing the installation position of a transformer pressure relief valve in the event of a high-energy arc fault. First, parametric modeling is performed to determine the parameters to be optimized and to sample points within the design space. Then, based on the ANSYS platform, internal transformer fault simulation calculations are conducted to obtain the input of each sample point and the corresponding simulated oil pressure response, constructing a surrogate model. Finally, a mathematical model for the pressure relief valve installation position optimization problem is established, and the optimal installation position of the pressure relief valve is found using a collaborative optimization method combining the surrogate model and an adaptive optimization algorithm.

[0024] This invention, based on transformer internal fault simulation methods, integrates parametric modeling, random sampling, surrogate models, and optimization algorithms to quantitatively optimize the position of transformer pressure relief valves. It overcomes the empirical shortcomings of traditional layout methods, improves the explosion-proof capability of oil-immersed transformers when facing internal arc faults, and the surrogate model can efficiently approximate the complex relationship between variables and responses, avoiding the need to solve a large number of simulation models ergonomically, thus significantly improving computational efficiency and having great practical application value. Attached Figure Description

[0025] Figure 1 This is a flowchart of the transformer pressure relief valve installation position optimization method of the present invention; Figure 2 The transformer geometric model is an embodiment of the present invention; Figure 3 This is a schematic diagram of a two-dimensional coordinate system established on the top cover of the transformer, as an embodiment of the present invention. Figure 4 This is a schematic diagram of the sampling point distribution according to an embodiment of the present invention; Figure 5 The following are simplified transformer solid domain models and insulating oil fluid domain models according to embodiments of the present invention; Figure 6 This is a comparison diagram of oil pressure at measuring points before and after optimizing the installation position of the pressure relief valve in an embodiment of the present invention. Detailed Implementation

[0026] To facilitate understanding of the present invention, a more comprehensive and detailed description of the invention is provided below with reference to the accompanying drawings and embodiments. It should be noted that the present invention can be implemented in many different forms, and the embodiments described herein are only for better explaining the invention and making its content easier to understand, and are not the only limitations.

[0027] This invention discloses a method for optimizing the installation position of a transformer pressure relief valve. It combines internal transformer fault simulation calculations with single-objective optimization based on a surrogate model to optimize the valve's position, thereby maximizing its pressure relief effect. This embodiment uses a ZZDFPZ-406000 / 500-400 converter transformer as the research object to optimize the pressure relief valve's installation position. (See [link to relevant documentation]). Figure 1 The optimization method includes the following steps: (1) Draw a three-dimensional geometric model of the transformer and pressure relief valve based on the actual transformer structure; Specifically, a 1:1 full-size 3D model of the ZZDFPZ-406000 / 500-400 converter transformer was created using Solidworks software, including components such as the pressure relief valve, transformer body, windings, core, and riser. Figure 2 As shown.

[0028] (2) Based on the top cover dimensions and pressure relief valve size in the three-dimensional geometry of the transformer, determine the optimization parameters (i.e., pressure relief valve installation position) and the variable design space of the optimization parameters for the pressure relief valve installation position. Specifically, in this embodiment, the transformer top cover is a rectangle with dimensions of 9.16m × 3.33m, the pressure relief valves have a pressure relief port diameter of 150mm, and the outer diameter of the part in contact with the top cover is 170mm.

[0029] Establish a two-dimensional xy coordinate system with the center of the top cover as the center point, such as... Figure 3 As shown. Because the pressure relief valve cannot extend beyond the top cover boundary and cannot overlap with other components installed on the top cover, the coordinate position of pressure relief valve 1 is... Coordinate position of pressure relief valve 2 The geometric constraints, i.e., the design space of the variables for the parameters to be optimized, are as follows: , In the formula, the unit of measurement is mm.

[0030] (3) Sample points are collected within the variable design space of the parameters to be optimized; Specifically, the initial sample set is generated using Latin hypercube sampling (LHS), including the following steps: For the installation position parameters of the two pressure relief valves to be optimized in this embodiment, the variable dimension is determined to be 4. 50 sets of samples are collected in the effective installation area of ​​the transformer top cover within the plane of the top cover. The number of samples is selected to be 50 sets to satisfy the empirical criterion N≥10 times the variable dimension. The range of values ​​for each variable is evenly divided into 50 non-overlapping sub-intervals; Subsequently, each variable's 50 sub-intervals are independently and randomly permuted. The Fisher-Yates shuffle algorithm is used to generate the random permutation sequence. Within each permuted sub-interval, a sample value is randomly selected according to a uniform distribution, with each sub-interval being sampled only once.

[0031] After completing univariate sampling, the four independently generated sample sequences are randomly combined into 50 groups of 4D samples. This ensures that the one-dimensional projection of each variable uniformly covers the variable design space, and that the distribution of sample points in the multidimensional space satisfies the no-cluster property. The sample point generation results are as follows: Figure 4 As shown.

[0032] To verify the sampling quality, a spatial coverage quantification evaluation method was adopted. All variables had a spatial coverage of no less than 90%, indicating good sampling quality.

[0033] (4) Based on the sample set, perform internal fault simulation calculations of the transformer on the sample points to obtain the simulation calculation results of all sample points; Specifically, the three-dimensional geometric model of the transformer established in (1) is imported into ANSYS SpaceClaim, and the model is then analyzed based on the coordinates of the extracted sample points. and The transformer model corresponding to each sample point is obtained by moving the installation position of the pressure relief valve.

[0034] Simplify minor structures with minimal impact on simulation results, such as bolt holes and nameplates, to reduce computational load. In the transformer solid-state model of SpaceClaim, perform Boolean subtraction operations on solid domains such as windings and cores to extract and generate the insulating oil fluid domain, such as... Figure 5 The figure shows the transformer solid domain model and insulating oil fluid domain model for a sample point in this embodiment.

[0035] After completing the geometric processing, the insulating oil fluid domain model is imported into the ANSYS Fluent module. In Meshing, a tetrahedral mesh is selected, with the global mesh size set to 10 mm and the near-wall surface mesh size to 2 mm.

[0036] The transient solver is activated, a second-order implicit time scheme is used, and the RSM turbulence model is selected. High-precision turbulence simulation is achieved by solving the transport equations for the Reynolds stress components. The transport equations for the Reynolds stress are as follows: , In the formula, The average density of the fluid; Represents the Reynolds stress tensor components; This is the turbulent dissipation rate component; For turbulent viscosity; These are spatial coordinate components; For speed; The component is the rotational angular velocity. For the correction of the Reynolds stress by the Coriolis force caused by rotation, For pressure.

[0037] The insulating oil material has the following properties: density 870 kg / m³, dynamic viscosity 0.02 Pa·s, and bulk modulus 1.2 × 10⁻⁶. 9 Pa.

[0038] The flow field boundary conditions are set as follows: the contact surface between the insulating oil and the tank wall is set as a fixed wall surface; after the pressure relief valve reaches the action threshold pressure of 70 kPa, the outlet surface changes from a wall surface to an outlet boundary condition; the oil conservator is set as an outlet boundary condition connected to atmospheric pressure; the motion conditions of the fault arc bubble boundary are defined by a user-defined function, i.e., the bubble dynamics model. , In the formula, Where is the bubble radius; The rate of change of the bubble radius; This refers to the acceleration due to the change in bubble radius. For fluid density; This represents the distance from the bubble to the fluid boundary. It is the energy of the electric arc; Energy dissipated by viscous forces; Energy conversion rate; This represents the pressure at the boundary of the fluid domain.

[0039] Pressure measurement points are placed in the fluid domain to monitor pressure changes during faults. The specific location of the measurement point is the center of the transformer side wall, and the spatial coordinates of the measurement point are accurately input through the Named Expressions function.

[0040] The fault duration is set to 80ms, the calculation step size is 0.1ms, and the residual of the continuity equation is less than 10. -4 The residual of the energy equation is less than 10. -6 As a convergence criterion, the pressure change at the measuring point during the internal arc fault of the transformer in this embodiment was obtained.

[0041] Repeat the above process to obtain the simulation results for all sample points.

[0042] (5) Based on the simulation results of each sample point, a Kriging surrogate model is constructed to fit the functional relationship between the parameters to be optimized at the installation position of the pressure relief valve and the peak oil pressure inside the transformer. The mathematical model of the Kriging surrogate model is as follows: , In the formula, To provide a global approximate simulation of the design space, For a Gaussian steady-state process, To optimize the objective fitting function, we will... Using constants If we express this as an expression, then the mathematical model can be represented as: , Its covariance matrix can be expressed as: , In the formula, It is a relation matrix. For any two points x i and x j Related functions.

[0043] Using a Gaussian correlation function, the expression is as follows: , In the formula, and They represent and The One element, This represents the relevant parameter vector.

[0044] The response estimates for points outside the sample points are: , In the formula, The correlation vector between the initial sample data is as follows: , The regression coefficients can be obtained using the generalized least squares regression method. and random variance The estimated value: , , After obtaining the above parameters, the model is used to predict the response of unknown sample points in the design space.

[0045] (6) The mathematical description of the optimization problem is constructed with minimizing the peak oil pressure at the measuring point inside the transformer as the objective function and the variable design space of the parameters to be optimized at the installation position of the pressure relief valve as the constraint condition. Specifically, with the objective function being minimizing the peak oil pressure at the internal measuring points of the transformer, and the constraint condition being the coordinate range of the two pressure relief valves' installable positions on the top cover in this embodiment, the mathematical description of the transformer pressure relief valve installation position optimization problem is as follows: , In the formula, For time, Design space for the variables in step (2), that is: .

[0046] (7) Based on the construction of the Kriging proxy model, the improved simulated annealing algorithm is used to optimize the coordinate parameters of the pressure relief valve globally.

[0047] During the algorithm initialization phase, the initial temperature is set. Annealing coefficient Maximum number of iterations Maximum neighborhood search times ; Based on the simulation results of the sample points, the sample point with the best performance is selected as the current solution. It is also the current optimal solution. ;

[0048] Randomly select a neighborhood search operator to perform a neighborhood search on the current solution, and store the searched neighborhood solutions in... In the middle, if the number of neighborhood searches reaches Then The optimal solution in the solution is used as the new solution generated at that temperature. ; like The objective function value is better than Then assign the new solution to and Otherwise, update x0 according to the Metropolis criterion; Cool down Return to the previous iteration for recalculation; The calculation terminates when the maximum number of iterations is reached, and the optimal solution is output. .

[0049] In this embodiment, the optimized coordinates for pressure relief valve 1 and pressure relief valve 2 are (-2042, -528) and (1334, 419), respectively. Specifically, under an arc energy of 2 MJ, before and after the optimization of the transformer pressure relief valve installation position in this embodiment, the opening time of the pressure relief valve is shortened from 20.5 ms and 31.2 ms to 15.7 ms and 24.4 ms, respectively. The oil pressure change curve at the measuring point is shown in the figure. Figure 6 As shown, the peak oil pressure before optimization was 425 kPa, and the peak oil pressure after optimization was 381 kPa, a reduction of 10.3%. This result indicates that the transformer pressure relief valve installation position optimization method provided by this invention, applicable to high-energy arc faults, can enable the pressure relief valve to respond more quickly when an arc fault occurs inside the transformer, thereby more effectively suppressing the rise in oil pressure and reducing the risk of equipment explosion due to arc faults.

[0050] The above description is a further detailed explanation of the present invention in conjunction with specific preferred embodiments. It should not be considered that the specific embodiments of the present invention are limited to this. For those skilled in the art, several simple deductions or substitutions can be made without departing from the present invention, and all of these should be considered to fall within the scope of patent protection determined by the submitted claims.

Claims

1. A method for optimizing the installation position of a transformer pressure relief valve suitable for high-energy arc faults, characterized in that, Includes the following steps: Step 1: Draw a three-dimensional geometric model of the transformer and pressure relief valve based on the actual transformer structure; Step 2: Based on the dimensions of the top cover and the size of the pressure relief valve in the three-dimensional geometric model of the transformer, determine the optimization parameters for the installation position of the pressure relief valve and the variable design space for the optimization parameters. Step 3: Sample points are collected within the variable design space of the parameters to be optimized to obtain a sample set; Step 4: Based on the sample set, perform pressure simulation calculations for the sample points under internal arc fault conditions in the transformer, and obtain the simulation calculation results for all sample points. Step 5: Based on the simulation results of each sample point, construct a surrogate model. The surrogate model is used to fit the functional relationship between the parameters to be optimized at the installation position of the pressure relief valve and the peak value of the internal oil pressure of the transformer. Step 6: Construct a mathematical description of the optimization problem with the objective function of minimizing the peak oil pressure at the internal measuring point of the transformer and the variable design space of the parameters to be optimized for the installation position of the pressure relief valve as the constraint condition; Step 7: Based on the constructed proxy model, use an optimization algorithm to solve the mathematical description of the optimization problem and obtain the installation coordinates of the pressure relief valve when the peak oil pressure inside the transformer is at its minimum under the internal arc fault.

2. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 1, characterized in that, The optimization parameters for determining the installation location of the pressure relief valve and the variable design space for the optimization parameters include: determining the range of possible installation locations for the pressure relief valve in a parametric form based on the spatial structure of the top cover and the physical dimensions of the pressure relief valve in the drawn transformer geometric model.

3. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 2, characterized in that, The method for determining the range of possible installation locations for the pressure relief valve using the parameterized form is as follows: establish a two-dimensional xy coordinate system with the center of the transformer top cover as the origin, and determine the feasible region of the pressure relief valve installation location coordinates based on the top cover boundary coordinates and the pressure relief valve diameter.

4. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 1, characterized in that, The method for sampling sample points in step three is to perform spatial sampling within the design space.

5. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 1, characterized in that, Step four involves performing pressure simulation calculations for the sample points during an internal arc fault in the transformer, including: Import the three-dimensional geometric model drawn in step one into the Spaceclaim software of the ANSYS platform. According to the coordinate position of the pressure relief valve in each sample point, move and change the position of the pressure relief valve in the model accordingly, and simplify the transformer to improve the simulation calculation efficiency. Based on the simplified solid domain model of the transformer, the fluid domain geometric model of the insulating oil part inside the transformer is drawn. Import the insulating oil fluid domain model into the ANSYS Fluent module and mesh the fluid domain; The simulation solver is set to a transient solver. The insulating oil material properties and flow field boundary conditions are defined. The RSM turbulence model is used to simulate the fluid motion. The RSM is closed by solving the Reynolds stress transport equation and dissipation rate equation. The Reynolds stress transport equation is as follows: ， In the formula, The average density of the fluid; Represents the Reynolds stress tensor components; This is the turbulent dissipation rate component; For turbulent viscosity; These are spatial coordinate components; For speed; The component is the rotational angular velocity. For the correction of the Reynolds stress by the Coriolis force caused by rotation, For pressure; Place pressure measurement points in the fluid domain to monitor pressure changes during a fault; By setting the fault duration and calculation step size, the pressure change at the measuring point during the internal arc fault of the transformer is obtained. Repeat the above process to obtain the simulation calculation results for all sample points.

6. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 5, characterized in that, The flow field boundary conditions specifically include: wall boundary conditions in contact with the oil tank wall and windings, pressure relief valve outlet boundary conditions, oil conservator outlet boundary conditions, and boundary motion conditions between insulating oil and fault arc bubbles. The boundary motion conditions between the insulating oil and the fault arc bubble are updated using a user-defined function, which is described by bubble dynamics. ， In the formula, Where is the bubble radius; The rate of change of the bubble radius; This refers to the acceleration due to the change in bubble radius. For fluid density; This represents the distance from the bubble to the fluid boundary. For electric arc energy; W Energy dissipated by viscous forces; Energy conversion rate; This represents the pressure at the boundary of the fluid domain.

7. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 1, characterized in that, The method for constructing the surrogate model in step five is as follows: Based on the design variables and stress simulation results of all sample points, a Kriging surrogate model is constructed, the mathematical expression of which is: ， In the formula, For a global approximate simulation of the design space, Z(X) is a Gaussian steady-state process, and Y(X) is the optimization objective fitting function. Using constants If we express this as an expression, then the mathematical model can be represented as: ， The response estimates for points outside the sample points are: ， In the formula, This is the correlation vector between the initial sample data. This is the correlation matrix; where the regression coefficients are... and random variance The estimated value is: ， 。 8. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 1, characterized in that, The mathematical description of the optimization problem is as follows: ， In the formula, The coordinates of the center of the pressure relief valve; This refers to the number of pressure relief valves; and These are the start and end times of the fault, respectively. For time; This refers to the feasible domain for the installation location.

9. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 1, characterized in that, In step seven, the optimization algorithm is used to solve the optimization problem as follows: based on the constructed surrogate model, the improved simulated annealing algorithm is used to find the optimal solution to the optimization problem, thereby obtaining the optimal installation position of each pressure relief valve.

10. The method for optimizing the installation position of a transformer pressure relief valve for high-energy arc faults according to claim 9, characterized in that, The improved simulated annealing algorithm steps are as follows: initialize basic parameters; construct an initial solution; perform a neighborhood search on the current solution to obtain a neighborhood solution; update the current solution; if the termination condition is met, the algorithm ends and outputs the optimal solution.

Images