Power transformer creepage path analysis method and system based on Grey Wolf optimization algorithm

Abstract

The present invention proposes a method and system for analyzing the creepage path of a power transformer based on the Gray Wolf optimization algorithm, which performs numerical discretization processing on the transformer equivalent body structure diagram to obtain a map numerical matrix; obtains the electric field gradient of each point of the power transformer in the map numerical matrix based on electric field simulation calculation; uses the heuristic Gray Wolf optimization algorithm to search the map numerical matrix along the direction of the transformer electric field gradient, analyzes the creepage path of the power transformer, and automatically outputs the potential power transformer creepage path; and the output potential power transformer creepage path is post-processed and displayed in the original CAD drawing of the transformer equivalent body structure diagram. In view of the current situation that the mainstream graph theory algorithm cannot embed rules to meet the needs of the creepage path analysis of power transformers, the patent of the present invention effectively constructs a numerical map of the internal body of the power transformer through digital discretization processing combined with the internal insulation characteristics of the transformer, so that it can be effectively analyzed and solved.

Description

Technical Field

[0001] The invention belongs to the technical field of transformer electric field path optimization and heuristic algorithms, and relates to a method and system for analyzing creepage paths of power transformers based on a Grey Wolf optimization algorithm. Background Art

[0002] The transformer main longitudinal insulation design is the core element to ensure the long-term stable operation, high reliability and extended service life of the power transformer. The insulation performance is directly related to whether the transformer can work safely and efficiently. Therefore, in the preliminary stage of insulation design, it is particularly important to conduct an in-depth and detailed analysis of the local creepage path inside the transformer. This can not only help designers fully understand the possible flow paths of current in the insulation structure, but also effectively identify potential discharge paths and those weak points with low insulation margins and prone to failure. Through such analysis, the insulation design can be optimized in a targeted manner, the overall performance of the transformer can be improved, and its excellent performance in actual operation can be ensured, providing a solid guarantee for the stable operation of the power system.

[0003] However, the composition of the transformer internal winding insulation structure is quite complex. It is usually carefully constructed by key components such as multiple layers of insulating cardboard, electrostatic rings, pads, insulating oil, and pressing blocks. These components not only each bear specific insulation and support functions, but also the spatial layout between them is extremely complex. Insulating cardboard is used to isolate windings with different potentials, electrostatic rings are used to balance the electric field distribution, pads and pressing blocks play a fixing and supporting role, and insulating oil is filled as a cooling and insulating medium. This complex spatial layout requires designers to fully consider the relative position relationship of each component during insulation design to ensure the overall performance and reliability of the insulation structure, thereby ensuring the safe operation of the transformer. In addition, there are many types of insulation materials inside the transformer, each of which has different dielectric constants, which has an important influence on the discharge path. The difference in dielectric constants may lead to uneven electric field distribution, which in turn causes safety hazards such as partial discharge or insulation breakdown. Therefore, in the insulation design, the dielectric properties of various insulating materials must be fully considered to ensure the rationality of the electric field distribution. At the same time, in the process of path optimization, we also need to closely combine the actual characteristics of the project. This includes considering many factors such as the size, structure, working environment and operating requirements of the transformer. In order to find the optimal discharge path, we need to embed various special rules in the optimization process. These rules may involve electrical safety, mechanical strength, thermal stability and other aspects. However, given the starting point and end point, there are often thousands of possible implementation paths. Manual troubleshooting is not only time-consuming and labor-intensive, but also difficult to avoid omissions. This undoubtedly poses a huge hidden danger to the safety of product design. Therefore, we need to use advanced computing methods and tools to assist in path optimization and insulation design to ensure the safe and reliable operation of the transformer. However, the current mainstream path optimization algorithms are mostly based on graph theory principles. They can only be used for simple maps (feasible domains and infeasible domains) and cannot embed more complex map rules (space layout of pads, insulating cardboard) to meet the actual transformer discharge path requirements. Summary of the invention

[0004] In view of the problems existing in the prior art, the purpose of the present invention is to propose a method and system for analyzing the creepage path of power transformers based on the Gray Wolf optimization algorithm, which can effectively combine the internal insulation characteristics of the transformer to obtain a map data matrix after discrete processing; obtain the potential gradient information of each point in the map data matrix through the electric field simulation calculation software, and provide a basis for subsequent search directions. The map data matrix can be effectively solved by the Gray Wolf optimization algorithm, and the potential low insulation margin points inside the power transformer can be output, providing an effective basis for insulation strength verification, and meeting the actual transformer discharge path requirements.

[0005] The present invention is achieved through the following technical solutions: A method for analyzing creepage paths of power transformers based on a Grey Wolf optimization algorithm, comprising: Perform numerical discretization on the transformer equivalent body structure diagram to obtain a map numerical matrix; Obtain the electric field gradient at each point of the power transformer in the map numerical matrix based on electric field simulation calculation; According to the set starting point and end point, the heuristic gray wolf optimization algorithm is used to search the map numerical matrix along the transformer electric field gradient direction, analyze the creepage path of the power transformer, and automatically output the potential power transformer creepage path; The output potential power transformer creepage path is post-processed and then displayed in the original CAD drawing of the transformer equivalent body structure diagram.

[0006] Preferably, the transformer equivalent body structure diagram is numerically discretized, and the specific process is as follows: According to the discretization precision k, the CAD graphics of the drawn transformer equivalent body structure diagram are numerically discretized based on the structural parts type of the transformer, and the region without graphic elements in the transformer equivalent body structure diagram in the map numerical matrix is ​​assigned a value of 0, the insulating paper tube in the transformer equivalent body structure diagram is assigned a value of 1, and the pad object in the transformer equivalent body structure diagram is assigned a value of 2, thereby generating a map numerical matrix; Preferably, during the discretization process, the arcs and straight lines in the CAD graphics of the transformer equivalent body structure diagram are discretized, and the arc length is l The arc is equivalent to l / k straight lines connected in sequence; the discretization accuracy k is obtained by segmenting any primitive line and any arc length in the CAD drawing of the transformer equivalent body structure diagram.

[0007] Preferably, the area assigned with a value of 0 in the map numerical matrix represents a searchable area. If the area is reached during the search, the search can be performed in any direction from this point; The area assigned with a value of 1 in the map numerical matrix represents a global impenetrable area. If it is reached during the search process, only the search along the outline is performed to reach its boundary; The area assigned a value of 2 in the map numerical matrix represents a penetrable area. If this area is reached during the search process, a penetration search is performed along the diagonals and edges by combining the actual characteristics of the dielectric material in the transformer body.

[0008] Preferably, the starting point and the end point include the high voltage winding of the transformer to the low voltage winding, the high voltage winding of the transformer to the ground, and the low voltage winding of the transformer to the ground.

[0009] Preferably, the electric field gradient information of each point of the power transformer in the map numerical matrix is ​​obtained based on the electric field simulation calculation, specifically: The finite element method is used to simulate the two-dimensional electric field of the power transformer in the map numerical matrix. The electric field area is discretized into a finite number of grid cells. Based on the boundary conditions and medium parameters, the By solving Poisson's equation, the electric field gradient information of each point in the electric field area can be effectively obtained; The formula for Poisson's equation is:

[0010] Where ∇ is the gradient operator, ε is the dielectric constant, V is the electric potential, and ρ is the charge density.

[0011] Preferably, according to the set starting point and end point, the heuristic gray wolf optimization algorithm is used to search the map numerical matrix along the transformer electric field gradient direction, analyze the creepage path of the power transformer, and automatically output the potential discharge path. The specific process is: The creepage path of the transformer in the map numerical matrix is ​​searched by the gray wolf optimization algorithm, and the search is performed along the direction of the electric field gradient. According to the electric field gradient search characteristics, during the search process, each individual of the gray wolf population updates its position along the direction of the electric field gradient according to the step vector; in each round of search iteration, the first three high-quality solutions are selected from the N gray wolf populations, named α, β and γ respectively, and the other N-3 individual solutions are guided by these three high-quality solutions to perform path iterative optimization, and the remaining N-3 individual solutions are updated based on the three high-quality solution individuals of α, β and γ until the optimal creepage path is obtained; among them, the creepage path of the transformer is taken as an individual in the gray wolf population.

[0012] Preferably, the initial solution of the grey wolf optimization algorithm is obtained by: Based on the MAINLINK graph theory algorithm, the individual initial solutions in the gray wolf population of the power transformer creepage path are generated, and the initial solutions are recorded as {V1, V2, …, Vn}. Then, based on the initial solutions, the nodes of the power transformer creepage path are discretized and divided into k sub-routes. The input data dimension used by the optimization function of the gray wolf optimization algorithm is a sequence of length n×k. Then, the initial solution of the sequence is used as the initial state solution of the gray wolf optimization algorithm for subsequent iterative optimization.

[0013] Preferably, the post-processing is: converting the potential power transformer creepage path output by the Grey Wolf optimization algorithm into a format that can be displayed in the CAD graphics of the transformer equivalent body structure diagram.

[0014] A power transformer creepage path analysis system based on the Grey Wolf optimization algorithm, comprising: A discretization processing module is used to perform numerical discretization processing on the transformer equivalent body structure diagram to obtain a map numerical matrix; A simulation calculation module, used for obtaining the electric field gradient of each point of the power transformer in the map numerical matrix based on electric field simulation calculation; Gray Wolf Optimization Algorithm Search Module, which is used to search the map numerical matrix along the transformer electric field gradient direction using the heuristic Gray Wolf Optimization Algorithm according to the set starting point and end point, analyze the creepage path of the power transformer, and automatically output the potential creepage path of the power transformer; The post-processing module is used for post-processing the output potential power transformer creepage path and displaying it back in the original CAD drawing of the transformer equivalent body structure diagram.

[0015] Compared with the prior art, the present invention has the following beneficial technical effects: The purpose of the present invention is to propose a method and system for analyzing the creepage path of a power transformer based on the gray wolf optimization algorithm, which can effectively combine the internal insulation characteristics of the transformer to obtain a map data matrix after discrete processing; the potential gradient information of each point in the map data matrix is ​​obtained through the electric field simulation calculation software to provide a basis for the subsequent search direction. The map data matrix can be effectively solved by the gray wolf optimization algorithm, and the potential low insulation margin points inside the power transformer are output, providing an effective basis for insulation strength verification. In view of the current situation that the mainstream graph theory algorithm cannot embed rules to meet the needs of creepage path analysis of power transformers, the patent of this invention effectively constructs a numerical map of the internal body of the power transformer through digital discretization processing combined with the internal insulation characteristics of the transformer, so that it can be effectively analyzed and solved. In view of the problems that the internal creepage path planning of the power transformer is a nonlinear discrete problem and the traditional operations research solution is inefficient, this patent proposes a heuristic solution algorithm based on the gray wolf optimization algorithm. The algorithm simulates the ladder system and hunting process of wolves in nature. As a bionic algorithm, it has the characteristics of strong robustness and low computational complexity. In the planning process of this creepage path, the actual distribution characteristics of the electric field inside the transformer are combined. The specific implementation is to solve the two-dimensional electric field through electric field simulation calculation software and extract the internal electric field gradient information. This process is more in line with the actual situation. The patent of this invention can automatically output the low insulation margin combination in the specific creepage path, providing an effective means for the creepage margin analysis and verification of power transformers.

[0016] Furthermore, the gray wolf optimization algorithm proposed in this patent updates the state position based on the first three optimal solutions during the population update process. However, the mainstream genetic algorithm and particle swarm optimization algorithm only update the state based on the optimal solution. Therefore, this algorithm has better search breadth and robustness during the optimization process.

[0017] Furthermore, according to the set working conditions (high voltage to ground, low voltage to ground and high voltage to low voltage), the patent of the present invention can automatically carry out the internal creepage path analysis of the transformer, which has strong versatility and avoids the potential problems of low efficiency and high error rate of manual search. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1Flowchart of creepage path analysis of power transformers for Gray Wolf optimization algorithm; Figure 2 The 0 area in the discretized map data matrix represents the searchable area in any direction search map; Figure 3 The 1 area in the discretized map data matrix represents the linear boundary search map of the insulating paper tube; Figure 4 A step search map for the connected regions of the 2-region representative compact in the map data matrix; Figure 5 Update the schematic diagram for the individual status of the gray wolf; Figure 6 Process diagram for discretization of map data matrix; Figure 7 The creepage path of a power transformer output for the Grey Wolf algorithm is displayed in a CAD drawing. DETAILED DESCRIPTION

[0019] The present invention is further described in detail below in conjunction with specific embodiments, which are intended to explain the present invention rather than to limit it.

[0020] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0021] In order to overcome the many defects and bottlenecks in the current power transformer electric field creepage path analysis, the purpose of the present invention is to propose a power transformer creepage path analysis method based on the Gray Wolf optimization algorithm. First, the map can be effectively constructed in combination with the internal insulation characteristics of the transformer to construct the model to be solved; secondly, the potential information of each point in the map is obtained through the electric field simulation calculation software to provide a basis for the subsequent search direction. Finally, the model can be effectively solved by the Gray Wolf optimization algorithm, and the potential low insulation margin points inside the power transformer can be output, providing an effective basis for insulation strength verification.

[0022] like Figure 1 As shown, a creepage path analysis method for a power transformer based on a grey wolf optimization algorithm comprises: S1. Perform numerical discretization processing on the transformer equivalent body structure diagram to obtain a map numerical matrix; Among them, the transformer equivalent body structure diagram is numerically discretized, and the specific process is as follows: like Figure 6As shown, according to the discretization accuracy k, the CAD graphics of the drawn transformer equivalent body structure diagram are numerically discretized based on the structural parts type of the transformer, and the region without graphic elements in the transformer equivalent body structure diagram in the map numerical matrix is ​​assigned a value of 0, the insulating paper tube in the transformer equivalent body structure diagram is assigned a value of 1, and the pad object in the transformer equivalent body structure diagram is assigned a value of 2, thereby generating a map numerical matrix; like Figure 2 As shown, the area assigned with a value of 0 in the map value matrix represents a searchable area. If you reach this area during the search process, you can search in any direction from this point; like Figure 3 As shown, the area assigned with a value of 1 in the map value matrix represents a global impenetrable area. If it is reached here during the search process, only the search along the outline is performed to reach its boundary; like Figure 4 As shown, the area assigned a value of 2 in the map numerical matrix represents a penetrable area. If this area is reached during the search process, a penetration search is performed along the diagonals and edges by combining the actual characteristics of the dielectric material in the transformer body.

[0023] Furthermore, in the discretization process, the arcs and straight lines in the CAD graphics of the transformer equivalent body structure diagram are discretized, and the arc length is l The arc is equivalent to l / k straight lines connected in sequence; the discretization accuracy k is obtained by segmenting any primitive line and any arc length in the CAD drawing of the transformer equivalent body structure diagram.

[0024] Furthermore, the map numerical matrix is ​​modified by parameters, specifically: Combined with the actual characteristics of the dielectric material in the transformer body, the node connectivity is judged in the map numerical matrix to determine whether each entity is a special pad. If so, diagonal penetration can be performed and the map numerical matrix can be parameter corrected.

[0025] S2. Obtain the electric field gradient of each point of the power transformer in the map numerical matrix based on the electric field simulation calculation software, specifically: The finite element method is used to simulate the two-dimensional electric field of the power transformer in the map numerical matrix. The electric field area is discretized into a finite number of grid cells. Based on the boundary conditions and medium parameters, the By solving Poisson's equation, the electric field gradient information of each point in the electric field area can be effectively obtained; The formula for Poisson's equation is:

[0026] Where ∇ is the gradient operator, ε is the dielectric constant, V is the electric potential, and ρ is the charge density.

[0027] Thereby obtaining the global electric field strength, divergence and gradient information under various operating conditions of the power transformer.

[0028] like Figure 6 As shown, S3, according to the set starting point and end point, the heuristic gray wolf optimization algorithm is used to search the map numerical matrix along the direction of the transformer electric field gradient, analyze the creepage path of the power transformer, and automatically output the potential power transformer creepage path; According to the set starting point and end point, the heuristic gray wolf optimization algorithm is used to search the map numerical matrix along the direction of the transformer electric field gradient, analyze the creepage path of the power transformer, and automatically output the potential discharge path. The specific process is as follows: Based on the initial state solution of the gray wolf optimization algorithm, the creepage path of the transformer in the map numerical matrix is ​​searched by the gray wolf optimization algorithm, and the search is performed along the direction of the electric field gradient. During the search process, each individual of the gray wolf population updates its position along the direction of the electric field gradient according to the step vector; in each round of search iteration, the top three high-quality solutions are selected from the N gray wolf populations, named α, β and γ respectively, and the other N-3 individual solutions are guided by these three high-quality solutions to perform path iterative optimization. The positions of the remaining N-3 individual solutions are updated based on the three high-quality solution individuals of α, β and γ until the optimal solution creepage path is obtained; among them, the creepage path of the transformer is taken as an individual in the gray wolf population.

[0029] Among them, the initial solution of the gray wolf optimization algorithm is obtained in the following specific process: Based on the MAINLINK graph theory algorithm, the individual initial solutions in the gray wolf population of the power transformer creepage path are generated, and the initial solutions are recorded as {V1, V2, …, Vn}. Then, based on the initial solutions, the nodes of the power transformer creepage path are discretized and divided into k sub-routes. The input data dimension used by the optimization function of the gray wolf optimization algorithm is a sequence of length n×k. Then, the initial solution of the sequence is used as the initial state solution of the gray wolf optimization algorithm for subsequent iterative optimization.

[0030] The remaining N-3 individual status updates can be performed using the following formula.

[0031]

[0032] Where: parameter is A 1 , A 2 , A 3 are random vectors with values ​​in [-1,1], simulating the approach and distance operations of wolves in the hunting process. When they are in [0,1], they ensure the individual convergence, and when they are in [-1,0], they expand the search domain. Among them C 1 , C 2 and C 3is a set of random vectors that decay linearly in [0,2], simulating the uncertainty of each step forward of the wolf pack.

[0033] The values ​​of each dimension in vector C are obtained based on the following formula, where t is the current iteration number, T is the total iteration number, and r1 is a random number generated by a standard normal distribution with a mean of 1 and a variance of 0.

[0034]

[0035] During the search process, each individual has a certain probability to follow the step vector , update the position along the gradient direction, specifically:

[0036] In the formula, is the Hamiltonian operator; E is the finite element field extractor Vector dynamic attenuation vector, which can effectively ensure that individuals search effectively in the feasible domain of the map and find potential feasible solutions at the beginning of the iteration; and ensure effective convergence at the end of the iteration; The starting point and the end point include the high voltage winding of the transformer to the low voltage winding, the high voltage winding of the transformer to the ground, and the low voltage winding of the transformer to the ground.

[0037] When analyzing the creepage path of the power transformer of the gray wolf optimization algorithm, the creepage path distance is defined as the Euclidean distance, which becomes the individual with the smallest path distance among the individuals of the gray wolf population, thereby guiding the rest of the population to update their positions.

[0038] like Figure 7 As shown, S4, the output potential power transformer creepage path is post-processed and then displayed in the original CAD drawing of the transformer equivalent body structure diagram. The post-processing is: converting the potential power transformer creepage path output by the gray wolf optimization algorithm into a format that can be displayed in the CAD drawing of the transformer equivalent body structure diagram; Embodiment 1, Discretize the transformer 2D equivalent insulation diagram into a matrix and give a given discretization accuracy (how long any line or arc in CAD is divided) k mm, the completed CAD file of the device body is numerically discretized, and the specific rules are: the searchable domain is assigned a value of 0, the global impenetrable area is assigned a value of 1, and the penetrable area is assigned a value of 2. After the numerical map matrix is ​​generated, a redrawing and echoing operation is performed to review whether various features are complete.

[0039] Combined with the actual characteristics of the device, node connectivity is judged to determine whether each entity is a three-dimensional pad, diagonal penetration can be performed, and parameter corrections are made to the numerical map matrix.

[0040] In order to reduce the computational complexity, the initial solution of the Gray Wolf Optimization Algorithm is generated based on the "MAINLINK" graph theory algorithm, and the initial feasible solution is recorded as {V1, V2, ..., Vn}, and then the nodes are discretized and divided into k sub-routes. At this point, the input data dimension of the optimization function is a one-dimensional sequence of length n × k; The electric field simulation calculation software can effectively obtain the electric field gradient information of each point inside the map. The electric field gradient information can be effectively considered during the algorithm search process, and the search will be performed in the gradient direction first.

[0041] Set the gray wolf optimization algorithm, the number of iterations T and the number of wolves N, and start iteration: generate the initial population based on the random strategy; evaluate the fitness of N populations based on the Euclidean distance and select α, β and γ individuals; start iteration; update the individual status; make a probability criterion judgment, if the requirements are met, update the path based on the gradient information; make a judgment on the update of α, β and γ individuals; make a termination condition judgment, if not met, return to the step of evaluating the fitness of N populations based on the Euclidean distance, select α, β and γ individuals, iterate again, and output the final population individuals; The final feasible path solution is output, and the corresponding path is displayed in CAD after post-processing. This algorithm can automatically calculate the insulation margin of each point on the discharge path, providing an effective basis for subsequent insulation design and effectiveness analysis.

[0042] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any ordinary technician in the industry can smoothly implement the present invention as shown in the drawings and described above. However, any equivalent changes, modifications and evolutions made by technicians familiar with the profession without departing from the scope of the technical solution of the present invention using the technical content disclosed above are all equivalent embodiments of the present invention. At the same time, any equivalent changes, modifications and evolutions made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the technical solution of the present invention.

Claims

1. A method for analyzing creepage paths of power transformers based on the Grey Wolf optimization algorithm, characterized in that: include: Perform numerical discretization on the transformer equivalent body structure diagram to obtain a map numerical matrix; Obtain the electric field gradient at each point of the power transformer in the map numerical matrix based on electric field simulation calculation; According to the set starting point and end point, the Gray Wolf optimization algorithm is used to search the map numerical matrix along the transformer electric field gradient direction, analyze the creepage path of the power transformer, and automatically output the potential power transformer creepage path; The output potential power transformer creepage path is post-processed and then displayed in the original CAD drawing of the transformer equivalent body structure diagram.

2. A method for analyzing creepage paths of power transformers based on the Grey Wolf optimization algorithm according to claim 1, characterized in that: The transformer equivalent body structure diagram is numerically discretized, and the specific process is as follows: According to the discretization accuracy k, the CAD graphics of the drawn transformer equivalent body structure diagram are numerically discretized based on the structural parts type of the transformer, and the area without graphic elements in the transformer equivalent body structure diagram in the map numerical matrix is ​​assigned a value of 0; the insulating paper tube in the transformer equivalent body structure diagram is assigned a value of 1; and the pad object in the transformer equivalent body structure diagram is assigned a value of 2, thereby generating a map numerical matrix.

3. A method for analyzing creepage paths of power transformers based on the Grey Wolf optimization algorithm according to claim 2, characterized in that: During the discretization process, the arcs and straight lines in the CAD graphics of the transformer equivalent body structure diagram are discretized, and the arc length is l The arc is equivalent to l / k straight lines connected in sequence; the discretization accuracy k is obtained by segmenting any primitive line and any arc length in the CAD drawing of the transformer equivalent body structure diagram.

4. A method for analyzing creepage paths of power transformers based on the Grey Wolf optimization algorithm according to claim 2, characterized in that: The area assigned with a value of 0 in the map numerical matrix represents a searchable area. If the search process reaches this area, the search can be performed in any direction from this point; The area assigned with a value of 1 in the map numerical matrix represents a global impenetrable area. If it is reached during the search process, only the search along the outline is performed to reach its boundary; The area assigned a value of 2 in the map numerical matrix represents a penetrable area. If this area is reached during the search process, a penetration search is performed along the diagonals and edges by combining the actual characteristics of the dielectric material in the transformer body.

5. The method for analyzing creepage path of power transformer based on Grey Wolf Optimization Algorithm according to claim 1, characterized in that: The starting point and the end point include the high voltage winding of the transformer to the low voltage winding, the high voltage winding of the transformer to the ground, and the low voltage winding of the transformer to the ground.

6. The method for analyzing creepage path of power transformer based on Grey Wolf Optimization Algorithm according to claim 1, characterized in that: Based on the electric field simulation calculation, the electric field gradient information of each point of the power transformer in the map numerical matrix is ​​obtained, specifically: The finite element method is used to simulate the two-dimensional electric field of the power transformer in the map numerical matrix. The electric field area is discretized into a finite number of grid cells. Based on the boundary conditions and medium parameters, the By solving Poisson's equation, the electric field gradient information of each point in the electric field area can be effectively obtained; Among them, Poisson's equation is: Where ∇ is the gradient operator, ε is the dielectric constant, V is the electric potential, and ρ is the charge density.

7. The method for analyzing creepage paths of power transformers based on the Grey Wolf optimization algorithm according to claim 1, characterized in that: According to the set starting point and end point, the Gray Wolf optimization algorithm is used to search the map numerical matrix along the electric field gradient direction of the transformer, analyze the creepage path of the power transformer, and automatically output the potential discharge path. The specific process is as follows: The creepage path of the transformer in the map numerical matrix is ​​searched by the gray wolf optimization algorithm along the direction of the electric field gradient. During the search process, based on the initial solution of the gray wolf optimization algorithm, each individual of the gray wolf population updates its position along the direction of the electric field gradient according to the step vector; in each round of search iteration, the first three high-quality solutions are selected from the N gray wolf populations, named α, β and γ respectively, and the other N-3 individual solutions are guided by these three high-quality solutions to perform path iterative optimization. The positions of the remaining N-3 individual solutions are updated based on the three high-quality solution individuals of α, β and γ until the optimal creepage path is obtained; among them, the creepage path of the transformer is taken as an individual in the gray wolf population.

8. A method for analyzing creepage paths of power transformers based on the Grey Wolf optimization algorithm according to claim 7, characterized in that: The initial solution of the Grey Wolf Optimization Algorithm is obtained as follows: Based on the MAINLINK graph theory algorithm, the individual initial solutions in the gray wolf population of the power transformer creepage path are generated, and the initial solutions are recorded as {V1, V2, …, Vn}. Then, based on the initial solutions, the nodes of the power transformer creepage path are discretized and divided into k sub-routes. The input data dimension used by the optimization function of the gray wolf optimization algorithm is a sequence of length n×k. Then, the initial solution of the sequence is used as the initial state solution of the gray wolf optimization algorithm for subsequent iterative optimization.

9. The method for analyzing creepage path of power transformer based on Grey Wolf Optimization Algorithm according to claim 1, characterized in that: The post-processing is to convert the potential power transformer creepage path output by the Gray Wolf optimization algorithm into a format that can be displayed in the CAD graphics of the transformer equivalent body structure diagram.

10. A power transformer creepage path analysis system based on the Grey Wolf optimization algorithm, characterized in that: include: A discretization processing module is used to perform numerical discretization processing on the transformer equivalent body structure diagram to obtain a map numerical matrix; A simulation calculation module, used for obtaining the electric field gradient of each point of the power transformer in the map numerical matrix based on electric field simulation calculation; Gray Wolf Optimization Algorithm Search Module, which is used to search the map numerical matrix along the transformer electric field gradient direction using the heuristic Gray Wolf Optimization Algorithm according to the set starting point and end point, analyze the creepage path of the power transformer, and automatically output the potential creepage path of the power transformer; The post-processing module is used for post-processing the output potential power transformer creepage path and displaying it back in the original CAD drawing of the transformer equivalent body structure diagram.

Images