Pole mounted circuit breaker testing device and testing method

Abstract

The present application relates to the technical field of pole-mounted circuit breaker testing, and discloses a pole-mounted circuit breaker testing device and testing method. The method comprises: constructing a coupled three-dimensional simulation model of the pole-mounted circuit breaker and performing finite element simulation analysis to obtain power loss simulation data and insulation resistance simulation data; performing change calculation to obtain relationship coefficients; performing feature fusion and feature screening on the power loss change data and the insulation resistance change data based on the relationship coefficients to obtain fused change feature data and using the fused change feature data as the feature nodes of the Markov network to obtain an initial Markov network model; performing multiple iterations and model optimization through a genetic algorithm to obtain a target Markov network model; real-time monitoring of the operation status data of the pole-mounted circuit breaker and performing operation fault prediction through the target Markov network model to obtain real-time fault prediction results. The present application improves the fault detection accuracy of the pole-mounted circuit breaker.

Description

Technical Field

[0001] The present application relates to the technical field of pole-mounted circuit breaker testing, and in particular to a pole-mounted circuit breaker testing device and testing method. Background Art

[0002] Pole-mounted circuit breakers are critical components in power systems, and their operational reliability and stability directly impact the safety of the entire power grid. With the continuous growth of power demand and the increasing complexity of power grid structures, fault detection and diagnosis of pole-mounted circuit breakers are becoming increasingly important. However, traditional circuit breaker testing methods often rely on manual inspections and simple electrical parameter monitoring, which is not only time-consuming and labor-intensive, but also difficult to achieve real-time monitoring and accurate fault location.

[0003] Existing circuit breaker testing methods have several major issues. Traditional methods lack the timeliness and accuracy of fault detection, making it difficult to provide early warnings in the early stages of a fault, leading to further expansion of the fault and increasing maintenance costs and the risk of power outages in the power system. Secondly, traditional methods lack advanced algorithmic support for data analysis and processing, making it impossible to fully utilize large amounts of power data for deep mining and feature extraction, resulting in limited reliability and accuracy of test results. Traditional circuit breaker testing methods often ignore the impact of the equipment operating environment and complex operating conditions, making it difficult to fully reflect the actual working status of pole-mounted circuit breakers under different operating conditions. Summary of the Invention

[0004] The present application provides a pole-mounted circuit breaker testing device and a testing method, which are used to improve the fault detection accuracy of the pole-mounted circuit breaker.

[0005] In a first aspect, the present application provides a pole-mounted circuit breaker testing method, the pole-mounted circuit breaker testing method comprising:

[0006] Obtaining geometric dimensions, material properties, and electrical parameters of a pole-mounted circuit breaker and constructing a coupled three-dimensional simulation model of the pole-mounted circuit breaker;

[0007] defining electric field distribution and magnetic field distribution under different operating conditions, and performing finite element simulation analysis on the insulation resistance and fault location in the coupled three-dimensional simulation model to obtain power loss simulation data and insulation resistance simulation data;

[0008] Performing change calculations on the power loss simulation data and the insulation resistance simulation data respectively to obtain power loss change data and insulation resistance change data, and performing change relationship analysis to obtain a relationship coefficient between power loss change and insulation resistance change;

[0009] Performing feature fusion and feature screening on the power loss change data and the insulation resistance change data based on the relationship coefficient to obtain fused change feature data, and using the fused change feature data as a feature node of a Markov network to obtain an initial Markov network model;

[0010] Performing multiple iterations and model optimization on the initial Markov network model through a genetic algorithm to obtain a target Markov network model;

[0011] The operating status data of the pole-mounted circuit breaker is monitored in real time, and the operating status data is input into the target Markov network model to perform operating fault prediction to obtain a real-time fault prediction result.

[0012] In a second aspect, the present application provides a pole-mounted circuit breaker testing device, the pole-mounted circuit breaker testing device comprising:

[0013] An acquisition module, configured to acquire geometric dimensions, material properties, and electrical parameters of a pole-mounted circuit breaker and construct a coupled three-dimensional simulation model of the pole-mounted circuit breaker;

[0014] a simulation module for defining electric field distribution and magnetic field distribution under different operating conditions, and performing finite element simulation analysis on the insulation resistance and fault location in the coupled three-dimensional simulation model to obtain power loss simulation data and insulation resistance simulation data;

[0015] a calculation module, configured to perform change calculations on the power loss simulation data and the insulation resistance simulation data, respectively, to obtain power loss change data and insulation resistance change data, and to perform change relationship analysis to obtain a relationship coefficient between the power loss change and the insulation resistance change;

[0016] a screening module, configured to perform feature fusion and feature screening on the power loss change data and the insulation resistance change data based on the relationship coefficient to obtain fused change feature data, and use the fused change feature data as a feature node of a Markov network to obtain an initial Markov network model;

[0017] An optimization module is used to perform multiple iterations and model optimization on the initial Markov network model through a genetic algorithm to obtain a target Markov network model;

[0018] The prediction module is used to monitor the operating status data of the pole-mounted circuit breaker in real time, and input the operating status data into the target Markov network model to perform operating fault prediction to obtain a real-time fault prediction result.

[0019] A third aspect of the present application provides a pole-mounted circuit breaker testing device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory to enable the pole-mounted circuit breaker testing device to execute the above-mentioned pole-mounted circuit breaker testing method.

[0020] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned pole-mounted circuit breaker testing method.

[0021] The technical solution provided in this application constructs a coupled three-dimensional simulation model by acquiring the geometric dimensions, material properties, and electrical parameters of a pole-mounted circuit breaker, enabling accurate modeling of the circuit breaker. Finite element simulation technology is used to conduct detailed analysis of the electric and magnetic field distributions under different operating conditions, accurately simulating the electromagnetic behavior of the circuit breaker during actual operation. By calculating and analyzing the relationship between power loss and insulation resistance simulation data, the relationship coefficient between power loss and insulation resistance changes is obtained, improving the efficiency and accuracy of data analysis. Time series decomposition and differential calculation methods are used to fine-tune data processing across multiple time scales, ensuring the accuracy and reliability of the analysis results. Based on a Markov network model and genetic algorithm optimization, real-time monitoring of circuit breaker operating status data and fault prediction are achieved, improving the timeliness and accuracy of fault detection. Through feature fusion and feature screening, power loss and insulation resistance change data are effectively integrated, enhancing the model's predictive performance and robustness. This facilitates accurate judgment of fault type and severity, locates faulty components, and provides comprehensive fault information. The fault severity and fault component location information are integrated to generate fault alarm information and send it to the monitoring center to ensure the timeliness and effectiveness of fault handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This is a schematic diagram of an embodiment of a pole-mounted circuit breaker testing method in an embodiment of the present application;

[0024] Figure 2 Schematic diagram of an embodiment of a pole-mounted circuit breaker testing device in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application provide a pole-mounted circuit breaker testing device and testing method. The terms "first," "second," "third," "fourth," and so on (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0026] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In one embodiment of the present application, a method for testing a pole-mounted circuit breaker includes:

[0027] Step S101: obtaining geometric dimensions, material properties, and electrical parameters of a pole-mounted circuit breaker and constructing a coupled three-dimensional simulation model of the pole-mounted circuit breaker;

[0028] It is understandable that the execution subject of the present application can be a pole-mounted circuit breaker test device, or a terminal or a server, which is not limited here. The embodiment of the present application is described by taking the server as the execution subject as an example.

[0029] Specifically, a 3D scan of the pole-mounted circuit breaker is performed to obtain its geometric dimensions. Based on the geometric dimensions, a 3D internal structure model of the pole-mounted circuit breaker is constructed to obtain a 3D internal structure model. This 3D internal structure model accurately reflects the internal structure of the pole-mounted circuit breaker, including the locations and connections of its components. The material properties of the pole-mounted circuit breaker are obtained and input into the 3D internal structure model to obtain a 3D material property model. Material properties include electrical conductivity, magnetic permeability, and dielectric constant, which are crucial for simulating the distribution of electric and magnetic fields. Inputting the material property parameters into the 3D internal structure model yields a more accurate 3D material property model. The electrical parameters of the pole-mounted circuit breaker are obtained and, based on the electrical parameters, electric and magnetic field boundary conditions are set for the 3D material property model to obtain a 3D electromagnetic boundary model. Electrical parameters include voltage, current, and frequency, which determine the electromagnetic characteristics of the pole-mounted circuit breaker during operation. By setting electric and magnetic field boundary conditions on the 3D material property model, a 3D electromagnetic boundary model that accurately reflects the operating state of the pole-mounted circuit breaker is constructed. The 3D electromagnetic boundary model is meshed to obtain a finite element mesh model. This divides the complex 3D model into many small elements for numerical calculations. By meshing the 3D electromagnetic boundary model, a finite element mesh model suitable for numerical solutions is obtained. Electrical parameters are input into the finite element mesh model to establish the electromagnetic governing equations. These governing equations are fundamental equations that describe the distribution of electric and magnetic fields. These equations are used to simulate the electromagnetic characteristics of the pole-mounted circuit breaker under different operating conditions. Electrical parameters are input into the finite element mesh model to establish governing equations that reflect the electromagnetic characteristics of the pole-mounted circuit breaker. Solving the electromagnetic governing equations yields the first electric and magnetic field distribution data for the pole-mounted circuit breaker in steady state. Solving the electromagnetic governing equations uses numerical calculations to determine the electric and magnetic field distributions of the pole-mounted circuit breaker under specific operating conditions. These distributions reflect the steady-state electromagnetic characteristics of the pole-mounted circuit breaker. The first electric and magnetic field distribution data are coupled with the finite element mesh model to obtain a coupled 3D simulation model.

[0030] Step S102: defining electric field distribution and magnetic field distribution under different operating conditions, and performing finite element simulation analysis on the insulation resistance and fault location in the coupled three-dimensional simulation model to obtain power loss simulation data and insulation resistance simulation data;

[0031] Specifically, the rated voltage and rated current of the pole-mounted circuit breaker are obtained, and the electric field strength and magnetic induction strength thresholds under normal operating conditions are defined, obtaining the normal operating thresholds. In a coupled three-dimensional simulation model, the normal operating thresholds are used as simulation boundary conditions to perform electric and magnetic field simulations on the pole-mounted circuit breaker, obtaining the second electric and magnetic field distribution data under normal operating conditions. Through simulation analysis, a model of the electromagnetic characteristics of the pole-mounted circuit breaker under normal operating conditions is established. The insulation level of the pole-mounted circuit breaker is obtained, and multiple insulation resistance values ​​of different levels are defined to form an insulation resistance value set. In the coupled three-dimensional simulation model, each insulation resistance value in the insulation resistance value set is sequentially assigned to each insulation component of the pole-mounted circuit breaker, obtaining simulation models under multiple different insulation levels. By varying the resistance values ​​of the insulation components, the electromagnetic behavior of the pole-mounted circuit breaker under different insulation conditions is simulated. Electric and magnetic field simulations are performed on the simulation model, obtaining the third electric and magnetic field distribution data under different insulation levels. These data reflect the impact of varying insulation levels on the electromagnetic characteristics of the pole-mounted circuit breaker. The structural characteristics of the pole-mounted circuit breaker are obtained, and multiple possible fault locations are defined to form a fault location set. In a coupled 3D simulation model, the material parameters of each fault location in the fault location set are modified, and the conductivity is set to infinity, resulting in simulation models of multiple different fault points. By simulating the conductive characteristics of different fault locations, a model of the electromagnetic characteristics of the pole-mounted circuit breaker during a fault is established. Electric and magnetic field simulations are performed on the simulation model to obtain fourth electric and magnetic field distribution data at different fault locations. These data reflect the impact of the fault location on the electromagnetic characteristics of the pole-mounted circuit breaker. Power calculations are performed on the second, third, and fourth electric and magnetic field distribution data to generate power loss datasets under different operating conditions. The power calculations, based on the electric and magnetic field distribution data, evaluate the power loss of the pole-mounted circuit breaker under different operating conditions. The insulation resistance values ​​are matched to the power loss dataset to obtain power loss simulation data and insulation resistance simulation data. By matching the insulation resistance values ​​with the power loss data, the power loss characteristics under different insulation states and fault conditions are analyzed.

[0032] Step S103: performing change calculations on the power loss simulation data and the insulation resistance simulation data respectively to obtain power loss change data and insulation resistance change data, and performing change relationship analysis to obtain a relationship coefficient between the power loss change and the insulation resistance change;

[0033] Specifically, the power loss simulation data is decomposed into time series to obtain power loss component data at different time scales. Time series decomposition helps to break down the power loss data into short-term, medium-term, and long-term variation components, allowing for a clearer analysis of its variation patterns. The power loss component data is differentially calculated to obtain power loss change rate data at each time scale. Differential calculations highlight the changing trends in the time series data, making the change rate data easier to analyze and compare. The insulation resistance simulation data is decomposed into time series to obtain insulation resistance component data at different time scales. Differential calculations are performed on the insulation resistance component data to obtain insulation resistance change rate data at each time scale. The power loss change rate data and the insulation resistance change rate data are matched to construct an initial scatter plot of the power loss change rate and the insulation resistance change rate. The initial scatter plot is normalized to obtain a normalized scatter plot. Data normalization eliminates the influence of different dimensions, making the data more comparable. Curve fitting is performed on the normalized scatter plot to obtain the fitting curve equation for the power loss change rate and the insulation resistance change rate. Curve fitting can reveal the functional relationship between the two. After obtaining the fitted curve equation, the first-order derivative of the fitted curve equation is taken to determine the response coefficient of the power loss change rate to the insulation resistance change rate. This derivative process reveals the sensitivity between the change rates, known as the response coefficient. Using the response coefficient as the independent variable, a weighted average of the response coefficients at multiple time scales is performed to determine the relationship coefficient between the power loss change and the insulation resistance change. This weighted average comprehensively accounts for the influence of different time scales, making the relationship coefficient more representative and accurate.

[0034] Step S104: performing feature fusion and feature screening on the power loss change data and the insulation resistance change data based on the relationship coefficient to obtain fused change feature data, and using the fused change feature data as feature nodes of the Markov network to obtain an initial Markov network model;

[0035] Specifically, based on the correlation coefficient, a correlation analysis is performed on the power loss change data and the insulation resistance change data to obtain a correlation matrix. By calculating the correlations between the individual data points, the intrinsic connection between the power loss change and the insulation resistance change is revealed. The correlation matrix is ​​subjected to singular value decomposition to obtain a left singular matrix, a singular value vector, and a right singular matrix. The singular value vector is then subjected to dimensionality reduction to obtain a reduced singular value vector. Singular value decomposition is a matrix decomposition method that reduces data complexity while preserving its key features. Matrix multiplication is performed on the left singular matrix and the reduced singular value vector to obtain a first fused feature matrix for power loss change. Matrix multiplication is performed on the reduced singular value vector and the right singular matrix to obtain a second fused feature matrix for insulation resistance change. The first and second fused feature matrices are concatenated to obtain a target feature matrix. The concatenation operation horizontally concatenates the two matrices to form a new matrix, allowing the power loss change and insulation resistance change features to be presented in the same matrix. Feature screening is performed on the target feature matrix to obtain fused change feature data. By selecting features with important information and removing redundant or noisy features, the efficiency and accuracy of data analysis are improved. The topology of the Markov network is initialized, and the fused change feature data is mapped to the corresponding feature nodes of the Markov network. A Markov network is a probabilistic graphical model that represents variables and their dependencies through nodes and edges. The fused change feature data is used as the feature nodes of the Markov network to construct an initial model representing the relationship between power loss changes and insulation resistance changes. Parameters of the transition probability matrix and emission probability matrix in the Markov network are estimated to obtain the initial Markov network model. Parameter estimation determines the transition probabilities between nodes and the emission probabilities of nodes in the Markov network by learning from historical data, allowing the network to accurately represent the statistical characteristics of the data.

[0036] Step S105: performing multiple iterations and model optimization on the initial Markov network model using a genetic algorithm to obtain a target Markov network model;

[0037] Specifically, the transition probability matrix and the emission probability matrix are encoded to obtain chromosome coding sequences. This encoding process represents the parameters of the Markov network in a form that can be processed by a genetic algorithm, allowing these parameters to be optimized through genetic operations. Based on a preset population size, multiple chromosome coding sequences are randomly generated to form an initial population. The multiple chromosome coding sequences in the initial population are then decoded to obtain multiple corresponding candidate Markov network models. Using the fused variation feature data as a test sample, the likelihood of each candidate Markov network model is calculated and used as the individual's first fitness. The likelihood measures the model's ability to explain the data; higher fitness indicates a more accurate description of the data characteristics. Based on the first fitness, a roulette wheel selection algorithm is used to select several individuals from the initial population to obtain parent individuals. The roulette wheel selection algorithm randomly selects individuals with higher fitness as parents based on their fitness to increase the probability of transmitting superior genes. A crossover operation is performed on the parent individuals to obtain offspring individuals, and a mutation operation is performed on the offspring individuals to obtain mutant individuals. The crossover operation generates new offspring individuals by exchanging partial gene sequences of parent individuals. The mutation operation randomly changes some of an individual's genes to increase population diversity and prevent premature convergence to a local optimum. The mutated individuals are added to the initial population, the second fitness of the individuals in the population is updated, and the initial population is sorted according to the second fitness to obtain the current population. This process recalculates fitness and sorts individuals to select individuals with higher fitness to form a new population. The optimal individual in the current population is determined to meet the preset convergence criteria. If so, it is decoded into the target Markov network model. The convergence criteria are usually a set fitness threshold or number of iterations. When the optimal individual in the population reaches or exceeds this threshold, the genetic algorithm terminates and outputs the optimized target model.

[0038] Step S106: monitor the operating status data of the pole-mounted circuit breaker in real time, and input the operating status data into the target Markov network model to perform operating fault prediction to obtain a real-time fault prediction result.

[0039] Specifically, the operating status data of the pole-mounted circuit breaker is monitored in real time and preprocessed to obtain real-time status data. This preprocessing step includes data cleaning, denoising, and normalization. Feature extraction is performed on the real-time status data based on the input nodes of the target Markov network model to obtain operating feature data. The feature extraction process analyzes the real-time status data to extract key features that reflect the operating status of the pole-mounted circuit breaker, such as current, voltage, and temperature. This feature data is passed as input to the target Markov network model. The operating feature data is input into the target Markov network model, and the probability distribution of each hidden state node is calculated using a forward algorithm to obtain a hidden state probability distribution. The forward algorithm, a common algorithm in Markov networks, recursively calculates the hidden state probability at each time step to provide a probability estimate of the current system state. Based on a preset fault state threshold, a fault diagnosis is performed on the hidden state probability distribution to obtain a real-time fault prediction result, including the fault type and severity. The fault diagnosis process compares the hidden state probability with the preset threshold to determine the presence of a fault and its type and severity. The results of this step can help promptly identify potential problems and prevent further equipment damage. Based on the fault type and the emission probability matrix of the target Markov network model, the Viterbi algorithm is used to decode the fault location and obtain the location information of the faulty component. The Viterbi algorithm is a dynamic programming algorithm that uses an optimal path search to find the most likely fault location sequence, thereby accurately locating the faulty component. The fault severity and faulty component location information are combined to generate a fault alarm message, which is sent to the pole-mounted circuit breaker monitoring center to prompt operation and maintenance personnel to address the fault. The fault alarm information includes detailed information about the fault type, severity, and specific location.

[0040] In the embodiments of the present application, a coupled three-dimensional simulation model is constructed by obtaining the geometric dimensions, material properties, and electrical parameters of a pole-mounted circuit breaker, enabling accurate modeling of the circuit breaker. Finite element simulation technology is used to conduct a detailed analysis of the electric and magnetic field distributions under different operating conditions, accurately simulating the electromagnetic behavior of the circuit breaker during actual operation. By calculating and analyzing the relationship between power loss simulation data and insulation resistance simulation data, the relationship coefficient between power loss change and insulation resistance change is obtained, improving the efficiency and accuracy of data analysis. Time series decomposition and differential calculation methods are used to fine-tune data processing of changes at multiple time scales, ensuring the accuracy and reliability of the analysis results. Based on a Markov network model and genetic algorithm optimization, real-time monitoring of circuit breaker operating status data and fault prediction are achieved, improving the timeliness and accuracy of fault detection. Through feature fusion and feature screening, power loss change data and insulation resistance change data are effectively integrated, enhancing the model's predictive performance and robustness. This facilitates accurate judgment of fault type and severity, locates faulty components, and provides comprehensive fault information. The fault severity and fault component location information are integrated to generate fault alarm information and send it to the monitoring center to ensure the timeliness and effectiveness of fault handling.

[0041] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0042] (1) Performing a three-dimensional scan on the pole-mounted circuit breaker to obtain the geometric dimensions of the pole-mounted circuit breaker, and performing a three-dimensional modeling of the internal structure of the pole-mounted circuit breaker based on the geometric dimensions to obtain a three-dimensional model of the internal structure;

[0043] (2) Obtaining the material properties of the pole-mounted circuit breaker and inputting the material properties into the internal structure three-dimensional model to obtain a material property three-dimensional model;

[0044] (3) Obtain the electrical parameters of the pole-mounted circuit breaker, and set the electric field and magnetic field boundary conditions for the three-dimensional model of material characteristics based on the electrical parameters to obtain the three-dimensional electromagnetic boundary model;

[0045] (4) Meshing the electromagnetic boundary three-dimensional model to obtain a finite element mesh model, and inputting electrical parameters into the finite element mesh model to establish the electromagnetic control equations;

[0046] (5) Solve the electromagnetic control equations to obtain the first electric field distribution and magnetic field distribution data of the pole-mounted circuit breaker in steady state;

[0047] (6) The first electric field distribution and magnetic field distribution data are coupled with the finite element mesh model to obtain a coupled three-dimensional simulation model.

[0048] Specifically, a 3D scan of the pole-mounted circuit breaker is performed to obtain its external geometric dimensions. 3D scanning technology uses laser scanning or structured light scanning to accurately capture the shape and dimensions of the pole-mounted circuit breaker, generating point cloud data. This point cloud data is then processed using reverse engineering software such as Geomagic or PolyWorks to extract the circuit breaker's geometric dimensions and construct a 3D CAD model. This model details the circuit breaker's external shape. Based on the acquired geometric dimensions, a 3D model of the pole-mounted circuit breaker's internal structure is constructed. This internal structure modeling is typically performed using CAD software such as SolidWorks, CATIA, or AutoCAD. Each component of the circuit breaker is accurately modeled according to its actual structure. This includes the positions and dimensions of various components, such as switch contacts, insulation material, and conductive rods, to form a 3D internal structure model. After constructing the 3D internal structure model, the material properties of the pole-mounted circuit breaker are determined. These properties include electrical conductivity, magnetic permeability, and dielectric constant, depending on the material used to manufacture the circuit breaker. These parameters can be obtained from material databases or experimental measurements. These material properties are then input into the 3D internal structure model to produce a 3D material property model. Obtain the electrical parameters of the pole-mounted circuit breaker. These parameters include rated voltage, rated current, and operating frequency. Based on these electrical parameters, set electric and magnetic field boundary conditions for the 3D material property model to obtain a 3D electromagnetic boundary model. These boundary conditions are based on electromagnetic field theory and simulate the electrical environment of the circuit breaker under actual operating conditions. Electric and magnetic field boundary conditions, such as the Dirichlet and Neumann conditions, can be described by the following formulas:

[0049] ;

[0050] ;

[0051] in, is the electric field strength, is the electric potential, is the electric displacement vector, is the charge density. The electromagnetic boundary three-dimensional model is meshed to obtain a finite element mesh model. Meshing is the process of decomposing a complex three-dimensional model into a finite number of simple units, such as tetrahedrons or hexahedrons. The generated mesh model makes it possible to numerically solve the electromagnetic field. After obtaining the finite element mesh model, the electrical parameters are input into the model to establish the electromagnetic control equations. The electromagnetic control equations are based on Maxwell's equations and describe the distribution of the electromagnetic field in the circuit breaker. The following are two basic equations in Maxwell's equations:

[0052] ;

[0053] ;

[0054] in, is the magnetic field strength, is the magnetic induction intensity, is the current density, is the electric displacement vector. By solving the electromagnetic control equations, the first electric field distribution and magnetic field distribution data of the pole-mounted circuit breaker in steady state are obtained. The solution process usually uses numerical methods, such as the finite element method (FEM) or the finite difference method (FDM), to simulate on a computer to obtain the electric field and magnetic field distribution of the circuit breaker under various electrical conditions. These distribution data provide detailed electromagnetic characteristics of the circuit breaker under operation. The first electric field distribution and magnetic field distribution data are coupled with the finite element mesh model to obtain a coupled three-dimensional simulation model. The coupling process is to combine the solved electromagnetic field distribution data with the original mesh model to form a comprehensive three-dimensional simulation model. For example, a three-dimensional scan of a certain model of pole-mounted circuit breaker shows that its external dimensions are 50 cm long, 30 cm wide, and 25 cm high. The internal structure includes copper contacts, ceramic insulators, and steel casing. Material properties such as the electrical conductivity of copper are S / m, the dielectric constant of ceramic is 6, and the magnetic permeability of steel is The electrical parameters were set to a rated voltage of 10 kV, a rated current of 100 A, and an operating frequency of 50 Hz. In the simulation, the electric field boundary conditions were defined as a contact voltage of 10 kV and an insulator boundary potential of 0. A finite element mesh model with approximately 1 million cells was generated. Solving the Maxwell equations yielded a maximum electric field strength of 5 kV / cm and a maximum magnetic induction strength of 0.2 T. By coupling these data, an accurate three-dimensional simulation model was formed.

[0055] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0056] (1) Obtain the rated voltage and rated current of the pole-mounted circuit breaker, define the electric field strength threshold and magnetic induction strength threshold under normal operating conditions of the pole-mounted circuit breaker, and obtain the normal operating condition threshold;

[0057] (2) In the coupled three-dimensional simulation model, the normal operating threshold is used as the simulation boundary condition to perform electric field simulation and magnetic field simulation on the pole-mounted circuit breaker to obtain the second electric field distribution and magnetic field distribution data under normal operating conditions;

[0058] (3) Obtain the insulation level of the pole-mounted circuit breaker, define multiple insulation resistance values ​​of different levels, obtain an insulation resistance value set, and in the coupled three-dimensional simulation model, assign each insulation resistance value in the insulation resistance value set to each insulation part of the pole-mounted circuit breaker in turn to obtain simulation models under multiple different insulation levels;

[0059] (4) Perform electric field simulation and magnetic field simulation on simulation models at multiple different insulation levels to obtain third electric field distribution and magnetic field distribution data at different insulation levels;

[0060] (5) Obtain the structural characteristics of the pole-mounted circuit breaker, define multiple possible fault locations, obtain a fault location set, and modify the material parameters of each fault location in the fault location set in the coupled three-dimensional simulation model, set the conductivity to infinity, and obtain simulation models of multiple different fault points;

[0061] (6) Perform electric field simulation and magnetic field simulation on the simulation models of multiple different fault points to obtain the fourth electric field distribution and magnetic field distribution data at different fault points;

[0062] (7) performing power calculation on the second electric field distribution and magnetic field distribution data, the third electric field distribution and magnetic field distribution data, and the fourth electric field distribution and magnetic field distribution data to obtain power loss data sets under different working conditions;

[0063] (8) Match the insulation resistance value with the power loss data set to obtain power loss simulation data and insulation resistance simulation data.

[0064] Specifically, the rated voltage and rated current of the pole-mounted circuit breaker are obtained. These electrical parameters form the basis for simulation analysis. For example, assume the rated voltage of the pole-mounted circuit breaker is 12 kV and the rated current is 200 A. Based on these parameters, the electric field strength threshold and magnetic induction strength threshold are defined under the pole-mounted circuit breaker's normal operating conditions. These thresholds ensure the rationality of the simulation results. The electric field strength threshold can be defined as 20 kV / m, and the magnetic induction strength threshold can be defined as 0.5 T / m. These thresholds are used to determine the normal operating condition thresholds, which are used to set boundary conditions during the simulation. In the coupled 3D simulation model, the normal operating condition thresholds are used as simulation boundary conditions to perform electric and magnetic field simulations of the pole-mounted circuit breaker. Using finite element analysis software such as ANSYS or COMSOL, the electric and magnetic fields of the pole-mounted circuit breaker under these boundary conditions are simulated and calculated to obtain the secondary electric and magnetic field distribution data under normal operating conditions. For example, in the simulation, the electric field strength may reach 18 kV / m in some areas, while the magnetic induction strength may reach 0.4 T in some areas. Obtain the insulation level of the pole-mounted circuit breaker, including the resistance of the insulation material. Define multiple insulation resistance values ​​of different levels to obtain an insulation resistance value set. For example, insulation resistance can be categorized into high, medium, and low levels: 100 MΩ, 50 MΩ, and 10 MΩ, respectively. In a coupled 3D simulation model, assign each insulation resistance value from these insulation resistance value sets to each insulation location of the pole-mounted circuit breaker, generating simulation models at different insulation levels. Perform electric and magnetic field simulations on the simulation models at different insulation levels to obtain third-party electric and magnetic field distribution data for each insulation level. For example, in the case of high insulation resistance, the electric field strength may be uniformly distributed, while in the case of low insulation resistance, the electric field strength may be concentrated in certain areas, forming hot spots. Obtain the structural characteristics of the pole-mounted circuit breaker and define multiple possible fault locations to obtain a fault location set. For example, fault locations may include contact points, connections, and cable ends. In the coupled three-dimensional simulation model, the material parameters of each fault location in the fault location set are modified, and the conductivity is set to infinity to obtain simulation models of multiple different fault points, thereby simulating the electromagnetic characteristics under the fault state. Electric field simulation and magnetic field simulation are performed on the simulation models of these different fault points to obtain the fourth electric field distribution and magnetic field distribution data at different fault points. For example, at a certain fault point, the electric field strength may increase significantly, and the magnetic induction intensity may also change significantly. These data help identify and analyze the fault location and type. Power calculation is performed on the second electric field distribution and magnetic field distribution data, the third electric field distribution and magnetic field distribution data, and the fourth electric field distribution and magnetic field distribution data to obtain power loss data sets under different working conditions. The calculation of power loss can be achieved using the following formula:

[0065] ;

[0066] in, is the power loss, is the current density, is the electric field strength, is the integration region. By integrating the electric field and current density data under various operating conditions, the power loss in each case is obtained. For example, the power loss under normal operating conditions may be 10W, while under a certain fault condition, the power loss may increase to 50W. The insulation resistance values ​​are matched with the power loss dataset to obtain power loss simulation data and insulation resistance simulation data. Through matching, the changes in power loss under different insulation levels and fault conditions are analyzed, and the performance indicators of the pole-mounted circuit breaker under different operating conditions are derived. For example, the matching results show that high insulation resistance corresponds to lower power loss, while low insulation resistance or fault conditions result in significantly increased power loss.

[0067] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0068] (1) Decompose the power loss simulation data into time series to obtain the power loss component data at different time scales, and perform differential calculation on the power loss component data to obtain the power loss change rate data at each time scale;

[0069] (2) Decompose the insulation resistance simulation data into time series to obtain the insulation resistance component data at different time scales, and perform differential calculation on the insulation resistance component data to obtain the insulation resistance change rate data at each time scale;

[0070] (3) Match the power loss change rate data and the insulation resistance change rate data to construct an initial scatter plot of the power loss change rate and the insulation resistance change rate;

[0071] (4) Performing data normalization on the initial scatter plot to obtain a normalized scatter plot, and performing curve fitting on the normalized scatter plot to obtain the fitting curve equation of the power loss change rate and the insulation resistance change rate;

[0072] (5) The first-order derivative of the fitting curve equation is calculated to obtain the response coefficient of the power loss change rate to the insulation resistance change rate. The response coefficient is used as the independent variable, and the weighted average of the response coefficients at multiple time scales is performed to obtain the relationship coefficient between the power loss change and the insulation resistance change.

[0073] Specifically, the power loss simulation data is decomposed into a time series to obtain power loss component data at different time scales. Methods such as wavelet transform or empirical mode decomposition are used. Wavelet transform is an effective time series decomposition method that can decompose the original signal into different frequency components, thereby revealing the characteristics of the signal at different time scales. For example, through wavelet transform, the power loss data is decomposed into components at multiple scales, each component corresponding to a specific frequency range. Assume that three main time scale components are decomposed: high-frequency component (second-level changes), medium-frequency component (minute-level changes) and low-frequency component (hour-level changes). The power loss component data is differentially calculated to obtain the power loss change rate data at each time scale. The differential calculation can be achieved by the following formula:

[0074] ;

[0075] in, represents the rate of change of power loss at the i-th moment, and Represent the power loss values ​​at the i+1th and ith moments respectively. In this way, the power loss change rate data at each time scale is obtained. Similarly, the insulation resistance simulation data is decomposed into time series to obtain the insulation resistance component data at different time scales. Assume that the insulation resistance data is decomposed using wavelet transform to obtain three time scale components: high frequency, medium frequency and low frequency. Perform differential calculation on the insulation resistance component data to obtain the insulation resistance change rate data at each time scale. Match the power loss change rate data and the insulation resistance change rate data to construct an initial scatter plot of the power loss change rate and the insulation resistance change rate. Pair the power loss change rate and the insulation resistance change rate at the same time scale and plot them on the scatter plot. Perform data normalization on the initial scatter plot to eliminate the influence of different dimensions and ranges to make the data more comparable. Perform curve fitting on the normalized scatter plot to obtain the fitting curve equation of the power loss change rate and the insulation resistance change rate. Common curve fitting methods include linear regression, polynomial regression, etc. For example, if a quadratic polynomial is selected for fitting, the fitting curve equation can be expressed as:

[0076] ;

[0077] in, represents the rate of change of power loss, Indicates the rate of change of insulation resistance, 、 and are the fitting parameters. Using the least squares method, we can determine the values ​​of these parameters so that the fitting curve is as close as possible to the scattered data. Taking the first-order derivative of the fitting curve equation yields the response coefficient of the power loss change rate to the insulation resistance change rate. This derivative can be achieved using the following formula:

[0078] ;

[0079] in, The response coefficient represents the sensitivity of the rate of change of power loss to the rate of change of insulation resistance. The weighted average of the response coefficients at multiple time scales is used as the independent variable to obtain the coefficient of relationship between power loss change and insulation resistance change.

[0080] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0081] (1) Based on the relationship coefficient, the power loss change data and the insulation resistance change data are subjected to correlation analysis to obtain the correlation matrix;

[0082] (2) Perform singular value decomposition on the correlation matrix to obtain the left singular matrix, singular value vector and right singular matrix, and perform dimensionality reduction on the singular value vector to obtain the reduced-dimensional singular value vector;

[0083] (3) Performing matrix multiplication on the left singular matrix and the dimensionality-reduced singular value vector to obtain the first fused characteristic matrix of power loss change, and performing matrix multiplication on the dimensionality-reduced singular value vector and the right singular matrix to obtain the second fused characteristic matrix of insulation resistance change;

[0084] (4) performing a juxtaposition operation on the first fusion feature matrix and the second fusion feature matrix to obtain a target feature matrix, and performing feature screening on the target feature matrix to obtain fusion change feature data;

[0085] (5) Initialize the topological structure of the Markov network, map the fused change feature data to the corresponding feature nodes of the Markov network, and perform parameter estimation on the transition probability matrix and emission probability matrix in the Markov network to obtain the initial Markov network model.

[0086] Specifically, based on the correlation coefficient, a correlation analysis is performed on the power loss change data and the insulation resistance change data. The Pearson correlation coefficient is used to calculate the correlation between each pair of variables. The formula for the Pearson correlation coefficient is:

[0087] ;

[0088] in, Representing variables and The correlation coefficient between express and The covariance of and They are and The correlation matrix is ​​obtained by calculating , where each element Indicates the The power loss change data of the The correlation between the insulation resistance change data. Perform singular value decomposition to obtain the left singular matrix , singular value vector and right singular matrices The formula for singular value decomposition is:

[0089] ;

[0090] in, and is an orthogonal matrix, Is a diagonal matrix containing singular values. Perform dimensionality reduction, retain the main singular values, and discard the smaller singular values. For example, choose to retain the previous Singular values, which can usually capture the main features of the data. and the singular value vector after dimensionality reduction Perform matrix multiplication to obtain the first fused feature matrix of power loss change

[0091] ;

[0092] Similarly, the singular value vector after dimensionality reduction is and right singular matrices Perform matrix multiplication to obtain the second fused characteristic matrix of insulation resistance change

[0093] ;

[0094] For the first fusion feature matrix and the second fusion feature matrix Perform the concatenation operation to obtain the target feature matrix The juxtaposition operation is to horizontally splice the two matrices together to form a new matrix, so that the power loss variation characteristics and insulation resistance variation characteristics can be presented in the same matrix:

[0095] ;

[0096] The target feature matrix is ​​subjected to feature screening to obtain fused change feature data. After obtaining the fused change feature data, the topology of the Markov network is initialized. A Markov network is a probabilistic graphical model that represents variables and their dependencies through nodes and edges. The fused change feature data is mapped to the corresponding feature nodes of the Markov network. These nodes represent different states or observations of the system. Parameters of the transition probability matrix and emission probability matrix in the Markov network are estimated. The transition probability matrix describes the probability of the system transitioning from one state to another, while the emission probability matrix describes the probability of observing a certain feature in a certain state. Parameter estimation typically uses the maximum likelihood estimation method to estimate these probability values ​​using training data.

[0097] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0098] (1) Encode the transition probability matrix and the emission probability matrix to obtain the chromosome coding sequence;

[0099] (2) According to the preset population size, multiple chromosome coding sequences are randomly generated to form an initial population, and multiple chromosome coding sequences in the initial population are decoded to obtain corresponding multiple candidate Markov network models;

[0100] (3) Using the fused change feature data as the test sample, the likelihood probability of multiple candidate Markov network models is calculated and the likelihood probability is used as the first fitness of the individual;

[0101] (4) According to the first fitness, a roulette wheel selection algorithm is used to select several individuals from the initial population to obtain parent individuals, and a crossover operation is performed on the parent individuals to obtain offspring individuals, and a mutation operation is performed on the offspring individuals to obtain mutant individuals;

[0102] (5) Add the mutant individuals to the initial population, update the second fitness of the individuals in the population, and sort the initial population according to the second fitness to obtain the current population;

[0103] (6) Determine whether the optimal individual of the current population meets the preset convergence conditions. If so, decode the optimal individual into the target Markov network model.

[0104] Specifically, the transition probability matrix and the emission probability matrix are encoded. Assume that the transition probability matrix and the emission probability matrix The dimensions are and ,in is the number of hidden states, is the number of observed states. Each element of the matrix is ​​a real value in the range [0, 1], and these real values ​​can be discretized with a certain precision. For example, assuming two decimal places are used, each real value can be represented as a two-digit decimal. In this way, each element in the matrix is ​​converted into a gene, and the entire matrix is ​​converted into a chromosome sequence. According to the preset population size, multiple chromosome coding sequences are randomly generated to form the initial population. Assume that the population size is , you need to generate chromosome coding sequences, each sequence represents a candidate Markov network model. The chromosome coding sequence is decoded and restored to the corresponding transition probability matrix and emission probability matrix to obtain multiple candidate Markov network models. The fusion change feature data is used as the test sample to calculate the likelihood probability of multiple candidate Markov network models, and the likelihood probability is used as the first fitness of the individual. The calculation of the likelihood probability can be achieved through the forward algorithm. The forward algorithm is used to calculate the likelihood probability of the model under a given observation sequence. The formula is as follows:

[0105] ;

[0106] in, Indicated in the model Next observation sequence The likelihood probability, It's in time At this moment, the system is in state The forward probability of the candidate Markov network model is calculated. By calculating the likelihood probability of each candidate Markov network model, the first fitness of each individual is obtained. Based on the first fitness, a roulette wheel selection algorithm is used to select several individuals from the initial population to obtain parent individuals. The roulette wheel selection algorithm is a probability-based selection method, in which the selection probability is proportional to the individual's fitness. The fitness of each individual is summed, and the selection probability of each individual is defined as the proportion of its fitness to the total fitness. Under this selection mechanism, individuals with higher fitness have a greater probability of being selected. A crossover operation is performed on the selected parent individuals to generate offspring individuals. Crossover is a key step in genetic algorithms, producing new offspring individuals by exchanging partial gene segments of parent individuals. For example, single-point crossover or multi-point crossover can be used to exchange gene segments of parent individuals at random locations, thereby generating diverse offspring individuals. Mutation is performed on the offspring individuals to obtain mutant individuals. Mutation is another important step in genetic algorithms, increasing population diversity and preventing regression into local optima by randomly changing partial gene segments of individuals. Randomly select a gene from the offspring and randomly change its value, for example, by adding or subtracting a small random number. The mutated individual is added to the initial population, and the secondary fitness of the individuals in the population is updated. The likelihood probability of each individual is recalculated, the fitness value is updated, and the initial population is sorted according to fitness to obtain the current population. The purpose of sorting is to select the optimal individual based on fitness value, ensuring that the population gradually converges to the optimal solution. Determine whether the optimal individual in the current population meets the preset convergence condition. The convergence condition can be a set fitness threshold or the fitness of the population has not changed significantly over several generations. If the convergence condition is met, the optimal individual is decoded into the target Markov network model; if not, the cycle of selection, crossover, mutation, and fitness update continues until the convergence condition is met.

[0107] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0108] (1) Real-time monitoring of the operating status data of the pole-mounted circuit breaker and pre-processing of the operating status data to obtain real-time status data;

[0109] (2) According to the input nodes of the target Markov network model, the real-time state data is feature extracted to obtain the operation feature data;

[0110] (3) Input the running feature data into the target Markov network model, calculate the probability distribution of each hidden state node through the forward algorithm, and obtain the hidden state probability distribution;

[0111] (4) Based on the preset fault state threshold, the hidden state probability distribution is used to make fault judgments and obtain real-time fault prediction results, including fault type and fault severity;

[0112] (5) According to the fault type and the emission probability matrix of the target Markov network model, the Viterbi algorithm is used to decode the fault location and obtain the location information of the faulty component;

[0113] (6) The fault severity and the location information of the faulty component are integrated to generate fault alarm information and send it to the monitoring center of the pole-mounted circuit breaker to prompt the operation and maintenance personnel to handle the fault.

[0114] Specifically, the operating status data of the pole-mounted circuit breaker is monitored in real time. These data include voltage, current, temperature and other related parameters. The operating status data is preprocessed to obtain real-time status data. Preprocessing includes operations such as data cleaning, denoising and normalization. Data cleaning aims to remove obvious outliers, denoising can use filters to smooth the data, and normalization adjusts the data to the same scale range for subsequent processing. According to the input node requirements of the target Markov network model, feature extraction is performed on the real-time status data. Key features that can reflect the operating status of the circuit breaker are extracted from the preprocessed data. For example, frequency domain features are extracted through Fourier transform, or time domain features such as mean and standard deviation are calculated through sliding windows. These operating feature data will be passed as input to the target Markov network model. After the operating feature data is input into the target Markov network model, the probability distribution of each hidden state node is calculated through the forward algorithm. The forward algorithm is a dynamic programming method used to calculate the probability distribution of the observation sequence. The formula is as follows:

[0115] ;

[0116] in, Indicates that at time t, the state is The forward probability of is the observation sequence, is a parameter of the Markov network model. By recursively calculating the forward probability, the probability distribution of each hidden state node at different time points is obtained, that is, the hidden state probability distribution. According to the preset fault state threshold, the hidden state probability distribution is judged as a fault. Set a threshold. If the probability of a hidden state exceeds the threshold, the system is judged to be in a fault state. The result of the fault judgment includes not only whether a fault occurs, but also the type of fault and the severity of the fault. For example, assuming that the threshold is set to 0.8, if the probability of a hidden state exceeds 0.8, it is judged as a serious fault; if it is between 0.5 and 0.8, it is a general fault. According to the fault type and the emission probability matrix of the target Markov network model, the Viterbi algorithm is used to decode the fault location. The Viterbi algorithm is a dynamic programming algorithm that can find the most likely state sequence for a given observation sequence. Its formula is as follows:

[0117] ;

[0118] in, Indicates that the state at time t is The probability of the maximum-probability path is calculated, and a backtracking algorithm is used to determine the most likely location of the faulty component. The fault severity and location information are combined to generate a fault alarm, which is then sent to the circuit breaker monitoring center. This alarm includes detailed information about the fault type, severity, and location, facilitating rapid response and resolution by maintenance personnel. For example, if the hidden state probability distribution and the Viterbi algorithm indicate an abnormal voltage at a contact point with a probability exceeding a threshold, a fault alarm is generated, indicating that the contact point may be experiencing poor contact or damage and requires repair.

[0119] The above describes the method for testing a pole-mounted circuit breaker in an embodiment of the present application. The following describes the apparatus for testing a pole-mounted circuit breaker in an embodiment of the present application. Figure 2 In one embodiment of the present application, a pole-mounted circuit breaker testing device includes:

[0120] An acquisition module 201 is used to acquire geometric dimensions, material properties, and electrical parameters of a pole-mounted circuit breaker and to construct a coupled three-dimensional simulation model of the pole-mounted circuit breaker;

[0121] The simulation module 202 is used to define the electric field distribution and magnetic field distribution under different operating conditions, and to change the insulation resistance and fault location in the coupled three-dimensional simulation model and perform finite element simulation analysis to obtain power loss simulation data and insulation resistance simulation data;

[0122] The calculation module 203 is used to perform change calculation on the power loss simulation data and the insulation resistance simulation data respectively, obtain power loss change data and insulation resistance change data, and perform change relationship analysis to obtain a relationship coefficient between the power loss change and the insulation resistance change;

[0123] A screening module 204 is configured to perform feature fusion and feature screening on the power loss change data and the insulation resistance change data based on the relationship coefficient to obtain fused change feature data, and use the fused change feature data as feature nodes of a Markov network to obtain an initial Markov network model;

[0124] The optimization module 205 is used to perform multiple iterations and model optimization on the initial Markov network model through a genetic algorithm to obtain a target Markov network model;

[0125] The prediction module 206 is used to monitor the operating status data of the pole-mounted circuit breaker in real time, and input the operating status data into the target Markov network model to perform operating fault prediction to obtain real-time fault prediction results.

[0126] Through the collaborative efforts of these components, a coupled 3D simulation model is constructed by acquiring the geometric dimensions, material properties, and electrical parameters of the pole-mounted circuit breaker, achieving accurate modeling of the circuit breaker. Finite element simulation technology is used to conduct detailed analysis of the electric and magnetic field distributions under different operating conditions, accurately simulating the electromagnetic behavior of the circuit breaker in actual operation. By calculating and analyzing the relationship between power loss and insulation resistance simulation data, the correlation coefficient between power loss and insulation resistance changes is obtained, improving the efficiency and accuracy of data analysis. Time series decomposition and differential calculation methods enable precise processing of data changes at multiple time scales, ensuring the accuracy and reliability of the analysis results. Based on a Markov network model and genetic algorithm optimization, real-time monitoring of circuit breaker operating status data and fault prediction are achieved, improving the timeliness and accuracy of fault detection. Through feature fusion and feature screening, power loss and insulation resistance change data are effectively integrated, enhancing the model's predictive performance and robustness. This facilitates accurate determination of fault type and severity, locates the faulty component, and provides comprehensive fault information. The fault severity and fault component location information are integrated to generate fault alarm information and send it to the monitoring center to ensure the timeliness and effectiveness of fault handling.

[0127] The present application also provides a pole-mounted circuit breaker testing device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the steps of the pole-mounted circuit breaker testing method in the above-mentioned embodiments.

[0128] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, which, when executed on a computer, cause the computer to execute the steps of the pole-mounted circuit breaker testing method.

[0129] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0130] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0131] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for testing a pole-mounted circuit breaker, characterized in that: The method comprises: Obtaining geometric dimensions, material properties and electrical parameters of a pole-mounted circuit breaker and constructing a coupled three-dimensional simulation model of the pole-mounted circuit breaker; Defining electric field distribution and magnetic field distribution under different operating conditions, and performing finite element simulation analysis on the insulation resistance and fault location in the coupled three-dimensional simulation model to obtain power loss simulation data and insulation resistance simulation data; Performing change calculations on the power loss simulation data and the insulation resistance simulation data respectively to obtain power loss change data and insulation resistance change data and performing change relationship analysis to obtain a relationship coefficient between power loss change and insulation resistance change; Based on the relationship coefficient, the power loss change data and the insulation resistance change data are subjected to feature fusion and feature screening to obtain fused change feature data, and the fused change feature data is used as the feature node of the Markov network to obtain an initial Markov network model; specifically comprising: according to the relationship coefficient, the power loss change data and the insulation resistance change data are subjected to correlation analysis to obtain a correlation matrix; the correlation matrix is ​​subjected to singular value decomposition to obtain a left singular matrix, a singular value vector and a right singular matrix, and the singular value vector is subjected to dimensionality reduction processing to obtain a singular value vector after dimensionality reduction; the left singular matrix and the singular value vector after dimensionality reduction are subjected to dimensionality reduction processing to obtain a singular value vector after dimensionality reduction; Perform matrix multiplication to obtain a first fused feature matrix of power loss change, and perform matrix multiplication on the dimensionality-reduced singular value vector and the right singular matrix to obtain a second fused feature matrix of insulation resistance change; perform juxtaposition operation on the first fused feature matrix and the second fused feature matrix to obtain a target feature matrix, and perform feature screening on the target feature matrix to obtain fused change feature data; initialize the topological structure of the Markov network, map the fused change feature data to the feature nodes corresponding to the Markov network, and perform parameter estimation on the transition probability matrix and the emission probability matrix in the Markov network to obtain an initial Markov network model; Performing multiple iterations and model optimization on the initial Markov network model through a genetic algorithm to obtain a target Markov network model; The operating status data of the pole-mounted circuit breaker is monitored in real time, and the operating status data is input into the target Markov network model for operating fault prediction to obtain a real-time fault prediction result.

2. The pole mounted circuit breaker testing method according to claim 1, characterized in that: The step of obtaining geometric dimensions, material properties and electrical parameters of the pole-mounted circuit breaker and constructing a coupled three-dimensional simulation model of the pole-mounted circuit breaker includes: Performing a three-dimensional scan on the pole-mounted circuit breaker to obtain geometric dimensions of the pole-mounted circuit breaker, and performing three-dimensional modeling of the internal structure of the pole-mounted circuit breaker according to the geometric dimensions to obtain a three-dimensional model of the internal structure; Acquiring material properties of the pole-mounted circuit breaker, and inputting the material properties into the internal structure three-dimensional model to obtain a material property three-dimensional model; Acquire electrical parameters of the pole-mounted circuit breaker, and set electric field and magnetic field boundary conditions for the three-dimensional model of material characteristics according to the electrical parameters to obtain a three-dimensional electromagnetic boundary model; Meshing the electromagnetic boundary three-dimensional model to obtain a finite element mesh model, and inputting the electrical parameters into the finite element mesh model to establish an electromagnetic control equation; Solving the electromagnetic control equation to obtain first electric field distribution and magnetic field distribution data of the pole-mounted circuit breaker in a steady state; The first electric field distribution and magnetic field distribution data are coupled with the finite element mesh model to obtain a coupled three-dimensional simulation model.

3. The method for testing a pole mounted circuit breaker according to claim 1, characterized in that: The electric field distribution and magnetic field distribution under different operating conditions are defined, and the insulation resistance and fault location in the coupled three-dimensional simulation model are changed and analyzed by finite element simulation to obtain power loss simulation data and insulation resistance simulation data, including: Obtaining the rated voltage and rated current of the pole-mounted circuit breaker, defining the electric field strength threshold and magnetic induction strength threshold under normal operating conditions of the pole-mounted circuit breaker, and obtaining the normal operating condition threshold; In the coupled three-dimensional simulation model, the normal operating condition threshold is used as a simulation boundary condition, and electric field simulation and magnetic field simulation are performed on the pole-mounted circuit breaker to obtain second electric field distribution and magnetic field distribution data under normal operating conditions; Acquire the insulation level of the pole-mounted circuit breaker, define multiple insulation resistance values ​​of different levels, obtain an insulation resistance value set, and in the coupled three-dimensional simulation model, sequentially assign each insulation resistance value in the insulation resistance value set to each insulation part of the pole-mounted circuit breaker to obtain simulation models under multiple different insulation levels; Performing electric field simulation and magnetic field simulation on the simulation models at the multiple different insulation levels to obtain third electric field distribution and magnetic field distribution data at the different insulation levels; Acquire structural characteristics of the pole-mounted circuit breaker, define multiple possible fault locations, obtain a fault location set, and modify material parameters of each fault location in the fault location set in the coupled three-dimensional simulation model, set the conductivity to infinity, and obtain simulation models of multiple different fault points; Performing electric field simulation and magnetic field simulation on the simulation models of the multiple different fault points to obtain fourth electric field distribution and magnetic field distribution data at different fault points; Performing power calculation on the second electric field distribution and magnetic field distribution data, the third electric field distribution and magnetic field distribution data, and the fourth electric field distribution and magnetic field distribution data to obtain power loss data sets under different working conditions; The insulation resistance value is matched with the power loss data set to obtain power loss simulation data and insulation resistance simulation data.

4. The method for testing a pole mounted circuit breaker according to claim 3, characterized in that: The respectively performing change calculation on the power loss simulation data and the insulation resistance simulation data to obtain power loss change data and insulation resistance change data and performing change relationship analysis to obtain the relationship coefficient between power loss change and insulation resistance change includes: Performing time series decomposition on the power loss simulation data to obtain power loss component data at different time scales, and performing differential calculation on the power loss component data to obtain power loss change rate data at each time scale; Performing time series decomposition on the insulation resistance simulation data to obtain insulation resistance component data at different time scales, and performing differential calculation on the insulation resistance component data to obtain insulation resistance change rate data at each time scale; Matching the power loss change rate data and the insulation resistance change rate data to construct an initial scatter plot of the power loss change rate and the insulation resistance change rate; Performing data normalization processing on the initial scatter plot to obtain a normalized scatter plot, and performing curve fitting on the normalized scatter plot to obtain a fitting curve equation of the power loss change rate and the insulation resistance change rate; The first-order derivative of the fitting curve equation is calculated to obtain the response coefficient of the power loss change rate to the insulation resistance change rate, and the response coefficient is used as the independent variable to perform weighted averaging on the response coefficients at multiple time scales to obtain the relationship coefficient between the power loss change and the insulation resistance change.

5. The method for testing a pole mounted circuit breaker according to claim 1, characterized in that: The initial Markov network model is iterated and optimized multiple times by a genetic algorithm to obtain a target Markov network model, including: Encoding the transfer probability matrix and the emission probability matrix to obtain a chromosome coding sequence; According to a preset population size, a plurality of chromosome coding sequences are randomly generated to form an initial population, and the plurality of chromosome coding sequences in the initial population are decoded to obtain a plurality of corresponding candidate Markov network models; Taking the fused change feature data as a test sample, calculating the likelihood probabilities of the multiple candidate Markov network models and using the likelihood probabilities as the first fitness of the individuals; According to the first fitness, a roulette wheel selection algorithm is used to select a number of individuals from the initial population to obtain parent individuals, and a crossover operation is performed on the parent individuals to obtain child individuals, and a mutation operation is performed on the child individuals to obtain mutant individuals; Adding the mutant individuals to the initial population, updating the second fitness of the individuals in the population, and sorting the initial population according to the second fitness to obtain the current population; It is determined whether the optimal individual of the current population meets a preset convergence condition, and if so, the optimal individual is decoded into a target Markov network model.

6. The method for testing a pole mounted circuit breaker according to claim 1, characterized in that: The real-time monitoring of the operating status data of the pole-mounted circuit breaker and inputting the operating status data into the target Markov network model for operating fault prediction to obtain a real-time fault prediction result include: real-time monitoring of the operating status data of the pole-mounted circuit breaker, and preprocessing of the operating status data to obtain real-time status data; According to the input nodes of the target Markov network model, feature extraction is performed on the real-time status data to obtain operation feature data; Inputting the operation characteristic data into the target Markov network model, calculating the probability distribution of each hidden state node through a forward algorithm, and obtaining the hidden state probability distribution; According to a preset fault state threshold, a fault judgment is performed on the hidden state probability distribution to obtain a real-time fault prediction result, including a fault type and a fault severity; According to the fault type and the emission probability matrix of the target Markov network model, a Viterbi algorithm is used to decode the fault position to obtain the fault component position information; The fault severity and the fault component location information are integrated to generate fault alarm information and send it to the monitoring center of the pole-mounted circuit breaker to prompt the operation and maintenance personnel to handle the fault.

7. A pole-mounted circuit breaker testing device, used to perform the pole-mounted circuit breaker testing method according to any one of claims 1 to 6, characterized in that: The device comprises: An acquisition module, used to acquire geometric dimensions, material properties and electrical parameters of the pole-mounted circuit breaker and construct a coupled three-dimensional simulation model of the pole-mounted circuit breaker; A simulation module, used to define electric field distribution and magnetic field distribution under different operating conditions, and to change and perform finite element simulation analysis on the insulation resistance and fault location in the coupled three-dimensional simulation model to obtain power loss simulation data and insulation resistance simulation data; A calculation module, used to perform change calculations on the power loss simulation data and the insulation resistance simulation data respectively, obtain power loss change data and insulation resistance change data, and perform change relationship analysis to obtain a relationship coefficient between power loss change and insulation resistance change; A screening module, used for performing feature fusion and feature screening on the power loss change data and the insulation resistance change data based on the relationship coefficient to obtain fused change feature data, and using the fused change feature data as a feature node of a Markov network to obtain an initial Markov network model; An optimization module, used for performing multiple iterations and model optimization on the initial Markov network model through a genetic algorithm to obtain a target Markov network model; The prediction module is used to monitor the operating status data of the pole-mounted circuit breaker in real time, and input the operating status data into the target Markov network model to perform operating fault prediction to obtain a real-time fault prediction result.

8. A pole mounted circuit breaker testing device, characterized in that: The pole-mounted circuit breaker testing device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the pole-mounted circuit breaker testing device to perform the pole-mounted circuit breaker testing method according to any one of claims 1-6.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by a processor, a pole-mounted circuit breaker testing method according to any one of claims 1 to 6 is implemented.

Images