Method and system for evaluating withstand strength state of primary and secondary fused column circuit breaker

Abstract

The application discloses a method and system for evaluating the pressure strength state of a primary and secondary fusion column breaker, and relates to the field of intelligent operation and maintenance of power equipment. The method comprises the following steps: extracting multi-dimensional factor data of the primary and secondary fusion column breaker; inputting a breaker state influence analysis model to perform influence analysis and construct a degradation influence graph; determining a plurality of weight coefficients corresponding to a plurality of evaluation dimensions based on the structured information of the degradation influence graph; and weighting and scoring the insulation strength layer, the secondary compatibility layer and the trend prediction layer of the primary and secondary fusion column breaker according to the plurality of weight coefficients to obtain the pressure strength state evaluation result. The application solves the technical problems of incomplete description of the multi-factor influence relationship and insufficient state determination accuracy in the existing evaluation of the pressure strength state of the primary and secondary fusion column breaker, and achieves the technical effects of comprehensively describing the multi-factor coupling influence relationship and improving the determination accuracy of the pressure strength state of the equipment.

Description

Technical Field

[0001] This application relates to the field of intelligent operation and maintenance of power equipment, and in particular to a method and system for assessing the withstand voltage strength of pole-mounted circuit breakers with integrated primary and secondary circuits. Background Technology

[0002] As a key control and protection device for power distribution lines, the withstand voltage condition of pole-mounted circuit breakers (POPs) directly affects the safety of power grid operation and the reliability of power supply. The industry typically conducts condition assessments using a combination of periodic offline withstand voltage tests and routine insulation parameter testing. Equipment condition identification is achieved through manual on-site testing, comparison of individual indicators, and empirical threshold judgments. However, this conventional testing and assessment method relies on discrete test data and manual experience, making it difficult to comprehensively reflect the insulation degradation patterns under the coupled effects of multiple factors. This can easily lead to biased condition assessments and missed potential hazards.

[0003] Currently, the relevant technologies for assessing the withstand voltage strength of primary and secondary integrated pole-mounted circuit breakers suffer from technical problems such as incomplete characterization of the influence of multiple factors and insufficient accuracy in condition determination. Summary of the Invention

[0004] This application provides a method and system for assessing the withstand voltage strength status of primary and secondary integrated pole-mounted circuit breakers. By extracting multi-dimensional factor data of the primary and secondary integrated pole-mounted circuit breakers and inputting this data into a pre-trained circuit breaker status influence analysis model for influence analysis, an insulation degradation influence diagram is constructed with influencing factors as nodes and causal transmission relationships with confidence weights as directed edges. Based on the structured information of this degradation influence diagram, the weight coefficients corresponding to multiple assessment dimensions—insulation strength layer, secondary compatibility layer, and trend prediction layer—are determined. The three assessment dimensions are then weighted and scored according to each weight coefficient to obtain the circuit breaker withstand voltage strength status assessment result. These technical means solve the existing technical problems of incomplete characterization of multi-factor influence relationships and insufficient accuracy in status determination in the withstand voltage strength status assessment of primary and secondary integrated pole-mounted circuit breakers. This achieves the technical effect of comprehensively characterizing the coupled influence relationships of multiple factors and improving the accuracy of equipment withstand voltage strength status determination.

[0005] This application provides a method for assessing the withstand voltage strength status of a primary and secondary integrated pole-mounted circuit breaker, comprising: extracting multi-dimensional factor data of the primary and secondary integrated pole-mounted circuit breaker; inputting the multi-dimensional factor data into a pre-trained circuit breaker status influence analysis model for influence analysis, and constructing a degradation influence diagram, wherein in the degradation influence diagram, nodes represent each influencing factor, directed edges represent the causal transmission relationship between factors and are attached with confidence weights, and the degradation influence diagram is used to characterize the insulation degradation influence relationship; determining multiple weight coefficients corresponding to multiple assessment dimensions based on the structured information of the degradation influence diagram; and weighting and scoring the insulation strength layer, secondary compatibility layer, and trend prediction layer of the primary and secondary integrated pole-mounted circuit breaker according to the multiple weight coefficients to obtain the withstand voltage strength status assessment result.

[0006] In a possible implementation, the following processing is performed: the multidimensional factor data includes physical connection relationship data, time aging line data, equipment profile data, and monitoring record data.

[0007] In possible implementations, the following processing is performed: the physical connection relationship data includes primary circuit topology information, electronic voltage transformer configuration information, and primary-secondary electrical coupling path information; the time aging line data includes operating years information, cumulative operation count information, and time series information of all test data; the equipment profile data includes equipment model information, insulation material type information, and installation environment parameter information; the monitoring record data includes partial discharge amplitude, phase distribution information, leakage current trend information, historical insulation resistance value information, and dielectric loss factor information.

[0008] In a possible implementation, the multidimensional factor data is input into a pre-trained circuit breaker state impact analysis model for impact analysis, a degradation impact map is constructed, and the following processing is performed: Constructing structured prompts, wherein the structured prompts include multidimensional factor data and preset domain causal prior knowledge; inputting the structured prompts into the circuit breaker state impact analysis model to guide the model to output formatted causal inference results; constructing an initial degradation impact map based on the causal relationship edges and confidence weights between factors in the causal inference results; and optimizing the initial degradation impact map based on monitoring record data to obtain a final degradation impact map.

[0009] In a possible implementation, the initial degradation impact map is optimized based on monitoring and recording data to obtain a degradation impact map, and the following processing is performed: initializing a seed node set; for each seed node in the seed node set, extracting its preceding related factors within a preset time window, and calculating the causal association strength value based on the causal inference result output by the circuit breaker state impact analysis model; when the causal association strength value is greater than a preset threshold, adding the corresponding causal edge to the initial degradation impact map to obtain a supplementary degradation impact map; recursively expanding upstream factor nodes until exceeding a preset maximum search depth, and performing redundant path pruning on the supplementary degradation impact map to obtain a degradation impact map.

[0010] In a possible implementation, the following process is performed: the seed node set contains the nodes corresponding to the current abnormal monitoring indicators in the monitoring record data.

[0011] In a possible implementation, the following processing is performed: the domain causal prior knowledge includes at least three of the following: information on the causal mechanism that the capacitance distribution effect of the primary and secondary fusion pole-mounted circuit breaker leads to an abnormal increase in the floating potential of the casing when the break withstand voltage is reached; information on the causal mechanism that the capacitance asymmetry of the high-voltage arm of the electronic voltage transformer exacerbates the uneven potential distribution; information on the causal chain that partial discharge is a key intermediate variable for insulation degradation; information on the causal mechanism that ambient temperature and humidity indirectly affect the withstand voltage level by influencing the dielectric properties of materials and surface creepage; and information on the causal coupling mechanism that primary-side partial discharge or operational overvoltage leads to secondary-side electromagnetic interference.

[0012] In a possible implementation, the following processing is performed: the insulation strength layer includes power frequency withstand voltage test pass indicators, insulation resistance change rate indicators, and partial discharge level indicators; the secondary compatibility layer includes secondary equipment functional integrity indicators during the test, post-impact data recovery capability indicators, and port residual voltage level indicators; the trend prediction layer includes insulation resistance change trend indicators, partial discharge change trend indicators, and dielectric loss factor change trend indicators.

[0013] In a possible implementation, the following process is performed: when the power frequency withstand voltage test pass index in the insulation strength layer is determined to be failed, the withstand voltage strength status assessment result is set to an abnormal state.

[0014] This application also provides a system for assessing the withstand voltage strength status of a primary and secondary integrated pole-mounted circuit breaker, comprising: a multi-dimensional factor data extraction module for extracting multi-dimensional factor data of the primary and secondary integrated pole-mounted circuit breaker; a causal reasoning analysis module for inputting the multi-dimensional factor data into a pre-trained circuit breaker status influence analysis model for influence analysis and constructing a degradation influence diagram, wherein nodes represent each influencing factor, directed edges represent the causal transmission relationship between factors and are assigned confidence weights, and the degradation influence diagram is used to characterize the insulation degradation influence relationship; a weight coefficient determination module for determining multiple weight coefficients corresponding to multiple assessment dimensions based on the structured information of the degradation influence diagram; and a weighted scoring module for weighted scoring of the insulation strength layer, secondary compatibility layer, and trend prediction layer of the primary and secondary integrated pole-mounted circuit breaker according to the multiple weight coefficients to obtain the withstand voltage strength status assessment result.

[0015] The proposed method and system for assessing the withstand voltage condition of a primary-secondary integrated pole-mounted circuit breaker, as described in this application, first extracts multi-dimensional factor data of the primary-secondary integrated pole-mounted circuit breaker. Then, this multi-dimensional factor data is input into a pre-trained circuit breaker condition influence analysis model for influence analysis, constructing a degradation influence diagram. In this degradation influence diagram, nodes represent influencing factors, and directed edges represent causal transmission relationships between factors, each with a confidence weight. This degradation influence diagram characterizes the insulation degradation influence relationship. Then, based on the structured information of the degradation influence diagram, multiple weight coefficients corresponding to multiple assessment dimensions are determined. Finally, the insulation strength layer, secondary compatibility layer, and trend prediction layer of the primary-secondary integrated pole-mounted circuit breaker are weighted and scored according to these weight coefficients to obtain the withstand voltage condition assessment result. The method and system proposed in this application achieve the technical effect of comprehensively depicting the coupled influence relationships of multiple factors and improving the accuracy of equipment withstand voltage condition determination. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a flowchart illustrating the method for assessing the withstand voltage strength of a primary and secondary fusion pole-mounted circuit breaker, as provided in an embodiment of this application.

[0018] Figure 2This is a schematic diagram of the structure of the withstand voltage strength assessment system for primary and secondary integrated pole-mounted circuit breakers provided in an embodiment of this application.

[0019] Figure labeling: Multidimensional factor data extraction module 10, causal reasoning analysis module 20, weight coefficient determination module 30, weighted scoring module 40. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments; however, it is understood that "some embodiments" may be the same or different subsets of all possible embodiments and may be combined with each other without conflict. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server 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 modules not explicitly listed or inherent to such processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0023] This application provides a method for assessing the withstand voltage condition of a primary and secondary integrated pole-mounted circuit breaker, such as... Figure 1 As shown, the method includes:

[0024] Step S100: Extract multi-dimensional factor data of the primary and secondary integrated pole-mounted circuit breaker. The multi-dimensional factor data includes physical connection relationship data, time aging line data, equipment profile data, and monitoring record data. The physical connection relationship data includes primary circuit topology information, electronic voltage transformer configuration information, and primary-secondary electrical coupling path information. The time aging line data includes operating years information, cumulative operation count information, and time series information of all test data. The equipment profile data includes equipment model information, insulation material type information, and installation environment parameter information. The monitoring record data includes partial discharge amplitude, phase distribution information, leakage current trend information, historical insulation resistance value information, and dielectric loss factor information.

[0025] Specifically, a multi-dimensional factor data of the primary and secondary integrated pole-mounted circuit breaker is extracted by linking the primary and secondary integrated pole-mounted circuit breaker status monitoring terminal, topology acquisition device and background database. The specific execution is as follows: start the monitoring terminal and topology acquisition device, establish a communication connection with the circuit breaker to be evaluated, and use the IEC61850 standard protocol for communication to ensure the stability and compatibility of data transmission.

[0026] Four types of data were collected. Physical connection data was acquired using a topology acquisition device. Primary circuit topology information was obtained by measuring the impedance values ​​between each node in the primary circuit of the circuit breaker. The circuit connection relationship was determined based on the impedance distribution. For example, if the impedance between node A and node B is less than 5Ω, they are considered to be directly connected. The wiring method of the primary circuit, such as single busbar connection or double busbar connection, was also recorded. Information on the configuration of electronic voltage transformers was obtained by reviewing equipment installation records and conducting on-site verification to determine the installation location, quantity, and wiring method of the transformers. Primary-secondary electrical coupling path information was detected using an electromagnetic coupling tester. The distribution and strength of the coupling path were determined by measuring the electromagnetic coupling coefficient between the primary and secondary sides. For example, a coupling coefficient greater than 0.8 was considered a strong coupling path. Data on the aging time line is obtained by retrieving data from the backend database and exporting historical data from the monitoring terminal. Operating years information is directly retrieved from equipment installation and acceptance records, recording the cumulative operating time from the commissioning date to the current date. Cumulative operation count information is obtained by exporting circuit breaker opening and closing operation records from the monitoring terminal, calculating the total number of operations since commissioning, including opening operations, closing operations, and cumulative operations. Time series information for past test data is retrieved from the backend database, including data on power frequency withstand voltage tests and insulation resistance tests, organized chronologically by test date. Equipment profile data is obtained by consulting the equipment manufacturer's manual, installation logs, and conducting on-site environmental monitoring. Equipment model information is directly read from the equipment nameplate; insulation material type information is determined from the manufacturer's manual; installation environmental parameters are collected using environmental monitoring sensors, continuously for 72 hours, with data collected hourly, and the average value is used as the final parameters, including ambient temperature, humidity, and altitude. Monitoring and recording data are obtained through real-time acquisition from monitoring terminals and historical data export. Partial discharge amplitude is acquired using an ultrasonic partial discharge detector, and the maximum amplitude acquired is taken as the partial discharge amplitude. Phase distribution information is obtained by simultaneously acquiring partial discharge signals and voltage phase signals through a partial discharge detector. The voltage phase is divided into 0 degrees to 360 degrees, and the number and amplitude of partial discharges in each phase interval are counted to form a phase distribution curve. Leakage current trend information is acquired through a leakage current tester. Leakage current values ​​are collected once a day for 30 consecutive days. The collected data are arranged by date, and a trend curve is plotted using a linear fitting method. Historical insulation resistance values ​​are retrieved from the background database, and historical data sequences are formed by organizing the data according to the test date. Dielectric loss factor information is acquired using a dielectric loss tester. The test voltage is set to 1.05 times the rated voltage, and the acquisition time is 5 minutes. The average value is taken as the dielectric loss factor.

[0027] After data collection, all multidimensional factor data are standardized using the min-max standardization method, which maps all types of data to the range of 0 to 1. The processed data is then stored in the backend database as input data for subsequent steps.

[0028] Step S200: Input the multidimensional factor data into the pre-trained circuit breaker state influence analysis model for influence analysis and construct a degradation influence diagram. In the degradation influence diagram, nodes represent each influencing factor, and directed edges represent the causal transmission relationship between factors and are assigned confidence weights. The degradation influence diagram is used to characterize the insulation degradation influence relationship.

[0029] Specifically, the circuit breaker condition impact analysis model adopts a hybrid architecture based on the fusion of Bayesian networks and structural equation models. The model input layer consists of feature vectors extracted from multi-dimensional factor data through feature engineering. The hidden layer consists of a directed acyclic graph structure layer of the Bayesian network and a path coefficient calculation layer of the structural equation model. The output layer consists of causal relationship edges between factors and their corresponding confidence weights. The model construction process is as follows: historical data of 1000 identical primary and secondary integrated pole-mounted circuit breakers are selected as modeling samples, specifically ZW32-12F type primary and secondary integrated pole-mounted circuit breakers. The equipment operation years cover the full cycle of 1-10 years. The historical data of each device includes multi-dimensional factor data and the corresponding actual insulation degradation situation. The dataset labeling standard is based on DL / T 596-2021 "Preventive Test Procedure for Power Equipment" and DL / T 2859-2024 "12kV Primary and Secondary Integrated Pole-Mounted Switches". The insulation degradation level and the true value of the contribution of each factor to the degradation are labeled in combination with the equipment disassembly inspection and withstand voltage test results. A Bayesian network structure learning algorithm (such as hill climbing combined with BIC scoring) is used to learn the initial network structure from the training samples. Then, structural equation modeling is used to estimate the path coefficients of causal relationships in the network, and these path coefficients are used as confidence weights to complete the training of the circuit breaker state impact analysis model. During model validation, the Bootstrap method is used to test the stability of the confidence weights. The model passes validation when the significance level of all causal edges is less than 0.05. The specific execution process is as follows: Standardized multidimensional factor data is retrieved from the backend database and converted into a feature vector format recognizable by the model. The converted feature vectors are input into the circuit breaker state impact analysis model. The model infers the conditional dependencies between factors through the Bayesian network structure, and then calculates the effect strength of each causal relationship as a confidence weight using the structural equation model. The model outputs the causal relationships between factors and their corresponding confidence weights, where the confidence weights range from 0 to 1. The larger the weight value, the stronger the corresponding causal relationship. Finally, a degradation impact map is constructed based on the causal relationships and confidence weights output by the model. The degradation impact diagram adopts a node-edge structure. Each node corresponds to a multi-dimensional factor data index, such as the partial discharge amplitude node, the service life node, and the environmental humidity node. The directed edges point from the causal factor node to the result factor node. For example, the service life node points to the partial discharge amplitude node, indicating that the service life is a causal factor affecting the partial discharge amplitude. The corresponding confidence weights are marked on the directed edges. The degradation impact diagram is used to characterize the influence relationship and influence intensity of each factor on the insulation degradation of the circuit breaker.

[0030] In one possible implementation, the multidimensional factor data is input into a pre-trained circuit breaker state influence analysis model for influence analysis, and a degradation influence diagram is constructed. Step S200 further includes step S210, constructing structured prompts, wherein the structured prompts include multidimensional factor data and preset domain causal prior knowledge. The domain causal prior knowledge includes at least three of the following: the causal mechanism information that the capacitance distribution effect of the primary and secondary fusion pole-mounted circuit breaker leads to an abnormal increase in the floating potential of the shell when the withstand voltage is broken; the causal mechanism information that the capacitance asymmetry of the high-voltage arm of the electronic voltage transformer exacerbates the uneven potential distribution; the causal chain information that partial discharge is a key intermediate variable for insulation degradation; the causal mechanism information that the ambient temperature and humidity indirectly affect the withstand voltage level by affecting the dielectric properties of the material and surface creepage; and the causal coupling mechanism information that primary side partial discharge or operational overvoltage leads to secondary side electromagnetic interference.

[0031] Specifically, structured prompt words are constructed. These structured prompt words use a fixed-format text description, consisting of two parts: multi-dimensional factor data and pre-defined domain causal prior knowledge. The specific construction is as follows: The multi-dimensional factor data collected and standardized in step S100 is converted into text descriptions. Each data indicator clearly indicates its data type, value, and unit. For example, physical connection relationship data: the primary circuit topology is a single busbar connection, the electronic voltage transformer is installed on the incoming side, using a star connection, and the primary-secondary electrical coupling coefficient is 0.85; time aging line data: 6 years of operation, 640 cumulative operations, and the time series of insulation resistance tests is 2200MΩ in 2021. 2022: 2000MΩ; 2023: 1800MΩ; 2024: 1700MΩ; 2025: 1600MΩ; 2026: 1500MΩ. Equipment profile data: Equipment model ZW32-12F / 630-20; Insulation material: epoxy resin shell, silicone rubber insulating rod; Installation environment temperature: 25℃; Humidity: 60%; Altitude: 100 meters. Monitoring record data: Partial discharge amplitude: 500pC; Phase distribution: 12 partial discharges in the 0-90 degree range; Average amplitude: 300pC; Leakage current increased from 5μA to 8μA within 30 days; Historical insulation resistance values: 2200-1500MΩ; Dielectric loss factor: 0.005.At least three types of causal prior knowledge from a preset domain are selected; this embodiment selects five types, namely: 1. Causal mechanism information regarding the abnormal increase in the floating potential of the casing during withstand voltage testing of the primary and secondary pole-mounted circuit breakers due to the capacitance distribution effect. Specifically, during the withstand voltage testing of the circuit breaker, uneven capacitance distribution between the primary and secondary circuits leads to charge accumulation on the casing surface, resulting in an abnormal increase in the floating potential. When the capacitance distribution deviation is greater than 10%, the floating potential increases by more than 500V; 2. Causal mechanism information regarding the asymmetry of the high-voltage arm capacitance of electronic voltage transformers exacerbating the uneven potential distribution. Specifically, when the capacitance value deviation of the high-voltage arm capacitance of the electronic voltage transformer is greater than 5%, it leads to a transformer output voltage deviation exceeding 3%, thereby exacerbating the uneven potential distribution between the primary and secondary circuits and affecting the withstand voltage level; 3. Causal chain information regarding partial discharge as a key intermediate variable in insulation degradation. Specifically described... The following information is provided: 1. The aging of insulation materials, excessive ambient humidity, and prolonged service life can all lead to an increase in partial discharge amplitude. When the partial discharge amplitude continuously exceeds 400 pC, it will accelerate the aging rate of the insulation materials, thereby reducing the withstand voltage of the circuit breaker; 2. The causal mechanism information of ambient temperature and humidity indirectly affecting the withstand voltage level by affecting the dielectric properties of materials and surface creepage is described as follows: When the ambient temperature exceeds 35℃ or the humidity exceeds 80%, the dielectric constant of the insulation material decreases by more than 5%, and the surface creepage distance shortens by more than 10%, thereby leading to a decrease in the withstand voltage level of the circuit breaker by more than 8%; 3. The causal coupling mechanism information of primary-side partial discharge or operational overvoltage leading to secondary-side electromagnetic interference is described as follows: When the primary-side partial discharge amplitude exceeds 600 pC or the operational overvoltage exceeds 1.2 times the rated voltage, a strong electromagnetic signal will be generated, causing electromagnetic interference to the secondary-side equipment, resulting in a data acquisition error of more than 2% for the secondary equipment. Finally, the textual descriptions of the multidimensional factor data are integrated with the selected domain causal prior knowledge text, using the format of multidimensional factor data: [specific data description]; domain causal prior knowledge: [specific knowledge description] to form structured prompt words. These structured prompt words are used to guide the circuit breaker state impact analysis model to accurately perform causal reasoning.

[0032] Step S220: Input the structured prompts into the circuit breaker state impact analysis model to guide the model to output formatted causal inference results. Specifically, start the circuit breaker state impact analysis model and set the model's output format to a fixed template. An example of the template content is as follows: Causal Factor: [Factor Name]; Result Factor: [Factor Name]; Causal Relationship Description: [Detailed Description]; Confidence Weight: [Specific Value], used to ensure the standardization and consistency of the output results. The confidence weight is calculated based on the path coefficients output by the structural equation model and the lower limit of the Bootstrap confidence interval, using the formula: Confidence Weight W = γ × P + δ × B lowerWhere P is the standardized path coefficient output by the structural equation model, located in the interval 0-1, and B... lower The lower bound of the causal effect confidence interval, calculated by Bootstrap after 500 resampling iterations, is located in the 0-1 range. γ and δ are fixed weight coefficients, with γ=0.7 and δ=0.3, and the calculation result is rounded to two decimal places. A weighted filtering threshold of 0.3 is also set; causal relationships with calculation results below 0.3 are considered weak associations and are not included in the output. Structured prompts are input into the model in text form. After receiving the prompts, the model first parses the multidimensional factor data and domain causal prior knowledge within the prompts. Combining this with the Bayesian network structure and path coefficients already learned within the model, it infers the causal relationships between various factors. Then, according to a preset output template, the model outputs all inferred causal relationships one by one, forming a formatted causal inference result, with each causal relationship corresponding to one output record.

[0033] Step S230: Construct an initial degradation influence graph based on the causal relationship edges and confidence weights between factors in the causal inference results. Specifically, retrieve the formatted causal inference results, extract the causal factors, result factors, and confidence weights from each causal inference result, and establish a causal relationship edge list. Each record in the list contains three parameters: causal factor node, result factor node, and confidence weight. Determine the node set of the initial degradation influence graph. The node set consists of all factor nodes involved in all causal inference results, i.e., all indicators in the multidimensional factor data, and assign a unique identifier to each node. Based on the causal relationship edge list, draw directed edges between nodes, with the starting point of the directed edge being the causal factor node and the ending point being the result factor node. Label the corresponding confidence weight next to the directed edge.

[0034] Step S240 involves optimizing the initial degradation impact map based on monitoring data to obtain a new degradation impact map. Specifically, the confidence weights output by the circuit breaker state impact analysis model correspond to the common causal relationships of all primary and secondary integrated pole-mounted circuit breakers of the same type. The initial degradation impact map constructed based on these confidence weights includes all causal edges that may affect the insulation degradation of the circuit breaker. These causal edges only reflect general rules and do not incorporate the real-time operating status and abnormal conditions of the circuit breaker currently being evaluated. The core of this optimization process is to selectively screen and strengthen the causal edges in the initial degradation impact map. By calculating the causal correlation strength value, key causal edges that contribute significantly to the abnormal state of the current circuit breaker are selected, and these causal edges are retained and strengthened. At the same time, causal edges that have no significant impact on the current abnormal state are removed. Then, through recursive expansion of upstream factors and redundant path pruning, an optimized degradation impact map that fits the actual state of the current equipment is finally obtained, ensuring the accuracy and specificity of the withstand voltage strength assessment and solving the problem of mismatch between general causal rules and the actual state of specific equipment.

[0035] In one possible implementation, the initial degradation impact map is optimized based on the monitoring record data to obtain a degradation impact map. Step S240 further includes step S241, initializing a seed node set, which contains nodes corresponding to the current abnormal monitoring indicators in the monitoring record data. Specifically, the monitoring record data collected in step S100 is retrieved. For each monitoring indicator, a corresponding abnormal judgment threshold is preset. The abnormal judgment threshold is set based on the industry standard and equipment factory parameters of the primary and secondary integrated pole-mounted circuit breaker. The current value of each monitoring indicator is compared with the corresponding abnormal judgment threshold. If the current value exceeds the abnormal threshold, the monitoring indicator is determined to be an abnormal monitoring indicator. All nodes corresponding to the abnormal monitoring indicators are used as seed nodes and included in the seed node set. Each seed node uses the standard name of the monitoring indicator as its node name.

[0036] Step S242: For each seed node in the seed node set, extract the associated factors within its preceding preset time window, and calculate the causal association strength value based on the causal inference results output by the circuit breaker state impact analysis model. Specifically, the preset time window is 30 days, that is, extract factors that have data correlation with the monitoring indicators corresponding to the seed node in the past 30 days. The criteria for determining the associated factors are: the change in the value of the factor will cause the value of the monitoring indicator corresponding to the seed node to change. Combined with the causal inference results output in step S220, filter out the factors that have a causal relationship with the seed node as associated factors. Retrieve the numerical data of each associated factor and the numerical data of the monitoring indicators corresponding to the seed node in the past 30 days, calculate the numerical change rate of each associated factor. The formula for calculating the numerical change rate is (current value - value 30 days ago) / value 30 days ago. Then, retrieve the causal inference results output by the circuit breaker state impact analysis model, extract the confidence weight between each associated factor and the seed node, and finally calculate the causal association strength value: causal association strength value = confidence weight × numerical change rate of associated factor.

[0037] Step S243: When the causal correlation strength value is greater than a preset threshold, the corresponding causal edge is added to the initial degradation influence graph to obtain a supplementary degradation influence graph. Specifically, the preset causal correlation strength value threshold is determined by analyzing historical data from 100 circuit breakers of the same type. Each causal correlation strength value calculated in step S242 is compared with the preset threshold, and it is determined whether it is greater than the threshold. For causal relationships where the causal correlation strength value is greater than the preset threshold, the corresponding causal edge is extracted, and these causal edges are added to the initial degradation influence graph constructed in step S230 to obtain a supplementary degradation influence graph.

[0038] Step S244: Recursively expand upstream factor nodes until the preset maximum search depth is exceeded. Redundant paths are pruned in the supplementary degradation impact graph to obtain the degradation impact graph. Specifically, the preset maximum search depth is 3 layers. Upstream factors exceeding 3 layers have a negligible impact on insulation degradation. Starting from the seed node of the supplementary degradation impact graph, upstream factor nodes are recursively expanded. Upstream factor nodes are the causal factor nodes that cause the current node's anomaly. If an upstream factor node for a node does not exist or the expansion depth reaches 3 layers, the recursive expansion of that path is stopped. Next, the recursively expanded paths are pruned. The total confidence weight of each path is calculated. The total confidence weight is the product of the confidence weights of all directed edges on the path. A preset total confidence weight threshold is set. The total confidence weight of each path is compared with the threshold. If the total confidence weight is less than the threshold, the path is determined to be redundant and removed from the supplementary degradation impact graph to obtain the final degradation impact graph.

[0039] Step S300: Determine multiple weight coefficients corresponding to multiple evaluation dimensions based on the structured information of the degradation impact map.

[0040] Specifically, the nodes in the degradation impact diagram are derived from the multi-dimensional factor data extracted in step S100, including nodes such as partial discharge amplitude, historical insulation resistance, dielectric loss factor, leakage current, primary-secondary electrical coupling path, electronic voltage transformer configuration, and installation environment parameters. Each node forms a causal transmission relationship through directed edges with confidence weights. The nodes in the degradation impact diagram are categorized into three evaluation dimensions according to their impact targets, with the following specific categorization relationships: the insulation strength layer corresponds to the nodes of partial discharge amplitude, historical insulation resistance, and dielectric loss factor in the degradation impact diagram; the secondary compatibility layer corresponds to the nodes of primary-secondary electrical coupling path, electronic voltage transformer configuration, and partial discharge amplitude in the degradation impact diagram; and the trend prediction layer corresponds to the time-series change-related nodes of the nodes of historical insulation resistance, partial discharge amplitude, and dielectric loss factor in the degradation impact diagram. For each evaluation dimension, all nodes belonging to that dimension in the degradation impact graph are traversed, and the confidence weights of all incoming edges pointing to that type of node are extracted. The arithmetic mean of all incoming edge confidence weights under the same dimension is calculated to obtain the initial impact weight of that dimension. The initial impact weights of the insulation strength layer, secondary compatibility layer, and trend prediction layer are summed to obtain the total initial weight. Then, the initial impact weight of each dimension is divided by the total initial weight for normalization to obtain the final weight coefficient of that dimension.

[0041] Step S400: The insulation strength layer, secondary compatibility layer, and trend prediction layer of the primary and secondary integrated pole-mounted circuit breaker are weighted and scored according to multiple weighting coefficients to obtain a withstand voltage strength status assessment result. The insulation strength layer includes indicators for passing the power frequency withstand voltage test, the rate of change of insulation resistance, and the partial discharge level. The secondary compatibility layer includes indicators for the functional integrity of secondary equipment during the test, the data recovery capability after impact, and the port residual voltage level. The trend prediction layer includes indicators for the trend of insulation resistance change, the trend of partial discharge change, and the trend of dielectric loss factor change. When the power frequency withstand voltage test passing indicator in the insulation strength layer is determined to be failed, the withstand voltage strength status assessment result is set to an abnormal state.

[0042] Specifically, each indicator within the three evaluation dimensions is scored independently according to a preset percentage system: For the insulation strength layer, the power frequency withstand voltage test involves continuously applying the rated power frequency withstand voltage for one minute; no breakdown or flashover is considered a pass, scoring 100 points for a pass and 0 points for a failure. The insulation resistance change rate is calculated by dividing the current insulation resistance by the insulation resistance of the previous cycle, multiplying by 100%, and so on. A smaller change rate results in a higher score; specifically: a change rate ≥ -5% earns 90-1. 0.00 points; -10% ≤ change rate < -5% = 70-89 points; -20% ≤ change rate < -10% = 40-69 points; change rate < -20% = 0-39 points; Partial discharge level is scored in segments according to amplitude, with smaller amplitudes receiving higher scores: amplitude ≤ 200pC = 90-100 points; 200pC < amplitude ≤ 400pC = 70-89 points; 400pC < amplitude ≤ 600pC = 40-69 points; amplitude > 600pC = 0-39 points. The arithmetic mean of the three indicators is taken as the score for the insulation strength layer.

[0043] In the secondary compatibility layer, the functional integrity of secondary equipment is scored based on the number of functional anomalies during the test. No anomalies are scored as 100 points; one minor anomaly that can be automatically recovered is scored as 80 points; one unrecoverable anomaly is scored as 50 points; two or more functional anomalies are scored as 0 points. Data recovery capability after a voltage surge is scored based on the data recovery time after the surge, with faster recovery resulting in higher scores: ≤10s for 90-100 points; 10s < ≤30s for 70-89 points; 30s < ≤60s for 40-69 points; >60s or no recovery results in 0-39 points. Port residual voltage level is scored based on the remaining voltage at the port after the test, with lower residual voltage resulting in higher scores: ≤50V for 90-100 points; 50V < ≤100V for 70-89 points; 100V < ≤200V for 40-69 points; >200V for 0-39 points. The arithmetic mean of the three indicators is taken as the score of the secondary compatibility layer.

[0044] In the trend prediction layer, the trends of insulation resistance, partial discharge, and dielectric loss factor are all scored according to their annual variation. The more stable the change, the higher the score. Specifically: insulation resistance variation trend with a continuous 12-month fluctuation range ≤10% scores 90-100 points; 10%-20% scores 70-89 points; 20%-30% scores 40-69 points; >30% scores 0-39 points; partial discharge variation trend with a continuous 12-month fluctuation range... For amplitude fluctuation increments ≤ 50 pC, 90-100 points are awarded; 50-100 pC, 70-89 points; 100-200 pC, 40-69 points; > 200 pC, 0-39 points. For the dielectric loss factor change trend, a continuous 12-month numerical fluctuation ≤ 0.002 is awarded 90-100 points; 0.002-0.005, 70-89 points; 0.005-0.01, 40-69 points; > 0.01, 0-39 points. The arithmetic mean of the three indicators is taken as the score for the trend prediction layer.

[0045] The scores for each dimension are multiplied by the corresponding weighting coefficients obtained in step S300 and then summed to obtain the total withstand voltage strength score. A total score ≥ 90 is considered excellent, 80 ≤ score < 90 is good, 60 ≤ score < 80 is a warning, and a score < 60 is a fault. A mandatory judgment rule is also set: when the power frequency withstand voltage test pass indicator in the insulation strength layer is determined to be failed, regardless of the total score value, the withstand voltage strength status assessment result is directly set to an abnormal state.

[0046] The following is a complete evaluation process and results for a specific embodiment of this application: Example 1: The equipment to be evaluated is a ZW32-12F / 630-20 primary and secondary integrated pole-mounted circuit breaker in a distribution substation. It has been in operation for 6 years, installed in a normal temperature and humidity environment, and has no major fault records. Standardized multi-dimensional factor data core indicators are: operating years (6 years), cumulative operation count (640 times), ambient temperature (25℃), humidity (60%), partial discharge amplitude (500 pC), 30-day leakage current increasing from 5 μA to 8 μA, annual insulation resistance decay rate (8%), dielectric loss factor (0.005), and primary-secondary coupling coefficient (0.85). Structured prompts were input into the circuit breaker state-effect analysis model, outputting 18 valid causal relationships. Typical results include: causal factor (operating years) → outcome factor (partial discharge amplitude), causal description: for every additional year of operation, the average partial discharge amplitude increases by 8%, with a confidence weight of 0.82; causal factor (ambient humidity) → outcome factor (dielectric loss factor), with a confidence weight of 0.75. After seed node optimization and path pruning, the final degraded influence graph retains 12 core causal edges. The initial weights of the insulation strength layer are 0.81, the secondary compatibility layer is 0.72, and the trend prediction layer is 0.78. After normalization, the final weight coefficients are 0.36, 0.32, and 0.32, respectively. Insulation strength layer: Power frequency withstand voltage test passed (100 points), insulation resistance change rate -8% (82 points), partial discharge 500pC (55 points), average score 79 points; Secondary compatibility layer: No functional abnormalities (100 points), data recovery time 8s (95 points), port residual voltage 42V (96 points), average score 97 points; Trend prediction layer: Insulation resistance fluctuation 8% (85 points), partial discharge increment 90pC (78 points), dielectric loss fluctuation 0.005 (80 points), average score 81 points; Total score = 79×0.36+97×0.32+81×0.32=85.48 points, final evaluation result: Good.

[0047] Example 2: Fault sample verification: The same model of equipment, in operation for 9 years, in a long-term high-humidity environment, had a partial discharge amplitude of 720pC, an annual insulation resistance decay of 25%, and failed the power frequency withstand voltage test for flashover. According to mandatory rules, it was directly judged as a serious abnormal condition, which is completely consistent with the results of on-site disassembly and inspection of insulation aging and minor creepage faults in the casing.

[0048] The comparative verification results of this application and existing evaluation methods are as follows: Fifty pole-mounted circuit breakers with integrated primary and secondary circuits in different operating states were selected, and comparative tests were conducted using the traditional threshold evaluation method, the conventional machine learning evaluation method, and the method of this application, respectively. Based on the true values ​​of on-site withstand voltage tests and equipment disassembly failure results, the accuracy rates were verified as follows: 1. Traditional threshold method: evaluation accuracy 78%, false positive rate 22%, unable to identify implicit insulation degradation trends; 2. Conventional machine learning method: evaluation accuracy 88%, false positive rate 12%, lacking causal mechanism support; 3. Causal reasoning evaluation method of this application: evaluation accuracy 98%, false positive rate 2%, accurately identifying explicit faults and implicit degradation trends, and possessing interpretability. The comparative results demonstrate that the method of this application significantly improves evaluation accuracy and reliability compared to existing technologies, solving the technical defects of traditional methods such as one-sided evaluation, lack of causal basis, and missed detection of implicit faults.

[0049] This application embodiment extracts multi-dimensional factor data of the primary and secondary integrated pole-mounted circuit breaker, inputs the data into a pre-trained circuit breaker state influence analysis model for influence analysis, constructs an insulation degradation influence diagram with influencing factors as nodes and causal transmission relationships with confidence weights as directed edges, determines the weight coefficients corresponding to multiple evaluation dimensions of the insulation strength layer, secondary compatibility layer, and trend prediction layer based on the structured information of the degradation influence diagram, and performs weighted scoring on the three evaluation dimensions according to each weight coefficient to obtain the circuit breaker withstand voltage strength state evaluation result. This technical means solves the technical problems of incomplete characterization of multi-factor influence relationships and insufficient accuracy of state determination in existing primary and secondary integrated pole-mounted circuit breaker withstand voltage strength state evaluation, and achieves the technical effect of comprehensively characterizing multi-factor coupled influence relationships and improving the accuracy of equipment withstand voltage strength state determination.

[0050] In the above text, refer to Figure 1 A method for assessing the withstand voltage condition of a primary and secondary integrated pole-mounted circuit breaker according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 A system for assessing the withstand voltage condition of a primary and secondary fusion pole-mounted circuit breaker is described according to an embodiment of the present invention.

[0051] The withstand voltage strength condition assessment system for primary and secondary integrated pole-mounted circuit breakers according to embodiments of the present invention addresses the technical problems of incomplete characterization of multi-factor influence relationships and insufficient accuracy of condition determination in existing withstand voltage strength condition assessment methods for primary and secondary integrated pole-mounted circuit breakers. It achieves the technical effect of comprehensively characterizing the coupled influence relationships of multiple factors and improving the accuracy of equipment withstand voltage strength condition determination. The withstand voltage strength condition assessment system for primary and secondary integrated pole-mounted circuit breakers includes: a multi-dimensional factor data extraction module 10, a causal reasoning analysis module 20, a weight coefficient determination module 30, and a weighted scoring module 40.

[0052] The multi-dimensional factor data extraction module 10 is used to extract multi-dimensional factor data of the primary and secondary integrated pole-mounted circuit breaker; the causal reasoning analysis module 20 is used to input the multi-dimensional factor data into a pre-trained circuit breaker state influence analysis model for influence analysis and construct a degradation influence diagram, wherein nodes represent each influencing factor, directed edges represent the causal transmission relationship between factors and are attached with confidence weights, and the degradation influence diagram is used to characterize the insulation degradation influence relationship; the weight coefficient determination module 30 is used to determine multiple weight coefficients corresponding to multiple evaluation dimensions based on the structured information of the degradation influence diagram; the weighted scoring module 40 is used to perform weighted scoring on the insulation strength layer, secondary compatibility layer and trend prediction layer of the primary and secondary integrated pole-mounted circuit breaker according to multiple weight coefficients to obtain the withstand voltage strength state evaluation result.

[0053] The detailed description of the specific configuration of the multidimensional factor data extraction module 10 is explained as follows: As mentioned above, the multidimensional factor data extraction module 10 may further include: the multidimensional factor data includes physical connection relationship data, time aging line data, equipment profile data and monitoring record data.

[0054] The multi-dimensional factor data extraction module 10 may further include: the physical connection relationship data including primary circuit topology information, electronic voltage transformer configuration information, and primary-secondary electrical coupling path information; the time aging line data including operating years information, cumulative operation times information, and time series information of all test data; the equipment profile data including equipment model information, insulation material type information, and installation environment parameter information; and the monitoring record data including partial discharge amplitude, phase distribution information, leakage current trend information, historical insulation resistance value information, and dielectric loss factor information.

[0055] The causal reasoning analysis module 20 is described in detail below: As mentioned above, the multi-dimensional factor data is input into a pre-trained circuit breaker state influence analysis model for influence analysis, and a degradation influence graph is constructed. The causal reasoning analysis module 20 may further include: a structured prompt word construction unit for constructing structured prompt words, wherein the structured prompt words include multi-dimensional factor data and preset domain causal prior knowledge; a causal reasoning unit for inputting the structured prompt words into the circuit breaker state influence analysis model, guiding the circuit breaker state influence analysis model to output formatted causal reasoning results; an initial degradation influence graph construction unit for constructing an initial degradation influence graph based on the causal relationship edges and confidence weights between factors in the causal reasoning results; and an influence graph optimization unit for optimizing the initial degradation influence graph based on monitoring record data to obtain a degradation influence graph.

[0056] The optimization of the initial degradation impact map based on monitoring and recorded data yields a further degradation impact map. The impact map optimization unit may include: a seed node set initialization subunit for initializing the seed node set; a causal association strength value calculation subunit for extracting the associated factors within a preset time window for each seed node in the seed node set, and calculating the causal association strength value based on the causal inference results output by the circuit breaker state impact analysis model; a causal edge supplementation subunit for adding the corresponding causal edge to the initial degradation impact map when the causal association strength value exceeds a preset threshold, thus obtaining a supplementary degradation impact map; and a redundant path pruning subunit for recursively expanding upstream factor nodes until a preset maximum search depth is exceeded, performing redundant path pruning on the supplementary degradation impact map to obtain a degradation impact map.

[0057] The seed node set initialization subunit may further include: the seed node set contains nodes corresponding to the current abnormal monitoring indicators in the monitoring record data.

[0058] The structured prompt word construction unit may further include: the domain causal prior knowledge includes at least three of the following: the causal mechanism information that the capacitance distribution effect of the primary and secondary fusion pole-mounted circuit breaker leads to an abnormal increase in the floating potential of the shell when the break withstand voltage is reached; the causal mechanism information that the capacitance asymmetry of the high-voltage arm of the electronic voltage transformer exacerbates the uneven potential distribution; the causal chain information that partial discharge is a key intermediate variable for insulation degradation; the causal mechanism information that the ambient temperature and humidity indirectly affect the withstand voltage level by affecting the dielectric properties of the material and surface creepage; and the causal coupling mechanism information that primary side partial discharge or operational overvoltage leads to secondary side electromagnetic interference.

[0059] The specific configuration of the weighted scoring module 40 is described in detail below: As mentioned above, the weighted scoring module 40 may further include: the insulation strength layer includes power frequency withstand voltage test pass indicators, insulation resistance change rate indicators, and partial discharge level indicators; the secondary compatibility layer includes secondary equipment functional integrity indicators, post-impact data recovery capability indicators, and port residual voltage level indicators; the trend prediction layer includes insulation resistance change trend indicators, partial discharge change trend indicators, and dielectric loss factor change trend indicators.

[0060] The weighted scoring module 40 may further include: when the power frequency withstand voltage test pass index in the insulation strength layer is determined to be failed, the withstand voltage strength status evaluation result is set to an abnormal state.

[0061] The withstand voltage strength assessment system for primary and secondary integrated pole-mounted circuit breakers provided in this invention can execute the withstand voltage strength assessment method for primary and secondary integrated pole-mounted circuit breakers provided in any embodiment of this invention, and has the corresponding functional modules and beneficial effects of the method.

[0062] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0063] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for assessing the withstand voltage strength condition of a primary and secondary integrated pole-mounted circuit breaker, characterized in that, The method includes: Extract multi-dimensional factor data from primary and secondary integrated pole-mounted circuit breakers; The multidimensional factor data is input into a pre-trained circuit breaker state influence analysis model for influence analysis, and a degradation influence diagram is constructed. In the degradation influence diagram, nodes represent each influencing factor, and directed edges represent the causal transmission relationship between factors and are attached with confidence weights. The degradation influence diagram is used to characterize the insulation degradation influence relationship. Based on the structured information of the degradation impact map, multiple weight coefficients corresponding to multiple evaluation dimensions are determined; The insulation strength layer, secondary compatibility layer, and trend prediction layer of the primary and secondary integrated pole-mounted circuit breaker are weighted and scored according to multiple weighting coefficients to obtain the withstand voltage strength status assessment result.

2. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 1, characterized in that, The multidimensional factor data includes physical connection relationship data, time aging line data, equipment profile data, and monitoring record data.

3. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 2, characterized in that, The physical connection relationship data includes primary circuit topology information, electronic voltage transformer configuration information, and primary-secondary electrical coupling path information; The time aging line data includes information on the number of years of operation, the cumulative number of operations, and the time series information of data from each test. The equipment profile data includes equipment model information, insulation material type information, and installation environment parameter information; The monitoring and recording data includes partial discharge amplitude, phase distribution information, leakage current trend information, historical insulation resistance information, and dielectric loss factor information.

4. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 1, characterized in that, The multidimensional factor data is input into a pre-trained circuit breaker state-of-the-art model for impact analysis, and a degradation impact map is constructed, including: Construct structured prompt words, wherein the structured prompt words include multidimensional factor data and pre-defined domain causal prior knowledge; The structured prompts are input into the circuit breaker state impact analysis model, which then guides the model to output formatted causal reasoning results. Based on the causal relationship edges and confidence weights among the factors in the causal inference results, an initial degradation influence graph is constructed. The initial degradation impact map is optimized based on the monitoring and recording data to obtain a new degradation impact map.

5. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 4, characterized in that, The initial degradation impact map is optimized based on the monitoring record data to obtain a new degradation impact map, including: Initialize the seed node set; For each seed node in the seed node set, extract the associated factors within its preceding preset time window, and calculate the causal association strength value based on the causal inference results output by the circuit breaker state influence analysis model. When the causal correlation strength value is greater than a preset threshold, the corresponding causal edge is added to the initial degradation influence graph to obtain a supplementary degradation influence graph; The upstream factor nodes are recursively expanded until the preset maximum search depth is exceeded. Redundant paths are pruned on the supplementary degradation impact map to obtain the degradation impact map.

6. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 5, characterized in that, The seed node set contains the nodes corresponding to the current abnormal monitoring indicators in the monitoring record data.

7. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 4, characterized in that, The causal prior knowledge in the domain includes at least three of the following: information on the causal mechanism that the capacitance distribution effect of a primary and secondary fusion pole-mounted circuit breaker leads to an abnormal increase in the floating potential of the casing during withstand voltage; information on the causal mechanism that the capacitance asymmetry of the high-voltage arm of an electronic voltage transformer exacerbates the uneven potential distribution; information on the causal chain that partial discharge is a key intermediate variable for insulation degradation; information on the causal mechanism that ambient temperature and humidity indirectly affect the withstand voltage level by influencing the dielectric properties of materials and surface creepage; and information on the causal coupling mechanism that primary-side partial discharge or operational overvoltage leads to secondary-side electromagnetic interference.

8. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 1, characterized in that, The insulation strength layer includes indicators for passing the power frequency withstand voltage test, indicators for the rate of change of insulation resistance, and indicators for the level of partial discharge. The secondary compatibility layer includes the functional integrity index of the secondary equipment during the test, the data recovery capability index after impact, and the port residual voltage level index. The trend prediction layer includes trends in insulation resistance, partial discharge, and dielectric loss factor.

9. The method for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker as described in claim 8, characterized in that, When the power frequency withstand voltage test pass index in the insulation strength layer is determined to be failed, the withstand voltage strength status assessment result is set to an abnormal state.

10. A system for assessing the withstand voltage strength of a primary and secondary integrated pole-mounted circuit breaker, characterized in that, The system is used to implement the method for assessing the withstand voltage condition of a primary and secondary integrated pole-mounted circuit breaker as described in any one of claims 1-9, the system comprising: The multi-dimensional factor data extraction module is used to extract multi-dimensional factor data of primary and secondary integrated pole-mounted circuit breakers; The causal reasoning analysis module is used to input the multidimensional factor data into a pre-trained circuit breaker state influence analysis model for influence analysis and to construct a degradation influence diagram. In the degradation influence diagram, nodes represent each influencing factor, and directed edges represent the causal transmission relationship between factors and are attached with confidence weights. The degradation influence diagram is used to characterize the insulation degradation influence relationship. The weight coefficient determination module is used to determine multiple weight coefficients corresponding to multiple evaluation dimensions based on the structured information of the degradation impact map. The weighted scoring module is used to perform weighted scoring on the insulation strength layer, secondary compatibility layer and trend prediction layer of the primary and secondary integrated pole-mounted circuit breaker according to multiple weight coefficients to obtain the withstand voltage strength status assessment result.

Images