A pole-mounted circuit breaker aging test method and system

Abstract

This invention discloses an aging test method and system for pole-mounted circuit breakers. The method includes: acquiring historical operating condition data and real-time load characteristics of the pole-mounted circuit breaker to generate a basic input set for accelerated aging testing; generating an accelerated aging loading strategy based on the basic input set; conducting electrothermal cyclic loading tests on the pole-mounted circuit breaker according to the accelerated aging loading strategy to form a time-domain aligned performance degradation dataset; dynamically adjusting the electrothermal stress intensity and duration in the test cycle using the performance degradation dataset to generate an adaptive cyclic control command that conforms to the target aging curve; executing the adaptive cyclic control command until the predetermined test cycle is completed, and outputting the aging test results of the pole-mounted circuit breaker. Using this invention, the accuracy and efficiency of aging testing can be improved, providing a scientific basis for the life assessment and reliability prediction of pole-mounted circuit breakers.

Description

Technical Field

[0001] This invention belongs to the technical field of pole-mounted circuit breakers, and in particular, it relates to an aging test method and system for pole-mounted circuit breakers. Background Technology

[0002] With the increasing intelligence level of power distribution networks, the long-term operational reliability of pole-mounted circuit breakers is crucial to grid stability. Traditional aging test methods often employ constant or simple cyclic stress loading, which struggles to accurately simulate the complex, non-uniform, and dynamically changing operating conditions faced by circuit breakers in actual operation. These methods often neglect the coupling effect between historical load characteristics, frequent opening and closing actions, and temperature changes, leading to significant differences between test conditions and the actual aging process, thus limiting the reference value of the obtained life assessment data. Furthermore, existing test processes are mostly open-loop control, unable to dynamically adjust test parameters based on the real-time performance degradation state of the tested object, resulting in low test efficiency and difficulty in accurately predicting its remaining life and reliability evolution trend. Therefore, there is an urgent need for an accelerated aging test method that closely reflects actual operating conditions and can adaptively adjust to scientifically evaluate the long-term service performance of pole-mounted circuit breakers. Summary of the Invention

[0003] The purpose of this invention is to provide an aging test method and system for pole-mounted circuit breakers to overcome the shortcomings of the prior art, improve the accuracy and efficiency of aging tests, and provide a scientific basis for the life assessment and reliability prediction of pole-mounted circuit breakers.

[0004] One embodiment of this application provides an aging test method for pole-mounted circuit breakers, the method comprising:

[0005] Acquire historical operating condition data and real-time load characteristics of pole-mounted circuit breakers to generate a basic input set for accelerated aging tests that includes current load distribution and opening and closing frequencies.

[0006] Based on the aforementioned accelerated aging test input set, a non-uniform cyclic electrothermal stress loading spectrum is constructed to generate an accelerated aging loading strategy that integrates dynamic current, opening and closing action and interval time.

[0007] According to the accelerated aging loading strategy, the pole-mounted circuit breaker was subjected to an electrothermal cyclic loading test. The mechanical characteristic parameters and insulation medium state parameters of the circuit breaker were collected in real time and synchronously to form a time-domain aligned performance degradation dataset.

[0008] The electrothermal stress intensity and duration in the test cycle are dynamically adjusted using the performance degradation dataset to generate adaptive cycle control commands that conform to the target aging curve.

[0009] The adaptive cyclic control command is executed until the predetermined test cycle is completed, and the aging test results of the pole-mounted circuit breaker are output, including the life assessment curve and the reliability degradation report.

[0010] Another embodiment of this application provides an aging test system for pole-mounted circuit breakers, the system comprising:

[0011] The acquisition module is used to acquire historical operating condition data and real-time load characteristics of pole-mounted circuit breakers, and generate a basic input set for accelerated aging tests that includes current load distribution and opening and closing frequencies.

[0012] The construction module is used to construct a non-uniform cyclic electrothermal stress loading spectrum based on the accelerated aging test basic input set, and generate an accelerated aging loading strategy that integrates dynamic current, opening and closing action and interval time.

[0013] The test module is used to conduct electrothermal cyclic loading tests on pole-mounted circuit breakers according to the accelerated aging loading strategy, and to collect the mechanical characteristic parameters and insulation medium state parameters of the circuit breaker in real time to form a time-domain aligned performance degradation dataset.

[0014] The adjustment module is used to dynamically adjust the intensity and duration of electrothermal stress in the test cycle using the performance degradation dataset, and generate adaptive cycle control commands that conform to the target aging curve.

[0015] The output module is used to execute the adaptive cyclic control command until the predetermined test cycle is completed, and output the aging test results of the pole-mounted circuit breaker, including the life assessment curve and the reliability degradation report.

[0016] Another embodiment of this application provides a storage medium storing a computer program, wherein the computer program is configured to execute the method described in any of the preceding claims when running.

[0017] Another embodiment of this application provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the method described in any of the preceding claims.

[0018] Compared with existing technologies, the aging test method for pole-mounted circuit breakers provided by this invention can improve the accuracy and efficiency of aging tests, and provide a scientific basis for the life assessment and reliability prediction of pole-mounted circuit breakers. Attached Figure Description

[0019] Figure 1 A hardware structure block diagram of a computer terminal for an aging test method for pole-mounted circuit breakers provided in an embodiment of the present invention;

[0020] Figure 2A flowchart illustrating an aging test method for a pole-mounted circuit breaker provided in an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of the structure of an aging test system for pole-mounted circuit breakers provided in an embodiment of the present invention. Detailed Implementation

[0022] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0023] The present invention first provides an aging test method for pole-mounted circuit breakers, which can be applied to electronic devices, such as computer terminals, specifically ordinary computers.

[0024] The following detailed explanation uses a computer terminal as an example. Figure 1 This is a hardware structure block diagram of a computer terminal for an aging test method for pole-mounted circuit breakers provided in an embodiment of the present invention. Figure 1 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0025] See Figure 2 The present invention provides an aging test method for pole-mounted circuit breakers, which may include the following steps:

[0026] S201: Obtain historical operating condition data and real-time load characteristics of the pole-mounted circuit breaker, and generate a basic input set for accelerated aging test that includes current load distribution and opening and closing frequency.

[0027] Specifically, the power monitoring system can collect the operation logs and SCADA data of the pole-mounted circuit breaker in the past set period, and extract historical raw data including the effective value of three-phase current, peak current, voltage waveform and opening and closing timestamps.

[0028] The core of this step is to comprehensively capture all operational data from the circuit breaker, providing a complete data source for subsequent analysis. The specific implementation method is as follows:

[0029] The data collection period is set to the past year (365 days), covering typical operating conditions such as different seasons and peak and off-peak loads to ensure data representativeness. Data acquisition relies on the remote data acquisition capabilities of the power monitoring system, simultaneously retrieving circuit breaker operation logs and SCADA (Supervisory Control and Data Acquisition) system data. These two types of data complement each other: operation logs record event-related information such as equipment fault alarms, manual operations, and maintenance records; SCADA data provides continuous monitoring values ​​of electrical and status quantities. The sampling frequency is set to 1 minute / time to balance data granularity and storage pressure.

[0030] The extracted core data items are categorized into electrical parameters and event parameters: electrical parameters include the effective value of three-phase current (unit: A, accuracy: ±0.1A), peak current (unit: A, taking the maximum value within each sampling period), and three-phase voltage waveform (sampling points: 1024 points / period, frequency: 50Hz); the event parameters focus on extracting the opening and closing timestamps (accurate to milliseconds, format: YYYY-MM-DDHH:MM:SS.XXX), while also associating them with the reasons for the opening and closing operations (fault tripping, normal switching, manual operation).

[0031] In the example, data from a pole-mounted circuit breaker is collected: SCADA data shows that at 14:30 on July 15, 2025 (peak load period), the effective value of the A-phase current was 820A, and the peak current was 1250A; the effective value of the B-phase current was 810A, and the peak current was 1230A; the effective value of the C-phase current was 815A, and the peak current was 1240A. The voltage waveform was a standard sine wave with no distortion. The operation log records a tripping operation at 14:35:20.123 on the same day, with a timestamp of 2025-07-15 14:35:20.123. The reason for the operation was a line overload protection trip, and the circuit breaker was closed and power restored at 14:40:15.345. All extracted data is temporarily stored in the format of "timestamp-data type-value-related event" to generate the initial historical raw dataset.

[0032] The historical raw data is cleaned and preprocessed to remove abnormal and invalid data points, and the current data is aligned and correlated with the opening and closing events based on the timestamp to generate a normalized historical dataset.

[0033] The core of this step is to eliminate data noise and bias, establish a time-series correlation between electrical data and event data, and improve data quality. The specific implementation method is as follows:

[0034] Data cleaning employs a classification strategy: For numerical data such as current and voltage, outliers are removed using the 3σ criterion, which calculates the mean μ and standard deviation σ of the data. Data exceeding the range [μ-3σ, μ+3σ] are identified as outliers and replaced with the linear interpolation result of two adjacent valid data points. Invalid data (such as null values ​​due to sampling failure or data format errors) are treated as missing values. Short-term missing data (≤5 consecutive sampling points) are filled using linear interpolation, while long-term missing data (>5 consecutive sampling points) are marked as data gaps and avoided in subsequent analyses.

[0035] In the example, three consecutive missing values ​​appeared in the phase A current data for a certain period. The valid data before and after were 820A (14:30) and 818A (14:33). Through linear interpolation, the corresponding values ​​for 14:31 were 819.33A and for 14:32 were 818.67A, and the missing values ​​were filled. Another data point of 1500A exceeded [μ-3σ,μ+3σ] (μ=780A, σ=120A, the range is 420A-1140A), and was judged as an outlier. It was replaced with the average of the adjacent data 820A and 818A, which was 819A.

[0036] Time alignment and correlation are based on millisecond-level timestamps to construct a time sequence correlation table. The opening and closing events are used as time nodes to correlate current and voltage data within 30 minutes before and after the event, clarifying the changes in electrical parameters before and after the event. At the same time, timestamp deviations are corrected. Time differences (≤1 second) caused by clock synchronization errors in different systems are calibrated using the SCADA system timestamp to ensure data timing consistency.

[0037] In the example, the opening and closing timestamp 2025-07-15 14:35:20.123 is used as the core node, and the current data from 14:05:20.123 to 15:05:20.123 is correlated, clearly showing the trend of continuously high current before opening, zero current after opening, and current gradually returning to the normal range after closing. After all cleaning and correlation operations are completed, a normalized historical dataset is generated, with the data arranged in chronological order, and each record containing complete electrical parameters and related event information.

[0038] Statistical analysis was performed using a normalized historical dataset to calculate the probability distribution of current duration in different load ranges and to count the frequency of opening and closing operations per unit time, thereby generating current load distribution characteristics and opening and closing frequency characteristics.

[0039] The core of this step is to extract key statistical features from the normalized data, quantify the circuit breaker's operating load and operational intensity, and provide a basis for subsequent loading strategy construction. The specific implementation method is as follows:

[0040] The current load distribution characteristic analysis first divides the load into intervals. Based on the rated current of the pole-mounted circuit breaker (set to 1250A in this example), the load is divided into three levels: high, medium, and low. The high load interval is 80%-100% of the rated current (1000A-1250A), the medium load interval is 40%-80% (500A-1000A), and the low load interval is 0%-40% (0A-500A). Based on the normalized dataset, the cumulative duration of each load interval is statistically analyzed, and the distribution probability is calculated using the formula P_i=T_i / T_total, where P_i is the distribution probability of the i-th interval, T_i is the cumulative duration of that interval, and T_total is the total data collection period (1 year, 8760 hours).

[0041] In the example, the cumulative duration of the high load range was 2190 hours, with a probability distribution P_high = 2190 / 8760 = 0.25; the cumulative duration of the medium load range was 4380 hours, with a probability distribution P_mid = 4380 / 8760 = 0.5; and the cumulative duration of the low load range was 2190 hours, with a probability distribution P_low = 2190 / 8760 = 0.25, forming a current load distribution characteristic: 25% for high load, 50% for medium load, and 25% for low load. Simultaneously, the peak current distribution characteristic was supplemented by statistically analyzing the frequency of peak currents ≥1200A within the high load range, serving as a basis for subsequent stress loading.

[0042] The frequency characteristics of circuit breaker opening and closing are statistically analyzed on a unit time dimension, calculating the daily average frequency, hourly average frequency, and peak period frequency (10:00-14:00, 17:00-21:00) separately. Manual opening and closing operations due to maintenance and repairs are excluded from the statistics, retaining only fault tripping and normal operation opening and closing events. The formula is F_h=N_total / T_total, where F_h is the hourly average frequency and N_total is the total number of effective opening and closing events.

[0043] In the example, there were a total of 438 valid opening and closing events within one year, with an hourly average frequency of F_h = 438 / 8760 = 0.05 times / hour (i.e., once every 20 hours on average); the daily average frequency was 1.2 times / day; during peak hours (8 hours per day), there were 219 opening and closing events, accounting for 50% of the total, which was significantly higher than during off-peak hours, forming the following opening and closing frequency characteristics: an hourly average of 0.05 times, concentrated during peak hours, accounting for 50%.

[0044] By combining the circuit breaker model and specifications, the current load distribution characteristics and the opening and closing frequency characteristics are normalized and structured, and finally a basic input set for accelerated aging tests is generated.

[0045] The core of this step is to transform statistical characteristics into standardized test inputs to match the parameter requirements of the accelerated aging test system. The specific implementation method is as follows:

[0046] Normalization is applied to characteristic parameters of different dimensions, mapping them to the [0,1] interval to eliminate the impact of dimensional differences on the subsequent loading strategy construction. The current load distribution probability is already dimensionless and requires no further normalization; the opening and closing frequencies are linearly normalized using the formula F_norm=(F-F_min) / (F_max-F_min), where F is the actual frequency, F_min is the minimum frequency (0 times / hour), and F_max is the statistically obtained maximum hourly frequency (0.2 times / hour in the example). In the example, the hourly average frequency of 0.05 times / hour is normalized to F_norm=(0.05-0) / (0.2-0)=0.25.

[0047] At the same time, supplementary basic parameters are added based on the circuit breaker model and specifications, including rated current, rated voltage, rated voltage of opening and closing coils, and insulation medium type, as auxiliary information for feature encapsulation, to ensure that the loading strategy is adapted to the performance limits of the equipment itself.

[0048] The structured encapsulation adopts a hierarchical structure, dividing the basic input set into three modules: a basic parameter module (model, rated current, rated voltage, etc.), a current load characteristic module (load range distribution probability, peak current distribution, normalized load coefficient), and a switching frequency characteristic module (hourly average frequency, peak period proportion, normalized frequency coefficient). The encapsulated data is stored in a standardized format to facilitate subsequent analysis and load spectrum construction. Each feature item is labeled with its statistical basis and unit to ensure traceability.

[0049] In the example, the final generated basic input set fragment includes: basic parameters (rated current 1250A, rated voltage 10kV); current load characteristics (high load 25%, medium load 50%, low load 25%, peak current ≥1200A 10%, normalized load factor 0.5); and opening / closing frequency characteristics (hourly average 0.05 times, peak percentage 50%, normalized frequency factor 0.25). This input set fully reflects the actual operating conditions of the circuit breaker, providing a core basis for constructing a practical accelerated aging loading strategy.

[0050] S202, Based on the basic input set of the accelerated aging test, construct a non-uniform cyclic electrothermal stress loading spectrum, and generate an accelerated aging loading strategy that integrates dynamic current, opening and closing action and interval time.

[0051] Specifically, it can parse the basic input set of accelerated aging test, determine the three typical load levels of high, medium and low and their corresponding duration probabilities based on the current load distribution characteristics, and generate a load level and proportion mapping table.

[0052] The core of this step is to extract the core features of the load from the basic input set, quantify the level classification and proportion, and provide a quantitative basis for the subsequent construction of the iterative model. The specific implementation method is as follows:

[0053] The analysis process first focuses on the current load distribution characteristics, combining the load range division and probability data from the basic input set, and simultaneously associating the circuit breaker's rated parameters (rated current 1250A, rated voltage 10kV in the example) to clarify the specific current amplitude range and corresponding duration probability for each level. The division must consider both actual operating conditions and accelerated aging requirements; high load levels select ranges that can trigger significant aging effects, while low load levels cover equipment standby and light load states, ensuring the mapping table closely matches real operating characteristics.

[0054] Based on the aforementioned basic input set data, the high load level is set to 80%-100% of the rated current (1000A-1250A), corresponding to a continuous probability of 25%; the medium load level is set to 40%-80% (500A-1000A), corresponding to a continuous probability of 50%; and the low load level is set to 0%-40% (0A-500A), corresponding to a continuous probability of 25%. Peak current parameters are also added: the peak current within the high load level is set to 1200A-1250A (accounting for 10% of the high load duration), the peak current for the medium load level is set to 800A-1000A, and the peak current for the low load level is set to 300A-500A, enhancing the targeting of electrothermal stress.

[0055] The generated load level and percentage mapping table must clearly define four core items: level name, current amplitude range, peak current range, and duration probability. The mapping table does not need to be presented concretely; instead, it uses structured data to describe the fixed relationships, providing a basis for subsequent cycle unit duration allocation and current settings. In the example, the mapping relationships are: high load (1000A-1250A, peak 1200A-1250A, percentage 25%), medium load (500A-1000A, peak 800A-1000A, percentage 50%), and low load (0A-500A, peak 300A-500A, percentage 25%). It is also noted that this mapping is based on one year of historical operating condition statistics to ensure traceability.

[0056] Based on the load level and proportion mapping table and the characteristics of opening and closing frequency, a non-uniform cyclic unit model is constructed. This model specifies the current loading amplitude sequence, the number of opening and closing triggers, and the holding time of each stage contained in each cyclic unit.

[0057] The core of this step is to construct the minimum cyclic unit, integrate load and opening / closing characteristics, and achieve basic modeling of non-uniform stress loading. The specific implementation method is as follows:

[0058] The duration of the non-uniform cyclic unit model needs to be calculated based on the frequency characteristics of circuit breaker opening and closing. The average hourly frequency of circuit breaker opening and closing in the basic input set mentioned earlier is 0.05 times (20 hours / time). To ensure that each cyclic unit includes complete circuit breaker opening and closing actions, the unit duration is set to 40 hours, corresponding to two circuit breaker opening and closing actions triggered within each unit (fitting the average frequency). The unit duration needs to balance acceleration efficiency and data acquisition completeness; too long a duration can lead to a prolonged test cycle, while too short a duration will fail to reflect the load continuity effect. 40 hours is the optimal value after balancing these factors.

[0059] Based on the load level percentage mapping table, the duration of each load phase is allocated as follows: total unit duration is 40 hours, high load phase duration = 40 × 25% = 10 hours, medium load phase duration = 40 × 50% = 20 hours, and low load phase duration = 40 × 25% = 10 hours. The current loading amplitude sequence is set according to phases. The high load phase uses a constant amplitude of 1100A (88% of rated current, in the middle of the high load range, balancing aging effects and equipment safety), interspersed with two peak currents of 1230A (each lasting 10 minutes, 5 hours apart); the medium load phase uses a constant amplitude of 750A (60% of rated current), interspersed with one peak current of 900A (lasting 15 minutes); the low load phase uses a constant amplitude of 300A (24% of rated current), with no additional peak current.

[0060] The number of circuit breaker opening and closing triggers is set to twice based on the unit duration. The triggering timing is reasonably arranged in conjunction with the load switching nodes to avoid triggering during peak current periods (to prevent stress superposition damage to equipment). The first circuit breaker opening and closing is set at the end of the high load phase and before the start of the medium load phase (10 hours). After opening, it is held for 5 minutes (simulating a fault isolation interval) before closing, and then smoothly transitions to the medium load amplitude. The second circuit breaker opening and closing is set at the end of the medium load phase and before the start of the low load phase (30 hours). After opening, it is held for 3 minutes before closing, transitioning to the low load amplitude. The circuit breaker opening and closing actions need to be associated with mechanical stress loading to ensure that the opening and closing speeds are consistent with the actual operating conditions (opening speed ≥ 1.2 m / s, closing speed ≥ 0.8 m / s).

[0061] In the example, the specific parameters of the non-uniform cyclic unit model are as follows: unit duration 40 hours, current loading amplitude sequence (0-10 hours: 1100A, 1230A peak triggered for 10 minutes each at the 2.5 and 7.5 hours; 10 hours 05 minutes-30 hours: 750A, 900A peak triggered for 15 minutes at the 20th hour; 30 hours 03 minutes-40 hours: 300A); circuit breaker tripping and closing triggered twice (tripping at the 10th hour, tripping and closing at the 10th hour 05; tripping at the 30th hour, tripping and closing at the 30th hour 03); the holding time of each stage is 10 hours for high load, 19 hours 55 minutes for medium load, and 9 hours 57 minutes for low load, fully integrating load, peak value and circuit breaker tripping and closing characteristics.

[0062] Based on the accelerated aging target, multiple non-uniform cyclic unit models are combined according to a specific rule to construct an electrothermal stress loading spectrum for a complete test cycle. The current amplitude in this loading spectrum exhibits time-varying characteristics with the opening and closing actions.

[0063] The core of this step is to combine cyclic units to construct a full-cycle loading spectrum, and simulate operating condition fluctuations through time-varying characteristics to achieve the goal of accelerated aging. The specific implementation method is as follows:

[0064] The accelerated aging target is set to compress the actual aging effect of 1 year (8760 hours) into a 1000-hour test cycle, with an acceleration factor of 8.76 (actual duration / test duration). To achieve this target, the stress intensity needs to be appropriately increased during unit assembly, while maintaining the operating condition sequence regularity to avoid overloading leading to abnormal equipment failure (deviation from the true aging mechanism). The assembly regularity adopts a "unit repetition + gradient reinforcement" mode, that is, first repeating the basic cycle unit, and then gradually increasing the high load amplitude and the frequency of opening and closing to simulate the cumulative aging effect during long-term operation of the equipment.

[0065] The complete test cycle is 1000 hours, calculated at 40 hours per unit, which can be divided into 25 cyclic units, divided into three stages: initial stage (units 1-10, 400 hours), intermediate stage (units 11-20, 400 hours), and final stage (units 21-25, 200 hours). In the initial stage, basic cyclic unit parameters are used without stress enhancement, allowing the equipment to adapt to the loading rhythm. In the intermediate stage, the high load amplitude is enhanced, increasing the constant high load amplitude from 1100A to 1150A, the peak current from 1230A to 1240A, and the number of circuit breakers / closes increases to 3 times per unit (shortening the interval to 13 hours / time). In the final stage, further enhancement is achieved, increasing the constant high load amplitude to 1200A, the peak current to 1250A, maintaining the number of circuit breakers / closes at 3 times per unit, while shortening the low load stage duration (reducing its proportion to 20%) and extending the high load proportion (increasing it to 30%).

[0066] The time-varying characteristics of the electrothermal stress loading spectrum are reflected in three dimensions: time-varying current amplitude (the high load amplitude gradually increases with each unit stage, and the amplitude switches according to each stage within each unit), time-varying opening and closing action (the frequency increases with each stage, and the triggering timing is staggered within the unit to avoid synchronous action throughout the entire cycle), and time-varying peak current (the amplitude gradually increases, and the duration is extended from 10 minutes to 15 minutes). During construction, it is necessary to ensure smooth connection between units, and that the current amplitude transitions with a linear slope (slope ≤ 5A / second) during load switching to avoid impact stress caused by sudden current changes. The triggering time difference for opening and closing actions in different units should be ≥ 1 hour to prevent mechanical stress concentration.

[0067] In the example, the complete loading spectrum segment is as follows: Initial stage, Unit 1 (0-40 hours): Loaded according to the basic model; Mid-stage, Unit 11 (400-440 hours): High load amplitude 1150A, peak 1240A (12 minutes each time), 3 opening and closing cycles (413 hours, 426 hours, 439 hours), medium load 20 hours, high load 12 hours, low load 8 hours; Final stage, Unit 21 (800-840 hours): High load amplitude 1200A, peak 1250A (15 minutes each time), 3 opening and closing cycles, medium load 20 hours, high load 12 hours, low load 8 hours. The overall loading spectrum shows a trend of gradually increasing stress, with current amplitude and opening / closing actions dynamically changing over time, closely matching the accelerated aging target and the evolution law of actual operating conditions.

[0068] The electrothermal stress loading spectrum is transformed into an executable sequence of control parameters, specifying the current setpoint, opening and closing commands, and interval time for each time step, ultimately generating a specific accelerated aging loading strategy.

[0069] The core of this step is to convert the loading spectrum into control parameters that the experimental system can recognize, refine the operation instructions, and ensure that the loading strategy is implemented effectively. The specific implementation method is as follows:

[0070] The conversion process begins by setting the time step. Considering the test system's response accuracy and data acquisition requirements, the time step is set to 1 second per step, balancing control accuracy and data storage pressure. Each time step's control parameters include three core items: current setpoint (unit: A, accuracy ±1A), circuit breaker / closing command (no command / opening / closing, trigger duration ±10ms), and interval time (holding time after circuit breaker / closing action, unit: seconds). During conversion, the stage duration, amplitude sequence, and circuit breaker / closing nodes in the loading spectrum need to be decomposed into second-by-second parameters, while simultaneously supplementing equipment protection parameters (such as overcurrent threshold of 1300A and circuit breaker / closing timeout threshold of 5 seconds).

[0071] The current setpoint transition is linear and phased. At load switching nodes (such as from high load to medium load), the current amplitude decreases to the target amplitude at a slope of 5A / second. The current value at each time step during the transition is calculated according to the slope. In the example, at the 10th hour (36000 seconds), the current switches from 1100A to 750A from high load to medium load, with a transition time of 70 seconds ((1100-750) / 5=70). From the 36000th to the 36069th second, the current value decreases by 5A per second from 1100A, stabilizing at 750A at the 36070th second. The peak current phase is set with a constant amplitude and the duration is accurate to the second. In the example, the peak current of 1230A lasts for 10 minutes (600 seconds), and the current setpoint is fixed at 1230A within the corresponding time step.

[0072] The conversion of opening and closing commands requires clearly defining the trigger time step, action duration, and interval. After the opening command is triggered, the opening state is maintained until the set interval ends, and then the closing command is triggered. The action duration is set to 2 seconds (1 second for opening and 1 second for closing), which conforms to the mechanical action characteristics of the circuit breaker. In the example, the opening command is triggered at the 10th hour (36000 seconds), the opening action is performed at time step 36000-36001 seconds, and the interval is 300 seconds (5 minutes). The closing command is triggered at the 36302nd second, and the closing action is performed at 36302-36303 seconds. After closing, the current transition phase begins. At the same time, a pre-warning parameter is added to the opening and closing commands to stabilize the current at the target amplitude 10 seconds before triggering, avoiding current fluctuations during operation.

[0073] The final accelerated aging loading strategy is encapsulated in a structured format, divided into four main modules: strategy description, control parameter sequence, protection mechanism, and execution flow. The strategy description clearly defines the acceleration target, loading cycle, and source of core parameters; the control parameter sequence stores second-by-second control instructions in time steps, which can be directly imported into the test system controller; the protection mechanism marks the abnormal handling logic such as overcurrent and timeout (e.g., immediate tripping and recording of fault nodes in case of overcurrent); the execution flow clearly defines the loading start conditions, parameter update method, and data synchronization frequency.

[0074] In the example, the loading strategy segment includes: Strategy description (acceleration factor 8.76, test cycle 1000 hours, time step 1 second); control parameter sequence (time step 36000: current 1100A, command to open, interval 300 seconds; time step 36001: current 1100A, command none, interval 299 seconds; ... time step 36302: current 750A, command to close, interval 0 seconds); protection mechanism (overcurrent threshold 1300A, timeout threshold 5 seconds, emergency tripping triggered in case of abnormality); execution process (30-minute warm-up before startup, parameter synchronization every 24 hours, real-time performance data collection). This strategy fully covers the entire test execution process, with parameters refined to each time step, ensuring the test system accurately executes the loading operation.

[0075] S203, according to the accelerated aging loading strategy, an electrothermal cyclic loading test is carried out on the pole-mounted circuit breaker, and the mechanical characteristic parameters and insulation medium state parameters of the circuit breaker are collected in real time to form a time-domain aligned performance degradation dataset.

[0076] Specifically, an electrothermal cyclic loading test platform can be built, the pole-mounted circuit breaker can be installed on the test bench, and a high current generator, an actuator driver, and a network of multiple types of sensors can be configured to obtain a ready test system.

[0077] The core of this step is to build an integrated test platform adapted to pole-mounted circuit breakers, realizing the coordination of electrothermal stress loading, action driving, and multi-parameter acquisition, providing hardware support for the test. The specific implementation method is as follows:

[0078] The test bench adopts a modular design, with the main body being a steel structure load-bearing platform equipped with an insulated support base (insulation strength ≥20kV to prevent ground discharge during testing). The bench has reserved mounting interfaces for circuit breakers and sensor installation points, taking into account the versatility of different equipment models. During installation, the pole-mounted circuit breaker is horizontally fixed to the base, and the circuit breaker's posture and wiring angle are adjusted to ensure symmetrical three-phase wiring and no mechanical interference during opening and closing operations. At the same time, a heat insulation protective layer is added to the circuit breaker housing to prevent high temperatures during loading from affecting surrounding equipment.

[0079] The core loading and drive equipment configuration must match the requirements of the accelerated aging loading strategy: the high current generator should be AC ​​type, with a rated output current of 0-2000A (covering the 1250A rated current and peak demand of the circuit breaker), current regulation accuracy of ±1A, response time ≤10ms, and can realize constant current loading, ramp loading and peak superposition loading according to control commands; the actuator driver should be adapted to the circuit breaker opening and closing coil (rated voltage 220V DC), with adjustable output power, and can accurately control the opening and closing speed (adjustment range 0.5-2.0m / s) and action sequence, and synchronously feedback the action completion signal.

[0080] A multi-type sensor network is laid out according to parameter type to ensure comprehensive capture of performance data: Mechanical characteristic acquisition is configured with two vibration sensors and two displacement sensors. The vibration sensor is attached to the circuit breaker operating mechanism housing (acquisition frequency 10kHz, range 0-50g, accuracy ±0.01g) to monitor mechanical vibration during opening and closing. The displacement sensor is laser-type and installed directly opposite the contacts (measurement range 0-50mm, accuracy ±0.01mm) to capture contact travel and speed in real time. Insulation characteristic acquisition is configured with a partial discharge detector and a dielectric loss meter. The partial discharge detector uses an ultra-high frequency sensor (detection frequency band 300MHz-1.5GHz, sensitivity ≤1pC) and is installed at the circuit breaker insulating bushing. The dielectric loss meter has an accuracy of ±0.0001 and is connected to the three-phase line through a dedicated terminal block to simultaneously measure the dielectric loss tangent of the insulating medium.

[0081] All devices are interconnected with the main control system via industrial Ethernet. The main control system is equipped with real-time data acquisition and control software, which supports multi-device collaborative linkage and real-time data storage. After the platform is built, debugging and calibration are carried out: the high current generator outputs a constant current of 1100A to calibrate the current measurement accuracy; drive the opening and closing actions 3 times to verify the effectiveness of sensor data acquisition and the accuracy of action timing, ensuring no data loss or signal distortion. After the debugging is qualified, the test system enters the ready state.

[0082] The test system executes loading cycles according to the accelerated aging loading strategy. At the same time, mechanical vibration signals, contact stroke and speed curves during the opening and closing process are collected by vibration sensors and displacement sensors. Partial discharge signals and dielectric loss tangents of the insulating medium are collected by partial discharge detectors and dielectric loss measuring instruments.

[0083] The core of this step is to execute loading according to a preset strategy, simultaneously collect mechanical and insulation characteristic parameters, and capture performance changes during the aging process. The specific implementation method is as follows:

[0084] The main control system imports the control parameter sequence of the accelerated aging loading strategy and drives each device to work together according to the time step: the high current generator realizes the smooth switching of high, medium and low load stages and the superposition of peak current based on the current setting value every second. During the loading process, the output current is monitored in real time. If it deviates from the set value (deviation > ±2A), the closed-loop regulation is automatically started to ensure the stability of the current. The actuator driver accurately controls the coil energization time and voltage according to the opening and closing command to realize the opening and closing action at the set speed. At the same time, the action completion status is fed back. If the action timeout (>5 seconds) or jamming occurs, an alarm is immediately triggered and the loading is suspended, and the fault node is recorded.

[0085] Mechanical characteristic parameters are acquired synchronously with loading and opening / closing actions: Vibration sensors continuously acquire signals throughout the entire loading cycle, focusing on capturing the vibration peak and frequency characteristics at the moment of opening / closing (1 second before and after the action), distinguishing the vibration characteristics corresponding to different faults such as mechanical structure wear and component loosening; Displacement sensors acquire contact stroke data in real time, store it at a frequency of 1kHz, calculate the contact movement speed through data differentiation, generate stroke-time curves and speed-time curves, and extract key indicators such as maximum opening / closing speed, number of closing bounces (allowed ≤2 times), and contact overtravel.

[0086] In the example, the opening and closing actions of the first cycle unit in the 10th hour were collected: the vibration sensor captured a peak vibration of 3.2g at the moment of opening, corresponding to a frequency of 120Hz, and a peak vibration of 2.8g at the moment of closing, corresponding to a frequency of 110Hz. There was no abnormal high-frequency vibration (excluding component breakage); the displacement sensor collected a contact stroke of 25mm, a maximum opening speed of 1.3m / s, a maximum closing speed of 0.9m / s, and a closing bounce of 1 time, with an overtravel of 5mm, all within the normal range.

[0087] The acquisition of insulation characteristic parameters takes into account the entire loading process and key nodes: the dielectric loss measuring instrument collects the dielectric loss tangent value every 5 minutes, focusing on monitoring parameter fluctuations during high load stages (temperature rise) and after opening and closing operations (insulation stress changes). Under normal operating conditions, the dielectric loss tangent value of the pole-mounted circuit breaker insulation medium is ≤0.005, and anything exceeding this value is marked as abnormal; the partial discharge detector continuously monitors, with a discharge threshold of 10pC set. When a discharge signal is detected, the discharge amplitude, frequency, occurrence time, and corresponding load status are recorded to distinguish between normal partial discharge and abnormal discharge caused by insulation degradation.

[0088] In the example, during the high load phase (1100A, lasting 2 hours), the dielectric loss tangent value remained stable at 0.0032-0.0035 with no significant fluctuations. During the peak current loading of 1230A, the partial discharge detector captured two weak discharge signals with an amplitude of 12pC and a frequency of 1 time / minute. These were determined to be slight discharges under normal load with no risk of insulation degradation. The corresponding timestamps and current parameters were recorded simultaneously.

[0089] Add a unified time stamp to all collected mechanical characteristic parameters and insulating medium state parameters to ensure that all data streams are strictly synchronized in the time dimension, and generate raw performance data streams with time stamps;

[0090] The core of this step is to achieve time-domain alignment of multi-source data streams, eliminate time skew, and provide a time-consistent data foundation for subsequent periodic analysis of performance degradation. The specific implementation method is as follows:

[0091] The unified time stamp is generated using a GPS-synchronized clock with millisecond-level accuracy (±1ms). The main control system acts as the time stamp reference node, distributing the unified time stamp in real-time to all sensors, data acquisition devices, and actuators via a clock synchronization protocol. This ensures strict clock synchronization across all devices and avoids data timing deviations caused by clock drift. The time stamp format is YYYY-MM-DDHH:MM:SS.XXX, precisely marking the acquisition time of each data point.

[0092] During data acquisition, each device binds the acquired parameter values ​​with real-time timestamps and uploads them to the main control system: vibration signals and displacement signals are bound point-by-point according to the acquisition frequency (10kHz, 1kHz), dielectric loss tangent values ​​are bound according to the acquisition cycle (5 minutes), and partial discharge signals are bound according to the occurrence time. Simultaneously, the trigger timestamps and completion timestamps of the opening and closing action commands are also synchronously associated with the corresponding data streams. After receiving the data, the main control system categorizes and stores it according to device type and parameter type, constructing a time-series data queue.

[0093] The time-scale synchronization verification mechanism operates synchronously. The main control system periodically (every hour) extracts data from different devices using the same time-scale for comparison and calculates the time deviation. If the time-scale deviation of a device's data from the reference time-scale is greater than 3ms, clock calibration is automatically triggered to ensure data timing consistency throughout the entire test cycle. Simultaneously, for critical action nodes such as opening and closing, the timing of the action trigger is verified to match the peak time-scale of the corresponding vibration and displacement signals. In the example, the time-scale for the opening command in the 10th hour is 2025-10-01 10:00:00.000, the vibration peak time-scale is 2025-10-01 10:00:00.050, and the displacement signal abrupt change time-scale is 2025-10-01 10:00:00.048. The deviation is ≤2ms, meeting the synchronization requirements.

[0094] All data bound to a unified time stamp are integrated into a raw performance data stream in time series. The data stream contains four main modules: time stamp information, mechanical characteristic parameters (vibration amplitude / frequency, contact stroke / velocity), insulation characteristic parameters (dielectric loss tangent, partial discharge data), and equipment status information (loaded current, opening and closing status). The data format is standardized, and each record contains complete time stamp and parameter association information to avoid data fragmentation and lay the foundation for subsequent processing.

[0095] The raw performance data stream with time stamps is segmented and features are extracted according to the test cycle. The characteristic statistics of mechanical and insulation parameters in each cycle are calculated to form a performance degradation dataset arranged in time series.

[0096] The core of this step is to process the raw data stream, extract periodic performance features, quantify performance degradation trends, and generate a structured dataset. The specific implementation method is as follows:

[0097] Data segmentation is performed according to the duration of the test cycle unit (40 hours). The main control system divides the raw performance data stream into data segments corresponding to the cycle unit based on the time stamp. Each data segment corresponds to one test cycle, and the segmentation node is precisely aligned with the cycle unit switching time (e.g., 0-40 hours for cycle 1, 40-80 hours for cycle 2). After segmentation, the integrity of each unit data segment is checked. If there is missing data (missing rate > 5%), it is supplemented by interpolation of adjacent valid data. If the missing rate is too high (> 20%), the data of that cycle unit is marked as invalid, and the corresponding cycle test needs to be repeated.

[0098] Feature extraction employs time-domain feature analysis to extract statistical features from mechanical and insulation characteristic parameters, quantifying the performance status of each cycle unit. Mechanical characteristic feature extraction targets vibration signals and contact motion parameters: vibration signals are extracted to obtain four statistical quantities—peak value, mean value, root mean square value, and peak factor—reflecting mechanical vibration intensity and stability; contact motion parameters are extracted to obtain maximum opening and closing speeds, average speeds, stroke deviations, and closing bounce counts, reflecting mechanical action accuracy and structural condition.

[0099] In the example, the mechanical characteristics of the first cycle unit are as follows: vibration signal peak value 3.5g, mean value 0.8g, root mean square value 1.2g, peak factor 2.9; maximum opening speed 1.3m / s, average speed 0.7m / s; maximum closing speed 0.9m / s, average speed 0.5m / s; stroke deviation 0.1mm; closing bounce 1 time; all characteristic quantities are within the normal reference range (the reference value is set based on the test data of the new equipment).

[0100] Insulation characteristic feature extraction targets the dielectric loss tangent and partial discharge signals: For the dielectric loss tangent, the average value, maximum value, and fluctuation range (maximum value - minimum value) are extracted to reflect the stability of the insulation dielectric loss; for the partial discharge signal, the total number of discharges, maximum discharge amplitude, and average discharge amplitude are extracted to reflect the insulation state. In the example, the insulation characteristic statistics for the first cycle unit are as follows: average dielectric loss tangent value: 0.0034, maximum value: 0.0036, fluctuation range: 0.0002; total number of partial discharges: 8, maximum discharge amplitude: 15 pC, average discharge amplitude: 12 pC, no abnormal discharge characteristics.

[0101] The characteristic statistics of all cycle units are arranged in order of cycle number to construct a performance degradation dataset. The dataset contains four core parts: cycle number, mechanical characteristic feature group, insulation characteristic feature group, and data validity label. Each characteristic is labeled with its statistical basis and unit, and is also associated with the corresponding cycle unit's loading parameters (current amplitude, number of opening and closing cycles), forming a "loading parameter-performance characteristic" relationship. The dataset is stored in time series, clearly showing the changing trend of each characteristic with the number of test cycles, providing core data support for subsequent dynamic adjustment of loading strategies and life assessment.

[0102] S204, using the performance degradation dataset, dynamically adjust the electrothermal stress intensity and duration in the test cycle to generate an adaptive cycle control command that conforms to the target aging curve;

[0103] Specifically, trend analysis can be performed on the performance degradation dataset to establish a degradation trajectory model of key performance parameters relative to the number of test cycles;

[0104] The core of this step is to extract key degradation features from the dataset, quantify the performance variation with the number of iterations through mathematical modeling, and provide a baseline trajectory for subsequent bias assessment. The specific implementation method is as follows:

[0105] First, key performance parameters were screened. Based on the aging failure mechanism of pole-mounted circuit breakers, 2-3 sensitive parameters were selected from both mechanical and insulation characteristics as core indicators. For mechanical characteristics, the maximum opening speed and the root mean square value of the vibration signal were selected; for insulation characteristics, the mean tangent of the dielectric loss angle and the maximum partial discharge amplitude were selected. These parameters accurately reflect the core trends of mechanical structure wear and insulation medium degradation. After screening, invalid cyclic data within the dataset were removed, and the key parameter sequences of valid cycles were retained and arranged in order of cycle number (1 to 25 times).

[0106] Trend analysis employs a time-domain trend fitting method, selecting an appropriate model based on parameter degradation characteristics: mechanical parameters (opening speed, root mean square of vibration) exhibit a linear degradation trend (accumulated wear leads to uniform performance decline), and a univariate linear regression model is used for modeling; insulation parameters (dielectric loss tangent, partial discharge amplitude) exhibit an exponential degradation trend (accelerated decay in the later stages of deterioration), and an exponential fitting model is used for modeling. The linear regression model is expressed as y = a × x + b, where x is the number of cycles, y is the parameter value, a is the degradation rate coefficient (negative values ​​indicate performance degradation), and b is the initial parameter value; the exponential fitting model is expressed as y = b × e^(k × x), where k is the exponential degradation coefficient (positive values ​​indicate accelerated deterioration), and b is the initial parameter value.

[0107] The least squares method is used to solve for the model parameters during the fitting process. The least squares method determines the optimal parameters by minimizing the sum of squared residuals between the fitted values ​​and the actual values. The formula for the sum of squared residuals is: ,in These are actual parameter values. These are the fitted values ​​for the model. A smaller value indicates a higher goodness of fit, and the required goodness of fit is... (Ensure the model can explain more than 90% of trend changes).

[0108] In the example, modeling is based on data from the first 10 cycles: the actual sequence of maximum tripping speeds (mechanical parameters) is 1.3, 1.29, 1.28, 1.27, 1.26, 1.25, 1.24, 1.23, 1.22, 1.21 m / s, and the fitted linear model is y = -0.01 × x + 1.31, R0. 2=0.99, a=-0.01 (decreases by 0.01 m / s per cycle), b=1.31 (initial velocity); the actual sequence of the mean tangent of dielectric loss angle (insulation parameter) is 0.0034, 0.0035, 0.0037, 0.0039, 0.0042, 0.0045, 0.0049, 0.0054, 0.0060, 0.0067, and the fitted exponential model is y=0.0033×e^(0.05×x), R 2 =0.98, k=0.05, b=0.0033. All key parameter models are integrated into a real-time degradation trajectory model, clearly showing the degradation pattern of each parameter with the number of cycles.

[0109] The real-time degradation trajectory model is compared with the preset target aging curve to calculate the deviation between the current aging rate and the target rate, and to generate an aging state deviation assessment result.

[0110] The core of this step is to quantify the deviation in aging progress through trajectory comparison, providing a basis for subsequent parameter adjustments and ensuring that the experiment conforms to the accelerated aging target. The specific implementation method is as follows:

[0111] The preset target aging curve is based on the accelerated aging target (1000 hours of testing equivalent to 1 year of actual aging), and is derived by combining the new equipment's baseline parameters and actual aging patterns. The target curve and the real-time degradation trajectory model use the same mathematical form. The linear parameter target model is set according to the equivalent aging rate, and the exponential parameter target model is set according to the equivalent exponential coefficient. In the example, the target curve for the maximum tripping speed is y = -0.012 × x + 1.31 (target degradation rate 0.012 m / s / cycle, slightly faster than the actual rate, with room for adjustment); the target curve for the mean tangent of the dielectric loss angle is y = 0.0033 × e^(0.055 × x) (target exponential coefficient 0.055, slightly higher than the actual rate).

[0112] Deviation calculation is performed in two steps: First, calculate the absolute deviation between the current aging rate and the target rate for each key parameter. For linear parameters, the rate deviation is Δa = |a_actual - a_target|, and for exponential parameters, the rate deviation is Δk = |k_actual - k_target|. Then, calculate the parameter value deviation by taking the relative deviation between the actual fitted value of the current cycle (e.g., the 10th cycle) and the target curve value. The formula is δ = |y_actual - y_target| / y_target × 100%. The relative deviation reflects the degree to which the current aging progress deviates from the target.

[0113] In the example, the deviation calculation for the 10th cycle is as follows: actual degradation rate of the tripping speed a_actual = -0.01, target a_target = -0.012, Δa = 0.002 m / s / cycle; current fitted value y_actual = 1.21 m / s, target value y_target = -0.012 × 10 + 1.31 = 1.19 m / s, relative deviation δ = |1.21 - 1.19| / 1.19 × 100% ≈ 1.68%. The actual tangent of the medium loss angle is k_actual=0.05, the target is k_target=0.055, and Δk=0.005; the current fitted value is y_actual=0.0067, the target value is y_target=0.0033×e^(0.055×10)≈0.0062, and the relative deviation is δ=|0.0067-0.0062| / 0.0062×100%≈8.06%.

[0114] The aging condition deviation assessment results are characterized by a comprehensive deviation index. The comprehensive deviation is calculated by combining the weights of each parameter (both mechanical and insulation parameters have a weight of 0.5), using the formula ε = 0.5 × δ_mech + 0.5 × δ_ins, where δ_mech is the average relative deviation of the mechanical parameters and δ_ins is the average relative deviation of the insulation parameters. In the example, δ_mech = 1.68%, δ_ins = 8.06%, and the comprehensive deviation ε = 0.5 × 1.68% + 0.5 × 8.06% ≈ 4.87%. The direction of the deviation is also clearly defined: the actual values ​​of both mechanical and insulation parameters are greater than the target values, indicating that the current aging rate is slower than the target rate, and the aging progress is lagging. The final assessment result includes the deviation values ​​of each parameter, the comprehensive deviation index, the direction of the deviation, and a cause analysis (insufficient current stress intensity leads to slower aging).

[0115] Based on the aging state deviation assessment results, the control parameters in subsequent test cycles are dynamically corrected using a feedback control algorithm. If the aging is too slow, the current stress intensity is increased or the cycle interval is shortened. If the aging is too fast, the stress is appropriately reduced, and the control parameter adjustment amount is generated.

[0116] The core of this step is to achieve adaptive parameter correction through feedback control, balancing the aging rate and equipment safety, and ensuring that the test conforms to the target curve. The specific implementation method is as follows:

[0117] The feedback control algorithm employs a proportional-integral-derivative (PID) control algorithm to meet the dynamic adjustment requirements of aging tests. The PID algorithm uses the proportional, integral, and derivative actions to collaboratively correct parameters and output the control parameter adjustment amount, expressed as ΔU = Kp × ε + Ki × ∫εdt + Kd × dε / dt, where ΔU is the control parameter adjustment amount, Kp is the proportional coefficient (amplifying the deviation signal), Ki is the integral coefficient (eliminating static deviation), and Kd is the derivative coefficient (suppressing overshoot). Based on the test characteristics, the PID parameters are set as follows: Kp = 5.0 (fast response to deviation), Ki = 0.1 (slow elimination of static deviation), and Kd = 0.5 (stable adjustment process).

[0118] The control parameter correction priority is ordered according to the degree of impact on the aging rate: first adjust the high load current amplitude (which has the most significant impact on aging), then adjust the high load duration, and finally adjust the cycle interval (to avoid frequent changes in the cycle affecting data continuity). The correction rules are clear: when the overall deviation ε > 3% and aging is slow, increase the high load current amplitude (each adjustment ≤ 50A, not exceeding the rated current of 1250A); when ε is between 1% and 3%, extend the high load duration (each adjustment ≤ 10%); when aging is fast (ε is negative and its absolute value > 2%), decrease the high load current amplitude (each adjustment ≤ 30A) or shorten the high load duration.

[0119] In the example, the overall deviation ε = 4.87% > 3%, indicating slow aging. Therefore, the high-load current amplitude is adjusted first. The current constant high-load amplitude is 1100A (before the mid-term enhancement). The adjustment amount is calculated using the PID algorithm: ΔU = 5.0 × 4.87% + 0.1 × ∫4.87%dt (the integral value of the first 10 cycles is 0.487) + 0.5 × dε / dt (the deviation change rate is 0.2% / cycle) ≈ 0.2435 + 0.0487 + 0.001 ≈ 0.2932. The corresponding current adjustment ΔI = 0.2932 × 50 ≈ 14.66A, rounded to 15A. Therefore, the constant high-load amplitude is adjusted from 1100A to 1115A, while maintaining a synchronous increase of 15A in the peak current (from 1230A to 1245A). The high-load duration remains unchanged at 10 hours, and the cycle interval remains 40 hours.

[0120] If aging is too rapid (example: overall deviation ε = -3.2%, actual aging is faster than the target), then reduce the high-load current amplitude. The calculated adjustment amount ΔI = -3.2% × 5.0 × 30 ≈ -4.8A, rounded to -5A, reduces the high-load amplitude from 1150A to 1145A. Simultaneously, shorten the high-load duration by 10% (from 12 hours to 10.8 hours) to slow the aging rate. All corrected parameters must be verified to meet equipment limits (current not exceeding 2000A, opening and closing speed between 0.5-2.0m / s) to ensure safe and feasible adjustments. Finally, generate a list of control parameter adjustments, clearly defining the adjustment items, original parameters, new parameters, and adjustment basis.

[0121] Based on the adjustment of control parameters, the subsequent electrothermal stress loading strategy to be executed is updated online, generating adaptive cyclic control instructions for real-time control of the test system.

[0122] The core of this step is to transform the parameter adjustment values ​​into an updated loading strategy, generate executable instructions, and realize online adaptive control of the experimental process. The specific implementation method is as follows:

[0123] The online update process adjusts the core parameters of the loading strategy for subsequent loop units (such as the 11th loop and later) according to the control parameter adjustment amount: high load current amplitude, peak current, high load duration, etc., while maintaining the timing consistency and operational safety of the strategy. During the update, the second-by-second data of the control parameter sequence must be corrected synchronously to ensure that the current setpoint, opening and closing commands at each time step match the adjusted parameters, and the current transition slope during load switching remains at 5A / second to avoid impact caused by sudden current changes.

[0124] In the example, the update strategy for cycles 11-20 (mid-term phase) is as follows: the original high load amplitude of 1150A is increased by 15A to 1165A; the original peak current of 1240A is updated to 1255A, and the peak duration is extended from 12 minutes to 13 minutes (simultaneously enhancing the aging effect); the high load duration remains 12 hours, the number of opening and closing operations remains 3 times / unit, and the trigger timing is fine-tuned (avoiding peak current periods). The control parameter sequence is updated synchronously: the second-by-second current setting value for the high load phase of cycle 11 (400-412 hours) is corrected from 1150A to 1165A, the second-by-second setting value for the peak current periods (406 hours, 410 hours) is corrected to 1255A, and the current value for the transition phase is recalculated based on the new amplitude.

[0125] The updated loading strategy needs to undergo two rounds of verification: first, a logic verification, checking whether the total duration of each stage after parameter adjustment is still 40 hours, whether the opening and closing commands and current loading are coordinated, and whether there are any timing conflicts; second, a safety verification, calculating the temperature rise of the equipment after adjustment (temperature rise ≤80K at high load 1165A, meeting the insulation medium tolerance requirements) and mechanical stress (opening and closing speed is still 1.2-1.3m / s, with no overload risk). After passing the verification, the updated strategy will be converted into adaptive cyclic control commands. The command format is consistent with the original loading strategy, including time steps, current setpoints, opening and closing commands, interval times, and protection parameters, and will be marked with an "adaptive update" label and an update version number.

[0126] After the control command is generated, it is immediately sent to the main control unit of the test system via industrial Ethernet. Upon receiving the command, the main control unit first compares it with the current execution status. If it is in a loop gap (no loading or opening / closing actions), the original command sequence is immediately replaced. If it is in the middle of a loop execution, it waits until the current stage ends (such as the end of the high-load stage) before switching to the updated command to avoid test interruption or data anomalies caused by mid-process switching. At the same time, the control parameter adjustment amount, the updated strategy, and the command issuance record are stored in the test log to achieve traceability of the adaptive adjustment process. In the example, after the 10th loop (400-hour node), the main control unit receives and switches to the updated 11th loop command, starts adaptive loading, and synchronously collects new parameters and incorporates them into the performance degradation dataset, forming a closed-loop control of "acquisition-modeling-correction-execution".

[0127] S205, execute the adaptive cyclic control command until the predetermined test cycle is completed, and output the aging test results of the pole-mounted circuit breaker, the results including the life assessment curve and the reliability degradation report.

[0128] Specifically, the test system receives and executes adaptive cyclic control commands to continuously perform electrothermal cyclic loading, while constantly incorporating newly acquired performance data into the performance degradation dataset to generate a cumulative performance degradation dataset for the entire test cycle.

[0129] The core of this step is to ensure the continuous execution of adaptive loading, and to synchronously complete the collection and integration of full-cycle performance data, providing a complete data source for subsequent lifetime assessment and reliability analysis. The specific implementation method is as follows:

[0130] After receiving the adaptive loop control command, the main control unit of the test system distributes the command to the high-current generator, actuator driver, and sensor network via industrial Ethernet, and advances the loading process using a "staged execution + real-time verification" mode. During execution, the main control unit monitors the operating status of each device in real time, including current loading accuracy, opening and closing action sequence, and sensor data acquisition validity. If abnormalities such as excessive current deviation, action stagnation, or data loss occur, the current loop is immediately paused, a local alarm is triggered, and fault information is recorded. After the fault is investigated and repaired, loading resumes from the abnormal node, ensuring the continuity of the test cycle.

[0131] The execution of adaptive commands strictly follows the updated loading strategy. For example, in the mid-term stage (cycles 11-20), the high-load current amplitude is adjusted to 1165A with a peak value of 1255A, and the circuit breaker is switched on and off 3 times per unit. In the final stage (cycles 21-25), the current is further strengthened to 1200A with a peak value of 1250A. During load switching in each stage, a current transition slope of 5A / second is maintained to avoid impact stress damage to the equipment. During the loading process, the sensor network continuously collects mechanical and insulation characteristic parameters at a predetermined frequency, updates the raw performance data stream every second, and synchronously binds it to a unified time scale to ensure the temporal consistency between the newly collected data and the historical data.

[0132] The data integration adopts an "incremental update + redundancy verification" mechanism. The newly collected characteristic statistics of each cycle (root mean square of vibration, tripping speed, mean tangent of dielectric loss angle, etc.) are appended to the performance degradation dataset in order of cycle number. At the same time, the deviation between the newly added data and the adjacent cycle data is calculated. If the deviation exceeds the normal fluctuation range (e.g., mechanical parameter deviation > 5%, insulation parameter deviation > 10%), it is marked as suspicious data. By tracing back the original data stream, it is confirmed whether it is a sudden change in equipment aging or a collection error. The error data is corrected, and the sudden change data is retained and marked as anomaly samples for subsequent reliability analysis.

[0133] After the complete test cycle (1000 hours), a cumulative performance degradation dataset is generated. This dataset covers all valid data from 25 cycle units and includes six modules: cycle number, equivalent running time (converted using an acceleration factor of 8.76, 1 hour of testing is equivalent to 8.76 hours of actual operation), mechanical characteristic feature groups, insulation characteristic feature groups, loading parameters (current amplitude, number of opening and closing cycles), and anomaly records. In the example, a segment of the cumulative dataset: Cycle 25 (960-1000 hours), equivalent actual running time 8760 hours (1 year), maximum opening speed 1.05 m / s, mean dielectric loss tangent 0.012, maximum partial discharge amplitude 35 pC, loading parameters: high load 1200 A, 3 opening and closing cycles, no anomaly records. The dataset is ultimately stored in a standardized format after integrity verification (missing rate <3%) and consistency verification (timescale synchronization deviation <3 ms), laying the foundation for subsequent analysis.

[0134] Based on the cumulative performance degradation dataset over the entire test cycle, an extrapolation algorithm is used to fit the curves of key performance parameters as a function of equivalent running time, predict the time when they reach the failure threshold, and generate a lifetime assessment curve.

[0135] The core of this step is to quantify the performance degradation trend through data extrapolation and predict the equipment failure time. The key lies in the selection of the algorithm and the setting of the failure threshold. The specific implementation method is as follows:

[0136] First, the failure thresholds for key performance parameters are clearly defined. Based on industry standards for pole-mounted circuit breakers and actual operating experience, the mechanical parameters are set with the maximum opening speed as the core indicator, and the failure threshold is set at 1.0 m / s (below this value, arc extinguishing performance cannot be guaranteed, leading to opening failure). The insulation parameters are set with the mean tangent of dielectric loss angle as the core indicator, and the failure threshold is set at 0.015 (exceeding this value results in excessive insulation dielectric loss, which can easily lead to breakdown faults). At the same time, the maximum amplitude of partial discharge of 30 pC is used as an auxiliary alarm threshold. If it is exceeded, it indicates accelerated insulation degradation.

[0137] The extrapolation algorithm is selected based on the parameter degradation characteristics. The mechanical parameter (opening speed) shows a linear degradation trend, so a linear extrapolation algorithm is used. Based on the linear regression model y=a×t_eq+b (t_eq is the equivalent running time) of the cumulative dataset, the full-cycle degradation curve is fitted, and the extrapolation is performed to t_eq when y equals the failure threshold, which is the predicted lifetime. The insulation parameter (dielectric loss tangent) shows an exponential degradation trend, so an exponential extrapolation algorithm is used. Based on the model y=b×e^(k×t_eq), the extrapolation is performed to t_eq corresponding to the failure threshold. At the same time, the confidence interval (confidence level 95%) is calculated to reflect the reliability of the prediction results.

[0138] The extrapolation process needs to eliminate outlier data points (such as parameter mutations caused by mid-experiment failures), and the least squares method should be used to optimize the fitting accuracy, requiring a goodness of fit R0. 2 ≥0.92. In the example, the cumulative data of the tripping speed is fitted to obtain a linear model y=-0.00003×t_eq+1.31, R 2 =0.93, extrapolating to y=1.0m / s, we calculate t_eq=(1.0-1.31) / (-0.00003)≈10333 hours (approximately 1.18 years); fitting the tangent of the dielectric loss angle yields the exponential model y=0.0033×e^(0.000012×t_eq), R 2 =0.94, extrapolating to y=0.015, t_eq=ln(0.015 / 0.0033) / 0.000012≈14267 hours (approximately 1.63 years).

[0139] Lifetime assessment curves are plotted separately for each parameter type. The horizontal axis represents equivalent operating time (hours), and the vertical axis represents parameter values. Each curve includes measured data points throughout the entire lifecycle, the fitted degradation curve, the extrapolated curve, the failure threshold line, and confidence intervals. Predicted lifetime values ​​are also labeled (the minimum predicted lifetime for each parameter is used to ensure safety; in this example, 10333 hours is used). The curves should clearly show the transition between the measured degradation stage and the extrapolated prediction stage. Confidence intervals are indicated by shading to reflect the extrapolation error range, making the lifetime prediction results intuitive and quantifiable.

[0140] By comprehensively analyzing the synergistic degradation relationship between mechanical and insulation properties, assessing the occurrence probability and evolution law of different failure modes, and generating a reliability degradation analysis report;

[0141] The core of this step is to uncover the multi-parameter collaborative degradation mechanism, quantify failure risk, and provide a key basis for equipment reliability assessment. The specific implementation method is as follows:

[0142] The synergistic degradation relationship analysis, based on a cumulative performance degradation dataset, employs correlation analysis to calculate the Pearson correlation coefficient between mechanical and insulation parameters. A coefficient with an absolute value ≥ 0.7 is considered a strong correlation, revealing the mutual influence between the two. In the example, the correlation coefficient between the degradation rate of the opening speed and the growth rate of the dielectric loss tangent is 0.78, showing a strong positive correlation. This indicates that mechanical wear (decreased opening speed) exacerbates the degradation of the insulation medium (increased dielectric loss) because contact wear leads to increased contact resistance, increased local temperature rise, and accelerated insulation aging. Conversely, partial discharge caused by insulation degradation can also erode the surface of mechanical components, accelerating wear and creating a vicious cycle.

[0143] Failure Mode and Effects (FMED) identification, combined with the aging mechanism of pole-mounted circuit breakers, identifies three typical failure modes: mechanical jamming (opening speed below the threshold), insulation breakdown (dielectric loss tangent exceeding the threshold or excessive partial discharge), and contact erosion (abnormally increased vibration signal accompanied by excessive travel deviation). Using Failure Mode and Effects Analysis (FMEA), and combining anomaly records and parameter degradation trends from the cumulative dataset, the probability of each mode's occurrence is assessed. Probability values ​​are calculated by statistically analyzing the number of warnings and parameter deviations for each mode throughout the entire cycle. Simultaneously, the transformation relationships between modes are analyzed (e.g., contact erosion can lead to mechanical jamming and insulation breakdown).

[0144] In the example, the reliability analysis results show that: the probability of mechanical jamming is 35%, mainly concentrated after 10,000 hours of equivalent operating time, with the risk increasing as the opening speed continues to decrease; the probability of insulation breakdown is 45%, with the risk accelerating after 8,000 hours, significantly affected by the combined effects of high load stress and mechanical wear; the probability of contact erosion is 20%, often accompanied by a root mean square vibration value exceeding 1.5g, which is a precursor to other failure modes. Regarding the evolution pattern, in the initial stage (0-4000 hours), the risk of each failure mode is low (<5%); in the middle stage (4000-8000 hours), the risk of contact erosion increases first, leading to an increase in the risk of mechanical and insulation failures; in the final stage (after 8000 hours), the risk of insulation breakdown becomes dominant, and the overall failure probability increases exponentially.

[0145] The reliability degradation analysis report presents the analysis results in a structured manner, including the synergistic degradation mechanism, failure mode list, probability assessment, evolution law, risk level classification (high / medium / low) and prevention and control recommendations. Each conclusion is supported by specific data from the accumulated dataset, ensuring the scientific validity and operability of the report.

[0146] Integrate life assessment curves and reliability degradation analysis reports to form a structured final test result document, along with a summary of the original data and key process descriptions, and output the aging test results of the pole-mounted circuit breaker.

[0147] The core of this step is to systematically integrate test data and analysis results to form standardized and traceable test documentation that meets engineering application and archiving requirements. The specific implementation method is as follows:

[0148] The structured final test results document adopts a hierarchical architecture, divided into five modules to ensure complete content and clear logic. The first module is the test summary, briefly describing the test object (circuit breaker model, rated parameters), test objective (accelerated aging equivalent to one year of actual operation), test cycle (1000 hours), and core conclusions (predicted lifespan, dominant failure mode). The second module is the lifespan assessment description, including lifespan assessment curves (combining text and graphics, labeling predicted lifespan, failure threshold, and confidence interval), extrapolation algorithm principles, parameter fitting process, and accuracy verification results. The third module is the reliability degradation analysis, fully referencing the previous analysis report and supplementing the assessment of the impact of each failure mode. The fourth module is the raw data summary, selecting core parameter values ​​for key cycles, corresponding equivalent operating times, and an anomaly record list to avoid raw data redundancy. The fifth module is the key process description, recording the number of adaptive control command adjustments (5 in the example), parameter adjustment amounts, fault handling processes, and equipment calibration records to ensure the traceability of the test process.

[0149] During document preparation, it is necessary to standardize terminology and parameter units, and to ensure standardized chart descriptions (clear labeling of the horizontal and vertical axes of curves, and explicit failure threshold lines). The conclusions should be objectively quantified to avoid vague expressions. Example document summary excerpt: "The test object is a pole-mounted circuit breaker with a rated current of 1250A and a rated voltage of 10kV. After 1000 hours of accelerated aging testing (equivalent to 8760 hours of actual operation), the predicted lifespan is 10333 hours (approximately 1.18 years). The dominant failure modes are mechanical jamming (35% probability) and insulation breakdown (45% probability). After 8000 hours, the risk of insulation degradation needs to be carefully controlled."

[0150] The original data summary selects the core parameters of typical cycles (cycles 1, 10, 20, and 25), annotates the equivalent running time and loading parameters, and details the reasons, adjustments, and effects of each adaptive adjustment in the key process description. For example, "In cycle 10, due to the slow aging rate, the high load current amplitude was adjusted from 1100A to 1115A. After the adjustment, the aging rate in cycle 11 increased by 18%, matching the target curve." The document is finally proofread and verified to ensure data consistency and rigorous conclusions. It is also exported in a standardized format, which can be directly used for equipment evaluation, archiving, or subsequent optimization design reference.

[0151] Another embodiment of the present invention provides an aging test system for pole-mounted circuit breakers, see [link to relevant documentation]. Figure 3 The system may include:

[0152] The acquisition module 301 is used to acquire historical operating condition data and real-time load characteristics of pole-mounted circuit breakers, and generate a basic input set for accelerated aging test that includes current load distribution and opening and closing frequency.

[0153] The construction module 302 is used to construct a non-uniform cyclic electrothermal stress loading spectrum based on the accelerated aging test basic input set, and generate an accelerated aging loading strategy that integrates dynamic current, opening and closing action and interval time.

[0154] Test module 303 is used to conduct electrothermal cyclic loading tests on pole-mounted circuit breakers according to the accelerated aging loading strategy, and to collect the mechanical characteristic parameters and insulation medium state parameters of the circuit breaker in real time to form a time-domain aligned performance degradation dataset.

[0155] The adjustment module 304 is used to dynamically adjust the intensity and duration of electrothermal stress in the test cycle using the performance degradation dataset, and generate an adaptive cycle control command that conforms to the target aging curve.

[0156] The output module 305 is used to execute the adaptive cyclic control command until the predetermined test cycle is completed, and output the aging test results of the pole-mounted circuit breaker, the results including the life assessment curve and the reliability degradation report.

[0157] This invention also provides a storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0158] This invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0159] Specifically, the aforementioned electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the aforementioned processor, and the input / output device is connected to the aforementioned processor.

[0160] The above description, based on the embodiments shown in the figures, details the structure, features, and effects of the present invention. The above description is only a preferred embodiment of the present invention, but the present invention is not limited to the scope of implementation shown in the figures. Any changes made in accordance with the concept of the present invention, or equivalent embodiments modified to have equivalent changes, that do not exceed the spirit covered by the specification and figures, should be within the protection scope of the present invention.

Claims

1. A method for aging test of pole-mounted circuit breakers, characterized in that, The method includes: Acquire historical operating condition data and real-time load characteristics of pole-mounted circuit breakers to generate a basic input set for accelerated aging tests that includes current load distribution and opening and closing frequencies. Based on the aforementioned accelerated aging test input set, a non-uniform cyclic electrothermal stress loading spectrum is constructed to generate an accelerated aging loading strategy that integrates dynamic current, opening and closing action and interval time. According to the accelerated aging loading strategy, the pole-mounted circuit breaker was subjected to an electrothermal cyclic loading test. The mechanical characteristic parameters and insulation medium state parameters of the circuit breaker were collected in real time and synchronously to form a time-domain aligned performance degradation dataset. The electrothermal stress intensity and duration in the test cycle are dynamically adjusted using the performance degradation dataset to generate adaptive cycle control commands that conform to the target aging curve. The adaptive cyclic control command is executed until the predetermined test cycle is completed, and the aging test results of the pole-mounted circuit breaker are output, including the life assessment curve and the reliability degradation report.

2. The method according to claim 1, characterized in that, The process involves acquiring historical operating condition data and real-time load characteristics of the pole-mounted circuit breaker to generate a basic input set for accelerated aging tests, including current load distribution and opening / closing frequencies. The power monitoring system collects the operation logs and SCADA data of the pole-mounted circuit breaker within the past set period, and extracts historical raw data including the effective value of three-phase current, peak current, voltage waveform and opening and closing timestamps. The historical raw data is cleaned and preprocessed to remove abnormal and invalid data points, and the current data is aligned and correlated with the opening and closing events based on the timestamp to generate a normalized historical dataset. Statistical analysis was performed using a normalized historical dataset to calculate the probability distribution of current duration in different load ranges and to count the frequency of opening and closing operations per unit time, thereby generating current load distribution characteristics and opening and closing frequency characteristics. By combining the circuit breaker model and specifications, the current load distribution characteristics and the opening and closing frequency characteristics are normalized and structured, and finally a basic input set for accelerated aging tests is generated.

3. The method according to claim 2, characterized in that, Based on the accelerated aging test input set, a non-uniform cyclic electrothermal stress loading spectrum is constructed to generate an accelerated aging loading strategy that integrates dynamic current, opening and closing actions, and interval duration, including: The basic input set of accelerated aging test is analyzed, and three typical load levels (high, medium and low) and their corresponding duration probabilities are determined according to the current load distribution characteristics. A load level and percentage mapping table is generated. Based on the load level and proportion mapping table and the characteristics of opening and closing frequency, a non-uniform cyclic unit model is constructed. This model specifies the current loading amplitude sequence, the number of opening and closing triggers, and the holding time of each stage contained in each cyclic unit. Based on the accelerated aging target, multiple non-uniform cyclic unit models are combined according to a specific rule to construct an electrothermal stress loading spectrum for a complete test cycle. The current amplitude in this loading spectrum exhibits time-varying characteristics with the opening and closing actions. The electrothermal stress loading spectrum is transformed into an executable sequence of control parameters, specifying the current setpoint, opening and closing commands, and interval time for each time step, ultimately generating a specific accelerated aging loading strategy.

4. The method according to claim 3, characterized in that, The electrothermal cyclic loading test is conducted on the pole-mounted circuit breaker according to the accelerated aging loading strategy. The mechanical characteristic parameters and insulation medium state parameters of the circuit breaker are collected in real time and synchronously to form a time-domain aligned performance degradation dataset, including: An electrothermal cyclic loading test platform was built, the pole-mounted circuit breaker was installed on the test bench, and a high current generator, an actuator driver, and a network of multiple types of sensors were configured to obtain a ready test system. The test system executes loading cycles according to the accelerated aging loading strategy. At the same time, mechanical vibration signals, contact stroke and velocity curves during the opening and closing process are collected by vibration sensors and displacement sensors. Partial discharge signals and dielectric loss tangents of the insulating medium are collected by partial discharge detectors and dielectric loss measuring instruments. Add a unified time stamp to all collected mechanical characteristic parameters and insulating medium state parameters to ensure that all data streams are strictly synchronized in the time dimension, and generate raw performance data streams with time stamps; The raw performance data stream with time stamps is segmented and features are extracted according to the test cycle. The characteristic statistics of mechanical and insulation parameters in each cycle are calculated to form a performance degradation dataset arranged in time series.

5. The method according to claim 4, characterized in that, The method of dynamically adjusting the electrothermal stress intensity and duration in the test cycle using the performance degradation dataset to generate adaptive cyclic control commands that conform to the target aging curve includes: Perform trend analysis on the performance degradation dataset and establish a degradation trajectory model of key performance parameters relative to the number of test cycles; The real-time degradation trajectory model is compared with the preset target aging curve to calculate the deviation between the current aging rate and the target rate, and to generate an aging state deviation assessment result. Based on the aging state deviation assessment results, the control parameters in subsequent test cycles are dynamically corrected using a feedback control algorithm. If the aging is too slow, the current stress intensity is increased or the cycle interval is shortened. If the aging is too fast, the stress is appropriately reduced, and the control parameter adjustment amount is generated. Based on the adjustment of control parameters, the subsequent electrothermal stress loading strategy to be executed is updated online, generating adaptive cyclic control instructions for real-time control of the test system.

6. The method according to claim 5, characterized in that, The adaptive cyclic control command is executed until the predetermined test cycle is completed, and the aging test results of the pole-mounted circuit breaker are output. The results include a life assessment curve and a reliability degradation report, including: The test system receives and executes adaptive cyclic control commands to continuously perform electrothermal cyclic loading, while constantly incorporating newly acquired performance data into the performance degradation dataset to generate a cumulative performance degradation dataset for the entire test cycle. Based on the cumulative performance degradation dataset within the complete test cycle, an extrapolation algorithm is used to fit the curves of key performance parameters changing with equivalent running time, predict the time when they reach the failure threshold, and generate a lifetime assessment curve. By comprehensively analyzing the synergistic degradation relationship between mechanical and insulation properties, assessing the occurrence probability and evolution law of different failure modes, and generating a reliability degradation analysis report; Integrate life assessment curves and reliability degradation analysis reports to form a structured final test result document, along with a summary of the original data and key process descriptions, and output the aging test results of the pole-mounted circuit breaker.

7. An aging test system for pole-mounted circuit breakers, characterized in that, The system includes: The acquisition module is used to acquire historical operating condition data and real-time load characteristics of pole-mounted circuit breakers, and generate a basic input set for accelerated aging tests that includes current load distribution and opening and closing frequencies. The construction module is used to construct a non-uniform cyclic electrothermal stress loading spectrum based on the accelerated aging test basic input set, and generate an accelerated aging loading strategy that integrates dynamic current, opening and closing action and interval time. The test module is used to conduct electrothermal cyclic loading tests on pole-mounted circuit breakers according to the accelerated aging loading strategy, and to collect the mechanical characteristic parameters and insulation medium state parameters of the circuit breaker in real time to form a time-domain aligned performance degradation dataset. The adjustment module is used to dynamically adjust the intensity and duration of electrothermal stress in the test cycle using the performance degradation dataset, and generate adaptive cycle control commands that conform to the target aging curve. The output module is used to execute the adaptive cyclic control command until the predetermined test cycle is completed, and output the aging test results of the pole-mounted circuit breaker, including the life assessment curve and the reliability degradation report.

8. The system according to claim 7, characterized in that, The acquisition module is specifically used for: The power monitoring system collects the operation logs and SCADA data of the pole-mounted circuit breaker within the past set period, and extracts historical raw data including the effective value of three-phase current, peak current, voltage waveform and opening and closing timestamps. The historical raw data is cleaned and preprocessed to remove abnormal and invalid data points, and the current data is aligned and correlated with the opening and closing events based on the timestamp to generate a normalized historical dataset. Statistical analysis was performed using a normalized historical dataset to calculate the probability distribution of current duration in different load ranges and to count the frequency of opening and closing operations per unit time, thereby generating current load distribution characteristics and opening and closing frequency characteristics. By combining the circuit breaker model and specifications, the current load distribution characteristics and the opening and closing frequency characteristics are normalized and structured, and finally a basic input set for accelerated aging tests is generated.

9. A storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the method of any one of claims 1-6 when it is run.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method of any one of claims 1-6.

Images