Self-adaptive protection setting method and system for primary and secondary fusion pole-mounted circuit breaker

Abstract

The invention discloses a self-adaptive protection setting method and system for primary and secondary fusion pole-mounted circuit breakers, and relates to the related field of power system emergency protection devices, and the method comprises the steps: obtaining a plurality of primary and secondary fusion pole-mounted circuit breakers; identifying a branch relation, and determining a circuit breaker group including a superior-subordinate relation; setting a short-time activation window, extracting abnormal operation monitoring parameters when the main circuit breaker is abnormal, carrying out recovery time window calculation on the abnormal operation monitoring parameters, and judging whether a linkage protection instruction is sent to the branch circuit breaker or not by comparing the recovery time window with the short-time activation window; if the recovery time window is smaller than the short-time activation window, the linkage protection instruction is not activated, and if the recovery time window is larger than the short-time activation window, the linkage protection instruction is activated. The technical problem of poor protection accuracy and economical efficiency of the existing protection setting for the primary and secondary fusion pole-mounted circuit breaker is solved, and the technical effect of improving the accuracy and economical efficiency of power grid protection is achieved.

Description

Technical Field

[0001] The present application relates to the field of emergency protection devices for power systems, and in particular to an adaptive protection setting method and system for a primary-secondary fusion pole-mounted circuit breaker. Background Art

[0002] The protection settings of integrated primary and secondary pole-mounted circuit breakers are crucial for ensuring grid reliability and power supply continuity. Currently, when a main circuit breaker anomaly is detected, protection commands are typically sent immediately to all branch circuit breakers to quickly isolate the faulty area. However, this approach fails to consider the recovery time of the main circuit breaker anomaly, meaning it cannot distinguish between transient and permanent faults. If the main circuit breaker recovers within a very short period of time, the previously issued protection commands will cause unnecessary branch circuit breaker activation, wasting protection resources and potentially triggering a chain reaction, such as power outages.

[0003] In the current related technologies, the protection setting for primary and secondary integrated pole-mounted circuit breakers has technical problems such as poor protection accuracy and economy. Summary of the Invention

[0004] The present application provides an adaptive protection setting method and system for a primary-secondary integrated pole-mounted circuit breaker, which obtains multiple circuit breaker devices, identifies the branch relationship between them and groups them, each group includes a main circuit breaker and several branch circuit breakers, and sets a short-time activation window time. When an abnormality is detected in the main circuit breaker, the abnormal operation parameters are extracted and the time window required for its recovery is calculated. This recovery time is compared with the preset short-time activation window. If the recovery time is shorter than the activation window, the system does not trigger the linkage protection and continues to operate; if the recovery time is longer than the activation window, a linkage protection instruction is immediately sent to the branch circuit breaker to disconnect the main circuit and branch circuit breakers synchronously. The present application solves the technical problems of poor protection accuracy and economy in the existing protection setting for primary-secondary integrated pole-mounted circuit breakers, and achieves the technical effect of improving the accuracy and economy of power grid protection.

[0005] The present application provides an adaptive protection setting method for a primary-secondary fusion pole-mounted circuit breaker, comprising: obtaining a plurality of primary-secondary fusion pole-mounted circuit breakers; identifying branch relationships between the plurality of primary-secondary fusion pole-mounted circuit breakers, and determining circuit breaker groups including a superior-subordinate relationship, wherein each circuit breaker group includes a main circuit breaker and at least one branch circuit breaker; setting a short-time activation window, extracting abnormal operation monitoring parameters of the main circuit breaker, performing a recovery time window calculation on the abnormal operation monitoring parameters, and determining whether to send a linkage protection instruction to the branch circuit breaker by comparing the recovery time window with the short-time activation window; wherein, if the recovery time window is less than the short-time activation window, the linkage protection instruction is not activated; and if the recovery time window is greater than the short-time activation window, the linkage protection instruction is activated.

[0006] In a possible implementation, the following processing is performed: each of the multiple primary-secondary fusion pole-mounted circuit breakers includes a primary breaking layer and a secondary intelligent control layer; wherein, the primary breaking layer is used to execute the breaking protection of the branch circuit breaker according to the activated linkage protection instruction, and the secondary intelligent control layer is used to perform abnormal operation monitoring and receive or send the linkage protection instruction.

[0007] In a possible implementation, the following processing is performed: a short-term activation window is set, and the short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker; wherein the historical abnormal event samples include abnormal event operation monitoring parameter samples of the main circuit breaker, abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend.

[0008] In a possible implementation, the short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker, and performing the following processing: screening effective recovery event samples of the historical abnormal event samples, wherein the effective recovery event samples are abnormal events marked as restored to normal; extracting the recovery time distribution of the effective recovery event samples, and analyzing the recovery time distribution to obtain a basic activation threshold; calculating a threshold adjustment factor based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend of the historical abnormal event samples; and updating the basic activation threshold according to the threshold adjustment factor to obtain an adaptive short-term activation window of the main circuit breaker.

[0009] In a possible implementation, the recovery time distribution of the effective recovery event sample is extracted, the recovery time distribution is analyzed to obtain a basic activation threshold, and the following processing is performed: density clustering analysis is performed according to the recovery time distribution to obtain multiple clustering results; the first clustering result is extracted according to the density value of each clustering result, and the quantile value of the first clustering result is used as the basic activation threshold.

[0010] In a possible implementation, a threshold adjustment factor is calculated based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend of the historical abnormal event samples, and the following processing is performed: a frequency adjustment factor, a continuous adjustment factor, and a trend adjustment factor are defined based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend; and a cumulative calculation is performed according to the frequency adjustment factor, continuous adjustment factor, and trend adjustment factor to comprehensively obtain the threshold adjustment factor.

[0011] In a possible implementation, a circuit breaker group including a superior-subordinate relationship is determined, and the following processing is further performed: if each circuit breaker group includes a main circuit breaker, at least one primary branch circuit breaker, and at least one secondary branch circuit breaker; a first short-time activation window and a second short-time activation window are set, and it is determined based on the first short-time activation window whether the main circuit breaker sends a primary linkage protection instruction to the primary branch circuit breaker, and based on the second short-time activation window whether the primary branch circuit breaker sends a secondary linkage protection instruction to the secondary branch circuit breaker.

[0012] In a possible implementation, the following processing is performed: the first short-time activation window is determined by adaptively analyzing the historical abnormal event samples of the main circuit breaker, and the second short-time activation window is determined by adaptively analyzing the historical abnormal event samples of the first-level branch circuit breaker; wherein, the multi-level linkage protection instruction consisting of the first-level linkage protection instruction and the second-level linkage protection instruction is sent sequentially in hierarchical order.

[0013] In a possible implementation, a recovery time window is calculated for the abnormal operation monitoring parameters, and the following processing is performed: an abnormal recovery time prediction model is constructed, and the abnormal recovery time prediction model is obtained by supervised training of historical abnormal event samples of the main circuit breaker; the abnormal operation monitoring parameters are input into the abnormal recovery time prediction model, and the predicted recovery time window is output.

[0014] The present application also provides an adaptive protection setting system for a primary-secondary fusion pole-mounted circuit breaker, comprising: a pole-mounted circuit breaker acquisition module, for acquiring a plurality of primary-secondary fusion pole-mounted circuit breakers; a circuit breaker group identification module, for identifying the branch relationships of the plurality of primary-secondary fusion pole-mounted circuit breakers, and determining a circuit breaker group including a superior-subordinate relationship, wherein each circuit breaker group includes a main circuit breaker and at least one branch circuit breaker; a recovery time window calculation module, for setting a short-time activation window, extracting abnormal operation monitoring parameters of the main circuit breaker, performing a recovery time window calculation on the abnormal operation monitoring parameters, and determining whether to send a linkage protection instruction to the branch circuit breaker by comparing the recovery time window with the short-time activation window; a protection instruction sending judgment module, for not activating the linkage protection instruction if the recovery time window is less than the short-time activation window, and activating the linkage protection instruction if the recovery time window is greater than the short-time activation window.

[0015] The adaptive protection setting method and system for primary and secondary fusion pole-mounted circuit breakers proposed in this application first obtains multiple primary and secondary fusion pole-mounted circuit breakers, then identifies the branch relationships of the multiple primary and secondary fusion pole-mounted circuit breakers, determines the circuit breaker groups containing the upper and lower relationships, each circuit breaker group includes a main circuit breaker and at least one branch circuit breaker, then sets a short-term activation window, extracts the abnormal operation monitoring parameters of the main circuit breaker, calculates the recovery time window for the abnormal operation monitoring parameters, and determines whether to send a linkage protection instruction to the branch circuit breaker by comparing the recovery time window with the short-term activation window. If the recovery time window is smaller than the short-term activation window, the linkage protection instruction is not activated. If the recovery time window is larger than the short-term activation window, the linkage protection instruction is activated. The technical effect of improving the accuracy and economy of power grid protection is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0017] Figure 1 A flow chart of an adaptive protection setting method for a primary-secondary integrated pole-mounted circuit breaker provided in an embodiment of the present application.

[0018] Figure 2 A schematic structural diagram of an adaptive protection setting system for a primary-secondary fusion pole-mounted circuit breaker provided in an embodiment of the present application.

[0019] Description of reference numerals: pole-mounted circuit breaker acquisition module 10 , circuit breaker group identification module 20 , restoration time window calculation module 30 , protection instruction sending judgment module 40 . DETAILED DESCRIPTION

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0021] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0022] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0023] The embodiment of the present application provides an adaptive protection setting method for a primary and secondary fusion pole mounted circuit breaker, such as Figure 1 As shown, the method includes: Step S100, obtain multiple primary and secondary fusion pole-mounted circuit breakers, each of the multiple primary and secondary fusion pole-mounted circuit breakers includes a primary fault layer and a secondary intelligent control layer, wherein the primary fault layer is used to execute the breaking protection of the branch circuit breaker according to the activated linkage protection instruction, and the secondary intelligent control layer is used to perform abnormal operation monitoring and receive or send linkage protection instructions.

[0024] Specifically, multiple primary and secondary integrated pole-mounted circuit breakers are connected to the master station system using power system communication networks (such as fiber-optic and wireless networks). Each pole-mounted circuit breaker is installed on a distribution network pole. Its location is determined by the geographical distribution of the distribution network and is identified in the master station system.

[0025] The primary circuit breaker includes components such as the operating mechanism and the arc extinguishing chamber. The operating mechanism uses mechanical transmission to open and close the circuit breaker. Its operating power can come from a spring or electromagnetic drive. For example, a spring-charged operating mechanism uses an electric motor to store spring energy. When the circuit breaker is required to open or close, the spring energy is released to actuate the circuit breaker. The arc extinguishing chamber is filled with an arc-extinguishing medium, such as sulfur hexafluoride (SF6), to extinguish the arc when the circuit breaker interrupts current.

[0026] The secondary intelligent control layer primarily consists of a microprocessor, memory, and communication modules. A microprocessor, such as an ARM chip, is responsible for running various control algorithms and monitoring programs. The memory is used to store information such as device parameters and historical operating data. The communication module supports multiple protocols (such as DL / T634.5104 and IEC61850) for data exchange with the master system.

[0027] In the master station system, a device profile is created for each integrated primary and secondary pole-mounted circuit breaker, including information such as device model, manufacturer, date of manufacture, installation location, and rated parameters (such as rated voltage and rated current). Software in the secondary intelligent control layer is initialized, loading basic monitoring and control programs. For example, the sampling rate for current and voltage monitoring is set, typically to hundreds to thousands of samples per second, to ensure accurate capture of abnormal operating signals.

[0028] Step S200: identifying the branch relationships of the plurality of primary and secondary fusion column mounted circuit breakers, and determining circuit breaker groups having a superior-subordinate relationship, each circuit breaker group including a main circuit breaker and at least one branch circuit breaker.

[0029] Specifically, the branch relationship in the distribution network refers to the connection between the main pole-mounted circuit breaker and the branch pole-mounted circuit breaker. The main pole-mounted circuit breaker is close to the substation outgoing line and provides power to the branch pole-mounted circuit breaker, which in turn supplies power to its respective load area.

[0030] Using the network topology analysis software in the master station system, combined with the primary wiring diagram of the distribution network and the connection relationship between the pole-mounted circuit breakers, the upper and lower level circuit breaker groups can be identified. For example, in a radial distribution network, the substation outgoing line is first connected to a primary and secondary integrated pole-mounted circuit breaker as the main circuit breaker, which is then connected to multiple pole-mounted branch circuit breakers.

[0031] In actual scenarios, by assigning a unique identification code to each pole-mounted circuit breaker (such as an equipment code based on power industry standards), the master station system can build a hierarchical relationship of circuit breaker groups based on these identification codes and the wiring information of the distribution network.

[0032] The master station system verifies the accuracy of the topology by communicating with the secondary intelligent control layer of each pole-mounted circuit breaker to obtain circuit breaker status information (such as open / close status, current level, etc.). For example, when the master station system sends a command to a branch circuit breaker, it observes its response and the impact on other related circuit breakers to verify the correctness of the branch relationship.

[0033] Step S300, set a short-time activation window, extract the abnormal operation monitoring parameters of the main circuit breaker, calculate the recovery time window for the abnormal operation monitoring parameters, and determine whether to send a linkage protection instruction to the branch circuit breaker by comparing the recovery time window and the short-time activation window.

[0034] Specifically, the secondary intelligent control layer uses built-in current and voltage sensors to collect real-time operating parameters of the main circuit breaker. These sensors, which can be electromagnetic or electronic transformers, convert high-voltage, high-current signals into low-voltage signals suitable for microprocessor processing. For example, electromagnetic current transformers, based on the principle of electromagnetic induction, convert the high primary current into a low secondary current signal at a specific ratio. The collected abnormal operation monitoring parameters undergo preprocessing, including filtering (such as using a low-pass filter to remove high-frequency interference signals) and data normalization (converting data of different dimensions to the same dimensional range, such as converting current and voltage data to per-unit values ​​relative to the rated value). Abnormal operation monitoring parameters, such as the amplitude and rate of change of current and voltage, reflect the normal operation of the circuit breaker. When these parameters exceed or fall below the normal range, it indicates an abnormal operating condition such as a short circuit or overcurrent.

[0035] A short-term activation window is set. This short-term activation window is a pre-set time interval used to determine whether a main circuit breaker anomaly is a transient disturbance or a persistent fault. In the distribution network protection strategy, the relationship between the anomaly recovery time and the short-term activation window is compared to determine whether to activate branch circuit breakers. The length of the short-term activation window can be set based on the specific conditions of the distribution network (such as power supply reliability requirements and the duration of common faults). For example, in a distribution network in a city core area with high power supply reliability requirements, the short-term activation window can be set to 100-300 milliseconds. The recovery time window is calculated by monitoring the recovery process of the main circuit breaker after an anomaly occurs. For example, after a short-circuit fault occurs, the time interval from the onset of the anomaly to the return of the short-circuit current to the normal range is the recovery time window. The recovery time window reflects the duration of the anomaly and is used to determine whether to activate branch circuit breakers for protection.

[0036] A comparison algorithm is used to compare the recovery time window with the short-time activation window. A comparison program runs in the microprocessor. If the recovery time window is smaller than the short-time activation window, the linkage protection instruction is deemed unnecessary. Conversely, if the recovery time window is larger than the short-time activation window, the linkage protection instruction is deemed necessary. This decision logic can be implemented, for example, by programming a conditional statement (e.g., "if recovery time window > short-time activation window, then activate linkage protection instruction").

[0037] In one possible implementation, a short-term activation window is set, and step S300 further includes: the short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker; wherein the historical abnormal event samples include abnormal event operation monitoring parameter samples of the main circuit breaker, abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend.

[0038] Specifically, historical abnormal event samples refer to records of various abnormal conditions that occurred during the past operation of the main circuit breaker, including information such as the time of abnormality occurrence, the time of restoration to normal, and operating monitoring parameters. Abnormal event operating monitoring parameter samples record the various operating parameters monitored during the occurrence of historical abnormal events, such as the specific values ​​of voltage, current, and temperature at the time of the abnormality. These parameters directly reflect the operating status of the circuit breaker during the abnormality. Abnormal event recovery frequency refers to the frequency with which the main circuit breaker recovered from abnormal conditions during historical abnormal events. For example, within a certain period of time, the number of abnormalities and the number of times the circuit breaker recovered after each abnormality reflects the pattern of abnormality recovery. Abnormal event duration refers to the duration of each abnormal event of the main circuit breaker. Different types and severity of abnormalities may have different durations, and the duration reflects the severity and development trend of the abnormality. Abnormal event fluctuation trend describes the fluctuation of various operating parameters of the main circuit breaker during the occurrence of abnormal events. For example, whether the voltage gradually increases or decreases, whether the current fluctuates violently, etc., reflects the development dynamics of the abnormal event.

[0039] By adaptively analyzing historical abnormal event samples, an appropriate short-term activation window can be intelligently determined based on the main circuit breaker's historical operating characteristics and abnormal patterns. This window better reflects the actual situation of the main circuit breaker and provides a more accurate and reasonable basis for subsequent decisions on whether to issue linkage protection instructions.

[0040] In one possible implementation, the short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker, further including: screening valid recovery event samples of the historical abnormal event samples, wherein the valid recovery event samples are abnormal events marked as restored to normal; extracting the recovery time distribution of the valid recovery event samples, and analyzing the recovery time distribution to obtain a basic activation threshold; calculating a threshold adjustment factor based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend of the historical abnormal event samples; and updating the basic activation threshold according to the threshold adjustment factor to obtain an adaptive short-term activation window of the main circuit breaker.

[0041] Specifically, each abnormal event in the stored sample library of historical abnormal events for main circuit breakers is traversed and checked. Using preset tag recognition rules, abnormal events marked as restored to normal are selected, forming a sample set of valid restoration events. This tag recognition rule is based on a combination of criteria, including the return of the circuit breaker's operating monitoring parameters to a normal range and the reset signal from the relevant protective device. For example, if the circuit breaker's current parameter gradually falls back to within ±5% of the rated current after an abnormality, and the protective device issues a reset signal, the abnormal event is marked as restored to normal.

[0042] For the selected valid recovery event samples, the recovery time for each event is extracted—the interval from the time the anomaly occurred to the time it was marked as restored to normal. This recovery time data is organized to form a recovery time distribution. Statistical analysis methods, such as calculating the mean and median of the recovery time, are used to determine the basic activation threshold. For example, the mean recovery time of all valid recovery event samples can be calculated and used as the initial value for the basic activation threshold. Furthermore, the basic activation threshold can be adjusted based on factors such as the standard deviation of the recovery time to improve its rationality and stability.

[0043] The threshold adjustment factor is a coefficient calculated based on the abnormal event recovery frequency, the duration of the abnormal event, and the fluctuation trend of the abnormal event. It is used to adjust the basic activation threshold to obtain an adaptive short-term activation window that is more in line with the actual situation. Specifically, for the abnormal event recovery frequency, the total number of abnormal events that occur in the main circuit breaker within a period of time and the number of valid recovery event samples can be counted. The abnormal event recovery frequency is calculated as the ratio of the number of valid recovery events to the total number of abnormal events. If the abnormal event recovery frequency is high, it means that the circuit breaker can recover on its own in most abnormal situations. In this case, the short-term activation window can be made longer to give the circuit breaker more time to recover on its own. Conversely, if the recovery frequency is low, it means that the abnormal situation is more complex and the circuit breaker is difficult to recover on its own, requiring more timely linkage protection.

[0044] In terms of the duration of abnormal events, we can group valid recovery event samples into intervals from short to long according to the duration of the abnormality. We can count the number of valid recovery event samples in each interval, and calculate the proportion of samples in each interval to the total number of valid recovery event samples to obtain the event proportion of each interval. Analyzing these interval proportions, if the proportion shows a trend of gradually increasing with the duration, it means that among the abnormalities that can be recovered on their own, there are relatively more abnormalities with longer durations, and the abnormalities are relatively stable. Conversely, if the distribution of the proportions is relatively scattered, with no obvious trend of increasing with the duration, or even shows a situation where the proportion decreases with the duration, it means that the duration of the abnormalities that can be recovered on their own is mostly short, and the abnormalities are not very stable.

[0045] For each valid recovery event sample, we can analyze the rate of change of operating parameters (such as voltage and current) during the abnormal event. If the rate of change is large, it indicates that the abnormal situation is unstable and requires increasing the waiting time to more accurately observe the abnormal development trend. Conversely, if the rate of change is small, it indicates that the abnormal situation is relatively stable.

[0046] Taking these three factors into account, a weighted average or other method is used to calculate the final threshold adjustment factor. This calculated threshold adjustment factor is multiplied or added (depending on the specific calculation model) to the basic activation threshold to obtain the updated main circuit breaker adaptive short-time activation window.

[0047] In one possible implementation, the recovery time distribution of the effective recovery event sample is extracted, and the recovery time distribution is analyzed to obtain a basic activation threshold, further including: performing density clustering analysis according to the recovery time distribution to obtain multiple clustering results; extracting a first clustering result according to the density value of each clustering result, and using the quantile value of the first clustering result as the basic activation threshold.

[0048] Specifically, the recovery times of all valid recovery event samples are aggregated to form a recovery time distribution dataset. For example, if 100 valid recovery event samples are collected, and their recovery times are 50 milliseconds, 80 milliseconds, 120 milliseconds, etc., these data constitute the recovery time distribution.

[0049] A density clustering algorithm (such as DBSCAN) is used to analyze the recovery time distribution dataset. This algorithm clusters data points based on their density, assigning densely connected data points to the same cluster. In the recovery time distribution, density clustering can group events with similar recovery times together. For example, after density clustering analysis, three clusters may be obtained: Cluster A contains events with recovery times between 0 and 100 milliseconds, Cluster B contains events with recovery times between 100 and 200 milliseconds, and Cluster C contains events with recovery times between 200 and 300 milliseconds. The density value of each cluster is calculated. The density value is measured by the ratio of the number of data points in the cluster to the spatial extent of the cluster. The cluster with the largest density value is selected as the first cluster. This is because clusters with large density values ​​represent a more concentrated and frequently occurring type of recovery time in the recovery time distribution and better reflect the characteristics of the majority of valid recovery events.

[0050] Sort the recovery time data in the first cluster and calculate its 90th percentile. The 90th percentile indicates that 90% of the recovery times in the first cluster are less than this value. This 90th percentile serves as the base activation threshold, meaning that the majority (90%) of valid recovery events will return to normal within this threshold time.

[0051] This implementation uses density clustering analysis to group events with similar characteristics in the recovery time distribution. The cluster with the highest density is selected as the first cluster, which better represents the actual situation of the majority of valid recovery events. Using the 90th percentile as the base activation threshold ensures that it covers the majority of valid recovery events and avoids the influence of individual extreme values ​​(such as events with extremely long or short recovery times), thereby improving the accuracy of the base activation threshold.

[0052] In one possible implementation, the threshold adjustment factor is calculated based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend of the historical abnormal event samples, further including: defining a frequency adjustment factor, a continuous adjustment factor, and a trend adjustment factor based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend; and performing cumulative calculations according to the frequency adjustment factor, continuous adjustment factor, and trend adjustment factor to comprehensively obtain the threshold adjustment factor.

[0053] Specifically, one possible method for calculating the frequency adjustment factor is as follows: historical abnormal event sample data within a preset time period is periodically read from a database, and the total number of abnormal event recovery events within this preset time period is counted. A baseline recovery frequency is also pre-set. This baseline value can be determined based on historical long-term stable operating data or industry experience. The frequency adjustment factor is calculated based on the difference between the actual recovery frequency and the baseline recovery frequency and multiplied by an adjustment coefficient. The adjustment coefficient ranges from 0 to 1 and can be adjusted based on system sensitivity and actual needs. For example, if the system is sensitive to changes in the abnormal event recovery frequency and you want the frequency adjustment factor to more clearly reflect these changes, a larger value can be used. If the system is relatively stable and the threshold does not need to be adjusted too frequently, a smaller value can be used. The frequency adjustment factor is calculated by first calculating the difference between the actual recovery frequency and the baseline recovery frequency, dividing it by the baseline recovery frequency to obtain the relative rate of change, and finally multiplying it by the adjustment coefficient.

[0054] One possible calculation method for the continuous adjustment factor is as follows: When reading historical abnormal event sample data, record the duration of each historical abnormal event sample and calculate the average of these durations. At the same time, set a baseline duration. This baseline value can be determined based on the system's normal operating parameters or industry standards. The calculation of the continuous adjustment factor is also related to the difference between the actual average duration and the baseline duration and is multiplied by an adjustment coefficient. The adjustment coefficient also ranges from 0 to 1, and its value depends on the system's sensitivity to the duration of abnormal events. If the system is more sensitive to changes in duration, a larger value can be used; otherwise, a smaller value can be used. The continuous adjustment factor is calculated by first calculating the difference between the actual average duration and the baseline duration, dividing it by the baseline duration to obtain the relative rate of change, and finally multiplying it by the adjustment coefficient.

[0055] One possible calculation method for the trend adjustment factor is as follows: Use time series analysis to analyze the fluctuation trend of abnormal events. After reading historical abnormal event sample data, calculate the rate of change in the duration of abnormal events in adjacent samples. Calculate the rate of change for all adjacent samples in sequence, and then calculate the average of these rates of change. The trend adjustment factor is related to this average rate of change and multiplied by an adjustment coefficient. The adjustment coefficient ranges from 0 to 1, and its value depends on the system's sensitivity to the fluctuation trend of abnormal events. If the system wants the trend adjustment factor to reflect changes in the fluctuation trend more promptly, a larger value can be used; if the system is relatively stable and less sensitive to trend changes, a smaller value can be used.

[0056] After calculating the frequency adjustment factor, continuous adjustment factor, and trend adjustment factor, these three adjustment factors are accumulated to obtain the final threshold adjustment factor. An accurate threshold adjustment factor enables the main circuit breaker to more accurately determine whether to send a linkage protection command to the branch circuit breaker. Inaccurate threshold adjustment factor calculations can lead to improper short-term activation window settings, potentially causing the system to malfunction (sending a linkage command when protection is not required) or miss actuation (not sending a linkage command when protection is required). This implementation method, by comprehensively considering multiple factors to calculate the threshold adjustment factor, can effectively reduce the occurrence of such malfunctions and misses, thereby enhancing system stability and reliability.

[0057] In one possible implementation, a recovery time window is calculated for the abnormal operation monitoring parameters. Step S300 further includes step S310, in which an abnormal recovery time prediction model is constructed. The abnormal recovery time prediction model is obtained through supervised training of historical abnormal event samples of the main circuit breaker. Specifically, a large number of historical abnormal event samples of the main circuit breaker are collected. These samples include various abnormal conditions that occurred under different operating conditions of the main circuit breaker, including but not limited to short circuit faults, overload faults, and leakage faults. For each abnormal event, detailed information such as the time of occurrence, the type of abnormality, the various operating parameters at the time of the abnormality (such as current, voltage, power, etc.), the fault handling measures, and the final recovery time are recorded.

[0058] Because collected historical data may contain noise, missing values, or outliers, preprocessing is necessary to improve data quality. Noise is removed using methods such as smoothing filters. Missing values ​​are filled using interpolation methods (such as linear interpolation and mean interpolation) based on the data's distribution characteristics and correlations. Outliers are identified and corrected through statistical analysis methods (such as those based on standard deviation). Preprocessed data is more accurate and reliable, providing a sound foundation for model training.

[0059] Select an appropriate machine learning or deep learning algorithm to build an anomaly recovery time prediction model. Divide the preprocessed historical anomaly event samples into a training set and a test set according to a specific ratio (e.g., 7:3 or 8:2). Based on the training set data, supervised training is performed on the model using the operating parameters at the time of the anomaly as input features and the corresponding recovery time as the output label. During training, the model parameters are continuously adjusted so that the model learns the mapping between input features and output labels, thereby improving its recovery time prediction capabilities. After training, the model is evaluated using the test set data. Model performance is measured by calculating error metrics (such as mean squared error and mean absolute error) between the predicted recovery time and the actual recovery time. If the model performance does not meet the requirements, adjust the model structure or parameters and retrain and evaluate until a satisfactory model is obtained.

[0060] Step S320: Input the abnormal operation monitoring parameters into the abnormal recovery time prediction model, and output the predicted recovery time window. Specifically, when the main circuit breaker experiences abnormal operation, the current abnormal operation monitoring parameters are collected in real time. These parameters correspond to the input features used in model training, including operating data such as current, voltage, and power. The collected parameters are preprocessed as necessary (such as normalization to bring the data to the same order of magnitude for ease of model processing) and then input into the trained abnormal recovery time prediction model. The model quickly calculates the predicted recovery time window based on the input abnormal operation monitoring parameters and the mapping relationship learned during training.

[0061] This implementation method constructs an abnormal recovery time prediction model through supervised training, which can more accurately consider the combined impact of multiple factors, thereby improving the accuracy of recovery time prediction.

[0062] In some possible embodiments, the abnormal recovery time prediction model is trained based on supervised training using historical abnormal event samples. However, this may present the following issues: Static models trained using historical samples are difficult to adapt to the real-time changes in the grid's operating state, such as load fluctuations caused by the integration of distributed energy resources, topology adjustments, or sudden environmental changes. When the main circuit breaker detects a new abnormal pattern (such as a high-frequency oscillation fault), the model's lack of a real-time learning mechanism will cause the predicted recovery time window to deviate significantly from the actual value. This can lead to two risks: if the predicted recovery time is too short, it may be misjudged as a transient fault, delaying the activation of protection and expanding the fault range; if the predicted recovery time is too long, the branch circuit breaker may be triggered prematurely, causing unnecessary power outages.

[0063] Therefore, as an implementation method, in order to solve this defect, this embodiment adds an edge training module and a real-time data pipeline to the secondary intelligent control layer, and the module is driven by the FPGA chip built into the main circuit breaker. When an abnormality occurs in the main circuit breaker, the real-time operating parameters collected by the current sensor and the voltage sensor are synchronously input into the abnormal recovery time prediction model and the edge training module. The edge training module dynamically updates the model weight by comparing the deviation between the predicted recovery time and the actual recovery time. For example, if the actual recovery time is 20% longer than the predicted value, the module automatically triggers the incremental learning algorithm and uses the new samples in the sliding time window to fine-tune the model parameters. At the same time, the model update parameters are exchanged with adjacent circuit breakers through the communication module to form a collaborative learning network. This embodiment enables the prediction model to have environmental adaptability, solves the failure problem of static models in dynamic power grids, and does not require changing the primary fault layer structure of the original device.

[0064] Furthermore, self-verification is achieved through a closed-loop data loop between devices. When the linkage protection instruction is activated, the result of the branch circuit breaker's primary fault-breaking action (such as tripping time and fault isolation status) is transmitted back to the main circuit breaker through the communication module. The secondary intelligent control layer cross-validates this result with the output of the prediction model. If a significant deviation occurs for three consecutive times (such as an error > 15%), the model reconstruction mode is automatically started: the latest data from the historical abnormal event sample library is called, and full retraining is performed in the edge training module. The reconstructed model is switched to the online state after simulation verification to ensure continuous optimization of the prediction accuracy. The technical solution of this embodiment integrates historical experience with real-time learning, significantly improving the scenario adaptability of the model.

[0065] Step S400: If the restoration time window is smaller than the short-time activation window, the linkage protection instruction is not activated; if the restoration time window is larger than the short-time activation window, the linkage protection instruction is activated.

[0066] Specifically, a coordinated protection command is a control command sent by the main circuit breaker to a branch circuit breaker when it detects an anomaly and determines that the branch circuit breaker needs to cooperate with the protection. This command causes the branch circuit breaker to perform operations such as opening and closing to achieve the protection function of the distribution network.

[0067] When it is determined that a linkage protection instruction needs to be activated, the secondary intelligent control layer generates a linkage protection instruction based on the preset protection strategy. The instruction content includes the action object (such as the specific branch circuit breaker number) and the action type (such as opening, closing, etc.). For example, a command "branch circuit breaker 001 open" is generated and encoded according to the communication protocol. The linkage protection instruction is sent to the corresponding branch circuit breaker through the communication module. During the transmission process, data verification mechanisms (such as cyclic redundancy check CRC) are used to ensure the accuracy and integrity of the instruction. For example, when the instruction is transmitted via fiber optic communication, a CRC check code is added at the sending end, and the received instruction is verified at the receiving end. If a verification error is found, it is requested to be resent. The primary disconnection layer of the branch circuit breaker performs the disconnection operation based on the received instruction through an electromagnetic mechanism or spring mechanism.

[0068] In one possible implementation, determining a circuit breaker group including a hierarchical relationship, the method further includes: if each circuit breaker group includes a main circuit breaker, at least one primary branch circuit breaker, and at least one secondary branch circuit breaker; setting a first short-time activation window and a second short-time activation window, determining whether the main circuit breaker sends a primary linkage protection instruction to the primary branch circuit breaker based on the first short-time activation window, and determining whether the primary branch circuit breaker sends a secondary linkage protection instruction to the secondary branch circuit breaker based on the second short-time activation window.

[0069] Specifically, in this scenario, the main circuit breaker is at the top level. When a serious fault occurs in the main line, such as a short circuit or overload, and the fault current exceeds the main circuit breaker's setting, the main circuit breaker must quickly shut off the power supply to protect the entire distribution system from further damage. The primary branch circuit breaker is responsible for monitoring and managing the electrical parameters of the primary branch in which it is located. When an abnormality occurs in that branch, it will promptly operate to prevent the fault from spreading to the main line or other branches. The secondary branch circuit breaker similarly protects smaller circuits, ensuring the safe operation of individual devices or small areas. For example, in the power distribution system of a large commercial building, the main circuit breaker controls the power supply to the entire building, the primary branch circuit breakers control the power supply to different floors, and the secondary branch circuit breakers control the power supply to different businesses or functional areas on each floor.

[0070] The first short-term activation window is a time interval set for the coordinated protection between the main circuit breaker and the primary branch circuit breaker. During system operation, electrical parameters of the main circuit breaker and primary branch circuit breakers, such as current and voltage, are monitored in real time. When the main circuit breaker detects a fault signal (such as an overcurrent signal), it does not immediately send a first-level coordinated protection command to the primary branch circuit breaker. Instead, the first short-term activation window is activated. During this window, the system continuously monitors the fault condition of the main circuit breaker and the electrical status of the branch circuit where the primary branch circuit breaker is located. If the main circuit breaker fault persists at the end of the first short-term activation window, and the preset coordinated protection logic determines that the primary branch circuit breaker needs to cooperate in power disconnection to prevent the fault from escalating, the main circuit breaker will send a first-level coordinated protection command to the primary branch circuit breaker. Upon receiving the command, the primary branch circuit breaker will quickly activate and disconnect the power to the branch circuit where it is located.

[0071] The second short-time activation window is the time interval for the linkage protection between the primary branch circuit breaker and the secondary branch circuit breaker. When the primary branch circuit breaker detects a fault signal or receives a primary linkage protection instruction from the main circuit breaker, it will not immediately send a secondary linkage protection instruction to the secondary branch circuit breaker. Instead, it will start the second short-time activation window. During the window time, the system monitors the fault condition of the primary branch circuit breaker and the electrical status of the branch where the secondary branch circuit breaker is located. If at the end of the second short-time activation window, according to the preset logic, the secondary branch circuit breaker needs to be activated to further isolate the fault, the primary branch circuit breaker will send a secondary linkage protection instruction to the secondary branch circuit breaker, and the secondary branch circuit breaker will execute the power cut operation.

[0072] This implementation approach's clear division of upper and lower level circuit breaker groups and precise judgment of coordinated protection commands enable faults to be quickly located and isolated to the smallest possible extent. This precise fault isolation method effectively reduces the scope of power outages, improves power system reliability, and ensures the power needs of critical users.

[0073] In one possible implementation, the method further includes: the first short-time activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker, and the second short-time activation window is determined by adaptively analyzing historical abnormal event samples of the first-level branch circuit breaker; wherein, the multi-level linkage protection instruction consisting of the first-level linkage protection instruction and the second-level linkage protection instruction is sent sequentially in hierarchical order.

[0074] Specifically, the first short-time activation window and the second short-time activation window are consistent with the principle of determining the short-time activation window through the aforementioned adaptive analysis, and will not be repeated here. The multi-level linkage protection instruction consists of a first-level linkage protection instruction and a second-level linkage protection instruction, and is sent sequentially in hierarchical order. That is, when the main circuit breaker detects a fault signal, it starts the first short-time activation window and continuously monitors the fault situation within this window. If the fault still exists at the end of the window and meets the first-level linkage protection conditions, the main circuit breaker sends a first-level linkage protection instruction to the first-level branch circuit breaker. After receiving the instruction, the first-level branch circuit breaker will not immediately send an instruction to the second-level branch circuit breaker, but will start the second short-time activation window to further judge the fault situation within its own monitoring range. Only when the second short-time activation window ends and it is determined according to the preset logic that the second-level branch circuit breaker needs to be activated, the first-level branch circuit breaker will send the second-level linkage protection instruction to the second-level branch circuit breaker. This hierarchical method of sending instructions in sequence ensures that circuit breakers at all levels can participate in fault protection in a reasonable order and time, avoids confusion and conflict of instructions, ensures that faults are gradually isolated according to predetermined strategies, controls the scope of impact to a minimum, improves the stability and reliability of the entire power distribution system, and reduces the duration and scope of power outages.

[0075] The embodiment of the present application adopts the method of obtaining multiple circuit breaker devices, identifying the branch relationship between them and grouping them, each group contains a main circuit breaker and several branch circuit breakers, and setting a short-time activation window time. When an abnormality is detected in the main circuit breaker, the abnormal operation parameters are extracted and the time window required for its recovery is calculated, and this recovery time is compared with the preset short-time activation window. If the recovery time is shorter than the activation window, the system does not trigger the linkage protection and continues to operate; if the recovery time is longer than the activation window, a linkage protection instruction is immediately sent to the branch circuit breaker to disconnect the main circuit and branch circuit breakers synchronously. Such technical means solve the technical problems of poor protection accuracy and economy existing in the existing protection setting of primary and secondary fusion pole-mounted circuit breakers, and achieve the technical effect of improving the accuracy and economy of power grid protection.

[0076] In the above, refer to Figure 1 The adaptive protection setting method for a primary and secondary fusion column mounted circuit breaker according to an embodiment of the present invention is described in detail. Figure 2 An adaptive protection setting system for a primary / secondary fusion pole mounted circuit breaker according to an embodiment of the present invention is described.

[0077] An adaptive protection setting system for a primary / secondary integrated pole-mounted circuit breaker according to an embodiment of the present invention is designed to address the technical issues of poor protection accuracy and cost-effectiveness associated with existing protection setting systems for primary / secondary integrated pole-mounted circuit breakers, thereby improving the accuracy and cost-effectiveness of power grid protection. The adaptive protection setting system for a primary / secondary integrated pole-mounted circuit breaker includes a pole-mounted circuit breaker acquisition module 10, a circuit breaker group identification module 20, a restoration time window calculation module 30, and a protection instruction transmission determination module 40.

[0078] A pole-mounted circuit breaker acquisition module 10 is used to acquire multiple primary and secondary fusion pole-mounted circuit breakers; a circuit breaker group identification module 20 is used to identify the branch relationships of the multiple primary and secondary fusion pole-mounted circuit breakers, and determine circuit breaker groups containing superior and subordinate relationships, each circuit breaker group including a main circuit breaker and at least one branch circuit breaker; a recovery time window calculation module 30 is used to set a short-time activation window, extract abnormal operation monitoring parameters of the main circuit breaker, perform a recovery time window calculation on the abnormal operation monitoring parameters, and determine whether to send a linkage protection instruction to the branch circuit breaker by comparing the recovery time window with the short-time activation window; a protection instruction sending judgment module 40 is used to not activate the linkage protection instruction if the recovery time window is less than the short-time activation window, and activate the linkage protection instruction if the recovery time window is greater than the short-time activation window.

[0079] The specific configuration of the pole-mounted circuit breaker acquisition module 10 will be described in detail below. As described above, the pole-mounted circuit breaker acquisition module 10 may further include: each of the plurality of primary / secondary fused pole-mounted circuit breakers includes a primary tripping layer and a secondary intelligent control layer; wherein the primary tripping layer is used to execute branch circuit breaker tripping protection according to activated linkage protection instructions, and the secondary intelligent control layer is used to monitor abnormal operation and receive or send linkage protection instructions.

[0080] The specific configuration of the recovery time window calculation module 30 will be described in detail below. As described above, a short-term activation window is set. The recovery time window calculation module 30 may further include: The short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker; wherein the historical abnormal event samples include abnormal event operation monitoring parameter samples of the main circuit breaker, abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend.

[0081] In which, the short-time activation window is determined by adaptively analyzing the historical abnormal event samples of the main circuit breaker, and the recovery time window calculation module 30 may further include: an effective recovery event sample screening unit for screening effective recovery event samples of the historical abnormal event samples, wherein the effective recovery event samples are abnormal events marked as restored to normal; a recovery time distribution extraction unit for extracting the recovery time distribution of the effective recovery event samples, and analyzing the recovery time distribution to obtain a basic activation threshold; a threshold adjustment factor calculation unit for calculating a threshold adjustment factor based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend of the historical abnormal event samples; and an updating unit for updating the basic activation threshold according to the threshold adjustment factor to obtain an adaptive short-time activation window of the main circuit breaker.

[0082] Among them, the recovery time distribution of the effective recovery event sample is extracted, and the recovery time distribution is analyzed to obtain the basic activation threshold. The recovery time distribution extraction unit may further include: a density clustering analysis subunit for performing density clustering analysis according to the recovery time distribution to obtain multiple clustering results; a basic activation threshold determination subunit for extracting the first clustering result according to the density value of each clustering result, and using the quantile value of the first clustering result as the basic activation threshold.

[0083] Among them, the threshold adjustment factor is calculated based on the abnormal event recovery frequency, abnormal event duration and abnormal event fluctuation trend of the historical abnormal event samples. The threshold adjustment factor calculation unit may further include: an adjustment factor definition subunit is used to define the frequency adjustment factor, the continuous adjustment factor and the trend adjustment factor according to the abnormal event recovery frequency, the abnormal event duration and the abnormal event fluctuation trend; the accumulation calculation subunit is used to perform accumulation calculation according to the frequency adjustment factor, the continuous adjustment factor and the trend adjustment factor to comprehensively obtain the threshold adjustment factor.

[0084] Among them, determining the circuit breaker group including the upper and lower relationships, the system may further include: if each circuit breaker group includes a main circuit breaker, at least one primary branch circuit breaker and at least one secondary branch circuit breaker, setting a first short-time activation window and a second short-time activation window, judging whether the main circuit breaker sends a primary linkage protection instruction to the primary branch circuit breaker according to the first short-time activation window, and judging whether the primary branch circuit breaker sends a secondary linkage protection instruction to the secondary branch circuit breaker according to the second short-time activation window.

[0085] In which, the system may further include: the first short-time activation window is determined by adaptively analyzing the historical abnormal event samples of the main circuit breaker, and the second short-time activation window is determined by adaptively analyzing the historical abnormal event samples of the first-level branch circuit breaker; wherein, the multi-level linkage protection instructions consisting of the first-level linkage protection instructions and the second-level linkage protection instructions are sent in sequence according to the hierarchical order.

[0086] Among them, the recovery time window calculation is performed on the abnormal operation monitoring parameters, and the recovery time window calculation module 30 may further include: an abnormal recovery time prediction model construction unit is used to construct an abnormal recovery time prediction model, and the abnormal recovery time prediction model is obtained by supervised training of historical abnormal event samples of the main circuit breaker; the abnormal recovery time prediction unit is used to input the abnormal operation monitoring parameters into the abnormal recovery time prediction model, and output the predicted recovery time window.

[0087] The adaptive protection setting system for a primary-secondary fusion pole-mounted circuit breaker provided in an embodiment of the present invention can execute the adaptive protection setting method for a primary-secondary fusion pole-mounted circuit breaker provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0088] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0089] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. The adaptive protection setting method for primary and secondary fusion column mounted circuit breakers is characterized in that: The method comprises: Obtain multiple primary and secondary fusion pole-mounted circuit breakers; Identify the branch relationships of the multiple primary and secondary fusion column mounted circuit breakers, and determine circuit breaker groups including a superior-subordinate relationship, each circuit breaker group including a main circuit breaker and at least one branch circuit breaker; Setting a short-time activation window, extracting abnormal operation monitoring parameters of the main circuit breaker when an abnormality occurs, calculating a recovery time window for the abnormal operation monitoring parameters, and determining whether to send a linkage protection instruction to the branch circuit breaker by comparing the recovery time window with the short-time activation window; If the restoration time window is smaller than the short-time activation window, the linkage protection instruction is not activated; if the restoration time window is larger than the short-time activation window, the linkage protection instruction is activated.

2. The adaptive protection setting method for a primary / secondary fusion column mounted circuit breaker according to claim 1, characterized in that: Each of the plurality of primary and secondary fusion column mounted circuit breakers comprises a primary breaking layer and a secondary intelligent control layer; The primary breaking layer is used to execute the breaking protection of the branch circuit breaker according to the activated linkage protection instruction, and the secondary intelligent control layer is used to perform abnormal operation monitoring and receive or send the linkage protection instruction.

3. The adaptive protection setting method for a primary / secondary fusion column mounted circuit breaker according to claim 1, characterized in that: Setting a short-term activation window, wherein the short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker; The historical abnormal event samples include abnormal event operation monitoring parameter samples of the main circuit breaker, abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend.

4. The adaptive protection setting method for a primary / secondary fusion column mounted circuit breaker according to claim 3, characterized in that: The short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker, and the method includes: Screening the historical abnormal event samples for valid recovery event samples, wherein the valid recovery event samples are abnormal events marked as returning to normal; Extracting the recovery time distribution of the effective recovery event sample, and analyzing the recovery time distribution to obtain a basic activation threshold; Calculating a threshold adjustment factor based on the abnormal event recovery frequency, abnormal event duration, and abnormal event fluctuation trend of the historical abnormal event samples; The basic activation threshold is updated according to the threshold adjustment factor to obtain an adaptive short-time activation window of the main circuit breaker.

5. The adaptive protection setting method for a primary / secondary fusion pole mounted circuit breaker according to claim 4, characterized in that: Extracting the recovery time distribution of the effective recovery event sample and analyzing the recovery time distribution to obtain a basic activation threshold, the method includes: Performing density cluster analysis according to the recovery time distribution to obtain multiple clustering results; A first clustering result is extracted according to the density value of each clustering result, and a quantile value of the first clustering result is used as a basic activation threshold.

6. The adaptive protection setting method for a primary / secondary fusion column mounted circuit breaker according to claim 5, characterized in that: The threshold adjustment factor is calculated based on the abnormal event recovery frequency, the duration of the abnormal event, and the abnormal event fluctuation trend of the historical abnormal event samples, and the method includes: Define a frequency adjustment factor, a duration adjustment factor, and a trend adjustment factor based on the abnormal event recovery frequency, the abnormal event duration, and the abnormal event fluctuation trend; The frequency adjustment factor, the continuous adjustment factor and the trend adjustment factor are cumulatively calculated to obtain a threshold adjustment factor.

7. The adaptive protection setting method for a primary / secondary fusion pole mounted circuit breaker according to claim 1, characterized in that: Determining a circuit breaker group including a superior-subordinate relationship, the method further includes: If each circuit breaker group includes a main circuit breaker, at least one primary branch circuit breaker and at least one secondary branch circuit breaker; A first short-time activation window and a second short-time activation window are set. Based on the first short-time activation window, it is determined whether the main circuit breaker sends a first-level linkage protection instruction to the first-level branch circuit breaker. Based on the second short-time activation window, it is determined whether the first-level branch circuit breaker sends a second-level linkage protection instruction to the second-level branch circuit breaker.

8. The adaptive protection setting method for a primary / secondary fusion pole mounted circuit breaker according to claim 7, characterized in that: The first short-term activation window is determined by adaptively analyzing historical abnormal event samples of the main circuit breaker, and the second short-term activation window is determined by adaptively analyzing historical abnormal event samples of the primary branch circuit breaker; Among them, the multi-level linkage protection instruction consisting of the first-level linkage protection instruction and the second-level linkage protection instruction is sent in sequence according to the hierarchical order.

9. The adaptive protection setting method for a primary / secondary fusion pole mounted circuit breaker according to claim 1, characterized in that: Calculating a recovery time window for the abnormal operation monitoring parameter includes: Constructing an abnormality recovery time prediction model, wherein the abnormality recovery time prediction model is obtained by performing supervised training on historical abnormal event samples of the main circuit breaker; The abnormal operation monitoring parameters are input into the abnormal recovery time prediction model, and the predicted recovery time window is output.

10. The adaptive protection setting system for primary and secondary fusion column mounted circuit breakers is characterized by: The system is used to implement the adaptive protection setting method for a primary-secondary fusion pole-mounted circuit breaker according to any one of claims 1 to 9, and the system includes: Pole-mounted circuit breaker acquisition module, used to obtain multiple primary and secondary fusion pole-mounted circuit breakers; A circuit breaker group identification module is used to identify the branch relationship of the multiple primary and secondary fusion column mounted circuit breakers and determine the circuit breaker groups containing the upper and lower hierarchical relationships, each circuit breaker group including a main circuit breaker and at least one branch circuit breaker; a recovery time window calculation module, configured to set a short-time activation window, extract abnormal operation monitoring parameters of the main circuit breaker, calculate a recovery time window for the abnormal operation monitoring parameters, and determine whether to send a linkage protection instruction to the branch circuit breaker by comparing the recovery time window with the short-time activation window; The protection instruction sending judgment module is used to not activate the linkage protection instruction if the recovery time window is smaller than the short-time activation window, and to activate the linkage protection instruction if the recovery time window is larger than the short-time activation window.

Images