A reclosing control method and device, a storage medium and an electronic device

Abstract

This application provides a reclosing control method, device, storage medium, and electronic device, relating to the field of control technology. The method is applied to a power distribution system, which includes a power source and a reclosing circuit breaker. The input terminal of the reclosing circuit breaker is connected to the power source, and the output terminal is connected to the load. The method includes: acquiring electrical parameters and load characteristic data of the power distribution system; determining whether an anomaly has occurred in the power distribution system based on the electrical parameters; if an anomaly is determined based on the electrical parameters, controlling the reclosing circuit breaker to perform a tripping operation to disconnect the power source from the load; inputting the electrical parameters and load characteristic data within a preset time window before the anomaly occurs into a pre-trained fault type identification model to obtain a fault type identification result; and, if the fault type identification result indicates a momentary fault on the power source side or an abnormal disturbance on the load side, controlling the reclosing circuit breaker to perform a reclosing operation according to the control strategy corresponding to the fault type identification result. This method enables precise control of the reclosing circuit breaker.

Description

Technical Field

[0001] This application relates to the field of control technology, and in particular to a reclosing control method, device, storage medium and electronic equipment. Background Technology

[0002] Reclosing is an important device in power distribution systems used to improve power supply reliability. After a circuit breaker trips due to a line fault, it automatically attempts to reclose the circuit after a preset delay to restore power supply. Traditional reclosing control is mainly based on overcurrent protection principles, detecting whether the current amplitude exceeds a threshold to determine the fault and performing the reclosing operation with a fixed delay. This control method can meet basic requirements in traditional power distribution networks.

[0003] As distribution network structures become increasingly complex, the integration of distributed power sources, the prevalence of nonlinear loads, and the increase in impulsive loads lead to diverse characteristics in line operation. Existing reclosing control methods struggle to accurately identify the nature of faults under complex operating conditions, resulting in unnecessary tripping or missed reclosing opportunities, thus affecting power supply reliability. Summary of the Invention

[0004] In view of this, this application provides a reclosing control method, device, storage medium, and electronic device, which can accurately control reclosing. The specific solution is as follows: A reclosing control method is applied to a power distribution system, the power distribution system including a power source and a reclosing device, wherein the input terminal of the reclosing device is connected to the power source and the output terminal is connected to a load, the method comprising: Obtain the electrical parameters and load characteristic data of the power distribution system; Based on the electrical parameters, determine whether the power distribution system has experienced an abnormality; If an abnormality is determined to have occurred in the power distribution system based on the electrical parameters, the reclosing circuit breaker is controlled to perform a tripping operation to disconnect the power supply from the load. The electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly are input into a pre-trained fault type identification model to obtain a fault type identification result. The anomaly type identification result includes one of the following fault types: power supply side instantaneous fault, power supply side permanent fault, and load side abnormal disturbance. If the fault type identification result is a momentary fault on the power supply side or an abnormal disturbance on the load side, the reclosing device is controlled to perform a reclosing operation according to the control strategy corresponding to the fault type identification result.

[0005] Optionally, in the above method, determining whether an abnormality has occurred in the power distribution system based on the electrical parameters includes: Obtain the reference current waveform of the power distribution system under normal operating conditions; Calculate the waveform similarity between the current current waveform in the electrical parameters and the reference current waveform; If the waveform similarity is lower than a preset similarity threshold, the power distribution system is determined to be abnormal.

[0006] Optionally, the above method, based on the electrical parameters, determines that the power distribution system has malfunctioned, including: If the voltage in the electrical parameters is detected to be less than a preset start-up threshold, it is determined that a voltage drop event has occurred in the power distribution system. Generate event features corresponding to the voltage drop event, the event features including voltage drop depth and event duration; Cluster analysis is performed based on the event characteristics and the abnormal event characteristics of each historical abnormal event in the pre-set historical voltage reduction feature library; If the event characteristics fall within any of the clusters of the historical abnormal events, then the configuration system is determined to be abnormal.

[0007] Optionally, in the above method, inputting the electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly into a pre-trained fault type identification model to obtain the fault type identification result includes: Based on the electrical parameters within a preset time window before the occurrence of the anomaly, calculate the power supply side fault contribution weight and the load side disturbance contribution weight. The power supply side fault contribution weight, the load side disturbance contribution weight, the electrical parameters, and the load characteristic data are input into a pre-trained fault type identification model, and the fault type identification result is output.

[0008] Optionally, in the above method, calculating the power supply-side fault contribution weight and the load-side disturbance contribution weight based on electrical parameters within a preset time window before the anomaly occurs includes: Modal decomposition was performed on the electrical parameters within a preset time window before the occurrence of the anomaly to obtain multiple intrinsic mode function components; Calculate the correlation coefficients between each intrinsic mode function component and the typical fault characteristic template of the power supply side and the typical disturbance characteristic template of the load side, respectively. The proportions of each intrinsic mode function component belonging to power supply side faults and load side disturbances are determined based on the correlation coefficients, and these proportions are used as the contribution weights of power supply side faults and load side disturbances.

[0009] Optionally, the training process of the fault type identification model in the above method includes: Obtain a training sample set and an initial recognition model. The training sample set includes multiple sample data and a sample label for each sample data. The sample data includes electrical parameters and load characteristics at the time of occurrence of historical abnormal events. The initial recognition model is trained using the training sample set; If the initial identification model meets the preset training completion conditions, the initial identification model that meets the training completion conditions will be used as the fault type identification model.

[0010] Optionally, in the above method, controlling the reclosing circuit to perform a reclosing operation based on the control strategy corresponding to the fault type identification result includes: If the fault type identification result is a power supply side instantaneous fault, then the reclosing operation is controlled to be performed according to the first reclosing strategy, the first reclosing strategy including the first reclosing delay time; If the fault type identification result is an abnormal disturbance on the load side, then the reclosing operation is controlled to be performed according to the second reclosing strategy, which includes a second reclosing delay time. The second reclosing delay time is shorter than the first reclosing delay time.

[0011] A reclosing control device is applied to a power distribution system, the power distribution system including a power source and a reclosing device, the input terminal of the reclosing device being connected to the power source and the output terminal being connected to a load, the device comprising: The acquisition unit is used to acquire electrical parameters and load characteristic data of the power distribution system; The judgment unit is used to determine whether an abnormality has occurred in the power distribution system based on the electrical parameters; The first control unit is configured to control the reclosing device to perform a tripping operation to disconnect the power supply from the load if it is determined that the power distribution system is abnormal based on the electrical parameters. The identification unit is used to input the electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly into a pre-trained fault type identification model to obtain a fault type identification result. The anomaly type identification result includes one of the following fault types: power supply side instantaneous fault, power supply side permanent fault, and load side abnormal disturbance. The second control unit is used to control the reclosing device to perform a reclosing operation according to the control strategy corresponding to the fault type identification result when the fault type identification result is a momentary fault on the power supply side or an abnormal disturbance on the load side.

[0012] A storage medium includes storage instructions, wherein, when the instructions are executed, the device in which the storage medium is located is controlled to perform the reclosing control method as described above.

[0013] An electronic device includes a memory and one or more instructions, wherein one or more instructions are stored in the memory and configured to be executed by one or more processors as described above for the reclosing control method.

[0014] Based on the above, this application provides a reclosing control method, device, storage medium, and electronic device, applied to a power distribution system. The power distribution system includes a power supply and a reclosing device. The input terminal of the reclosing device is connected to the power supply, and the output terminal is connected to the load. The method includes: acquiring electrical parameters and load characteristic data of the power distribution system; determining whether an abnormality has occurred in the power distribution system based on the electrical parameters; if an abnormality is determined based on the electrical parameters, controlling the reclosing device to perform a tripping operation to disconnect the power supply from the load; inputting the electrical parameters and load characteristic data within a preset time window before the abnormality occurs into a pre-trained fault type identification model to obtain a fault type identification result. The fault type identification result includes one of the following fault types: power supply-side instantaneous fault, power supply-side permanent fault, and load-side abnormal disturbance; if the fault type identification result is a power supply-side instantaneous fault or a load-side abnormal disturbance, controlling the reclosing device to perform a reclosing operation according to the control strategy corresponding to the fault type identification result. This application can improve the accuracy of fault identification in power distribution systems and effectively distinguish between power supply-side faults and load-side disturbances. This allows for differentiated reclosing control strategies to be implemented for different types of abnormal events, achieving precise control of reclosing and reducing unnecessary tripping operations or missed reclosing opportunities. This effectively improves power supply reliability and reduces equipment impact and power outage losses caused by fault misjudgment. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0016] Figure 1 A flowchart of a reclosing control method provided in this application; Figure 2 A flowchart of a process for obtaining fault type identification results is provided in this application; Figure 3A flowchart illustrating the process of calculating the power supply-side fault contribution weight and the load-side disturbance contribution weight provided in this application; Figure 4 This application provides a schematic diagram of the structure of a reclosing control system; Figure 5 This is a schematic diagram of the structure of an electronic device provided in this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] In this application, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0019] This application provides a reclosing control method, which can be applied to a power distribution system. The power distribution system includes a power source and a reclosing device. The input terminal of the reclosing device is connected to the power source, and the output terminal is connected to the load. The method flowchart is shown below. Figure 1 As shown, it specifically includes: S101: Obtain electrical parameters and load characteristic data of the power distribution system.

[0020] In this embodiment, electrical parameters refer to electrical quantities that characterize the operating state of the power distribution system, including but not limited to at least one of the following: effective current value, effective voltage value, current waveform, voltage waveform, phase difference, active power, and reactive power. Load characteristic data refer to data that characterizes the operating characteristics of the load side, including but not limited to at least one of the following: load current waveform, harmonic content, power factor, load change rate, and load type identification characteristics.

[0021] In practice, the above data can be collected in real time by a sampling unit installed in the reclosing circuit. The sampling unit may include current transformers and voltage transformers, etc., to obtain the original current and voltage signals from the power distribution line, and then process them through signal conditioning circuits and analog-to-digital conversion circuits to obtain electrical parameters and load characteristic data.

[0022] The power distribution system consists of three parts: the power supply side, the power grid lines, and the load side. The recloser, as a protective device, is connected in series with the power grid lines, with its input connected to the power supply side and its output connected to the load side. Therefore, the data collected by the sampling unit built into the recloser can simultaneously reflect the operating status of both the power supply side and the load side.

[0023] S102: Determine whether an abnormality has occurred in the power distribution system based on electrical parameters.

[0024] In this embodiment, an anomaly refers to an event in which the operating state of the power distribution system deviates from the normal range, including but not limited to overload, short circuit, voltage drop, waveform distortion, etc. Anomalies can be determined using either a threshold comparison method or a pattern matching method.

[0025] S103: If an abnormality is detected in the power distribution system based on electrical parameters, the reclosing circuit breaker is controlled to perform a tripping operation to disconnect the power supply from the load.

[0026] In this embodiment, the tripping operation refers to the tripping mechanism within the reclosing circuit driving the operating mechanism to separate the contacts, thereby cutting off the circuit. The specific implementation of controlling the reclosing circuit to perform the tripping operation is as follows: the main control circuit board sends a tripping command to the tripping mechanism, and the tripping mechanism, upon receiving the command, drives the operating mechanism to perform the tripping operation. Through the tripping operation, the fault point can be isolated from the power distribution system, preventing the fault from escalating.

[0027] S104: Input the electrical parameters and load characteristic data within the preset time window before the anomaly occurs into the pre-trained fault type identification model to obtain the fault type identification result. The anomaly type identification result includes one of the following fault types: power supply side instantaneous fault, power supply side permanent fault, and load side abnormal disturbance.

[0028] In this embodiment, the fault type identification model is a classification model trained based on machine learning algorithms, used to distinguish different fault sources and natures based on data characteristics prior to the occurrence of the anomaly. The preset time window refers to a continuous time interval before the anomaly occurs, such as 5 power frequency cycles before the anomaly occurs.

[0029] Specifically, a power supply-side transient fault refers to a fault originating on the power supply side and characterized by its transient nature, such as a situation where the electric arc extinguishes itself after a lightning strike. A power supply-side permanent fault refers to a fault originating on the power supply side and characterized by its permanent nature, such as a broken line or insulation breakdown, requiring manual intervention for recovery. A load-side abnormal disturbance refers to an event with the fault originating on the load side, such as a current surge caused by the starting of a high-power motor or the switching of a nonlinear load. Such events are not true line faults but may be misjudged as faults by traditional protection devices.

[0030] The input data for the fault type identification model includes electrical parameters and load characteristic data. Electrical parameters primarily reflect the operating status of the power supply side, while load characteristic data primarily reflects the operating status of the load side. By jointly inputting these two types of data into the model, the model can learn to distinguish between power supply side faults and load side disturbances.

[0031] S105: If the fault type identification result is a power supply side instantaneous fault or a load side abnormal disturbance, control the reclosing to perform a reclosing operation according to the control strategy corresponding to the fault type identification result.

[0032] In this embodiment, the reclosing operation refers to the electric operating mechanism within the reclosing circuit driving the operating mechanism to reclose the contacts and restore circuit continuity. The specific implementation of controlling the reclosing circuit to perform the reclosing operation is as follows: the main control circuit board sends a closing command to the electric operating mechanism, and the electric operating mechanism, upon receiving the command, drives the operating mechanism to perform the closing operation.

[0033] Different control strategies are adopted based on the different fault type identification results.

[0034] In some embodiments, if the fault type identification result is a permanent fault on the power supply side, the reclosing function is blocked and the reclosing operation is no longer performed. At the same time, an alarm signal is output to notify the operation and maintenance personnel to handle the situation.

[0035] This application improves the accuracy of fault identification in power distribution systems, effectively distinguishing between power supply-side faults and load-side disturbances. This allows for differentiated reclosing control strategies to be implemented for different types of abnormal events, achieving precise control of reclosing and reducing unnecessary tripping operations or missed reclosing opportunities. This effectively improves power supply reliability and reduces equipment impact and power outage losses caused by fault misjudgment.

[0036] In one embodiment provided in this application, based on the above-described solution, optionally, determining whether an abnormality has occurred in the power distribution system based on electrical parameters includes: Obtain the reference current waveform of the power distribution system under normal operating conditions; Calculate the waveform similarity between the current current waveform and the reference current waveform in the electrical parameters; If the waveform similarity is lower than the preset similarity threshold, the power distribution system is determined to be abnormal.

[0037] In this embodiment, the reference current waveform refers to the current waveform collected and stored by the sampling unit when the power distribution system is in normal operation, used to characterize the current change characteristics of the load under normal operating conditions. In practice, when the power distribution system is first put into operation or confirmed to be in a stable normal operating state, current waveforms of multiple consecutive power frequency cycles can be collected, processed to generate the reference current waveform, and stored in the reclosing storage unit.

[0038] The current waveform refers to the current waveform of the current power frequency cycle that is collected in real time. It is obtained in real time by the sampling unit through the current transformer and is used to reflect the current operating status of the power distribution system.

[0039] Waveform similarity is used to characterize the degree of similarity in shape between the current current waveform and the reference current waveform. It can be calculated using a preset similarity algorithm, such as cross-correlation analysis or dynamic time warping, with the specific algorithm chosen based on the actual application scenario. Higher similarity indicates that the current waveform is closer to the normal operating state.

[0040] The similarity threshold can be determined based on statistical analysis of historical normal operation data.

[0041] Traditional anomaly detection typically uses amplitude threshold comparison, which struggles to identify anomalies where the amplitude is normal but the waveform is distorted. This embodiment utilizes waveform similarity analysis to identify anomalies where the waveform shape changes, thus expanding the scope of anomaly detection.

[0042] In one alternative embodiment, before calculating waveform similarity, the current current waveform and the reference current waveform can be preprocessed, including amplitude normalization and phase alignment, to eliminate the influence of amplitude differences and phase shifts on the similarity calculation.

[0043] In another optional embodiment, the reference current waveform can be set according to the load type. The corresponding reference current waveform is stored for different load types. When calculating the similarity, the corresponding reference waveform is retrieved according to the current load type to adapt to the waveform feature differences of different loads.

[0044] In one embodiment provided in this application, based on the above-described solution, optionally, determining an anomaly in the power distribution system based on electrical parameters includes: If the voltage in the electrical parameters is detected to be less than a preset start-up threshold, a voltage drop event is determined to have occurred in the power distribution system. Generate event characteristics corresponding to voltage drop events, including voltage drop depth and event duration; Cluster analysis is performed based on the event characteristics and the abnormal event characteristics of each historical abnormal event in the pre-set historical voltage reduction feature library; If the event characteristics fall within a cluster of any historical abnormal events, then it is determined that the configuration system has malfunctioned.

[0045] In this embodiment, voltage refers to the effective voltage value of the power distribution system, which is collected in real time by the sampling unit through a voltage transformer. When the voltage is lower than the activation threshold, it indicates that the power distribution system may experience a voltage drop-type anomaly. The activation threshold can be set according to relevant standards or historical operating data, for example, it can be set to 90% of the rated voltage.

[0046] Optionally, a voltage drop event refers to an event in which the voltage remains below the start-up threshold for a certain period of time. When the voltage is detected to be below the start-up threshold, recording of this event begins and continues until the voltage recovers above the start-up threshold or exceeds the preset maximum recording time.

[0047] The event characteristics can include voltage drop depth and event duration. Voltage drop depth refers to the difference or ratio between the lowest voltage value during the event and the rated voltage, reflecting the severity of the voltage drop. Event duration refers to the length of time from when the voltage first falls below the start-up threshold to when the voltage recovers above the start-up threshold, reflecting the duration of the voltage drop.

[0048] In this embodiment, the historical voltage drop feature library stores the abnormal event features of multiple historical abnormal events. The features of each historical abnormal event also include the voltage drop depth and event duration, and are associated with a corresponding abnormal type label (such as power supply side transient fault, power supply side permanent fault, load side abnormal disturbance, etc.).

[0049] In this embodiment, the feature points of the current voltage drop event are clustered with feature points in the historical abnormal event feature database to determine whether the current event feature falls within any cluster of historical abnormal events. In practice, clustering algorithms such as DBSCAN or K-means can be used to cluster the historical abnormal event features, generating multiple clusters. Each cluster corresponds to a class of abnormal events with similar features. Then, the distance between the current event feature and the center point of each cluster is calculated. If the distance is less than the cluster radius threshold, it is determined that the current event feature is within that cluster, i.e., an anomaly has occurred in the power distribution system.

[0050] In one embodiment provided in this application, based on the above-described scheme, optionally, the process of inputting electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly into a pre-trained fault type identification model to obtain the fault type identification result is as follows: Figure 2 As shown, it includes: S201: Calculate the power supply side fault contribution weight and the load side disturbance contribution weight based on the electrical parameters within a preset time window before the anomaly occurs.

[0051] In this embodiment, the power supply-side fault contribution weight is used to characterize the proportion of factors originating from the power supply side in the current abnormal event, and the load-side disturbance contribution weight is used to characterize the proportion of factors originating from the load side in the current abnormal event. The sum of the two weights is 1, and their numerical distribution reflects the tendency of the fault source.

[0052] S202: Input the power supply side fault contribution weight, load side disturbance contribution weight, electrical parameters and load characteristic data into the pre-trained fault type identification model, and output the fault type identification result.

[0053] In this embodiment, the fault type identification model is a classification model trained based on a machine learning algorithm, used to output fault type identification results based on the input multidimensional features. The fault type identification results include one of the following: power-side transient fault, power-side permanent fault, and load-side abnormal disturbance.

[0054] The inputs to the fault type identification model include: power supply-side fault contribution weights, load-side disturbance contribution weights, electrical parameters, and load characteristic data. The contribution weights provide prior information about the fault source's tendency, while the electrical parameters and load characteristic data provide the raw data features. Inputting the contribution weights and raw data together into the model allows the model to pay more attention to the contribution information of the fault source during inference, which helps improve classification accuracy.

[0055] For example, a fault type identification model can employ a lightweight gradient boosting machine or a deep neural network. The input layer dimension of the model matches the number of input features, and the output layer provides a three-class classification, corresponding to three fault types.

[0056] In one optional embodiment, feature engineering can be performed on electrical parameters and load characteristic data before inputting the data into the model. For example, time-frequency transformation can be performed on the current and voltage waveforms in the electrical parameters to extract frequency domain features; statistical features can be extracted from the load characteristic data to obtain time domain statistical features. The extracted frequency domain features, time domain statistical features, and contribution weights are then fused to generate a fused feature vector, which is then input into the fault type identification model.

[0057] As is easily understood, the introduction of contribution weights is equivalent to incorporating prior knowledge of fault source decoupling into the model input, enabling the model to better learn and distinguish the decision boundaries between power supply-side faults and load-side disturbances. Compared to directly inputting raw electrical parameters and load characteristic data into the model, this embodiment, guided by contribution weights, can reduce the model's dependence on data volume and improve classification accuracy in small sample scenarios.

[0058] In one embodiment provided in this application, based on the above-described scheme, optionally, the process of calculating the power supply-side fault contribution weight and the load-side disturbance contribution weight based on electrical parameters within a preset time window before the anomaly occurs, such as... Figure 3 As shown, it includes: S301: Perform modal decomposition on electrical parameters within a preset time window before the occurrence of the anomaly to obtain multiple intrinsic mode function components.

[0059] In this embodiment, the electrical parameters include current waveforms and voltage waveforms, which are continuously collected by the sampling unit within a preset time window before the anomaly occurs.

[0060] In this embodiment, empirical mode decomposition is used to decompose nonlinear, non-stationary current or voltage waveforms into several intrinsic mode function components and a residual component. Each intrinsic mode function component represents an oscillation mode at different time scales in the original signal and satisfies two conditions: first, the number of extreme points is equal to or differs from the number of zero-crossing points by at most one in the entire data sequence; second, at any given time, the mean of the upper envelope defined by local maxima and the lower envelope defined by local minima is zero.

[0061] In practice, taking current waveforms as an example, empirical mode decomposition (EMD) is performed on the current waveform data within a preset time window before the anomaly occurs. First, all local maxima and minima of the current waveform are identified. Then, interpolation methods are used to fit the upper and lower envelopes respectively. The mean of the upper and lower envelopes is then calculated, and this mean is subtracted from the original current waveform to obtain a candidate component. It is then determined whether this candidate component satisfies the conditions for intrinsic mode functions (IMFs). If it does, it is extracted as an IMF component; otherwise, the above process is repeated until the conditions are met. The extracted IMF component is subtracted from the original current waveform, and the above decomposition process is repeated for the remaining part until the remaining part is a monotonic function or less than a preset threshold, at which point the decomposition stops.

[0062] Through the above decomposition, multiple intrinsic mode function components are obtained. These components are arranged in descending order of frequency and characterize the oscillation characteristics of the current waveform at different time scales. Among them, high-frequency components usually reflect rapidly changing transient processes, while low-frequency components reflect slowly changing trend components. The same modal decomposition is performed on the voltage waveform to obtain the corresponding intrinsic mode function components.

[0063] S302: Calculate the correlation coefficients between each intrinsic mode function component and the typical fault characteristic template of the power supply side and the typical disturbance characteristic template of the load side.

[0064] In this embodiment, the typical fault feature template on the power supply side is a feature reference template pre-constructed based on historical power supply side fault data, used to characterize the typical feature distribution of power supply side faults on each intrinsic mode function component. The typical disturbance feature template on the load side is a feature reference template pre-constructed based on historical load side disturbance data, used to characterize the typical feature distribution of load side disturbances on each intrinsic mode function component.

[0065] The process of constructing a typical fault feature template for the power supply side includes: collecting historical data of multiple known types of power supply side fault events; performing modal decomposition on the electrical parameters within a preset time window before the abnormal occurrence of each fault event to obtain the intrinsic mode function components corresponding to each fault event; averaging or extracting cluster centers from all fault event components in the same frequency band to obtain the typical fault features of the power supply side corresponding to that frequency band; and combining the typical features of all frequency bands to form a typical fault feature template for the power supply side.

[0066] The correlation coefficient is used to measure the similarity between each intrinsic mode function component and the corresponding template. Its value ranges from negative one to positive one. The closer the absolute value is to one, the stronger the correlation. The closer it is to zero, the weaker the correlation.

[0067] In implementation, for each intrinsic mode function component obtained from S301 decomposition, its correlation coefficient with the corresponding frequency band feature in the typical fault feature template of the power supply side is calculated, and its correlation coefficient with the corresponding frequency band feature in the typical disturbance feature template of the load side is also calculated. For example, for the first intrinsic mode function component, its correlation coefficient with the first frequency band feature in the power supply side template and its correlation coefficient with the first frequency band feature in the load side template are calculated; for the second intrinsic mode function component, its correlation coefficient with the second frequency band feature in the power supply side template and its correlation coefficient with the second frequency band feature in the load side template are calculated, and so on, and the same processing is performed on all components.

[0068] S303: Determine the proportion of power supply side faults and the proportion of load side disturbances in each intrinsic mode function component based on the correlation coefficient, and use them as the contribution weights of power supply side faults and load side disturbances.

[0069] In this embodiment, based on the correlation coefficients of each intrinsic mode function component calculated in S302, the degree of attribution of each component to power supply side faults and load side disturbances is further determined, and then the overall contribution weight is obtained by summing them up.

[0070] In one optional embodiment, for each intrinsic mode function component, its correlation coefficient with the power supply side template and its correlation coefficient with the load side template are compared, and the power supply side attribution and load side attribution of the component are obtained through normalization. The normalization process is to perform an exponential operation on the two correlation coefficients, and then divide each of them by the sum of the exponential operation results to obtain two values ​​between zero and one that add up to one, representing the probability that the component belongs to a power supply side fault and the probability that it belongs to a load side disturbance, respectively.

[0071] In another optional embodiment, the magnitudes of the two correlation coefficients can be directly compared. If the correlation coefficient with the power supply side template is greater than that with the load side template, the component is considered to be mainly attributed to a power supply side fault, with a power supply side attribution score of one and a load side attribution score of zero. Conversely, if the correlation coefficient with the power supply side template is greater than that with the load side template, the component is considered to be mainly attributed to a load side disturbance, with a load side attribution score of one and a power supply side attribution score of zero. If the two are equal or the difference is very small, the component is considered to contribute equally to both sides, with each attribution score of 0.5.

[0072] After obtaining the attribution degree of each component, the proportion of the energy of each intrinsic mode function component to the total energy of all components is calculated, and this proportion is used as the weighting factor for that component. The energy can be calculated by summing the squares of the amplitudes of each component. The power supply-side attribution degree of each component is multiplied by its corresponding weighting factor and then summed to obtain the power supply-side fault contribution weight; the load-side disturbance contribution weight is obtained by multiplying the load-side attribution degree of each component by its corresponding weighting factor and then summing these two weights. The sum of these two weights is one.

[0073] In another optional embodiment, if the energy difference of each component is not considered, the average value of the belonging degree of each component can be directly taken to obtain the power supply side fault contribution weight and the load side disturbance contribution weight.

[0074] In one embodiment provided in this application, based on the above-described scheme, optionally, the training process of the fault type identification model includes: Obtain a training sample set and an initial recognition model. The training sample set includes multiple sample data and the sample label for each sample data. The sample data includes the electrical parameters and load characteristics at the time of occurrence of historical abnormal events. The initial recognition model is trained using the training sample set; If the initial identification model meets the preset training completion conditions, the initial identification model that meets the training completion conditions will be used as the fault type identification model.

[0075] In this embodiment, the training sample set is a dataset used for model training, containing multiple sample data and corresponding sample labels. Each sample data comes from a historical abnormal event, specifically including electrical parameters and load characteristic data at the time the abnormal event occurred. The sample label is used to identify the fault type to which the historical abnormal event belongs; for example, the fault type includes one of the following: power supply-side transient fault, power supply-side permanent fault, and load-side abnormal disturbance.

[0076] The initial identification model refers to a machine learning model that is untrained or insufficiently trained, used to learn from a training sample set and eventually acquire the ability to identify fault types. The initial identification model can employ classification models such as lightweight gradient boosting machines, random forests, support vector machines, or deep neural networks; the specific model structure can be selected based on the actual application scenario and computing resources.

[0077] The process of training the initial recognition model using the training sample set includes: using electrical parameters and load feature data from the sample data as input to the model, and using the corresponding sample labels as the model's expected output; iteratively optimizing the model to make its predicted output approximate the sample labels. During training, the model continuously adjusts its internal parameters using the backpropagation algorithm to minimize the difference between the predicted output and the sample labels.

[0078] Training completion criteria are preset standards used to determine whether a model has been trained successfully. For example, training completion criteria can be set to the number of training epochs reaching a preset number, such as 100 iterations; or to the model's recognition accuracy on the validation set exceeding a preset threshold, such as an accuracy of 95% or higher; or to the loss function value converging to a preset range, such as the loss value no longer decreasing after multiple consecutive training epochs.

[0079] In one embodiment provided in this application, based on the above-described scheme, optionally, according to the control strategy corresponding to the fault type identification result, the reclosing circuit is controlled to perform a reclosing operation, including: If the fault type identification result is a power supply side instantaneous fault, then the reclosing operation is executed according to the first reclosing strategy, which includes the first reclosing delay time. If the fault type identification result is an abnormal disturbance on the load side, the reclosing operation is executed according to the second reclosing strategy, which includes the second reclosing delay time. The second reclosing delay time is shorter than the first reclosing delay time.

[0080] In this embodiment, the fault type identification result is output by the aforementioned fault type identification model, which indicates the specific type of the current abnormal event, including one of the following: power supply side transient fault, power supply side permanent fault, and load side abnormal disturbance.

[0081] The first reclosing strategy refers to the reclosing control scheme adopted when the fault type identification result is a transient fault on the power supply side. This strategy includes at least a first reclosing delay time, which is the waiting time between the completion of the opening operation and the execution of the reclosing operation. Transient faults on the power supply side are usually caused by short-lived events such as lightning strikes or tree branches touching the wire. The arc at the fault point can extinguish itself within a short time, and the insulating medium can restore its insulation strength. Therefore, a certain delay time needs to be reserved to ensure that the fault point is completely deionized, avoiding secondary impact on the system when reclosing with a permanent fault. For example, the first reclosing delay time can be set between 0.5 seconds and 1 second, and the specific value can be determined according to factors such as the voltage level of the power distribution system, the line type, and environmental conditions.

[0082] The second reclosing strategy refers to the reclosing control scheme adopted when the fault type identification result is an abnormal disturbance on the load side. This strategy includes at least a second reclosing delay time. Abnormal disturbances on the load side refer to non-faulty events caused by load-side factors, such as the starting of a high-power motor, transformer inrush current, and nonlinear load switching. Although these events may cause current surges or voltage fluctuations, they are not permanent faults of the line itself. Since the disturbance source is located on the load side, and the fault point is not on the line, there is no need to wait for the arc to extinguish and the insulation to recover. Therefore, a shorter reclosing delay can be used to restore power supply as quickly as possible and reduce the impact on load-side equipment.

[0083] In an optional embodiment, the first reclosing strategy and the second reclosing strategy may further include the number of reclosing attempts. For example, for a transient fault on the power supply side, the number of reclosing attempts can be set to once, meaning that no further attempts will be made after the first reclosing attempt fails; for abnormal disturbances on the load side, the number of reclosing attempts can be set to multiple, allowing multiple reclosing attempts to be performed in multiple consecutive disturbance events. The number of reclosing attempts can also be dynamically adjusted according to actual conditions.

[0084] and Figure 1 Corresponding to the method described above, this application also provides a reclosing control device applied to the aforementioned power distribution system. A schematic diagram of the device is shown below. Figure 4 As shown, it includes: Acquisition unit 401 is used to acquire electrical parameters and load characteristic data of the power distribution system; Judgment unit 402 is used to determine whether an abnormality has occurred in the power distribution system based on electrical parameters; The first control unit 403 is used to control the reclosing device to perform a tripping operation to disconnect the power supply from the load if an abnormality is determined to have occurred in the power distribution system based on electrical parameters. The identification unit 404 is used to input electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly into a pre-trained fault type identification model to obtain the fault type identification result. The anomaly type identification result includes one of the following fault types: power supply side instantaneous fault, power supply side permanent fault, and load side abnormal disturbance. The second control unit 405 is used to control the reclosing operation to be performed according to the control strategy corresponding to the fault type identification result when the fault type identification result is a power supply side instantaneous fault or a load side abnormal disturbance.

[0085] This application also provides an electronic device, the structural schematic diagram of which is shown below. Figure 5As shown, it specifically includes a memory 501 and one or more instructions 502, wherein one or more instructions 502 are stored in the memory 501 and are configured to be executed by one or more processors 403 to perform the above-mentioned reclosing control method.

[0086] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0087] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0088] For ease of description, the above system is described by dividing it into various functional units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0089] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0090] The solution provided in this application has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A reclosing control method characterized by, Applied to a power distribution system, the power distribution system including a power source and a reclosing device, the input terminal of the reclosing device is connected to the power source, and the output terminal is connected to the load, the method includes: Obtain the electrical parameters and load characteristic data of the power distribution system; Based on the electrical parameters, determine whether the power distribution system has experienced an abnormality; If an abnormality is determined to have occurred in the power distribution system based on the electrical parameters, the reclosing circuit breaker is controlled to perform a tripping operation to disconnect the power supply from the load. The electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly are input into a pre-trained fault type identification model to obtain a fault type identification result. The anomaly type identification result includes one of the following fault types: power supply side instantaneous fault, power supply side permanent fault, and load side abnormal disturbance. If the fault type identification result is a momentary fault on the power supply side or an abnormal disturbance on the load side, the reclosing device is controlled to perform a reclosing operation according to the control strategy corresponding to the fault type identification result.

2. The method of claim 1, wherein, The step of determining whether the power distribution system has experienced an abnormality based on the electrical parameters includes: Obtain the reference current waveform of the power distribution system under normal operating conditions; Calculate the waveform similarity between the current current waveform in the electrical parameters and the reference current waveform; If the waveform similarity is lower than a preset similarity threshold, the power distribution system is determined to be abnormal.

3. The method of claim 1, wherein, Determining an anomaly in the power distribution system based on the aforementioned electrical parameters includes: If the voltage in the electrical parameters is detected to be less than a preset start-up threshold, it is determined that a voltage drop event has occurred in the power distribution system. Generate event features corresponding to the voltage drop event, the event features including voltage drop depth and event duration; Cluster analysis is performed based on the event characteristics and the abnormal event characteristics of each historical abnormal event in the pre-set historical voltage reduction feature library; If the event characteristics fall within any of the clusters of the historical abnormal events, then the configuration system is determined to be abnormal.

4. The method of claim 1, wherein, The step of inputting the electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly into a pre-trained fault type identification model to obtain the fault type identification result includes: Based on the electrical parameters within a preset time window before the occurrence of the anomaly, calculate the power supply side fault contribution weight and the load side disturbance contribution weight. The power supply side fault contribution weight, the load side disturbance contribution weight, the electrical parameters, and the load characteristic data are input into a pre-trained fault type identification model, and the fault type identification result is output.

5. The method of claim 4, wherein, The calculation of the power supply-side fault contribution weight and the load-side disturbance contribution weight based on the electrical parameters within a preset time window before the anomaly occurs includes: Modal decomposition was performed on the electrical parameters within a preset time window before the occurrence of the anomaly to obtain multiple intrinsic mode function components; Calculate the correlation coefficients between each intrinsic mode function component and the typical fault characteristic template of the power supply side and the typical disturbance characteristic template of the load side, respectively. The proportions of each intrinsic mode function component belonging to power supply side faults and load side disturbances are determined based on the correlation coefficients, and these proportions are used as the contribution weights of power supply side faults and load side disturbances.

6. The method of claim 1, wherein, The training process of the fault type identification model includes: Obtain a training sample set and an initial recognition model. The training sample set includes multiple sample data and a sample label for each sample data. The sample data includes electrical parameters and load characteristics at the time of occurrence of historical abnormal events. The initial recognition model is trained using the training sample set; If the initial identification model meets the preset training completion conditions, the initial identification model that meets the training completion conditions will be used as the fault type identification model.

7. The method of claim 1, wherein, The control strategy corresponding to the fault type identification result, which controls the reclosing circuit to perform a reclosing operation, includes: If the fault type identification result is a power supply side instantaneous fault, then the reclosing operation is controlled to be performed according to the first reclosing strategy, the first reclosing strategy including the first reclosing delay time; If the fault type identification result is an abnormal disturbance on the load side, then the reclosing operation is controlled to be performed according to the second reclosing strategy, which includes a second reclosing delay time. The second reclosing delay time is shorter than the first reclosing delay time.

8. A reclosing control device, characterized by Applied to a power distribution system, the power distribution system including a power source and a reclosing device, the input terminal of the reclosing device is connected to the power source, and the output terminal is connected to the load. The device includes: The acquisition unit is used to acquire electrical parameters and load characteristic data of the power distribution system; The judgment unit is used to determine whether an abnormality has occurred in the power distribution system based on the electrical parameters; The first control unit is configured to control the reclosing device to perform a tripping operation to disconnect the power supply from the load if it is determined that the power distribution system is abnormal based on the electrical parameters. The identification unit is used to input the electrical parameters and load characteristic data within a preset time window before the occurrence of the anomaly into a pre-trained fault type identification model to obtain a fault type identification result. The anomaly type identification result includes one of the following fault types: power supply side instantaneous fault, power supply side permanent fault, and load side abnormal disturbance. The second control unit is used to control the reclosing device to perform a reclosing operation according to the control strategy corresponding to the fault type identification result when the fault type identification result is a momentary fault on the power supply side or an abnormal disturbance on the load side.

9. A storage medium, characterized by The storage medium includes storage instructions, wherein, when the instructions are executed, the device containing the storage medium is controlled to perform the reclosing control method as described in any one of claims 1 to 7.

10. An electronic device, comprising: It includes a memory and one or more instructions, wherein one or more instructions are stored in the memory and configured to be executed by one or more processors as described in any one of claims 1 to 7.

Images