Intelligent circuit breaker parameter updating reliability data guarantee method and system and circuit breaker

Abstract

The invention relates to the technical field of power system automation and power distribution network relay protection, in particular to an intelligent circuit breaker parameter updating reliability data guarantee method and system and a circuit breaker. According to the method, distortion index calculation and threshold comparison are performed on fault recording data after the intelligent circuit breaker is switched off, recorded waves are divided into a first quality level, a second quality level and a third quality level, and the recorded waves of the second quality level are reconstructed based on a steady-state segment and undistorted sample points to obtain a calibration waveform aligned with original recording time; writing a quality mark in the event entry and associating a corresponding calibration record; and selecting corresponding recorded waves to form a verification event set before parameter updating, performing offline verification on the to-be-updated protection setting value, and prohibiting downloading of versions which do not meet preset selectivity, sensitivity and safety criteria. The influence of waveform distortion of a measurement link on a data driving analysis result can be reduced, and the reliability of the intelligent circuit breaker protection constant value and operation control model updating process is improved.

Description

Technical Field

[0001] This invention relates to the field of power system automation and distribution network relay protection technology, specifically to a method, system and circuit breaker for ensuring the reliability of parameter updates for intelligent circuit breakers, and particularly to data screening and parameter update verification technology based on fault recording data quality assessment and calibration. Background Technology

[0002] With the large-scale integration of distributed power sources and the increasing complexity of distribution network operation, smart circuit breakers in power distribution systems not only perform primary equipment functions such as fault isolation and reclosing, but also generally integrate functions such as fault recording, event logging, communication, and remote parameter setting. After a smart circuit breaker experiences a short circuit, ground fault, or other fault and trips, its fault recording data is typically used for post-fault analysis, protection setting verification, and evaluation of the effectiveness of operation control strategies, becoming an important data source supporting distribution network operation status analysis and parameter optimization.

[0003] In actual operation, the measurement link containing the primary current transformer, sampling circuit, and waveform recording device of the intelligent circuit breaker is prone to distortion in fault waveform recordings due to factors such as high current surges, core saturation, nonlinear loads, and rapid transient changes in fault conditions. These distortions include current waveform clipping, flattening, phase distortion, amplitude imbalance, and abnormal sampling in individual channels, leading to discrepancies between the recorded data and the actual electrical quantities. Current technologies primarily check the quality of fault waveform recordings from intelligent circuit breakers by examining triggering conditions, recording length, or simple amplitude ranges. They lack quantitative assessment and hierarchical management mechanisms for waveform distortion in the measurement link, and there is a lack of clear identification and processing strategies for waveform data with calibrable distortions. This results in waveform data of inconsistent quality often being mixed together for subsequent analysis and parameter tuning.

[0004] Against this backdrop, the inventors further discovered that as protection settings and operation control models increasingly rely on historical fault recording data for offline evaluation and version updates, when the recording quality is not effectively differentiated and managed, measurement link distortion or abnormal recordings are directly used for parameter optimization and model evaluation. This makes it difficult to grasp the reliability of parameter update conclusions and is also not conducive to timely identification of possible anomalies or performance degradation trends in the measurement link. Summary of the Invention

[0005] (i) The technical problem to be solved by the present invention is that after the intelligent circuit breaker trips due to protection action, the existing technology is difficult to objectively evaluate and classify the waveform distortion caused by the fault recording caused by the measurement link. There is a lack of a mechanism to effectively link the waveform recording quality with the subsequent protection setting and operation control model parameter update process. As a result, the fault recording data with severe distortion or unreliability may be directly used for parameter setting and model verification, which is not conducive to ensuring the reliability of parameter update conclusions, nor is it conducive to timely detection of potential operational risks caused by long-term degradation or abnormal accumulation of the measurement link.

[0006] (II) Technical Solution To address the aforementioned technical problems, this invention provides a data assurance method for ensuring the reliability of parameter updates in intelligent circuit breakers, applicable to intelligent circuit breakers connecting the main grid side and the load side, comprising the following steps: S1. After the intelligent circuit breaker trips due to protection action, acquire the multi-phase voltage and / or current fault recording data corresponding to this tripping event. The fault recording data includes at least the steady-state data segment within the preset time window before tripping and the fault transient data segment after tripping. S2, based on the current waveforms of each phase in the fault transient data segment, calculate at least one set of distortion indices to characterize the waveform distortion of the measurement link, compare the distortion indices with a preset threshold, and divide the fault recording data corresponding to this tripping event into recording quality levels that include at least a first quality level, a second quality level, and a third quality level. S3, For fault recording data classified as the second quality level, construct a distortion-free reference waveform using the sample points in the steady-state data segment and the fault transient data segment that do not exhibit distortion characteristics, reconstruct the sample points that exhibit distortion characteristics, and generate a calibration fault waveform that is aligned with the original recording waveform on the time axis. S4. Create an event entry for each tripping event, write a quality flag indicating the waveform quality level in the event entry, and associate the original fault waveform corresponding to the first quality level and the calibration fault waveform corresponding to the second quality level in the event entry respectively, and mark the fault waveform corresponding to the third quality level as data that cannot be used for subsequent data-driven analysis. S5, when the intelligent circuit breaker intends to update the protection settings or operation control model, based on the quality marker in the event entry, events with recorded waveform quality levels of the first quality level and the second quality level are selected from the historical tripping events to form a verification event set. The verification event set is used as the verification dataset for the protection settings or operation control model to be updated. Before the parameter update, the version to be updated is verified based on the verification dataset. When the verification result does not meet the preset selectivity, sensitivity and safety criteria, the version to be updated is prohibited from taking effect on the intelligent circuit breaker.

[0007] This invention calculates distortion indices and compares thresholds on fault recording data after a smart circuit breaker trips, classifying the recordings into a first quality level that can be used directly, a second quality level that can be calibrated, and a third quality level that should be discarded. For the second quality level recordings, a calibration waveform aligned with the original recording time is reconstructed based on the steady-state segment and undistorted sample points. Simultaneously, a quality flag is written into the event entry and associated with the corresponding original / calibrated recording. Based on this, before parameter updates, only the recordings corresponding to the first and second quality levels are selected to form a verification event set for offline verification of the protection settings or operation control model to be updated. Versions that do not meet preset selectivity, sensitivity, and safety criteria are prohibited from downloading. Through the combined application of these technical features, this invention can perform quality screening and necessary calibration on the data used to drive the update before parameter updates, reducing the impact of measurement link waveform distortion on the data-driven analysis results, and improving the reliability of the smart circuit breaker protection setting and operation control model update process.

[0008] Further, in step S2, the distortion index includes at least: a first distortion index, a second distortion index, and a third distortion index, wherein: The first distortion index is the degree of flattening of the amplitude of each phase current near the peak in the fault transient data segment, relative to the reference sinusoidal waveform fitted based on the steady-state data segment before the fault; The second distortion index is the phase difference between the fundamental components of each phase current in the pre-fault steady-state data segment and the fault transient data segment; The third distortion index is the amplitude imbalance between the fundamental components of each phase current in the fault transient data segment. By comparing the first distortion index, the second distortion index, and the third distortion index with their corresponding preset thresholds, fault recordings that meet all threshold requirements are classified as the first quality level, fault recordings that exceed the first set of thresholds but do not exceed the second set of thresholds are classified as the second quality level, and fault recordings that exceed the second set of thresholds are classified as the third quality level.

[0009] The three types of distortion indices described above characterize the distortion features of the measurement link from three dimensions: waveform amplitude shape, fundamental phase consistency, and amplitude symmetry of each phase. The first distortion index focuses on identifying flat-top current waveforms caused by transformer saturation and clipping; the second distortion index reflects abnormal drift of the fundamental phase before and after a fault; and the third distortion index measures the degree of imbalance in current amplitude among phases during a fault. By setting grouped thresholds for each distortion index and performing segmented comparisons, fault recordings can be classified under a unified criterion into a first quality level that can be used directly, a second quality level that can still be used for analysis after reasonable calibration, and a third quality level that should be removed in subsequent data-driven analysis. This allows for hierarchical management of data with different reliability levels during the recording quality assessment stage, providing a more reliable data foundation for subsequent parameter tuning, model training, and operational performance evaluation.

[0010] Furthermore, the preset threshold is adjusted based on at least one of the type, capacity, and access location of the distributed power source connected to the load side, so that the ratio of the number of first quality level events and third quality level events obtained by step S2 remains within a preset range under different distributed power source access scenarios.

[0011] By adjusting preset thresholds based on the type, capacity, and access location of distributed power sources, appropriate distortion criteria can be dynamically matched to address differences in short-circuit current distortion, transient characteristics, and measurement link disturbance intensity under different access scenarios. This allows the "strictness" of waveform quality assessment to adaptively adjust with changing operating conditions. By using the ratio of first-quality-level events to third-quality-level events as a statistical indicator and constraining this ratio within a preset range, a closed-loop correction of the criterion sensitivity and tolerance is achieved: a high ratio indicates an overly broad threshold, requiring improved distortion detection capabilities; a low ratio indicates an overly strict threshold, necessitating the avoidance of excessive misjudgments of severe distortion. Therefore, a generally consistent quality grading effect can be maintained under different distributed power source access methods and operating conditions, preventing the same set of threshold parameters from being too lenient or too conservative in certain scenarios. This helps maintain the stability and comparability of waveform quality assessment, providing a more robust foundation for subsequent data screening based on quality levels.

[0012] Furthermore, in step S3, calibrating the fault recording data corresponding to the second quality level includes: Based on the sampled value of at least one electrical quantity in the steady-state data segment before the fault, construct a linear equivalent model representing the equivalent impedance of the power system at the installation point of the smart circuit breaker and / or a distortion-free reference waveform of the current in each phase. In the fault transient data segment, sample points that do not exhibit saturation or clipping characteristics are identified, and these sample points are used as fitting constraints to perform parameter correction on the linear equivalent model and / or the distortion-free reference waveform. For sample points identified as having distortion characteristics, alternative sample values ​​are calculated using a modified linear equivalent model and / or a distortion-free reference waveform to obtain the calibration fault waveform.

[0013] Through the aforementioned modeling and reconstruction process, without altering the fault occurrence time, fault type, or the actual system response sequence, a reference waveform reflecting the equivalent impedance characteristics of the installation point and constrained by undistorted samples can replace the sampling points in the original waveform recording that are significantly affected by transformer saturation and clipping. This ensures that the calibrated fault waveform remains aligned with the actual event on the time axis, while also more closely approximating the real electrical process in terms of amplitude and waveform morphology. In this way, the representativeness of the original operating conditions and system topology is preserved, and the reliability of electrical quantities used for subsequent analysis and parameter verification is improved. This allows waveform recordings marked as second quality level to still serve as high-quality data for protection setting optimization and model evaluation after calibration.

[0014] Furthermore, in step S4, the event entry includes at least an event timestamp, a tripping reason, a waveform quality level marker, and the original fault waveform and / or calibration fault waveform corresponding to the quality level marker; When the event entries are uploaded to the upper-level monitoring system or used for local analysis, the waveform quality level label is used as a filtering condition, and only the fault waveforms corresponding to the first quality level and the second quality level are allowed to be used for protection setting optimization, fault identification model training and operation performance evaluation.

[0015] By simultaneously recording the timestamp, tripping reason, and original or calibrated fault waveform corresponding to each quality level in the event entries, the operating background and data reliability of each tripping event can be quickly traced back during subsequent analysis, and the structured binding of waveform data with quality tags can be achieved. When calling event entries from the upper-level monitoring system or local analysis, the waveform quality level tag is used as a mandatory filtering condition. Only fault waveforms corresponding to the first and second quality levels are allowed to participate in protection setting optimization, fault identification model training, and operational performance evaluation. This helps to eliminate severely distorted and unreliable data at the source, reduce the interference of abnormal measurement links on parameter tuning and data-driven models, and retain necessary traceability information to facilitate the verification and management of parameter adjustment processes by operation and maintenance personnel.

[0016] Furthermore, the parameter update verification process in step S5 includes: Without altering the historical fault records and actual system response, based on the protection settings and / or operation control model to be updated, the action behavior of the intelligent circuit breaker in various verification events is simulated to obtain the simulation action results. The simulated action results are compared with historical actual action results and preset selectivity, sensitivity and security criteria. When all events in the verification event set meet the preset criteria, the version to be updated is allowed to take effect on the smart circuit breaker; otherwise, the original version remains unchanged and a parameter update failure alarm message is generated.

[0017] In some current intelligent power distribution terminals and protection devices, protection settings and operation control models are often centrally calculated and frequently distributed by a higher-level system, or periodically adaptively adjusted based on long-term operating data. If a local verification process tailored to specific devices and actual fault conditions is lacking before updates, the new version parameters may disrupt the original selectivity and sensitivity matching relationships in certain fault scenarios when network topology, load structure, or distributed power output changes, or even introduce risks of maloperation or failure to operate that are difficult to detect in a timely manner. Through offline calculation and comparison based on the set of verification events described in this paragraph, while keeping the historical fault records and actual system response unchanged, the behavior of the version to be updated under typical operating conditions is replayed and reviewed. This is equivalent to introducing a layer of "trial operation" verification strongly correlated with the field operating conditions before the parameters are officially downloaded. Therefore, on the one hand, it can identify and intercept parameters or model versions that do not meet the established criteria in advance at the terminal side, avoiding new safety hazards to the operation of the distribution network caused by direct commissioning of incompatible configurations. On the other hand, it also provides operating units with a tool to quantitatively evaluate the performance of different version strategies, which is conducive to steadily promoting the iterative upgrade of intelligent circuit breaker protection settings and operation control models while ensuring the safety and selectivity of protection actions.

[0018] Furthermore, while marking the fault waveforms corresponding to the third quality level as data that cannot be used for parameter tuning, model training, and online optimization, the number of occurrences of the third quality level event within a preset statistical time window is recorded. When the number of occurrences reaches or exceeds a preset alarm threshold, an abnormal risk alarm message for the measurement link is generated to indicate that at least one of the primary current transformer, sampling circuit, or waveform recording device of the smart circuit breaker has a risk of long-term high stress or performance degradation.

[0019] By uniformly marking third-quality-level fault waveforms as unusable for parameter tuning, model training, and online optimization, severely distorted samples in the measurement link can be directly isolated at the data usage level. This prevents such abnormal data from being mixed into the training or validation sets used to optimize protection settings and operational control models, reducing the interference of "faulty data" on data-driven results from the source. Furthermore, by accumulating the occurrence of third-quality-level events within a preset statistical time window and triggering an alarm for measurement link anomalies when the alarm threshold is reached or exceeded, a health monitoring mechanism for primary current transformers, sampling circuits, and waveform recording devices is added on top of waveform recording quality management. When a device is under high stress for an extended period, the transformer frequently approaches saturation, the sampling link accuracy degrades, or the wiring environment deteriorates, the proportion of severely distorted waveforms will increase significantly, thus being promptly exposed through this statistical alarm. This not only prevents abnormal measurement links from "quietly" affecting parameter updates but also provides maintenance personnel with a reference for locating and troubleshooting potential faults or performance degradation in the measurement link.

[0020] Furthermore, during operation, the number of fault recording events classified into first quality level and third quality level within a preset statistical time window is periodically counted. Based on the statistical results, the preset thresholds corresponding to the first distortion index, second distortion index and third distortion index are adjusted so that the ratio of first quality level events to third quality level events is maintained within a preset range.

[0021] By periodically counting the number of events at the first and third quality levels within a preset statistical time window during operation, and adjusting the preset thresholds corresponding to each distortion index accordingly, a feedback tuning mechanism based on result distribution is essentially superimposed on the distortion criteria: when the proportion of third quality level events is too high, the system can appropriately relax the threshold to avoid too many normal or calibrable waveforms being misjudged as severely distorted; when the proportion of first quality level events is too high and the proportion of third quality level events is significantly low, the threshold can be appropriately tightened to improve the detection capability of potential measurement link distortions. By constraining the proportion of first and third quality level events within a preset range, the discrimination characteristics of waveform quality assessment can be kept basically consistent under different operating conditions and different distributed power supply access scenarios, reducing the "too strict" or "too loose" problems caused by fixed thresholds, making waveform quality classification both stable and possessing a certain degree of self-adaptability.

[0022] This invention also provides a data assurance system for the reliability of intelligent circuit breaker parameter updates, applied to intelligent circuit breakers connecting the main grid side and the load side, comprising: The fault recording acquisition unit is used to acquire multi-phase voltage and / or current fault recording data corresponding to the tripping event when the intelligent circuit breaker trips due to protection action. The data processing unit is signal-connected to the fault recording acquisition unit and is used to calculate the distortion index, classify the recording quality level, and calibrate the fault recording corresponding to the second quality level for the fault recording data, and write a quality mark indicating the recording quality level into the event entry. A storage and communication unit is used to store event entries containing the quality tags and to provide the event entries to the upper-level monitoring system and / or the local analysis module when needed; The parameter management unit is used to call historical event entries with recording quality levels of first and second quality levels when it receives protection settings and / or operation control models to be updated. It works with the data processing unit to execute the parameter update verification process in the data assurance method for the reliability of parameter updates of any of the above-mentioned smart circuit breakers on the fault recording data corresponding to the historical event entries, and controls whether the version to be updated is effective on the smart circuit breaker based on the verification results. The data processing unit and the parameter management unit are configured to perform the steps of the data assurance method for ensuring the reliability of parameter updates of any of the above-described intelligent circuit breakers.

[0023] Through the integrated configuration of the above units within the intelligent circuit breaker itself or in conjunction with them, the data assurance system of this invention can centrally manage the data required for fault waveform acquisition, quality assessment, calibration processing, quality marking, and parameter update verification at the device side. This allows some functions that were originally scattered across upper-level monitoring and offline analysis software to be implemented at the terminal layer closer to the primary equipment. The collaborative work of the data processing unit and the parameter management unit creates a closed-loop correlation between the waveform quality grading results and the parameter update verification process within the system. This facilitates the modular embedding of the waveform quality management and parameter update reliability control strategies proposed in this invention into existing intelligent circuit breaker products or distribution automation systems without altering existing communication protocols and upper-level system structures, thereby improving the convenience and maintainability of engineering implementation.

[0024] The present invention also provides an intelligent circuit breaker, including an intelligent circuit breaker body and an intelligent circuit breaker parameter update reliability data assurance system electrically connected to the intelligent circuit breaker body, wherein the data assurance system is the aforementioned intelligent circuit breaker parameter update reliability data assurance system.

[0025] (III) Beneficial effects of the present invention: The present invention can perform quality screening and necessary calibration on the data used to drive the update before the parameter update, reduce the impact of measurement link waveform distortion on the data-driven analysis results, and help improve the reliability of the update process of the protection setting and operation control model of the intelligent circuit breaker. Attached Figure Description

[0026] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0027] Figure 1 A flowchart illustrating a data assurance method for ensuring the reliability of intelligent circuit breaker parameter updates, provided in an embodiment of the present invention. Detailed Implementation

[0028] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Specific Implementation

[0029] This embodiment provides a method, system, and intelligent circuit breaker for ensuring the reliability of parameter updates, specifically based on data screening and parameter update verification using fault recording data quality assessment and calibration. In specific application scenarios, the intelligent circuit breaker is preferably installed on overhead line towers or in switching stations / substations of the distribution network. It is used to electrically connect the main grid side of the upstream power supply side with the downstream distribution lines and their load sides, and to perform functions such as overload, short circuit, and grounding fault detection, as well as tripping and reclosing control. The intelligent circuit breaker body typically includes a primary conductive circuit (e.g., an arc-extinguishing unit containing a vacuum arc-extinguishing chamber, stationary contacts, and moving contacts, and its incoming and outgoing terminals), a drive mechanism (e.g., a spring-operated mechanism or an electric operating mechanism), a current transformer for acquiring the primary side current, and optionally a voltage transformer for monitoring the primary side voltage. The secondary sides of the current transformer and the voltage transformer are connected to the protection and control unit of the circuit breaker via a sampling circuit. The protection and control unit internally includes a microprocessor, power supply module, analog sampling and A / D conversion circuits, digital input / output interfaces, and communication interfaces. These are used to digitize the sampled current and voltage signals, execute protection criteria and reclosing logic, and exchange information with the upper-level monitoring system. In this embodiment, the data assurance system for ensuring the reliability of intelligent circuit breaker parameter updates is preferably integrated into the hardware platform of the protection and control unit and implemented as a software functional module. Alternatively, a separate processing board can be set up within the same control cabinet and exchange data with the protection and control unit via an internal bus, thus achieving the solution of this invention without changing the primary equipment structure.

[0030] In this embodiment, the fault recording acquisition unit can consist of an analog sampling circuit, an A / D converter, and a matching recording control program connected to the secondary sides of the current transformer and the voltage transformer. When the protection and control unit detects a protection action and issues a tripping command, the fault recording acquisition unit samples the current signals of each phase according to a preset sampling frequency. In scenarios where voltage analysis is required, the voltage signals of each phase can be acquired simultaneously. The steady-state data segment within a preset time window before tripping and the transient fault data segment after tripping are continuously stored in the internal buffer, and then organized into a multi-phase fault recording record, which is sent to the data processing unit via the internal bus. The data processing unit can be implemented using the same microprocessor as the protection function. The program includes dedicated recording quality assessment and calibration processing tasks. Its input receives the multi-phase fault recording data output by the fault recording acquisition unit, and its output is connected to the interface circuit of the storage and communication unit and the parameter management unit. After receiving the fault recording, the data processing unit calls the preset algorithm to calculate the distortion index in sequence, classify the recording quality level according to the preset threshold, perform calibration and reconstruction based on the reference model for the recording determined to be of the second quality level, and generate an event entry containing the quality level mark, the original fault recording and / or calibration fault waveform reference information internally, and then write it into the event database maintained by the storage and communication unit through the interface.

[0031] The storage and communication unit may include non-volatile memory (e.g., flash memory), a cache for temporarily storing recent events, and a field communication module. The non-volatile memory stores multiple event entries in chronological order. Each event entry is associated with an event timestamp, tripping reason, waveform quality level marker, the corresponding original fault waveform file index, and, if calibration results exist, the calibration fault waveform file index. The communication module establishes a connection with the upper-level monitoring system or local analysis terminal through a preset communication protocol (e.g., a commonly used protocol for distribution automation). When receiving an external query request, it can select data that meets the request conditions from the event entries and upload it, or periodically push event entries containing quality markers according to internal strategies. The parameter management unit can share the same processor as the data processing unit, with independent task modules in software; alternatively, it can use a separate processor board, interacting with the storage and communication unit and the data processing unit via an internal bus. In this embodiment, when the parameter management unit detects a parameter update command from the host system or the local human-machine interface, it first retrieves historical event entries with waveform quality levels of first and second quality levels from the storage and communication unit, and reads the corresponding fault waveform data set to form the verification event set required for parameter update verification. Subsequently, it performs offline simulation on the fault waveform corresponding to each verification event for the protection settings and / or operation control model to be updated, obtains the simulated action results, and compares the simulation results with the actual action records at that time and the preset selectivity, sensitivity, and safety criteria. Based on the comparison results, it generates a "allow update" or "prohibit update" decision signal and feeds it back to the parameter management interface of the protection and control unit. The data processing unit and the parameter management unit cooperate in the above manner to form a closed loop within the same device for fault waveform quality assessment and calibration, event entry recording and screening, and parameter update verification based on high-quality data, thereby achieving local data assurance for the reliability of intelligent circuit breaker parameter updates.

[0032] Through the above structural layout, this embodiment adds a clear hardware carrying path to the original protection and control architecture of the intelligent circuit breaker, enabling the data screening and parameter update verification method based on fault recording quality assessment to operate in a closed loop on the field equipment, thereby providing a sufficient system foundation for the implementation of subsequent specific method steps.

[0033] The data assurance method for ensuring the reliability of intelligent circuit breaker parameter updates in this embodiment includes the following steps: S1. After the intelligent circuit breaker trips due to protection action, acquire the multi-phase voltage and / or current fault recording data corresponding to this tripping event. The fault recording data includes at least the steady-state data segment within the preset time window before tripping and the fault transient data segment after tripping. S2, based on the current waveforms of each phase in the fault transient data segment, calculate at least one set of distortion indices to characterize the waveform distortion of the measurement link, compare the distortion indices with a preset threshold, and divide the fault recording data corresponding to this tripping event into recording quality levels that include at least a first quality level, a second quality level, and a third quality level. In step S2, the distortion index includes at least: a first distortion index, a second distortion index, and a third distortion index, wherein: The first distortion index is the degree of flattening of the amplitude of each phase current near the peak in the fault transient data segment, relative to the reference sinusoidal waveform fitted based on the steady-state data segment before the fault; The second distortion index is the phase difference between the fundamental components of each phase current in the pre-fault steady-state data segment and the fault transient data segment; The third distortion index is the amplitude imbalance between the fundamental components of each phase current in the fault transient data segment. By comparing the first distortion index, the second distortion index, and the third distortion index with their corresponding preset thresholds, fault recordings that meet all threshold requirements are classified as the first quality level, fault recordings that exceed the first set of thresholds but do not exceed the second set of thresholds are classified as the second quality level, and fault recordings that exceed the second set of thresholds are classified as the third quality level.

[0034] The preset threshold is adjusted based on at least one of the type, capacity, and access location of the distributed power source connected to the load side, so that the ratio of the number of first quality level events and third quality level events obtained by step S2 remains within a preset range under different distributed power source access scenarios.

[0035] S3, For fault recording data classified as the second quality level, construct a distortion-free reference waveform using the sample points in the steady-state data segment and the fault transient data segment that do not exhibit distortion characteristics, reconstruct the sample points that exhibit distortion characteristics, and generate a calibration fault waveform that is aligned with the original recording waveform on the time axis. After completing the waveform acquisition and quality level classification in steps S1 and S2, for fault waveforms classified as second quality level, this embodiment restores the repairable distortion portion to a state close to that when the measurement link is unsaturated through waveform calibration. Specifically, the data processing unit locally calls the calibration module, first using sample points without distortion characteristics in the already divided steady-state data segment and fault transient data segment to fit the equivalent electrical characteristics of the power system near the installation point of the smart circuit breaker and the ideal response of each phase current, and then reconstructs the sample points with distortion characteristics based on this. Correspondingly, step S3 can be specifically implemented as follows: S3, For fault recording data classified as the second quality level, construct a distortion-free reference waveform using the sample points in the steady-state data segment and the fault transient data segment that do not exhibit distortion characteristics, reconstruct the sample points that exhibit distortion characteristics, and generate a calibration fault waveform that is aligned with the original recording waveform on the time axis. In step S3, calibrating the fault recording data corresponding to the second quality level includes: Based on the sampled value of at least one electrical quantity in the steady-state data segment before the fault, construct a linear equivalent model representing the equivalent impedance of the power system at the installation point of the smart circuit breaker and / or a distortion-free reference waveform of the current in each phase. In the fault transient data segment, sample points that do not exhibit saturation or clipping characteristics are identified, and these sample points are used as fitting constraints to perform parameter correction on the linear equivalent model and / or the distortion-free reference waveform. For sample points identified as having distortion characteristics, alternative sample values ​​are calculated using a modified linear equivalent model and / or a distortion-free reference waveform to obtain the calibration fault waveform.

[0036] When constructing a distortion-free reference waveform, the data processing unit can first use the sampled values ​​of each phase current in the steady-state data segment before the fault to obtain the fundamental amplitude, phase, and corresponding equivalent voltage source parameters of each phase current through discrete Fourier transform or least-squares fitting. Based on these parameters, combined with the rated voltage and short-circuit capacity of the same substation or feeder, a linear equivalent model representing the equivalent impedance of the power system at the installation point of the smart circuit breaker can be established. For example, in the phasor domain, the system can be simplified to a structure of voltage source series equivalent impedance, with the current reference value determined by the voltage phasor and impedance phasor; alternatively, in the time domain, the distortion-free reference waveform can be represented as a fundamental sinusoidal form i_ref(t) = I1 * sin(ω*t + φ1), where I1 and φ1 are obtained by fitting the steady-state data segment. In the fault transient data segment, the distortion index calculated in step S2 is used to mark the sample points in each phase current waveform that do not show saturation or clipping characteristics. The positions of these sample points in time and phase angle are matched one-to-one with the linear equivalent model and the distortion-free reference waveform to form a constraint set for parameter correction.

[0037] During parameter correction, the data processing unit can set the system equivalent impedance, distributed source equivalent injection, and current transformer proportional coefficient as parameters to be optimized. It constructs an objective function constrained by undistorted sample points, such as minimizing the sum of squared errors between the actual sampled values ​​and the calculated values ​​of the undistorted reference waveform at these sample points. By solving this optimization problem, the corrected linear equivalent model parameters and the corresponding undistorted reference waveform are obtained. Subsequently, for sample points identified as having distortion characteristics (e.g., sample points close to the current peak and with clipped amplitude), the data processing unit no longer directly uses their original sampled values. Instead, it calls the corrected linear equivalent model or the undistorted reference waveform at the same timestamp to calculate the theoretically expected current value. This theoretical value is then used as a substitute sampled value and written into the calibration fault waveform. By using this method of "fitting the model with the healthy section and reconstructing the distortion point with the model", on the one hand, the calibration waveform is kept completely aligned with the original recording on the time axis, ensuring that the original event time markers can be used for subsequent protection simulation or feature extraction. On the other hand, it can also eliminate distortion caused by measurement link saturation or clipping as much as possible, thereby improving the usability of the second quality level fault recording in parameter tuning and model training without changing the actual fault conditions of the system.

[0038] After waveform calibration is completed, this embodiment records and manages each tripping event using a unified event entry, enabling subsequent data filtering and retrieval processes to be easily completed based on quality markers. Specifically, after each tripping process, the data processing unit writes a new event entry to the storage and communication unit, which includes basic information about the event, its quality level, and available waveform data. Correspondingly, step S4 can be further defined as follows: S4. Create an event entry for each tripping event, write a quality flag indicating the waveform quality level in the event entry, and associate the original fault waveform corresponding to the first quality level and the calibration fault waveform corresponding to the second quality level in the event entry respectively, and mark the fault waveform corresponding to the third quality level as data that cannot be used for subsequent data-driven analysis. In step S4, the event entry includes at least an event timestamp, the reason for tripping, a waveform quality level marker, and the original fault waveform and / or calibration fault waveform corresponding to the quality level marker. When the event entries are uploaded to the upper-level monitoring system or used for local analysis, the waveform quality level label is used as a filtering condition, and only the fault waveforms corresponding to the first quality level and the second quality level are allowed to be used for protection setting optimization, fault identification model training and operation performance evaluation.

[0039] In a specific implementation, event entries can be designed as structured records containing index information and data reference information. The event timestamp can be given by the protection action time or the trip coil action time, and the tripping reason can be represented by the action type, protection name, or fault category code given by the protection logic. The waveform quality level mark can use integer encoding, for example, "1" corresponds to the first quality level, "2" corresponds to the second quality level, and "3" corresponds to the third quality level. To avoid ambiguity caused by "and / or", in this embodiment, it is preferable to save the original fault waveform at all quality levels, and when the quality level is the second quality level, an additional set of calibration fault waveforms aligned with the time axis of the original waveform is associated. The event entries store the storage locations or reference pointers of the original waveform data and the calibration waveform data, respectively. For the third quality level, only the original waveform and the quality level mark are saved, and a mark bit that cannot be used for parameter tuning and model training is set in the entry. When reading event entries, the supervisory control system or local analysis module only needs to filter based on the quality level marker and marker bit. This automatically ignores third-quality-level events, prioritizes feature extraction and parameter verification using calibration fault waveforms for second-quality-level events, and directly uses the original fault waveform for first-quality-level events. Through this design, the event entry structure supports unified management of events of different quality levels and eliminates ambiguity at the interface level regarding whether data can be used for subsequent analysis. This facilitates the stable integration of the parameter update reliability assurance process provided by this invention into complex engineering systems.

[0040] In this embodiment, regarding step S5, the parameter management unit can trigger the parameter update verification process upon receiving a new protection setting version or a new operation control model version. Specifically, the parameter management unit first retrieves all event entries established in step S4 from the storage and communication unit. Based on the waveform recording quality level markers written in the event entries, it retains only events with waveform recording quality levels of the first and second quality levels, and organizes the multi-phase voltage fault waveform recording data and / or multi-phase current fault waveform recording data corresponding to these events into a verification event set. For each verification event, without modifying the original waveform recording and actual action records, the parameter management unit calls the protection operation logic or operation control logic consistent with the actual operation of the intelligent circuit breaker on site. It inputs the fault waveform corresponding to the event in the order of sampling time, sequentially calculates the fundamental components, active and reactive components, impedance, or current amplitude of each phase current and voltage, and executes existing criteria such as overcurrent protection, distance protection, and zero-sequence protection accordingly. This yields simulated action results such as whether the version to be updated issues a trip command under the verification event, the timing of the trip command, and whether an alarm is issued. Subsequently, the parameter management unit compares the simulated action results with the historical actual action results recorded in the verification event entry item by item: for fault events that should have been activated, it checks whether the version to be updated operates reliably within the allowed action time band; for disturbance events or non-level fault events that should not have been activated, it checks whether the version to be updated remains inactive; at the same time, it combines the pre-set selection criteria (such as the coordination relationship with upstream and downstream protection), sensitivity criteria (such as the response capability to low fault current levels) and safety criteria (such as the ability to prevent false tripping under load fluctuation conditions) to comprehensively judge whether the performance of the version on the entire set of verification events meets the requirements. When any verification event fails to meet the requirements under any of the above criteria, the parameter management unit will determine the verification result as unsuccessful, retain the current protection settings and operation control model version unchanged, and generate a parameter update failure alarm message in the event entry and operation and maintenance interface to remind operation and maintenance personnel to review the version to be updated. Only when all events in the verification event set meet the preset requirements under the selectivity, sensitivity, and safety criteria will the version to be updated be marked as a verified version and allowed to take effect on the smart circuit breaker. In this way, closed-loop verification based on high-quality historical events is implemented for each parameter update without changing the actual on-site operating conditions, reducing the possibility of operational risks introduced by improper settings or algorithm defects.

[0041] In this embodiment, for the fault recording data classified as the third quality level in step S2, the data processing unit, while writing the quality mark in step S4, counts the number of occurrences of these events within a preset statistical time window. The parameter management unit determines whether there is an abnormal trend in the measurement link based on the counting results. When the cumulative number of events of the third quality level within the statistical time window reaches or exceeds the preset alarm threshold, the parameter management unit generates an alarm message for abnormal risk of the measurement link and writes the alarm message and prompt content together into the event entry and the operation and maintenance interface. This is used to indicate that at least one of the primary current transformer, sampling circuit, or recording device of the smart circuit breaker may have long-term high stress or performance degradation risks such as insulation aging, frequent saturation, and poor contact, which facilitates the operation personnel to arrange targeted on-site inspections and maintenance. Meanwhile, throughout the entire operation, the parameter management unit can summarize the number of fault recording events divided into first and third quality levels within a preset statistical time window according to a fixed statistical period, calculate the ratio of first-quality-level events to third-quality-level events, and adjust the preset thresholds of the first, second, and third distortion indices accordingly based on the deviation between this ratio and the target ratio range. This ensures that the sensitivity and tolerance of the recording quality level classification remain at a reasonable level when the grid operating conditions, load structure, or distributed power source access situation change. This avoids long-term overly strict classification that leads to a large amount of valid data being misjudged as unusable, and also avoids overly lenient classification that leads to severely distorted recordings being mixed into the data set used for parameter tuning and model verification. Through the above-mentioned parameter update verification process, the third quality level event alarm mechanism, and the threshold adaptive adjustment based on statistical results, this embodiment not only ensures the reliability of intelligent circuit breaker parameter updates, but also enables online monitoring of the health status of the measurement link. This makes the closed loop of waveform data acquisition, screening, use and feedback more complete, which is conducive to maintaining the long-term stable operation of the protection device in scenarios with complex distribution networks and widespread access to distributed power sources.

[0042] The above are preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made to the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A data assurance method for the reliability of parameter updates in intelligent circuit breakers, applied to intelligent circuit breakers connecting the main grid side and the load side, characterized in that, Including the following steps: S1. After the intelligent circuit breaker trips due to protection action, acquire the multi-phase voltage and / or current fault recording data corresponding to this tripping event. The fault recording data includes at least the steady-state data segment within the preset time window before tripping and the fault transient data segment after tripping. S2, based on the current waveforms of each phase in the fault transient data segment, calculate at least one set of distortion indices to characterize the waveform distortion of the measurement link, compare the distortion indices with a preset threshold, and divide the fault recording data corresponding to this tripping event into recording quality levels that include at least a first quality level, a second quality level, and a third quality level. S3, For fault recording data classified as the second quality level, construct a distortion-free reference waveform using the sample points in the steady-state data segment and the fault transient data segment that do not exhibit distortion characteristics, reconstruct the sample points that exhibit distortion characteristics, and generate a calibration fault waveform that is aligned with the original recording waveform on the time axis. S4. Create an event entry for each tripping event, write a quality flag indicating the waveform quality level in the event entry, and associate the original fault waveform corresponding to the first quality level and the calibration fault waveform corresponding to the second quality level in the event entry respectively, and mark the fault waveform corresponding to the third quality level as data that cannot be used for subsequent data-driven analysis. S5, when the intelligent circuit breaker intends to update the protection settings or operation control model, based on the quality marker in the event entry, events with recorded waveform quality levels of the first quality level and the second quality level are selected from the historical tripping events to form a verification event set. The verification event set is used as the verification dataset for the protection settings or operation control model to be updated. Before the parameter update, the version to be updated is verified based on the verification dataset. When the verification result does not meet the preset selectivity, sensitivity and safety criteria, the version to be updated is prohibited from taking effect on the intelligent circuit breaker.

2. The data assurance method for reliable parameter updates of intelligent circuit breakers according to claim 1, characterized in that, In step S2, the distortion index includes at least: a first distortion index, a second distortion index, and a third distortion index, wherein: The first distortion index is the degree of flattening of the amplitude of each phase current near the peak in the fault transient data segment, relative to the reference sinusoidal waveform fitted based on the steady-state data segment before the fault; The second distortion index is the phase difference between the fundamental components of each phase current in the pre-fault steady-state data segment and the fault transient data segment; The third distortion index is the amplitude imbalance between the fundamental components of each phase current in the fault transient data segment. By comparing the first distortion index, the second distortion index, and the third distortion index with their corresponding preset thresholds, fault recordings that meet all threshold requirements are classified as the first quality level, fault recordings that exceed the first set of thresholds but do not exceed the second set of thresholds are classified as the second quality level, and fault recordings that exceed the second set of thresholds are classified as the third quality level.

3. The data assurance method for reliable parameter updates of intelligent circuit breakers according to claim 2, characterized in that, The preset threshold is adjusted based on at least one of the type, capacity, and access location of the distributed power source connected to the load side, so that the ratio of the number of first quality level events and third quality level events obtained by step S2 remains within a preset range under different distributed power source access scenarios.

4. The data assurance method for reliable parameter updates of intelligent circuit breakers according to claim 1, characterized in that, In step S3, calibrating the fault recording data corresponding to the second quality level includes: Based on the sampled value of at least one electrical quantity in the steady-state data segment before the fault, construct a linear equivalent model representing the equivalent impedance of the power system at the installation point of the smart circuit breaker and / or a distortion-free reference waveform of the current in each phase. In the fault transient data segment, sample points that do not exhibit saturation or clipping characteristics are identified, and these sample points are used as fitting constraints to perform parameter correction on the linear equivalent model and / or the distortion-free reference waveform. For sample points identified as having distortion characteristics, alternative sample values ​​are calculated using a modified linear equivalent model and / or a distortion-free reference waveform to obtain the calibration fault waveform.

5. The data assurance method for reliable parameter updates of intelligent circuit breakers according to claim 1, characterized in that, In step S4, the event entry includes at least an event timestamp, the reason for tripping, a waveform quality level marker, and the original fault waveform and / or calibration fault waveform corresponding to the quality level marker. When the event entries are uploaded to the upper-level monitoring system or used for local analysis, the waveform quality level label is used as a filtering condition, and only the fault waveforms corresponding to the first quality level and the second quality level are allowed to be used for protection setting optimization, fault identification model training and operation performance evaluation.

6. The data assurance method for reliable parameter updates of intelligent circuit breakers according to claim 5, characterized in that, The parameter update verification process in step S5 includes: Without altering the historical fault records and actual system response, based on the protection settings and / or operation control model to be updated, the action behavior of the intelligent circuit breaker in various verification events is simulated to obtain the simulation action results. The simulated action results are compared with historical actual action results and preset selectivity, sensitivity and security criteria. When all events in the verification event set meet the preset criteria, the version to be updated is allowed to take effect on the smart circuit breaker; otherwise, the original version remains unchanged and a parameter update failure alarm message is generated.

7. The data assurance method for reliable parameter updates of intelligent circuit breakers according to claim 1, characterized in that, While marking the fault waveforms corresponding to the third quality level as data that cannot be used for parameter tuning, model training, and online optimization, the number of occurrences of the third quality level event within a preset statistical time window is recorded. When the number of occurrences reaches or exceeds a preset alarm threshold, an abnormal risk alarm message for the measurement link is generated to indicate that at least one of the primary current transformer, sampling circuit, or waveform recording device of the smart circuit breaker has a risk of long-term high stress or performance degradation.

8. The data assurance method for the reliability of intelligent circuit breaker parameter updates according to any one of claims 1 to 7, characterized in that, During operation, the number of fault recording events classified into first quality level and third quality level within a preset statistical time window is periodically counted. Based on the statistical results, the preset thresholds corresponding to the first distortion index, second distortion index and third distortion index are adjusted so that the ratio of first quality level events to third quality level events is maintained within a preset range.

9. A data assurance system for the reliability of parameter updates of intelligent circuit breakers, applied to intelligent circuit breakers connecting the main grid side and the load side, characterized in that, include: The fault recording acquisition unit is used to acquire multi-phase voltage and / or current fault recording data corresponding to the tripping event when the intelligent circuit breaker trips due to protection action. The data processing unit is signal-connected to the fault recording acquisition unit and is used to calculate the distortion index, classify the recording quality level, and calibrate the fault recording corresponding to the second quality level for the fault recording data, and write a quality mark indicating the recording quality level into the event entry. A storage and communication unit is used to store event entries containing the quality tags and to provide the event entries to the upper-level monitoring system and / or the local analysis module when needed; The parameter management unit is used to call historical event entries with recording quality levels of first and second quality levels when it receives protection settings and / or operation control models to be updated, and to cooperate with the data processing unit to perform the parameter update verification process as described in any one of claims 1 to 8 on the fault recording data corresponding to the historical event entries, and to control whether the version to be updated is effective on the smart circuit breaker according to the verification result. The data processing unit and the parameter management unit are configured to perform the steps of the method as described in any one of claims 1 to 8.

10. An intelligent circuit breaker, characterized in that, The system includes a smart circuit breaker body and a data assurance system for ensuring the reliability of smart circuit breaker parameter updates, which is electrically connected to the smart circuit breaker body. The data assurance system is the data assurance system for ensuring the reliability of smart circuit breaker parameter updates as described in claim 9.

Images