An outdoor high-voltage vacuum circuit breaker operation sequence optimization method and system

Abstract

The application discloses an outdoor high-voltage vacuum circuit breaker operation sequence optimization method and system, which generates a candidate operation sequence by combining power grid conditions and power saving targets, obtains an initial operation sequence through maintenance plan screening and phase selection tripping verification, traces a history with a load mis-tripping record of the sequence, identifies an attempted mis-operation hidden danger and implements a live-lock interlocking verification to generate a dynamic reordering sequence, arranges a hierarchical execution instruction according to the reordering sequence, traces a tripping and transient fluctuation, generates a state confirmation signal through medium recovery evaluation and in-place confirmation, re-evaluates remaining steps, further identifies an execution deviation, backtracks and calibrates a long-term false alarm channel and sets a threshold, finally determines a circuit breaker reset feasible state, identifies a transferable path and counts line loss, and outputs an optimized control instruction after a synergistic trade-off between power saving gain and operation risk, so that device state sensing and safety verification are throughout the operation arrangement process, and economy and safety are taken into account.

Description

Technical Field

[0001] This invention relates to the field of power automation technology, and in particular to a method and system for optimizing the operation sequence of outdoor high-voltage vacuum circuit breakers. Background Technology

[0002] Outdoor high-voltage vacuum circuit breakers are key equipment in power distribution networks, responsible for load switching and fault isolation. The rational arrangement of their operation sequence directly affects the safety and economy of power grid operation. A complete switching operation often involves the sequential operation of multiple circuit breakers. The operation sequence must not only meet the dispatching objectives of reducing line losses and improving power supply economy, but also avoid safety risks such as misoperation, and take into account the operating status of the equipment itself. Existing arrangements mostly rely on the experience and judgment of dispatchers and fixed operation tickets, making it difficult to achieve dynamic coordination among multiple objectives.

[0003] Existing technologies have several shortcomings in addressing this problem. On the one hand, equipment condition assessment and operational sequence arrangement are disconnected. Information such as circuit breaker insulation recovery status, mechanical action deviations, and alarm system reliability are not effectively incorporated into operational decisions, making it difficult to quantify and grasp the safety margin of operational sequences. On the other hand, energy-saving targets and operational risks are usually handled separately, lacking a means to coordinate and balance the two on the same scale. Often, operational risks are increased in pursuit of economy, or considerable energy-saving potential is sacrificed for conservative stability. In addition, there is a lack of mechanisms to proactively identify hidden degradation accumulated by equipment during long-term operation, such as contact erosion and falsely low settings, from historical operational data, resulting in some equipment in critical condition not being identified in a timely manner. Summary of the Invention

[0004] This invention discloses a method and system for optimizing the operation sequence of outdoor high-voltage vacuum circuit breakers. The aim is to integrate the insulation, mechanical and electrical life status perception of the circuit breaker into the operation sequence arrangement, and to ensure operational safety step by step through phase selection trip verification, interlock conflict identification and medium recovery assessment. The alarm channels with long-term false alarms are retrospectively calibrated and the setting threshold is adjusted. The power saving gain and operation risk are synergistically balanced, and finally the optimized control command that takes into account both economy and safety is output, providing a reliable decision basis for the switching operation of the distribution network.

[0005] A first aspect of the present invention discloses a method for optimizing the operation sequence of an outdoor high-voltage vacuum circuit breaker, comprising: Real-time power grid operation data is matched and analyzed with energy-saving target parameters to generate a candidate operation sequence list. Based on the candidate operation sequence list, maintenance plan postponement flags are identified to obtain a preferred operation sequence group. The initial operation sequence is obtained by performing phase selection and circuit breaker verification through the preferred operation sequence group. For the initial operation sequence, a risk node table is generated by tracking the load-bearing erroneous trip records. Based on the risk node table, potential erroneous operation hazards are identified and hazard circuit breaker groups are obtained. Through the hazard circuit breaker groups, live interlocking and interlocking verification is performed to generate a dynamic rearrangement sequence. Based on the dynamic rearrangement sequence, hierarchical instruction arrangement is carried out to output the execution instruction group. The execution instruction group is used to track the transient fluctuations of the circuit breaker restart and generate a status confirmation signal. The remaining steps are re-estimated based on the status confirmation signal to construct a continued status record. An abnormal alarm signal is obtained by identifying the execution deviation from the continued status record. The abnormal alarm signal is used to identify false alarms and backtrack to obtain threshold calibration parameters. The threshold calibration parameters are used to reset the alarm threshold to form a pause alarm command. Based on the pause alarm command, the circuit breaker reset feasibility status is determined to obtain a reset criterion group. Based on the reset criterion group, the transferable path is identified, line loss is statistically analyzed, and power saving benefit data is constructed. Through the power saving benefit data, the power saving gain and risk of the operation sequence are weighed together to output optimized control commands.

[0006] A second aspect of the present invention discloses an outdoor high-voltage vacuum circuit breaker operation sequence optimization system, comprising: The sequence generation unit is used to acquire real-time power grid operating condition data and power saving target parameters, match and analyze them to generate a candidate operation sequence list, identify the maintenance plan postponement flag according to the candidate operation sequence list to obtain the preferred operation sequence group, and perform phase selection and circuit breaker verification through the preferred operation sequence group to obtain the initial operation sequence. The sequence construction unit is used to perform load-bearing erroneous tripping record tracking to generate a risk node table for the initial operation sequence, identify potential erroneous operation hazards based on the risk node table, obtain hazard circuit breaker groups, and generate a dynamic rearrangement sequence by implementing live interlocking and interlocking verification through the hazard circuit breaker groups. The execution confirmation unit is used to perform hierarchical instruction arrangement and output execution instruction group according to the dynamic rearrangement sequence, use the execution instruction group to track the transient fluctuation of the circuit breaker restart to generate a status confirmation signal, and re-estimate the remaining steps based on the status confirmation signal to construct a continued status record. An abnormal calibration unit is used to identify the execution deviation from the continued status record to obtain an abnormal alarm signal, use the abnormal alarm signal to identify false alarms and backtrack to obtain threshold calibration parameters, and use the threshold calibration parameters to set the alarm threshold reset to form a pause alarm command. The instruction output unit is used to determine the circuit breaker reset feasibility status based on the pause alarm instruction to obtain a reset criterion group, identify transferable paths based on the reset criterion group to statistically analyze line losses and construct power saving benefit data, and output optimized control instructions by performing a collaborative trade-off between power saving gain and risk in the operation sequence based on the power saving benefit data.

[0007] The beneficial effects of this invention are reflected in the following points: First, this invention moves the verification of the circuit breaker's own state to the operation sequence arrangement stage: through phase selection trip verification and maintenance plan screening, the mechanical wear, opening distance drift, and overvoltage suppression requirements of the circuit breaker are incorporated into the process of determining the operating phase angle; for historical records of erroneous tripping under load, potential hazards of attempted misoperation are identified, and live-line interlocking verification is implemented, dynamically rearranging conflicting operation steps. The operation sequence determined in this way is no longer solely based on grid operating conditions and dispatch objectives; the safety margin can be quantified and grasped at the beginning of the scheme determination. Second, whether the equipment state has changed after the operation is completed is difficult to know based solely on pre-execution judgment. This invention tracks the transient fluctuations of reignition between contacts after tripping, calculates the rate of change of reignition frequency, and cumulatively verifies contact erosion to obtain the equipment's dielectric recovery rating. Then, it verifies the degree of tripping completion and analyzes and confirms the interval. Steps with questionable execution states are promptly paused and the remaining steps are reassessed, enabling changes in equipment state to be detected and responded to in real time during execution, without waiting until the next maintenance. Finally, this invention does not prioritize either energy saving gains or operational risks: it identifies and adjusts alarm channels with long-term false alarms to suppress the interference of false alarms on subsequent judgments; it determines the feasibility of circuit breaker reset, identifies transferable power paths, and calculates line losses, mapping both objectives to the same trade-off scale for optimal selection. The final optimized control command output must meet requirements in both economic and safety dimensions, rather than sacrificing one for the other. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating an optimized operation sequence method for an outdoor high-voltage vacuum circuit breaker according to the present invention.

[0009] Figure 2 This is a schematic diagram of the gap drift and overvoltage suppression verification of the present invention.

[0010] Figure 3 This is a schematic diagram of the reignition frequency sequence and ablation cumulative rating of the present invention.

[0011] Figure 4 This is a structural block diagram of an outdoor high-voltage vacuum circuit breaker operation sequence optimization system according to the present invention.

[0012] Wherein: 1-Power frequency voltage waveform; 2-Theoretical contact separation time; 3-Actual contact separation time; 4-Open gap drift; 5-Theoretical expected arcing period; 6-Actual expected arcing period; 7-Dielectric recovery curve; 8-Transient recovery voltage curve; 9-Overvoltage margin; 10-Margin critical point; 11-Reignition frequency sequence; 12-Reignition frequency change rate; 13-Cumulative ablation amount; 14-Upper limit of normal range; 15-Scrap threshold; 16-Watch range; 17-Cumulative ablation increment; 18-Maintenance boundary marker; 19-Rate-ablation control range; 20-Dielectric recovery rating zone. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0015] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0016] The technical solutions of the embodiments of this application will be described below.

[0017] like Figure 1 As shown, this embodiment of the invention provides a method for optimizing the operation sequence of an outdoor high-voltage vacuum circuit breaker, including the following steps S101-S105: Step S101: Obtain real-time power grid operating condition data and match it with power saving target parameters to generate a candidate operation sequence table. Based on the candidate operation sequence table, identify the maintenance plan postponement flag to obtain the preferred operation sequence group. Perform phase selection and circuit breaker verification through the preferred operation sequence group to obtain the initial operation sequence.

[0018] Specifically, real-time grid operation data is acquired and matched with energy-saving target parameters to generate a candidate operation sequence table. Real-time grid operation data is collected by the dispatch automation system every 15 minutes, covering measured load current, bus voltage, and power flow direction for each outgoing line. The collection time is strictly synchronized with the operator console system clock to avoid data mismatch. The energy-saving target parameter is the line loss rate reduction indicator issued by the dispatch center for the current month, allocated to each outgoing line according to the annual measured line loss ratio. Matching analysis verifies the candidate operation sequence item by item: each candidate corresponds to a set of switching operation combinations planned to complete this switching task. The expected load transfer direction is first checked against the measured power flow direction of the same outgoing line in the real-time grid operation data. Opposite directions are directly eliminated; for those with the same direction, the expected loss reduction contribution is divided by the number of switching operations to obtain a matching score. A higher ratio indicates a more sufficient loss reduction per unit operation. This score is also constrained by the loss reduction indicator after the energy-saving target parameters are allocated to each outgoing line. The candidate operation sequence table includes all verified candidate records in descending order of matching scores. Each record carries the matching score, switch combination number, target time of switch operation, and expected loss reduction contribution. When scores are tied, they are sorted by the number of operations performed, from least to most. For individual outgoing lines, due to communication interruptions, real-time power grid operating condition data lacks power flow direction readings. These candidates are marked as data pending in the candidate operation sequence table and are not eliminated until the next cycle. For candidates whose switches are under maintenance lockout, there is no measured power flow data available. However, their matching scores are still included in the candidate operation sequence table. Their expected loss reduction contribution is changed to the difference between the theoretically calculated line loss before and after load transfer. The power flow data used for theoretical calculation and the real-time power grid operating condition data used for matching are taken from the same data collection.

[0019] The optimal operation sequence group is obtained by identifying the maintenance plan postponement flag based on the candidate operation sequence table. The postponement flag is determined by checking whether the target time of the switch operation for each record in the candidate operation sequence table falls within the power outage maintenance window registered in the maintenance plan system: if the target time overlaps with the start and end interval of an approved but not yet executed maintenance record, the flag is set to postpone; otherwise, it is set to normal. Approved but not yet effective maintenance records are not included in this overlap determination. The candidate operation sequence table is checked against the overlap between the target time and the maintenance window, using the original matching score sorting in the table without re-scoring. Records with pending data are not checked in this round and are skipped to the next record. Candidates marked as postponed are simultaneously marked with a postponement status in the candidate operation sequence table, and their operation target time is postponed to the most recent executable cycle after the maintenance window ends. The preferred operation sequence group consists of candidates marked as normal and those that, after postponement, still fall within the permitted operation period of the power grid dispatch. When constructing the group, two fields are added to the original record: a postponement flag and a target time after postponement. For records that are not postponed, this field is left blank. Candidates whose target time after postponement exceeds the permitted operation period are not included in the preferred operation sequence group. For candidates whose original target time overlaps with two maintenance windows, the postponement time is the later ending time of the two windows. If the candidate still overlaps with a third window, this rule is repeated to avoid secondary conflicts caused by falling into another window after postponement. The start and end times of the maintenance windows are taken from the latest approved records in the maintenance plan system. The judgment time and the reading time may differ by several days; the construction of the preferred operation sequence group always uses the latest reading value at the judgment time.

[0020] In some embodiments, obtaining the initial operation sequence by performing phase selection trip verification through the preferred operation sequence group includes: calculating the target phase angle of each circuit breaker tripping for the preferred operation sequence group to form a phase selection window table; performing mechanical wear opening distance drift tracking based on the phase selection window table to obtain the expected arcing period; performing overvoltage suppression verification according to the expected arcing period to output the phase selection verification conclusion; and filling the operation steps with the phase selection verification conclusion to obtain the initial operation sequence.

[0021] A phase selection window table is generated by calculating the target phase angle of each circuit breaker for tripping in the preferred operation sequence group. The calculation is based on the phases to be tripped contained in the corresponding switch operation combination for each record in the preferred operation sequence group. Only the phases to be tripped are calculated; phases already tripped or not involved in this operation are not calculated to avoid irrelevant phases consuming calculation resources. For example... Figure 2As shown, the phase angle is based on the zero-crossing point of the power frequency voltage waveform 1 of the phase to be opened. The zero-crossing point is calculated in real time from the waveform on the secondary side of the voltage transformer, and the refresh cycle is consistent with the instruction cycle issued by the preferred operation sequence group. The calculation is applied as θ=ω·(t_a-t_z): t_a is the planned opening time of the switch, i.e., the theoretical contact separation time 2, which is directly taken from the corresponding record of the preferred operation sequence group. If it has been postponed, the postponed time is taken; t_z is the most recent zero-crossing time of the same phase of the power frequency voltage waveform 1 before t_a; ω is the power frequency angular frequency, and the result is converted from radians to electrical angles and retained to one decimal place. In the phase selection window table, the phase angles of multiple phases to be opened involving the same switch combination are stored side by side according to phase, and the phase angles of each phase are calculated independently. When a 220kV outgoing line is tripped in three phases, if the voltage transformer of phase C temporarily loses synchronization and the zero-crossing point cannot be read, the target phase angle for that phase tripping will be temporarily missing in this record of the phase selection window table. The phase angles of phases A and B will still be calculated and added to the table as usual. The missing phase angle will be supplemented in the next calculation cycle after the transformers regain synchronization. Each record in the phase selection window table also carries the original sorting position of the corresponding switch combination in the preferred operation sequence group, which, together with the switch combination number, uniquely identifies the record's affiliation in the entire operation sequence. The reason why t_z is taken as the zero-crossing point before t_a rather than after it in the formula is to avoid θ being calculated as a negative value due to the wrong reference point selection, which would render the entire phase angle of that combination unusable and require remeasurement.

[0022] Based on the phase selection window table, mechanical wear drift tracking is used to obtain the expected arcing period. Each time the circuit breaker completes a trip, the wear of the operating mechanism linkage causes a slight increase in the actual arcing distance relative to the factory calibration value. The target phase angle given in the phase selection window table corresponds to the theoretical arcing start time under the factory calibration arcing distance, which is different from the actual start time after adding the wear increment. The tracking first extracts the cumulative number of trips for each switch combination recorded in the phase selection window table from the equipment ledger as the wear estimation input. The current expected arcing distance increases synchronously with the factory calibration distance by the product of the cumulative number of trips and the wear coefficient. The wear coefficient is calibrated by the factory mechanical life test and reflects the average contribution of each trip to the arcing distance. The current expected opening distance, relative to the factory-calibrated opening distance, is the opening distance drift 4. The delay of the actual contact separation time 3 relative to the theoretical contact separation time 2 is proportional to the opening distance drift 4. The larger the opening distance drift 4, the greater the delay. The delayed actual contact separation time 3 is the arc initiation reference point, which is the direct source of the starting point of the actual expected arc duration 6. The arc duration is taken as the difference between the arc extinguishing time corresponding to the current expected opening distance and the starting reference point on the medium recovery characteristic curve of this model. This curve is obtained from the factory type test at different opening distances and is retrieved by indexing the opening distance. The theoretical expected arcing period 5 is determined starting from the theoretical contact separation time 2 under the factory-calibrated opening distance; the expected arcing period (actual expected arcing period 6) is determined by the arcing start reference point and the arcing duration. The starting point is the delayed actual contact separation time 3, and the ending point is the starting point plus the duration. Both fall within the same power frequency cycle. When the cumulative number of trips exceeds the mid-term maintenance threshold given by the mechanical life test, the wear coefficient is changed to the recalibrated value after maintenance, and the expected arcing period is recalculated according to the new coefficient.

[0023] The overvoltage suppression verification outputs the phase selection verification conclusion based on the expected arcing period. The verification passes when the dielectric recovery curve 7 remains above the transient recovery voltage curve 8 throughout the actual expected arcing period 6. The judgment is based on whether the dielectric recovery curve 7 is higher than the transient recovery voltage curve 8 throughout the arc initiation to arc extinction interval. The transient recovery voltage curve 8 is calculated according to the transient recovery voltage standard method based on the line-to-ground capacitance current and the system-side short-circuit capacity. The dielectric recovery curve 7 is the characteristic curve under the corresponding opening distance during the actual expected arcing period 6. The two curves are compared and subtracted point by point according to the same time reference during the arc initiation to arc extinction interval. The overvoltage margin 9 is the difference between the dielectric recovery curve 7 and the transient recovery voltage curve 8 at the same moment. It is calculated point by point for all sampling points during the actual expected arcing period 6. The risk threshold is determined when the overvoltage margin 9 first drops to the statistical lower bound at the margin critical point 10. If the overvoltage margin 9 is negative at any sampling point, the phase is determined to have a risk of reignition, so as to leave a safety margin to cope with the fluctuation of operating conditions. The phase selection verification results are given separately for each switch combination, indicating whether the overvoltage suppression requirements are met or not. If met, the sampling time of the minimum margin is recorded for traceability, so as to check which time was closest to the critical point. If the dielectric recovery curve fails to catch up with the recovery voltage curve at the end of the expected arcing period, it is judged as unmet. When the same switch combination involves multiple phases to be opened, the verification is performed for each phase separately. The overall condition is considered met only if all phases meet the phase selection verification results. If any phase fails, the entire condition is judged as unmet, and the specific phase that failed is marked for subsequent phase angle remeasurement.

[0024] The initial operation sequence is obtained by backfilling the phase selection verification conclusion into the operation steps. Based on the original switch combinations and order of the preferred operation sequence group, the initial operation sequence backfills the phase selection verification conclusion corresponding to each switch combination into the operation steps for that combination. The backfilled content includes two items: the finally determined target phase angle for tripping and the overvoltage suppression judgment result. These two items are stored side-by-side within the same operation step. Taking a 220 kV outgoing circuit breaker as an example, its A-phase target phase angle after backfilling is 87.6 degrees, and the overvoltage suppression judgment result is satisfied. Both are placed in the record of this step for the execution terminal to read at once. For switch combinations whose phase selection verification conclusion is satisfied, the operation steps directly adopt the target phase angle given in the expected arcing period calculation stage without modification. For those determined not to be satisfied and whose phase angle has been recalculated, the operation steps adopt the recalculated phase angle. The phase angle corresponding to the original phase selection verification conclusion is retained in the operation step history field and does not participate in this execution, only for traceability. In the initial operation sequence, the target operation time of each switch combination corresponds one-to-one with the verified tripping phase angle. The arrangement order follows the original sorting position in the preferred operation sequence group. During the backfilling process, only the content is filled in, maintaining the original order. For switch combinations whose overvoltage margin is still negative after two consecutive rounds of recalculation, their operation steps are marked with a manual review mark and are temporarily excluded from the automatic execution scope of the initial operation sequence. They will be added after manual confirmation. The final operation time of each switch combination in the initial operation sequence is determined by positioning its target time in the preferred operation sequence group (taking the delayed time if it has been delayed) to the corresponding power frequency cycle. Then, the specific execution time is calculated within that cycle based on the verified tripping target phase angle. Only after both are determined is it written into this record. For combinations involving multiple phases, the execution time of each phase is calculated and written separately, with each phase forming an independent record.

[0025] Step S102: For the initial operation sequence, perform load-bearing erroneous trip record tracking to generate a risk node table, identify potential erroneous operation hazards based on the risk node table, obtain the hazard circuit breaker group, and generate a dynamic rearrangement sequence by implementing live interlocking and verification through the hazard circuit breaker group.

[0026] In some embodiments, the step of generating a risk node table by tracking load-bearing erroneous tripping records for the initial operation sequence includes: tracking historical load-bearing erroneous tripping records from the initial operation sequence to generate an erroneous tripping record library; identifying critical events approaching an action threshold based on the erroneous tripping record library to obtain a set of attempted erroneous operations; using the set of attempted erroneous operations to perform critical frequency weighting to form risk node weights; and classifying and archiving the risk node weights to generate a risk node table.

[0027] A database of erroneous tripping records is generated by tracing historical records of erroneous tripping under load from the initial operation sequence. Each switch combination in the initial operation sequence corresponds to a historical operation record in the equipment ledger, sourced from the operation event log automatically generated by the substation automation system each time a switch is operated, without relying on manual entry. Entries marked as erroneous tripping under load are screened one by one, covering all records of that combination within the past five years. For combinations put into operation less than five years ago, records are truncated based on their actual operation time. The determination of erroneous tripping under load requires the simultaneous fulfillment of two conditions: the corresponding circuit current is not zero when the tripping action occurs, and the reclosing or alarm signal of the protection device is triggered after the tripping. Both conditions are directly read from the log fields without secondary calculation. If either condition is missing, it is not considered an erroneous tripping under load. For example, if the circuit current is exactly zero when a switch combination is tripped, even if a reclosing signal is subsequently triggered, it will not be included in this determination because the current condition is not met. Conversely, if the current is not zero but no protection action is triggered, it will also not be considered. The erroneous tripping record database collects all historical entries that meet the judgment criteria, categorized by switch combination number. Each entry includes three fields: the time of occurrence, the measured value of the circuit current at the time of tripping, and the type of subsequent protection action. If multiple erroneous tripping records occur for the same combination on the same day, they are retained sequentially according to their time. Missing records in the ledger prior to the year the automation system was put into operation are not considered blank erroneous operation data; they only indicate that no electronic records were retained for that period, and are unrelated to whether the combination's operation is currently scheduled in the initial operation sequence. For switch combinations without records meeting the judgment criteria in the past five years, an empty record is created in the erroneous tripping record database as a placeholder. Empty records are not included in subsequent statistics; they are only used to prevent subsequent steps from mistakenly judging that the combination lacks ledger data.

[0028] A set of near-miss maloperation records is obtained by identifying critical events approaching the operating threshold based on the maloperation record database. During record-by-record comparison, the measured circuit current at the time of tripping in each record is compared to the rated operating threshold current of the switch combination. 85% of the threshold current is taken as the critical lower limit. This ratio originates from the unified regulations for maloperation warning intervals in relay protection procedures, and this unified ratio is shared across different voltage levels. The rated operating threshold current is read from the protection setting sheet according to the voltage level and model of the switch combination. The read value must be verified to be consistent with the currently used setting version of the combination to avoid misusing obsolete old settings. Records where the measured current falls between the critical lower limit and the rated threshold are judged as critical events; currents below the critical lower limit are not counted and are considered normal fluctuations; currents exactly equal to the critical lower limit are recorded as critical events. All records deemed critical events are included in the attempted misoperation set. In addition to retaining the original three fields in the erroneous tripping record database, a critical ratio field is added. The critical ratio is the ratio of the measured current to the rated threshold, used for manual review to determine how close the event is to the threshold. Critical events are still arranged by the time of occurrence, not by the critical ratio. For switch combinations whose rated thresholds have recently changed due to a change in protection settings, the latest changed settings are used for comparison. The update timestamp of the new settings must also be checked for consistency with the acquisition time of the record in the erroneous tripping record database to prevent cross-use of old and new settings. Switch combinations without any records in the attempted misoperation set are considered to have not experienced critical events in the past five years. This determination is unrelated to whether the combination has been included in the erroneous tripping record database. Even if the database contains records of the combination, it will not be included in the attempted misoperation set if the measured current has not reached the critical lower limit.

[0029] Risk node weights are generated by weighting the critical frequency using the set of near-miss operations. The near-miss operation set is grouped and statistically analyzed by switch combination. The number of critical events within the most recent two years for the same combination is recorded as its critical frequency. The statistical window is fixed at two years and does not expand with the current time. The start and end of the window are calculated backwards from the date of statistical execution, not aligned with calendar years. Records spanning maintenance periods are included in this window. The risk node weight is calculated as W = f × (1 + r): f is the critical frequency of the combination given by the near-miss operation set; r is the time decay correction ratio from the most recent critical event of the combination to the current time. The shorter the interval, the larger the value of r, and the value is zero when the interval is one year. The two-year window is used to statistically analyze the frequency level, while the one-year criterion is only used to characterize the recent concentration. The two reflect the cumulative amount and urgency of potential hazards, respectively, and are not interchangeable. W is the desired risk node weight, taking into account both frequency level and recent concentration. The specific values ​​of 'r' within one year are given according to a tiered reference table derived from the experience-based reference table provided in the equipment defect statistical reports of the past three years. The reference table is updated quarterly to keep track of the quarterly changes in equipment deterioration trends. Historical weights calculated before the quarterly update are not retrospectively modified due to changes in the reference table. Risk node weights are output for each switch combination. The higher the weight, the more concentrated the frequency of critical events in the recent period and the more obvious the hidden dangers are exposed. For switch combinations with a critical frequency of zero, the risk node weight is directly set to zero and does not enter the tiered correction process. For combinations where the judgment criteria for some near misses and malfunctions have changed due to protection setting adjustments during the statistical period, when recalculating the critical frequency, all historical records are rechecked using the adjusted standard, without mixing the old and new standards. The verification results are retained together with the critical frequencies before the adjustment for comparison.

[0030] Risk node weights are classified and archived to generate a risk node table. Risk node weights are divided into three levels: high, medium, and low, based on numerical ranges. The interval boundaries are determined by the classification standards given in the equipment defect statistical reports of the past three years, based on actual measured distribution. The three-level boundaries are set separately for different equipment types, and the same classification standard is not universally applicable across models. Boundary values ​​are checked periodically as the reports are updated. The risk node table includes all classified records by switch combination number. Each record carries two fields: the original risk node weight value and the corresponding classification result. The value and classification result are stored separately for easy retrieval at different granularities: coarse-grained classification for quick screening and fine-grained values ​​for in-depth analysis. Switch combinations classified as high are marked with a priority review indicator for priority review during maintenance planning; combinations classified as medium are not marked with this indicator, but the original values ​​are still fully retained for later reference. If a risk node weight falls exactly on the boundary value of two classification intervals, it is classified into the higher level according to the principle of "choose the higher level," and boundary values ​​are always assigned upwards. The rationale is that it is better to conduct an extra review than to miss potential hazards. The classification boundaries in the risk node table are updated quarterly with the latest equipment defect statistics report. After the boundary update, the classification results of archived records are recalculated synchronously, and the new boundary results shall prevail. The calculation timestamp is recorded in the corresponding record. The original values ​​of the risk node weights themselves do not change due to the boundary adjustment; only the classification corresponding to the values ​​changes. Switch combinations that have never appeared in the risk node weight statistics results before are classified as low-level in this update, and the corresponding data in the risk node table are included in the label for the first time.

[0031] Based on the risk node table, potential operational hazards are identified, and circuit breaker groups with potential hazards are obtained. Switch combinations with a risk node table classification of high or medium are considered to have potential operational hazards. Combinations with a low classification are not considered hazards and will not proceed to the subsequent interlocking verification stage, thus narrowing the range of combinations to be verified. No additional numerical threshold is set for hazard identification; the classification results already completed in the risk node table are directly used to avoid repeated verification of the same batch of data by two sets of standards, leading to contradictory conclusions. The potential hazard circuit breaker group includes all switch combinations identified as hazards. Each entry carries the classification result of the combination in the risk node table and a priority verification identifier (if any). The recording order follows the original numbering order in the risk node table without reordering, allowing for a line-by-line comparison with the risk node table. For switch combinations with multiple candidate operations in the initial operation sequence, once a hazard is identified, all its candidate operations are included in the subsequent verification scope of the potential hazard circuit breaker group, rather than being judged individually for each candidate. This ensures that no individual items under the same combination are missed during line-by-line screening. Combinations downgraded from medium to low are removed from the hazardous circuit breaker group. The removal record retains the original hazardous identification history for traceability and is not directly deleted. Combinations upgraded from low to medium or high are added to the hazardous circuit breaker group as new records, carrying the upgrade and subsequent classification changes for traceability. An empty hazardous circuit breaker group indicates that there are currently no combinations requiring interlocking verification; subsequent steps directly transmit this result without performing interlocking verification. An empty result is not the same as a zero-record number in the risk node table. Even if the risk node table has a large number of low-level records, as long as none reach medium level or above, the hazardous circuit breaker group will still be empty.

[0032] In some embodiments, the step of generating a dynamic rearrangement sequence by performing live-line interlocking verification through the circuit breaker group with potential hazards includes: tracking adjacent live states based on the circuit breaker group with potential hazards to obtain live-line interlocking conditions; identifying interlocking conflicts before and after the operation based on the live-line interlocking conditions to establish an interlocking verification list; performing step-by-step executability verification on the interlocking verification list to obtain an executable marker sequence; and reordering the executable marker sequence back to the operation steps to generate a dynamic rearrangement sequence.

[0033] Based on the tracking of adjacent energized states of circuit breaker groups with potential hazards, energization interlocking conditions are obtained. For each switch combination of the circuit breaker group with potential hazards, the adjacent switchgear is first found from the equipment topology table. The adjacent relationship is determined by the physical electrical connection of the same busbar segment or the same feeder segment, with electrical connection as the standard. The topology table is maintained by the substation wiring diagram management system and updated synchronously with the actual wiring on site. It is refreshed immediately after switching changes in wiring. For example, after adding a bypass switch during a busbar expansion, the adjacent relationship is re-determined according to the new wiring, and the range determined by the drawings before the expansion is no longer used. The energized state of adjacent switchgear is taken from the current reported value of the real-time monitoring system, divided into three types: energized, non-energized, and unknown. Adjacent equipment with an unknown state is treated as energized, adopting a strict principle to avoid misjudging as safe due to monitoring gaps. Unknown equipment counted as energized is included in the energization interlocking condition list along with the actual energized equipment and is listed together in this section. Live-line interlocking conditions are provided separately for each switch combination. The content includes a list of specific device numbers in adjacent equipment that are energized when the combination performs a tripping operation. An empty list indicates that there are currently no adjacent energized devices for that combination, and the operation can be directly permitted. If the topology of adjacent switchgear has recently undergone changes, this step uses the latest topology. Historical topology records before the change are only retained in the historical fields of the corresponding records for the circuit breaker group with potential hazards and are not included in this tracking, preventing old wiring relationships from misleading the determination of adjacent devices. When adjacent devices in the same combination of circuit breakers with potential hazards belong to different voltage levels, the live-line interlocking conditions are listed separately for each voltage level, with interlocking conditions for different voltage levels listed hierarchically. Interlocking determinations across voltage levels are reserved for verification in the next step.

[0034] Based on the live-line interlocking conditions, an interlocking verification list is established to identify interlocking conflicts before and after operation. The rule for determining interlocking conflicts is that if the time of the adjacent live equipment listed in the live-line interlocking conditions overlaps with the currently scheduled operation time of the switch combination, it is considered a conflict. The time overlap is determined by whether the operation windows of the two overlap. The operation window is a fixed margin interval reserved before and after the planned operation time of the combination. The margin value comes from the standard working hours in the on-site operation ticket management regulations. This working hour is usually set at 15 minutes. The operation window of a switch is the interval formed by extending 15 minutes before and after the planned operation time. The margin itself is independent of the combination type and is uniformly used throughout the station. The same combination has a corresponding live-line interlocking condition before and after the operation. The pre-state and post-state are checked separately. If any conflict exists, it is included in this clause. If there is no conflict, no record is generated. The pre-state check is based on the live-line status reported in real time before the operation, and the post-state check is based on the live-line status reported in real time after the operation. The two data are taken independently and do not share the same time reading. The interlock verification checklist records conflict determination results item by item for each switch combination. Each entry includes the corresponding operation step number, conflict type (preceding / following), adjacent equipment number, and the required energized state of that equipment. Combinations without conflicts are not recorded. Combinations judged as conflicting in both checks are retained as two separate records in the interlock verification checklist, not merged into one, to avoid mixing up two different types of conflicts (preceding and following). Each record carries an independent conflict type label, and they are processed separately during subsequent executability verification without interference. Switch combinations with an empty energized interlocking condition list are directly judged as conflict-free and no interlock verification checklist record is generated.

[0035] For example, the step-by-step executability verification of the interlock checklist to obtain an executable marker sequence includes: establishing a set of pre-conditions for each step of the interlock checklist based on the pre-conditions for the pre-operation; generating a real-time failure list by identifying dynamically failed constraints during execution based on the pre-conditions; forming a delayed candidate list by using the real-time failure list to check the preset duration observation window when the constraints have just recovered; and obtaining an executable marker sequence by backtracking the delayed candidate list for subsequent unexecuted steps.

[0036] A set of pre-conditions is established to track the pre-operation energization conditions for each step in the interlock checklist. Each conflict record in the interlock checklist carries a corresponding operation step number, which follows the original numbering of the initial operation sequence. The system progressively tracks the energization requirements that adjacent devices must meet before the execution of that step. The requirements are directly taken from the conflict type and adjacent device numbers marked in the record, without introducing additional criteria. Pre-conditions are extracted sequentially step by step. When multiple conflict records exist for the same step in the interlock checklist, their pre-conditions are merged into one group. During merging, all adjacent device numbers are retained and listed side-by-side. The merged list of numbers is determined solely by exact string matching. The pre-constraint set includes all pre-energized conditions according to the operation steps. Each step corresponds to a set of constraints, stored as key-value pairs of equipment number and required energized state. This facilitates item-by-item verification and quick retrieval. The key is the adjacent equipment number, and the value is the required energized state of that equipment. For example, if the pre-constraint for a certain step is that isolating switch No. 2 should be in a de-energized state and bypass switch No. 3 should be in a energized state, the real-time status on site must be checked against these two requirements before execution. Only if both are met can the step be allowed. Conflict records marked as post-constraint in the interlock verification list are not included in the pre-constraint set. Post-constraint conflicts are checked separately after the step is executed. The two types of conflicts use independent verification processes, and the post-constraint verification results are archived separately. The two records are linked by the step number. For operation steps without corresponding conflict records in the interlock verification list, there is an empty constraint in the pre-constraint set. An empty constraint is considered to have no pre-energized requirement and does not hinder the execution of the step.

[0037] A real-time failure list is generated based on constraints that dynamically fail during execution, identified from the pre-constraint set. For each constraint corresponding to a step in the pre-constraint set, the energized status reported by the real-time monitoring system is checked item by item during the actual execution of the sequence. The check frequency is consistent with the status reporting frequency of adjacent devices and is triggered synchronously. The check action itself is not written to the operation log; records are only generated when a failure is determined to reduce log load. When the constraint check finds that the actual energized status of an adjacent device is inconsistent with the requirements recorded in the pre-constraint set, the constraint is determined to be dynamically failed. Failure determination only considers whether the current status meets the requirements, and the determination is completed immediately during the check when the inconsistency is found. For example, if a pre-constraint for a certain step requires that switch 2 remain de-energized, and if this switch is closed due to misoperation during inspection and becomes energized, the system will determine that the constraint is dynamically failed in the next check, without waiting for multiple subsequent confirmations. The real-time failure list records failures for each constraint item. Each failure record compares the time of failure with the energized state before and after the failure. The state before failure is taken from the original requirement value in the preceding constraint set, and the state after failure is taken from the verification reading that triggered this judgment. The comparison of the two can reconstruct the specific process of the failure. If only some of the constraints in the same step fail, the real-time failure list records each failed constraint item separately, and those that do not fail are not included. For example, if a step is associated with three adjacent devices, and only one of them changes to an inconsistent state, the list only records the failure of that one device, while the other two are still processed as meeting the requirements. This avoids the states of different constraints in the same step being merged into one record, thus masking individual failures. Steps with empty constraints in the preceding constraint set do not participate in the verification of this stage, and no record is generated for them in the real-time failure list.

[0038] A delayed candidate list is generated by using a real-time failure list to identify newly recovered constraints within a preset observation window. Constraints corresponding to each failure record in the real-time failure list are retrieved one by one. If the status of adjacent equipment recovers to the required level during subsequent verification, it is considered recovered. The recovery time is the earliest verification time when the status meets the requirements; recovery determination only considers the status itself. A preset observation window is set after constraint recovery. The window duration is the historical statistical upper limit of the action return time of the protection device for that type of adjacent equipment. Different types of equipment correspond to different window durations, with values ​​derived from measured statistics of the action return time of that model of equipment over the past three years. For a certain type of isolating switch, the observation window is typically set to three minutes. Even if the status remains satisfied within the observation window, it is not considered a fully stable recovery; stability is only recognized after the window expires. The delayed candidate list includes all constraints within the observation window. Each constraint carries two fields: recovery time and observation window end time. The corresponding operation steps are not considered executable before the window ends; querying this step before the window ends will always return "to be observed." If the status of an adjacent device within the observation window changes to unsatisfactory again, the observation window for that record in the delay candidate list will be reset, and the recovery time will be updated to the latest time when it became satisfactory. For example, if a switch is accidentally closed again in the second minute of the observation window and becomes energized again, and then becomes de-energized again, the reset timer must start from the time when it became satisfactory again. The original recovery time is retained in the corresponding record in the real-time failure list for traceability. Constraints that have never been recovered in the real-time failure list will not enter the delay candidate list, and the corresponding steps will remain blocked. The blocked state requires manual intervention to resolve.

[0039] The executable marker sequence is obtained by back-ordering subsequent unexecuted steps from the delayed candidate list. For each record in the delayed candidate list, the corresponding operation step is taken, and its observation window end time is compared one by one with the original scheduled operation time of subsequent unexecuted steps. The comparison scope covers all unexecuted steps after that step, without limiting the immediately following next step. Subsequent steps whose original scheduled time is earlier than the observation window end time are postponed to the most recent executable cycle after the window ends. The executable marker sequence includes all steps in the original order of the operation steps. Each step is marked with either an executable or observation-pending state. Observation-pending steps are accompanied by the corresponding observation window end time, while executable steps are not. The record structures for the two states are inherently different, facilitating separate parsing and treatment by the execution terminal. If a step in the delayed candidate list corresponds to multiple constraints and has different observation window end times, the latest end time in all windows is taken as the actual executable time for that step during back-ranking. For example, if a step is associated with the observation windows of two devices, one ending at 9:05 and the other at 9:12, the actual executable time for that step is 9:12. Taking the latest value ensures that all relevant constraints have crossed their respective observation windows by the time of execution. Subsequent steps whose original operation time is later than the end time of the observation window in the delayed candidate list are directly marked as executable, and each step is judged independently according to its own constraints. Steps marked as pending observation in the executable marker sequence will have their downstream steps that depend on this step being marked as pending observation in the executable marker sequence simultaneously. This avoids downstream steps being mistakenly judged as being able to start early before upstream steps are actually executable, thus disrupting the order of operations.

[0040] The executable marker sequence is back-ordered to generate a dynamic rearranged sequence. Steps marked as executable in the executable marker sequence are back-ordered using their original operation times. Steps marked as pending observation are back-ordered using the end time of their corresponding observation window as new operation times. After back-ordering, the record format for both types of steps no longer distinguishes between the source of the markers and they are uniformly included in subsequent distribution. The dynamic rearranged sequence, based on the original switch combinations and phase angle results of the initial operation sequence, only adjusts the operation times of steps with interlocking conflicts. The operation times and phase angle results of steps without conflicts remain unchanged. Steps not adjusted are completely consistent with their corresponding records in the initial operation sequence in the dynamic rearranged sequence, preserving the original arrangement with minimal modification. If multiple phases of the same switch combination belong to different operating steps and the new operating times no longer fall within the same power frequency cycle, the phase angle results of each phase of the combination in the dynamic rearrangement sequence are returned to the phase selection and tripping verification stage for re-verification. Whether they are in the same cycle is determined by whether the difference between the new operating times of the two phases is less than one power frequency cycle. For example, if phase A is postponed to a new cycle due to being to be observed while phase B is still executed according to the original cycle, and the two are now in different power frequency cycles, then the criterion is exceeded and the phase angle must be re-verified. For steps marked as to be observed due to downstream dependence in the executable marking sequence, the new operating time is taken as the end time of the observation window of the upstream step it depends on, which is consistent with the marking logic of the executable marking sequence itself. The dynamic rearrangement sequence includes the final operating time, phase angle results, and classification results brought in along the hidden danger circuit breaker group for all switch combinations (this item is left blank for non-hidden danger combinations). The order of inclusion follows the original arrangement in the initial operating sequence. The re-arrangement process only changes the time values ​​and keeps the record arrangement position unchanged.

[0041] Step S103: Based on the dynamic rearrangement sequence, hierarchical instruction arrangement is carried out to output the execution instruction group. The execution instruction group is used to track the transient fluctuations of the circuit breaker restart and generate a status confirmation signal. The remaining steps are re-estimated based on the status confirmation signal to construct a continued status record.

[0042] Specifically, execution instruction groups are output and arranged hierarchically based on the dynamic rearrangement sequence. The final operation time of each switch combination in the dynamic rearrangement sequence is taken, and groups are formed according to their respective substations or busbars. Switch combinations within the same group are assigned to the same execution batch. Grouping is based on the current dispatch shift assignment table for that substation, with different groups corresponding to different execution shifts. The assignment table is updated in real-time with shift rotations. Switch combinations within a batch are then arranged in execution order according to their operation time. Multiple combinations planned for operation at the same time are sorted by switch combination number from smallest to largest. Instruction levels are divided into two levels: Level 1 and Level 2. Level 1 instructions correspond to switch combinations previously marked as high-level in hazard assessments, while Level 2 instructions correspond to all other combinations. Level 1 instruction execution results must be returned in real-time, while Level 2 instructions are returned uniformly by batch. The execution instruction group includes all instructions by execution batch. Each instruction carries four fields: switch combination number, target operation time, verified tripping phase angle, and instruction level. The field order is fixed, and all levels share the same structure. Level 1 and Level 2 instructions share the same field structure, differing only in the value of the instruction level field. This facilitates the distribution of the entire instruction group to the execution terminal in a unified format without requiring separate parsing. In the dynamic rearrangement sequence, combinations marked as unobservable due to downstream dependencies have their corresponding instruction's target time directly taken from the final time of that record in the dynamic rearrangement sequence. When the number of instructions within the same execution batch exceeds the single-execution limit for that shift, the excess is carried over to the next execution batch. The priority for carrying over is given to instructions with later target times within the same batch. The carried-over instructions are marked with a batch change record in the instruction group for tracing back to their original batch.

[0043] In some embodiments, the step of using the execution instruction group to track the transient fluctuations of re-ignition after circuit breaker tripping and generate a status confirmation signal includes: obtaining re-ignition characteristic parameters by tracking the transient fluctuations of re-ignition between contacts after circuit breaker tripping according to the execution instruction group; determining the media recovery rating by calculating the re-ignition frequency change rate through the re-ignition characteristic parameters; verifying the circuit breaker tripping confirmation degree based on the media recovery rating to obtain the tripping confirmation degree index; and judging the tripping confirmation interval based on the tripping confirmation degree index to generate a status confirmation signal.

[0044] Reignition characteristic parameters are obtained by tracking the transient fluctuations of re-ignition between contacts after the circuit breaker trips according to the execution instruction group. For each switch combination corresponding to each instruction in the execution instruction group, the voltage waveform between contacts is collected after the actual tripping action. The collection window is a fixed duration from the tripping action to the complete restoration of insulation. The window duration is taken as the longest historical time for dielectric recovery under full-phase opening distance of this type of circuit breaker. For a certain 40.5 kV model, this time is approximately 20 milliseconds. The starting point of the window is strictly aligned with the operation target time recorded in the execution instruction group, neither earlier nor later. The high-frequency oscillation segment appearing in the voltage waveform is a re-ignition. Re-ignition is identified by a sudden rise and fall in waveform amplitude within a very short time, and the oscillation amplitude reaches the lower limit of the criterion. This lower limit is calibrated by the oscillation amplitude corresponding to the re-ignition event in the factory type test of this type of circuit breaker. The re-ignition characteristic parameters are recorded for each switch combination, including the number of re-ignitions corresponding to this trip, the time of each re-ignition, and the corresponding oscillation amplitude. All three items are based on the waveform data actually read in the collection window. No re-ignition section switch combination was identified during this trip. The re-ignition count for this combination in the re-ignition characteristic parameters was recorded as zero. The other two fields were left blank, which does not affect the normal recording of this record. Blank fields and zero values ​​will be processed differently in subsequent statistical steps. For switch combinations marked as Level 1 commands in the execution command group, the re-ignition characteristic parameter acquisition window was extended. The extended window duration was amplified by a preset multiple based on the original full-phase opening distance consumption time. The additional waveform data generated by the extended portion was also used to identify re-ignition according to the same criteria.

[0045] For example, the step of determining the medium recovery rating by calculating the re-ignition frequency change rate through the re-ignition characteristic parameters includes: obtaining a re-ignition frequency sequence by statistically analyzing the number of re-ignitions occurring in a single circuit breaker trip using the re-ignition characteristic parameters; calculating the re-ignition frequency change rate by tracking the post-arc recovery changes between adjacent circuit breakers based on the re-ignition frequency sequence; establishing an ablation accumulation assessment table by performing an ablation accumulation verification using the re-ignition frequency change rate; and identifying the slow medium recovery item in the ablation accumulation assessment table to determine the medium recovery rating.

[0046] The re-ignition frequency sequence is obtained by statistically analyzing the number of re-ignitions occurring during a single circuit breaker trip using re-ignition characteristic parameters. Each record of the re-ignition characteristic parameters corresponds to one circuit breaker trip. The re-ignition count field carried by this record is directly read as the re-ignition count for this trip. During reading, it is checked against the switch combination number of the record to avoid misattribution across different combinations. For example... Figure 3As shown, the re-ignition frequency sequence 11 is established separately for each switch combination. The number of re-ignitions in each trip of the same combination is arranged in the order of tripping to form the re-ignition frequency sequence 11 for that combination. The sequence length continuously increases with the cumulative number of trips for that combination. Each position in the re-ignition frequency sequence 11 corresponds one-to-one with the operation target time in the execution instruction group for that trip, facilitating time-based lookup. For example, the value at the fifth position of a combination's sequence corresponds to the operation target time recorded at the time of its fifth trip, which allows direct location of a specific operation on a specific day. For the switch combination that trips for the first time after the method is put into operation, its re-ignition frequency sequence contains only one value and is not treated as an anomaly; this value also participates in the calculation of subsequent steps. Trips with zero re-ignition counts recorded in the re-ignition characteristic parameters are still included in the re-ignition frequency sequence and have a value of zero to ensure the continuity of the sequence in the time dimension. This continuity is indispensable for subsequent calculations of post-arc recovery changes between adjacent trips. If the interval between two consecutive trips in the re-ignition frequency sequence 11 of the same switch combination exceeds the upper limit of the maintenance cycle, the sequence shall be marked with maintenance boundary mark 18 at that point. Maintenance boundary mark 18 is only used to distinguish the data before and after maintenance when making subsequent statistics. If the maintenance record corresponding to the boundary mark involves contact replacement, the historical record of re-ignition characteristic parameters before this maintenance shall also be marked with the pre-maintenance mark to facilitate the verification of the two data segments separately. The medium recovery capability before and after replacement should not be regarded as a single curve.

[0047] The re-ignition frequency change rate is calculated by tracking the post-arc recovery changes between adjacent circuit breakers based on the re-ignition frequency sequence. The difference in the number of re-ignitions corresponding to two adjacent circuit breakers in the re-ignition frequency sequence reflects the direction of change in the insulation recovery capability of the switch combination after the arc between these two circuit breakers. A positive difference indicates that the number of re-ignitions increases and the dielectric recovery capability tends to decrease; a negative difference indicates that the number of re-ignitions decreases and the recovery capability tends to improve; and a zero difference is considered as no directional change in the recovery capability. The re-ignition frequency change rate 12 is calculated as v = Δn / Δt: Δn is the difference in the number of re-ignitions between two adjacent circuit breakers in the re-ignition frequency sequence 11; Δt is the actual time interval between these two circuit breakers, in hours; and v is the re-ignition frequency change rate 12, used to characterize the rate of degradation of the dielectric recovery capability. A positive v indicates that the re-ignition frequency increases over time, that is, the recovery capability continues to deteriorate. For example, if the number of re-ignitions between two adjacent circuit breakers increases from 1 to 3 times, with an interval of 100 hours, v is 0.02 times per hour. If the interval between two consecutive trips in the reignition frequency sequence is less than the shortest allowable operating interval for that type of circuit breaker, this interval will not be included in the calculation of the reignition frequency change rate and will be considered statistically insignificant due to the data points being too close. The shortest allowable operating interval is taken from the mechanical life protection index given in the equipment operating procedures. The reignition frequency change rate will output the latest calculation result for each switch combination. Historical calculation results will be retained in chronological order and archived alongside the latest result, along with the corresponding Δt interval, for tracing the rate change trajectory within a certain period. For switch combinations with a reignition frequency sequence length of less than two values, making it impossible to calculate adjacent differences, the reignition frequency change rate will not be output in this instance and will be calculated after the next trip to avoid filling the gap and contaminating the rate trajectory of that combination.

[0048] An ablation accumulation assessment table is established using the reignition frequency change rate to perform ablation accumulation verification. For switchgear with a positive reignition frequency change rate (12), the cumulative ablation increment (17) is calculated based on its value. The conversion relationship is determined according to the rate-ablation control section 19 given in the factory arc-extinguishing chamber ablation life test for this type of circuit breaker. The larger the reignition frequency change rate (12), the larger the corresponding cumulative ablation increment (17). The rate-ablation control section 19 is divided into rate ranges, each range corresponding to a fixed increment value, and the range granularity is taken from the rate gradient actually collected in the factory test and set according to the measured gradient. The upper limit of the rate covered by the control table is based on the maximum rate change value that has appeared in all type tests of this model. Rates exceeding the upper limit are all classified into the upper limit range and the corresponding increment is taken. For switchgear with a reignition frequency change rate of zero or negative, the cumulative ablation increment for this time is recorded as zero. The previously accumulated ablation amount is not deducted even if the recovery capability remains stable or improves, as ablation is an irreversible mechanical and electrical loss. The ablation cumulative assessment table is updated by adding an ablation cumulative increment of 17 for each switch combination. Each entry carries two fields: the current ablation cumulative total of 13 and the most recent ablation cumulative increment of 17. The ablation cumulative total of 13 is the sum of all previous ablation cumulative increments of 17 for that combination, added with equal weights. Increments from earlier years and recent years have the same weight in the total, as ablation damage is irreversible once it occurs. For switch combinations whose arc-extinguishing chambers have been recently replaced, the cumulative total for that combination in the ablation cumulative assessment table is reset to zero and re-added when the replacement record takes effect. The cumulative history before the replacement is archived separately, and the archived record is linked to the new cumulative record in the ablation cumulative assessment table through the replacement event timestamp.

[0049] The medium recovery rating is determined by identifying items with slow medium recovery in the cumulative erosion assessment table. Any switch combination whose total cumulative erosion exceeds a certain percentage of the factory-specified scrap threshold for that type of circuit breaker is considered to have slow medium recovery. The percentage is determined by the unified provisions of the equipment life management regulations regarding the warning range. The scrap threshold itself is consistent with the unit of the total cumulative amount recorded in the cumulative erosion assessment table, expressed in the unit of measurement of the factory arc-extinguishing chamber erosion life test, avoiding errors introduced by unit conversion. This threshold is updated synchronously with each major equipment overhaul as the arc-extinguishing chamber type changes; different types result in different scrap thresholds. The medium recovery rating is divided into three levels: normal, watch out, and warning. A cumulative erosion amount (13) below the upper limit (14) of the normal level corresponds to the normal level; reaching the upper limit (14) but not exceeding the scrap threshold (15) corresponds to the watch out level (16); exceeding the scrap threshold (15) corresponds to the warning level. The three levels correspond one-to-one with the original value of the cumulative erosion amount (13) recorded in the cumulative erosion assessment table and are archived according to the original value. The medium recovery rating provides results for each switch assembly. The rating result is not numerical and represents only one of three levels; the specific numerical value is still retained in the ablation accumulation assessment table for reference. The rating and numerical value are stored separately for easy retrieval in subsequent steps according to different granularities. The coarser-grained rating is used for rapid screening, while the finer-grained raw numerical value is used for in-depth analysis. Switch assemblies previously rated as "concerned" but whose cumulative ablation accumulation assessment table was reset to zero due to arc-extinguishing chamber replacement have their medium recovery rating simultaneously reverted to the normal level. The reversion record retains the start and end times of the original "concerned" rating for traceability; this historical period is valuable for manual review of lifespan trajectories.

[0050] The confirmation index for contact placement is obtained by verifying the contact placement confirmation based on the medium recovery rating. For switch combinations with a medium recovery rating of normal or watch out, the initial verification basis is directly whether the number of reignitions recorded in the reignition characteristic parameters is zero. A zero number of reignitions indicates initial placement, while a non-zero number indicates initial non-placement. The verification itself only uses the already collected reignition count field in the reignition characteristic parameters to reduce the data processing load during verification. For switch combinations with a medium recovery rating of warning, the verification standard is further raised, requiring that the number of reignitions in the current trip is zero and that the number of reignitions in the most recent three trips in the reignition frequency sequence is also zero. If the sequence has less than three reignitions, it is determined by whether all the current reignition counts are zero. The higher standard is because the warning level means that the combination is in a state of high erosion accumulation, and a single good result is not enough to indicate that the contact placement status has stabilized. Each switch combination has a corresponding record for the arrival confirmation index. The value is taken from the confidence level corresponding to the preliminary verification result. For switch combinations that have initially arrived and whose media recovery rating is normal, the highest confidence level is used, assigned a value of 0.95. For switch combinations that have initially arrived but are in the warning level, the medium confidence level is used, assigned a value of 0.7. For switch combinations that have not initially arrived, regardless of the rating, the lowest confidence level is used, assigned a value of 0.3. The confidence level is divided into three fixed values, and each level is directly taken. The three levels of values ​​directly constitute the value range of the arrival confirmation index for reference in the next stage. For switch combinations whose media recovery rating is updated at the same time during this verification process, the verification is based on the latest updated rating to avoid the arrival confirmation index being based on an outdated level due to a lag in rating updates. The order of the update time and the verification time is based on the system clock.

[0051] The system analyzes the confirmation interval of the tripping confirmation index to generate a status confirmation signal. The tripping confirmation index is divided into two categories based on numerical ranges: confirmed and pending verification. The interval boundary is dynamically determined by taking the statistical median of all tripping confirmation indices for the entire batch of switch combinations. For example, if a batch contains twenty switch combinations and the median falls around 0.7, then the boundary for that batch is approximately 0.7. The boundary is recalculated based on the actual index distribution for each batch. Since the confidence benchmarks for different models and years of operation within the same batch vary, a fixed value cannot adequately account for these differences. The status confirmation signal lists one of two results for each switch combination: confirmed or pending verification. Confirmed indicates that the tripping of that combination has been completed, while pending verification indicates that manual or subsequent steps are required for verification. The subsequent processing paths for each result are detailed in the next step. Switch combinations whose confirmation index is near the interval boundary and whose difference from the boundary is less than one-tenth of the standard deviation of the batch of indicators are all classified as pending review. This is to avoid misjudgment caused by boundary values. The one-tenth ratio is taken from the balance point of misjudgment rate and missed judgment rate in historical review data and is re-verified annually as review data accumulates. For example, after a certain year's verification, this ratio was adjusted from one-tenth to one-twelfth, and subsequent batches are judged according to the new ratio to determine the pending review range near the boundary. Switch combinations marked as pending review in the status confirmation signal also carry the specific reason for triggering the pending review. The reasons are divided into two categories: abnormal reignition number and rating warning, for subsequent manual judgment reference. When both types of reasons are true, both are marked. Combinations marked as confirmed in the status confirmation signal do not carry this reason field, as this field itself is the exclusive label for the pending review result.

[0052] For the remaining steps of the status confirmation signal reassessment, a continued status record is constructed. For switch combinations marked as confirmed in the status confirmation signal, their subsequent steps in the dynamic reordering sequence continue to execute according to the original time and order. Continuation does not require additional manual confirmation to avoid slowing down the progress of already confirmed combinations. For switch combinations marked as pending review in the status confirmation signal, all subsequent unexecuted steps are suspended. The suspension range extends from the switch combination to the steps corresponding to its next operation. For example, if a switch combination is judged to be pending review after this trip, its second operation scheduled for two hours later will be suspended until the current review is completed. This does not affect the subsequent steps of other combinations, and other combinations, even those in the same execution batch, are processed independently according to their respective status confirmation signal results. The continued status record includes the reassessment results for each switch combination. Each record carries three fields: confirmed or pending review result, suspension range (if any), and the time of this reassessment. The field structure remains consistent for both types of results, except that the suspension range field is left blank in confirmed results, as this field is not applicable to this type of result. If a switch combination already has a previous continuation status record, the current reassessment result will overwrite the old record, and the old record will be transferred to the historical archive. The historical archive is arranged chronologically according to the time of reassessment, allowing for tracing the complete trajectory of all reassessments for this combination. For switch combinations marked as paused in the continuation status record, the paused status will be lifted after manual verification. The lifting action itself will generate a new record in the continuation status record. The new record and the original paused record are linked through the switch combination number, and the two together constitute the complete process of this event for this combination.

[0053] Step S104: Identify the execution deviation from the continued status record to obtain the abnormal alarm signal, use the abnormal alarm signal to identify false alarms and backtrack to obtain the threshold calibration parameter, and use the threshold calibration parameter to set the alarm threshold reset to form a pause alarm command.

[0054] Specifically, abnormal alarm signals are obtained by identifying execution deviations from the follow-up status records. The determination of execution deviation is based on the difference between the actual operation time of the switch combination and the original scheduled operation time of the corresponding step in the follow-up status record. If the deviation exceeds the time tolerance allowed by the operation ticket for this type of circuit breaker, it is counted as a deviation. The tolerance value is derived from the on-site operation ticket management regulations. The tolerance itself does not distinguish between early or late directions; both directions share the same value. Currently, the tolerance is set at five minutes. If a switch combination is scheduled to open at 9:00 AM but actually completes the operation at 9:06 AM, this exceeds the tolerance and is judged as a late deviation. For records in the follow-up status record marked as pending review but already manually confirmed, the actual operation time is changed to the manually confirmed completion time to recalculate the deviation. The recalculated result replaces the original automatic judgment result. Abnormal alarm signals are generated separately for each switch combination. The signal content includes three items: the deviation value, the deviation direction (advanced / delayed), and the alarm channel number to which the combination belongs. Combinations whose deviation does not exceed the tolerance will not generate this signal, indicating that the operation was within the tolerance range, not that the combination did not perform an operation. If the same combination has multiple deviations within the same execution batch, the abnormal alarm signals are recorded sequentially according to the order of the deviations, each independent. This facilitates subsequent steps to check each specific deviation individually and prevents individual large deviations from being masked by the average. For example, if a combination in a batch is delayed by two minutes and then by seven minutes, the two deviations are recorded separately for manual verification of the cause. If, due to unforeseen circumstances, the corresponding actual execution data for the follow-up status record could not be collected during the operation, the deviation value is marked as missing data and is neither included in the normal range nor the deviation range. This situation itself does not generate an abnormal alarm signal because it is impossible to determine whether the operation truly exceeds the tolerance when the data is missing.

[0055] In some embodiments, the step of using the abnormal alarm signal to identify false alarms and backtrack to obtain threshold calibration parameters includes: identifying alarm channels with long-term and frequent false alarms from the abnormal alarm signal to form a set of candidate values ​​for low false alarms; tracking historical action threshold margins based on the set of candidate values ​​for low false alarms to obtain a list of false alarm features; verifying the sensitivity offset amplitude of each mode for the false alarm feature list to obtain an offset verification table; and using the offset verification table to back-calculate the alarm sensitivity correction value and output the threshold calibration parameters.

[0056] An alarm channel with a long history of frequent false alarms is identified from abnormal alarm signals to form a candidate set of falsely low setpoints. Abnormal alarm signals are grouped and statistically analyzed according to their corresponding alarm channels. The number of triggers for the same channel within the most recent year is summarized as the total number of triggers for that channel in the current period, for subsequent comparison with a benchmark. The benchmark is the annual average of the number of triggers for that channel in the past three years (excluding the current statistical period), measured on the same one-year cycle as the current period. When calculating the average, periods in which the channel underwent setpoint adjustments are excluded to avoid data from different setpoints before and after adjustments being mixed into the benchmark. Alarm channels with a total number of triggers in the current period exceeding three times the benchmark are identified as channels with a long history of frequent false alarms. This three-fold multiple is taken from the unified provisions of the relay protection regulations regarding the false alarm warning range. The determination only considers whether the total number of triggers exceeds the multiple, without distinguishing the direction of deviation. For example, if a channel has an average of 8 triggers per year over the past three years and 26 triggers in the current period, exceeding the three-fold threshold of 24, it is identified as a channel with a long history of frequent false alarms. The candidate set for falsely low setpoints includes all identified alarm channels. Each channel carries three fields: channel number, total number of triggers in the current period, and corresponding benchmark value. The number of triggers and the benchmark value are both taken from the actual values ​​within the current statistical period, and the field values ​​are directly derived from the statistical results of abnormal alarm signals. For alarm channels previously identified as having long-term frequent false alarms and whose number of triggers has further increased in the current statistical period, the trigger count of the channel recorded in the candidate set for falsely low setpoints is updated to the latest value, and the old period values ​​are archived. For alarm channels that have recently undergone setpoint adjustments, the trigger records before and after the adjustment are statistically analyzed separately according to the setpoint version. If the adjusted record meets the multiple criterion on its own, it is also included in the candidate set for falsely low setpoints as an independent record.

[0057] A false alarm feature list is obtained by tracing historical action threshold margins based on the candidate set of false low values. Historical action threshold margin refers to the difference between the measured value at each trigger moment of the alarm channel and the set action threshold at that time. The unit of the difference is consistent with the original measurement of the channel. The smaller the difference, the closer the trigger is to the threshold edge, and the more likely it is to cause a false alarm under normal fluctuations. For example, if the measured value of a trigger differs from the action threshold at that time by only two percent of the original range, it is considered a trigger close to the edge, and such records are the first to be considered in subsequent verification. For each alarm channel in the candidate set of false low values, all historical trigger records of that channel within the past three years are traced one by one, and the action threshold margin corresponding to each trigger is calculated, covering all traceable historical data of that channel. The tracing scope is different from the one-year statistical period in the first section of this step, and the two serve different judgment purposes. The false alarm feature list includes margin tracking results for each alarm channel. Each entry contains two fields: a sequence consisting of all historical margin values ​​for that channel and a sequence mean. The sequences are arranged chronologically by trigger time, and the sequence length corresponds to the number of historical trigger records for that channel when the false low threshold candidate set was established. If a historical margin sequence for the same alarm channel contains individual records with negative margins, these negative values ​​are retained and included in the false alarm feature list's sequence and mean calculations, considered valid samples. Negative values ​​precisely reflect that the trigger exceeded the threshold itself and represent the most important type of false alarm feature. For alarm channels with fewer than three historical trigger records, the false alarm feature list still calculates the sequence and mean based on the actual number of records. Records for this channel in the false low threshold candidate set are also retained, but are marked in the list as insufficient samples for further weighting during the next verification stage.

[0058] A deviation verification table is obtained by verifying the sensitivity deviation of each mode based on the false alarm feature list. Operating modes are divided into three categories: light load, heavy load, and maintenance. Mode determination is based on the real-time load status corresponding to the alarm channel's trigger time. The applicable sensitivity setting may differ for the same channel under different modes. The load boundaries for the three modes are uniformly defined according to the operating procedures of this equipment model. The load falls into the corresponding mode based on its range. For example, a certain equipment model classifies load rates below 70% as light load and above 70% as heavy load. Maintenance mode is handled separately according to the maintenance mode setting and is not subject to load rate constraints. Each record in the false alarm feature list is categorized according to its corresponding trigger time mode. Historical margins under different modes for the same channel are calculated as averages for each mode. The load status used for categorization is taken from load monitoring data at the same timestamp at the trigger time, using real-time values. The sensitivity deviation is the difference between the average margin of the same channel and mode and the factory-set margin benchmark for that mode. The larger the difference, the further the actual sensitivity deviates from the factory setting under that mode. The benchmark is also selected with reference to the factory sensitivity adjustment characteristic curve of this equipment model. The offset verification table records verification results at two levels: alarm channel and operating mode. Each entry carries three fields: mode name, average margin, and offset magnitude. A maximum of three records correspond to the same channel. The average margin value is directly derived from the statistical results of the corresponding mode in the false alarm feature list. For alarm channels that have never been traced back to historical trigger records in the false alarm feature list under a certain mode, the offset magnitude corresponding to that channel and mode in the offset verification table is recorded as empty and not substituted with a zero value for subsequent back-calculation. This empty status allows the next step to identify and skip this combination; the empty status is limited to only one cell for that channel and mode.

[0059] The alarm sensitivity correction value is calculated using the offset verification table to output the threshold calibration parameters. The sensitivity correction value is calculated as ΔS = -k × D: D is the sensitivity offset amplitude of the corresponding mode for that channel in the offset verification table; k is the correction coefficient, whose dimension is the same as the set value, and the value is determined according to the factory sensitivity adjustment characteristic curve of the alarm device model and segmented according to the range of the offset amplitude; ΔS is the desired sensitivity correction value, and the negative sign indicates that the correction direction is opposite to the offset direction, that is, if it is too high, adjust it lower, and if it is too low, adjust it higher. For example, if the offset amplitude D of a certain channel is four percent of the original range, which is biased to the higher side, the calculated ΔS is a negative value, and the set value of that channel is adjusted downward accordingly. For channel mode combinations with missing offset amplitudes in the offset verification table, no correction value is calculated in this case, and the set value corresponding to this combination remains unchanged from the factory setting. The threshold calibration parameters record all calculated correction values ​​for each alarm channel. Each channel includes its channel number, the relevant operating mode, and the corresponding correction value. Correction values ​​for multiple modes within the same channel are listed separately, and the sign of the correction value matches the original offset direction of the source mode in the offset verification table. If the correction values ​​for multiple modes within the same alarm channel have inconsistent directions (e.g., positive for light load mode and negative for heavy load mode), both correction values ​​are retained in the threshold calibration parameters. Each mode has its own independent setting adjustment, without interference, and is adjusted according to the operational needs of its respective mode. For channels where the absolute value of the correction value exceeds the upper limit of the adjustable setting range for that model, the threshold calibration parameters truncate this value to the upper limit and mark it with a truncation indicator for subsequent manual review. The original value is retained separately in the history field next to the indicator for manual review to determine whether the truncation is reasonable and whether the setting device needs to be replaced.

[0060] The alarm threshold reset setting of the threshold calibration parameters generates a pause alarm command. The alarm threshold reset setting includes two actions: first, adjusting the action threshold setting of the channel according to the correction value; then, generating a corresponding pause alarm command. Both actions are completed in the same setting process. The correction value used for adjustment is directly taken from the corresponding record of the threshold calibration parameters. Records in the threshold calibration parameters where the absolute value of the correction value is less than the preset minimum adjustable step size for the channel are not adjusted in this cycle. This is accumulated until the next calibration cycle and recalculated along with the newly generated offset. The minimum adjustable step size is taken from the mechanical adjustment accuracy of the setting device for this model. The new setting value after setting is equal to the sum of the original setting value and the correction value for the channel. The new setting value must fall within the adjustable range of the setting value for this model. For example, if the original setting value of a channel is 80% of the rated current and the correction value is -3%, the new setting value after setting will be reduced to 77%. Cases exceeding this range have already been handled in the previous step. Alarm channels not covered by the threshold calibration parameters are not adjusted in this cycle, and their setting values ​​remain unchanged. The alarm pause command is generated one by one for each alarm channel. The command content includes the pause duration and pause start time for that channel. The pause duration is the stable observation period required for the setpoints of the alarm device model to take effect. The stable observation period for a certain model of alarm device is usually set to 24 hours. Triggers generated by this channel during the observation period are not included in subsequent false alarm statistics. The pause start time of the alarm pause command is strictly aligned with the setpoint adjustment completion time. After the alarm pause command is executed and the pause duration expires, the corresponding alarm channel resumes normal alarm function. The recovery action is automatically triggered by the alarm device itself according to the pause duration. When the channel recovers, it is already operating stably according to the new setpoints, and the previous setpoints are no longer effective.

[0061] Step S105: Based on the pause alarm command, determine the circuit breaker reset feasibility status to obtain a reset criterion group. Based on the reset criterion group, identify transferable paths, statistically analyze line losses to construct power saving benefit data. Through the power saving benefit data, perform a collaborative trade-off between power saving gain and risk in the operation sequence and output optimized control commands.

[0062] In some embodiments, the step of determining the circuit breaker reset feasibility status based on the pause alarm command to obtain a reset criterion set includes: identifying the degree of energy storage reset of the circuit breaker operating mechanism based on the pause alarm command to obtain the energy storage reset margin; comparing the dielectric recovery level according to the energy storage reset margin to determine the dielectric tolerance margin; performing electrical life loss verification on the dielectric tolerance margin to generate reset feasibility weights; and performing tiered analysis on the reset feasibility weights to obtain a reset criterion set.

[0063] The energy storage reset margin is obtained by identifying the degree of energy storage reset of the circuit breaker operating mechanism based on the pause alarm command. The pause alarm command traces back to the circuit breaker corresponding to its channel number. After the circuit breaker enters the pause observation period, the operating mechanism immediately initiates energy storage reset. The degree of energy storage reset is reported in real time by the spring energy storage stroke sensor of the operating mechanism. The closer the stroke reading is to the full stroke, the more complete the reset. The reporting cycle and the pause duration of the pause alarm command share the same clock reference to facilitate timing alignment. The degree of energy storage reset is read according to the percentage of the full stroke. When the percentage reaches the lower limit of closing readiness specified for this model of mechanism, it is considered to be reset in place. The lower limit is calibrated by the factory mechanical characteristic test report of this model of operating mechanism. The lower limit of closing readiness for a certain model of circuit breaker is generally 95% of the full stroke. If there are differences in the lower limit between different batches of the same model, the actual factory report of the equipment shall prevail. The energy storage reset margin is output for each circuit breaker individually. The margin is the difference between the current energy storage travel percentage and the lower limit of the closing readiness threshold. A positive difference indicates that the energy storage has crossed the readiness threshold and has a margin remaining; a negative difference indicates that the readiness threshold has not yet been reached. The margin value is directly expressed as a percentage difference in travel percentage. For circuit breakers where the energy storage travel sensor reading jumps during the pause observation period, the smaller of the two consecutive stable readings before and after the jump is included in the energy storage reset margin. This avoids the judgment result of a falsely inflated reset margin due to instantaneous jumps. The selection of the smaller value is also for the sake of stricter reset judgment; underestimating the margin is more conducive to ensuring reset safety than overestimating it. For circuit breakers not covered by the pause alarm command, their energy storage reset is still performed according to the normal procedure, and the energy storage reset margin is read as usual to ensure that all circuit breakers to be reset have the same diameter.

[0064] The dielectric tolerance margin is determined by comparing the dielectric recovery level based on the energy storage reset margin. Only circuit breakers with a positive energy storage reset margin are included in this comparison step. The dielectric recovery level being compared refers to the current dielectric recovery state between the contacts of the arc-extinguishing chamber. The criterion is whether the dielectric recovery curve after this trip has returned to the allowable range of the initial factory curve. Circuit breakers with a negative energy storage reset margin are not compared in this step, and their dielectric tolerance margin is directly marked as not ready. This is because discussing dielectric tolerance capacity before energy storage is in place is meaningless, and the "not ready" label is still included in the dielectric tolerance margin. The dielectric withstand margin is calculated as the difference between the withstand voltage corresponding to the current dielectric recovery curve and the current recovery voltage of the power grid where the circuit breaker is located. A larger difference indicates a greater overvoltage redundancy that the dielectric can currently withstand. For example, if the measured withstand voltage of a circuit breaker is 150 kV and the current recovery voltage of the power grid is 90 kV, the dielectric withstand margin is 60 kV. The withstand voltage is taken as the steady-state value on the measured recovery curve after this trip, with the measured value being the standard. This is because a positive energy storage reset margin only indicates mechanical readiness; the actual state of the dielectric still needs to be judged separately according to the current curve. Each circuit breaker has a corresponding record for the dielectric withstand margin, carrying three items: the current withstand voltage value, the current recovery voltage value, and the difference between the two. A negative difference indicates that the dielectric's current withstand capability is lower than the power grid recovery voltage level, and reset at this time carries the risk of re-breakdown. For circuit breakers whose current recovery voltage changes due to temporary adjustments in system operation, the adjusted real-time recovery voltage is used for comparison to avoid deriving the dielectric withstand margin from outdated power grid conditions. The time of data acquisition for the real-time recovery voltage and the time of sampling the dielectric recovery curve are kept within the same tripping event.

[0065] Electrical life wear verification is performed on circuit breakers with positive dielectric margins to generate reset feasibility weights. The verification checks whether the total contact wear corresponding to the cumulative breaking current is still within the allowable lifespan range. The cumulative breaking current is obtained by summing up the equipment log, covering all breaking records since the circuit breaker was put into operation. For example, if a circuit breaker has recorded twenty breaking events over eight years, its cumulative breaking current is the sum of these twenty measured values. For circuit breakers with negative dielectric margins or marked as not ready, electrical life wear verification is not performed, and the reset feasibility weight is directly set to the lowest level. These circuit breakers do not currently meet the reset conditions, regardless of whether their electrical life is sufficient. The lowest level value is the lower bound of a preset range to distinguish them from cases with missing data. The reset feasibility weight is determined by both the dielectric tolerance margin and the remaining electrical life. The more abundant both are, the higher the weight. The weight comprehensively reflects the degree of certainty that the circuit breaker will operate reliably after a current reset. The relative proportions of these two factors are derived from the statistical analysis of historical reset failure cases for this type of circuit breaker, with dielectric and electrical life factors allocated according to their respective frequencies of failure. Each circuit breaker generates its own reset feasibility weight, falling within a preset range. The value itself is dimensionless, representing only a relative level. Weights from the same batch of circuit breakers can be compared horizontally, but weights from different models should not be directly compared due to different proportion configurations. For circuit breakers whose cumulative breaking current is close to the upper limit of their allowable lifespan, even with a positive dielectric tolerance margin, the reset feasibility weight is correspondingly lowered. The lowering effect of insufficient remaining electrical life on the weight is not offset by an abundant dielectric tolerance margin.

[0066] A reset criterion group is obtained by classifying the reset feasibility weights. The reset feasibility weights are divided into three levels: resettable, pending observation, and prohibited reset, based on their numerical values. The boundary between these levels is taken from the historical reset success rate statistics for that circuit breaker model, set according to the point where the success rate drops. Historical reset success rates are statistically analyzed for different weight ranges. The point where the success rate drops significantly is the boundary between adjacent levels. For example, statistics for a certain model show that the success rate is above 90% for weights above 0.6, but drops sharply to below 60% after falling below 0.6; this point is taken as the boundary between the resettable and pending observation levels. Circuit breakers with reset feasibility weights in the resettable level are marked as allowing immediate reset in the reset criterion group; those in the pending observation level are marked as requiring delayed reset with a suggested observation period; and those in the prohibited reset level are marked as not resetting for the time being, with the suggested observation period being the historical average time required for the circuit breaker model to recover from the current weight to the resettable level. The reset criterion group collects the classification results for each circuit breaker. Each result includes the original value of the reset feasible weight, the corresponding classification level, and the suggested observation time for the classification level to be observed. The original value is for traceability, the classification level is for direct judgment, and the suggested observation time is for subsequent path continuation. Circuit breakers whose reset feasible weight falls exactly on the boundary value between two levels are classified into the lower level according to the principle of choosing the lower value. Boundary values ​​are all classified downwards. Reset is related to equipment and personal safety, so strict handling is more prudent than lenient. The results of the lower-level classification are also fully included in the reset criterion group, with boundary hit indicators so that people know that the circuit breaker is in a critical state. Circuit breakers whose reset feasible weight cannot be calculated due to missing data are marked as criterion to be supplemented in the reset criterion group and no classification level is given for the time being. They will be reclassified after the data is supplemented. Before the data is supplemented, they will not be archived. Circuit breakers with criteria to be supplemented will be skipped in the subsequent transferable path identification and will be included after reclassification.

[0067] Based on the reset criterion group, transferable power supply paths are identified, and line losses are statistically analyzed to construct energy-saving benefit data. Transferable power supply paths refer to power supply routes that can be put back into operation after a circuit breaker is reset, thus assuming load transfer. Path identification only applies to circuit breakers marked as resettable or under observation in the reset criterion group. Circuit breakers in the "prohibited reset" position are not included in path identification, and their corresponding routes are considered currently unavailable. For routes containing circuit breakers marked as under observation in the reset criterion group, the suggested observation period for that circuit breaker is included during identification. The actual usability start time of the route is postponed until the observation period expires. The postponed start time is also recorded in the corresponding record of that route in the energy-saving benefit data for subsequent calculation based on the actual usability time. Each transferable path corresponds to a load transfer scheme. The line loss of each scheme is calculated segment by segment based on the product of the square of the current and the resistance of each line segment after the transfer. The resistance used in the calculation is a corrected value based on the current temperature of the line, adjusted according to the actual measured temperature. For example, if the rated resistance of a certain section of conductor is 0.2 ohms per kilometer at 20 degrees Celsius, and the actual measured line temperature is 35 degrees Celsius, it is used after adjustment according to the temperature coefficient. For sections without real-time temperature monitoring, the resistance is uniformly adjusted according to the ambient temperature of the day to be as close as possible to the actual resistance of the conductor. The energy-saving benefit data is listed separately for each path. The benefit is the difference in line loss before and after the transfer. A positive difference indicates that the line loss decreases after the transfer and there is room for energy saving. Paths with negative or zero differences are also included, but only indicate that the path does not bring energy-saving benefits, for subsequent comparison of all candidate paths. For a transferable power path that relies on the reset of two circuit breakers, the energy saving benefit data is immediately included when both circuit breakers are in the resettable position. For paths that are under observation, the data is included after the recommended observation period of the later circuit breaker. If either circuit breaker is in the reset-prohibited position, the entire path is unusable.

[0068] The power saving benefit data is used to optimize control commands by balancing the power saving gain and risk of the operation sequence. The power saving gain is the sum of the power loss reduction values ​​of each transferable path in the power saving benefit data, expressed in kilowatts. The risk is obtained by inverse mapping of the reset feasible weight values ​​of each circuit breaker participating in the operation sequence; the lower the weight value, the higher the risk. The synergistic trade-off is calculated as J = G - λ·R_risk: G is the power saving gain; R_risk is the operation sequence risk, obtained by inverse mapping of the reset feasible weight values ​​of each circuit breaker; λ is the risk penalty coefficient, whose dimension is the same as that of λ·R_risk, and its value comes from the grid dispatching's setting of the relative priority between power saving and safety; when safety is prioritized, λ takes a larger value; J is the comprehensive trade-off score. The same λ is used for the same batch of operation sequences to ensure that each sequence is compared under the same trade-off scale. The G used for scoring is directly taken from the sum of the line loss reduction values ​​of each path in the power saving benefit data. Operation sequences with positive energy-saving gains but corresponding high path risks in the energy-saving benefit data may not have positive J scores. Sequences with negative scores are not included in the optimized control instructions, and even if energy-saving potential exists, they will not be executed at the expense of increased risk. Optimized control instructions select operation sequences from high to low J scores. When scores are tied, the sequence with lower risk is prioritized to choose the more reliable one under the same energy-saving conditions. Operation sequences with positive J scores but containing circuit breakers to be added according to criteria are temporarily suspended from inclusion in optimized control instructions until the criteria are completed and the scores are re-evaluated. Each operation sequence given in the final optimized control instructions must simultaneously meet the conditions of positive energy-saving gain and positive comprehensive score. A sequence must be truly valid in both dimensions. If either dimension fails to meet the standard, the sequence will not be included in the final optimized control instructions.

[0069] To implement the above-described method embodiment, an outdoor high-voltage vacuum circuit breaker operation sequence optimization method is proposed to achieve the corresponding functional and technical effects. See also... Figure 4 , Figure 4 This paper presents a structural block diagram of an outdoor high-voltage vacuum circuit breaker operation sequence optimization system 400 according to an embodiment of this application, including: The sequence generation unit 401 is used to acquire real-time power grid operating condition data and match and analyze power saving target parameters to generate a candidate operation sequence table, identify maintenance plan postponement flags based on the candidate operation sequence table to obtain a preferred operation sequence group, and perform phase selection and circuit breaker verification through the preferred operation sequence group to obtain the initial operation sequence. The sequence construction unit 402 is used to perform load-bearing erroneous tripping record tracking to generate a risk node table for the initial operation sequence, identify potential erroneous operation hazards based on the risk node table, obtain hazard circuit breaker groups, and generate a dynamic rearrangement sequence by implementing live interlocking and interlocking verification through the hazard circuit breaker groups. The execution confirmation unit 403 is used to perform hierarchical instruction arrangement and output an execution instruction group according to the dynamic rearrangement sequence, use the execution instruction group to track the transient fluctuation of the circuit breaker restart to generate a status confirmation signal, and re-estimate the remaining steps on the status confirmation signal to construct a continued status record. An abnormal calibration unit 404 is used to identify the execution deviation from the continued status record to obtain an abnormal alarm signal, use the abnormal alarm signal to identify false alarms and backtrack to obtain threshold calibration parameters, and use the threshold calibration parameters to set the alarm threshold reset to form a pause alarm command. The instruction output unit 405 is used to determine the circuit breaker reset feasibility status based on the pause alarm instruction to obtain a reset criterion group, identify transferable paths based on the reset criterion group to construct power saving benefit data by statistically analyzing line losses, and output optimized control instructions by performing a collaborative trade-off between power saving gain and risk in the operation sequence based on the power saving benefit data.

[0070] The aforementioned outdoor high-voltage vacuum circuit breaker operation sequence optimization system 400 can implement the outdoor high-voltage vacuum circuit breaker operation sequence optimization method of the above method embodiment. The options in the above method embodiments are also applicable to this embodiment, and will not be detailed here. The remaining contents of this application embodiment can be referred to the contents of the above method embodiments, and will not be repeated in this embodiment.

[0071] The purpose of the above embodiments is to reproduce and derive the technical solution of the present invention by way of example, and to fully describe the technical solution, purpose and effect of the present invention. The purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosure of the present invention, and not to limit the scope of protection of the present invention.

[0072] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for optimizing the operation sequence of an outdoor high-voltage vacuum circuit breaker, characterized in that, include: Real-time power grid operation data is matched and analyzed with energy-saving target parameters to generate a candidate operation sequence list. Based on the candidate operation sequence list, maintenance plan postponement flags are identified to obtain a preferred operation sequence group. The initial operation sequence is obtained by performing phase selection and circuit breaker verification through the preferred operation sequence group. For the initial operation sequence, a risk node table is generated by tracking the load-bearing erroneous trip records. Based on the risk node table, potential erroneous operation hazards are identified and hazard circuit breaker groups are obtained. Through the hazard circuit breaker groups, live interlocking and interlocking verification is performed to generate a dynamic rearrangement sequence. Based on the dynamic rearrangement sequence, hierarchical instruction arrangement is carried out to output the execution instruction group. The execution instruction group is used to track the transient fluctuations of the circuit breaker restart and generate a status confirmation signal. The remaining steps are re-estimated based on the status confirmation signal to construct a continued status record. An abnormal alarm signal is obtained by identifying the execution deviation from the continued status record. The abnormal alarm signal is used to identify false alarms and backtrack to obtain threshold calibration parameters. The threshold calibration parameters are used to reset the alarm threshold to form a pause alarm command. Based on the pause alarm command, the circuit breaker reset feasibility status is determined to obtain a reset criterion group. Based on the reset criterion group, the transferable path is identified, line loss is statistically analyzed, and power saving benefit data is constructed. Through the power saving benefit data, the power saving gain and risk of the operation sequence are weighed together to output an optimized control command.

2. The method according to claim 1, characterized in that, The initial operation sequence obtained by performing phase selection and tripping verification through the preferred operation sequence group includes: For the preferred operation sequence group, the target phase angle of each circuit breaker is calculated to form a phase selection window table; Based on the phase selection window table, mechanical wear gap drift tracking is performed to obtain the expected arcing period; Based on the expected arcing period, overvoltage suppression verification is performed to output phase selection verification conclusions. The initial operation sequence is obtained by backfilling the phase selection verification conclusion into the operation steps.

3. The method according to claim 1, characterized in that, The step of generating a risk node table by tracking load-bearing erroneous circuit breakers in the initial operation sequence includes: A fault trip record library is generated by tracing historical overload fault trip records from the initial operation sequence; Based on the erroneous tripping record library, identify critical events approaching the action threshold to obtain a set of attempted erroneous operations; The risk node weights are generated by performing a critical frequency weighting on the set of attempted erroneous operations. The risk node weights are classified and archived to generate a risk node table.

4. The method according to claim 1, characterized in that, The process of generating a dynamic rearrangement sequence through live-line interlocking verification of the circuit breaker group with potential hazards includes: Based on the aforementioned circuit breaker group with potential hazards, adjacent energized states are tracked to obtain energized interlocking conditions. Based on the described live interlocking conditions, identify interlock conflicts before and after the operation and establish an interlocking verification list; Perform step-by-step executability verification on the interlock checklist to obtain an executable marker sequence; The executable tag sequence back-rowing operation step generates a dynamically rearranged sequence.

5. The method according to claim 1, characterized in that, The step of using the execution instruction group to track transient fluctuations during circuit breaker restart and generate a status confirmation signal includes: The reignition characteristic parameters are obtained by tracking the transient fluctuations of reignition between contacts after the circuit breaker trips according to the execution instruction group. The media recovery rating is determined by calculating the rate of change of the re-ignition frequency using the aforementioned re-ignition characteristic parameters. Based on the aforementioned medium recovery rating, the confirmation degree of the circuit breaker tripping position is verified to obtain the confirmation degree index. The arrival confirmation index is used to determine the arrival confirmation interval and generate a status confirmation signal.

6. The method according to claim 1, characterized in that, The step of using the abnormal alarm signal to identify false alarms and backtrack to obtain threshold calibration parameters includes: The abnormal alarm signals are used to identify alarm channels that frequently and repeatedly report false alarms to form a candidate set of falsely low set values; Based on the fixed low candidate set, a list of false alarm features is obtained by tracking historical action threshold margins. The offset verification table is obtained by verifying the sensitivity offset of each mode against the false alarm feature list. The alarm sensitivity correction value is calculated using the offset verification table, and the threshold calibration parameter is output.

7. The method according to claim 1, characterized in that, The reset criterion group obtained by determining the circuit breaker reset feasibility status based on the pause alarm command includes: Based on the pause alarm command, the degree of energy storage reset of the circuit breaker operating mechanism is identified to obtain the energy storage reset margin. The dielectric bearing capacity is determined by comparing the dielectric recovery level based on the energy storage reset margin. The electrical life loss is checked to generate a reset feasible weight value based on the dielectric's tolerance margin; The reset feasibility weights are analyzed in a tiered manner to obtain a reset criterion group.

8. The method according to claim 4, characterized in that, The stepwise executability verification of the interlock checklist to obtain the executable marker sequence includes: For each step of the interlock verification list, a set of pre-conditions is established to track the pre-energized conditions. Based on the aforementioned pre-constraint set, identify constraints that dynamically fail during execution and generate a real-time failure list; The real-time failure list is used to identify a delayed candidate list by observing a preset duration window after the constraints have just been restored. The list of delayed candidates is back-ordered to obtain an executable tag sequence of subsequent unexecuted steps.

9. The method according to claim 5, characterized in that, The determination of the medium recovery rating by calculating the rate of change of reignition frequency using the reignition characteristic parameters includes: The re-ignition frequency sequence is obtained by statistically analyzing the number of re-ignitions occurring during a single circuit breaker trip using the aforementioned re-ignition characteristic parameters. The rate of change of the re-ignition frequency is calculated by tracking the post-arc recovery changes between adjacent circuit breakers based on the re-ignition frequency sequence. A cumulative ablation assessment table is established by performing ablation accumulation verification using the aforementioned re-ignition frequency change rate. The medium recovery rating is determined by identifying the slow-recovery item in the ablation accumulation assessment table.

10. An outdoor high-voltage vacuum circuit breaker operation sequence optimization system, characterized in that, include: The sequence generation unit is used to acquire real-time power grid operating condition data and power saving target parameters, match and analyze them to generate a candidate operation sequence list, identify the maintenance plan postponement flag according to the candidate operation sequence list to obtain the preferred operation sequence group, and perform phase selection and circuit breaker verification through the preferred operation sequence group to obtain the initial operation sequence. The sequence construction unit is used to perform load-bearing erroneous tripping record tracking to generate a risk node table for the initial operation sequence, identify potential erroneous operation hazards based on the risk node table, obtain hazard circuit breaker groups, and generate a dynamic rearrangement sequence by implementing live interlocking and interlocking verification through the hazard circuit breaker groups. The execution confirmation unit is used to perform hierarchical instruction arrangement and output execution instruction group according to the dynamic rearrangement sequence, use the execution instruction group to track the transient fluctuation of the circuit breaker restart to generate a status confirmation signal, and re-estimate the remaining steps based on the status confirmation signal to construct a continued status record. An abnormal calibration unit is used to identify the execution deviation from the continued status record to obtain an abnormal alarm signal, use the abnormal alarm signal to identify false alarms and backtrack to obtain threshold calibration parameters, and use the threshold calibration parameters to set the alarm threshold reset to form a pause alarm command. The instruction output unit is used to determine the circuit breaker reset feasibility status based on the pause alarm instruction to obtain a reset criterion group, identify transferable paths based on the reset criterion group to statistically analyze line losses and construct power saving benefit data, and output optimized control instructions by performing a collaborative trade-off between power saving gain and risk in the operation sequence based on the power saving benefit data.

Images