Method for dynamic collection of centralized feeder faults based on topology and communication characteristics

Abstract

The present application relates to power distribution network fault processing technical field, especially based on topology and communication feature's centralized type feeder fault dynamic collection method, the general technical thought of this method is: system after starting, according to the current operating mode generation feeder automation terminal sequence, the sequence only includes the automation switch, FTU / DTU, protection device or fault indication terminal with the main station can be collected fault detection capability;Ordinary non-automation switch, manual switch, no communication ability equipment is not as fault feature chain convergence judgment object.System checks automation terminal according to power supply direction step by step.The current most downstream confirmed fault feature terminal is only temporary boundary node, only when its downstream adjacent can be judged terminal is confirmed without fault feature, the signal collection is allowed to end;If the downstream adjacent can be judged terminal has not completed checking, it is a to-be-confirmed node, executes terminal waiting, active inquiry or degradation processing.

Description

Technical Field

[0001] This invention relates to the field of power distribution network fault handling technology, and in particular to a centralized feeder fault dynamic collection method based on topology and communication characteristics. Background Technology

[0002] In centralized feeder automation systems for distribution networks, when a fault such as a short circuit occurs on a line, the distribution automation master station needs to collect the fault signals sent by each terminal within a very short time to locate the fault and restore power to non-faulty areas. In current engineering practices, fault signal collection generally adopts a strategy of fixed waiting time or maximum waiting time; that is, after a fault is initiated, a preset waiting period is observed before proceeding with subsequent self-healing logic. This method is simple to implement and was widely used in early systems with low levels of automation.

[0003] However, when using a fixed waiting time, if a critical fault signal arrives early, the system still needs to wait for the preset time to end, resulting in invalid delays and reducing self-healing efficiency. Conversely, if a critical terminal delays uploading due to communication congestion or link interruption, fixed waiting may cause the system to enter the location logic before all signals are received, affecting the accuracy of fault location. When a terminal is offline, the maximum waiting mechanism will cause the system to wait for a long time until timeout, severely slowing down the self-healing process. In addition, existing methods generally lack expected signal set constraints based on feeder topology, making it difficult for the master station to distinguish between signals related to the current fault and alarms from unrelated branches, making it susceptible to alarm storm interference. Different terminals have significant differences in communication media and quality (such as fiber optic and wireless public networks), and a unified waiting strategy cannot balance speed and completeness: shortening the waiting time may cause late signals to be missed, while extending the waiting time sacrifices the self-healing efficiency of fast lines. Finally, when the boundary switch signal of a suspected faulty section is missing, existing systems mostly passively wait or ignore it after timeout, lacking an active compensation mechanism, leading to blurred fault location boundaries or even location failure.

[0004] Therefore, how to dynamically and intelligently control the acquisition of fault signals during the self-healing process of centralized feeders, ensuring that critical terminal signals are not missed and avoiding unnecessary waiting, thereby achieving the optimal balance between accuracy, real-time performance and reliability, is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] Technical Objective: To overcome the shortcomings of existing technologies, this invention provides a centralized dynamic collection method for feeder faults based on topology and communication characteristics. Based on feeder topology prediction, terminal communication characteristics, and active inquiry mechanism, it improves the accuracy, real-time performance, and interpretability of fault signal acquisition.

[0006] Technical Solution: To achieve the above objectives, this invention discloses a centralized feeder fault dynamic collection method based on topology and communication characteristics, comprising the following steps: S1 generates an automated terminal sequence: Based on the fault characteristic activation signal, the power supply direction and automated terminal sequence of the current feeder are determined. The automated terminal sequence is a set of automated terminals arranged along the power supply direction and equipped with the main station's fault detection capability. S2 forms a strong convergence fault boundary: Scan the terminal status in the automated terminal sequence along the power supply direction, identify the current downstream fault characteristic terminal U, continue to scan downstream, find the fault-free characteristic terminal D adjacent to U, and form a strong fault convergence boundary UD by U and D. S3 Delayed Signal Verification: After the strong convergence fault boundary UD is output, the system continuously receives delayed signals from the automation terminal and uses these signals to verify or re-determine the strong convergence fault boundary UD.

[0007] Further, S1 includes: S11 identifies fault start signals: using overcurrent start, protection action, fault indication or switch trip signals sent from each terminal of the feeder, it determines the feeder to which the fault belongs, the fault judgment time window and the source of the initial signal; S12 analyzes the current operating mode: Based on the feeder topology, power supply point, switch on / off status, interconnection operating mode, and branch relationship, determine the current power supply direction and actual energized path; S13 Screening of Automated Terminals: Select automated terminals from all terminals that have fault detection capabilities and can provide protection actions, fault indications, telemetry or SOE information to the master station. The automated terminals include automated switches, feeder terminal units (FTUs), distribution terminal units (DTUs), protection devices or fault indication terminals. Ordinary non-automated switches are not included in this sequence. S14 Forming an automated terminal sequence: The selected automated terminals are sorted along the power supply direction to form the automated terminal sequence.

[0008] Furthermore, S1 also includes: S15 marking each terminal according to the signal status of the terminals in the automated terminal sequence, specifically as follows: Mark the terminals corresponding to overcurrent initiation, protection action, and fault indication that have been received and passed validity verification as "fault characteristics confirmed"; Terminals that are confirmed to have no protection action, no fault indication, and no fault current characteristics through active calling or cache query will be marked as "confirmed to have no fault characteristics"; Terminals that have not yet completed verification are marked as "pending confirmation"; Terminals that are unreachable and whose calls have failed are marked as "unreachable".

[0009] Furthermore, the specific method for forming the strongly convergent fault boundary UD in S2 includes: S21: Scan each terminal from the power supply side to the load side along the automated terminal sequence; S22: If the current terminal is marked as "fault characteristics confirmed", then this terminal is the current downstream fault characteristic terminal U. Continue scanning downstream to find the downstream adjacent verifiable terminal D' of the current downstream fault characteristic terminal U, and determine: If the downstream adjacent terminal D' has been marked as "faulty feature confirmed", then update the downstream adjacent terminal D' to the new current downstream faulty feature terminal U, and continue scanning downstream; If the downstream adjacent verifiable terminal D' has been marked as "confirmed to be fault-free", then the downstream adjacent verifiable terminal D' is determined as the downstream adjacent fault-free terminal D, and U and D form a strong convergent fault boundary UD; If the downstream adjacent terminal D' can be determined to be marked as "pending confirmation", then D' is a blocked node, and node-level waiting or active querying is performed on the blocked node; If the downstream neighboring terminal D' is marked as "unreachable", or if a valid response cannot be obtained after waiting or actively querying at the node level, then the waiting for that node is stopped. Based on the received terminal signal, the most recent valid remote signaling status, and the topology connection relationship, the candidate fault segment for degradation is inferred and output. At the same time, the candidate fault segment for degradation is marked with the "unconfirmed" attribute and the corresponding confidence level. It is not regarded as a strong convergence conclusion, and the strong convergence determination process is exited, and the process is transferred to the manual confirmation or secondary status verification stage.

[0010] Furthermore, based on the communication characteristics of the blocked node, a node-level waiting threshold and an active query trigger time are determined. The communication characteristics are parameter information describing the terminal's communication reachability, communication quality, and data retrieval response success rate.

[0011] Furthermore, based on the terminal's real-time online status, message response latency, and packet loss rate, the communication characteristics of the blocked node are divided into four levels: normal communication status, communication jitter status, abnormal communication status, and offline status.

[0012] Furthermore, based on the communication characteristic level of the blocked node, the corresponding acquisition control strategy is executed: When the communication characteristics of the blocked node are in a normal communication state, the remote signaling and protection signal active call is initiated first. If the call is successful and there are no fault characteristics, the collection waiting for the node ends. When the communication characteristics of the blocked node are in a communication jitter state, the acquisition waiting time of the node is extended, and an active call is triggered simultaneously to wait for the late signal; When the communication characteristics of the blocked node are in an abnormal communication state or an offline state, the waiting for the node is terminated, and the process of outputting the downgraded candidate fault section is initiated.

[0013] Furthermore, in S3, the late signal verification standard is as follows: If the terminal location and fault characteristic state corresponding to the late signal match the strong convergence fault boundary UD, then the late signal is recorded as a supplement and the reliability of the current fault segment is improved. If the terminal location or fault characteristic state corresponding to the late signal conflicts with the strong convergence fault boundary UD, the fault section re-judgment is triggered and the review procedure is started.

[0014] Furthermore, the verification procedure includes signal timescale verification, topology verification, and communication verification.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: 1. This invention does not employ a fixed global waiting time. Instead, it generates an automated terminal sequence based on the feeder topology and checks fault characteristics level by level along the power supply direction. When the downstream adjacent terminal of the terminal with the most downstream fault characteristic has been confirmed to have no fault characteristics, the system can immediately terminate signal acquisition and output the fault boundary without waiting for other unrelated terminals. Compared to traditional fixed waiting or maximum waiting strategies, this significantly shortens the decision time for fault location and self-healing recovery, making it particularly suitable for high-speed lines with good communication quality.

[0016] 2. When critical terminals delay sending signals due to communication congestion, link jitter, or momentary failures, existing methods are prone to missing signals due to insufficient waiting time, leading to incorrect fault location. This invention introduces a "collection blocking node" mechanism: as long as a downstream adjacent terminal has not yet confirmed the presence of fault characteristics, the system is blocked, collection cannot be prematurely terminated, and node-level waiting or active querying is performed on that terminal until a clear conclusion is obtained or it is confirmed that the terminal is unreachable. This mechanism fundamentally avoids misjudging the absence of fault characteristics due to delayed signal arrival, ensuring the accuracy of fault location.

[0017] 3. This invention adopts a differentiated acquisition and control strategy for different terminals: for congested nodes with good communication, priority is given to quickly calling for confirmation; for nodes with communication jitter or poor quality, the waiting time is appropriately extended and multiple polling methods are used; for offline or continuously unreachable terminals, the output is downgraded in a timely manner and the credibility is marked to avoid long-term invalid waiting.

[0018] 4. After the faulty section is output, this invention continues to receive delayed signals from the automated terminal. If the delayed signal matches the already output boundary, it is used to supplement the record and improve reliability; if the delayed signal conflicts with the already output conclusion (such as when a downstream terminal sends up fault characteristics late), it automatically triggers signal time stamp verification, topology verification, communication verification, and faulty section re-judgment procedures. This mechanism effectively prevents long-term erroneous location caused by delayed signals and enhances the self-healing reliability of the system in complex communication environments. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the feeder in the embodiment. Detailed Implementation

[0020] The following is in conjunction with the appendix Figure 1 The principles and features of the present invention are described, and the examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0021] The overall technical concept of the fault dynamic collection method of this invention can be summarized as follows: topology and protection logic determine the verification order, the automated terminal sequence determines the fault boundary, communication characteristics determine the acquisition control strategy, and late signals are used for verification and correction. After a fault is initiated, the system does not directly presume the location of the fault point, nor does it directly presume which switches will necessarily send protection signals. Instead, it generates a feeder automated terminal sequence based on the current operating mode. This sequence only includes automated switches, FTUs / DTUs, protection devices, or fault indication terminals with fault detection capabilities that can be acquired by the master station; ordinary non-automated switches, manual disconnectors, and equipment without communication capabilities are not considered as objects for fault feature chain convergence judgment. The system verifies the automated terminals level by level according to the power supply direction. The current downstream terminal with confirmed fault characteristics is only a temporary boundary node. Signal acquisition is only allowed to end when its downstream adjacent verifiable terminal is confirmed to have no fault characteristics; if the downstream adjacent verifiable terminal has not yet completed verification, it is a node awaiting confirmation, and terminal waiting, active querying, or downgrading processing is performed. The specific steps of the fault dynamic collection method of this invention are now described in detail.

[0022] (a) Step S1: Generate automated terminal sequence Step S1 includes: determining the power supply direction and automation terminal sequence of the current feeder based on the fault characteristic activation signal. The automation terminal sequence is a set of automation terminals arranged along the power supply direction, capable of acquiring fault detection data from the main station. Specifically, it includes the following sub-steps S11 to S15: S11 identifies fault start signals: using overcurrent start, protection action, fault indication or switch trip signals sent from each terminal of the feeder, it determines the feeder to which the fault belongs, the fault judgment time window and the source of the initial signal.

[0023] Specifically, when the distribution automation master station receives an overcurrent start, protection action, fault indication, or switch trip signal from any terminal, it identifies the signal as a fault characteristic start signal. Based on this, it determines the feeder to which the fault belongs, the fault identification time window, and the first terminal to send the fault characteristic start signal.

[0024] S12 analyzes the current operating mode: Based on the feeder topology, power supply point, switch on / off status, interconnection operating mode, and branch relationship, determine the current power supply direction and actual energized path.

[0025] Specifically, after S11 determines the feeder to which the fault belongs, S12 retrieves the network topology model of that feeder from the topology database of the distribution automation master station. Since the distribution network usually operates in an open-loop state, and the operating mode may change dynamically due to load transfer caused by the fault, S12 must obtain the switch opening and closing status before the fault occurred, and calculate the actual power supply direction and energized path of the feeder at the current moment by combining the power supply point distribution of the feeder, the operating mode of the tie switch, and the feeder branch relationship.

[0026] S13 Screening of Automated Terminals: Screening out automated terminals from all terminals that have fault detection capabilities and can provide protection actions, fault indications, telemetry or SOE (Sequence of Events) information to the master station. The automated terminals include automated switches, feeder terminal units (FTUs), distribution terminal units (DTUs), protection devices or fault indication terminals. Ordinary non-automated switches are not included in this sequence. Specifically, various types of equipment terminals are distributed along the actual energized path determined by S12. Ordinary non-automatic switches (such as manually operated switches and disconnectors without communication capabilities) lack the ability to send fault detection information to the master station and cannot be used as a basis for fault characteristic identification; therefore, they are excluded. Only automated terminals with fault detection capabilities and the ability to provide protection action signals, fault indication signals, telemetry data, or SOE information to the master station are selected for subsequent processing.

[0027] S14 Forming an automated terminal sequence: The selected automated terminals are sorted along the power supply direction to form the automated terminal sequence.

[0028] S15 marks each terminal according to its signal status in the automated terminal sequence. Within the fault diagnosis time window, as the master station continuously receives signals from each terminal, it marks each terminal in the sequence based on its current signal reception status, including the following four marking types: (1) Mark the terminals corresponding to overcurrent initiation, protection action, and fault indication that have been received and passed the validity verification as "fault characteristics confirmed". The validity verification includes message integrity verification, signal timestamp verification, and signal type and protection action logic consistency verification. For duplicate or conflicting signals, deduplication and conflict identification are performed. Only signals that pass the above verification are adopted as valid fault characteristics.

[0029] (2) Terminals that have been confirmed to have no protection action, no fault indication, and no fault current characteristics through active calling or cache query are marked as "confirmed to have no fault characteristics". When the master station needs to confirm whether a terminal is indeed without fault characteristics, it can initiate an active call to the terminal or query the historical data of the terminal cached by the master station. If it is confirmed that the terminal has no fault-related records within the fault judgment time window, it is marked as "confirmed to have no fault characteristics".

[0030] (3) Mark terminals that have not yet completed verification as “pending confirmation”. If a terminal neither sends a fault characteristic signal nor completes the active call verification of it by the master station, the terminal’s status is temporarily marked as “pending confirmation”.

[0031] (4) Terminals that are unreachable and whose calls fail are marked as “unreachable”. If the master station initiates an active call to a terminal but cannot obtain any effective response due to reasons such as the terminal being offline or the communication channel being unreachable, the terminal is marked as “unreachable”.

[0032] (ii) Step S2: Forming a strong convergence fault boundary Step S2 includes: scanning the terminal states in the automated terminal sequence along the power supply direction, identifying the current downstream fault-feature terminal U, continuing to scan downstream to find the fault-free terminal D adjacent to U, and forming a strong fault convergence boundary UD between U and D. Step S2 specifically includes the following sub-steps S21 to S22: S21: Scan each terminal from the power supply side to the load side along the automated terminal sequence. The master station scans each terminal in the sequence in the downstream direction according to the order of the automated terminal sequence, starting from the first terminal on the power supply side, and reads the marking status of each terminal.

[0033] S22: Judge step by step and form a strong convergence fault boundary.

[0034] If the current terminal is marked as "fault characteristics confirmed", then this terminal is the current downstream terminal U with the most downstream fault characteristics. Then, the scan continues downstream to find the next downstream terminal D' that can be judged, and the process is divided into different cases based on the marking status of D': Scenario 1: If the downstream adjacent terminal D' has been marked as "fault feature confirmed", then update the downstream adjacent terminal D' to the new current downstream fault feature terminal U, and continue scanning downstream. This means that the fault feature chain has not been interrupted and needs to be extended downstream to find the real end of the fault feature chain.

[0035] Scenario 2: If the downstream adjacent verifiable terminal D' has been marked as "confirmed fault-free," then the downstream adjacent verifiable terminal D' is determined as the downstream adjacent fault-free terminal D, and U and D form a strongly convergent fault boundary UD. This is the ideal situation: a clear fault segment boundary is formed between the fault-feature terminal and its first downstream fault-free terminal. At this point, the signal acquisition process can end, and the strongly convergent fault boundary UD is output.

[0036] Scenario 3: If the downstream neighboring terminal D' is marked as "pending confirmation," then D' is a blocked node, and node-level waiting or active polling is performed on the blocked node. This is the case where the signal has not yet fully arrived. The system must not directly use the current U as the final boundary; it must wait or actively poll D'. In this case, the system determines the node-level waiting threshold and the active polling trigger time based on D''s communication characteristics. After D's status is updated to "confirmed fault characteristics" or "confirmed no fault characteristics," the system returns to Scenario 1 or Scenario 2 to continue processing; if the waiting or polling times out, it proceeds to Scenario 4.

[0037] After identifying a blocked node, the system needs to determine specific waiting and polling strategies based on the node's communication characteristics. These characteristics are categorized into four levels: normal communication state, communication jitter state, communication abnormal state, and offline state.

[0038] Normal communication status means that the terminal is online and can be successfully initiated within a specified time limit.

[0039] Communication jitter refers to a situation where the terminal is still online, but there are significant fluctuations in message response latency or packet loss rate, which may result in delayed or intermittent responses.

[0040] An abnormal communication state refers to a situation where, although the terminal is not determined to be offline, it continues to experience high latency, high packet loss, repeated failed calls, or is unable to obtain a reliable response within a specified time limit.

[0041] Offline status refers to the situation where the main station is unable to maintain a communication connection with the terminal, the communication link is interrupted, and an effective active call cannot be implemented.

[0042] Based on the communication characteristic level of the blocked node, execute the corresponding acquisition control strategy: Strategy 1: When the communication characteristics of the blocked node are in a normal communication state, prioritize initiating a remote signaling and protection signal activation. If the activation is successful and no fault characteristics are found, the data acquisition waiting for that node ends. Specifically, if the activation is successful and verification confirms that the terminal has no protection action, no fault indication, and no fault current characteristics within the fault identification time window, the data acquisition waiting for that node ends, its mark is updated to "confirmed no fault characteristics," and the process returns to scenario 2. If the activation is successful and a fault characteristic signal is received, its mark is updated to "confirmed fault characteristics," and the process returns to scenario 1.

[0043] Strategy Two: When the communication characteristics of the blocked node are in a jittery state, extend the node's acquisition waiting time and simultaneously trigger an active call to wait for the late signal. Specifically, based on the normal waiting threshold, dynamically increase the waiting time according to the current communication quality, and immediately initiate an active call for remote signaling and protection signals to avoid misjudging the late signal as no signal due to communication delay. If a valid response is successfully received within the extended waiting time, update the flag according to the logic of Strategy One and return to the corresponding branch; if there is still no valid response after the extended waiting timeout, update the terminal flag to "unreachable" and proceed to Strategy Three for processing.

[0044] Strategy 3: When the communication characteristics of the blocked node are in an abnormal communication state or an offline state, the waiting for that node is terminated, and the process proceeds to the output of the downgraded candidate fault segment. Specifically, following the logic of step S22, case four, the downgraded candidate fault segment is inferred and output based on the received terminal signal, the most recent valid remote signaling status, and the topology connection relationship. It is marked with an "unconfirmed" attribute and a reliability level, exits the strong convergence judgment process, and proceeds to the manual confirmation or secondary status verification stage. For both the abnormal communication state and the offline state, the cause of the communication anomaly and the data quality are recorded respectively, and the reliability level of the downgraded candidate fault segment is determined accordingly.

[0045] Scenario 4: If the downstream neighboring terminal D' is marked as "unreachable," or if a valid response cannot be obtained after waiting or active inquiry at the node level, then the waiting for that node stops. Based on the received terminal signals, the most recent valid remote signaling status, and the topology connections, a candidate fault segment for degradation is inferred and output. This candidate fault segment is marked with an "unconfirmed" attribute and its corresponding confidence level. It is not considered a strong convergence conclusion, and the process exits the strong convergence determination process, proceeding to manual confirmation or a secondary status verification stage. This is the degradation handling situation under extreme communication anomalies. When critical boundary nodes are completely unreachable, the system cannot wait indefinitely; it must infer based on existing information. The inference result is marked with an "unconfirmed" attribute and a confidence level (e.g., high, medium, low). It is not output as a strong convergence conclusion but is instead proceeded to a secondary status verification stage or manual confirmation.

[0046] (III) Step S3: Late arrival signal verification Step S3 includes: after the strong convergence fault boundary UD is output, continuously receiving delayed signals from the automation terminal, and using the delayed signals to review or re-determine the strong convergence fault boundary UD. In actual distribution network operation, due to communication delays, terminal processing delays, etc., some terminal fault signals may arrive at the master station only after the strong convergence fault boundary UD has been output. These "delayed signals" may contain information consistent with the already output conclusions, or they may contain information that conflicts with the already output conclusions. Therefore, step S3 provides a post-processing mechanism for delayed signals.

[0047] The criteria for reviewing late arrival signals are as follows: (1) If the terminal position and fault characteristic state corresponding to the late signal match the strong convergence fault boundary UD, then the late signal is recorded as a supplement and the credibility of the current fault segment is improved.

[0048] (2) If the terminal location or fault characteristic state corresponding to the late signal conflicts with the strong convergence fault boundary UD, the fault segment re-judgment is triggered, and the review procedure is started. The review procedure includes signal time stamp review, topology review, and communication review. Among them, the signal time stamp review is used to verify whether the time stamp of the late signal is indeed within the fault judgment time window (if the time stamp exceeds the window, it may be a signal of other events and should not be used as the basis for re-judgment); the topology review is used to verify whether the position of the signal source terminal in the topology is consistent with the current sequence; the communication review is used to verify whether the communication status of the terminal at the time of the fault is reliable.

[0049] After review, if the delayed signal is confirmed to be valid and conflicting with the boundary, step S2 is executed again to form a new fault boundary; if the review finds that the delayed signal is invalid, the original strong convergence fault boundary UD is maintained unchanged.

[0050] The following is the method execution process when a specific feeder fails: Case 1: For example Figure 1 As shown, a short-circuit fault occurs on a 10kV feeder, and the feeder outlet switch CB1 first sends an overcurrent start signal. Along the power supply direction, the feeder has five automated terminals: CB1, S1, S2, S3, and S4, each equipped with fault detection or fault information acquisition capabilities.

[0051] The system successively receives overcurrent or fault indication signals from CB1, S1, and S2. After confirming their validity through time stamping and topology verification, CB1, S1, and S2 are marked as "fault characteristics confirmed." At this point, the downstream terminal with the most downstream fault characteristic is S2. Its downstream neighboring terminal, S3, has not yet sent any signals and is marked as "pending confirmation," becoming a blocked node. The system queries the communication profile of S3 and confirms that S3 uses fiber optic communication with a communication characteristic of "normal communication status." Therefore, it prioritizes initiating remote signaling and protection signal activation. S3 quickly returns the query results, confirming that it has no protection action, no fault indication, and no fault current characteristics within the fault identification time window. The system updates the S3 mark to "no fault characteristics confirmed," forming a strongly convergent fault boundary S2-S3, and terminates signal acquisition. After the boundary output, if the downstream S4 sends a fault characteristic signal late, it will not affect the boundary conclusion because it is located downstream of the boundary. It will only be used as a supplementary record for accident tracing. If the downstream S3 sends a fault characteristic signal late, it will conflict with the boundary conclusion and trigger the signal time stamp verification, topology verification and communication verification procedures. After the verification confirms that the signal is valid, the fault boundary determination will be re-executed.

[0052] Case 2: After a feeder failure, the system receives fault feature signals from CB1, S1, and S2, and marks CB1, S1, and S2 as "fault features confirmed". The current downstream fault feature terminal is S2. Its downstream adjacent terminal S3 has not yet sent any valid information and is marked as "pending confirmation", becoming a blocked node.

[0053] Unlike Case 1, in this embodiment, S3 uses wireless public network communication. Based on its recent message response latency and packet loss rate, the system determines its communication characteristic to be "communication jitter state." According to this communication characteristic level, the system executes a differentiated acquisition control strategy: extending the node-level waiting threshold for S3 and simultaneously triggering an active call. After the delay, S3 successfully returns a response, confirming its fault-free characteristics. The system updates S3's flag to "confirmed fault-free characteristics," forming a strong convergence fault boundary S2-S3, and ends the acquisition. If S3 consistently fails to respond within the extended waiting time, the system updates S3's flag to "unreachable." Based on the received fault characteristic signals of CB1, S1, and S2, S3's most recent valid remote signaling status, and topology connections, the system infers and outputs downgraded candidate fault segments, marking them with an "unconfirmed" attribute and credibility level. This is not considered a strong convergence conclusion and is transferred to the manual confirmation or secondary status verification stage. This embodiment demonstrates the invention's ability to differentiate the handling of communication jitter terminals. By dynamically adjusting the waiting threshold and actively querying, it avoids both signal loss due to communication delays and efficiency loss caused by excessively long fixed waiting times.

[0054] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A centralized feeder fault dynamic collection method based on topology and communication characteristics, characterized in that, Includes the following steps: S1 generates an automated terminal sequence: Based on the fault characteristic activation signal, the power supply direction and automated terminal sequence of the current feeder are determined. The automated terminal sequence is a set of automated terminals arranged along the power supply direction and equipped with the main station's fault detection capability. S2 forms a strong convergence fault boundary: Scan the terminal status in the automated terminal sequence along the power supply direction, identify the current downstream fault characteristic terminal U, continue to scan downstream, find the fault-free characteristic terminal D adjacent to U, and form a strong fault convergence boundary UD by U and D. S3 Delayed Signal Verification: After the strong convergence fault boundary UD is output, the system continuously receives delayed signals from the automation terminal and uses these signals to verify or re-determine the strong convergence fault boundary UD.

2. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 1, characterized in that, S1 includes: S11 identifies fault start signals: using overcurrent start, protection action, fault indication or switch trip signals sent from each terminal of the feeder, it determines the feeder to which the fault belongs, the fault judgment time window and the source of the initial signal; S12 analyzes the current operating mode: Based on the feeder topology, power supply point, switch on / off status, interconnection operating mode, and branch relationship, determine the current power supply direction and actual energized path; S13 Screening of Automated Terminals: Select automated terminals from all terminals that have fault detection capabilities and can provide protection actions, fault indications, telemetry or SOE information to the master station. The automated terminals include automated switches, feeder terminal units (FTUs), distribution terminal units (DTUs), protection devices or fault indication terminals. Ordinary non-automated switches are not included in this sequence. S14 Forming an automated terminal sequence: The selected automated terminals are sorted along the power supply direction to form the automated terminal sequence.

3. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 2, characterized in that, S1 further includes: S15 marking each terminal according to the signal status of the terminals in the automated terminal sequence, specifically as follows: Mark the terminals corresponding to overcurrent initiation, protection actions, and fault indications that have been received and passed validity verification as "fault characteristics confirmed"; Terminals that are confirmed to have no protection action, no fault indication, and no fault current characteristics through active calling or cache query will be marked as "confirmed to have no fault characteristics"; Terminals that have not yet completed verification are marked as "pending confirmation"; Terminals that are unreachable and whose calls have failed are marked as "unreachable".

4. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 3, characterized in that, The specific methods for forming the strongly convergent fault boundary UD in S2 include: S21: Scan each terminal from the power supply side to the load side along the automated terminal sequence; S22: If the current terminal is marked as "fault characteristics confirmed", then this terminal is the current downstream fault characteristic terminal U. Continue scanning downstream to find the downstream adjacent verifiable terminal D' of the current downstream fault characteristic terminal U, and determine: If the downstream adjacent terminal D' has been marked as "fault characteristics confirmed", then update the downstream adjacent terminal D' to the new current downstream fault characteristic terminal U, and continue scanning downstream; If the downstream adjacent verifiable terminal D' has been marked as "confirmed to be fault-free", then the downstream adjacent verifiable terminal D' is determined as the downstream adjacent fault-free terminal D, and U and D form a strong convergent fault boundary UD; If the downstream adjacent terminal D' can be determined to be marked as "pending confirmation", then D' is a blocked node, and node-level waiting or active querying is performed on the blocked node; If the downstream neighboring terminal D' has been marked as "unreachable", or if a valid response cannot be obtained after waiting or actively querying at the node level, then the waiting for that node will stop. Based on the received terminal signals, the most recent valid remote signaling status, and the topology connection relationship, the candidate fault segment for degradation will be inferred and output. At the same time, the candidate fault segment for degradation will be marked with the "unconfirmed" attribute and the corresponding confidence level. It will not be regarded as a strong convergence conclusion, and the strong convergence determination process will be exited, and the process will be transferred to the secondary state verification stage.

5. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 4, characterized in that, Based on the communication characteristics of the blocked node, the node-level waiting threshold and the active query trigger time are determined. The communication characteristics are parameter information describing the terminal's communication reachability, communication quality, and data retrieval response success rate.

6. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 5, characterized in that, The communication characteristics of the blocked nodes are divided into four levels: normal communication state, communication jitter state, communication abnormal state, and offline state.

7. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 6, characterized in that, Based on the communication characteristic level of the blocked node, execute the corresponding acquisition control strategy: When the communication characteristics of the blocked node are in a normal communication state, the remote signaling and protection signal active call is initiated first. If the call is successful and there are no fault characteristics, the collection waiting for the node ends. When the communication characteristics of the blocked node are in a communication jitter state, the acquisition waiting time of the node is extended, and an active call is triggered simultaneously to wait for the late signal; When the communication characteristics of the blocked node are in an abnormal communication state or an offline state, the waiting for the node is terminated, and the process of outputting the downgraded candidate fault section is initiated.

8. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 1, characterized in that, In S3, the criteria for verifying late signals are as follows: If the terminal location and fault characteristic state corresponding to the late signal match the strong convergence fault boundary UD, then the late signal is recorded as a supplement and the reliability of the current fault segment is improved. If the terminal location or fault characteristic state corresponding to the late signal conflicts with the strong convergence fault boundary UD, the fault section re-judgment is triggered and the review procedure is started.

9. The centralized feeder fault dynamic collection method based on topology and communication characteristics according to claim 8, characterized in that, The verification process includes signal timescale verification, topology verification, and communication verification.

Images