Centralized station terminal multi-granularity exception handling method based on sliding window
By adopting a multi-granularity anomaly handling method based on sliding windows, the problem of equipment malfunction when multiple modules of a centralized station terminal are in coordination with anomalies is solved, realizing comprehensive monitoring and hierarchical response, and improving the operational stability and robustness of the terminal.
Patent Information
- Application Number
- CN202510872937.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-27
- Publication Date
- 2025-10-17
AI Technical Summary
When a centralized station terminal encounters anomalies during operation due to multi-module coordination problems, it may cause the protection equipment to malfunction and issue false alarms, resulting in line tripping and power outages, affecting the normal power supply order of the power system.
A multi-granularity exception handling method based on sliding windows is adopted. By establishing a multi-granularity monitoring architecture, terminal security monitoring is divided into independent units, an abnormality objective function is constructed, and the total abnormality target value is calculated through the sliding window time. The abnormality trigger threshold is preset and a hierarchical processing response is implemented, including device restart, unit reset and micro-anomaly identification.
It enables comprehensive and detailed anomaly monitoring of centralized station terminals, improving their operational stability and reliability, ensuring normal operation even under abnormal conditions, and enhancing the robustness of safe operation.
Smart Images
Figure CN120803778A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of data monitoring, in particular to a centralized station terminal multi-granularity abnormality processing method based on a sliding window. BACKGROUND
[0002] Compared with a decentralized station terminal and a single feeder terminal device, a centralized station terminal has the advantages of more pipe control, centralized data management, low management and maintenance cost, and low data transmission cost, and is a new generation of microcomputer station terminal integrating functions such as telemetry, remote signaling, electric energy metering, and protection. It is suitable for multi-loop centralized monitoring application occasions such as distribution rooms, ring network cabinets, or switching stations, and cooperates with distribution substation and main station to realize functions such as operation state monitoring, fault identification, fault isolation, and non-fault area power supply recovery of distribution network automation.
[0003] However, the centralized station terminal faces the problem of multi-module cooperation (such as communication module, metering unit, data acquisition unit, etc.) in operation. When a fault occurs, the abnormality of any module will cause the device to not function properly, resulting in false operation of the protection device, false alarm, line trip, and power outage. This will interfere with the normal power supply order of the power system and cause unnecessary power outage trouble to the users on the power supply side. SUMMARY
[0004] To solve the above problems, the present application provides a centralized station terminal multi-granularity abnormality processing method based on a sliding window, comprising:
[0005] Establishing a multi-granularity monitoring architecture: according to the hardware architecture of the centralized station terminal, the terminal safety monitoring is divided into several independent units; the abnormal factors of each independent unit are determined, and the cumulative time of the abnormal factors is collected to trigger the monitoring of the corresponding independent unit;
[0006] Building an abnormality degree target function: according to the multi-granularity monitoring architecture, the monitoring abnormality of each independent unit is taken as a single granularity, and an abnormality degree target function of each granularity is established;
[0007] Calculating the total abnormality degree target value;
[0008] Hierarchical processing response: preset different abnormality trigger thresholds, calculate the total abnormality degree target value through the sliding window time, and trigger the hierarchical processing response of the centralized station terminal according to the total abnormality degree target value.
[0009] The formula of the abnormality degree target function of each granularity is:
[0010]
[0011] In the formula, S xa weight ratio representing a single granularity calculation abnormality target value; G x a weight ratio representing a single granularity S x abnormality target value; Kz represents an impact factor of an impact attribute of different granularity; T y represents a sliding window time; is a sparse matrix.
[0012] In a specific implementation of the first aspect, the abnormal trigger threshold includes M1, M2, M3, and M4, and M1>M2>M3>M4.
[0013] The hierarchical processing response of the centralized station terminal triggered according to the total abnormality target value specifically includes:
[0014] If the total abnormality target value is greater than the abnormal trigger threshold M1, the centralized station terminal performs device restart, and confirms whether to recover to normal operation;
[0015] If the abnormal trigger threshold M2 is less than the total abnormality target value and the total abnormality target value is less than the abnormal trigger threshold M1, the single independent unit is reset and restarted according to the abnormality target value of each granularity, and after the restart, it is reconfirmed whether the abnormality target value of the granularity in the current sliding window time recovers to normal;
[0016] If the abnormal trigger threshold M3 is less than the total abnormality target value and the total abnormality target value is less than the abnormal trigger threshold M2, the chip or module in the single independent unit is individually reset, and after the restart, it is reconfirmed whether the abnormality target value recovers to normal in the current sliding window time;
[0017] If the abnormal trigger threshold M4 is less than the total abnormality target value and the total abnormality target value is less than the abnormal trigger threshold M3, micro-abnormality identification is performed, and micro-abnormality identification elimination is performed according to the total abnormality target value in the next sliding window time;
[0018] If the total abnormality target value is less than the abnormal trigger threshold M4, the total abnormality target value of each sliding window time is obtained in real time, and threshold judgment is performed.
[0019] The hierarchical processing response of the centralized station terminal triggered according to the total abnormality target value further includes stopping triggering the hierarchical processing response based on an iteration stopping condition, and specifically includes:
[0020] If the centralized station terminal device restart or single independent unit reset restart cannot recover to normal, the sliding window time is adjusted, and the restart number accumulation is performed; when the restart number accumulation reaches a preset number, the operation is stopped, and an operation abnormality prompt is reported.
[0021] The six independent units include a metering and monitoring unit, a local and remote communication monitoring unit, a data acquisition monitoring unit, a data recording monitoring unit, a data encryption monitoring unit, and a time monitoring unit.
[0022] The abnormal factors of the data acquisition monitoring unit include abnormal data signals, abnormal feedback signals and / or abnormal operation of the analog-to-digital conversion chip.
[0023] The second aspect provides a centralized station terminal multi-granularity abnormality processing system based on a sliding window, which is used to implement the centralized station terminal multi-granularity abnormality processing method based on a sliding window.
[0024] The multi-granularity monitoring module is configured to divide terminal security monitoring into a plurality of independent units, determine abnormal factors of each independent unit, and collect cumulative time of the abnormal factors to trigger monitoring of the independent unit.
[0025] The target function module is configured to establish an abnormality degree target function of each granularity according to the multi-granularity monitoring architecture, and take the monitoring abnormality of each independent unit as a single granularity.
[0026] The target function module is further configured to calculate a total abnormality degree target value.
[0027] The trigger response module is configured to process responses in stages, preset different abnormal trigger thresholds, calculate the total abnormality degree target value through a sliding window time, and trigger a hierarchical processing response of the centralized station terminal according to the total abnormality degree target value.
[0028] The third aspect provides a centralized station terminal multi-granularity abnormality processing device based on a sliding window, which includes a processor and a memory.
[0029] The fourth aspect provides a computer readable storage medium for storing a computer program, wherein the computer program is executed by a processor to implement the centralized station terminal multi-granularity abnormality processing method based on a sliding window.
[0030] Beneficial effects: the application is a centralized station terminal multi-granularity abnormality processing method based on a sliding window, a multi-granularity monitoring architecture is designed, terminal security monitoring is divided into six independent units of metering monitoring unit, local and remote communication monitoring unit, data acquisition monitoring unit, data recording monitoring unit, data encryption monitoring unit and time monitoring unit, which comprehensively covers all aspects of possible abnormality of station terminal. Each unit determines abnormal factors respectively, realizing all-around and detailed abnormality monitoring of station terminal. Moreover, the multi-granularity monitoring architecture progressively calculates abnormality degree target value in the sliding window, according to the abnormality degree target value and abnormality trigger threshold, corresponding abnormality processing strategies are provided, different trigger response processing modes such as device restart, single independent unit reset and restart, chip or module individual reset, micro-abnormality identification, etc. are taken for different degrees of abnormality, which guarantees the centralized station terminal to run normally as much as possible under various abnormal conditions, improves its running stability and reliability, and enhances the robustness of the safe operation of the centralized station terminal. BRIEF DESCRIPTION OF DRAWINGS
[0031] The solutions and advantages of the present application will become clear to those ordinarily skilled in the art from a review of the detailed description of the preferred embodiments. The drawings are merely provided for the purpose of illustrating the preferred embodiments and are not considered as limiting the present application.
[0032] In the drawings:
[0033] Figure 1 Flow chart of the centralized station terminal multi-granularity abnormality processing method;
[0034] Figure 2 Structure diagram of the independent unit of the centralized station terminal. DETAILED DESCRIPTION
[0035] The exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.
[0036] EMBODIMENT
[0037] Reference Figure 1 The present embodiment provides a centralized station terminal multi-granularity abnormality processing method based on a sliding window, and the specific implementation steps are as follows:
[0038] Establish a multi-granularity monitoring architecture: according to the hardware architecture of the centralized station terminal, divide terminal security monitoring into several independent units; determine the abnormal factors of each independent unit respectively, and collect the cumulative time of the abnormal factors to trigger the monitoring of the corresponding independent unit.
[0039] The centralized station terminal comprises a bus plate, six-way interval plates, detection transformers, filter conditioning circuits, analog-to-digital conversion chips, the bus plate is integrally installed with a main processor, a metering module, a historical data storage module, a time synchronization module, a log saving module, a data encryption module and a communication module; each way interval plate is integrally provided with a coprocessor. Each way interval plate realizes the collection and calculation of voltage and current signals through a data bus, the bus plate collects data from the six-way interval plates through SPI communication, and stores, records and encrypts the collected data and metering data respectively.
[0040] In the embodiment, as shown in Figure 2 To realize the safe operation of the multi-module cooperative centralized station terminal, the terminal safety monitoring of the centralized station terminal is divided into six independent units, including a metering monitoring unit, a local and remote communication monitoring unit, a data collection monitoring unit, a data recording monitoring unit, a data encryption monitoring unit and a time monitoring unit.
[0041] The data collection monitoring unit monitors the data signals collected by each coprocessor and the data signals collected by the main processor, and the feedback signals output by the digital quantity of the centralized station terminal.
[0042] The data collection monitoring unit is used to determine whether there is an abnormality in the collected data signals and / or an abnormality in the feedback signals and / or an abnormality in the operation of the analog-to-digital conversion chip.
[0043] The data recording monitoring unit monitors the data recording, including the data recording stored by the historical data storage module and the data recording of the log saving module.
[0044] The data recording monitoring unit is used to determine whether there is an abnormality and / or an error in the data recording.
[0045] The data encryption monitoring unit monitors the data interaction of encrypted data, the data format of non-encrypted and encrypted data, and the parsed data of encrypted data.
[0046] The data encryption monitoring unit is used to determine whether there is an abnormality in the operation of the data encryption module and / or an abnormality in the encrypted data.
[0047] The metering monitoring unit monitors the voltage and current signals collected by the six-way coprocessors and detects the metering data obtained by the six-way coprocessors.
[0048] The metering monitoring unit is used to determine whether there is an abnormality in the metering data and / or an abnormality in the operation of the metering module.
[0049] The time monitoring unit monitors the time communication signal. The time monitoring unit is used to determine whether there is an abnormality in the time communication signal.
[0050] The local and remote communication monitoring units are used to determine whether there are communication data abnormalities and / or local communication module communication abnormalities and / or remote communication module communication abnormalities.
[0051] An abnormality degree target function is constructed: according to the multi-granularity monitoring architecture, the monitoring abnormalities of each independent unit are taken as a single granularity, and an abnormality degree target function of each granularity is established.
[0052] Because the abnormal factors of the monitoring abnormalities of the six independent units are different, the abnormality degree target functions of each granularity are constructed based on a multi-dimensional weight matrix, which not only considers the weight of a single factor but also calculates the mutual influence between factors.
[0053] The abnormality degree target function formula is:
[0054]
[0055] In the formula, S x represents the weight proportion of the abnormality degree target value of a single granularity, i.e., the weight proportion of the monitoring abnormalities of the metering monitoring unit, the local and remote communication monitoring unit, the data acquisition monitoring unit, the data recording monitoring unit, the data encryption monitoring unit, and the time monitoring unit;
[0056] G x represents the abnormality degree target value of a single granularity S x ; Kz represents the influence factor of the influence attribute of different granularities; T y represents the sliding window time. is a sparse matrix, and the numerical values of each sparse matrix are the influence factors of different influence attributes.
[0057] The total abnormality degree target value is calculated, including:
[0058] The modules and chips of the centralized station terminal are initialized, and the influence attributes of each granularity are determined.
[0059] Each module and chip of the centralized station terminal is initialized, the module and chip initialization is recorded, and the values are assigned to the influence attributes of each granularity. For example, the data acquisition monitoring granularity, the influence attribute F1 of this granularity is whether the chip initialization is normal; F2 is whether the data reading is overtime / error; F3 is whether the data detection is abnormal. Similarly, the influence data of each granularity is initialized and assigned.
[0060] According to the emergency degree of the monitoring abnormalities, the weight proportions of different granularities are set;
[0061] The abnormality degree target values of each granularity are added to obtain the total abnormality degree target value of the centralized station terminal.
[0062] The total abnormality degree target value formula is:
[0063]
[0064] In the formula, G is a total abnormality target value; Gx represents an abnormality target value of different granularity.
[0065] The hierarchical processing response comprises: presetting different abnormality trigger thresholds, calculating the total abnormality target value through a sliding window time, and triggering the hierarchical processing response of the centralized station terminal according to the total abnormality target value.
[0066] The abnormality trigger thresholds M1, M2, M3 and M4 are preset, and M1>M2>M3>M4. The total abnormality target value G is calculated through a sliding window time.
[0067] The total abnormality target value G is compared with each abnormality trigger threshold, and the hierarchical processing response of the centralized station terminal is triggered according to the total abnormality target value G, specifically as follows:
[0068] If the total abnormality target value G is greater than the abnormality trigger threshold M1, it indicates that the centralized station terminal cannot be safely operated, the device is restarted, and whether the operation is restored to normal is confirmed, and the number of device restarts is recorded.
[0069] If the abnormality trigger threshold M2 is less than the total abnormality target value G and the total abnormality target value G is less than the abnormality trigger threshold M1, the single independent unit is reset and restarted according to the abnormality target value of each granularity, and whether the abnormality target value of the granularity in the current sliding window time is restored to normal is confirmed again after the restart.
[0070] If the abnormality trigger threshold M3 is less than the total abnormality target value G and the total abnormality target value G is less than the abnormality trigger threshold M2, the chip or module in the single independent unit is individually reset, and whether the abnormality target value in the current sliding window time is restored to normal is confirmed again after the restart.
[0071] If the abnormality trigger threshold M4 is less than the total abnormality target value G and the total abnormality target value G is less than the abnormality trigger threshold M3, it indicates that the safe operation of the centralized station terminal is not affected, a micro-abnormality identification is performed, and the micro-abnormality identification is eliminated according to the total abnormality target value in the next sliding window time.
[0072] If the total abnormality target value G is less than the abnormality trigger threshold M4, it indicates that the centralized station terminal is normally operated, and the total abnormality target value of each sliding window time is acquired in real time, and threshold judgment is performed.
[0073] In addition, the hierarchical processing response triggered by the centralized station terminal according to the total abnormality target value further comprises stopping triggering the hierarchical processing response based on an iteration stop condition, specifically as follows:
[0074] If the centralized station terminal device cannot recover to normal after restarting or resetting, the sliding window time is adjusted, and the number of restarts is accumulated; when the number of restarts accumulates to a preset number, the operation is stopped, and an operation exception prompt is reported.
[0075] In addition, a centralized station terminal multi-granularity abnormality processing system based on a sliding window is also provided, which is used to implement the centralized station terminal multi-granularity abnormality processing method based on a sliding window as described above, and includes:
[0076] A multi-granularity monitoring module is configured to divide terminal security monitoring into a plurality of independent units, determine abnormal factors of each independent unit, and collect the cumulative time of the abnormal factors to trigger monitoring of the abnormal units.
[0077] A target function module is configured to establish an abnormality degree target function of each granularity according to the multi-granularity monitoring architecture, and take the monitoring abnormality of each independent unit as a single granularity.
[0078] The target function module is further configured to calculate a total abnormality degree target value.
[0079] A trigger response module is configured to process responses in stages, preset different abnormal trigger thresholds, calculate the total abnormality degree target value through the sliding window time, and trigger the hierarchical processing response of the centralized station terminal according to the total abnormality degree target value.
[0080] Then, a centralized station terminal multi-granularity abnormality processing device based on a sliding window is also provided, which includes a processor and a memory, wherein the processor implements the centralized station terminal multi-granularity abnormality processing method based on a sliding window as described above when executing the computer program saved in the memory.
[0081] Finally, a computer readable storage medium is provided for storing a computer program, wherein the computer program is executed by a processor to implement the centralized station terminal multi-granularity abnormality processing method based on a sliding window as described above.
Claims
1. A method for handling multi-granularity exceptions at centralized stations and terminals based on sliding windows, characterized in that: include: Establish a multi-granularity monitoring architecture: Based on the hardware architecture of the centralized station terminal, terminal security monitoring is divided into several independent units. The abnormal factors of each independent unit are determined separately, and the accumulated time of the abnormal factors is collected to trigger the monitoring abnormality of the corresponding independent unit. Constructing anomaly objective functions: Based on the multi-granularity monitoring architecture, the monitoring anomalies of each independent unit are treated as a single granularity, and an anomaly objective function is established for each granularity. Calculate the total abnormality target value; Hierarchical processing response: Different abnormality trigger thresholds are preset, the total abnormality target value is calculated through the sliding window time, and the hierarchical processing response of the centralized station terminal is triggered according to the total abnormality target value.
2. The centralized station terminal multi-granularity exception handling method according to claim 1 is characterized in that: The formula of the abnormality objective function of each granularity is: Where S x Indicates the weight ratio of the abnormality target value calculated at a single granularity; G x Represents a single granularity S x The abnormality target value; Kz represents the influencing factors of the influencing attributes of different granularities; T y Indicates the sliding window time; is a sparse matrix.
3. The centralized station terminal multi-granularity exception handling method according to claim 1, characterized in that: The abnormal triggering thresholds include M1, M2, M3, and M4, and M1>M2>M3>M4.
4. The centralized station terminal multi-granularity exception handling method according to claim 3 is characterized in that: The triggering of the hierarchical processing response of the centralized station terminal according to the total abnormality target value is specifically as follows: If the total abnormality target value is greater than the abnormality trigger threshold M1, the centralized station terminal will restart the equipment and confirm whether it has recovered to normal operation; If the abnormality trigger threshold M2 is less than the total abnormality target value and less than the abnormality trigger threshold M1, reset and restart the individual independent units according to the abnormality target value of each granularity, and reconfirm whether the abnormality target value of the granularity returns to normal within the current sliding window time after restart; If the abnormality trigger threshold M3 is less than the total abnormality target value and less than the abnormality trigger threshold M2, the chip or module in the single independent unit is reset separately, and after restarting, it is reconfirmed whether the abnormality target value within the current sliding window time returns to normal; If the abnormality trigger threshold M4 is less than the total abnormality target value and less than the abnormality trigger threshold M3, a minor abnormality mark is performed and the minor abnormality mark is eliminated according to the total abnormality target value within the next sliding window time; If the total abnormality target value is less than the abnormality trigger threshold M4, the total abnormality target value of each sliding window time is obtained in real time to perform threshold judgment.
5. The centralized station terminal multi-granularity exception handling method according to claim 1, characterized in that: The triggering of the hierarchical processing response of the centralized station terminal according to the total abnormality target value further includes stopping the triggering of the hierarchical processing response based on an iteration stop condition, specifically: If the centralized station terminal equipment cannot return to normal after restart or a single independent unit is reset and restarted, the sliding window time will be adjusted and the number of restarts will be accumulated; when the number of restarts accumulates to the preset number, the operation will be stopped and an abnormal operation prompt will be reported.
6. The centralized station terminal multi-granularity exception handling method according to claim 1, characterized in that: The six independent units include a metering monitoring unit, a local and remote communication monitoring unit, a data acquisition monitoring unit, a data recording monitoring unit, a data encryption monitoring unit and a time monitoring unit.
7. The centralized station terminal multi-granularity exception handling method according to claim 6, characterized in that: The abnormal factors of the data acquisition and monitoring unit include abnormal collected data signals and / or abnormal feedback signals and / or abnormal operation of the analog-to-digital conversion chip.
8. A sliding window-based centralized station terminal multi-granularity exception handling system, used to implement the sliding window-based centralized station terminal multi-granularity exception handling method according to claim 1, characterized in that: include: Multi-granularity monitoring module, used to divide terminal security monitoring into several independent units; Determine the abnormal factors of each independent unit respectively, and collect the accumulated time of the abnormal factors to trigger the monitoring abnormality of the corresponding independent unit; The objective function module is used to treat the monitoring anomalies of each independent unit as a single granularity based on the multi-granularity monitoring architecture and establish the abnormality objective function of each granularity; The objective function module is also used to calculate the total abnormality target value; The trigger response module is used for hierarchical processing responses, presets different abnormal trigger thresholds, calculates the total abnormality target value through the sliding window time, and triggers the hierarchical processing response of the centralized station terminal according to the total abnormality target value.
9. A centralized station terminal multi-granularity exception processing device based on sliding window, characterized in that: The method comprises a processor and a memory, wherein when the processor executes the computer program stored in the memory, the method for handling multi-granularity exceptions of a centralized station terminal based on a sliding window according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that Used to store a computer program, wherein when the computer program is executed by a processor, the sliding window-based centralized station terminal multi-granularity exception processing method according to any one of claims 1 to 7 is implemented.