A cloud-edge collaborative architecture fault recording master station system and its operation method

Abstract

The present invention discloses a fault recording master station system and operation method of a cloud-edge collaborative architecture. The cloud-edge collaborative system includes: an edge computing module and a recorder communication service module arranged at the recorder; and a recording master station communication service module, a cloud computing module, a cloud-edge collaborative module, an information publishing module, a clock synchronization module and a system interface module arranged at the fault recording master station. The method changes from the previous "centralized management" to the "cloud-edge collaborative" mode. The present invention can reduce the occupation of non-fault files on the recording master station data processing and storage resources, and improve the fault analysis efficiency of the recording master station system.

Description

Technical Field

[0001] The present invention relates to the field of relay protection technology, and in particular to a fault recording master station system and operation method of a cloud-edge collaborative architecture. Background Art

[0002] At present, the traditional fault recording master station mainly adopts the centralized processing mode to process fault recording files:

[0003] ① When the disturbance of the power grid exceeds the start threshold set in the recorder setting, the recorder is triggered to start recording and generate a recording file;

[0004] ② The fault recording master station periodically polls the recorder to see if there is a new recording file;

[0005] ③ The fault recording master station finds the new recording file and sends a new file retrieval instruction to the recorder;

[0006] ④ The fault recording master station performs fault analysis on the new recording file and generates a fault report for the recording file containing the fault;

[0007] ⑤ Notify the fault recording station operation and maintenance personnel through system alarms and text messages;

[0008] ⑥ The operation and maintenance personnel of the fault recording master station download the fault recording file and fault report on the workstation and send them to the relay protection professionals via email;

[0009] ⑦ Relay protection professionals manually check the fault report;

[0010] ⑧ The relay protection professionals inform the dispatcher of the fault report content by telephone;

[0011] ⑨ The dispatcher integrates data from other systems to make accident handling decisions.

[0012] With the overall acceleration of smart grid construction, the rapid development of new energy and the in-depth promotion of power market reform, the number of power grid equipment and the scale of data are also increasing. The new types of monitoring data derived from intelligent substations are also increasing. When dealing with this situation, the traditional fault recording master station has reduced data processing capabilities due to system architecture defects, which affects the efficiency of power grid fault monitoring and analysis functions. In addition, there are the following problems: ① Indiscriminate reception of data: With the continuous expansion of the scale of the power grid, the massive data generated in the monitoring process of relay protection equipment has exploded. There is no effective filtering mechanism for the current recording data. All data with a value density of less than 2% are sent to the master station system. The data surge not only has a great impact on the communication efficiency and transmission stability of the master station system, but also instantly impacts the storage, computing and other resources of the master station system, causing the system to Excessive resource load causes unpredictable problems such as freezes and even crashes, making it difficult to ensure the continued stability of the system; ② Centralized data processing: As the information support point for dispatching accident handling, the recording master station uniformly pre-processes, extracts features, analyzes faults, and releases information for the received raw recording data. All operations are performed in the master station system, and the master station's response speed to fault data processing slows down, affecting the timeliness and accuracy of fault analysis and diagnosis; ③ Manual intervention is required: Since fault handling involves operation and maintenance personnel, relay protection personnel, and dispatching personnel, the manual review and transmission of fault information is not only labor-intensive and time-consuming, but also difficult to ensure the correctness of information transmission. It is difficult to resolve the contradiction between the increasing expansion of the power grid and the shortage of human resources, and it is easy to cause large economic losses due to delayed fault decision-making time and long accident power outages. Summary of the invention

[0013] The present invention discloses a fault recording master station system and operation method of a cloud-edge collaborative architecture, which transforms the previous "centralized management" into a "cloud-edge collaboration" mode, can reduce the occupation of non-fault files on the data processing and storage resources of the recording master station, and improve the fault analysis efficiency of the fault recording master station.

[0014] To achieve the above object, the technical solution of the present invention is:

[0015] A fault recording master station system with cloud-edge collaborative architecture, comprising:

[0016] The edge computing module and the communication service module of the recorder are arranged in the recorder; and the communication service module of the recording master station, the cloud computing module, the cloud-edge collaboration module, the information publishing module, the clock synchronization module and the system interface module are arranged in the fault recording master station; wherein:

[0017] An edge computing module is used to perform fault analysis and identification on-site on the recording file, and generate a new recording file list, wherein the new recording file list includes the recording file and a feature identifier, and the feature identifier distinguishes the recording file into a fault, a suspected fault, and a non-fault;

[0018] The wave recorder communication service module is used to transmit the new wave recording file list to the wave recording master station communication service module;

[0019] The recording master station communication service module is used to receive a new recording file list and selectively send a call instruction to the recorder communication service module to call the recording files with feature identification as faults and suspected faults;

[0020] The cloud computing module is used to perform fault analysis on the fault recording files in the new recording file list, perform double-end distance measurement analysis on the fault line after the data correlation verification of the opposite-side recording files retrieved by the cloud-edge collaboration module is successful, and perform comprehensive analysis of relay protection based on the fault information of the recording master station, security master station, and traveling wave master station matched after the data correlation verification of the cloud-edge collaboration module is successful;

[0021] The cloud-edge collaboration module is used to uniformly manage the edge computing module and schedule data processing tasks, and perform homologous data requests, data correlation verification, and data matching based on the information reflected in the recording files determined as faults by the cloud computing module;

[0022] Information publishing module, used to publish fault models and fault reports;

[0023] Clock synchronization module, used to achieve time synchronization of all modules in the fault recording master station;

[0024] The system interface module is responsible for the protocol customization service for communicating with the target system interface.

[0025] Furthermore, the new recording file list includes serial number, bureau ID, station ID, device ID, file name, start time, size and feature identifier, and the feature identifier distinguishes the recording files into faults, suspected faults and non-faults.

[0026] Furthermore, the cloud-edge collaboration module includes an edge computing module management submodule, a data processing task scheduling submodule, a homologous data request submodule, a data relevance verification submodule and a data matching submodule; wherein:

[0027] The edge computing module management submodule is used to manage the edge computing module ledger, operation status monitoring, and algorithm version in a unified manner, and provide reminder services including module operation abnormalities and algorithm version upgrades;

[0028] The data processing task scheduling submodule is used to perform message parsing, data conversion, and submit the recording files to the cloud computing module in a certain order for fault analysis;

[0029] The same-source data request submodule is used to request the recording file of the fault confirmed by the cloud computing module to be retrieved from the recording device on the opposite side of the fault line; and to request the protection action report from the security signal master station and the traveling wave ranging report from the traveling wave master station;

[0030] Data correlation verification submodule: used for data correlation verification, including correlation verification of double-end recording files, correlation verification of protection action reports, and correlation verification of traveling wave ranging reports;

[0031] Data matching submodule: parses the data in the recording files, protection action reports and traveling wave ranging reports that have passed the homologous data correlation verification, and writes them into the database for use by the cloud computing module.

[0032] Furthermore, the cloud computing module includes a power grid fault analysis submodule, a double-terminal ranging analysis submodule, and a relay protection comprehensive analysis submodule, wherein:

[0033] The power grid fault analysis submodule is used to perform fault analysis on the recorded files that are retrieved in a targeted manner and are marked as faults or suspected faults;

[0034] The dual-end distance measurement analysis submodule is used to perform fault distance measurement based on the dual-end recording file of the fault line after the data correlation verification of the opposite-side recording file retrieved by the cloud-edge collaboration module. If the recording file of one end is missing, single-end fault distance measurement is performed.

[0035] The relay protection comprehensive analysis submodule is used to perform comprehensive analysis of relay protection based on the fault information of the recording master station, the security master station, and the traveling wave master station matched after the data correlation verification of the cloud-edge collaborative module is successful. The comprehensive analysis of relay protection includes fault judgment, protection action timing analysis, and fault location.

[0036] Furthermore, the cloud computing module further includes an OCS fault model generation submodule and an OMS fault report generation submodule.

[0037] in:

[0038] The OCS fault model generation submodule is used to obtain the power grid model from the OCS system through the system interface module, generate the OCS fault model according to the fault analysis result and send it to the OCS system through the information release module;

[0039] The OMS fault report generation submodule is used to generate a fault report of the OMS system specification and send it to the OMS system through the information publishing module.

[0040] Furthermore, the system interface module is used to connect with the OCS system, the traveling wave master station system, the security master station system, the OMS system, the WEB platform and the SMS platform.

[0041] The operation method of the above system comprises the following steps:

[0042] S1. When the disturbance of the power grid exceeds the start threshold set in the recorder setting, the recorder is triggered to start recording and generate a recording file;

[0043] S2. The edge computing module performs fault analysis and identification on the new recording file and generates a new recording file list;

[0044] S3. When the recording file identified as a fault and a suspected fault is found, the edge computing module initiates an active uploading instruction to the recorder communication service module, and sends the new recording file list to the fault recording master station. The recording master station communication service module receives the new recording file list and performs directional retrieval based on the characteristics in the list that identify the fault and the suspected fault recording file;

[0045] S4. The cloud computing module performs grid fault analysis on the recording file retrieved in step S3. If it is confirmed to be a fault recording file, the fault information is written into the database and a fault report is generated;

[0046] S5. The cloud-edge collaboration module makes a homologous data request and searches for the recording file in the recorder on the opposite side of the fault line according to the fault information in the database. If it is found that the recorder on the opposite side has a fault recording file at the same time of the fault, the cloud computing module obtains the recording file of the recorder on the opposite side for dual-end ranging analysis, writes the dual-end ranging results into the database, and generates a new fault report;

[0047] S6. The cloud-edge collaboration module initiates a homologous data request to the protection signal master station and the traveling wave master station according to the fault time, obtains the protection action report and the traveling wave ranging report of the same fault time or the fault time with a similar time series, performs data correlation verification, parses the data that passes the data correlation verification, matches the fault information model, and writes it into the database for use by the cloud computing module;

[0048] S7. The cloud computing module combines the protection action report and the traveling wave ranging report to conduct a comprehensive analysis of the relay protection, and forms an automatically verified OCS fault model and OMS fault report;

[0049] S8. The information publishing module publishes the generated OCS fault model and OMS fault report.

[0050] Furthermore, in step S5, the homologous data request includes the following steps:

[0051] S5.1 Search the fault recording master station for the recording file generated by the recorder on the opposite side of the fault line at the time of the fault. If the recording file is not available in the fault recording master station, retrieve it from the recorder on the opposite side of the fault line through the communication service module.

[0052] S5.2 requests a protection action report from the protection signal master station and a traveling wave ranging report from the traveling wave master station through the system interface module.

[0053] S5.3 If no recording file, protection action report or traveling wave ranging report with the same fault time is found, the request time will be relaxed to ±15 seconds of the fault time.

[0054] Furthermore, in step S6, the data relevance verification includes the following steps:

[0055] S6.1 Double-end recording file correlation check is divided into two steps: the first step is fault time check. The recording files with the same fault time generated by the recorders at both ends of the line are initially judged to be a set of double-end recording files, and the second step of check is carried out; the second step is waveform shape check. The curve shape change rate of the current channel of the recording file within ±15 seconds before and after the fault time is compared. If the consistency exceeds 95%, it is judged to be a set of double-end recording files;

[0056] S6.2 Protection action report correlation check is divided into two steps: the first step is fault time check. The protection action report with the same fault time as the fault recording file is initially judged to be a group of homologous data and the second step of check is carried out; the second step is protection action timing check. When the recorder has clock deviation, the protection action timing in the recording file and the protection action report within ±15 seconds is compared, ignoring the initial time, and only the sequence and time interval of various protection actions and circuit breaker opening and closing states are compared. If the comparison result meets the protection action information timing consistency recorded in the recording file of 100%, it is judged to be a group of homologous data;

[0057] S6.3 Traveling wave ranging report correlation check is divided into two steps: the first step is fault time check, the principle is the same as S4.2; the second step is ranging range check: compare the ranging results in the wave recording file and the traveling wave ranging report within ±15 seconds. If the comparison result satisfies: (wave recording ranging result - traveling wave ranging result) ÷ total line length × 100% ≤ ±5%, it can be determined to be a set of homologous data;

[0058] S6.4 When any data correlation check fails in S6.1, S6.2, or S6.3, the homologous data is discarded.

[0059] The fault recording master station system and method using the cloud-edge collaborative architecture described above transforms the fault handling work from the previous "centralized processing" to the "cloud-edge collaborative" mode, which can achieve the following functions:

[0060] 1. Data transmission on demand: The fault recording master station does not need to regularly poll the recorders in the entire network. The edge computing module can actively notify the master station to retrieve the specified recording file when there is a fault or suspected fault, avoiding the problem of massive data instantly impacting the storage, computing and other resources of the master station system, which is conducive to ensuring the continuous stability of the system.

[0061] 2. Distributed management: The fault recording master station only needs to manage the fault recording files that account for about 2% of the recording files, and the other 98% of non-fault files are still managed by the recorder itself, reducing the occupation of non-fault files on the data processing and storage resources of the recording master station, and improving the fault analysis efficiency of the fault recording master station.

[0062] 3. Automatic report submission: The fault recording master station system can automatically complete the horizontal verification of fault information, and automatically submit fault reports through the OCS system and OMS system, so that accident handling decision support can directly complete the vertical extension from relay protection to dispatching.

[0063] Through the combination of on-demand data transmission, distributed management and automatic reporting of fault reports, the system and method of the present invention can greatly improve the timeliness and accuracy of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a structural schematic diagram of the system of the present invention.

[0065] Figure 2 It is a structural diagram of the cloud-edge collaboration module.

[0066] Figure 3 It is a structural diagram of the cloud computing module. DETAILED DESCRIPTION

[0067] The present invention is further described below in conjunction with specific embodiments, but the protection scope of the present invention is not limited to the following embodiments.

[0068] A fault recording master station system with cloud-edge collaborative architecture, such as Figure 1 As shown, including:

[0069] The edge computing module and the communication service module of the recorder are arranged in the recorder; and the communication service module of the recording master station, the cloud computing module, the cloud-edge collaboration module, the information publishing module, the clock synchronization module and the system interface module are arranged in the fault recording master station; wherein:

[0070] The edge computing module is installed in the recorder management machine and is used to perform fault analysis and identification on-site for the recording files and generate a new recording file list. The new recording file list includes recording files and feature identifiers. The feature identifiers distinguish the recording files into faults, suspected faults, and non-faults. The fault recording files are actively sent to the fault recording master station communication service module through the recorder communication service module.

[0071] The recorder communication service module is installed in the recorder management machine and is used to transmit the new recording file list to the recording master station communication service module.

[0072] The recording master station communication service module is a service program installed in the fault recording master station communication server, which is used to receive a new recording file list and selectively send a retrieval instruction to the recorder communication service module to retrieve the recording files with characteristics identified as faults and suspected faults; compare with the recording file list received last time, and when new recording files identified as faults or suspected faults are found, send a retrieval instruction to the recorder communication service program to retrieve the above recording files to the master station side; recording files identified as non-faults are not automatically retrieved.

[0073] The cloud computing module is an algorithm service program installed in the fault recording master station analysis server, which is used to perform fault analysis on the fault recording files in the new recording file list, perform double-end ranging analysis on the fault line after the data correlation verification of the opposite side recording files retrieved by the cloud-edge collaboration module is successful, and perform comprehensive analysis of relay protection based on the fault information of the recording master station, security master station, and traveling wave master station matched after the data correlation verification of the cloud-edge collaboration module is successful.

[0074] The cloud-edge collaborative module is a collaborative service program installed in the fault recording main station analysis server. It is used to uniformly manage the edge computing module and schedule data processing tasks. It also performs homologous data requests, data correlation verification, and data matching based on the information reflected in the recording files determined to be faulty by the cloud computing module.

[0075] The information publishing module is used to publish fault models and fault reports.

[0076] The clock synchronization module is a time synchronization service program installed in all servers such as the fault recording master station communication, analysis, and publishing, and is used to achieve time synchronization of all modules in the fault recording master station. It is connected to the time synchronization device timing interface of the dispatching room to achieve time synchronization with the entire network equipment.

[0077] The system interface module is a communication service program installed in the fault recording master station communication and publishing server, which is responsible for the protocol customization service of the communication with the target system interface.

[0078] The communication service module of the recording master station is connected to the cloud computing module and the cloud-edge collaboration module, the cloud-edge collaboration module is connected to the cloud computing module, the cloud computing module is connected to the information release module, the communication service module, the cloud computing module, the cloud-edge collaboration module and the information release module are connected to the clock synchronization module, and the information release module and the communication service module of the recording master station are connected to the system interface module.

[0079] The recording master station communication service module is connected to the recorder communication service module, and the recorder communication service module is connected to the edge computing module.

[0080] Preferably, the new recording file list includes serial number, bureau ID, station ID, device ID, file name, start time, size and identification, and the characteristic identification distinguishes the recording files into faults, suspected faults and non-faults.

[0081] The format of the recording file list is shown in Table 1:

[0082] Serial number Bureau ID Station ID Device ID file name Start time size Feature Identification 1 1 1 1 2021-07-07@14-38-59.#FD833 2021-07-07 14:38:59 1.2M F

[0083] In Table 1:

[0084] “Serial number” indicates the order of the recording file in the list;

[0085] "Device ID", "Station ID" and "Office ID" respectively represent the recorder number, station number and power supply bureau number of the recorder to which the recording file belongs. ID is a number, which is uniformly allocated by the fault recording master station when establishing the recorder account.

[0086] "File name" refers to the file name of the recording file stored in the computer. The file naming format varies depending on the recorder manufacturer.

[0087] "Start time" means the time when the recorder starts recording. In the fault recording file, it is generally equivalent to the "fault time".

[0088] “Size” indicates the file size of the recording file stored in the computer.

[0089] “Feature Identifier” indicates the fault condition of the recording file: “F” indicates fault; “A” and “T” both indicate suspected fault, where “A” indicates no fault is detected but contains switch position change information; “T” indicates test data; and “N” indicates non-fault.

[0090] Preferably, a cloud-edge collaboration module, such as Figure 2 As shown, it includes an edge computing module management submodule, a data processing task scheduling submodule, a homologous data request submodule, a data relevance verification submodule and a data matching submodule; wherein:

[0091] The edge computing module management submodule is used to uniformly manage the edge computing module ledger, operation status monitoring, and algorithm version, and provide reminder services including module operation abnormalities and algorithm version upgrades.

[0092] The data processing task scheduling submodule is used to perform message parsing, data conversion, and submit the recording files to the cloud computing module in a certain order for fault analysis; the data processing task scheduling is for the recording files sent by the recorders of the entire network. According to the recording file identifier, the priority of the data processing task is F>A>T.

[0093] The same source data request submodule is used to identify the recording file of the fault in the cloud computing module, and request the recording file generated at the time of the fault from the recorder on the opposite side of the fault line; and request the protection action report from the security master station and the traveling wave ranging report from the traveling wave master station; execute the following requests:

[0094] a. Search the fault recording master station for the recording file generated by the recorder on the opposite side of the fault line at the time of the fault. If the fault recording master station does not have the file, retrieve it from the recorder on the opposite side of the fault line through the communication service module.

[0095] b. Request protection action report from the security master station and request traveling wave ranging report from the traveling wave master station through the system interface module.

[0096] c. If no recording file, protection action report or traveling wave ranging report at the same fault time is found, the request time will be relaxed to ±15 seconds of the fault time.

[0097] Data correlation verification submodule: used to perform correlation verification of homologous data, including correlation verification of double-end recording files, correlation verification of protection action reports, and correlation verification of traveling wave ranging reports;

[0098] Data matching submodule: parses the recorded file data that has passed the homology data verification and writes it into the database for use by the cloud computing module.

[0099] The edge computing module management submodule is connected to the data processing task scheduling submodule, the data processing task scheduling submodule is connected to the homologous data request submodule, the homologous data request submodule is connected to the data relevance verification submodule, and the data relevance verification submodule is connected to the data matching submodule.

[0100] Preferably, the cloud computing module, such as Figure 3As shown, it includes a power grid fault analysis submodule, a two-terminal ranging analysis submodule, and a relay protection comprehensive analysis submodule, among which: the power grid fault analysis submodule is used to perform fault analysis on the recording files that are directionally retrieved and marked as faults and suspected faults; determine the fault line, phase, time, type and other information; verify and evaluate the fault analysis results of the edge computing module; and provide a basis for the same-source data request and data correlation verification functions of the cloud-edge collaboration module.

[0101] The dual-end ranging analysis submodule is used to perform fault ranging on the fault line based on the dual-end recording file after the data correlation verification of the opposite-side recording file retrieved by the cloud-edge collaboration module is successful. If the recording file on one end is missing, single-end fault ranging is performed. This provides a basis for the data correlation verification function of the cloud-edge collaboration module.

[0102] The relay protection comprehensive analysis submodule is used to perform comprehensive analysis on relay protection based on the fault information of the recording master station, the security master station, and the traveling wave master station matched after the successful correlation verification of the data in the cloud-edge collaborative module, replacing relay protection professionals to complete comprehensive fault analysis such as fault determination, fault location, and protection action timing analysis.

[0103] More preferably, the cloud computing module also includes an OCS fault model generation submodule and an OMS fault report generation submodule, wherein: the OCS fault model generation submodule is used to obtain the power grid model from the OCS system through the system interface module, generate the OCS fault model according to the fault analysis results, and send it to the OCS system through the information release module; the fault alarm and its topological relationship can be intuitively reflected on the dispatching screen. The OMS fault report generation submodule is used to generate a fault report of the OMS system specification, including the fault condition, fault distance, protection action condition and correctness evaluation, as well as the data upload status of the recording master station, the security master station and the traveling wave master station, and send it to the OMS system through the information release module. The fault details and equipment operation evaluation can be reported to the dispatch through the electronic interface to help the dispatching personnel make decisions directly.

[0104] The power grid fault analysis submodule is connected to the double-terminal ranging analysis submodule, the double-terminal ranging analysis submodule is connected to the relay protection comprehensive analysis submodule, and the relay protection comprehensive analysis submodule is respectively connected to the OCS fault model generation submodule and the OMS fault report generation submodule.

[0105] Preferably, the system interface module is used to connect with the OCS system, the traveling wave master station system, the security master station system, the OMS system, the WEB platform and the SMS platform.

[0106] The operation method of the system described above includes the following steps: S1. When the disturbance of the power grid exceeds the starting threshold set in the recorder setting, the recorder is triggered to start recording and generate a recording file; S2. The edge computing module performs fault analysis and identification on the new recording file to generate a new recording file list; S3. When a recording file identified as a fault or suspected fault is found, the edge computing module initiates an active upload instruction to the recorder communication service module, and sends the new recording file list to the fault recording master station. The recording master station communication service module receives the new recording file list and performs targeted retrieval based on the characteristics in the list. The recording file is identified as a fault or suspected fault; S4. The cloud computing module performs power grid fault analysis on the recording file retrieved in step S3. If it is confirmed to be a fault recording file, the fault information is written into the database and a fault report is generated; S5. The cloud-edge collaborative module performs a homologous data request, searches for the recording file in the recorder on the opposite side of the fault line according to the fault information in the database, and if it is found that the recorder on the opposite side has a fault recording file at the same time of the fault, the cloud computing module obtains the fault file. Take the recording file of the opposite side recorder for double-end ranging analysis, write the double-end ranging results into the database, and generate a new fault report; S6. The cloud-edge collaboration module initiates a homologous data request to the Baoxin master station and the traveling wave master station according to the fault time, obtains the protection action report and the traveling wave ranging report of the same fault time or the fault time with a similar time series, performs data correlation verification, parses the data that passes the data correlation verification, matches the fault information model, and writes it into the database for use by the cloud computing module; S7. The cloud computing module combines the protection action report and the traveling wave ranging report to perform a comprehensive analysis of the relay protection, and forms an automatically verified OCS fault model and OMS fault report; S8. The information release module publishes the generated OCS fault model and OMS fault report. In addition, fault reports can also be sent to the WEB platform and SMS platform.

[0107] Among them, the fault analysis of the edge computing module adopts the existing technology, specifically: In 2018, China Southern Power Grid promoted an intelligent wave recorder consisting of a collection unit and a management unit, which integrated the functions of fault wave record analysis, network record analysis, secondary system visualization, and intelligent operation and maintenance, and can realize the on-site fault feature extraction and preliminary diagnosis of the wave record data, and divide the fault into F, A, T and N types, corresponding to faults, suspected faults and non-faults. Please refer to the technical specifications of the intelligent wave recorder proposed by China Southern Power Grid.

[0108] The cloud computing module in step S4 performs power grid fault analysis, specifically: the cloud computing module uses the fault waveform analysis software to lock the fault range of the received recording data according to the data processing priority F>A>T, and extracts information such as the fault line, fault type and fault time. The extracted fault information is used to retrieve the line side recording, protection action information and traveling wave ranging information for the cloud-edge collaborative module to provide a basis for the correlation verification of the same source data.

[0109] For the comprehensive analysis of relay protection in step S6, the cloud computing module conducts a comprehensive analysis based on the wave recording fault diagnosis results, protection action report and traveling wave ranging results in step S5 to confirm whether the fault line judgment is correct, whether the protection action is implemented correctly, whether the ranging results are reliable, etc. The previous method of relying on single information is upgraded to a comprehensive diagnosis method combining wave recording, protection and traveling wave information, and directly outputs the results of the comprehensive analysis of the three-station information. It no longer relies on manual comprehensive fault judgment, thus reducing human intervention.

[0110] Preferably, in step S5, the homologous data request includes the following steps:

[0111] S5.1 Search the fault recording master station for the recording file generated by the recorder on the opposite side of the fault line at the time of the fault. If the recording file is not available in the fault recording master station, retrieve it from the recorder on the opposite side of the fault line through the communication service module.

[0112] S5.2 requests a protection action report from the protection signal master station and a traveling wave ranging report from the traveling wave master station through the system interface module.

[0113] S5.3 If no recording file, protection action report or traveling wave ranging report with the same fault time is found, the request time will be relaxed to ±15 seconds of the fault time.

[0114] Preferably, in step S6, the homologous data correlation verification includes the following steps:

[0115] S6.1 Double-end recording file correlation check is divided into two steps: the first step is fault time check. The recording files with the same fault time generated by the recorders at both ends of the line are initially judged to be a set of double-end recording files, and the second step of check is carried out; the second step is waveform shape check. The recorder may cause slight clock deviation due to uneven component quality. At this time, the curve shape change rate of the current channel of the recording file within ±15 seconds at one cycle before and after the fault time is compared. If the consistency exceeds 95%, it is judged to be a set of double-end recording files;

[0116] S6.2 Protection action report correlation check is divided into two steps: the first step is fault time check. The protection action report with the same fault time as the fault recording file is preliminarily judged to be a group of homologous data and the second step of check is carried out; the second step is protection action timing check. The protection device is protected by multiple clock synchronization measures. The probability of clock deviation is much lower than that of the recorder. When the recorder has clock deviation, the protection action timing in the recording file and the protection action report within ±15 seconds is compared. The initial time is not counted. Only the sequence and time interval of various protection actions and circuit breaker opening and closing states are compared. If the comparison result meets the protection action information timing consistency of 100% recorded in the recording file, it is judged to be a group of homologous data.

[0117] S6.3 Traveling wave ranging report correlation verification is divided into two steps: the first step is the fault time verification, the principle is the same as S4.2; the second step is the ranging range verification: both the wave recorder and the traveling wave device may have clock deviations. At this time, the ranging results in the wave recording file and the traveling wave ranging report within ±15 seconds are compared. Traveling wave ranging has the following characteristics: when the fault time is selected correctly, the traveling wave ranging error is slightly smaller than the recording ranging error; when the fault time is selected incorrectly, the traveling wave ranging error is significantly larger than the recording ranging error. Therefore, if the comparison result satisfies: (recording ranging result - traveling wave ranging result) ÷ total line length × 100% ≤ ±5%, it can be determined to be a set of homologous data;

[0118] S6.4 When any data correlation check fails in S6.1, S6.2, or S6.3, the homologous data is discarded.

[0119] The purpose of the correlation check of the same-source data is to confirm that the requested opposite-side recording data, protection data and ranging result data are data monitored and collected from the same fault occurrence in terms of time, shape, ranging range, etc., and the data is reflected in different ways from the recording, protection and traveling waves.

[0120] When S6.1 verification is successful, the wave recording files at both ends of the same line are found; when S6.2 verification is successful, the protection action report of the same fault is found; when S6.3 verification is successful, the traveling wave ranging report of the same fault is found. The more related files found based on the same fault, the greater the amount of fault information, which is more helpful for the comprehensive analysis of the cloud computing module, and the more complete the content of the fault report output in the end.

[0121] The specific implementation cases are:

[0122] by Figure 2 For example, a 220kV substation is generally equipped with two 220kV recorders, two main transformer recorders, and one 110kV recorder. When a 220kV line fails, the accident handling process is as follows:

[0123] ① A 220kV line fault caused a large increase in the current of the 220kV line, which caused a small increase in the current of the adjacent 220kV line; the fault caused the 220kV bus voltage to drop, causing the 110kV side bus voltage to drop. In summary, the 5 recorders in the substation started recording at the same time, generating at least 5 new recording files.

[0124] ②The edge computing module of the 220kV#1 recorder performs fault analysis on the new recording file, marks the fault recording file as "F", and uploads the new recording file list to the fault recording master station communication server through the recorder communication service program; the edge computing modules of the other four recorders perform fault analysis on the new recording file, mark the non-fault recording file as "N", and do not initiate active upload.

[0125] ③ The fault recording master station communication service module compares the new recording file list with the previous recording file list, and retrieves a new fault recording file marked as "F" to the master station side. Almost at the same time, the recording master station communication service module receives the new recording file list of the opposite side recorder, and retrieves a new fault recording file marked as "F" on the opposite side to the master station side.

[0126] ④ The cloud computing module of the fault recording master station performs power grid fault analysis on the new fault recording file, confirms that it is a fault recording file, writes the fault information into the database and generates a fault report.

[0127] ⑤ The cloud-edge collaboration module of the fault recording master station searches for the opposite recording file based on the fault information in the database. At this time, it is found that the opposite recording device has a fault recording file at the same fault time, and has completed the power grid fault analysis and entered the request queue for searching the opposite recording file. The cloud-edge collaboration module directly matches the two fault recording files at the same fault time in the queue. The purpose of this step is to retrieve the opposite recording file through the cloud-edge collaboration module to perform a double-end recording file correlation check. Only after the verification is successful can it be used for double-end ranging to ensure the accuracy of double-end ranging.

[0128] The cloud computing module of the fault recording master station performs dual-end distance measurement analysis, writes the dual-end distance measurement results into the database, and generates a new fault report.

[0129] ⑥ The cloud-edge collaborative module of the fault recording master station initiates a homologous data request to the protection signal master station and the traveling wave master station, and receives the protection action report and traveling wave ranging report at the same fault time.

[0130] Obtain protection action reports and traveling wave ranging reports at the same fault time or at fault times close to the time series, perform homologous data verification, parse the data that passes the data correlation verification, match the fault information model, and write it into the database for use by the cloud computing module;

[0131] ⑦ The cloud computing module of the fault recording master station performs comprehensive analysis of relay protection to form an automatically verified OCS fault model and OMS fault report.

[0132] ⑧The information publishing module publishes the generated OCS fault model and OMS fault report.

Claims

1. A fault recording master station system with cloud-edge collaborative architecture, characterized in that include: An edge computing module and a communication service module of the wave recorder are arranged in the wave recorder; And a recording master station communication service module, a cloud computing module, a cloud-edge collaboration module, an information publishing module, a clock synchronization module and a system interface module are arranged in the fault recording master station; wherein: An edge computing module is used to perform fault analysis and identification on-site on the recording file, and generate a new recording file list, wherein the new recording file list includes the recording file and a feature identifier, and the feature identifier distinguishes the recording file into a fault, a suspected fault, and a non-fault; The wave recorder communication service module is used to transmit the new wave recording file list to the wave recording master station communication service module; The recording master station communication service module is used to receive a new recording file list and selectively send a call instruction to the recorder communication service module to call the recording files with feature identification as faults and suspected faults; The cloud computing module is used to perform fault analysis on the fault recording files in the new recording file list, perform double-end distance measurement analysis on the fault line after the data correlation verification of the opposite-side recording files retrieved by the cloud-edge collaboration module is successful, and perform comprehensive analysis of relay protection based on the fault information of the recording master station, security master station, and traveling wave master station matched after the data correlation verification of the cloud-edge collaboration module is successful; The cloud-edge collaboration module is used to uniformly manage the edge computing module and schedule data processing tasks, and perform homologous data requests, data correlation verification, and data matching based on the information reflected in the recording files determined as faults by the cloud computing module; Information publishing module, used to publish fault models and fault reports; Clock synchronization module, used to achieve time synchronization of all modules in the fault recording master station; The system interface module is responsible for the protocol customization service of communicating with the target system interface; The cloud-edge collaboration module includes an edge computing module management submodule, a data processing task scheduling submodule, a homologous data request submodule, a data relevance verification submodule and a data matching submodule; wherein: The edge computing module management submodule is used to manage the edge computing module ledger, operation status monitoring, and algorithm version in a unified manner, and provide reminder services including module operation abnormalities and algorithm version upgrades; The data processing task scheduling submodule is used to perform message parsing, data conversion, and submit the recording files to the cloud computing module in a certain order for fault analysis; The same-source data request submodule is used to request the recording file generated at the time of the fault to be retrieved from the recording device on the opposite side of the fault line after the cloud computing module confirms it as the fault; and to request the protection action report from the security signal master station and the traveling wave ranging report from the traveling wave master station; Data correlation verification submodule: used for data correlation verification, including correlation verification of double-end recording files, correlation verification of protection action reports, and correlation verification of traveling wave ranging reports; Data matching submodule: parses the data in the recording files, protection action reports and traveling wave ranging reports that have passed the homologous data correlation verification, and writes them into the database for use by the cloud computing module; The cloud computing module includes a power grid fault analysis submodule, a double-terminal ranging analysis submodule, and a relay protection comprehensive analysis submodule, wherein: The power grid fault analysis submodule is used to perform fault analysis on the recorded files that are retrieved in a targeted manner and are marked as faults or suspected faults; The dual-end distance measurement analysis submodule is used to perform fault distance measurement based on the dual-end recording file of the fault line after the data correlation verification of the opposite-side recording file retrieved by the cloud-edge collaboration module. If the recording file of one end is missing, single-end fault distance measurement is performed. The relay protection comprehensive analysis submodule is used to perform comprehensive analysis of relay protection based on the fault information of the recording master station, the security master station, and the traveling wave master station matched after the data correlation verification of the cloud-edge collaboration module is successful. The comprehensive analysis of relay protection includes fault judgment, protection action timing analysis, and fault location; The cloud computing module further includes an OCS fault model generation submodule and an OMS fault report generation submodule, wherein: The OCS fault model generation submodule is used to obtain the power grid model from the OCS system through the system interface module, generate the OCS fault model according to the fault analysis result and send it to the OCS system through the information release module; The OMS fault report generation submodule is used to generate a fault report of the OMS system specification and send it to the OMS system through the information publishing module.

2. The system according to claim 1, characterized in that: The new recording file list includes serial number, bureau ID, station ID, device ID, file name, start time, size and feature identifier. The feature identifier distinguishes the recording files into faults, suspected faults and non-faults.

3. The system according to claim 1, characterized in that: The system interface module is used to connect with the OCS system, the traveling wave master station system, the security master station system, the OMS system, the WEB platform and the SMS platform.

4. The method of operating the system as claimed in claim 1, characterized in that The following steps are involved: S1. When the disturbance of the power grid exceeds the start threshold set in the recorder setting, the recorder is triggered to start recording and generate a recording file; S2. The edge computing module performs fault analysis and identification on the new recording file and generates a new recording file list; S3. When a recording file marked as a fault or suspected fault is found, the edge computing module initiates an active upload instruction to the recorder communication service module, and sends the new recording file list to the fault recording master station. The recording master station communication service module receives the new recording file list and performs directional retrieval based on the characteristics in the list that identify the recording file as a fault or suspected fault; S4. The cloud computing module performs grid fault analysis on the recording file retrieved in step S3. If it is confirmed to be a fault recording file, the fault information is written into the database and a fault report is generated; S5. The cloud-edge collaboration module makes a homologous data request and searches for the recording file in the recorder on the opposite side of the faulty line according to the fault information in the database. If it is found that the recorder on the opposite side has a fault recording file at the same time of the fault, the cloud computing module obtains the recording file of the recorder on the opposite side for dual-end ranging analysis, writes the dual-end ranging results into the database, and generates a new fault report; S6. The cloud-edge collaboration module initiates a homologous data request to the protection signal master station and the traveling wave master station according to the fault time, obtains the protection action report and the traveling wave ranging report of the same fault time or the fault time with a similar time series, performs data correlation verification, parses the data that passes the data correlation verification, matches the fault information model, and writes it into the database for use by the cloud computing module; S7. The cloud computing module combines the protection action report and the traveling wave ranging report to conduct a comprehensive analysis of the relay protection, and forms an automatically verified OCS fault model and OMS fault report; S8. The information publishing module publishes the generated OCS fault model and OMS fault report; In step S5, the homologous data request includes the following steps: S5.1 searching the fault recording master station for the recording file generated by the recorder on the opposite side of the fault line at the time of the fault. If the recording file is not available in the fault recording master station, the recording file is retrieved from the recorder on the opposite side of the fault line through the communication service module; S5.2 requests a protection action report from the security master station and a traveling wave ranging report from the traveling wave master station through the system interface module; S5.3 If no recording file, protection action report or traveling wave ranging report with the same fault time is found, the request time will be relaxed to ±15 seconds of the fault time; In step S6, the data relevance verification includes the following steps: S6.1 Double-end recording file correlation check is divided into two steps: the first step is fault time check. The recording files with the same fault time generated by the recorders at both ends of the line are initially judged to be a set of double-end recording files, and the second step of check is carried out; the second step is waveform shape check. The curve shape change rate of the current channel of the recording file within ±15 seconds before and after the fault time is compared. If the consistency exceeds 95%, it is judged to be a set of double-end recording files; S6.2 Protection action report correlation check is divided into two steps: the first step is fault time check. The protection action report with the same fault time as the fault recording file is initially judged to be a group of homologous data and the second step of check is carried out; the second step is protection action timing check. When the recorder has clock deviation, the protection action timing in the recording file and the protection action report within ±15 seconds is compared, ignoring the initial time, and only the sequence and time interval of various protection actions and circuit breaker opening and closing states are compared. If the comparison result meets the protection action information timing consistency recorded in the recording file of 100%, it is judged to be a group of homologous data; S6.3 Traveling wave ranging report correlation check is divided into two steps: the first step is fault time check, the principle is the same as S4.2; the second step is ranging range check: compare the ranging results in the wave recording file and the traveling wave ranging report within ±15 seconds. If the comparison result satisfies: (wave recording ranging result - traveling wave ranging result) ÷ total line length × 100% ≤ ±5%, it can be determined to be a set of homologous data; S6.4 When any data correlation check fails in S6.1, S6.2, or S6.3, the homologous data is discarded.

Images