CN108037415A - Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data - Google Patents
Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data Download PDFInfo
- Publication number
- CN108037415A CN108037415A CN201711350231.5A CN201711350231A CN108037415A CN 108037415 A CN108037415 A CN 108037415A CN 201711350231 A CN201711350231 A CN 201711350231A CN 108037415 A CN108037415 A CN 108037415A
- Authority
- CN
- China
- Prior art keywords
- data
- information
- event
- scada
- rule
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data, pre- collection rule is formulated firstly for distribution network failure event, data mining is carried out, then the multi-source heterogeneous information of power distribution network is pre-processed and merged, multi-source data is finally based on and carries out Fault Diagnosis of Distribution Network.Present invention fusion production management system, breakdown repair and SCADA data, establish SCADA system and collect rule, extraction and the relevant information of failure in advance with TCM system events, be stored in big data platform；The relevance of multi-source heterogeneous data is excavated, diagnoses the information and attribute of event in distribution network systems, and combines auxiliary information verification diagnostic result, technical support is provided for the high-quality of power distribution network, safety and stablization operation.
Description
Technical field
The invention belongs to technical field of electric power, is related to the excavation and diagnosis of distribution network failure information, and multi-source is based on for one kind
The distribution network failure information excavating and diagnostic method of isomeric data.
Background technology
With the development of national economy and the raising of living standards of the people, power consumer is to power supply reliability and power quality
Requirement it is higher and higher.But influenced by adverse circumstances and uncertain human factor, power distribution network especially overhead transmission line occurs
The probability of failure is very high, and in the process of running, excitation surge current, overload, lightning stroke, stealing are when various abnormal operating conditions
There is generation[1].How power distribution network responds actively accident, not only need to be grasped the local distribution of covering it is accurate, it is reliable, comprehensively,
Timely status information, will also be directed to the analysis and diagnosis that these multi-source heterogeneous complex datas carry out specialty[2,3]。
Distribution Network Failure information excavating is horizontal in order to lift distribution safety operation level and good service with diagnosis, realizes peace
Full risk prevention and control.Obtainable base data type is various in power distribution network, production management system (Production
Management System, PMS) data, breakdown repair (Trouble Call Management, TCM) data, SCADA numbers
According to etc., such as fault type, failure distribution, failure are associated with load relation, failure season, trend, switch state etc.
Statistics and operation data.In addition, distribution construction retrofit data, overhaul of the equipments data or even the post-installation review data of distribution, distribution
Tripping data also can be as power distribution network running state analysis and the important sources excavated.Since Distributing network structure is complicated, distribution basis
Data not only wide variety, but also data volume is big；By taking OPEN3000 systems as an example, the tables of data of its Database Systems has hundreds of
As many as, data type complexity, structure disunity, storage is scattered, relevance is poor.Mass data is provides well with Analysis of Isolated Net Running
Data basis, while also considerably increase the difficulty of data processing and information excavating, it is therefore desirable to reference to data with existing, establish and close
The storage organization of reason, excavates multi-source heterogeneous data intension, extracted valid data information.
Traditional fault diagnosis technology be to circuit-breaker status change information, protective device warning message and action message and
The special of the electrical quantity such as voltage, electric current claims to be analyzed, and possible event is judged according to protection act logical AND operations staff experience
Hinder position and fault type[1-2].As the increase of data collecting system, data type, data scale constantly increase in power distribution network,
Fault diagnosis index is developed into from traditional fault location and failure sizing to be sentenced comprising fault zone and element positioning, wrong components
The composite target of the contents such as fixed and failure reconfiguration, traditional fault diagnosis technology are difficult to be applicable in[2-3].With expert system, manually god
Become main method under the new situation through network, petri net, rough set theory for the Intelligent Diagnosis Technology of representative.Expert system
System includes several parts such as knowledge base, inference machine, dynamic data base, man-machine interface, explanation module and knowledge base management system, it is led
Failure is diagnosed according to existing experts database logical relation based on switch and the information for protecting equipment[4-6].Artificial neuron
Network can find optimal hypersurface in higher dimensional space and carry out simulation input, defeated by specific learning method training sample set
The functional relation gone out.Its most typical diagnostic model is the state conduct input of switch and protection, with the failure thing that may occur
Part is as output, and construction is according to the global BP algorithm approached or the radial basis function neural network of partial approximation[7-9].Petri network
Network states the relation of input quantity and output quantity with figure, and then according to system and corresponding rule, one-to-one corresponding is established with it
Relation, by judging whether to meet rule condition, draw corresponding conclusion[10-13].Petri net is due to its good calculating
Ability and intuitively graphic representation capability, are widely used in electric network fault field[14-17].Rough set theory is effectively to locate
Imperfection and the new mathematical tool of uncertain problem are managed, effective mould can be extracted from data using special algorithm
Formula, excavates potential rule and tacit knowledge from mass data[18-20]。
Above-mentioned intelligent diagnostics are widely used in Fault Diagnosis of Distribution Network, but are still come with some shortcomings.Specially
Although family's system can obtain the diagnostic result for meeting human thinking's custom, with the increase of power grid scale and complexity, switch
It is significantly increased with protection quantity, action logic is more complicated, and it is extremely difficult to establish complete knowledge base[21].Neutral net
The a large amount of representational samples of study are needed before application, and usually complete sufficient sample is difficult to obtain, this is just
The commonly used of neutral net forms difficulty, and the convergence rate of learning algorithm is usually slower[22].Petri net is carrying out
When large-scale distribution network models, it can face and multiple shot array be in the presence of, and model strong and fault-tolerant to the dependence of network structure
Property is poor[23].Rough set theory can only be used for handling discrete data, and sliding-model control must be carried out first for continuous data, this
During it is possible that errored message or loss[24].In conclusion complete experts database, typical sample set and rationally
Data structure be Intelligent Diagnosis Technology application key.This merges proposition with multi-system information to distribution network failure Feature Selection
The requirement of higher.
Bibliography
[1] Guo innovates, and height is revitalized, Liu Yi, waits electric network failure diagnosis method [J] high of using layering Multi-source Information Fusion
Voltage Technique, 2010 (12):2976-2983.
[2] Fault Diagnosis Method for Distribution Networks research [D] Northeastern University of the Jing using multi-source information is opened, 2013.
[3] Song Kai, Liu Runhua, Kang Zhong are good for Fault Diagnosis of Distribution Network technology [J] electricity of the based on blended data method for digging
Power science and technology journal, 2010,25 (2):68-72.
[4] the electric network failure diagnosis systematic research [D] that Duan Qiaojia rough set theories are combined with expert system, Beijing：
North China electricity is university, 2003
[5]Minakawa T,Ichikawa Y,Kunugi M,et al.Development and
implementation of a power system fault diagnosis expert system[J].IEEE
Transactions on Power Systems,1995,10(2):932-940.
[6]Styvaktakis E,Bollen M H J,Gu I Y H.Expert system for
classification and analysis of power system events[J].IEEE Transactions on
Power Delivery,2002,17(2):423-428.
[7] Bi Tianshu, Ni Yixin, Wu Fuli, wait power grids of the based on radial basis function neural network and Fuzzy control system
New Fault Diagnosis Method [J] Proceedings of the CSEEs, 2005,25 (14):12-18.
[8] application [J] the relays of Liu Yan swallows based on the information fusion of fuzzy neural network in electric network failure diagnosis,
2005,33(9):9-11.
[9] Bi Tianshu, Ni Yixin, Wu Fuli, wait electric network failure diagnosis method [J] China of the based on new neural network
Electrical engineering journal, 2002,22 (2):73-78.
[10] behavioral theory of prosperous person of outstanding talent .Petri nets and its application [M] Higher Education Publishing House, 2003.
[11] research [D] of electric network failure diagnosis methods of the refined of high body based on failure information system and Petri network theory
Shandong University, 2009.
[12] Gao Zhanjun, Chen Qing, Wang Tao, wait electric network failure diagnosis model [J] electricity of the based on relay protection Time And Space Parameters
Force system automates, 2012 (2012 13):61-66+91.
[13] electric network failure diagnosis methods of the Shi Lu based on Petri network theory and technology [D] Shandong University, 2010.
[14] Zeng Qingfeng, He Zhengyou, Yang Jian tie up power system failure diagnostic scale-model investigations of the based on colored Petri network
[J] electric power system protection and controls, 2010 (14):5-11.
[15] Xie Hongtao, child's dawn electric network fault error comprehensive diagnosis method [J] the power grid skills of sun based on hierarchical fuzzy Petri network
Art, 2012 (2012 01):246-252.
[16] Pan Chao, Yue Jianping, Liu Bing, wait electric network failure diagnosis method [J] the power grid skills of based on adaptive Petri net
Art, 2008 (1):46-50.
[17]Chen S M.Weighted fuzzy reasoning using weighted fuzzy Petri nets
[J].IEEE Transactions on Knowledge and Data Engineering,2002,14(2):386-397.
[18] Sun Qiuye, Zhang Huaguang, wear Jing based on the distribution system on-line fault diagnosis for improving Rough Set Reduction algorithm
[J] Proceedings of the CSEEs, 2007,27 (7):58-64.
[19] trembling, Zhang Lieyong, Gu Xueping, wait to use the distributed power grid failure of rough set union rule mining algorithm
Diagnose [J] Proceedings of the CSEEs, 2010 (4):28-34.
[20] Chao is into Liu Wenying, Liu Yongzhi, wait grid alarm rules of the based on rough set theory to automatically extract and apply
[J] electric power system protection and controls, 2011,39 (8):95-99.
[21] application [D] of the high fine jade complex event processing techniques in Fault Diagnosis of Distribution Network, Beijing：North China electricity is big
Learn, 2016
[22] Fault Diagnosis Method for Distribution Networks research [D] Wuhan University Of Technologies of the Li Yao based on FUZZY H NETS, 2014.
[23] fault diagnosis method of electric power system and application [D] Zhejiang University of the Xu Bing using multi-source information, 2017.
[24] research of Zhao Yuan rocs Fault Diagnosis Method for Distribution Networks and realization [D] Xi'an Petroleum Universities, 2013.
The content of the invention
The technical problem to be solved in the present invention is：Distribution magnanimity operation data provide good base for its operating analysis
How plinth, excavate multi-source heterogeneous data, troubleshooting, is the key for lifting distribution safety operation level and good service level.
The technical scheme is that：Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data, bag
Include with lower part：
1) distribution network failure event formulates pre- collection rule and data mining, and the pre- rule that collects refers to " breakdown judge set in advance
Regular collection ", data are excavated according to rule：
1.1) the direct events of SCADA collect rule in advance：The direct events of SCADA refer to the accident point from distribution automation system crawl
Lock event, distribution automation system timing acquisition breaker type are circuit, main transformer, mother/segmentation/bypass and unknown open circuit
Device accident separating brake records, and relevant auxiliary data is captured respectively for separating brake record, and auxiliary data divides position, secondary distant including remote signalling
Believe alarm, remote signalling SOE, telemetry and remote-control data, breakdown judge rule is set according to auxiliary data；
1.2) the indirect events of SCADA collect rule in advance：The indirect events of SCADA are distribution automation systems based on remote signalling point position
The remote signalling of index divides position information, and distribution automation system timing acquisition breaker type is circuit, main transformer, mother/segmentation/bypass
Recorded with unknown breaker remote signalling point position, according to following Rules Filtering, the remote signalling point position after screening is recorded and is taken over a job as between
Part is preserved：
Regular 1.2.1)：Remote signalling divides before and after the time of origin of position is not present Automation System of Power Network accident separating brake in setting time
Record；
Regular 1.2.2)：Remote signalling divides secondary remote signalling protection or remote signalling the SOE data in setting time before and after the time of origin of position
In have the record for including protection act keyword；
Regular 1.2.3)：In front and rear 30 minutes, there is non-zero status in remote measurement current value, i.e. current value is more than or equal to 5 amperes
Situation；
All remote signalling displacement information for meeting above-mentioned regular automated system is used as indirect event, and can be inquired about,
Conjugated for the remote signalling being screened in indirect event, system captures its corresponding auxiliary information, root from SCADA system at the same time
According to auxiliary data, breakdown judge rule is set；
1.3) TCM events collect matching association in advance；TCM events refer to from production management system crawl failure report information,
And the information is matched and associated with SCADA event, form breakdown judge rule；
2) the multi-source heterogeneous information pre-processing of power distribution network is with merging：
2.1) pretreatment of the multi-source heterogeneous data of distribution, from business decision logic, information show and event handling is several
Aspect carries out specificity analysis, data cleansing and extraction to data, then using data source, purposes and data attribute as dimension,
Collect rule in advance with reference to the data in 1), sort merge is carried out to multi-source heterogeneous data, is put in order by rule in independent region；
After the completion of pretreatment, the association of cross-platform data will be carried out with merging with attribute according to the source of pre- collection data；
2.2) distribution multi-source heterogeneous data storage, the multi-source heterogeneous data of distribution it is preprocessed with merge after, transform into
Structural data, builds data model, and structural data batch is imported big data platform is stored；
3) Fault Diagnosis of Distribution Network based on multi-source data, is to accident separating brake in SCADA system and TCM systems first
Judgement is handled, and confirms to break down；Second step be excavated from SCADA system and TCM systems with the relevant protection act of failure,
Remote measurement unusual fluctuation, phenomena such as accident is total and associated processing outcomes, clear failure object, content and time, the 3rd step combine auxiliary
Informix validation fault, there are manual intervention is submitted at ambiguity.
Step 3) is specially：
The judgement processing of the first step and second step is followed successively by：
3.1) diagnostic rule 1:
If there is " capacitor " or " capacitance " or " reactance " or " capacitive reactance " in event content, event is non-faulting；It is right
Fault message carries out the judgement of diagnostic rule 2 or 3；
3.2) diagnostic rule 2：
If occurs " ground connection " or " 3U0 " or " 3V0 " or " overvoltage " or " overvoltage " and not in secondary remote signalling alarm record
Comprising " ground connection become ", " grounding switch ", " earthing switch ", " device alarm ", " resistance ", " switch ", " protection act mark puts action
Position is 1 ", then event is small current neutral grounding；
3.3) diagnostic rule 3：
If there is " tripping " or " coincidence " or " separating brake " or " action " in event content, event is short trouble；
If event corresponds to appearance " tripping " in trip condition, event is short trouble；
If reported for repairment, appearance " signal " and plant stand name include " station " in content or " switch " or implementor name occurs in implementor name
There is breaker, then event is short trouble；
3.4) judgement of diagnostic rule 4 is carried out according to diagnostic rule 2：
If SCADA busbar grounding event matches, event in 1min before and after the ID and event time of small current neutral grounding point
SCADA busbar grounding information is replicated as fault message；
If without above-mentioned matching, but when SCADA 2 is small it is interior have busbar grounding event, then 1min before and after the match event moment
Interior SCADA busbar groundings event simultaneously replicates ground connection information as fault message；
If above-mentioned be unsatisfactory for, secondary remote signalling alarm, remote signalling SOE, remote signalling point position, remote-control data are captured, then is excavated
Information, according to pre- collection rule judgment fault message；
3.5) judgement of diagnostic rule 5 is carried out according to diagnostic rule 3：
Failure is the direct events of SCADA, replicates the fault message of the direct event correlations of SCADA as a result；
Failure is TCM information, such as the direct event successes of TCM information matches association SCADA, then replicates the direct events of SCADA
Associated fault message is as a result；Match unsuccessful, if TCM information matches association SCADA indirect event successes, excavate
With data in plant stand and equipment 1min/8h, fault message is filled up, it is crawl remote signalling alarm, remote signalling SOE, distant if also unsuccessful
Letter divides position, remote control and telemetry, is excavated with reference to the pre- collection rule of the direct or indirect events of SCADA, failure judgement information；
The verification and manual intervention of 3rd step be specially：
3.6) auxiliary information verifies：
Power distribution network data source further includes power information acquisition system, regional Meteorological Information System in addition to production system, matches somebody with somebody
Grid topology data, geographical location information and distributed generation resource operational monitoring information, according to auxiliary judgment needs, to other information
Pretreatment is assessed and filtered to system data, establishes auxiliary correlation model, and auxiliary letter is provided for the fault diagnosis of production system
Breath；
3.7) manual intervention：
For the critical data deletion condition occurred during fault diagnosis, processing is compensated to the criterion lacked, according to
SOE signals are changed and accident resultant signal, obtain the displacement signal lacked originally, then carry out the logical process of next step.
In the case of current mass data, multiple systems, present invention analysis and research propose a kind of fault information mining
With diagnostic method, production management system, breakdown repair and SCADA data are merged first, establishes SCADA system and TCM systems
Event collects rule, extraction and the relevant information of failure in advance, is stored in big data hadoop platforms, establishes sending for data mining；
Next, excavating the relevance of multi-source heterogeneous data, the information and attribute of event in distribution network systems are diagnosed, and combines auxiliary information
Verify diagnostic result, technical support is provided for the high-quality of power distribution network, safety and stablization operation.
The advantage of the invention is that：
(1) power distribution network information, including SCADA system, TCM systems, PMS systems etc. are made full use of, it is independent to establish each system
Event of failure collect rule in advance, excavate the related information of different system failure, provide the foundation for the fusions of multi-source heterogeneous data；
(2) using different system event attribute as core, establish the matchings of power distribution network multisystem data and associate, establish compared with
For complete distribution event base and sample set, establish multi-source data for fault diagnosis and excavate basis；
(3) excavation and the diagnostic rule of power distribution network multi-source data are established, the fault message for coming from different channels is sent out
Raw time, defect content (description), fault object, breakdown judge auxiliary information (protection act, SOE, curent change etc.) etc.
Various dimensions are verified, are set it automatically when unambiguously as determinate fault, can be adjusted in allowed limits when there is ambiguity
The whole secondary judgement of parameter, such as still with the presence of ambiguity, then submit manual intervention, with reference to remote control, maintenance and scene investigate to failure into
Row assert or removes that the statistic analysis result after the inspection of this method has high accuracy, and extremely low failure is omitted
Probability.
Brief description of the drawings
Fig. 1 is distribution multi-source data of the present invention processing and the flow diagram of fault diagnosis.
Fig. 2 is TCM time matchs in the present invention with associating flow chart.
Fig. 3 is the multi-source heterogeneous data prediction of distribution in the present invention and Stored Procedure schematic diagram.
Fig. 4 is the storage of big data platform and the process flow schematic diagram of the multi-source heterogeneous data of the present invention.
Fig. 5 is the Fault Diagnosis of Distribution Network flow diagram of the invention based on multi-source heterogeneous data.
Embodiment
It is contemplated that the mass data of the different platform such as fusion scheduling, production, establishes SCADA system and TCM system things
Therefore pre- collection, screen distribution network failure feature and information.Analyze distribution network systems information attribute, establish the association of different system information with
Matching relationship, studies the storage method of multi-source heterogeneous information.Fault Diagnosis of Distribution Network based on multi-source heterogeneous data is first to thing
Part judged, excavates, clear failure object, content and attribute total etc. with the relevant protection act of failure, remote measurement, accident, and
Diagnostic result is verified with reference to auxiliary information.
The present invention has following characteristics：
(1) present invention establish SCADA system, the event of failure of TCM systems collect in advance rule, event include SCADA directly with
Indirect event, TCM events, by excavating the related information of different system failure, the fusion for multi-source heterogeneous data provides base
Plinth；
(2) present invention establishes power distribution network SCADA, TCM and PMS multisystem data using different system event attribute as core
Matching with associate, establish more complete distribution event base and sample set, for fault diagnosis establish multi-source data excavation
Basis；
(3) present invention establishes excavation and the diagnostic rule of power distribution network multi-source data, the fault message to coming from different channels
Carry out time of origin, defect content (description), fault object, breakdown judge auxiliary information (protection act, SOE, curent change
Deng) etc. carry out various dimensions excavation.
Lower mask body introduces the implementation of the present invention.
1 distribution network failure event collects rule and data mining in advance
1.1 multi-source fault message sources
Distribution multi-source data source includes the distribution line of all 10kV, 35kV, 110kV voltage class in PMS systems
Switch trip information, associated remote signalling displacement information, protection act information (secondary remote signalling), telemetry, and manual operation
Record, including straighforward operation, trouble hunting record, the accident separating brake information of automated system and remote signalling displacement information, SCADA letters
Breath etc., as shown in Figure 1.By taking OPEN3000 systems as an example, the tables of data of its Database Systems have it is as many as hundreds of, each tables of data it
Between logic index association it is extremely complex, original of key data and the auxiliary judgment information that Distribution Network Failure needs in OPEN3000 is
There is no related thread can use in system.Distribution multi-source data presentation type complexity, structure disunity, storage are scattered, relevance difference
, moreover, in system also there are many interference data, i.e. bad data in feature, this allow for data excavation it is more complicated with
Cumbersome, there is an urgent need to more efficient data processing method.
For distribution multi-source data is effectively classified, the degree of association and the degree of polymerization of fault message are improved, it is necessary to pin
Pre- collection rule is formulated system multi-source data, and the pre- rule that collects refers to " regular collection of breakdown judge set in advance ", the present invention point
Not Shai Xuan different pieces of information source event of failure information, for associated storage, excavation with diagnosis basis is provided.The pre- set analysis of event is simultaneously
Collect the accident separating brake information and remote signalling displacement information of automated system, and TCM failures make a report on information.These data are carried out
Auxiliary judgment, including displacement alarm, secondary remote signalling alarm, remote signalling SOE, remote measurement sampled data and straighforward operation record.Event is pre-
Collection function is made of three parts：The direct events of SCADA, the indirect events of SCADA and TCM events.These three parts have corresponded to three
The different types of data of kind, and corresponding data processing and exhibition method.Data grabber mode, processing decision logic also it is each not
It is identical.
The direct events of 1.2 SCADA collect rule in advance
The direct events of SCADA are primarily referred to as the accident separating brake event from distribution automation system crawl, automated system thing
Therefore separating brake processing background process timing crawl breaker type is circuit, main transformer, mother/segmentation/bypass and unknown breaker
Accident separating brake records.Capture relevant auxiliary data respectively for separating brake record and handled, auxiliary data including remote signalling divide position,
Secondary remote signalling alarm, remote signalling SOE, telemetry and remote-control data.
Line-breaker can be used for flag line maintenance, failure and overlap situation.For the thing that breaker type is circuit
Therefore separating brake, confirmation is not such as audited, is divided position and telemetry according to remote signalling first, is judged whether accident separating brake meets maintenance situation,
If meeting maintenance condition, set accident separating brake to be recorded as " suspected of maintenance ", handled for professional's judgement, judgment rule such as table
Shown in 1.If non-inspecting state, then the judgment rule as shown in table 1, excavates the mark that breaker overlaps situation.
1 line-breaker state mining rule of table
* remote measurement current value is determined as zero less than 5A
The indirect events of 1.3 SCADA collect rule in advance
The indirect events of SCADA are remote signalling point position information of the system using remote signalling point position as master index, and back-end data captures process
Timing acquisition breaker type is recorded for circuit, main transformer, mother/segmentation/bypass and unknown breaker remote signalling point position, according to
Lower Rules Filtering, records the remote signalling point position after screening as indirect event and is preserved.
Regular (1)：Remote signalling divides before and after the time of origin of position is not present Automation System of Power Network in setting time (default 20 seconds)
Accident separating brake records；
Regular (2)：Remote signalling divides secondary the remote signalling protection or remote signalling in setting time (default 20 seconds) before and after the time of origin of position
There is the record for including protection act keyword in SOE data；
Regular (3)：In front and rear 30 minutes, non-zero status occurs in remote measurement current value, and (current value is more than or equal to 5 amperes of feelings
Condition).
The remote signalling displacement information of all automated systems for meeting above-mentioned regular (1)-(3) is used as indirect event, and can be into
Row inquiry.Conjugated for the remote signalling being screened in indirect event, system captures its corresponding auxiliary from SCADA system at the same time
Information, rule are similar with direct event above.Auxiliary information can be showed with list, as long as choosing desired to take over a job when checking
Part.
1.4 TCM events collect matching association in advance
TCM events refer to report information from the failure of production management system crawl, and by the information and SCADA accident separating brakes
Information is matched with being associated, and TCM events and the matching of SCADA events are as shown in Figure 2 with associating flow.It is first direct in SCADA
Plant stand and the identical accident separating brake record of switch object are searched in logout, if it is found, then successful match, and establishes pass
Connection, otherwise by time range be expanded to 8 it is small when search；If successful match, both are associated and indicia matched result；
If it fails to match, matched with the indirect events of SCADA, if successful match, both are associated and marked
With as a result, otherwise factory can be searched in automated system by the plant stand title and breaker title in TCM accident separating brake data
Stand and record and breaker record, if it is possible to find record, then PMS accident separating brakes are inserted into event handling, and set
Put incidence relation.Do not matched in the case of remaining or not associated data are, it is necessary to which artificial judgment can be simultaneously handled at the event of being inserted into
In reason.
In view of TCM failures are made a report on, artificial fill substance is more, and station name, switch name, defect content and time etc. exist not
With the uncertainty of degree, need to carry out screening and filtering from the data of TCM crawls, real accident separating brake information found out,
System needs also to have carried out fuzzy matching processing to the defect content for making a report on information while station name, implementor name is checked, and improves
The integrality of TCM data.
The multi-source heterogeneous information pre-processing of 2 power distribution networks is with merging
The multi-source heterogeneous data prediction of 2.1 distributions
The automated system that the status information of equipment of distribution at present is adjusted with mostling come from, these information adhere to different lifes separately
Production department and production system, trans-departmental, cross-platform data distribution state are unsatisfactory for grasping the requirement with network operation data comprehensively.
Data are pre-processed first before data fusion with storage, as shown in Figure 3.The pretreatment of the multi-source heterogeneous data of distribution is
Refer to from business decision logic, information show and event handling etc. it is several from the aspect of to carry out specificity analysis, data to data clear
Wash and extract, be then based on the different dimensions such as data source, purposes, data attribute, consider data and collect rule in advance, to more
Source isomeric data carries out sort merge.
As shown in figure 3, the structure and attribute of core data trans-departmental to power distribution network first, multi-platform are analyzed, use
Data in SCADA, PMS especially TCM systems are modified, reject with supplementing by data cleansing with data filtering method, carry
The integrality of high multi-source heterogeneous data, is that basis is established in the extraction of data matrix and restructuring.Event based on foundation collects excavation in advance
Rule extracts different system initial data, and it is put in order in independent region by rule with associating matching process；The number that need to be extracted
According to including SCADA events and TCM events, SCADA events the inside further comprises in addition to direct accident separating brake is drawn by remote signalling separating brake
The indirect emergency stop valve trip event of hair.
After the completion of pretreatment, the association of cross-platform data will be carried out with merging with attribute according to the source of pre- collection data.
Due to pre- collection data since source is different, data attribute can be variant, and system has carried out this to compare analysis, retains general character category
Property, collect the attribute for refining differentiation, eventually form the complete or collected works of an attribute, including event content description, generation event, equipment
Object, device type and electric company, substation/switchyard etc., establish the distribution network failure data fusion according to property index.
After data fusion, system can auto-associating event information, such as remote signalling displacement, protection act, remote measurement change, provided for event
Comprehensively, the multidimensional data of structuring, fast reaction event generating process.
The multi-source heterogeneous data storage of 2.2 distributions
The multi-source heterogeneous data of distribution it is preprocessed with merge after, can transform into structural data, build data model,
Its basic structure is as shown in table 2.The foundation of data model be for data storage, retrieval, Distribution Network Failure diagnosis etc. service, in
The data model that heart database is set up includes status information, fault message and prototype data information, and status information, fault message are used
In state dynamic evaluation, prototype data information is used to dock with initial data, can receive data, also can be by having received storage
Data it is counter trace back to data source, this has played important function in practical applications.
2 Distribution Network Failure original data basic model of table
Huge in view of data volume, the technology of big data, which can provide, more accurately to be analyzed, improve operational efficiency and reduce into
This, and reduce business risk.In order to using the advantage of big data, store the multi-source heterogeneous number of distribution using big data platform herein
According to.For structural data, big data platform is imported using Sqoop batches, is handled in big data platform ETL technologies, number
It is as shown in Figure 4 with Stored Procedure according to handling.Data access is buffered region layer first, do slightly collect, Unified coding and cleaning
After be put into unified view area, being then aggregated into big wide table according to each business model falls in data warehouse layer, after final analysis calculates
Data be placed on Data Mart can also shift onto online storehouse for upper layer application call.
3 Fault Diagnosis of Distribution Network based on multi-source data
The essence of Fault Diagnosis of Distribution Network is to the logic judgment with Running State, its process is not just for single source
Failure and auxiliary information, and be to rely on the extraction and analysis of multi-source data.Multi-source data is recognized respectively, certification
Basis on, to come from different channels fault message carry out time of origin, defect content (description), fault object, failure
Judge the various dimensions verification of auxiliary information (protection act, SOE, curent change etc.) etc., it is true to set it automatically when unambiguously
Qualitative failure, when there is ambiguity can the secondary judgement of adjusting parameter in allowed limits, such as still with the presence of ambiguity, then submit
Manual intervention, investigates with reference to remote control, maintenance and scene and failure is assert or removed, the statistical analysis after the inspection of this method
As a result there is high accuracy, and extremely low failure omits probability.
Power distribution network carries out fault diagnosis using multi-source data and is divided into three steps, is to SCADA system and TCM systems first
The judgement processing of middle accident separating brake, confirms to break down；Second step main line is to be excavated from SCADA system and TCM systems and event
Hinder relevant protection act, remote measurement unusual fluctuation, phenomena such as accident is total and associated processing outcomes, clear failure object, content and when
Between, the 3rd step combination auxiliary information comprehensive verification failure, there are manual intervention is submitted at ambiguity.Fault Diagnosis of Distribution Network is according to institute
The information of acquisition, the decision logic according to design are handled to obtain the diagnosis situation of distribution, its overall procedure is as shown in Figure 5.
As shown in figure 5, it is made of based on the Fault Diagnosis of Distribution Network of multi-source data different diagnostic rules, these diagnosis rule
Then on the one hand consider SCADA and the data type and feature of TCM systems, on the other hand make full use of keyword in system.Examine
Disconnected rule is as follows：
(1) diagnostic rule 1:
There is " capacitor " or " capacitance " or " reactance " or " capacitive reactance " in if event contents
Then events are non-faulting
endif
(2) diagnostic rule 2：
There is " ground connection " or " 3U0 " or " 3V0 " or " overvoltage " or " overvoltage " in bis- remote signalling of if alarm record and do not wrap
Containing " ground connection become ", " grounding switch ", " earthing switch ", " device alarm ", " resistance ", " switch ", " protection act mark puts act bit
For 1 "
Then events are small current neutral grounding
endif
(3) diagnostic rule 3：
(4) diagnostic rule 4：
(5) diagnostic rule 5：
(6) auxiliary information verifies：
Power distribution network data source further includes power information acquisition system, regional Meteorological Information System in addition to production system, matches somebody with somebody
Grid topology data, geographical location information, distributed generation resource operational monitoring information etc., cannot meet to analyze in production system data
It is required that when, data source can be expanded.According to auxiliary judgment needs, other information systems data are assessed, are filtered etc. with pre- place
Reason, establishes auxiliary correlation model, auxiliary information is provided for the fault diagnosis of production system.
(7) manual intervention：
Due to the unstability of distribution network data, the missing of critical data usually occurs in fault diagnosis, causes event
Barrier analysis is obstructed, even stagnate.One kind typically conjugate signal deletion, this can be caused by many reasons, as substation,
The integrated system level of switchyard is uneven, dropout or is not uploaded at all in signals transmission, and signal capture occurs
One of the reason for exception etc., substation/switchyard is in during construction retrofit and such phenomenon occurs.Need for this pair
The criterion lacked compensates processing, and one of method is exactly that the conversion of SOE signals and accident are total, thus obtains the change lacked originally
Position signal, then carries out the logical process of next step.
In conclusion the present invention is for data volume is big, data type is complicated, structure disunity, storage is scattered, relevance is poor
Distribution mass data, it is proposed that one kind excavate and diagnostic method, solve problems with：
(1) using production management system, breakdown repair and SCADA data as object, analysis multi-source information forms and attribute,
The pre- collection rule of SCADA events is established, and matches relevant fault repairing event and SCADA data.
(2) based on the different dimensions such as data source, attribute, consider data and collect rule in advance, to multi-source heterogeneous data into
Row sort merge establishes complete sample set, and be stored in big data hadoop platforms with merging.
(3) the Fault Diagnosis of Distribution Network rule based on multi-source heterogeneous data is established, event is judged, excavation and failure
Relevant protection act, remote measurement, accident are total etc., clear failure object, content and attribute, and combine auxiliary information verification diagnosis knot
Fruit.
Claims (2)
1. distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data, it is characterized in that including with lower part：
1) distribution network failure event formulates pre- collection rule and data mining, and the pre- rule that collects refers to " the rule of breakdown judge set in advance
Then gather ", data are excavated according to rule：
1.1) the direct events of SCADA collect rule in advance：The direct events of SCADA refer to the accident separating brake thing from distribution automation system crawl
Part, distribution automation system timing acquisition breaker type are circuit, main transformer, mother/segmentation/bypass and unknown breaker thing
Therefore separating brake records, relevant auxiliary data is captured respectively for separating brake record, auxiliary data divides position, secondary remote signalling to accuse including remote signalling
Police, remote signalling SOE, telemetry and remote-control data, breakdown judge rule is set according to auxiliary data；
1.2) the indirect events of SCADA collect rule in advance：The indirect events of SCADA are distribution automation systems using remote signalling point position as master index
Remote signalling divide position information, distribution automation system timing acquisition breaker type is circuit, main transformer, mother/segmentation/bypass and not
The breaker remote signalling point position known records, according to following Rules Filtering, using the remote signalling after screening divide position record as indirect event into
Row preserves：
Regular 1.2.1)：The interior Automation System of Power Network accident separating brake that is not present of setting time is remembered before and after remote signalling divides position time of origin
Record；
Regular 1.2.2)：Have in secondary remote signalling protection or remote signalling SOE data before and after the time of origin of remote signalling point position in setting time
Include the record of protection act keyword；
Regular 1.2.3)：In front and rear 30 minutes, there is non-zero status in remote measurement current value, i.e. current value is more than or equal to 5 amperes of feelings
Condition；
All remote signalling displacement information for meeting above-mentioned regular automated system is used as indirect event, and can be inquired about, for
The remote signalling displacement being screened in indirect event, system captures its corresponding auxiliary information from SCADA system at the same time, according to auxiliary
Help data that breakdown judge rule is set；
1.3) TCM events collect matching association in advance；TCM events refer to that the failure from production management system crawl reports information, and will
The information is matched and associated with SCADA event, forms breakdown judge rule；
2) the multi-source heterogeneous information pre-processing of power distribution network is with merging：
2.1) pretreatment of the multi-source heterogeneous data of distribution, shows and in terms of event handling is several from business decision logic, information
Specificity analysis, data cleansing and extraction are carried out to data, then using data source, purposes and data attribute as dimension, with reference to
1) data in collect rule in advance, carry out sort merge to multi-source heterogeneous data, are put in order by rule in independent region；Locating in advance
After the completion of reason, the association of cross-platform data will be carried out with merging with attribute according to the source of pre- collection data；
2.2) distribution multi-source heterogeneous data storage, the multi-source heterogeneous data of distribution it is preprocessed with merge after, transform into structure
Change data, build data model, structural data batch is imported big data platform is stored；
3) Fault Diagnosis of Distribution Network based on multi-source data, is the judgement to accident separating brake in SCADA system and TCM systems first
Processing, confirms to break down；Second step is excavated and the relevant protection act of failure, remote measurement from SCADA system and TCM systems
Phenomena such as unusual fluctuation, total accident and associated processing outcomes, clear failure object, content and time, the 3rd step combination auxiliary information
Comprehensive verification failure, there are manual intervention is submitted at ambiguity.
2. distribution network failure information excavating and diagnostic method according to claim 1 based on multi-source heterogeneous data, it is special
Sign is that step 3) is specially：
The judgement processing of the first step and second step is followed successively by：
3.1) diagnostic rule 1:
If there is " capacitor " or " capacitance " or " reactance " or " capacitive reactance " in event content, event is non-faulting；To failure
Information carries out the judgement of diagnostic rule 2 or 3；
3.2) diagnostic rule 2：
If there is " ground connection " or " 3U0 " or " 3V0 " or " overvoltage " or " overvoltage " in secondary remote signalling alarm record and do not include
" ground connection become ", " grounding switch ", " earthing switch ", " device alarm ", " resistance ", " switch ", " protection act mark put act bit into
1 ", then event is small current neutral grounding；
3.3) diagnostic rule 3：
If there is " tripping " or " coincidence " or " separating brake " or " action " in event content, event is short trouble；
If event corresponds to appearance " tripping " in trip condition, event is short trouble；
If reported for repairment, appearance " signal " and plant stand name in content include " station " or implementor name " switch " occurs or implementor name occurs
Breaker, then event is short trouble；
3.4) judgement of diagnostic rule 4 is carried out according to diagnostic rule 2：
If SCADA busbar grounding event matches in 1min before and after the ID and event time of small current neutral grounding point, event replicates
SCADA busbar grounding information is as fault message；
If without above-mentioned matching, but when SCADA 2 is small it is interior have busbar grounding event, then before and after the match event moment in 1min
SCADA busbar groundings event simultaneously replicates ground connection information as fault message；
If above-mentioned be unsatisfactory for, secondary remote signalling alarm, remote signalling SOE, remote signalling point position, remote-control data are captured, then mined information,
According to pre- collection rule judgment fault message；
3.5) judgement of diagnostic rule 5 is carried out according to diagnostic rule 3：
Failure is the direct events of SCADA, replicates the fault message of the direct event correlations of SCADA as a result；
Failure is TCM information, such as the direct event successes of TCM information matches association SCADA, then replicates the direct event correlations of SCADA
Fault message as a result；Match it is unsuccessful, if TCM information matches association SCADA indirect event successes, excavate matching factory
Stand and data in equipment 1min/8h, fill up fault message, if also unsuccessful, crawl remote signalling alarm, remote signalling SOE, remote signalling point
Position, remote control and telemetry, are excavated, failure judgement information with reference to the pre- collection rule of the direct or indirect events of SCADA；
The verification and manual intervention of 3rd step be specially：
3.6) auxiliary information verifies：
Power distribution network data source further includes power information acquisition system, regional Meteorological Information System, power distribution network in addition to production system
Topological data, geographical location information and distributed generation resource operational monitoring information, according to auxiliary judgment needs, to other information systems
Pretreatment is assessed and filtered to data, establishes auxiliary correlation model, auxiliary information is provided for the fault diagnosis of production system；
3.7) manual intervention：
For the critical data deletion condition occurred during fault diagnosis, processing is compensated to the criterion lacked, is believed according to SOE
Number conversion and accident resultant signal, obtain the displacement signal that lacks originally, then carry out the logical process of next step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711350231.5A CN108037415A (en) | 2017-12-15 | 2017-12-15 | Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711350231.5A CN108037415A (en) | 2017-12-15 | 2017-12-15 | Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108037415A true CN108037415A (en) | 2018-05-15 |
Family
ID=62103195
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711350231.5A Pending CN108037415A (en) | 2017-12-15 | 2017-12-15 | Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108037415A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109406943A (en) * | 2018-11-16 | 2019-03-01 | 国网江苏省电力有限公司盐城供电分公司 | A kind of active distribution network monitoring method based on big data |
| CN109445306A (en) * | 2018-10-26 | 2019-03-08 | 湖南磁浮技术研究中心有限公司 | Automatic associated parameter interpretation method and system based on rule configuration analysis |
| CN109490707A (en) * | 2018-11-13 | 2019-03-19 | 国网江苏省电力有限公司南通供电分公司 | The automatic analysis method of electric network fault tripping based on multidimensional multi-source grid operation data |
| CN109494882A (en) * | 2018-12-29 | 2019-03-19 | 上海南华兰陵电气有限公司 | A kind of diagnostic method and system of transformer substation switch equipment state |
| CN109617001A (en) * | 2018-12-28 | 2019-04-12 | 广西电网有限责任公司防城港供电局 | A kind of multi-source heterogeneous information intelligent processing system of relay protection |
| CN109813999A (en) * | 2019-01-22 | 2019-05-28 | 山东大学 | A kind of Fault Diagnosis of Distribution Network algorithm automatically testing platform, method and application |
| CN110210632A (en) * | 2019-05-31 | 2019-09-06 | 国网河北省电力有限公司沧州供电分公司 | Failure emergency processing method, device and terminal based on ubiquitous electric power Internet of Things |
| CN110516929A (en) * | 2019-08-09 | 2019-11-29 | 国网浙江省电力有限公司 | Power transmission and transformation closed-loop data processing method based on multi-source information |
| CN111062633A (en) * | 2019-12-24 | 2020-04-24 | 广东电网有限责任公司 | Power transmission and transformation line and equipment state evaluation system based on multi-source heterogeneous data |
| CN111125074A (en) * | 2019-12-12 | 2020-05-08 | 深圳供电局有限公司 | Power distribution Internet of things data processing method and device |
| CN111257686A (en) * | 2020-01-15 | 2020-06-09 | 国家电网有限公司 | Intelligent tripping analysis system based on data sharing analysis |
| CN111313355A (en) * | 2020-03-02 | 2020-06-19 | 国网江苏省电力有限公司南京供电分公司 | Method for updating monitoring signal event rule under manual supervision |
| CN111366814A (en) * | 2020-03-31 | 2020-07-03 | 上海电力大学 | Power grid fault diagnosis method based on multi-source data and multi-dimensional fault coding space |
| CN111400295A (en) * | 2020-03-13 | 2020-07-10 | 国电南瑞科技股份有限公司 | Power distribution network power failure event analysis method and device and storage medium |
| CN111999605A (en) * | 2020-09-16 | 2020-11-27 | 珠海许继芝电网自动化有限公司 | Power distribution network fault tolerance judgment method and device based on fault correlation analysis |
| CN112147459A (en) * | 2020-08-12 | 2020-12-29 | 国电南瑞科技股份有限公司 | Power grid fault analysis device and method based on SCADA system |
| CN112507227A (en) * | 2020-12-15 | 2021-03-16 | 北京中科智营科技发展有限公司 | Intelligent perception search platform |
| CN113537415A (en) * | 2021-09-17 | 2021-10-22 | 中国南方电网有限责任公司超高压输电公司广州局 | Convertor station inspection method and device based on multi-information fusion and computer equipment |
| CN113837423A (en) * | 2020-06-24 | 2021-12-24 | 国家电网有限公司华东分部 | Power grid operation situation prediction method based on energy internet electric power big data |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5343155A (en) * | 1991-12-20 | 1994-08-30 | The Research And Development Institute, Inc. At Montana State University | Fault detection and location system for power transmission and distribution lines |
| CN103941158A (en) * | 2014-04-17 | 2014-07-23 | 国家电网公司 | Power distribution network fault diagnosis system and method based on multi-source information |
| CN104463712A (en) * | 2014-12-22 | 2015-03-25 | 国网上海市电力公司 | Intelligent distribution network fault information statistical analysis system |
| CN104459474A (en) * | 2014-12-22 | 2015-03-25 | 国网上海市电力公司 | Intelligent distribution network fault recognition method |
| CN105353702A (en) * | 2015-11-17 | 2016-02-24 | 国家电网公司 | High voltage equipment intelligent monitoring system |
| CN106339509A (en) * | 2016-10-26 | 2017-01-18 | 国网山东省电力公司临沂供电公司 | Power grid operation data sharing system based on large data technology |
| CN107368932A (en) * | 2017-08-09 | 2017-11-21 | 国网山东省电力公司经济技术研究院 | A kind of load Analysis forecasting system suitable for power network development specialty |
-
2017
- 2017-12-15 CN CN201711350231.5A patent/CN108037415A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5343155A (en) * | 1991-12-20 | 1994-08-30 | The Research And Development Institute, Inc. At Montana State University | Fault detection and location system for power transmission and distribution lines |
| CN103941158A (en) * | 2014-04-17 | 2014-07-23 | 国家电网公司 | Power distribution network fault diagnosis system and method based on multi-source information |
| CN104463712A (en) * | 2014-12-22 | 2015-03-25 | 国网上海市电力公司 | Intelligent distribution network fault information statistical analysis system |
| CN104459474A (en) * | 2014-12-22 | 2015-03-25 | 国网上海市电力公司 | Intelligent distribution network fault recognition method |
| CN105353702A (en) * | 2015-11-17 | 2016-02-24 | 国家电网公司 | High voltage equipment intelligent monitoring system |
| CN106339509A (en) * | 2016-10-26 | 2017-01-18 | 国网山东省电力公司临沂供电公司 | Power grid operation data sharing system based on large data technology |
| CN107368932A (en) * | 2017-08-09 | 2017-11-21 | 国网山东省电力公司经济技术研究院 | A kind of load Analysis forecasting system suitable for power network development specialty |
Non-Patent Citations (2)
| Title |
|---|
| 宋杰 等: "基于多源异构数据挖掘的配电网故障信息统计分析", 《电力系统保护与控制》 * |
| 庄伟明 等: "《计算机技术基础》", 30 September 2017, 上海大学出版社 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109445306A (en) * | 2018-10-26 | 2019-03-08 | 湖南磁浮技术研究中心有限公司 | Automatic associated parameter interpretation method and system based on rule configuration analysis |
| CN109445306B (en) * | 2018-10-26 | 2022-01-25 | 湖南磁浮技术研究中心有限公司 | Automatic associated parameter interpretation method and system based on rule configuration analysis |
| CN109490707A (en) * | 2018-11-13 | 2019-03-19 | 国网江苏省电力有限公司南通供电分公司 | The automatic analysis method of electric network fault tripping based on multidimensional multi-source grid operation data |
| CN109406943A (en) * | 2018-11-16 | 2019-03-01 | 国网江苏省电力有限公司盐城供电分公司 | A kind of active distribution network monitoring method based on big data |
| CN109617001A (en) * | 2018-12-28 | 2019-04-12 | 广西电网有限责任公司防城港供电局 | A kind of multi-source heterogeneous information intelligent processing system of relay protection |
| CN109617001B (en) * | 2018-12-28 | 2021-03-30 | 广西电网有限责任公司防城港供电局 | Relay protection multi-source heterogeneous information intelligent processing system |
| CN109494882B (en) * | 2018-12-29 | 2020-11-24 | 上海南华兰陵电气有限公司 | Method and system for diagnosing state of substation switch equipment |
| CN109494882A (en) * | 2018-12-29 | 2019-03-19 | 上海南华兰陵电气有限公司 | A kind of diagnostic method and system of transformer substation switch equipment state |
| CN109813999A (en) * | 2019-01-22 | 2019-05-28 | 山东大学 | A kind of Fault Diagnosis of Distribution Network algorithm automatically testing platform, method and application |
| CN110210632A (en) * | 2019-05-31 | 2019-09-06 | 国网河北省电力有限公司沧州供电分公司 | Failure emergency processing method, device and terminal based on ubiquitous electric power Internet of Things |
| CN110516929A (en) * | 2019-08-09 | 2019-11-29 | 国网浙江省电力有限公司 | Power transmission and transformation closed-loop data processing method based on multi-source information |
| CN111125074A (en) * | 2019-12-12 | 2020-05-08 | 深圳供电局有限公司 | Power distribution Internet of things data processing method and device |
| CN111062633A (en) * | 2019-12-24 | 2020-04-24 | 广东电网有限责任公司 | Power transmission and transformation line and equipment state evaluation system based on multi-source heterogeneous data |
| CN111257686A (en) * | 2020-01-15 | 2020-06-09 | 国家电网有限公司 | Intelligent tripping analysis system based on data sharing analysis |
| CN111313355B (en) * | 2020-03-02 | 2022-06-10 | 国网江苏省电力有限公司南京供电分公司 | Method for updating monitoring signal event rule under manual supervision |
| CN111313355A (en) * | 2020-03-02 | 2020-06-19 | 国网江苏省电力有限公司南京供电分公司 | Method for updating monitoring signal event rule under manual supervision |
| CN111400295B (en) * | 2020-03-13 | 2022-10-14 | 国电南瑞科技股份有限公司 | Power distribution network power failure event analysis method and device and storage medium |
| CN111400295A (en) * | 2020-03-13 | 2020-07-10 | 国电南瑞科技股份有限公司 | Power distribution network power failure event analysis method and device and storage medium |
| CN111366814A (en) * | 2020-03-31 | 2020-07-03 | 上海电力大学 | Power grid fault diagnosis method based on multi-source data and multi-dimensional fault coding space |
| CN113837423A (en) * | 2020-06-24 | 2021-12-24 | 国家电网有限公司华东分部 | Power grid operation situation prediction method based on energy internet electric power big data |
| CN112147459A (en) * | 2020-08-12 | 2020-12-29 | 国电南瑞科技股份有限公司 | Power grid fault analysis device and method based on SCADA system |
| CN111999605A (en) * | 2020-09-16 | 2020-11-27 | 珠海许继芝电网自动化有限公司 | Power distribution network fault tolerance judgment method and device based on fault correlation analysis |
| CN111999605B (en) * | 2020-09-16 | 2023-11-07 | 珠海许继芝电网自动化有限公司 | Power distribution network fault tolerance judging method and device based on fault correlation analysis |
| CN112507227A (en) * | 2020-12-15 | 2021-03-16 | 北京中科智营科技发展有限公司 | Intelligent perception search platform |
| CN112507227B (en) * | 2020-12-15 | 2024-03-01 | 北京中科智营科技发展有限公司 | Intelligent perception search platform |
| CN113537415A (en) * | 2021-09-17 | 2021-10-22 | 中国南方电网有限责任公司超高压输电公司广州局 | Convertor station inspection method and device based on multi-information fusion and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108037415A (en) | Distribution network failure information excavating and diagnostic method based on multi-source heterogeneous data | |
| CN106709580B (en) | Transformer substation secondary system operation and maintenance cloud platform | |
| CN105245185B (en) | A kind of area distribution formula photovoltaic fault diagnosis system and method for accessing power distribution network | |
| CN103001328B (en) | Fault diagnosis and assessment method of intelligent substation | |
| CN103336222B (en) | Power system fault diagnosis method based on fuzzy reasoning pulse neurolemma system | |
| CN103744850B (en) | A kind of electrical network disaster real-time monitoring device and method based on intuitionistic fuzzy-rough sets | |
| CN110674189B (en) | Method for monitoring secondary state and positioning fault of intelligent substation | |
| CN102638100B (en) | District power network equipment abnormal alarm signal association analysis and diagnosis method | |
| CN103901882B (en) | A kind of system and method for train dynamics system on-line monitoring fault diagnosis | |
| CN106655522A (en) | Master station system suitable for operation and maintenance management of secondary equipment of power grid | |
| CN102142716B (en) | Power grid online fault diagnosis method based on three-state data multidimensional cooperative processing | |
| CN102035202B (en) | Network reconfiguration system | |
| CN107346466A (en) | A kind of control method and device of electric power dispatching system | |
| CN107798395A (en) | A kind of power grid accident signal automatic diagnosis method and system | |
| CN106908690A (en) | Distributed intelligence warning system and its method for diagnosing faults between boss station | |
| CN106124935A (en) | Middle and low voltage network Fault Locating Method | |
| CN104459378B (en) | A kind of intelligent substation method for diagnosing faults | |
| CN108051709A (en) | Transformer state online evaluation analysis method based on artificial intelligence technology | |
| CN111768076B (en) | Monitoring alarm signal clustering method taking power grid event as center | |
| CN107834523B (en) | Extra-high voltage direct-current fault diagnosis system and working method based on model and rule base | |
| CN105974232B (en) | A kind of electric network failure diagnosis method suitable for grid | |
| CN105573283B (en) | The foundation in substation equipment function association storehouse and the filter method of event correlation information group | |
| CN104463712A (en) | Intelligent distribution network fault information statistical analysis system | |
| CN109474067A (en) | A kind of dispatching of power netwoks troubleshooting aid decision-making method | |
| CN108683187A (en) | A kind of EMS grid monitoring systems based on big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180515 |