CN109586239A - Intelligent substation real-time diagnosis and fault early warning method - Google Patents
Intelligent substation real-time diagnosis and fault early warning method Download PDFInfo
- Publication number
- CN109586239A CN109586239A CN201811505559.4A CN201811505559A CN109586239A CN 109586239 A CN109586239 A CN 109586239A CN 201811505559 A CN201811505559 A CN 201811505559A CN 109586239 A CN109586239 A CN 109586239A
- Authority
- CN
- China
- Prior art keywords
- data
- fault
- substation
- network
- machine learning
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000003745 diagnosis Methods 0.000 title claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 6
- 238000013480 data collection Methods 0.000 claims abstract description 5
- 238000012098 association analyses Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000005065 mining Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 5
- 238000003012 network analysis Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000272814 Anser sp. Species 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0092—Details of emergency protective circuit arrangements concerning the data processing means, e.g. expert systems, neural networks
-
- H02J13/0006—
Landscapes
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention discloses intelligent substation real-time diagnosis and fault early warning methods, it is characterized in that, described method includes following steps: S1: forming fault data in the electric equipment operation data, network flow data, server log data input database when substation fault is occurred, establishes mass data collection sample training space;S2: network packet is captured by network packet analysis tool, and the data upload database in network packet is analyzed and shown, the data in network packet are network data;S3: carrying out machine learning to data in database, carries out comparison of classification to fault data and network data by machine learning, obtains fault model;S4: being based on fault model above-mentioned, forms substation equipment operation troubles parameter fluctuation threshold value, and access Real Time Data of Substation, when the fluctuation threshold of real time data is more than the early warning value of fault model, carries out fault pre-alarming and shows.
Description
Technical field
The present invention relates to a kind of fault early warning methods, and in particular to intelligent substation real-time diagnosis and fault early warning method.
Background technique
IEC61850 specification realizes the digitlization and informationization of substation, is realized by GOOSE, SV and MMS message
Information exchange between full station equipment completes the functions such as the control, protection and monitoring of equipment.In order to help operation maintenance personnel and technology
Personnel fault point and can look for failure cause in time after transformer substation case generation, and intelligent substation is equipped with network analysis
System, existing network analysis system realize whole network data acquisition storage, the monitoring of communication link state, exception information announcement
Alert function, but lack the means that effectively various warning information are classified and screened, facing operation maintenance personnel
It is helpless and without any idea when a large amount of warning information.Exist independently of each other in addition, link is abnormal with information such as equipment alarms,
Between do not have association analysis ability, cause current network analysis system that can only show extremely, do not have intellectual analysis ability,
It cannot achieve Situation Awareness and active forewarning.
Summary of the invention
The technical problem to be solved by the present invention is to diagnostic result blurrings, and a large amount of abnormality alarming can only be prompted to believe
Breath, can not classify to warning information and provide defect elimination measure, be unable to Dynamic Display whole transformer station equipment and the network operation
State, technical staff can only analyze and trace to its source step by step by carrying out dismantling layer by layer to data packet, not only technical threshold
Height, working efficiency is also extremely inefficient, and it is an object of the present invention to provide intelligent substation real-time diagnosis and fault early warning method, solve above-mentioned
Problem.
The present invention is achieved through the following technical solutions:
Intelligent substation real-time diagnosis and fault early warning method, which is characterized in that described method includes following steps: S1:
In electric equipment operation data, network flow data, server log data input database when substation fault is occurred
Fault data is formed, mass data collection sample training space is established;S2: network packet is captured by network packet analysis tool, and will
Data in network packet upload database and are analyzed and shown, the data in network packet are network data;S3: in data
Machine learning is carried out to data in library, comparison of classification is carried out to fault data and network data by machine learning, obtains failure
Model;S4: being based on fault model above-mentioned, forms substation equipment operation troubles parameter fluctuation threshold value, and accesses substation's reality
When data, when the fluctuation threshold of real time data be more than fault model early warning value when, carry out fault pre-alarming show.
Existing network analysis system technology realizes step: 1) mirror image switch port;2) it acquires whole network data message and deposits
Storage is in local hard drive;3) according to the collected data message of IEC61850 protocol analysis;4) to abnormal conditions (communication disruption, survey
Control equipment does not receive signal etc.) it is alerted.It can lead to the problem of very much, if diagnostic result is blurred, can only prompt big in this way
The abnormality alarming information of amount can not classify to warning information and provide defect elimination measure;It is unable to Dynamic Display whole transformer station
Equipment and network operation state, technical staff can only be by carrying out dismantling analysis layer by layer and getting to the bottom of it step by step to trace back to data packet
Source, not only technical threshold is high, and working efficiency is also extremely inefficient.
In order to solve the above-mentioned technical problem, by way of to the real-time collection analysis of intelligent substation network data, system
It is made of front end data acquisition system, network transmitting sub-system and back-end analysis platform.The network data of substation is adopted in realization
Collection, processing, analysis, criterion and prediction etc., and whole network operating status is visualized.
Further, the machine learning in the step S3 includes Supervised machine learning and unsupervised machine learning, institute
Stating Supervised machine learning is that the failure and abnormal phenomenon common to substation are classified, and settling time sequence algorithm is come to working as
Preceding substation operation situation is analyzed and predicted;The unsupervised machine learning uses association analysis mode, passes through discovery frequency
Numerous item collection and Mining Association Rules carry out classification comparison to fault data and network data, then carry out analysis prediction.
Further, the unsupervised machine learning is associated operation using Apriori algorithm, passes through what is successively searched for
Iteration is associated classification to classification data.
Further, the electric equipment operation data in the step S1 include electric current, voltage, active Value Data, into
When row fault pre-alarming is shown, each data in electrical equipment are shown.
Compared with prior art, the present invention having the following advantages and benefits: being transported primarily directed to intelligent substation
The inefficient status of maintenance mode is tieed up, a kind of realization of a kind of Intelligence Diagnosis and Predicting Technique is proposed, has combined supervision
The advantages of technologies such as machine learning, unsupervised machine learning, fundamentally changes existing substation's O&M and diagnostic mode, mentions
High substation's maintenance work efficiency, real-time online diagnose substation " health status ", simultaneously to sensed in advance the problem of being likely to occur
Preventive guidance opinion is provided, is provided for a rainy day, nips off the symptom of a trend of failure, social and economic benefit with higher in time.
Detailed description of the invention
Attached drawing described herein is used to provide to further understand the embodiment of the present invention, constitutes one of the application
Point, do not constitute the restriction to the embodiment of the present invention.In the accompanying drawings:
Fig. 1 is process flow diagram of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below with reference to embodiment and attached drawing, to this
Invention is described in further detail, and exemplary embodiment of the invention and its explanation for explaining only the invention, are not made
For limitation of the invention.
Embodiment
Intelligent substation real-time diagnosis and fault early warning method, which is characterized in that described method includes following steps: S1:
In electric equipment operation data, network flow data, server log data input database when substation fault is occurred
Fault data is formed, mass data collection sample training space is established;S2: network packet is captured by network packet analysis tool, and will
Data in network packet upload database and are analyzed and shown, the data in network packet are network data;S3: in data
Machine learning is carried out to data in library, comparison of classification is carried out to fault data and network data by machine learning, obtains failure
Model;S4: being based on fault model above-mentioned, forms substation equipment operation troubles parameter fluctuation threshold value, and accesses substation's reality
When data, when the fluctuation threshold of real time data be more than fault model early warning value when, carry out fault pre-alarming show.By to intelligence
The mode of substation network real time data acquisition analysis, system is by front end data acquisition system, network transmitting sub-system and rear end
Analysis platform composition.It realizes to the network data acquisition of substation, processing, analysis, criterion and prediction etc., and whole network is transported
Row state is visualized.
(1) mass data collection sample training space is established
It collects substation's typical fault case and to record electric equipment operation data when failure occurs (electric current, voltage, active
Value etc.), network flow data, server log data etc., while can also be by being obtained in the different failure of laboratory simulation
Related data.
(2) packet capture
Sniffer, Net Flow and Wireshark are that the main function of classical network packet analysis tool is an attempt to capture net
Network packet, and attempt the situation as detailed as possible of display packet.Key property has following several parts: supporting that UNIX and Windows is flat
Platform can display the details of the detailed protocols information of packet, can open/save the packet of capture, can import in interface real-time capture packet
It exports other and captures the bag data format that program is supported, bag filter, various ways can search packet in several ways, pass through
Filter is shown with multicolour and is wrapped, and creates a variety of statistical analysis, captures multiple network interface, the text for supporting a variety of other programs to capture
Part supports multi-format outputting, decodes to various protocols and provides support, open source software.
(3) Supervised machine learning
Machine learning includes supervised learning and unsupervised learning, artificially normal to substation under supervised learning mode
The failure and abnormal phenomenon seen are classified, settling time sequence algorithm (regression algorithm) come to current transformer substation operating condition into
Row analysis and prediction.
(4) unsupervised machine learning
It is association analysis that unsupervised a few areas, which learn a most important ring, it can be used for concentrating discovery number in large-scale data
It is believed that relevance or correlation between breath.The target of association analysis mainly includes two: discovery frequent item set is associated with excavation
Rule.Apriori algorithm is a kind of efficient algorithm for finding frequent item set, it uses a kind of iteration side for being referred to as and successively searching for
Method, i.e. k item collection are for exploring (k+1) item collection: firstly, finding out the set of frequent 1 item collection, which is denoted as L1, and L1 is for looking for frequency
The set L2 of numerous 2 item collection, and L2 so goes down, for looking for L3 until that cannot find k item collection.A Lk is often looked for need a number
It is scanned according to library.To improve the efficiency that frequent item set successively generates, a kind of critical nature being referred to as Apriori is empty for compressing search
Between, basic principle is: if 1, some item collection is frequently, its all subsets certainly will be also frequent;2, such as
As soon as fruit item collection is nonmatching grids, then the superset corresponding to it is all nonmatching grids.Although Apriori algorithm can
To avoid the exponential increase of item collection number, so that frequent item set is calculated within the reasonable time, but its efficiency is still less managed
Think.In order to more rapidly find out frequent item set, researcher proposes FP-growth algorithm to make up Apriori algorithm
It is insufficient.It is compared with Apriori algorithm, FP-growth algorithm only needs to traverse database twice, thus efficiently discovery frequency
Numerous item collection.FP-growth algorithm is based on Apriori principle, by the way that data set is stored in FP (Frequent
Pattern frequent item set) is found on tree, but cannot find the correlation rule between data.FP-growth algorithm only needs logarithm
Twice sweep is carried out according to library, and Apriori algorithm requires run-down data set when seeking each potential frequent item set, institute
It is efficient with FP-growth algorithm.
(5) fault location/anticipation
Compared based on data training above-mentioned, unsupervised learning, association analysis, substation configuration description file SCD and etc., shape
At a set of failure criterion and forecasting system, current transformer substation operation data (electric current, has work value at voltage) and network data are inputted
(flow, data package size, time delay etc.), whether output is then normal for Substation Electric Equipment, and failure, there are the states such as hidden danger.
(6) it visualizes
A set of friendly interface is being formed from the background, and functional coverage comprehensively visualizes system, so that operation maintenance personnel is in master
Control grasp whole transformer station electrical equipment and the real-time diagnosis of network operation situation concrete intelligence substation that room can be comprehensive in real time
And fault early warning system overview flow chart is as shown in Figure 1.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not intended to limit the present invention the foregoing is merely a specific embodiment of the invention
Protection scope, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (4)
1. intelligent substation real-time diagnosis and fault early warning method, which is characterized in that described method includes following steps:
S1: electric equipment operation data, network flow data, server log data when substation fault is occurred input number
According to fault data is formed in library, mass data collection sample training space is established;
S2: network packet is captured by network packet analysis tool, and the data upload database in network packet is subjected to analysis and is gone forward side by side
Row shows that the data in network packet are network data;
S3: machine learning is carried out to data in database, is classified by machine learning to fault data and network data
Comparison obtains fault model;
S4: being based on fault model above-mentioned, forms substation equipment operation troubles parameter fluctuation threshold value, and it is real-time to access substation
Data carry out fault pre-alarming and show when the fluctuation threshold of real time data is more than the early warning value of fault model.
2. intelligent substation real-time diagnosis according to claim 1 and fault early warning method, which is characterized in that the step
Machine learning in S3 includes Supervised machine learning and unsupervised machine learning, and the Supervised machine learning is to substation
Common failure and abnormal phenomenon is classified, settling time sequence algorithm come to current transformer substation operating condition carry out analysis and
Prediction;The unsupervised machine learning uses association analysis mode, by discovery frequent item set and Mining Association Rules to failure
Data and network data carry out classification comparison, then carry out analysis prediction.
3. intelligent substation real-time diagnosis according to claim 2 and fault early warning method, which is characterized in that the no prison
It superintends and directs machine learning and operation is associated using Apriori algorithm, classification data is associated point by the iteration successively searched for
Class.
4. intelligent substation real-time diagnosis according to claim 1 and fault early warning method, which is characterized in that the step
Electric equipment operation data in S1 include electric current, voltage, active Value Data, when carrying out fault pre-alarming display, will electrically be set
Each data in standby are shown.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811505559.4A CN109586239B (en) | 2018-12-10 | 2018-12-10 | Real-time diagnosis and fault early warning method for intelligent substation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811505559.4A CN109586239B (en) | 2018-12-10 | 2018-12-10 | Real-time diagnosis and fault early warning method for intelligent substation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109586239A true CN109586239A (en) | 2019-04-05 |
CN109586239B CN109586239B (en) | 2020-03-31 |
Family
ID=65928749
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811505559.4A Active CN109586239B (en) | 2018-12-10 | 2018-12-10 | Real-time diagnosis and fault early warning method for intelligent substation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109586239B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110703743A (en) * | 2019-11-12 | 2020-01-17 | 深圳市亲邻科技有限公司 | Equipment failure prediction and detection system and method |
CN110866616A (en) * | 2019-11-01 | 2020-03-06 | 许继集团有限公司 | Fault early warning method and device for secondary equipment of transformer substation |
CN110941918A (en) * | 2019-12-30 | 2020-03-31 | 国网安徽省电力有限公司 | Intelligent substation fault analysis system |
CN111371180A (en) * | 2020-03-23 | 2020-07-03 | 国网黑龙江省电力有限公司鹤岗供电公司 | Substation patrol supervision and data analysis system |
CN112329914A (en) * | 2020-10-26 | 2021-02-05 | 华翔翔能科技股份有限公司 | Fault diagnosis method and device for buried transformer substation and electronic equipment |
CN112398218A (en) * | 2020-09-28 | 2021-02-23 | 国网山东省电力公司冠县供电公司 | Power grid safety and stability analysis device and method |
CN114442543A (en) * | 2021-10-29 | 2022-05-06 | 南京河海南自水电自动化有限公司 | Computer monitoring method suitable for early warning of hydropower station fault |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010011918A2 (en) * | 2008-07-24 | 2010-01-28 | University Of Cincinnati | Methods for prognosing mechanical systems |
CN103136587A (en) * | 2013-03-07 | 2013-06-05 | 武汉大学 | Power distribution network operating state classification recognition method based on support vector machine |
CN104102773A (en) * | 2014-07-05 | 2014-10-15 | 山东鲁能软件技术有限公司 | Equipment fault warning and state monitoring method |
CN106779505A (en) * | 2017-02-28 | 2017-05-31 | 中国南方电网有限责任公司 | A kind of transmission line malfunction method for early warning driven based on big data and system |
-
2018
- 2018-12-10 CN CN201811505559.4A patent/CN109586239B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010011918A2 (en) * | 2008-07-24 | 2010-01-28 | University Of Cincinnati | Methods for prognosing mechanical systems |
CN103136587A (en) * | 2013-03-07 | 2013-06-05 | 武汉大学 | Power distribution network operating state classification recognition method based on support vector machine |
CN104102773A (en) * | 2014-07-05 | 2014-10-15 | 山东鲁能软件技术有限公司 | Equipment fault warning and state monitoring method |
CN106779505A (en) * | 2017-02-28 | 2017-05-31 | 中国南方电网有限责任公司 | A kind of transmission line malfunction method for early warning driven based on big data and system |
Non-Patent Citations (2)
Title |
---|
CHEAGOUN: "机器学习简介", 《百度文库,HTTPS://WENKU.BAIDU.COM/VIEW/84747028FBD6195F312B3169A45177232F60E4E7.HTML?FROM=SEARCH》 * |
刘林凡: "基于机器学习的电力变压器故障诊断的研究进展", 《电子世界》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110866616A (en) * | 2019-11-01 | 2020-03-06 | 许继集团有限公司 | Fault early warning method and device for secondary equipment of transformer substation |
CN110703743A (en) * | 2019-11-12 | 2020-01-17 | 深圳市亲邻科技有限公司 | Equipment failure prediction and detection system and method |
CN110941918A (en) * | 2019-12-30 | 2020-03-31 | 国网安徽省电力有限公司 | Intelligent substation fault analysis system |
CN110941918B (en) * | 2019-12-30 | 2024-03-19 | 国网安徽省电力有限公司 | Intelligent substation fault analysis system |
CN111371180A (en) * | 2020-03-23 | 2020-07-03 | 国网黑龙江省电力有限公司鹤岗供电公司 | Substation patrol supervision and data analysis system |
CN112398218A (en) * | 2020-09-28 | 2021-02-23 | 国网山东省电力公司冠县供电公司 | Power grid safety and stability analysis device and method |
CN112329914A (en) * | 2020-10-26 | 2021-02-05 | 华翔翔能科技股份有限公司 | Fault diagnosis method and device for buried transformer substation and electronic equipment |
CN112329914B (en) * | 2020-10-26 | 2024-02-02 | 华翔翔能科技股份有限公司 | Fault diagnosis method and device for buried transformer substation and electronic equipment |
CN114442543A (en) * | 2021-10-29 | 2022-05-06 | 南京河海南自水电自动化有限公司 | Computer monitoring method suitable for early warning of hydropower station fault |
Also Published As
Publication number | Publication date |
---|---|
CN109586239B (en) | 2020-03-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109586239A (en) | Intelligent substation real-time diagnosis and fault early warning method | |
Zhao et al. | Multivariate time-series anomaly detection via graph attention network | |
CN110717665B (en) | System and method for fault identification and trend analysis based on scheduling control system | |
CN107577588A (en) | A kind of massive logs data intelligence operational system | |
CN112187514A (en) | Intelligent operation and maintenance system, method and terminal for data center network equipment | |
CN106371986A (en) | Log treatment operation and maintenance monitoring system | |
CN116796907A (en) | Water environment dynamic monitoring system and method based on Internet of things | |
CN108809974A (en) | A kind of Network Abnormal recognition detection method and device | |
CN115809183A (en) | Method for discovering and disposing information-creating terminal fault based on knowledge graph | |
CN116862442A (en) | Intelligent equipment management method and system based on big data analysis | |
CN107145959A (en) | A kind of electric power data processing method based on big data platform | |
CN109501834A (en) | A kind of point machine failure prediction method and device | |
CN106503439A (en) | A kind of method of the collection fault early warning system based on data mining | |
TWI721693B (en) | Network behavior anomaly detection system and method based on mobile internet of things | |
CN102420700A (en) | Network fault diagnosis system | |
JP2015095060A (en) | Log analysis device and method | |
Chen et al. | Graph-based incident aggregation for large-scale online service systems | |
CN115237717A (en) | Micro-service abnormity detection method and system | |
CN118200118A (en) | Substation communication network equipment monitoring and fault early warning method and system | |
CN115114856A (en) | Intelligent manufacturing production line operation system based on digital twin | |
CN117235169A (en) | Wisdom fortune dimension data storage platform | |
CN110816938B (en) | Big data analysis method based on comprehensive detection platform of cigarette packaging machine | |
CN112803587A (en) | Intelligent inspection method for state of automatic equipment based on diagnosis decision library | |
CN107454089A (en) | A kind of network safety situation diagnostic method based on multinode relevance | |
CN116155581A (en) | Network intrusion detection method and device based on graph neural network |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |