CN106199494A - A kind of intelligent diagnosis system based on metering device fault - Google Patents
A kind of intelligent diagnosis system based on metering device fault Download PDFInfo
- Publication number
- CN106199494A CN106199494A CN201610586569.XA CN201610586569A CN106199494A CN 106199494 A CN106199494 A CN 106199494A CN 201610586569 A CN201610586569 A CN 201610586569A CN 106199494 A CN106199494 A CN 106199494A
- Authority
- CN
- China
- Prior art keywords
- metaevent
- data
- diagnosis
- information
- metering device
- 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
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/04—Testing or calibrating of apparatus covered by the other groups of this subclass of instruments for measuring time integral of power or current
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The present invention relates to a kind of intelligent diagnosis system based on metering device fault, described system includes: data acquisition and monitoring modular, for carrying out the collection of related data according to frequency acquisition and transmission means and upload;Metaevent information generating module, it is connected with monitoring modular with data acquisition, for the related data of data acquisition Yu monitoring modular collection is comprehensively analyzed and pretreatment, according to User Profile information, the related data gathered is mated, generate metaevent information by metaevent diagnostic cast;Fault diagnosis and decision-making module, be connected with metaevent information generating module, and the metaevent information for providing according to metaevent information generating module is associated coupling, it is achieved the intelligent diagnostics of metering device fault.Compared with prior art, the present invention have effectively carry out information integration, reject redundancy, improve diagnosis efficiency and diagnostic accuracy advantages of higher.
Description
Technical field
The present invention relates to field of power, especially relate to a kind of intelligent diagnosis system based on metering device fault.
Background technology
According to State Grid Corporation of China about the construction object of power information acquisition system, company's constituent parts has carried out collection the most
The construction of system, State Grid Shanghai Electric Power Company on the basis of all-round construction power information acquisition system in 2010~2014,
Strengthening indices to promote and application dynamics, ended for the end of the year 2015, Shanghai company gathers and covers user 9,920,000 family, substantially realizes
All standing, gathers success rate and is maintained at more than 98.8%, system run all right, carrying out of sustainable every strengthened research.Gather
System has played significant economic benefit in terms of fall damage increases income, saves management and cost of labor;Supporting " marketing greatly " system
Build, resident's step price performs, take precautions against electricity charge risk, improve the aspect obvious management performances of acquirement such as recruitment efficiency;In intelligence
The aspects such as electrical network, good service, ordered electric, energy-saving and emission-reduction, distributed power source access provide strong technical support, obtain prominent
Go out social benefit.Accelerate acquisition system construction and will effectively strengthen marketing lean operating capability with strengthened research, may advantageously facilitate
Sales service pattern intensivization development, strength promotes marketing management to change and breaks through, General Promotion marketing management and the letter of service
Breathization, automatization, intellectuality, interactive level.
By the construction of five years, State Grid Shanghai Electric Power Company's power information acquisition system has been basically completed " complete
Cover, entirely gather " target, in this context, the application demand of acquisition system has been not limited solely to initial electric flux and has adopted
Collection, and then change to deeper data mining, strengthened research.The diagnostic system of existing metering device is difficult in a large number
The electric power meter run is managed, simultaneously because the information only pin such as electric energy meter and the event of acquisition terminal generation, data
For own situation, lack effective association, cause acquisition system to have collected substantial amounts of redundancy, and due to various uncertain
Reason cause information to there is the features such as imperfect, inconsistent so that only rely on them and event can not be made and judging accurately
And explanation.
Summary of the invention
It is an object of the invention to provide a kind of diagnosis efficiency intelligence based on metering device fault accurately for the problems referred to above
Can diagnostic system.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of intelligent diagnosis system based on metering device fault, described system includes:
Data acquisition and monitoring modular, for according to frequency acquisition and transmission means carry out related data collection and
Pass;
Metaevent information generating module, is connected with monitoring modular with data acquisition, for data acquisition and monitoring modular
The related data gathered comprehensively is analyzed and pretreatment, mates, according to User Profile information, the related data gathered,
Metaevent information is generated by metaevent diagnostic cast;
Fault diagnosis and decision-making module, be connected with metaevent information generating module, for generating mould according to metaevent information
The metaevent information that block provides is associated coupling, it is achieved the intelligent diagnostics of metering device fault.
Described related data includes electric energy meter logout, acquisition terminal logout, ac analog and electric quantity data.
Described metaevent includes:
Unique identification, is autonomously generated, as metering device fault for interval with day by data acquisition and monitoring modular
The unique identity of intelligent diagnostics;
Electric energy meter is numbered, for determining the essential information of metering device self;
Metaevent information, is generated by metaevent diagnostic cast, for the base of the intelligent diagnostics as metering device fault
This foundation.
Described metaevent information includes Time To Event, event end time, diagnosis generation Time And Event title.
Described metaevent information generating module includes:
Metaevent data pre-processing unit, is connected with monitoring modular with data acquisition, for data acquisition and monitoring mould
The related data of block collection carries out pretreatment;
Metaevent diagnostic cast call unit, respectively with metaevent data pre-processing unit and fault diagnosis and decision-making module
Connect, be used for calling metaevent diagnostic cast to generate metaevent information.
Described metaevent data pre-processing unit carries out pretreatment to the related data of data acquisition Yu monitoring modular collection
Specifically include the following step:
1) utilize the mode ignoring tuple or interpolation arithmetic that the related data of data acquisition with monitoring modular collection is carried out
Shield or fill corresponding attribute, completing metaevent data scrubbing;
2) metaevent of same entity is carried out metaevent data integration;
3) according to the intelligent diagnostics result of metering device fault, the demand of metaevent information is carried out metaevent data regularization;
4) metaevent data conversion is carried out by the method for smooth, Data generalization, data normalization and attribute construction.
Described metaevent diagnostic cast includes that electric meter fault diagnostic cast, electricity abnormity diagnosis model, voltage x current are abnormal
Diagnostic cast, abnormal electricity consumption diagnostic cast, load abnormity diagnosis model, clock abnormity diagnosis model, wiring abnormity diagnosis model
Abnormity diagnosis model is controlled with taking.
Described fault diagnosis includes with decision-making module:
Diagnostic knowledge base, for depositing examining of the history of metaevent information, equipment essential information and stoichiometric point essential information
Disconnected result;
Mining model sets up unit, is connected with diagnostic knowledge base, is used for setting up mining model, and by mining model to examining
Diagnostic result in disconnected knowledge base excavates, and then extracts diagnosis rule;
Inline diagnosis unit, sets up unit respectively and metaevent information generating module is connected with mining model, for online
Time metaevent information and diagnosis rule are mated, realize the intelligent diagnostics of metering device fault according to the result of coupling.
Described diagnostic result include equipment fault, doubtful stealing, loop exception, electrical network exception, on-site maintenance, error connection and
Promise breaking electricity consumption.
Described mining model includes basic association mining model, multilamellar association mining model, multidimensional association rule mould
Type and meta-rule guidance mining model.
Compared with prior art, the method have the advantages that
(1) present invention takes hierarchy, comprises the data acquisition being sequentially connected with and generates with monitoring modular, metaevent information
Module and fault diagnosis and decision-making module, by this hierarchy, effectively can integrate information and process, thus
Retain effective information and reject redundancy.
(2) by the event of fault diagnosis being decomposed into the single i.e. metaevent of independent event, Monitoring Data can be unified
Form, strengthen data versatility, it is therefore prevented that both there is redundant data, also exist again data imperfect, inconsistent and also time
Clock is asynchronous.
(3) by metaevent is carried out pretreatment, the various noises in metaevent data can be removed, improve fault and examine
Break and the quality of the mining model of foundation in decision-making module.
(4) metaevent data scrubbing can correct the inconsistent of metaevent data, and metaevent data integration can reduce unit
The redundancy of event data and inconsistent, promotes accuracy and the speed of follow-up excavation, and the reduction of metaevent data can reduce number
According to size, improving the efficiency that follow-up data excavates, the conversion of metaevent data can be with specification metaevent data, after improving further
The efficiency of continuous data mining.
(5) by setting up diagnostic knowledge base, can continually strengthen about recurrent abnormal knowledge and corresponding diagnosis
Result so that its support in mining process and confidence level improve constantly, and are more prone to be mined, it is simple to carry out fault
Diagnosis.
(6) by setting up multiple mining model, can be more profound excavate data, improves the efficiency excavated
And accuracy.
(7) association rule mining and the fault diagnosis of native system can asynchronous be carried out, and possesses off-line and excavates and on-line monitoring
Dual characteristics, improve the service efficiency of whole system, enhance the practicality of system.
Accompanying drawing explanation
Fig. 1 is the structural representation of the present embodiment;
Fig. 2 is the schematic flow sheet of the present embodiment;
Fig. 3 is the diagnostic process of intelligent diagnosis system based on metering device fault;
Wherein, 1 is data acquisition and monitoring modular, and 2 is metaevent information generating module, and 3 is fault diagnosis and decision model
Block, 21 is metaevent data pre-processing unit, and 22 is metaevent diagnostic cast call unit, and 31 is diagnostic knowledge base, and 32 for digging
Unit set up by pick model, and 33 is inline diagnosis unit.
Detailed description of the invention
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention
Premised on implement, give detailed embodiment and concrete operating process, but protection scope of the present invention be not limited to
Following embodiment.
As depicted in figs. 1 and 2, for intelligent diagnosis system based on metering device fault, it is divided into three levels: data acquisition
With monitoring modular 1, metaevent information generating module 2 and fault diagnosis and decision-making module 3, it is each responsible for monitoring and diagnoses data
Gather and upload, data process and comprehensive analysis, Rule Extraction and intelligent diagnostics.Data acquisition and monitoring modular 1 are responsible for electric energy
Table and acquisition terminal logout, the ac analog such as voltage, electric current, power, forward and reverse have the information such as capacity of idle power, by design
Frequency acquisition and transmission means be acquired and upload, for upper layer analysis provide data.Metaevent information generating module 2 includes
The metaevent pretreatment unit 21 being sequentially connected with and metaevent diagnostic cast call unit 22, be responsible for data collection layer collection
Data are comprehensively analyzed and pretreatment, mate gathering data according to User Profile information, diagnose mould by metaevent
Type generates metaevent information, and the intelligent diagnostics for upper strata provides information.Fault diagnosis and decision-making module 3 include diagnostic knowledge base
31, mining model sets up unit 32 and inline diagnosis unit 33, is responsible for maintenance and the excavation of correlation rule of diagnostic knowledge base 31,
The metaevent information provided according to information processing layer is associated coupling, it is achieved intelligent fault diagnosis.Due to metering device fault
Diagnosing the highest to the requirement of real-time, the time interval of the excavation of correlation rule can determine according to the cycle of troubleshooting,
Use off-line excavates, and intelligent diagnostics can be to use according to the actual requirements in real time, per diem or carry out in the way of the moon flexibly.
Wherein, the event being used for fault diagnosis is decomposed into single independent event, referred to as " metaevent ".Metering device
Each metaevent of fault diagnosis is represented by I={i1,i2,i3,…,in, i1,i2,i3,…,inFor the attribute of metaevent,
Structure is made up of unique identification (ID), electric energy meter numbering, metaevent information, and its structure and attribute are as shown in table 1.Wherein, only
One property mark is for representing metaevent identity of uniqueness in system;The basic letter of the corresponding metering device of electric energy meter numbering
Breath, for mating of metaevent and other dimensional information;Metaevent information represents the fault message formed after systematic analysis.
Table 1 metaevent structure and attribute
Metaevent unique identification by acquisition system with day for interval be autonomously generated, as system internal fault diagnosis unique
Identity, an only corresponding unique identification of metaevent.Electric energy meter numbering is for determining the basic of metering device self
Information, is associated in conjunction with facility information and stoichiometric point information in marketing and acquisition system data base, can extract multiple dimension
Data excavate, make Result more accurate, more targetedly.As associate device information can extract electric energy meter batch
Number, the information such as electric energy meter manufacturer, for the diagnosis of the Universal Faults features such as the bulk fault of electric energy meter, familial form hidden danger,
Association stoichiometric point essential information may determine that the information such as customer location, electricity capacity, region.Metaevent information is according to prison
The basic foundation for fault diagnosis that measurement information analysis draws, including Time To Event, event end time, diagnosis generation
Time, event title, logout that metaevent information is reported according to stoichiometric point by acquisition system or the electricity of collection, voltage,
The data analysiss such as electric current draw, are the information formed after system is processed.By setting up metaevent diagnostic cast, collection it is
System carries out Macro or mass analysis to data, draws metaevent information.
Metaevent information generating module 2 includes: metaevent data pre-processing unit 21 and metaevent diagnostic cast are set up single
Unit.
The Data Source of intelligent diagnostics metaevent includes the electric energy measurement data in electric energy meter and acquisition terminal, operating condition
The Various types of data such as data and logout, due to by various factors such as Data Source, acquisition quality, misprogramming, data mess codes
Impact, metaevent data exist various noise, affect the quality of Result, therefore, metaevent data are located in advance
The quality of reason will be directly connected to the quality of final result.It is clear that metaevent data pre-processing unit 21 sequentially passes through metaevent data
Reason, metaevent data integration, metaevent data regularization and metaevent data have converted preprocessing process.
Wherein, the purpose of metaevent data scrubbing seeks to fill missing values, smooth noise and identify outlier, correct number
Inconsistent according to.Acquisition system is in the gatherer process of event or data, due to the interference of all kinds of uncertain factors, always produces
More raw missing values, the dependency of specific attribute data is used by this project according to mining process ignores tuple or interpolation arithmetic
Mode carries out shielding or fill corresponding attribute.
Needing to consider two aspects during metaevent data integration, one is the integrated of metaevent structure: the composition letter of metaevent
Breath, from different information sources, needs to carry out integrated by the information of same entity, and this project passes through family number by the collection number of entity
Mate according to the essential information in marketing system, form the metaevent of a structural integrity;Two is first thing of same entity
Part is integrated: the object of metering device on-line monitoring is usually metering device itself or independent platform district, it is therefore desirable to by metaevent root
Being defined according to the scope of object, the data volume that acquisition system obtains is the hugest, if but decompose each monitoring object,
Then data volume and dimension will be substantially reduced, and efficiency and the accuracy of excavation are higher.This project limits monitoring station by terminal number
District, limits the metering device of monitoring, it is achieved the metaevent of same entity scope is integrated by family number and electric energy table number.Except needing to solve
Certainly outside Entity recognition problem, in addition it is also necessary to consider the problems such as data redundancy, owing to metering device fault type ratio is relatively decentralized, Duo Gexin
The information asymmetry of breath system, therefore the content to diagnostic message requires relatively stricter, causes a lot of content to there is implication relation,
And during practical operation, cannot directly delete this redundancy.By the accurate identification to entity, determine each item number that entity is corresponding
According to, extract corresponding attribute according to algorithm requirements during actual excavation and participate in computing, can effectively solve redundancy issue.
Fault diagnosis metaevent data attribute is more, is unfavorable for carrying out and excavates fast and effectively, and data regularization technology is permissible
The reduction being used for obtaining data set represents, it is more much smaller than set of source data, but remains close to keep the integrity of former data, is returning
Excavate more efficient on data set after about, and the analysis result of identical (or almost identical) can be produced.The strategy of hough transformation
Mainly there are dimension reduction, quantity reduction, data compression, discretization and Concept Hierarchies etc., by fault diagnosis result, metaevent believed
The demand of breath carries out hough transformation, detects and deletes uncorrelated, the weak relevant or attribute of redundancy or dimension, examine as carried out equipment fault
Time disconnected, such as the attribute such as user type, decompression defluidization event, overcurrent is uncorrelated with task, can be subtracted by data regularization
Less or delete incoherent attribute, thus reduce the scale of data set.Data regularization generally uses attribute set system of selection, its
Target is to find out minimal attribute set so that data attribute is minimum, on the property set after stipulations for the distortion factor excavating application
Excavate, not only reduce the number of the attribute occurred on discovery mode, and pattern is more readily understood.For often
One diagnostic result RjCorresponding all metaevent Ij={ ij1,ij2,ij3,…,ijn, one can be arranged and carry out hough transformation
Vectorial Αj={ α1,α2,α3,…,αn}Τ, wherein:
Metaevent after reduction is represented by Ij'=IjΑj={ ij1α1,ij2α2,ij3α3,…,ijnαn, it is one and comprises
The sparse metaevent of multiple null attributes, it is achieved the attribute relevant to diagnostic result is remained, and deletes unrelated attribute.
From structure and the attribute of metering device intelligent diagnostics metaevent it can be seen that metaevent comprises Asset Attributes, event
The data of the different dimensions such as record, collection data, data conversion converts the data into the form being suitable for excavating, conventional method
There are smooth, Data generalization, standardization, attribute construction etc., realize the data smoothing of metaevent, generalization and specification by diagnostic cast
Change.
Metaevent diagnostic cast sets up unit for setting up metaevent diagnostic cast and generating metaevent information, and metaevent is examined
Disconnected model can be divided into electric meter fault diagnosis, electricity abnormity diagnosis, voltage x current abnormity diagnosis, exception electrodiagnosis, load extremely
Diagnosis, clock abnormity diagnosis, wiring abnormity diagnosis, expense control abnormity diagnosis amount to 41 intelligent diagnostics of 8 class and analyze model.
Fault diagnosis includes with decision-making module 3: diagnostic knowledge base 31, is connected with metaevent information generating module 2, is used for depositing
Putting historical diagnostic result, diagnostic result includes equipment fault, doubtful stealing, loop exception, electrical network exception, on-site maintenance, misconnection
Line and promise breaking electricity consumption;Mining model sets up unit 32, is connected with diagnostic knowledge base 31, is used for setting up mining model, excavating diagnosis
Knowledge base 31, to extract diagnosis rule and self study fault signature, realizes the intelligent diagnostics of metering device fault then.
The defect elimination result of historical failure or abnormal information and correspondence is preserved by intelligent diagnosis system with unified data structure
Get off, form diagnostic knowledge base 31, as the Extracting Knowledge of consequent malfunction diagnosis.In the excavation of correlation rule, diagnostic knowledge
Transaction database is also in storehouse 31, is the basis of Mining Frequent Itemsets Based, along with scene is abnormal, the continuous of fault confirms and process, examines
Disconnected knowledge base 31 the most constantly expands, and kind is also enriched constantly, the most recurrent abnormal and corresponding diagnosis knot of some knowledge
Fruit is continually strengthened so that its support in mining process and confidence level improve constantly, and become and are easier to be mined
Rule.Fault signature can obtain by excavating fault sample, it is also possible to generates according to existing operating experience.Excavating system
The initial operating stage of system, can conclude some fault signatures, and constantly inspection and adjustment in running, online to adapt to metering
The reality of monitoring.This project is according to metering device failure modes over the years and the statistical analysis of feature, in conjunction with metering device O&M
The experience of defect elimination, generates initial diagnostic knowledge base 31.The basic armrest of knowledge in system initial operating stage, diagnostic knowledge base 31
Flowing mode is safeguarded, according to statistical conditions, the on-the-spot data such as operating condition and expertise of characteristic, sets up at the beginning of one
The knowledge data base begun.On the one hand, the startup of algorithm can be facilitated, initially excavate;On the other hand, can rule of thumb by
Diagnostic knowledge that some are important, that everybody is concerned about joins, it is to avoid miss the diagnosis of this type of information at first.Along with
The continuous operation of system, knowledge data base constantly evolved by study mechanism, perfect, in general, the dimension of diagnostic knowledge base 31
Protect and knowledge accumulation, study restriction and knowledge can be divided into eliminate three steps.
Mining model include basic association mining model, multilamellar association mining model, multidimensional association rule model and
Meta-rule guidance mining model.
As in figure 2 it is shown, the intelligent diagnostics flow process according to the formation of above-mentioned intelligent diagnosis system based on metering device fault can
Be divided into correlation rule to extract, metaevent generates and result three parts of diagnosis, and correlation rule uses offline mode to extract, i.e. intelligence
Diagnostic system is spaced (being defaulted as 1) according to set time and extracts the correlation rule meeting requirement from knowledge data base, as
The diagnostic rule of subsequent time period intelligent diagnostics.Initial rule is generated by the information excavating such as historical data, expertise, along with
The gradual perfection that system is run, the diagnostic knowledge of intelligent diagnosis system accumulation gets more and more, the rule excavated by knowledge data base
Then accuracy is more and more higher.
The main flow of intelligent diagnostics is from the beginning of metaevent generates, and acquisition system is analyzed place to data and the event of collection
Reason, after corresponding data cleansing, forms metaevent by metaevent diagnostic cast, and realizes metaevent by related information
With facility information and the association of stoichiometric point information, provide diagnostic message for result diagnosis.
Metaevent in certain time window is mated by result diagnosis with correlation rule, and provides each diagnostic result
Support and confidence level.Instructing the result after excavating according to meta-rule, form the correlation rule that 7 classes are relatively independent, result is examined
Need to judge respectively according to every the ageing of class diagnostic result time disconnected, the selection to metaevent time window should be according to ageing
Formulating respectively, such as: same metaevent, ageing in equipment fault diagnosis is 1, and timeliness in the diagnosis of doubtful stealing
Property is January.When just running in view of intelligent diagnosis system, for ensureing regular grid DEM and confidence level, to knowledge data base
Excavating and be still not enough to produce enough correlation rules, this flow process also add expert diagnostic system as auxiliary diagnostic tool, logical
Cross expertise and preset weight and the threshold values of diagnostic result judgement of each metaevent correspondence diagnostic result, it is achieved expert
Diagnosis, solves the starting problem of intelligent diagnostics.Expert diagnosis model is as follows:
In formula, EiFor the degree of association that i-th correspondence diagnostic result is corresponding, WijAnd RijIt is respectively i-th diagnostic result corresponding
The weight of jth metaevent and correlation coefficient, initial weight Wij=1, correlation coefficient is as shown in table 2.As 0.8≤EiDuring < 1, sentence
For doubtful exception, sustainable concern, if continuing K (K interval is 3~60, and acquiescence takes 3) day to be then judged to exception;Work as Ei≥1
Time, it is judged to exception, manual confirmation and defect elimination need to be carried out.
Table 2 metaevent and the correlation coefficient of diagnostic result
In the present embodiment this intelligent diagnosis system include data acquisition and monitoring modular 1, metaevent information generating module 2 and
Fault diagnosis and 3 three parts of decision-making module, can not only limit content and object that each layer processes easily, contributes to controlling letter
The complexity that breath processes;Simultaneously as matched rule information to be processed is all same level information, the process of diagnosis is the simplest
Single, and versatility also greatly enhances.Intelligent diagnostics model diagnoses according to metaevent, metaevent defined in the present embodiment
Structure and attribute, and have studied metaevent data scrubbing, integrated, stipulations and the preconditioning technique of conversion, it is used in intelligent diagnostics
Metaevent possess unified form, be that 41 events set up metaevent diagnostic cast respectively on this basis, and according to unit's thing
Diagnostic cast is classified by the character of part, it is simple to during intelligent diagnostics, the guidance of correlation rule is excavated.
Based on metering device fault and the statistics of practical operation situation, diagnostic result is divided into equipment fault, doubtful stealing,
Loop is abnormal waits 7 big classes, simultaneously by metaevent information, equipment essential information and the diagnosis of three dimensions of stoichiometric point essential information
Information carries out layer digging, is the association mining model of a kind of Multilayer multidimensional.In terms of the maintenance of diagnostic knowledge base 31, introduce and learn
Habit mechanism makes knowledge data base constantly evolve, perfect, improve the accuracy excavated.Excavation and fault diagnosis due to correlation rule
Can asynchronous carry out, native system possesses off-line and excavates and on-line monitoring dual characteristics.
Claims (10)
1. an intelligent diagnosis system based on metering device fault, it is characterised in that described system includes:
Data acquisition and monitoring modular, for carrying out the collection of related data according to frequency acquisition and transmission means and upload;
Metaevent information generating module, is connected with monitoring modular with data acquisition, for data acquisition and monitoring modular collection
Related data comprehensively analyze and pretreatment, according to User Profile information to gather related data mate, pass through
Metaevent diagnostic cast generates metaevent information;
Fault diagnosis and decision-making module, be connected with metaevent information generating module, for carrying according to metaevent information generating module
The metaevent information of confession is associated coupling, it is achieved the intelligent diagnostics of metering device fault.
Intelligent diagnosis system based on metering device fault the most according to claim 1, it is characterised in that described dependency number
According to including electric energy meter logout, acquisition terminal logout, ac analog and electric quantity data.
Intelligent diagnosis system based on metering device fault the most according to claim 1, it is characterised in that described metaevent
Including:
Unique identification, is autonomously generated, as the intelligence of metering device fault for interval with day by data acquisition and monitoring modular
The unique identity of diagnosis;
Electric energy meter is numbered, for determining the essential information of metering device self;
Metaevent information, is generated by metaevent diagnostic cast, for substantially depending on as the intelligent diagnostics of metering device fault
According to.
Intelligent diagnosis system based on metering device fault the most according to claim 1, it is characterised in that described metaevent
Information includes Time To Event, event end time, diagnosis generation Time And Event title.
Intelligent diagnosis system based on metering device fault the most according to claim 1, it is characterised in that described metaevent
Information generating module includes:
Metaevent data pre-processing unit, is connected with monitoring modular with data acquisition, for adopting data acquisition with monitoring modular
The related data of collection carries out pretreatment;
Metaevent diagnostic cast call unit, connects with decision-making module with metaevent data pre-processing unit and fault diagnosis respectively
Connect, be used for calling metaevent diagnostic cast to generate metaevent information.
Intelligent diagnosis system based on metering device fault the most according to claim 5, it is characterised in that described metaevent
Data pre-processing unit carries out pretreatment and specifically includes the following step the related data of data acquisition Yu monitoring modular collection:
1) utilize the mode ignoring tuple or interpolation arithmetic that the related data of data acquisition with monitoring modular collection is shielded
Or fill corresponding attribute, complete metaevent data scrubbing;
2) metaevent of same entity is carried out metaevent data integration;
3) according to the intelligent diagnostics result of metering device fault, the demand of metaevent information is carried out metaevent data regularization;
4) metaevent data conversion is carried out by the method for smooth, Data generalization, data normalization and attribute construction.
Intelligent diagnosis system based on metering device fault the most according to claim 5, it is characterised in that described metaevent
Diagnostic cast includes that electric meter fault diagnostic cast, electricity abnormity diagnosis model, voltage x current abnormity diagnosis model, abnormal electricity consumption are examined
Disconnected model, load abnormity diagnosis model, clock abnormity diagnosis model, wiring abnormity diagnosis model and take and control abnormity diagnosis model.
Intelligent diagnosis system based on metering device fault the most according to claim 1, it is characterised in that described fault is examined
Break and include with decision-making module:
Diagnostic knowledge base, for depositing the diagnostic result of history;
Mining model sets up unit, is connected with diagnostic knowledge base, is used for setting up mining model, and is known diagnosis by mining model
The diagnostic result knowing the history in storehouse excavates, and then extracts diagnosis rule;
Inline diagnosis unit, sets up unit, metaevent information generating module and diagnostic knowledge base respectively and is connected with mining model, uses
When online, metaevent information and diagnosis rule are mated, realize the intelligence of metering device fault according to the result of coupling
Diagnosis, and diagnostic result is stored in diagnostic knowledge base.
Intelligent diagnosis system based on metering device fault the most according to claim 8, it is characterised in that described diagnosis is tied
Fruit includes equipment fault, doubtful stealing, loop exception, electrical network exception, on-site maintenance, error connection and promise breaking electricity consumption.
Intelligent diagnosis system based on metering device fault the most according to claim 8, it is characterised in that described excavation
Model includes that basic association mining model, multilamellar association mining model, multidimensional association rule model and meta-rule guidance are dug
Pick model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610586569.XA CN106199494A (en) | 2016-07-25 | 2016-07-25 | A kind of intelligent diagnosis system based on metering device fault |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610586569.XA CN106199494A (en) | 2016-07-25 | 2016-07-25 | A kind of intelligent diagnosis system based on metering device fault |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106199494A true CN106199494A (en) | 2016-12-07 |
Family
ID=57491761
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610586569.XA Pending CN106199494A (en) | 2016-07-25 | 2016-07-25 | A kind of intelligent diagnosis system based on metering device fault |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106199494A (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106842106A (en) * | 2017-02-23 | 2017-06-13 | 广东电网有限责任公司茂名供电局 | Electrical energy meter fault Forecasting Methodology and device |
CN107392457A (en) * | 2017-07-17 | 2017-11-24 | 贵州电网有限责任公司电力科学研究院 | Metering automation intelligence operational system based on line loss anomaly analysis |
CN107703383A (en) * | 2017-09-21 | 2018-02-16 | 国网上海市电力公司 | A kind of method for building up of information acquisition system fault diagnosis knowledge base |
CN107741522A (en) * | 2017-09-21 | 2018-02-27 | 国网上海市电力公司 | A kind of electrical energy meter fault remote diagnosis method |
CN108171960A (en) * | 2017-12-14 | 2018-06-15 | 深圳市科陆电子科技股份有限公司 | A kind of self-diagnosing method and system of comprehensive energy management platform metering device exception |
CN108363364A (en) * | 2017-12-29 | 2018-08-03 | 武汉武钢众鹏信息***有限公司 | A kind of alarm method based on the driving of industrial big data |
CN109308766A (en) * | 2018-08-21 | 2019-02-05 | 国网辽宁省电力有限公司电力科学研究院 | A kind of remote bill control switching on executes the fault location system and processing method of failure |
CN109407039A (en) * | 2018-11-13 | 2019-03-01 | 许继集团有限公司 | Intelligent electric meter and its system, self-diagnosing method and fault detection method |
CN109885978A (en) * | 2019-03-27 | 2019-06-14 | 航天恒星科技有限公司 | A kind of remote sensing earth station fault diagnosis system and method |
CN109976318A (en) * | 2019-04-28 | 2019-07-05 | 郑州万特电气股份有限公司 | Electrical energy measurement fault diagnosis Internet-based checks expert system |
CN109993949A (en) * | 2019-04-14 | 2019-07-09 | 杭州拓深科技有限公司 | A kind of security against fire detection method based on Multi-sensor Fusion |
CN110599066A (en) * | 2019-09-23 | 2019-12-20 | 海南电网有限责任公司 | Power disaster monitoring method and device, computer equipment and storage medium |
CN110968703A (en) * | 2019-12-06 | 2020-04-07 | 国网天津市电力公司电力科学研究院 | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm |
CN111985558A (en) * | 2020-08-19 | 2020-11-24 | 安徽蓝杰鑫信息科技有限公司 | Electric energy meter abnormity diagnosis method and system |
CN112084220A (en) * | 2020-09-21 | 2020-12-15 | 国网重庆市电力公司营销服务中心 | Method and device for diagnosing abnormity of electric energy metering device and readable storage medium |
CN114236448A (en) * | 2021-11-23 | 2022-03-25 | 国网山东省电力公司日照供电公司 | Metering device troubleshooting system based on big data |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103499804A (en) * | 2013-09-24 | 2014-01-08 | 国家电网公司 | Anomaly analysis system of electric energy metering device and analysis method thereof |
CN104363106A (en) * | 2014-10-09 | 2015-02-18 | 国网辽宁省电力有限公司信息通信分公司 | Electric power information communication fault early warning analysis method based on big-data technique |
CN104656050A (en) * | 2012-10-29 | 2015-05-27 | 江苏省电力公司常州供电公司 | Manpower-saving remote monitoring and diagnosing method for electric energy metering device |
CN104679828A (en) * | 2015-01-19 | 2015-06-03 | 云南电力调度控制中心 | Rules-based intelligent system for grid fault diagnosis |
CN204719225U (en) * | 2015-07-08 | 2015-10-21 | 国家电网公司 | A kind of Watthour meter remote monitors check system automatically |
-
2016
- 2016-07-25 CN CN201610586569.XA patent/CN106199494A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104656050A (en) * | 2012-10-29 | 2015-05-27 | 江苏省电力公司常州供电公司 | Manpower-saving remote monitoring and diagnosing method for electric energy metering device |
CN103499804A (en) * | 2013-09-24 | 2014-01-08 | 国家电网公司 | Anomaly analysis system of electric energy metering device and analysis method thereof |
CN104363106A (en) * | 2014-10-09 | 2015-02-18 | 国网辽宁省电力有限公司信息通信分公司 | Electric power information communication fault early warning analysis method based on big-data technique |
CN104679828A (en) * | 2015-01-19 | 2015-06-03 | 云南电力调度控制中心 | Rules-based intelligent system for grid fault diagnosis |
CN204719225U (en) * | 2015-07-08 | 2015-10-21 | 国家电网公司 | A kind of Watthour meter remote monitors check system automatically |
Non-Patent Citations (1)
Title |
---|
李再华 等: "基于特征挖掘的电网故障诊断方法", 《中国电机工程学报》 * |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106842106A (en) * | 2017-02-23 | 2017-06-13 | 广东电网有限责任公司茂名供电局 | Electrical energy meter fault Forecasting Methodology and device |
CN107392457A (en) * | 2017-07-17 | 2017-11-24 | 贵州电网有限责任公司电力科学研究院 | Metering automation intelligence operational system based on line loss anomaly analysis |
CN107703383A (en) * | 2017-09-21 | 2018-02-16 | 国网上海市电力公司 | A kind of method for building up of information acquisition system fault diagnosis knowledge base |
CN107741522A (en) * | 2017-09-21 | 2018-02-27 | 国网上海市电力公司 | A kind of electrical energy meter fault remote diagnosis method |
CN108171960A (en) * | 2017-12-14 | 2018-06-15 | 深圳市科陆电子科技股份有限公司 | A kind of self-diagnosing method and system of comprehensive energy management platform metering device exception |
CN108363364A (en) * | 2017-12-29 | 2018-08-03 | 武汉武钢众鹏信息***有限公司 | A kind of alarm method based on the driving of industrial big data |
CN109308766B (en) * | 2018-08-21 | 2021-06-18 | 国网辽宁省电力有限公司电力科学研究院 | Fault positioning system and processing method for remote fee control switching on and off execution failure |
CN109308766A (en) * | 2018-08-21 | 2019-02-05 | 国网辽宁省电力有限公司电力科学研究院 | A kind of remote bill control switching on executes the fault location system and processing method of failure |
CN109407039A (en) * | 2018-11-13 | 2019-03-01 | 许继集团有限公司 | Intelligent electric meter and its system, self-diagnosing method and fault detection method |
CN109885978A (en) * | 2019-03-27 | 2019-06-14 | 航天恒星科技有限公司 | A kind of remote sensing earth station fault diagnosis system and method |
CN109885978B (en) * | 2019-03-27 | 2023-05-16 | 航天恒星科技有限公司 | Remote sensing ground station fault diagnosis system and method |
CN109993949B (en) * | 2019-04-14 | 2021-06-29 | 杭州拓深科技有限公司 | Fire safety detection method based on multi-sensor fusion |
CN109993949A (en) * | 2019-04-14 | 2019-07-09 | 杭州拓深科技有限公司 | A kind of security against fire detection method based on Multi-sensor Fusion |
CN109976318B (en) * | 2019-04-28 | 2021-07-02 | 郑州万特电气股份有限公司 | Internet-based electric energy metering fault diagnosis and troubleshooting expert system |
CN109976318A (en) * | 2019-04-28 | 2019-07-05 | 郑州万特电气股份有限公司 | Electrical energy measurement fault diagnosis Internet-based checks expert system |
CN110599066A (en) * | 2019-09-23 | 2019-12-20 | 海南电网有限责任公司 | Power disaster monitoring method and device, computer equipment and storage medium |
CN110968703A (en) * | 2019-12-06 | 2020-04-07 | 国网天津市电力公司电力科学研究院 | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm |
CN110968703B (en) * | 2019-12-06 | 2024-01-02 | 国网天津市电力公司电力科学研究院 | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm |
CN111985558A (en) * | 2020-08-19 | 2020-11-24 | 安徽蓝杰鑫信息科技有限公司 | Electric energy meter abnormity diagnosis method and system |
CN112084220A (en) * | 2020-09-21 | 2020-12-15 | 国网重庆市电力公司营销服务中心 | Method and device for diagnosing abnormity of electric energy metering device and readable storage medium |
CN112084220B (en) * | 2020-09-21 | 2024-05-14 | 国网重庆市电力公司营销服务中心 | Abnormality diagnosis method and device for electric energy metering device and readable storage medium |
CN114236448A (en) * | 2021-11-23 | 2022-03-25 | 国网山东省电力公司日照供电公司 | Metering device troubleshooting system based on big data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106199494A (en) | A kind of intelligent diagnosis system based on metering device fault | |
CN106557991B (en) | Voltage monitoring data platform | |
CN102130783B (en) | Intelligent alarm monitoring method of neural network | |
CN110674189B (en) | Method for monitoring secondary state and positioning fault of intelligent substation | |
CN103996077B (en) | A kind of electrical equipment malfunction Forecasting Methodology based on multidimensional time-series | |
CN108051709A (en) | Transformer state online evaluation analysis method based on artificial intelligence technology | |
CN102098175B (en) | Alarm association rule obtaining method of mobile internet | |
CN108345282B (en) | Pump station operation abnormity diagnosis method and system based on artificial intelligence | |
CN103765469B (en) | Dynamic infrastructure administration system | |
CN106301522A (en) | The Visual method of fault diagnosis of Remote Sensing Ground Station data receiver task and system | |
CN110766277A (en) | Health assessment and diagnosis system and mobile terminal for nuclear industry field | |
CN103902816A (en) | Electrification detection data processing method based on data mining technology | |
CN107085653A (en) | A kind of anti-electricity-theft real-time diagnosis method of data-driven | |
CN110296124A (en) | Remote failure diagnosis system and method based on expert system | |
CN111507013A (en) | Line loss fault positioning implementation method for power system | |
CN105930424A (en) | Method for realizing online asynchronous acquisition and intelligent mining of power distribution network data | |
CN106325162A (en) | Embedded intelligent electromechanical equipment condition monitoring system | |
CN109840864A (en) | A kind of building energy consumption management system | |
CN111766512A (en) | Generator fault maintenance system and method | |
CN115718472A (en) | Fault scanning and diagnosing method for hydroelectric generating set | |
CN112488502A (en) | Standard water resource management integrated management and control platform | |
CN108230183A (en) | A kind of processing method of the grid equipment various dimensions comprehensive warning based on time scale measurement | |
CN107741522A (en) | A kind of electrical energy meter fault remote diagnosis method | |
CN110210634A (en) | Based on big data driving Civil Aviation Engine fault diagnosis with it is health management system arranged | |
CN114942934A (en) | Intelligent park multi-dimensional space-time simulation platform based on digital twin |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161207 |
|
RJ01 | Rejection of invention patent application after publication |