CN110503131A - Wind-driven generator health monitoring systems based on big data analysis - Google Patents
Wind-driven generator health monitoring systems based on big data analysis Download PDFInfo
- Publication number
- CN110503131A CN110503131A CN201910658432.4A CN201910658432A CN110503131A CN 110503131 A CN110503131 A CN 110503131A CN 201910658432 A CN201910658432 A CN 201910658432A CN 110503131 A CN110503131 A CN 110503131A
- Authority
- CN
- China
- Prior art keywords
- wind
- big data
- driven generator
- data analysis
- module
- 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
- 238000007405 data analysis Methods 0.000 title claims abstract description 43
- 238000012544 monitoring process Methods 0.000 title claims abstract description 29
- 230000036541 health Effects 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 28
- 239000000843 powder Substances 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 238000012423 maintenance Methods 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000011160 research Methods 0.000 claims description 5
- 238000012706 support-vector machine Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000005065 mining Methods 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 238000003780 insertion Methods 0.000 claims description 2
- 230000037431 insertion Effects 0.000 claims description 2
- 230000006855 networking Effects 0.000 claims description 2
- 239000013307 optical fiber Substances 0.000 claims description 2
- 238000007619 statistical method Methods 0.000 claims description 2
- 238000003012 network analysis Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 15
- 238000007726 management method Methods 0.000 description 5
- 229910000831 Steel Inorganic materials 0.000 description 3
- 239000010959 steel Substances 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 230000008140 language development Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/288—Entity relationship models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Tourism & Hospitality (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses the wind-driven generator health monitoring systems based on big data analysis, which includes five modules, is respectively: characteristic extracting module, failure modes module, black powder rule base, big data analysis platform and system interface module.Aiming at the problem that continuous feature, the method for diagnosing faults based on EMD algorithm and data branch mailbox will be used, first pass through EMD algorithm and decompose to obtain IMF signal, extracted its amplitude domain parameter and hold sign as feature vector, and it is input in SVM and carries out failure modes, and verified by emulation early period and real data.Using big data analysis, based on open source big data platform Spark, the realization of Distributed and parallel structure has been carried out to EMD and statistics description feature operator respectively.The present invention carries out platform-type service to fan condition using interconnected network mode, has shared blower warning information, by network analysis early warning, substantially increases the speed that failure actively monitors, guarantees each Fan Equipment warning information real-time and accuracy.
Description
Technical field
The present invention relates to a kind of wind-driven generator health monitoring systems based on big data analysis, belong to wind-power electricity generation software
Monitoring technical field.
Background technique
In the process of running, long-time discontinuity will lead to its fatigue rupture, while dust storm, humidity to wind-driven generator
Etc. adverse circumstances also extreme influence unit durability.Main shaft bearing is easy to produce the failures such as abrasion, casting erosion, is typical equipment event
Hinder diagnosis problem；Generator magnet steel can also fall off, and show as typical complicated uncertain problem.In wind-driven generator, bearing,
Generator is as transmission system.Important component, by operation maintenance personnel, the very big concern of experts and scholars.Study wind-driven generator
The fault signature extraction process of main shaft bearing, falls off to generator magnet steel and carries out correlation analysis, is under industrial big data background
The typical case of data mining greatly helps business personnel to increase generator component mechanism the research of the problem and recognizes,
Solves the problems, such as the blind spot in wind-driven generator operational process to the monitoring.
By taking bearing as an example, the multiple wind fields in the statistics north are known, what cascading failure caused by bearing accounted for total failare 30% arrives
45%.In general, the O&M expenditure of Wind turbines has accounted for 10% to the 15% of unit total income.One 5000 dollars of replacement
Bearing, required ultimate cost be 25000 dollars, lose it is huge.Therefore, the shape of the wind turbine components such as bearing is carried out
State monitoring and fault diagnosis improves wind field economic benefit and has great importance for reducing unit failure rate.
It is the most common failure mode that the main shaft bearing failure and magnet steel of Wind turbines, which fall off, and having respectively represented has continuously
The typical problem of two kinds of data of feature and discrete features analysis.
With the development of technology of Internet of things, manufacturing industry has accumulated a large amount of isomeric data in process of production, for data
Storage tape carrys out great challenge.Around industrial big data, industry begins to focus on its abundant value contained, carries out Wang Ye from multi-angle
The copper of big data is examined and is studied.On this basis, being monitored by big data to wind-driven generator health is to have platform and reason
By support, has feasibility.
Summary of the invention
It is an object of the invention to the distributed parallel characteristics using big data analysis method, supervise to wind-driven generator health
The related data of survey is analyzed.
Aiming at the problem that continuous feature, the method for diagnosing faults based on EMD algorithm and data branch mailbox will be used, first passed through
EMD algorithm decomposes to obtain IMF signal, extracts its amplitude domain parameter and holds sign as feature vector, and is input in SVM and carries out failure point
Class, and verified by emulation early period and real data.
Using big data analysis, based on open source big data platform Spark, respectively to EMD and statistics description feature operator into
The realization of Distributed and parallel structure is gone.
Aiming at the problem that discrete features, correlation research analysis has been carried out from multi-angle with the method for big data analysis,
Reliable basis is provided for the health monitoring of wind-driven generator.
The technical solution adopted by the present invention is the wind-driven generator health monitoring systems based on big data analysis, the system packet
Five modules are included, are respectively: characteristic extracting module, failure modes module, black powder rule base, big data analysis platform and being
System interface module.
Specific module contents is as follows:
Characteristic extracting module uses the method for diagnosing faults based on EMD and data branch mailbox, first passes through EMD algorithm and decomposes to obtain
IMF signal extracts its amplitude domain parameter and holds sign as feature vector；
Failure modes module carries out fault reconstruction, SVM model foundation using SVM.
Black powder rule base carries out correlation analysis from multi-angle with the method for big data analysis；Derive black powder
Relevant rule establishes rule base, for the prediction of black powder phenomenon, provides reliable basis for wind-driven generator health monitoring.
Big data analysis platform is established using Spark distributed parallel frame and Map Reduce programming model；
System interface module is for data acquisition and report generation etc..
The related data of wind-driven generator health monitoring is connect by system interface module with characteristic extracting module, failure point
Generic module carries out classification to the associated arguments that characteristic extracting module is extracted and is sent to big data point by black powder rule base
It analyses in platform.
Big data analysis platform includes: device management module, wind field management module, INDEX MANAGEMENT module, statistical analysis mould
Block, regular library module and six part of system setup module.
Big data analysis platform is taken by two mutual standby application/database servers, a web server, several acquisitions
Business device and a GIS server distributed deployment are constituted, and equipment interconnection is completed by two networking switch, and application database takes
Business device accesses disk arrays and tape library by two optical fiber switch.There are two fire prevention between big data analysis platform and internet
Wall realizes network security policy.System acquisition module is deployed on wind-driven generator, and by network insertion internet, access is big
Data Analysis Platform.Monitor terminal is connected through the internet to firewall access big data analysis platform.
The institute of big data analysis platform is functional to carry out offer service by B/S structure.
Characteristic extracting module
The method for diagnosing faults based on EMD and data branch mailbox will be used, first passes through EMD algorithm and decompose to obtain IMF signal, mention
Its amplitude domain parameter is taken to hold sign as feature vector.
The feature extraction algorithm with data branch mailbox is decomposed based on EMD, the system being deployed on wind-driven generator is adopted first
The original signal for collecting module passes through denoising, then carries out empirical mode decomposition (Empirical Mode
Decomposition, abbreviation EMD, obtains MFQntrinsicModeFunction, abbreviation IMF) signal.H layers of MF letter before choosing
Number, the spy of four feature composition wind-driven generators such as mean value, root-mean-square value, standard deviation, RMSE (root-mean-square error) is extracted respectively
Levy vector matrix.To the smooth abnormal point of eigenvectors matrix maintenance data branch mailbox method, support vector machines is inputted
(SupportVector Machine, abbreviation SVM) carries out classification accuracy verifying.
Failure modes module
Fault reconstruction, SVM model foundation are carried out using SVM.
SVM training process: training sample is mapped to high dimensional feature by multinomial (or other) kernel function of the multiple side of selection
Space.The optimal separating hyper plane of feature samples of all categories Yu other feature samples is found out in sample characteristics space using SVM,
The supporting vector collection and its corresponding VC confidence level, formation for obtaining representing each sample characteristics judge the differentiation letter of each feature classification
Number.
SVM judging process: pixel to be sorted is mapped in feature space by kernel function effect, as discriminant function
Input obtains the result that two classes can divide using classification decision function.
The effect of kernel function is by each pixel of image, and conversion is input to the supporting vector collection of each training sample conversion and VC can
In the discriminant function that reliability is formed, classify.
Black powder rule base
Correlation research analysis has been carried out from multi-angle with the method for big data analysis；Derive the relevant rule of black powder
Rule base is then established, for the prediction of black powder phenomenon, provides reliable basis for the health monitoring of wind-driven generator.
Big data analysis platform
For the fault type with discrete features, has the characteristics that the black powder of complicated uncertainty by furtheing investigate
Problem designs the rule mining algorithms of closed loop, asks for black powder by calling the correlation rule operator in the library SparkMLib
Topic establishes the rule base based on current big data, convenient for the monitoring of wind-driven generator health monitoring situation.
Modeling analysis is carried out using ELK tool.ELK is the abbreviation of three open source softwares, is respectively indicated:
Elasticsearch, Logstash, Kibana.A FileBeat is increased newly, it is the log collection processing an of lightweight
Tool (Agent), Filebeat takes up less resources, and is suitable for being transferred to Logstash after collecting log on each server,
This tool is also recommended by official.
Elasticsearch is an open source distributed search engine, provides collection, analysis, storing data three zones.
Logstash be the collection for log, analysis, filtering log tool, support a large amount of data acquiring mode.
Working method is c/s framework, and the end client is mounted on the host for needing collector journal, and server is responsible at end each section that will be received
Point log is filtered, modify etc. to operate is being sent to elasticsearch up together.
Kibana is the web interface of log analysis close friend that Logstash and ElasticSearch are provided, help to summarize,
Analysis and search significant data log.
System interface module
System provides wind-driven generator data acquisition interface module, and to wind-driven generator operation data, wind-driven generator is accused
Alert data are acquired, parse, are put in storage.
By Reports module to the operating condition of wind-driven generator, the alarm situation of wind-driven generator carries out statistics and generates report
Table.
Compared with prior art, the present invention carries out platform-type service to fan condition using interconnected network mode, shares
Blower warning information substantially increases the speed of failure actively monitoring by network analysis early warning.Man power and material's money is saved
Source has reached expected requirement.Such as: equipment early warning, wind field management, INDEX MANAGEMENT valid data are comprehensive, substantially reduce correlation
The working strength of practitioner, for improving efficiency, service level have great help.
By Internet of Things mode, fan condition, equipment early warning etc. can be shown in real time and calculated.It can be effective
Make an announcement to equipment early warning, under the action of Internet of Things, internet, guarantee each Fan Equipment warning information real-time
And accuracy.
Platform uses static instruction html, and simple but powerful java language development.Since java language is good at
Quickly exploitation, mysql is the database freely increased income, so the development cost and maintenance cost on entire backstage can be reduced.
Detailed description of the invention
Fig. 1 is present system structure chart.
Fig. 2 is big data analysis platform structure schematic diagram.
Fig. 3 is system architecture diagram.
Specific embodiment
Below in conjunction with drawings and examples, the present invention is described in detail.
As shown in Figure 1-3, the technical solution adopted by the present invention is the wind-driven generator health monitoring based on big data analysis
System, the system include five modules, are respectively: characteristic extracting module, failure modes module, black powder rule base, big number
According to analysis platform and system interface module.
Characteristic extracting module
The method for diagnosing faults based on EMD and data branch mailbox will be used, first passes through EMD algorithm and decompose to obtain IMF signal, mention
Its amplitude domain parameter is taken to hold sign as feature vector.
The feature extraction algorithm with data branch mailbox is decomposed based on EMD.First to original signal pass through denoising, then into
(Empirical Mode Decomposition, abbreviation EMD are obtained row empirical mode decomposition
MFQntrinsicModeFunction, abbreviation IMF) signal.H layers of MF signal before choosing extract mean value, root-mean-square value, mark respectively
Four feature composition characteristic vector matrixs such as the poor, RMSE (root-mean-square error) of standard.To eigenvectors matrix maintenance data branch mailbox side
The smooth abnormal point of method, input support vector machines (Support Vector Machine, abbreviation SVM) carry out classification accuracy and test
Card.
Failure modes module
Fault reconstruction, SVM model foundation are carried out using SVM.
SVM training process:
Multinomial (or other) kernel function of the multiple side of selection, is mapped to high-dimensional feature space for training sample.Utilize SVM
The optimal separating hyper plane that feature samples of all categories Yu other feature samples are found out in sample characteristics space obtains representing various kinds
The supporting vector collection of eigen and its corresponding VC confidence level form the discriminant function for judging each feature classification.
SVM judging process:
Pixel to be sorted in image is mapped in feature space by kernel function effect, as the input of discriminant function,
The result that two classes can divide is obtained using classification decision function.
The effect of kernel function is by each pixel of image, and conversion is input to the supporting vector collection and VC confidence level of each sample conversion
In the discriminant function of formation, classify.
Black powder rule base
Correlation research analysis has been carried out from multi-angle with the method for big data analysis；Derive the relevant rule of black powder
Rule base is then established, for the prediction of black powder phenomenon, provides reliable basis for the health monitoring of wind-driven generator.
Big data analysis platform
For the fault type with discrete features, has the characteristics that the black powder of complicated uncertainty by furtheing investigate
Problem designs the rule mining algorithms of closed loop, asks for black powder by calling the correlation rule operator in the library SparkMLib
Topic establishes the rule base based on current big data, convenient for the monitoring of wind-driven generator health monitoring situation.
Modeling analysis is carried out using ELK tool.ELK is the abbreviation of three open source softwares, is respectively indicated:
Elasticsearch, Logstash, Kibana, they are all open source softwares.A FileBeat is increased newly, it is one light
The log collection handling implement (Agent) of magnitude, Filebeat takes up less resources, and is suitable for collecting log on each server
After be transferred to Logstash, this tool is also recommended by official.
Elasticsearch is an open source distributed search engine, provides collection, analysis, storing data three zones.It
The characteristics of have: distributed, zero configuration, it is automatic to find, index auto plate separation, index copy mechanism, restful style interface is more
Data source, automatic search overhead etc..
Logstash be primarily used to the collection of log, analysis, filtering log tool, support a large amount of data recipient
Formula.General work mode is c/s framework, and the end client is mounted on the host for needing collector journal, and server is responsible at end to receive
To each node log the operation such as be filtered, modify and go being sent on elasticsearch together.
Kibana is also an open source and free tool, and Kibana can mention for Logstash and ElasticSearch
The web interface of the log analysis close friend of confession, can help to summarize, analyze and search for significant data log.
System interface module
System provides wind-driven generator data acquisition interface module, and to wind-driven generator operation data, wind-driven generator is accused
Alert data are acquired, and are parsed, storage.
By Reports module, the operating condition of wind-driven generator, the alarm situation of wind-driven generator can be counted
Generate report.
Embodiment
The use-case of alarm inquiry is realized
User can be by input blower or the search condition of wind field information, to retrieve the wind field monitored required for oneself
Or blower.This function supports user using keywords such as blower number, blower regions to retrieve to blower alarm.With
After family uses this function, all blower activity warning information for meeting user search condition of user will be returned to.
The entity that system is related to includes: user, blower, alarm, region, grouping, rule, index etc..
Association attributes and the connection of each entity are shown below by way of each entity attributes figure and totality E-R figure.
Database logic structure is as follows
1) user's table: (user name, password, role).Major key in user's log form is user name
1 user's table of table
2) blower table: (blower ID, brand, model, blower number, affiliated area, affiliated group, owning user) blower table
Major key is ID:
2 blower table of table
3) warning watch (Alarm ID, blower ID, alarm level, alarm name, alarm description, alarm time of origin, alarm knot
The beam time) major key in warning watch is Alarm ID；
3 warning watch of table
| Field name | Field type | Field name | Remarks |
| Id | Char(10) | Alarm ID | Major key |
| Wind_id | Int | Blower ID | |
| Grade | VarChar(20) | Alarm level | |
| Name | VarChar(20) | Alarm name | |
| Status | VarChar(20) | Alarm description | |
| Start_time | Date | Alert time of origin | |
| End_time | Date | Alert the end time |
4) region table: (region ID, zone name, affiliated group, settling time, responsible person)；Major key in the table table of region is
Region ID:
4 region table of table
| Field name | Field type | Field name | Remarks |
| Id | Char(10) | Region ID | Major key |
| Name | VarChar(12) | Zone name | |
| Group_id | VarChar(12) | Affiliated group | |
| time | Date | Settling time | |
| manager | VarChar(20) | Responsible person |
5) grouping sheet: (region ID, zone name, settling time, responsible person)；Major key in grouping sheet is packet ID:
5 grouping sheet of table
6) rule list: (rule ID, rule name, blower ID, rule declaration, settling time, responsible person)；In rule list
Major key is rule ID:
6 rule list of table
| Field name | Field type | Field name | Remarks |
| Id | Char(10) | Rule ID | Major key |
| Name | VarChar(12) | Rule name | |
| Wind_id | VarChar(12) | Blower ID | |
| status | VarChar(20) | Rule declaration | |
| time | Date | Settling time | |
| manager | VarChar(20) | Responsible person |
7) index table: (index ID, index name, blower ID, index value, addition time, responsible person)；Major key in index
It is index ID:
7 index table of table
| Field name | Field type | Field name | Remarks |
| Id | Char(10) | Index ID | Major key |
| Name | VarChar(12) | Index name | |
| Wind_id | VarChar(12) | Blower ID | |
| Value | VarChar(20) | Index value | |
| time | Date | Add the time | |
| manager | VarChar(20) | Responsible person |
Claims (6)
1. the wind-driven generator health monitoring systems based on big data analysis, it is characterised in that: the system includes five modules, point
It is not: characteristic extracting module, failure modes module, black powder rule base, big data analysis platform and system interface module；
Specific module contents is as follows:
Characteristic extracting module uses the method for diagnosing faults based on EMD and data branch mailbox, first passes through EMD algorithm and decomposes to obtain IMF
Signal extracts its amplitude domain parameter and holds sign as feature vector；
Failure modes module carries out fault reconstruction, SVM model foundation using SVM；
Black powder rule base carries out correlation analysis from multi-angle with the method for big data analysis；It is related to derive black powder
Rule establish rule base, for the prediction of black powder phenomenon, provide reliable basis for wind-driven generator health monitoring；
Big data analysis platform is established using Spark distributed parallel frame and Map Reduce programming model；
System interface module is acquired for data and report generation；
The related data of wind-driven generator health monitoring is connect by system interface module with characteristic extracting module, failure modes mould
Block carries out classification and is sent to big data analysis by black powder rule base to put down to the associated arguments that characteristic extracting module is extracted
In platform；
Big data analysis platform includes: device management module, wind field management module, INDEX MANAGEMENT module, statistical analysis module, rule
Then six part of library module and system setup module；
Big data analysis platform is by two mutual standby application/database servers, a web servers, several acquisition servers
It is constituted with a GIS server distributed deployment, equipment interconnection is completed by two networking switch, application database server
Disk array and tape library are accessed by two optical fiber switch；There are two firewalls between big data analysis platform and internet,
Realize network security policy；System acquisition module is deployed on wind-driven generator, by network insertion internet, accesses big data
Analysis platform；Monitor terminal is connected through the internet to firewall access big data analysis platform；
The institute of big data analysis platform is functional to carry out offer service by B/S structure.
2. the wind-driven generator health monitoring systems according to claim 1 based on big data analysis, it is characterised in that: special
It levies in extraction module,
The method for diagnosing faults based on EMD and data branch mailbox will be used, first passes through EMD algorithm and decompose to obtain IMF signal, extract it
Amplitude domain parameter holds sign as feature vector；
The feature extraction algorithm with data branch mailbox is decomposed based on EMD, first to the system acquisition mould being deployed on wind-driven generator
The original signal of block passes through denoising, then carries out empirical mode decomposition EM, obtains IMF signal；H layers of MF signal before choosing,
The eigenvectors matrix that tetra- mean value, root-mean-square value, standard deviation, RMSE features form wind-driven generator is extracted respectively；To feature
The smooth abnormal point of vector matrix maintenance data branch mailbox method, input support vector machines carry out classification accuracy verifying.
3. the wind-driven generator health monitoring systems according to claim 1 based on big data analysis, it is characterised in that: therefore
Hinder in categorization module,
Fault reconstruction, SVM model foundation are carried out using SVM；
SVM training process: training sample is mapped to high-dimensional feature space by the Polynomial kernel function of the multiple side of selection；Utilize SVM
The optimal separating hyper plane that feature samples of all categories Yu other feature samples are found out in sample characteristics space obtains representing various kinds
The supporting vector collection of eigen and its corresponding VC confidence level form the discriminant function for judging each feature classification；
SVM judging process: pixel to be sorted is mapped in feature space by kernel function effect, as the defeated of discriminant function
Enter, obtains the result that two classes can divide using classification decision function；
The effect of kernel function is by each pixel of image, and conversion is input to the supporting vector collection and VC confidence level of each training sample conversion
In the discriminant function of formation, classify.
4. the wind-driven generator health monitoring systems according to claim 1 based on big data analysis, it is characterised in that: black
In color powder rule base,
Correlation research analysis has been carried out from multi-angle with the method for big data analysis；The relevant rule of black powder is derived to build
Vertical rule base provides reliable basis for the prediction of black powder phenomenon for the health monitoring of wind-driven generator.
5. the wind-driven generator health monitoring systems according to claim 1 based on big data analysis, it is characterised in that: big
In Data Analysis Platform,
For the fault type with discrete features, have the characteristics that the black powder of complicated uncertainty is asked by furtheing investigate
Topic designs the rule mining algorithms of closed loop, is black powder problem by calling the correlation rule operator in the library SparkMLib
The rule base based on current big data is established, convenient for the monitoring of wind-driven generator health monitoring situation；
Modeling analysis is carried out using ELK tool；ELK is the abbreviation of three open source softwares, is respectively indicated: Elasticsearch,
Logstash, Kibana；A FileBeat is increased newly, it is the log collection handling implement an of lightweight, Filebeat
It takes up less resources, is suitable for being transferred to Logstash after collecting log on each server, this tool is also recommended by official；
Elasticsearch is an open source distributed search engine, provides collection, analysis, storing data three zones；
Logstash be the collection for log, analysis, filtering log tool, support a large amount of data acquiring mode；Work
Mode is c/s framework, and the end client is mounted on the host for needing collector journal, and server is responsible at end each node day that will be received
Will is filtered, modifies operation is being sent to elasticsearch up together；
Kibana is the web interface for the log analysis close friend that Logstash and ElasticSearch are provided, and help summarizes, analyzes
With search significant data log.
6. the wind-driven generator health monitoring systems according to claim 1 based on big data analysis, it is characterised in that: be
System interface module
System provides wind-driven generator data acquisition interface module, and to wind-driven generator operation data, wind-driven generator alerts number
According to being acquired, parse, be put in storage；
By Reports module to the operating condition of wind-driven generator, the alarm situation of wind-driven generator carries out statistics and generates report.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910658432.4A CN110503131B (en) | 2019-07-22 | 2019-07-22 | Wind driven generator health monitoring system based on big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910658432.4A CN110503131B (en) | 2019-07-22 | 2019-07-22 | Wind driven generator health monitoring system based on big data analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110503131A true CN110503131A (en) | 2019-11-26 |
| CN110503131B CN110503131B (en) | 2023-10-10 |
Family
ID=68586748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910658432.4A Active CN110503131B (en) | 2019-07-22 | 2019-07-22 | Wind driven generator health monitoring system based on big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110503131B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101301A (en) * | 2020-11-03 | 2020-12-18 | 武汉工程大学 | Good sound stability early warning method and device for screw water cooling unit and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103472802A (en) * | 2013-09-13 | 2013-12-25 | 同济大学 | Wind generating set intelligent condition monitoring terminal and data processing method thereof |
| CN103809115A (en) * | 2014-02-27 | 2014-05-21 | 北京航空航天大学 | Online monitoring and fault diagnosis system based on vibration signal detection for circuit breaker |
| CN106844171A (en) * | 2016-12-27 | 2017-06-13 | 浪潮软件集团有限公司 | Mass operation and maintenance implementation method |
| CN108399199A (en) * | 2018-01-30 | 2018-08-14 | 武汉大学 | A kind of collection of the application software running log based on Spark and service processing system and method |
-
2019
- 2019-07-22 CN CN201910658432.4A patent/CN110503131B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103472802A (en) * | 2013-09-13 | 2013-12-25 | 同济大学 | Wind generating set intelligent condition monitoring terminal and data processing method thereof |
| CN103809115A (en) * | 2014-02-27 | 2014-05-21 | 北京航空航天大学 | Online monitoring and fault diagnosis system based on vibration signal detection for circuit breaker |
| CN106844171A (en) * | 2016-12-27 | 2017-06-13 | 浪潮软件集团有限公司 | Mass operation and maintenance implementation method |
| CN108399199A (en) * | 2018-01-30 | 2018-08-14 | 武汉大学 | A kind of collection of the application software running log based on Spark and service processing system and method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101301A (en) * | 2020-11-03 | 2020-12-18 | 武汉工程大学 | Good sound stability early warning method and device for screw water cooling unit and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110503131B (en) | 2023-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106371986A (en) | Log treatment operation and maintenance monitoring system | |
| CN113450065B (en) | Production operation management system and method for wind power plant | |
| CN107577588A (en) | A kind of massive logs data intelligence operational system | |
| CN109800127A (en) | A kind of system fault diagnosis intelligence O&M method and system based on machine learning | |
| CN102594620B (en) | Linkable distributed network intrusion detection method based on behavior description | |
| CN108964995A (en) | Log correlation analysis method based on time shaft event | |
| CN105553957A (en) | Network safety situation awareness early-warning method and system based big data | |
| CN111162949A (en) | Interface monitoring method based on Java byte code embedding technology | |
| CN115809183A (en) | Method for discovering and disposing information-creating terminal fault based on knowledge graph | |
| CN104574219A (en) | System and method for monitoring and early warning of operation conditions of power grid service information system | |
| CN116662989B (en) | Security data analysis method and system | |
| CN106897779A (en) | A kind of processing method of data center's operational system event | |
| CN111858251A (en) | Big data computing technology-based data security audit method and system | |
| CN115733762A (en) | Monitoring system with big data analysis capability | |
| CN115001934A (en) | Industrial control safety risk analysis system and method | |
| CN116257021A (en) | Intelligent network security situation monitoring and early warning platform for industrial control system | |
| CN111262734A (en) | Network security event emergency processing method | |
| CN110503131A (en) | Wind-driven generator health monitoring systems based on big data analysis | |
| CN117220917A (en) | Network real-time monitoring method based on cloud computing | |
| CN106649034A (en) | Visual intelligent operation and maintenance method and platform | |
| CN116232695A (en) | Network security operation and maintenance association analysis system | |
| CN115658772A (en) | Unmanned aerial vehicle photovoltaic inspection data asset management method and system | |
| CN111539642B (en) | Object-oriented power plant data acquisition and processing system and method thereof | |
| Wang et al. | LSTM-based alarm prediction in the mobile communication network | |
| CN112488873A (en) | Intelligent mining construction method for health codes and state tracks of power supply and utilization equipment |
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 |