CN103412220B - A kind of ship power station fault diagnosis system method based on data fusion - Google Patents
A kind of ship power station fault diagnosis system method based on data fusion Download PDFInfo
- Publication number
- CN103412220B CN103412220B CN201310350610.XA CN201310350610A CN103412220B CN 103412220 B CN103412220 B CN 103412220B CN 201310350610 A CN201310350610 A CN 201310350610A CN 103412220 B CN103412220 B CN 103412220B
- Authority
- CN
- China
- Prior art keywords
- data
- fusion
- fault diagnosis
- level
- parameter
- 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.)
- Active
Links
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides a kind of ship power station fault diagnosis system method based on data fusion, its different levels in fault diagnosis adopt different data anastomosing algorithms respectively, by the many-sided detection failure information of multiple sensor, classification process is carried out to multi-source information, accurately, the state of system is judged in time, provide the system failure whether and the correct judgement of fault mode, effective realization is to the intelligent monitoring of Ship Power Station unit, improve reliability and the security of unit operation, the phenomenon decrease false-alarm, report by mistake, failing to report; Multiple sensor is monitored respectively to failure message parameter, the Data Integration monitored is carried out first order fusion detection layer data, the data merged through the described first order carry out second level fusion, the data merged through the described second level carry out third level fusion, data after merging three grades are mated with in fault diagnosis knowledge base, export fault diagnosis result.
Description
Technical field
The present invention relates to artificial intelligence fault diagnosis technology field, be specially a kind of ship power station fault diagnosis system method based on data fusion.
Background technology
Ship Power Station is the important component part of boats and ships, it is the core of Ship Electrical Power System, the reliability service of Ship Power Station is significant to guarantee safety of ship, which is responsible for full marine electric equipment provide continuously, the reliable and task of electric energy that conforms to quality requirements, once it is quite serious that major accident consequence occurs, and along with boats and ships are to maximization, high speed and automation direction development, it is also proposed higher requirement to Ship Power Station, the main boats and ships unit method for diagnosing faults used mainly contains two kinds both at home and abroad at present: first vibration analysis; It two is the faults judging unit from electric parameter, but because Ship Power Station unit is a complicated nonlinear system, its fault observer is difficult to describe by simple mathematical model, so when carrying out fault diagnosis to it, is difficult to set up high-precision mathematical model.
Summary of the invention
For the problems referred to above, the invention provides a kind of ship power station fault diagnosis system method based on data fusion, its different levels in fault diagnosis adopt different data anastomosing algorithms respectively, by the many-sided detection failure information of multiple sensor, classification process is carried out to multi-source information, accurately, the state of system is judged in time, provide the system failure whether and the correct judgement of fault mode, effective realization is to the intelligent monitoring of Ship Power Station unit, improve reliability and the security of unit operation, the phenomenon decrease false-alarm, report by mistake, failing to report.
Technical scheme of the present invention is such: it is characterized in that: multiple sensor is monitored respectively to failure message parameter, the Data Integration monitored is carried out first order fusion detection layer data, the data merged through the described first order carry out second level fusion, the data merged through the described second level carry out third level fusion, data after merging three grades are mated with in fault diagnosis knowledge base, export fault diagnosis result.
It is further characterized in that: method for diagnosing faults step is:
A. the described first order merges employing adaptive weighting data fusion, is assumed to be the data that its n sensor detects
, find its corresponding flexible strategy factor W in an adaptive way according to the numerical value that each sensor obtains
1, W
2..., W
n, by adaptive weighting data fusion formula
,
, obtain the estimated value after merging
;
B. the grey Advantage Analysis adopting and improve is merged in the described second level, supposes often to detect 5 numerical value as one group to each sensor, and the value after often organizing detection fusion is designated as
,
,
, according to front four groups of data
,
,
,
obtain mean value
, wherein
, definition
for reference time array,
for the time series compared with it,
, then by calculating
with
the grey absolute correlation degree improved
, wherein,
,
，
，
，
，
，
，
Wherein
be grey correlation parameter,
for grey correlation corrected parameter,
According to the Grey Incidence Matrix improved
obtain grey correlation and compare parameter
,
, and obtain grey correlation and compare parameter
in minimum comparison parameter
if,
, the grey correlation in the 5th group of data is compared parameter and replaces the
grey correlation in group data compares parameter, exports the mean value of the front four groups of data after substituting
, wherein
, otherwise export
, output valve is just the fusion results of second level fusion;
C. the described third level merges employing D-S evidence theory, supposes that every 5 of the fusion results merged the described second level is one group, calculates the belief function often organizing data
and plausibility function
,
Definition elementary probability value
, identification framework
, function
,
According to
,
,
,
,
Wherein
for proposition
basic reliability distribution, represent proposition
accurate trusting degree,
it is identification framework
on belief function, represent proposition
the uncertain proposition sum of all subsets, be also to proposition
total trust,
expression can not deny proposition
degree,
Choose the belief function of each group of data
in closest to plausibility function
those data, be final diagnostic data.
The present invention adopt a kind of ship power station fault diagnosis system method based on data fusion, Data-Fusion theory is applied to the fault diagnosis of Ship Power Station unit by it, at three of fault diagnosis different levels, three kinds of different data anastomosing algorithms are proposed, failure message from multiple sensor is merged, obtain diagnostic result data, diagnostic result data are mated with the fault in fault diagnosis knowledge base the most at last, export fault diagnosis result, effective realization is to the intelligent monitoring of Ship Power Station unit, and be conducive to the reliability and the security that improve unit operation, reduce false-alarm, wrong report, fail to report phenomenon.
Accompanying drawing explanation
Fig. 1 is structured flowchart of the present invention.
Embodiment
As shown in Figure 1, the present invention includes multiple sensor to monitor respectively failure message parameter, the Data Integration monitored is carried out first order fusion detection layer data, and eliminate the unequal accuracy problem of sensor image data, the data merged through the first order carry out second level fusion to identify superior items and non-superior items, thus identify advantage fault eigenvalue, the data merged through the second level are carried out the third level and are merged to remove uncertain information, obtain fault diagnosis conclusion, diagnostic data after then merging three grades mates with in fault diagnosis knowledge base, export fault diagnosis result.
Method for diagnosing faults step is:
A. the first order merges employing adaptive weighting data fusion, is assumed to be the data that its n sensor detects
, find its corresponding flexible strategy factor W in an adaptive way according to the numerical value that each sensor obtains
1, W
2..., W
n, by adaptive weighting data fusion formula
,
, obtain the estimated value after merging
;
B. the grey Advantage Analysis adopting and improve is merged in the second level, supposes often to detect 5 numerical value as one group to each sensor, and the value after often organizing detection fusion is designated as
,
,
, according to front four groups of data
,
,
,
obtain mean value
, wherein
, definition
for reference time array,
for the time series compared with it,
, then by calculating
with
the grey absolute correlation degree improved
, wherein,
,
，
，
，
，
，
，
Wherein
be grey correlation parameter,
for grey correlation corrected parameter,
According to the Grey Incidence Matrix improved
obtain grey correlation and compare parameter
,
, and obtain grey correlation and compare parameter
in minimum comparison parameter
if,
, the grey correlation in the 5th group of data is compared parameter and replaces the
grey correlation in group data compares parameter, exports the mean value of the front four groups of data after substituting
, wherein
, otherwise export
, output valve is just the fusion results of second level fusion;
C. the third level merges employing D-S evidence theory, supposes that every 5 of the fusion results merged the second level is one group, calculates the belief function often organizing data
and plausibility function
,
Definition elementary probability value
, identification framework
, function
,
According to
,
,
,
,
Wherein
for proposition
basic reliability distribution, represent proposition
accurate trusting degree,
it is identification framework
on belief function, represent proposition
the uncertain proposition sum of all subsets, be also to proposition
total trust,
expression can not deny proposition
degree,
Choose the belief function of each group of data
in closest to plausibility function
those data, be final diagnostic data.
Claims (1)
1. the ship power station fault diagnosis system method based on data fusion, it is characterized in that: multiple sensor is monitored respectively to failure message parameter, the Data Integration monitored is carried out first order fusion detection layer data, the data merged through the described first order carry out second level fusion, the data merged through the described second level carry out third level fusion, data after merging three grades are mated with in fault diagnosis knowledge base, export fault diagnosis result;
Method for diagnosing faults step is:
A. the described first order merges employing adaptive weighting data fusion, is assumed to be the data that its n sensor detects
, find its corresponding flexible strategy factor W in an adaptive way according to the numerical value that each sensor obtains
1, W
2..., W
n, by adaptive weighting data fusion formula
,
, obtain the estimated value after merging
;
B. the grey Advantage Analysis adopting and improve is merged in the described second level, supposes often to detect 5 numerical value as one group to each sensor, and the value after often organizing detection fusion is designated as
,
,
, according to front four groups of data
,
,
,
obtain mean value
, wherein
, definition
for reference time array,
for the time series compared with it,
, then by calculating
with
the grey absolute correlation degree improved
, wherein,
,
，
，
，
，
，
，
Wherein
be grey correlation parameter,
for grey correlation corrected parameter,
According to the Grey Incidence Matrix improved
obtain grey correlation and compare parameter
,
, and obtain grey correlation and compare parameter
in minimum comparison parameter
if,
, the grey correlation in the 5th group of data is compared parameter and replaces the
grey correlation in group data compares parameter, exports the mean value of the front four groups of data after substituting
, wherein
, otherwise export
, output valve is just the fusion results of second level fusion;
C. the described third level merges employing D-S evidence theory, supposes that every 5 of the fusion results merged the described second level is one group, calculates the belief function often organizing data
and plausibility function
,
Definition elementary probability value
, identification framework
, function
,
According to
,
,
,
,
Wherein
for proposition
basic reliability distribution, represent proposition
accurate trusting degree,
it is identification framework
on belief function, represent proposition
the uncertain proposition sum of all subsets, be also to proposition
total trust,
expression can not deny proposition
degree,
Choose the belief function of each group of data
in closest to plausibility function
those data, be final diagnostic data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310350610.XA CN103412220B (en) | 2013-08-13 | 2013-08-13 | A kind of ship power station fault diagnosis system method based on data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310350610.XA CN103412220B (en) | 2013-08-13 | 2013-08-13 | A kind of ship power station fault diagnosis system method based on data fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103412220A CN103412220A (en) | 2013-11-27 |
| CN103412220B true CN103412220B (en) | 2016-03-02 |
Family
ID=49605246
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310350610.XA Active CN103412220B (en) | 2013-08-13 | 2013-08-13 | A kind of ship power station fault diagnosis system method based on data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103412220B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105792110A (en) * | 2016-03-30 | 2016-07-20 | 上海申腾信息技术有限公司 | Data fusion and intelligent searching processing method for multiple data sources |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104950875A (en) * | 2015-06-15 | 2015-09-30 | 西北工业大学 | Fault diagnosis method by combining correlation analysis and data fusion |
| CN105933388A (en) * | 2016-04-11 | 2016-09-07 | 贵州大学 | WSN data layered fusion method for plant growth monitoring |
| CN106981873B (en) * | 2017-04-25 | 2019-09-20 | 集美大学 | A kind of isolated island formula electric system based on dynamic behavior is hidden failure prediction method |
| CN108228535A (en) * | 2018-01-02 | 2018-06-29 | 佛山科学技术学院 | A kind of optimal weighting parameter evaluation method of unequal precision measurement data fusion |
| CN108319571A (en) * | 2018-01-02 | 2018-07-24 | 佛山科学技术学院 | A kind of multi-Dimensional parameters evaluation method of multiclass isomery model optimum fusion |
| CN109361728B (en) * | 2018-08-30 | 2021-01-29 | 中国科学院上海微系统与信息技术研究所 | Hierarchical event reporting system and method based on multi-source sensing data relevance |
| CN111307453B (en) * | 2020-03-20 | 2021-11-12 | 朗斯顿科技(北京)有限公司 | Transmission system fault diagnosis method based on multi-information fusion |
| CN111507429B (en) * | 2020-05-29 | 2023-08-01 | 智慧航海（青岛）科技有限公司 | Intelligent ship multisource perception data ship end fusion method, device and decision system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5586130A (en) * | 1994-10-03 | 1996-12-17 | Qualcomm Incorporated | Method and apparatus for detecting fault conditions in a vehicle data recording device to detect tampering or unauthorized access |
| CN101201370A (en) * | 2006-12-13 | 2008-06-18 | 上海海事大学 | Fault diagnosis system adopting circuit information amalgamation and implementing method thereof |
| CN103147972A (en) * | 2013-03-19 | 2013-06-12 | 北京化工大学 | Reciprocating-type compressor fault diagnosis method based on multi-sensor information fusion |
-
2013
- 2013-08-13 CN CN201310350610.XA patent/CN103412220B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5586130A (en) * | 1994-10-03 | 1996-12-17 | Qualcomm Incorporated | Method and apparatus for detecting fault conditions in a vehicle data recording device to detect tampering or unauthorized access |
| CN101201370A (en) * | 2006-12-13 | 2008-06-18 | 上海海事大学 | Fault diagnosis system adopting circuit information amalgamation and implementing method thereof |
| CN103147972A (en) * | 2013-03-19 | 2013-06-12 | 北京化工大学 | Reciprocating-type compressor fault diagnosis method based on multi-sensor information fusion |
Non-Patent Citations (2)
| Title |
|---|
| 基于证据理论的电机故障诊断研究;张松涛等;《仪器仪表学报》;20080430;第29卷(第4期);第101页 * |
| 船舶电站故障诊断中的数据融合算法;陈佳等;《电力自动化设备》;20060331;第26卷(第3期);第28-30页 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105792110A (en) * | 2016-03-30 | 2016-07-20 | 上海申腾信息技术有限公司 | Data fusion and intelligent searching processing method for multiple data sources |
| CN105792110B (en) * | 2016-03-30 | 2019-04-30 | 上海申腾信息技术有限公司 | A kind of data fusion of multiple data origin, the processing method of intelligent search |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103412220A (en) | 2013-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103412220B (en) | A kind of ship power station fault diagnosis system method based on data fusion | |
| CN108092623B (en) | A kind of photovoltaic array multisensor fault detecting and positioning method | |
| CN110398699A (en) | Power battery thermal runaway method for early warning combined of multi-sensor information | |
| CN113033011B (en) | Ship mechanical health state assessment method and system | |
| KR102156858B1 (en) | Method for diagnosing and predicting robot arm's failure | |
| CN115453356B (en) | Power equipment operation state monitoring and analyzing method, system, terminal and medium | |
| CN103617105B (en) | A kind of adaptive multi-layer flow model facility diagnosis method based on data-driven | |
| CN114977483A (en) | Fault diagnosis system for intelligent power grid regulation and control equipment | |
| CN109990803B (en) | Method and device for detecting system abnormity and method and device for sensor processing | |
| KR20210031179A (en) | System for predicting engine failure using big data | |
| Wang et al. | Improved methods of online monitoring and prediction in condensate and feed water system of nuclear power plant | |
| Guo et al. | Fault detection and diagnosis using statistic feature and improved broad learning for traction systems in high-speed trains | |
| KR101808461B1 (en) | Method and apparatus for predicting remaining life of a machine | |
| CN112039387A (en) | Fault diagnosis method for permanent magnet synchronous motor position sensor | |
| CN110399926B (en) | Street lamp fault diagnosis method and device | |
| CN108957315A (en) | Fault diagnosis method and equipment for wind generating set | |
| Liao et al. | Fault diagnosis of capacitance aging in DC link capacitors of voltage source inverters using evidence reasoning rule | |
| CN109268692B (en) | Heat supply pipe network well type compensator leakage monitoring system and monitoring method | |
| Zhang et al. | Applied sensor fault detection and validation using transposed input data PCA and ANNs | |
| CN115563868A (en) | Fault diagnosis method and device for oil circuit system of aviation alternating-current generator | |
| Pengbo et al. | Recent progress and prospective evaluation of fault diagnosis strategies for electrified drive powertrains: A comprehensive review | |
| KR102448187B1 (en) | the fault detection methods of PV panel using unit vector analysis for I-V curve | |
| Zia et al. | Locational Detection of False Data Injection Attack in Smart Grid Based on Multilabel Machine Learning Classification Methods | |
| Cheng et al. | A step-wise feature selection scheme for a prognostics and health management system in autonomous ferry crossing operation | |
| CN105068529B (en) | A kind of fault separating method and system compared based on weighted residual component |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |