CN104007343B - A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network - Google Patents
A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network Download PDFInfo
- Publication number
- CN104007343B CN104007343B CN201410222904.9A CN201410222904A CN104007343B CN 104007343 B CN104007343 B CN 104007343B CN 201410222904 A CN201410222904 A CN 201410222904A CN 104007343 B CN104007343 B CN 104007343B
- Authority
- CN
- China
- Prior art keywords
- transformer
- fault
- characteristic quantity
- bayesian network
- evidence
- 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 Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The invention discloses a kind of transformer dynamic fault diagnosis method based on Bayesian network, it is related to technical field of electric equipment.In the case where acquisition evidence is limited, resultant fault diagnostic model is extended to forward evidence and obtains the stage by the present invention, it is proposed dynamic fault diagnosis mechanism, evidence acquisition process is optimized according to certain principle, prioritizing selection is to the state characteristic quantity of transformer fault situation support maximum as evident information.Dynamic fault diagnosis mechanism, which is intended to prioritizing selection, influences transformer station high-voltage side bus failure diagnostic process input parameter of the maximum state characteristic quantity as model, and other unnecessary testing inspections are omitted, checkup item is reduced in the case where resource is limited can obtain accurate Risk parameter again.
Description
Technical field
The present invention relates to technical field of electric equipment, and in particular to a kind of transformer dynamic fault based on Bayesian network
Diagnostic method.
Background technology
Transformer fault is mostly the process slowly developed, and during which some state characteristic quantities can show exception, these exceptions
Amount is the important evidence of fault diagnosis.Oil chromatography can reflect inside transformer majority failure as one of important state characteristic quantity,
And oil colours modal data is easier to obtain relative to other state characteristic quantities, so the research weight as current transformer fault diagnosis
Point.But the Limited information that oil chromatography is covered, is only capable of that the malfunction of transformer is carried out tentatively and roughly to judge, in order to obtain
Detailed failure mode information, it is necessary to which resultant fault diagnosis is carried out to transformer.Resultant fault diagnosis is the stateful spy of fusion institute
Sign amount carries out transformer fault mode that may be present the process of complex reasoning diagnosis.Since transformer fault process is complicated,
State characteristic quantity has the characteristics that ambiguity, incompleteness, it is not corresponded with fault mode.Therefore, existing research is more
Problems are handled with Intelligent Diagnosis Technology, such as the methods of neutral net, Bayesian network, expert system and evidential reasoning,
So that diagnostic result is more accurate.
During the diagnosis based on resultant fault diagnostic model, the evidence of input model is more, and diagnostic result is closer
Truth, the estimation of probability of malfunction is also more accurate, if entire evidence information can be obtained, the estimation to risk has the most
Profit.But because live physical condition is limited, the evidence detected is all incomplete, the classification and quantity of these evidences directly affect
The accuracy of probability of malfunction estimation.The problem of general resultant fault diagnostic model is not related to evidence validity, but will inspection
The evidence measured directly inputs diagnostic model and the fault mode of transformer is judged, herein that this diagnosis mechanism is referred to as quiet
State fault diagnosis mechanism, existing resultant fault diagnostic model are nearly all static failure diagnosis mechanism.Static failure diagnostic machine
System does not optimize evidence acquisition process screening, and the evidence inputted during diagnosis, which may be omitted, can most reflect transformer
The state characteristic quantity of fault condition, this will directly affect the accuracy of probability of malfunction estimation.
The content of the invention
The purpose of the present invention is to propose to a kind of transformer dynamic fault diagnosis method based on Bayesian network, to overcome
The deficiency of the static failure diagnostic mode of faulty diagnostic method, makes result of calculation be consistent with actual transformer fault situation.
A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network of the embodiment of the present invention, including it is following
Step：
(1) fault case and expertise according to transformer establish the Bayesian network model of transformer fault；
(2) the partial status characteristic quantity that Transformer monitors is collected
Value, and be regarded as the evidence of transformer fault diagnosis, be input in the Bayesian network model established in step (1) into
Mobile state resultant fault diagnoses.
In one embodiment of the invention, the step (1) comprises the following steps：
(1-1) according to transformer fault case and expertise, compile transformer common failure pattern F and
State characteristic quantity S；
(1-2) are based between expertise and failure data analyzing the state characteristic quantity S and fault mode F of transformer
Causality, and establish corresponding Bayesian network model.
In one embodiment of the invention, the step (2) comprises the following steps：
(2-1) the partial status characteristic quantity value e that Transformer detects is collected, is regarded as evidence E=e, if
Determine the probability threshold value P during dynamic fault diagnosisth；
Evidence E is inputted Bayesian network model by (2-2), the posterior probability of all fault modes is asked for, by posterior probability
Maximum fault mode fiIt is considered as the fault mode that may occur, its posteriority probability tables is shown as Pmax；
(2-3) are to the posterior probability P that is obtained in step (2-2)maxJudged, if PmaxMore than or equal to PthThen turn to step
Suddenly (2-6)；If PmaxLess than Pth, then implement the steps of in order；
(2-4) fault mode f is assumediOccur, by evidence E=[e, fi] input Bayesian network model, obtain state spy
Sign amountThe state characteristic quantity s of middle posterior probability maximumj；
(2-5) state characteristic quantity s is implementedjVerification experimental verification, judge sjIt is whether abnormal, if sjThere is no abnormal then renewal card
According to for E=[e, sj=0], otherwiseRepeat step (2-3) to (2-5)；
(2-6) dynamic fault diagnosis process terminates, and it is f to finally obtain the current fault mode of transformer.
The present invention proposes a kind of transformer dynamic fault diagnosis method based on Bayesian network, its advantage is the present invention
The calculating process of method employs dynamic fault diagnosis mechanism, reduces the blindness of diagnosis process, can be with actual transformer
Fault condition is coincide more preferably, applicability higher.
Embodiment
The embodiment of the present invention is described below in detail, the example of the embodiment is shown, wherein identical from beginning to end or class
As label represent same or similar element or there is same or like element.It is below by the embodiment of description
Exemplary, it is intended to for explaining the present invention, and it is not considered as limiting the invention.
In the case where acquisition evidence is limited, resultant fault diagnostic model is extended to forward evidence and obtains rank by the present invention
Section, proposes dynamic fault diagnosis mechanism, optimizes evidence acquisition process according to certain principle, prioritizing selection is to transformer fault feelings
The state characteristic quantity of condition support maximum is as evident information.Dynamic fault diagnosis mechanism is intended to prioritizing selection to transformer station high-voltage side bus
Failure diagnostic process influences input parameter of the maximum state characteristic quantity as model, and omits other unnecessary experiment inspections
Survey, checkup item is reduced in the case where resource is limited can obtain accurate Risk parameter again.
A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network of the embodiment of the present invention, including it is following
Step：
(1) fault case and expertise according to transformer establish the Bayesian network model of transformer fault, specifically
Comprise the following steps：
(1-1) according to transformer fault case and expertise, compile transformer common failure pattern F and
State characteristic quantity S.
F=[f1, f2, f3, f4, f5, f6, f7, f8, f9, f10]；S=[s1, s2, s3, s4, s5, s6, s7, s8, s9]；Wherein, f1
For multipoint earthing of iron core, f2For insulation ag(e)ing, f3Generate heat for leakage field, f4For winding short circuit, f5For humidified insulation, f6For tap switch
Failure, f7For suspended discharge, f8Discharge for screen, f9For winding deformation, f10To discharge in oil；s1For iron core grounding current, s2For
Code of direct ratio is in Superheated steam drier feature, s3For the three-phase imbalance coefficient of winding D.C. resistance, s4For in transformer body oil
Water content, s5It is in discharging fault feature for code of direct ratio, s6For winding no-load voltage ratio deviation, s7For shelf depreciation, s8Fors9For the absorptance or polarization index of winding.
(1-2) are based between expertise and failure data analyzing the state characteristic quantity S and fault mode F of transformer
Causality, and establish corresponding Bayesian network model.
By analysis, F is S reasons, and S is F's as a result, Causal Strength matrix R is between F and S
Wherein, row represents F, and row represent S, RijValue represent fiKind fault mode causes sjKind state characteristic quantity hair
Raw abnormal probability, particularly, "-" represents noncausal relationship therebetween.
R is the connection relation of the two in Bayesian network, and Bayesian network can be directly established according to R.Bayesian network
Another important parameter of network, the prior probability P of Finitial, the fault statistics data based on transformer can obtain：
Pinitial=[0.45,0.11,0.13,0.12,0.10,0.26,0.16,0.28,0.24,0.14].
(2) value for the partial status characteristic quantity that Transformer monitors is collected And be regarded as the evidence of transformer fault diagnosis, be input in the Bayesian network model established in step (1) into
Mobile state resultant fault diagnoses.Wherein SpartWithComplementation, the two set added up are equal to S.Step (2) specifically include with
Lower step：
(2-1) the partial status characteristic quantity value e=[s that Transformer detects are collected2=1, s5=0], by it
It is considered as evidence E=e, sets the probability threshold value P during dynamic fault diagnosisth=0.8.
Wherein state characteristic quantity value represents not to be abnormal for 0, and 1 represents to be abnormal.
Evidence E is inputted Bayesian network model by (2-2), asks for the posterior probability of all fault modes, posterior probability is most
Big fault mode is f1, it is regarded as the fault mode that may occur, its posteriority probability Pmax=0.1965.
(2-3) are to the posterior probability P that is obtained in step (2-2)maxJudged, PmaxLess than Pth。
(2-4) fault mode f is assumed1Occur, evidence E=[e, f] is inputted into Bayesian network model, obtain state spy
Sign amountThe state characteristic quantity of middle posterior probability maximum is s1。
(2-5) state characteristic quantity s is implemented1Verification experimental verification, s1It is not abnormal, then more fresh evidence is E=[e, s1=
0]；
Repeat step (2-2) to (2-5), until PmaxMore than or equal to PthUntill.
(2-6) dynamic fault diagnosis process terminates, and it is f to finally obtain the current fault mode of transformer6。
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example description
Point is contained at least one embodiment of the present invention or example.In the present specification, schematic expression of the above terms is not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
Combined in an appropriate manner in one or more embodiments or example.In addition, without conflicting with each other, the skill of this area
Art personnel can be tied the different embodiments or example described in this specification and different embodiments or exemplary feature
Close and combine.
Although the embodiment of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is impossible to limitation of the present invention is interpreted as, those of ordinary skill in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, changes, replacing and modification.
Claims (1)
1. a kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network, it is characterised in that comprise the following steps：
(1) fault case and expertise according to transformer establish the Bayesian network model of transformer fault, specifically include
Following steps：
(1-1) fault case and expertise according to transformer, compiles the common failure pattern F and state of transformer
Characteristic quantity S, wherein, F=[f1, f2, f3, f4, f5, f6, f7, f8, f9, f10]；S=[s1, s2, s3, s4, s5, s6, s7, s8, s9], f1
For multipoint earthing of iron core, f2For insulation ag(e)ing, f3Generate heat for leakage field, f4For winding short circuit, f5For humidified insulation, f6For tap switch
Failure, f7For suspended discharge, f8Discharge for screen, f9For winding deformation, f10To discharge in oil；s1For iron core grounding current, s2For
Code of direct ratio is in Superheated steam drier feature, s3For the three-phase imbalance coefficient of winding D.C. resistance, s4For in transformer body oil
Water content, s5It is in discharging fault feature for code of direct ratio, s6For winding no-load voltage ratio deviation, s7For shelf depreciation, s8Fors9For the absorptance or polarization index of winding,
(1-2) are based on the cause and effect between expertise and failure data analyzing the state characteristic quantity S and fault mode F of transformer
Relation, and establish corresponding Bayesian network model；
(2) the partial status characteristic quantity that Transformer monitors is collected's
Value, and the evidence of transformer fault diagnosis is regarded as, it is input in the Bayesian network model established in step (1) and carries out
Dynamic comprehensive fault diagnosis, specifically includes following steps：
(2-1) collects the partial status characteristic quantity value e=[s that Transformer detects2=1, s5=0], it is regarded as demonstrate,proving
According to E=e, the probability threshold value P during dynamic fault diagnosis is setth, wherein, state characteristic quantity value represents not occur for 0
Abnormal, 1 represents to be abnormal,
Evidence E is inputted Bayesian network model by (2-2), asks for the posterior probability of all fault modes, and posterior probability is maximum
Fault mode fiIt is considered as the fault mode that may occur, its posteriority probability tables is shown as Pmax,
(2-3) is to the posterior probability P that is obtained in step (2-2)maxJudged, if PmaxMore than or equal to PthThen turn to step (2-
6), if PmaxLess than Pth, then implement the steps of in order,
(2-4) assumes fault mode fiOccur, by evidence E=[e, fi] input Bayesian network model, obtain state characteristic quantity
The state characteristic quantity s of middle posterior probability maximumj,
(2-5) implements state characteristic quantity sjVerification experimental verification, judge sjIt is whether abnormal, if sjIt is E there is no abnormal then more fresh evidence
=[e, sj=0], otherwise E=[e, sj=1], repeat step (2-2) to (2-5),
(2-6) dynamic fault diagnosis process terminates, and it is f to finally obtain the current fault mode of transformer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410222904.9A CN104007343B (en) | 2014-05-23 | 2014-05-23 | A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410222904.9A CN104007343B (en) | 2014-05-23 | 2014-05-23 | A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104007343A CN104007343A (en) | 2014-08-27 |
| CN104007343B true CN104007343B (en) | 2018-04-20 |
Family
ID=51368095
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410222904.9A Active CN104007343B (en) | 2014-05-23 | 2014-05-23 | A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104007343B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11899075B2 (en) | 2020-08-04 | 2024-02-13 | Maschinenfabrik Reinhausen Gmbh | Device for determining an error probability value for a transformer component and a system having such a device |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242129B (en) * | 2015-08-28 | 2018-03-13 | 广西电网有限责任公司电力科学研究院 | A kind of transformer winding fault probability determination method |
| CN105372528B (en) * | 2015-11-24 | 2018-10-09 | 湖南大学 | A kind of state maintenance method of Power Transformer Internal Faults and New Transformer |
| CN107907778B (en) * | 2017-10-31 | 2020-06-19 | 华北电力大学（保定） | Transformer comprehensive fault diagnosis method based on multiple characteristic parameters |
| CN107846016A (en) * | 2017-11-16 | 2018-03-27 | 中国南方电网有限责任公司 | A kind of Distribution Network Failure localization method and equipment based on Bayes and Complex event processing |
| CN109447511A (en) * | 2018-11-13 | 2019-03-08 | 南方电网科学研究院有限责任公司 | A kind of Diagnosis Method of Transformer Faults, system and relevant apparatus |
| CN111291245A (en) * | 2020-02-17 | 2020-06-16 | 广东电网有限责任公司电力科学研究院 | Case online generation system and method applied to PAS and computer equipment |
| CN111272222B (en) * | 2020-02-28 | 2021-06-25 | 西南交通大学 | Transformer fault diagnosis method based on characteristic quantity set |
| CN112182960A (en) * | 2020-09-22 | 2021-01-05 | 国网内蒙古东部电力有限公司呼伦贝尔供电公司 | Power transformer state risk assessment method based on Bayesian network |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102411106B (en) * | 2011-11-18 | 2014-01-15 | 广东电网公司广州供电局 | Fault monitoring method and device for power transformer |
| CN102779230B (en) * | 2012-06-14 | 2015-01-28 | 华南理工大学 | State analysis and maintenance decision judging method of power transformer system |
| CN103197177B (en) * | 2013-03-20 | 2015-09-23 | 山东电力集团公司济宁供电公司 | A kind of transformer fault diagnosis analytical approach based on Bayesian network |
| CN103245861B (en) * | 2013-05-03 | 2016-06-08 | 云南电力试验研究院（集团）有限公司电力研究院 | A kind of transformer fault diagnosis method based on Bayesian network |
-
2014
- 2014-05-23 CN CN201410222904.9A patent/CN104007343B/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11899075B2 (en) | 2020-08-04 | 2024-02-13 | Maschinenfabrik Reinhausen Gmbh | Device for determining an error probability value for a transformer component and a system having such a device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104007343A (en) | 2014-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104007343B (en) | A kind of transformer dynamic comprehensive method for diagnosing faults based on Bayesian network | |
| DE102012102770B9 (en) | System and method for error isolation and error mitigation based on network modeling | |
| Guo et al. | A simple reliability block diagram method for safety integrity verification | |
| CN110221145A (en) | Fault Diagnosis for Electrical Equipment method, apparatus and terminal device | |
| Abbasghorbani et al. | Reliability‐centred maintenance for circuit breakers in transmission networks | |
| Feng et al. | A technical framework of PHM and active maintenance for modern high-speed railway traction power supply systems | |
| CN104504248A (en) | Failure diagnosis modeling method based on designing data analysis | |
| CN109670610A (en) | Fault diagnosis optimization method based on fault propagation analysis | |
| CN109670611A (en) | A kind of power information system method for diagnosing faults and device | |
| CN111489539A (en) | Household appliance system fault early warning method, system and device | |
| CN102306244B (en) | Fault eliminating method based on evaluation of detecting points | |
| Beyza et al. | Characterising the security of power system topologies through a combined assessment of reliability, robustness, and resilience | |
| CN110361609A (en) | Extra-high voltage equipment monitors system and method | |
| CN111062569A (en) | Low-current fault discrimination method based on BP neural network | |
| Naderi et al. | Detection of false data injection cyberattacks: Experimental validation on a lab-scale microgrid | |
| CN103235206A (en) | Transformer fault diagnosis method | |
| CN117031201A (en) | Multi-scene topology anomaly identification method and system for power distribution network | |
| CN111654417A (en) | Evaluation method and device, storage medium, processor and train | |
| AL-Rubayi et al. | Simulation of line outage distribution factors (lodf) calculation for n-buses system | |
| Liu et al. | Fault diagnosis of electric railway traction substation with model-based relation guiding algorithm | |
| CN110238878A (en) | Self checking method and device for robot | |
| Davis et al. | Linear analysis of multiple outage interaction | |
| CN115372752A (en) | Fault detection method, device, electronic equipment and storage medium | |
| CN110907716B (en) | Method and device for judging voltage deviation isolation effect, power supply system, computer equipment and storage medium | |
| Overbye et al. | Human factors aspects of power system visualizations: An empirical investigation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |