CN106443259A - Transformer fault diagnosis new method based on Euclidean clustering and SPO-SVM - Google Patents
Transformer fault diagnosis new method based on Euclidean clustering and SPO-SVM Download PDFInfo
- Publication number
- CN106443259A CN106443259A CN201610865113.7A CN201610865113A CN106443259A CN 106443259 A CN106443259 A CN 106443259A CN 201610865113 A CN201610865113 A CN 201610865113A CN 106443259 A CN106443259 A CN 106443259A
- Authority
- CN
- China
- Prior art keywords
- svm
- transformer
- spo
- fault diagnosis
- new method
- 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
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Abstract
The invention discloses a transformer fault diagnosis new method based on Euclidean clustering and SPO-SVM. The method comprises the following steps: selecting sample data, dividing the sample data into a training sample and a test sample and carrying out normalization processing; dividing states of a transformer; constructing an Euclidean distance classifier; constructing an SVM multi-classifier; constructing a Gaussian radial basis kernel function as a kernel function; carrying out optimization on parameters of the Gaussian radial basis kernel function through a particle swarm optimization theory; inputting the training sample to a support vector machine for model learning, and establishing a learning model based on a least square support vector machine; inputting the test sample to the binary-class SVM for calculating; and obtaining fault category of the transformer. The beneficial effects are that optimization is carried out on the parameters of the kernel function of the support vector machine through particle swarm, and the support vector machine obtained after optimization carries out training and test on the sample data, thereby realizing quick and accurate diagnosis of the fault of the transformer.
Description
Technical field
The invention belongs to converting equipment operation troubles diagnostic field, particularly to a kind of based on European cluster and SPO-SVM
New Method of Power Transformer Fault Diagnosis.
Background technology
Power transformer is the visual plant of power system, and its running status directly affects the level of security of system, in time
Find latent transformer fault, can prevent from thus causing major accident.Due to Gases Dissolved in Transformer Oil content with
Ratio can reflect the running status of transformer to a great extent, based on this defined IEC recommendation three-ratio method,
The conventional methods such as Rogers method, but it is excessively absolute etc. to find in running at the scene that traditional algorithm has understaffed code, encoded limit
Major defect.
SVMs (SVM) needs training sample data few, and generalization ability is strong, has very in transformer fault diagnosis
Good application prospect, but it is more and exist and subregion, regularization coefficient can not determine the inherent limitations such as difficulty to there is grader.Cause
This, be optimized to grader classification effectiveness and kernel functional parameter, carries out fast and accurately transformer fault diagnosis to using SVM
There is very great meaning.
Content of the invention
For solving the defect of prior art, the invention particularly discloses a kind of transformation based on European cluster and SPO-SVM
Device New Fault Diagnosis Method, the method is based on least square method supporting vector machine (LS-SVM), and construction based on European cluster and is supported
The fault diagnosis model of vector machine, is optimized to the kernel functional parameter of SVMs using population, after optimizing
SVMs is trained to sample data and tests, and realizes the fault mode diagnosis to transformer.
For achieving the above object, the concrete scheme of the present invention is as follows:
Based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, including:
(1) choose oil dissolved gas as feature input vector, choose sample data, sample data is divided into training sample
Basis and test sample, and it is normalized;
(2) divide the state of transformer, and be respectively adopted the different conditions of the vector representation transformer of setting;
(3) construct euclid-distance classifier:According to the different state of transformer, calculate under each transformer state not
With Oil Dissolved Gases Concentration mean value as center vector, Oil Dissolved Gases Concentrations different in test sample data are divided
Do not carry out Euclidean distance calculating to the center vector of corresponding dissolved gas under each transformer state;
Choose one of Euclidean distance minimum as classification results, record the corresponding fault category of sub-minimum simultaneously;
(4) construct SVM multi-categorizer:According to the different state of transformer, using arranging in pairs or groups two-by-two, form the SVM of two classification;
(5) construction gaussian radial basis function is as kernel function;
(6) using particle group optimizing theory, parameter σ of gaussian radial basis function and punishment parameter C are optimized;
(7) training sample input SVMs is carried out model learning, set up based on least square method supporting vector machine
Learning model;
(8) SVM of two classification of construction in test sample input step (4) is calculated, carried out using temporal voting strategy
Decision-making;
(9) if using the testing data calculating in step (3) and center vector UkiApart from dkiFor minimum of a value, with
Heart vector UkjApart from dkjFor sub-minimum, then will be inner for SVM (i, j) that testing data is brought directly to be trained by classification i, j
Judged, the result obtaining is final judged result, the as fault category of transformer;Wherein, i, j represent gas classification, k
The state of indication transformer.
Further, in described step (1), choose H2、CH4、C2H6、C2H4、C2H2Five kinds reflection power transformer interior faults and
The characteristic gas that on-Line Monitor Device can monitor, remember H2、CH4、C2H6、C2H4、C2H2The ratio accounting for total hydrocarbon amount is designated as x respectively1、
x2,x3,x4,x5, as feature input variable.
Further, in described step (2), by transformer state C be divided into normally, low energy electric discharge, high-energy discharge, middle low temperature
Overheated, hyperthermia and superheating and six kinds of states of shelf depreciation, are respectively adopted vectorial [0,0,0,0,0,1]T、[0,0,0,0,1,0]T、[0,
0,0,1,0,0]T、[0,0,1,0,0,0]T、[0,1,0,0,0,0]T[1,0,0,0,0,0]TRepresent.
Further, in described step (6), using the target that particle group optimizing theory is optimized it is:K folding is selected to intersect
Corresponding parameter value when the average classification accuracy of checking maximizes.It is specially:
Wherein, liIt is the number of samples that i-th checking is concentrated, li TConcentrate the number correctly classified for this checking.
Further, in described step (6), the detailed process being optimized using particle group optimizing theory is as follows:
1) initialize;
2) pass through to calculate initial fitness value on search space for each particle in population, evaluate particle fitness;
3) speed according to iterative formula more new particle and position;
4) if optimizing reaches maximum evolutionary generation Tmax or evaluation of estimate is less than given accuracy, terminate searching process, no
Then proceed fitness evaluation.
Further, in described step (8), the concrete methods of realizing carrying out decision-making using temporal voting strategy is as follows:For survey
This x of sample, in the SVM of the i-th class and jth class composition, if test sample x belongs to i class, Jia 1 in the ballot of i class, no
Then in j class, Jia 1, until all classifier calculated complete, who gets the most votes's class is the class belonging to test sample.
Beneficial effect of the present invention:
The present invention passes through to be combined euclid-distance classifier with SVM algorithm, constructs European-SVM model, decreases SVM
Amount of calculation, and can effectively solve the problem that presence can not subregional problem;Using the kernel function to SVMs for the population
Parameter is optimized, and the SVMs after optimizing is trained to sample data and tests, and realizes the event to transformer
The quick and precisely diagnosis of barrier.
Brief description
Fig. 1 is particle group optimizing flow chart;
Fig. 2 is European-SVM model schematic;
Fig. 3 is transformer fault diagnosis flow chart.
Specific embodiment
The present invention is described in detail below in conjunction with the accompanying drawings:
Below the term being related in the present invention is described as follows:
SPO-SVM represents:SVMs based on particle swarm optimization algorithm;
SVM represents:SVMs;
European-SVM represents:European cluster and SVMs;
PSO represents particle swarm optimization algorithm;
LS-SVM represents least square method supporting vector machine;
Based on the New Method of Power Transformer Fault Diagnosis of European cluster and SPO-SVM, as shown in figure 3, comprising the following steps:
(1) choose oil dissolved gas as feature input vector, choose sample data, sample data is divided into training sample
Basis and test sample, and it is normalized;
Choose H2、CH4、C2H6、C2H4、C2H2Power transformer interior fault and most of on-Line Monitor Device can be reflected Deng five kinds
The characteristic gas that can monitor.Note H2、CH4、C2H6、C2H4、C2H2The ratio accounting for total hydrocarbon amount is designated as x respectively1、x2,x3,x4,x5, make
It is characterized input variable, the pattern of input vector is:[x1,x2,x3,x4,x5], choose sample data, sample data is returned
One change is processed, and sample data is divided into training set and test set.
(2) divide the state of transformer, and be respectively adopted the different conditions of the vector representation transformer of setting;
Transformer state C is divided into normally (N), low energy electric discharge (D1), high-energy discharge (D2), middle cryogenic overheating (T12), high
Overheated (the T of temperature3), six kinds of states of shelf depreciation (PD).And it is respectively adopted vectorial [0,0,0,0,0,1]T、[0,0,0,0,1,0]T、
[0,0,0,1,0,0]T、[0,0,1,0,0,0]T、[0,1,0,0,0,0]T、[1,0,0,0,0,0]TRepresent.
(3) construct euclid-distance classifier:According to the different state of transformer, calculate under each transformer state not
With Oil Dissolved Gases Concentration mean value as center vector, Oil Dissolved Gases Concentrations different in test sample data are divided
Center vector not and under each transformer state carries out Euclidean distance calculating, chooses a minimum conduct of Euclidean distance and divides
Class result, records the corresponding fault category of sub-minimum simultaneously;
It is illustrated below:
Step 1), calculate center vector UK, (k=1,2..m).Assume that transformer has m kind state, a) winding deformation, b) office
Discharge in portion, c) humidified insulation ..., and m) iron core short circuit, the mean value such as following table of different Oil Dissolved Gases Concentrations, then made
Centered on vector.
Table 1 center vector
Transformer state | H2 | CH4 | C2H6 | C2H4 | C2H2 |
a | 0.205 | 0.125 | 0.253 | 0.562 | 0.235 |
b | 0.125 | 0.251 | 0.134 | 0.189 | 0.168 |
c | 0.215 | 0.154 | 0.278 | 0.268 | 0.246 |
… | … | … | … | … | … |
m | 0.231 | 0.253 | 0.152 | 0.321 | 0.145 |
Step 2), Euclidean distance dkCalculate.Computing formula is as follows:
In formula, xiIt is test data, (i=1~5, x1~x5Respectively correspond to five kinds of gases), by test data respectively with respectively
The center vector of Status Type carries out Euclidean distance calculating.If test data is H2(0.165)、CH4(0.231)、C2H6
(0.215)、C2H4(0.214)、C2H2(0.189), carry out European calculating with a state first, obtain a result dka;Then successively with b
~m kind state carries out European calculating, obtains m Euclidean distance.
Step 3):Using one minimum in m Euclidean distance as classification results.As dkbMinimum, dkcSecondary minimum, then divide
Class result is b, and fault category is c.
According to six kinds of transformer states, calculate H in each class2、CH4、C2H6、C2H4、C2H2The mean value of content as in
Heart vector;Characteristic parameter H test data2、CH4、C2H6、C2H4、C2H2Content carries out European respectively with all kinds of center vectors
Distance calculates, as formula (1);Choose one of minimum as classification results in six Euclidean distances calculating, record simultaneously
The corresponding fault category of sub-minimum.
Wherein, dkFor Euclidean distance, i=1~5, x1~x5Correspond to five kinds of gases, x respectivelyiFor test data, xkiCentered on
Vector.
(4) construct SVM multi-categorizer:According to the different state of transformer, using arranging in pairs or groups two-by-two, form the SVM of two classification;
According to six kinds of transformer states, the SVM of 15 two classification of composition, carries out decision-making using temporal voting strategy altogether:For
Test sample x, in the SVM of the i-th class and jth class composition, if x belongs to i class, Jia 1, otherwise in j in the ballot of i class
Jia 1 in class, until all classifier calculated complete, who gets the most votes's class is the class belonging to test sample.I, j represent i-th respectively
Class, jth class gas, scope is 1~5.
(5) construct gaussian radial basis function (RBF) as kernel function, as formula (2).
Wherein, K (x, xi) represent kernel function, σ be not equal to zero constant;x-xiFor the distance between two vectors.
(6) adopt particle group optimizing theory (PSO) that parameter σ of RBF and punishment parameter C are optimized.Based on PSO's
LS-SVM fault diagnosis model needs to optimize 2 parameters C and σ, and that is, the dimension of PSO search space is 2 (d=2), then each particle
XiCan be by vector xi=(C, σ) is representing.The performance optimizing for evaluating, the fitness function of PSO is set to k folding intersection and tests
The average classification accuracy of card.When the target of PSO parameter optimization is that the average classification accuracy selecting k to roll over cross validation maximizes
Corresponding parameter value.
Wherein, liIt is the number of samples that i-th checking is concentrated, li TConcentrate the number correctly classified for this checking.
To realize step as follows for particle swarm optimization algorithm main:Initialize PSO first;Then pass through to calculate each in population
Initial fitness value on search space for the particle, evaluates particle fitness;Further according to iterative formula more new particle speed and
Position;If optimizing reaches maximum evolutionary generation Tmax or evaluation of estimate is less than given accuracy, terminate searching process, otherwise continue
Carry out fitness evaluation.Specific algorithm flow process is as shown in Figure 1.
(7) training sample input SVMs is carried out model learning, set up based on least square method supporting vector machine
Learning model;European-SVM the model building is as shown in Figure 2.
(8) SVM of 15 two classification of construction in test sample input step (4) is calculated, according to each SVM
Result of calculation is voted, and gained vote highest is final result;
(9) if using the testing data calculating in step (3) and center vector UkiApart from dkiFor minimum of a value, with
Heart vector UkjApart from dkjFor sub-minimum, then will be inner for SVM (i, j) that testing data is brought directly to be trained by classification i, j
Judged, the result obtaining is final judged result, the as fault category of transformer.K=1,2 ... m, represent that kth kind becomes
Depressor state, common m kind;I, j represent the i-th class, jth class gas respectively, and scope is 1~5.
Although the above-mentioned accompanying drawing that combines is described to the specific embodiment of the present invention, not model is protected to the present invention
The restriction enclosed, one of ordinary skill in the art should be understood that on the basis of technical scheme, and those skilled in the art are not
Need to pay the various modifications that creative work can make or deformation still within protection scope of the present invention.
Claims (8)
1. the New Method of Power Transformer Fault Diagnosis based on European cluster and SPO-SVM, is characterized in that, including:
(1) choose oil dissolved gas as feature input vector, choose sample data, by sample data be divided into training sample and
Test sample, and it is normalized;
(2) divide the state of transformer, and be respectively adopted the different conditions of the vector representation transformer of setting;
(3) construct euclid-distance classifier:According to the different state of transformer, calculate different oil under each transformer state
The mean value of middle dissolved gas content as center vector, by Oil Dissolved Gases Concentrations different in test sample data respectively with
Under each transformer state, the center vector of corresponding dissolved gas carries out Euclidean distance calculating;
Choose one of Euclidean distance minimum as classification results, record the corresponding fault category of sub-minimum simultaneously;
(4) construct SVM multi-categorizer:According to the different state of transformer, using arranging in pairs or groups two-by-two, form the SVM of two classification;
(5) construct gaussian radial basis function as kernel function, using particle group optimizing theory to gaussian radial basis function
Parameter σ and punishment parameter C are optimized;
(6) training sample input SVMs is carried out model learning, set up the study based on least square method supporting vector machine
Model;
(7) SVM of two classification of construction in test sample input step (4) is calculated, carried out decision-making using temporal voting strategy;
(8) if using the testing data calculating in step (3) and center vector UkiApart from dkiFor minimum of a value, with center vector
UkjApart from dkjFor sub-minimum, then testing data is brought directly to that the SVM (i, j) that trained by classification i, j is inner to be sentenced
Disconnected, the result obtaining is final judged result, the as fault category of transformer;Wherein, i, j represent gas classification, and k represents change
The state of depressor.
2. as claimed in claim 1 a kind of based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, its feature
It is, in described step (1), to choose H2、CH4、C2H6、C2H4、C2H2Five kinds of reflection power transformer interior faults and on-Line Monitor Device energy
The characteristic gas monitoring, remember H2、CH4、C2H6、C2H4、C2H2The ratio accounting for total hydrocarbon amount is designated as x respectively1、x2,x3,x4,x5, as
Feature input variable.
3. as claimed in claim 1 a kind of based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, its feature
Be, in described step (2), by transformer state C be divided into normally, low energy electric discharge, high-energy discharge, middle cryogenic overheating, hyperthermia and superheating
With six kinds of states of shelf depreciation, it is respectively adopted the vector representation of setting.
4. as claimed in claim 3 a kind of based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, its feature
It is that the vector of described setting is specially:[0,0,0,0,0,1]T、[0,0,0,0,1,0]T、[0,0,0,1,0,0]T、[0,0,1,0,
0,0]T、[0,1,0,0,0,0]T[1,0,0,0,0,0]T.
5. as claimed in claim 1 a kind of based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, its feature
It is, in described step (5), to be using the target that particle group optimizing theory is optimized:K is selected to roll over the average classification of cross validation
Corresponding parameter value when accuracy rate maximizes.
6. as claimed in claim 5 a kind of based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, its feature
It is to be specially using the desired value that particle group optimizing theory is optimized:
Wherein, liIt is the number of samples that i-th checking is concentrated, li TConcentrate the number correctly classified for this checking.
7. as claimed in claim 1 a kind of based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, its feature
It is that, in described step (5), the detailed process being optimized using particle group optimizing theory is as follows:
1) initialize;
2) pass through to calculate initial fitness value on search space for each particle in population, evaluate particle fitness;
3) speed according to iterative formula more new particle and position;
4) if optimizing reaches maximum evolutionary generation Tmax or evaluation of estimate is less than given accuracy, terminate searching process, otherwise continue
Continue and carry out fitness evaluation.
8. as claimed in claim 1 a kind of based on European cluster and SPO-SVM New Method of Power Transformer Fault Diagnosis, its feature
It is that, in described step (7), the concrete methods of realizing carrying out decision-making using temporal voting strategy is as follows:For test sample x, in the i-th class
In the SVM of jth class composition, if test sample x belongs to i class, the ballot of i class Jia 1, otherwise in j class, Jia 1, directly
Complete to all classifier calculated, who gets the most votes's class is the class belonging to test sample.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610865113.7A CN106443259A (en) | 2016-09-29 | 2016-09-29 | Transformer fault diagnosis new method based on Euclidean clustering and SPO-SVM |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610865113.7A CN106443259A (en) | 2016-09-29 | 2016-09-29 | Transformer fault diagnosis new method based on Euclidean clustering and SPO-SVM |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106443259A true CN106443259A (en) | 2017-02-22 |
Family
ID=58171141
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610865113.7A Pending CN106443259A (en) | 2016-09-29 | 2016-09-29 | Transformer fault diagnosis new method based on Euclidean clustering and SPO-SVM |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106443259A (en) |
Cited By (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107194068A (en) * | 2017-05-22 | 2017-09-22 | 中国石油大学(北京) | Shale gas fracturing process underground unusual service condition real-time estimate method for early warning and device |
CN107228913A (en) * | 2017-06-09 | 2017-10-03 | 广西电网有限责任公司电力科学研究院 | A kind of condition diagnosing system of transformer fault type |
CN107341504A (en) * | 2017-06-07 | 2017-11-10 | 同济大学 | A kind of Trouble Diagnostic Method of Machinery Equipment based on the popular study of time series data |
CN107463963A (en) * | 2017-08-10 | 2017-12-12 | 郑州云海信息技术有限公司 | A kind of Fault Classification and device |
CN107491783A (en) * | 2017-07-31 | 2017-12-19 | 广东电网有限责任公司惠州供电局 | Based on the transformer fault genre classification methods for improving density peaks clustering algorithm |
CN107656154A (en) * | 2017-09-18 | 2018-02-02 | 杭州安脉盛智能技术有限公司 | Based on the Diagnosis Method of Transformer Faults for improving Fuzzy C-Means Cluster Algorithm |
CN107884663A (en) * | 2017-10-27 | 2018-04-06 | 国网天津市电力公司电力科学研究院 | A kind of Diagnosis Method of Transformer Faults based on combination core Method Using Relevance Vector Machine |
CN108169583A (en) * | 2017-11-17 | 2018-06-15 | 国网湖南省电力有限公司 | Auto-transformer D.C. magnetic biasing method of discrimination and system of the neutral point through capacity earth |
CN108268905A (en) * | 2018-03-21 | 2018-07-10 | 广东电网有限责任公司电力科学研究院 | A kind of Diagnosis Method of Transformer Faults and system based on support vector machines |
CN108508318A (en) * | 2018-03-22 | 2018-09-07 | 国网湖南省电力有限公司 | A kind of method and system judging transformer unbalanced load operating status |
CN108519526A (en) * | 2018-03-22 | 2018-09-11 | 国网湖南省电力有限公司 | A kind of method and system judging transformer harmonic load operating status |
CN108663202A (en) * | 2018-05-03 | 2018-10-16 | 国家电网公司 | GIS mechanical failure diagnostic methods based on chaos cuckoo algorithm and system |
CN108693437A (en) * | 2018-03-22 | 2018-10-23 | 国网湖南省电力有限公司 | A kind of method and system judging deformation of transformer winding |
CN108717149A (en) * | 2018-05-25 | 2018-10-30 | 西安工程大学 | Diagnosis Method of Transformer Faults based on M-RVM fusion dynamic weightings AdaBoost |
CN109034646A (en) * | 2018-08-13 | 2018-12-18 | 东华大学 | A kind of method for diagnosing fault of power transformer and system of double class composite character selections |
CN109060892A (en) * | 2018-06-26 | 2018-12-21 | 西安交通大学 | SF based on graphene composite material sensor array6Decompose object detecting method |
CN109063734A (en) * | 2018-06-29 | 2018-12-21 | 广东工业大学 | The oil-immersed transformer malfunction appraisal procedure clustered in conjunction with multistage local density |
CN109270390A (en) * | 2018-09-14 | 2019-01-25 | 广西电网有限责任公司电力科学研究院 | Diagnosis Method of Transformer Faults based on Gaussian transformation Yu global optimizing SVM |
CN109342862A (en) * | 2018-12-14 | 2019-02-15 | 国网山东省电力公司电力科学研究院 | Based on Non-surveillance clustering with and svm classifier Diagnosis Method of Transformer Faults |
CN109490661A (en) * | 2018-10-23 | 2019-03-19 | 国网江苏省电力有限公司检修分公司 | Oil-immersed electric reactor method for diagnosing faults, apparatus and system based on PSO-SVM and Artificial Immune Algorithm |
CN110376458A (en) * | 2019-07-03 | 2019-10-25 | 东华大学 | Optimize the diagnosing fault of power transformer system of twin support vector machines |
CN110390419A (en) * | 2019-05-20 | 2019-10-29 | 重庆大学 | Freeway toll station method for predicting based on PSO-LSSVM model |
CN111598150A (en) * | 2020-05-12 | 2020-08-28 | 国网四川省电力公司电力科学研究院 | Transformer fault diagnosis method considering operation state grade |
CN111814989A (en) * | 2020-06-02 | 2020-10-23 | 西安工程大学 | Transformer fault diagnosis method for optimizing twin support vector machine based on locust algorithm |
CN112199890A (en) * | 2020-10-11 | 2021-01-08 | 哈尔滨工程大学 | System-level fault diagnosis method for integrated nuclear power device |
CN114295999A (en) * | 2021-12-30 | 2022-04-08 | 国网浙江省电力有限公司电力科学研究院 | Lithium ion battery SOH prediction method and system based on indirect health index |
CN115358353A (en) * | 2022-10-20 | 2022-11-18 | 众芯汉创(北京)科技有限公司 | Multi-source fusion transformer fault diagnosis method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101464964A (en) * | 2007-12-18 | 2009-06-24 | 同济大学 | Pattern recognition method capable of holding vectorial machine for equipment fault diagnosis |
CN101614775A (en) * | 2009-07-15 | 2009-12-30 | 河北科技大学 | Transformer State Assessment system and appraisal procedure thereof based on Multi-source Information Fusion |
CN103645249A (en) * | 2013-11-27 | 2014-03-19 | 国网黑龙江省电力有限公司 | Online fault detection method for reduced set-based downsampling unbalance SVM (Support Vector Machine) transformer |
CN103679263A (en) * | 2012-08-30 | 2014-03-26 | 重庆邮电大学 | Thunder and lightning approach forecasting method based on particle swarm support vector machine |
CN103942620A (en) * | 2014-04-18 | 2014-07-23 | 国家电网公司 | Wind power short-term prediction method using composite data source based on radial basis kernel function support vector machine |
US20140277599A1 (en) * | 2013-03-13 | 2014-09-18 | Oracle International Corporation | Innovative Approach to Distributed Energy Resource Scheduling |
-
2016
- 2016-09-29 CN CN201610865113.7A patent/CN106443259A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101464964A (en) * | 2007-12-18 | 2009-06-24 | 同济大学 | Pattern recognition method capable of holding vectorial machine for equipment fault diagnosis |
CN101614775A (en) * | 2009-07-15 | 2009-12-30 | 河北科技大学 | Transformer State Assessment system and appraisal procedure thereof based on Multi-source Information Fusion |
CN103679263A (en) * | 2012-08-30 | 2014-03-26 | 重庆邮电大学 | Thunder and lightning approach forecasting method based on particle swarm support vector machine |
US20140277599A1 (en) * | 2013-03-13 | 2014-09-18 | Oracle International Corporation | Innovative Approach to Distributed Energy Resource Scheduling |
CN103645249A (en) * | 2013-11-27 | 2014-03-19 | 国网黑龙江省电力有限公司 | Online fault detection method for reduced set-based downsampling unbalance SVM (Support Vector Machine) transformer |
CN103942620A (en) * | 2014-04-18 | 2014-07-23 | 国家电网公司 | Wind power short-term prediction method using composite data source based on radial basis kernel function support vector machine |
Non-Patent Citations (2)
Title |
---|
赵文清等: ""一种变压器故障诊断新方法"", 《计算机工程与应用》 * |
郑含博等: ""基于多分类最小二乘支持向量机和改进粒子群优化算法的电力变压器故障诊断方法"", 《高电压技术》 * |
Cited By (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107194068B (en) * | 2017-05-22 | 2020-01-10 | 中国石油大学(北京) | Real-time prediction and early warning method and device for underground abnormal working condition in shale gas fracturing process |
CN107194068A (en) * | 2017-05-22 | 2017-09-22 | 中国石油大学(北京) | Shale gas fracturing process underground unusual service condition real-time estimate method for early warning and device |
CN107341504A (en) * | 2017-06-07 | 2017-11-10 | 同济大学 | A kind of Trouble Diagnostic Method of Machinery Equipment based on the popular study of time series data |
CN107228913B (en) * | 2017-06-09 | 2023-08-08 | 广西电网有限责任公司电力科学研究院 | State diagnosis system for fault type of transformer |
CN107228913A (en) * | 2017-06-09 | 2017-10-03 | 广西电网有限责任公司电力科学研究院 | A kind of condition diagnosing system of transformer fault type |
CN107491783A (en) * | 2017-07-31 | 2017-12-19 | 广东电网有限责任公司惠州供电局 | Based on the transformer fault genre classification methods for improving density peaks clustering algorithm |
CN107491783B (en) * | 2017-07-31 | 2020-07-21 | 广东电网有限责任公司惠州供电局 | Transformer fault type classification method based on improved density peak value clustering algorithm |
CN107463963A (en) * | 2017-08-10 | 2017-12-12 | 郑州云海信息技术有限公司 | A kind of Fault Classification and device |
CN107656154A (en) * | 2017-09-18 | 2018-02-02 | 杭州安脉盛智能技术有限公司 | Based on the Diagnosis Method of Transformer Faults for improving Fuzzy C-Means Cluster Algorithm |
CN107884663A (en) * | 2017-10-27 | 2018-04-06 | 国网天津市电力公司电力科学研究院 | A kind of Diagnosis Method of Transformer Faults based on combination core Method Using Relevance Vector Machine |
CN108169583A (en) * | 2017-11-17 | 2018-06-15 | 国网湖南省电力有限公司 | Auto-transformer D.C. magnetic biasing method of discrimination and system of the neutral point through capacity earth |
CN108169583B (en) * | 2017-11-17 | 2020-11-20 | 国网湖南省电力有限公司 | Autotransformer direct-current magnetic bias discrimination method and system with neutral point grounded through capacitor |
CN108268905A (en) * | 2018-03-21 | 2018-07-10 | 广东电网有限责任公司电力科学研究院 | A kind of Diagnosis Method of Transformer Faults and system based on support vector machines |
CN108508318B (en) * | 2018-03-22 | 2020-12-18 | 国网湖南省电力有限公司 | Method and system for judging operation state of unbalanced load of transformer |
CN108693437A (en) * | 2018-03-22 | 2018-10-23 | 国网湖南省电力有限公司 | A kind of method and system judging deformation of transformer winding |
CN108693437B (en) * | 2018-03-22 | 2020-12-25 | 国网湖南省电力有限公司 | Method and system for judging deformation of transformer winding |
CN108508318A (en) * | 2018-03-22 | 2018-09-07 | 国网湖南省电力有限公司 | A kind of method and system judging transformer unbalanced load operating status |
CN108519526A (en) * | 2018-03-22 | 2018-09-11 | 国网湖南省电力有限公司 | A kind of method and system judging transformer harmonic load operating status |
CN108519526B (en) * | 2018-03-22 | 2020-12-25 | 国网湖南省电力有限公司 | Method and system for judging running state of harmonic load of transformer |
CN108663202A (en) * | 2018-05-03 | 2018-10-16 | 国家电网公司 | GIS mechanical failure diagnostic methods based on chaos cuckoo algorithm and system |
CN108717149A (en) * | 2018-05-25 | 2018-10-30 | 西安工程大学 | Diagnosis Method of Transformer Faults based on M-RVM fusion dynamic weightings AdaBoost |
CN109060892A (en) * | 2018-06-26 | 2018-12-21 | 西安交通大学 | SF based on graphene composite material sensor array6Decompose object detecting method |
CN109060892B (en) * | 2018-06-26 | 2020-12-25 | 西安交通大学 | SF based on graphene composite material sensor array6Method for detecting decomposition product |
CN109063734B (en) * | 2018-06-29 | 2022-02-25 | 广东工业大学 | Oil-immersed transformer fault state evaluation method combining multi-level local density clustering |
CN109063734A (en) * | 2018-06-29 | 2018-12-21 | 广东工业大学 | The oil-immersed transformer malfunction appraisal procedure clustered in conjunction with multistage local density |
CN109034646A (en) * | 2018-08-13 | 2018-12-18 | 东华大学 | A kind of method for diagnosing fault of power transformer and system of double class composite character selections |
CN109270390A (en) * | 2018-09-14 | 2019-01-25 | 广西电网有限责任公司电力科学研究院 | Diagnosis Method of Transformer Faults based on Gaussian transformation Yu global optimizing SVM |
CN109490661A (en) * | 2018-10-23 | 2019-03-19 | 国网江苏省电力有限公司检修分公司 | Oil-immersed electric reactor method for diagnosing faults, apparatus and system based on PSO-SVM and Artificial Immune Algorithm |
CN109342862A (en) * | 2018-12-14 | 2019-02-15 | 国网山东省电力公司电力科学研究院 | Based on Non-surveillance clustering with and svm classifier Diagnosis Method of Transformer Faults |
CN110390419A (en) * | 2019-05-20 | 2019-10-29 | 重庆大学 | Freeway toll station method for predicting based on PSO-LSSVM model |
CN110376458A (en) * | 2019-07-03 | 2019-10-25 | 东华大学 | Optimize the diagnosing fault of power transformer system of twin support vector machines |
CN111598150A (en) * | 2020-05-12 | 2020-08-28 | 国网四川省电力公司电力科学研究院 | Transformer fault diagnosis method considering operation state grade |
CN111598150B (en) * | 2020-05-12 | 2022-06-24 | 国网四川省电力公司电力科学研究院 | Transformer fault diagnosis method considering operation state grade |
CN111814989A (en) * | 2020-06-02 | 2020-10-23 | 西安工程大学 | Transformer fault diagnosis method for optimizing twin support vector machine based on locust algorithm |
CN112199890A (en) * | 2020-10-11 | 2021-01-08 | 哈尔滨工程大学 | System-level fault diagnosis method for integrated nuclear power device |
CN114295999A (en) * | 2021-12-30 | 2022-04-08 | 国网浙江省电力有限公司电力科学研究院 | Lithium ion battery SOH prediction method and system based on indirect health index |
CN115358353A (en) * | 2022-10-20 | 2022-11-18 | 众芯汉创(北京)科技有限公司 | Multi-source fusion transformer fault diagnosis method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106443259A (en) | Transformer fault diagnosis new method based on Euclidean clustering and SPO-SVM | |
CN112041693B (en) | Power distribution network fault positioning system based on mixed wave recording | |
CN102735999B (en) | Gas insulated substation (GIS) partial discharge online monitoring system and fault mode identifying method thereof | |
CN105550700B (en) | A kind of time series data cleaning method based on association analysis and principal component analysis | |
CN103914064B (en) | Based on the commercial run method for diagnosing faults that multi-categorizer and D-S evidence merge | |
CN107505133A (en) | The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM | |
CN103366123B (en) | Software hazard appraisal procedure based on defect analysis | |
CN108664010A (en) | Generating set fault data prediction technique, device and computer equipment | |
Zhang et al. | A new support vector machine model based on improved imperialist competitive algorithm for fault diagnosis of oil-immersed transformers | |
CN107656154A (en) | Based on the Diagnosis Method of Transformer Faults for improving Fuzzy C-Means Cluster Algorithm | |
CN106355030A (en) | Fault detection method based on analytic hierarchy process and weighted vote decision fusion | |
CN108268905A (en) | A kind of Diagnosis Method of Transformer Faults and system based on support vector machines | |
CN103324939B (en) | Skewed popularity classification and parameter optimization method based on least square method supporting vector machine technology | |
CN102074955A (en) | Method based on knowledge discovery technology for stability assessment and control of electric system | |
CN106897821A (en) | A kind of transient state assesses feature selection approach and device | |
CN104299115B (en) | Secondary system of intelligent substation state analysis method based on Fuzzy C-Means Cluster Algorithm | |
CN109298225B (en) | Automatic identification model system and method for abnormal state of voltage measurement data | |
CN106056235A (en) | Power transmission grid efficiency and benefit detection method based on Klee method and matter element extension model | |
CN110705887A (en) | Low-voltage transformer area operation state comprehensive evaluation method based on neural network model | |
CN107992880A (en) | A kind of optimal lump classification method for diagnosing faults of power transformer | |
CN109657720B (en) | On-line diagnosis method for turn-to-turn short circuit fault of power transformer | |
CN110133444A (en) | A kind of Fault Locating Method based on positive sequence voltage variable quantity, apparatus and system | |
Omar et al. | Fault classification on transmission line using LSTM network | |
CN109829627A (en) | A kind of safe confidence appraisal procedure of Electrical Power System Dynamic based on integrated study scheme | |
CN105117847A (en) | Method for evaluating transformer failure importance |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170222 |