CN107862332A - Insulation defect based on SVMs identification sulfur hexafluoride Characteristics of Partial Discharge - Google Patents
Insulation defect based on SVMs identification sulfur hexafluoride Characteristics of Partial Discharge Download PDFInfo
- Publication number
- CN107862332A CN107862332A CN201711049428.5A CN201711049428A CN107862332A CN 107862332 A CN107862332 A CN 107862332A CN 201711049428 A CN201711049428 A CN 201711049428A CN 107862332 A CN107862332 A CN 107862332A
- Authority
- CN
- China
- Prior art keywords
- support vector
- vector machine
- sulfur hexafluoride
- partial discharge
- formula
- 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
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
Abstract
The invention discloses a kind of insulation defect based on SVMs identification sulfur hexafluoride Characteristics of Partial Discharge, the identification process that it includes the insulation defect is as follows：S1：Construct support vector machine classifier；S2：The parameter of SVMs is optimized；S3：Obtain the decomposition data of sulfur hexafluoride；S4：The decomposition data of sulfur hexafluoride is identified SVMs.The beneficial effect that the present invention obtains is：Ensure SF6The safe and reliable operation of equipment, its discrimination to all kinds of defects is improved, the treatment effeciency to insulation fault can be improved.
Description
Technical Field
The invention relates to the technical field of support vector machines, in particular to a method for identifying insulation defects of sulfur hexafluoride partial discharge characteristics based on a support vector machine.
Background
Sulfur hexafluoride (SF) 6 ) Gas is widely used in gas-insulated equipment due to its excellent insulating and arc-extinguishing properties. However, SF 6 Gas-insulated apparatus (SF for short) 6 Electrical equipment, such as gas insulated switchgear GIS, gas insulated circuit breaker GCB, gas insulated transformer GIT, and gas insulated line or pipeline GIL) during manufacturing, transportation, installation, inspection, and operation, various insulation defects inevitably occur inside the equipment, such as metal burrs on the conductor, loose parts or poor contact, air gaps formed by peeling the conductor from the support insulator, remnants after inspection, and metal micro-scale in the cavityGranules, etc., all of which cause SF 6 Insulation defects are formed in different degrees inside the device, so that the electric field inside the device is distorted, and Partial Discharge (PD) is generated.
When severe PD occurs, on one hand, PD can accelerate further damage to the internal insulation of equipment, finally, insulation failure causes power failure accident, and SF in operation 6 The equipment is a potential hidden trouble and is called as insulated tumor; on the other hand, PD is a characteristic quantity for effectively representing the insulation condition by the pair of SF 6 The PD of the electrical equipment is detected and the pattern recognition is carried out, so that SF can be found to a great extent 6 Insulation defects and types present inside the device. Thus, identifying the occurrence of insulation defects versus ensuring SF 6 The safe and reliable operation of the electrical equipment has important practical significance.
Disclosure of Invention
In view of the above-mentioned defects in the prior art, the present invention provides an insulation defect for identifying the partial discharge characteristics of sulfur hexafluoride based on a support vector machine, so as to ensure SF 6 The equipment is safe and reliable to operate, the discrimination of the equipment to various defects is improved, and the treatment efficiency of the insulation fault can be improved.
The invention aims to realize the technical scheme that the method for identifying the insulation defect of the sulfur hexafluoride partial discharge characteristic based on the support vector machine comprises the following steps: the identification process of the insulation defect comprises the following steps:
s1: constructing a support vector machine classifier;
s2: optimizing parameters of a support vector machine;
s3: acquiring decomposition data of sulfur hexafluoride;
s4: and the support vector machine identifies the decomposition data of the sulfur hexafluoride.
Further, the construction process of the support vector machine classifier in the step S1 is as follows:
s11: assume a sample set ofWherein x is i ∈R n For the input vector, N represents the number of samples, y i E { +/-1 } is a class label, and the corresponding vector of the input vector in the high-dimensional space F is phi (x);
s12: constructing a hyperplane F (x) = (w · phi (x)) + b in an F space, wherein w is a normal vector of the hyperplane and b is an offset;
s13: there is a hyperplane to maximize the separation of the two classes of samples, this hyperplane is called the optimal classification plane, when w = w * ,b＝b * 。
Further, S14: solving the optimal classification plane f (x) = (w) of step S13 * ·φ(x))+b * The problem of (2) can be converted into:
wherein ξ i Is a relaxation factor, C&0 is called a penalty factor, the optimal classification surface solving problem is a quadratic programming problem with linear constraint, and the solved expression is as follows:
s.t.α i ≥0,β i ≥0 (4)
wherein alpha is i ，β i Is a lagrange multiplier.
Further, S15: converting step S14 into a dual problem may result:
determining alpha from the formula (5) and the formula (6) i Andwherein, K (x) i ,x j )＝φ(x i )·φ(x j ) Referred to as a kernel function.
Further, S16: the kernel function in step S15 adopts a gaussian radial basis kernel, and its expression is:
s17: from the KKT condition, the formula (8) can be obtained, from which b can be obtained * ；
S18: finally, an optimal classification surface f (x) = (w) can be determined * ·φ(x))+b * And obtaining a classification discriminant function according to the optimal classification surface as follows:
further, the classifier construction of the support vector machine in step S1 further includes:
s101: if the samples have k types, constructing k (k-1)/2 classifiers;
s102: when classifying the samples, inputting the samples into the k (k-1)/2 classifiers at the same time;
s103: and counting the classification result, and taking the category with the most votes in the result as the category to which the sample belongs.
Further, the optimization process of the support vector machine parameters in step S2 is as follows:
s21: let the search space be D-dimensional, where n particles are in flight and the position of the ith particle in space be represented as vector X i ＝(x i1 ,x i2 ,...,x iD )；
S22: let P be the optimal position where the ith particle has been in the past i ＝(p i1 ,p i2 ,...,p iD ) Where the past optimal position P of the g-th particle g Is all P i (i = 1.... N), the flight speed of the i-th particle is the vector V i ＝(v i1 ,v i2 ,...,v iD )；
S23: in the process of searching for the optimal position, each particle updates the current flight speed and position according to the best position found by the particle and the best position found by the whole group, and the position formula of the particle is as follows:
the velocity equation for the particle is:
in the formula, c 1 ，c 2 Is a normal number, called acceleration factor; r is 1 ，r 1 Is [0,1]A random number in between; w is called an inertia factor; the position variation range of the D (1. Ltoreq. D. Ltoreq.D) dimension is [ -X dmax ,X dmax ]The speed variation range is [ -V ] dmax ,V dmax ]；
S24: and if the position and the speed exceed the boundary range in the iteration, taking a boundary value, randomly generating the initial position and the speed of the particle swarm, and then iterating according to a formula (10) and a formula (11) until a solution is solved.
Further, the sulfur hexafluoride decomposition data obtained in step S3 includes training data of the support vector machine, test data of the support vector machine, the number of protrusions in the sample, the number of air gaps in the sample, and the number of particles in the sample.
Due to the adoption of the technical scheme, the invention has the following advantages:
(1) Ensuring SF 6 Safe and reliable operation of the equipment, for grasping SF 6 The equipment insulation operation condition and the construction state maintenance system have important academic and practical values;
(2) The discrimination of the insulating defect identifying device on various defects is improved, so that the insulating defect identifying device can better represent the characteristics of various insulating defects;
(3) The support vector machine is utilized to establish the insulation defect intelligent diagnosis system based on the decomposition components, so that the treatment efficiency of the insulation fault can be improved.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof.
Drawings
The drawings of the invention are illustrated as follows:
fig. 1 is a flow chart of insulation defect identification according to the present invention.
FIG. 2 is a diagram of a multi-class support vector machine according to the present invention.
Fig. 3 is a fitness curve graph of the optimization process of the present invention.
Detailed Description
The invention is further illustrated by the following examples in conjunction with the drawings.
Example (b): as shown in fig. 1-3; an insulation defect based on the partial discharge characteristic of sulfur hexafluoride is identified by a support vector machine, which comprises the following components: the identification process of the insulation defect comprises the following steps:
s1: constructing a support vector machine classifier;
s2: optimizing parameters of a support vector machine;
s3: acquiring decomposition data of sulfur hexafluoride;
s4: and the support vector machine identifies the decomposition data of the sulfur hexafluoride.
A Support Vector Machine (SVM) is used as a classifier for identifying insulation defects, the SVM is proposed by Vapnik, and the basic idea is to map an input Vector to a high-dimensional space F through nonlinear mapping phi and then find an optimal hyperplane in the high-dimensional space so as to maximize the interval between samples. The construction process of the support vector machine classifier in the step S1 is as follows:
s11: assume a sample set ofWherein x is i ∈R n For the input vector, N represents the number of samples, y i E { +/-1 } is a class label, and the corresponding vector of the input vector in the high-dimensional space F is phi (x);
s12: constructing a hyperplane F (x) = (w · phi (x)) + b in an F space, wherein w is a normal vector of the hyperplane and b is an offset;
s13: there is a hyperplane to maximize the separation of the two classes of samples, this hyperplane is called the optimal classification plane, when w = w * ,b＝b * 。
S14: solving the optimal classification plane f (x) = (w) of step S13 * ·φ(x))+b * The problem of (2) can be converted into:
wherein xi is i Is a relaxation factor, C&0 is called a penalty factor, the optimal classification surface solving problem is a quadratic programming problem with linear constraint, and can be solved by a Lagrange multiplier method, namely:
s.t.α i ≥0,β i ≥0 (4)
wherein alpha is i ，β i Is a lagrange multiplier.
S15: converting step S14 into a dual problem may result:
determining alpha from the formula (5) and the formula (6) i Andwherein, K (x) i ,x j )＝φ(x i )·φ(x j ) Referred to as a kernel function.
S16: the kernel function in step S15 adopts a gaussian radial basis kernel, and its expression is:
s17: from the KKT condition, the formula (8) can be obtained, from which b can be obtained * ；
S18: finally, the optimal classification plane f (x) = (w) can be obtained * ·φ(x))+b * And obtaining a classification discriminant function according to the optimal classification surface as follows:
the principle of SVM can see that SVM can only be used for two-class problem, if the multi-class problem needs to be solved, a plurality of SVM classifiers must be constructed, and the currently commonly used method for constructing multi-classifiers has algorithms such as one-to-one, one-to-many, binary tree and the like, and the invention adopts a one-to-one method, and the basic idea is as follows: an SVM classifier is constructed between any two classes.
The classifier construction of the support vector machine in the step S1 further includes:
s101: if the samples have k types, constructing k (k-1)/2 classifiers;
s102: when classifying the samples, inputting the samples into the k (k-1)/2 classifiers at the same time;
s103: and counting the classification result, and taking the category with the most votes in the result as the category to which the sample belongs.
The partial discharge types identified by gas in the invention have three types, and 3 SVM two classifiers are required to be constructed, the schematic diagram of which is shown in FIG. 2, wherein the SVM is shown in the figure NG The support vector machine is used for classifying two types of insulation defects, namely metal protrusions and air gaps; SVM in the figure NP The support vector machine is used for classifying two types of insulation defects, namely metal protrusions and free metal particles; SVM in the figure GP Is a support vector machine for classifying two types of insulation defects, free metal particles and air gaps.
It can be seen from the above-constructed support vector machine that two parameters need to be determined in advance, that is, the penalty factor C and the kernel function parameter σ greatly affect the recognition performance of the SVM, and if the selection is not proper, the classification performance of the SVM is greatly reduced.
The basic principle of the PSO algorithm is as follows:
s21: let the search space be D-dimensional, where n particles are in flight and the position of the ith particle in space be represented as vector X i ＝(x i1 ,x i2 ,...,x iD ) (ii) a Each position represents a possible solution in the space, the fitness corresponding to the position can be obtained by substituting the position vector into the fitness function, and the quality of the position can be evaluated according to the fitness.
S22: let P be the optimal position where the ith particle has been in the past i ＝(p i1 ,p i2 ,...,p iD ) Where the past optimal position P of the g-th particle g Is all P i (i = 1.... N), the flight speed of the i-th particle is the vector V i ＝(v i1 ,v i2 ,...,v iD )；
S23: in the process of searching for the optimal position, each particle updates the current flight speed and position according to the best position found by the particle and the best position found by the whole group, and the position formula of the particle is as follows:
the velocity formula for the particles is:
x id (t+1)＝x id (t)+v id (t+1) (11)
1≤i≤n1≤d≤D
in the formula, c 1 ，c 2 Is a normal number, called acceleration factor; r is a radical of hydrogen 1 ，r 1 Is [0,1]A random number in between; w is called an inertia factor; w is large for large-scale exploration (exploration) of the solution space and small for small-scale excavation (exploration). The position variation range of the D (1. Ltoreq. D. Ltoreq.D) dimension is [ -X dmax ,X dmax ]The speed variation range is [ -V ] dmax ,V dmax ]；
S24: and if the position and the speed exceed the boundary range in the iteration, taking a boundary value, randomly generating the initial position and the speed of the particle swarm, and then iterating according to a formula (10) and a formula (11) until a solution is solved.
Acquiring data; the sulfur hexafluoride decomposition data obtained in the step S3 include training data of the support vector machine, test data of the support vector machine, the number of protrusions in the sample, the number of air gaps in the sample, and the number of particles in the sample.
The invention carries out partial discharge experiments under three insulation defects of a needle plate, particles and air gaps, and measures SO in the decomposed gas under each partial discharge 2 F 2 、SOF 2 、CO 2 And CF 4 The contents of four characteristic components, 72 groups of data are collected under each defect, and 216 groups of SF under three defects are obtained 6 Decomposing component content data, wherein 108 groups of data are used for training a classifier, the other 108 groups are used for testing the performance of the classifier, and the distribution of a data set is shown in table 1;
TABLE 1 sample composition
Training and testing a classifier; as previously mentioned, the present invention selects c (SO) 2 F 2 )/c(SOF 2 )、c(CF 4 )/c(CO 2 ) And C (CO) 2 +CF 4 )/c(SOF 2 +SO 2 F 2 ) And taking the content ratio of the three components as characteristic quantity, and training the constructed support vector machine classifier by utilizing a training set.
During training, k-fold cross validation is adopted to measure the performance of the classifier, according to the size of a sample, k =3 is taken, namely the training sample is equally divided into three groups, wherein two groups are used as a training set for training the classifier, the other group is used as a test set for testing the classification accuracy of the classifier, then the data groups corresponding to the test set and the training set are replaced, and the process is circulated for three times until each group of data is used as the test set.
Using POS algorithm to support in training classifierThe penalty factor C and the kernel function parameter sigma of the vector machine are optimized, the (C, delta) is used as a search space, and the search range of the C is [0.1,100 ]]The search range of σ is taken [0.01,100]The fitness function is selected as the classification accuracy obtained by 3-fold cross validation, the highest accuracy obtained by three tests in the cross validation under each group (C, delta) is used as the optimal fitness, the average value of the obtained three accuracies is used as the average fitness, the inertia factor w =1, and the acceleration factor C is used as 1 、c 2 The population number n =20 and the iteration number t =200 are respectively taken as 1.5 and 1.7, the fitness curve obtained in the optimization process is shown in fig. 3, the final optimization result is (C, δ) = (73.68,44.36), and the corresponding optimal fitness at this time is 88.89%.
TABLE 2 results of the classification
The SVM classifier obtained by the above method was tested using 108 test samples, and the test results are shown in Table 2, which indicates that the overall classification accuracy reaches 97.72%, indicating that SF is used 6 The decomposition components can identify different partial discharge types to achieve good identification effect.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (8)
1. The method for identifying the insulation defect of the sulfur hexafluoride based on the partial discharge characteristic of the sulfur hexafluoride by the support vector machine is characterized in that the identification process of the insulation defect is as follows:
s1: constructing a support vector machine classifier;
s2: optimizing parameters of a support vector machine;
s3: acquiring decomposition data of sulfur hexafluoride;
s4: and the support vector machine identifies the decomposition data of the sulfur hexafluoride.
2. The support vector machine-based method for identifying the insulation defect of the sulfur hexafluoride partial discharge characteristic, as claimed in claim 1, wherein the support vector machine classifier in the step S1 is constructed by the following process:
s11: assume a sample set ofWherein x is i ∈R n For the input vector, N represents the number of samples, y i E { +/-1 } is a class label, and the corresponding vector of the input vector in the high-dimensional space F is phi (x);
s12: constructing a hyperplane F (x) = (w · phi (x)) + b in an F space, wherein w is a normal vector of the hyperplane and b is an offset;
s13: there is a hyperplane to maximize the separation of the two classes of samples, this hyperplane is called the optimal classification plane, when w = w * ,b＝b * 。
3. The support vector machine-based method for identifying sulfur hexafluoride partial discharge characteristics insulation defects of claim 2,
s14: solving the optimal classification plane f (x) = (w) of step S13 * ·φ(x))+b * The problem of (2) can be converted into:
wherein ξ i Is a relaxation factor, C>, 0 is called penalty factor, the problem of solving the optimal classification surface is a quadratic rule with linear constraintDividing the problem, and solving the following expression:
s.t.α i ≥0,β i ≥0 (4)
wherein alpha is i ，β i Is a lagrange multiplier.
4. The support vector machine-based method for identifying sulfur hexafluoride partial discharge characteristics insulation defects of claim 3,
s15: converting step S14 into a dual problem may result:
determining alpha from the formula (5) and the formula (6) i Andwherein, K (x) i ,x j )＝φ(x i )·φ(x j ) Referred to as a kernel function.
5. The support vector machine-based method for identifying sulfur hexafluoride partial discharge characteristics insulation defects of claim 4,
s16: the kernel function in step S15 adopts a gaussian radial basis kernel, and its expression is:
s17: from the KKT condition, the formula (8) can be obtained, from which b can be obtained * ；
S18: finally, an optimal classification surface f (x) = (w) can be determined * ·φ(x))+b * And obtaining a classification discriminant function according to the optimal classification surface as follows:
6. the support vector machine-based method for identifying sulfur hexafluoride partial discharge characteristics insulation defects in claim 1, wherein the support vector machine classifier structure in step S1 further includes:
s101: if the samples have k types, constructing k (k-1)/2 classifiers;
s102: when classifying the samples, inputting the samples into the k (k-1)/2 classifiers at the same time;
s103: and counting the classification result, and taking the category with the most votes in the result as the category to which the sample belongs.
7. The support vector machine-based method for identifying the insulation defect of the sulfur hexafluoride partial discharge characteristic, as claimed in claim 1, wherein the optimization process of the support vector machine parameters in the step S2 is as follows:
s21: let the search space be D-dimensional, where n particles are in flight and the position of the ith particle in space be represented as vector X i ＝(x i1 ,x i2 ,...,x iD )；
S22: let P be the optimal position where the ith particle has been in the past i ＝(p i1 ,p i2 ,...,p iD ) Wherein the past optimal position P of the g-th particle g Is all P i (i = 1.... N), the flight speed of the i-th particle is the vector V i ＝(v i1 ,v i2 ,...,v iD )；
S23: in the process of searching for the optimal position, each particle updates the current flight speed and position according to the best position found by the particle and the best position found by the whole group, and the position formula of the particle is as follows:
the velocity formula for the particles is:
in the formula, c 1 ，c 2 Is a normal number, called acceleration factor; r is 1 ，r 1 Is [0,1]A random number in between; w is called inertia factor; the position variation range of the D (1. Ltoreq. D. Ltoreq.D) dimension is [ -X dmax ,X dmax ]The speed variation range is [ -V ] dmax ,V dmax ]；
S24: and if the position and the speed exceed the boundary range in the iteration, taking a boundary value, randomly generating the initial position and the speed of the particle swarm, and then iterating according to a formula (10) and a formula (11) until a solution is solved.
8. The support vector machine-based method for identifying the insulation defect of the sulfur hexafluoride partial discharge characteristic, as claimed in claim 1, wherein the sulfur hexafluoride decomposition data obtained in step S3 includes training data of the support vector machine, test data of the support vector machine, the number of protrusions in the sample, the number of air gaps in the sample, and the number of particles in the sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711049428.5A CN107862332A (en) | 2017-10-31 | 2017-10-31 | Insulation defect based on SVMs identification sulfur hexafluoride Characteristics of Partial Discharge |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711049428.5A CN107862332A (en) | 2017-10-31 | 2017-10-31 | Insulation defect based on SVMs identification sulfur hexafluoride Characteristics of Partial Discharge |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107862332A true CN107862332A (en) | 2018-03-30 |
Family
ID=61697035
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711049428.5A Pending CN107862332A (en) | 2017-10-31 | 2017-10-31 | Insulation defect based on SVMs identification sulfur hexafluoride Characteristics of Partial Discharge |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107862332A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926620A (en) * | 2021-01-21 | 2021-06-08 | 国网山东省电力公司检修公司 | Transformer bushing partial discharge defect type identification method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103439594A (en) * | 2013-08-02 | 2013-12-11 | 广东电网公司电力科学研究院 | System and method for diagnosing faults of SF6 electric device |
| CN205920187U (en) * | 2015-12-01 | 2017-02-01 | 国网电力科学研究院武汉南瑞有限责任公司 | Direct current gas -insulated electrical equipment partial discharge decomposes simulation experiment device |
-
2017
- 2017-10-31 CN CN201711049428.5A patent/CN107862332A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103439594A (en) * | 2013-08-02 | 2013-12-11 | 广东电网公司电力科学研究院 | System and method for diagnosing faults of SF6 electric device |
| CN205920187U (en) * | 2015-12-01 | 2017-02-01 | 国网电力科学研究院武汉南瑞有限责任公司 | Direct current gas -insulated electrical equipment partial discharge decomposes simulation experiment device |
Non-Patent Citations (1)
| Title |
|---|
| 刘帆: "局部放电下六氟化硫分解特性与放电类型辨识及影响因素校正", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926620A (en) * | 2021-01-21 | 2021-06-08 | 国网山东省电力公司检修公司 | Transformer bushing partial discharge defect type identification method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zheng et al. | Fault diagnosis of power transformers using multi-class least square support vector machines classifiers with particle swarm optimisation | |
| Tang et al. | Partial discharge recognition through an analysis of SF 6 decomposition products part 2: feature extraction and decision tree-based pattern recognition | |
| Lazzaretti et al. | Novelty detection and multi-class classification in power distribution voltage waveforms | |
| Mitiche et al. | 1D‐CNN based real‐time fault detection system for power asset diagnostics | |
| Cai et al. | Intelligent DC arc-fault detection of solar PV power generation system via optimized VMD-based signal processing and PSO–SVM classifier | |
| Tang et al. | Assessment of PD severity in gas‐insulated switchgear with an SSAE | |
| Janani et al. | Classification of simultaneous multiple partial discharge sources based on probabilistic interpretation using a two-step logistic regression algorithm | |
| CN108764265A (en) | A kind of method for diagnosing faults based on algorithm of support vector machine | |
| Khodsuz et al. | Monitoring and identification of metal–oxide surge arrester conditions using multi‐layer support vector machine | |
| Zeng et al. | Feature extraction and severity assessment of partial discharge under protrusion defect based on fuzzy comprehensive evaluation | |
| Zhu et al. | Semi-supervised ensemble learning framework for accelerating power system transient stability knowledge base generation | |
| Herath et al. | Comparison of supervised machine learning techniques for PD classification in generator insulation | |
| Zheng et al. | Applications of support vector machine and improved k-Nearest Neighbor algorithm in fault diagnosis and fault degree evaluation of gas insulated switchgear | |
| Ali et al. | Reliable IoT paradigm with ensemble machine learning for faults diagnosis of power transformers considering adversarial attacks | |
| Dong et al. | Wafer yield prediction using derived spatial variables | |
| CN110888025B (en) | GIS equipment fault judgment method based on machine learning | |
| Gao et al. | Pattern recognition of unknown partial discharge based on improved SVDD | |
| Wang et al. | Wavelet transform‐based feature extraction for detection and classification of disturbances in an islanded micro‐grid | |
| Wang et al. | Novel metric-based meta-learning model for few-shot diagnosis of partial discharge in a gas-insulated switchgear | |
| Cheemala et al. | Transformer incipient fault diagnosis using machine learning classifiers | |
| Bin et al. | Identification of ultra‐high‐frequency PD signals in gas‐insulated switchgear based on moment features considering electromagnetic mode | |
| Ye et al. | A corona recognition method based on visible light color and machine learning | |
| CN107862332A (en) | Insulation defect based on SVMs identification sulfur hexafluoride Characteristics of Partial Discharge | |
| CN112132226A (en) | Method for diagnosing defects of +/-1100 kV composite insulator | |
| Mittal et al. | Optimal feature selection for SVM based fault diagnosis in power transformers |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180330 |