CN113899997A - Transformer insulation state diagnosis method based on improved support vector machine - Google Patents
Transformer insulation state diagnosis method based on improved support vector machine Download PDFInfo
- Publication number
- CN113899997A CN113899997A CN202111322970.XA CN202111322970A CN113899997A CN 113899997 A CN113899997 A CN 113899997A CN 202111322970 A CN202111322970 A CN 202111322970A CN 113899997 A CN113899997 A CN 113899997A
- Authority
- CN
- China
- Prior art keywords
- data
- transformer
- support vector
- vector machine
- svm
- 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
- 238000012706 support-vector machine Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000009413 insulation Methods 0.000 title claims abstract description 22
- 238000003745 diagnosis Methods 0.000 title claims abstract description 16
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 9
- 239000011087 paperboard Substances 0.000 claims abstract description 7
- 238000006116 polymerization reaction Methods 0.000 claims abstract description 7
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims abstract description 4
- 239000013598 vector Substances 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 3
- 230000009977 dual effect Effects 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 claims description 2
- 238000012887 quadratic function Methods 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims 1
- 230000007613 environmental effect Effects 0.000 abstract description 5
- 239000000123 paper Substances 0.000 abstract description 4
- 238000011084 recovery Methods 0.000 abstract description 4
- 230000004044 response Effects 0.000 abstract description 2
- 238000004804 winding Methods 0.000 abstract 1
- 230000032683 aging Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000009467 reduction Effects 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 4
- 238000013501 data transformation Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000004927 fusion Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004868 gas analysis Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005201 scrubbing Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Images
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
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
-
- 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
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1263—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
Landscapes
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Housings And Mounting Of Transformers (AREA)
Abstract
The invention provides a transformer insulation state diagnosis method based on an improved Support Vector Machine (SVM), which is mainly based on an SVM theory and improves a traditional SVM algorithm. The method simulates a transformer winding structure, namely an XY model, through a time domain dielectric response, namely a return voltage method, and adopts the return voltage method to measure the characteristic quantity of the existing transformer oil paper insulation system: the recovery voltage peak, the peak time, and the recovery voltage initial slope. And (3) extracting samples which accord with training for preprocessing by using a large number of measured experimental characteristic quantities, and predicting the polymerization degree of the unknown oil-immersed paperboard by combining environmental factors and adopting SVM theoretical modeling according to the polymerization degree of the existing oil-immersed paperboard so as to achieve the purpose of diagnosing the insulation state of the unknown transformer.
Description
Technical Field
The invention designs a transformer insulation state diagnosis method based on an improved support vector machine, and belongs to the technical field of power equipment insulation state evaluation.
Background
In the modern times, electric energy is an important dependence of basic life and work of people, and the scale of a power grid is larger and larger, so that the basis for ensuring the safe and stable operation of power equipment is the safety of the power grid. The failure of the transformer oil paper insulation system due to aging can cause power failure, and huge economic loss is caused to the society.
In early days, China generally waits until a transformer has a certain fault and then carries out processing and maintenance. Therefore, faults are difficult to predict, the position of the fault and the degree of influence of the fault are judged, and many unsafe factors and hidden dangers are generated for the operation of the power grid. Furthermore, it is not feasible to shut down the operating electrical equipment and perform the relevant checks regularly. This method is not preferable because it is blind and mandatory, does not fully consider the actual operating state of the power equipment, and wastes a lot of manpower and material resources.
In the middle of the twentieth century, real-time online monitoring technology and maintenance technology become mature day by day, online data monitoring of transformers for fault diagnosis is scientific, and online fault diagnosis of transformers becomes a development trend. By analyzing the working state and the operating parameters of the transformer and calling the overhaul history and the factory information of the transformer, the transformer can be judged to be in which fault by owning the data. At present, the mainstream method for diagnosing the fault of the transformer is to determine the local state or even the whole state of the transformer insulation by performing chromatographic analysis on the gas dissolved in the transformer oil according to the type of the gas dissolved in the transformer oil and the content of the gas dissolved in the transformer oil, so that the fault type can be reflected in time, and thus, maintenance personnel can conveniently and timely remove the fault and maintain the fault. However, due to randomness, ambiguity and uncertainty of transformer fault causes, it is limited to diagnose the fault state of the transformer by the content of gas in the transformer oil only. Because a single dissolved gas analysis information source cannot completely reflect the fault state of the transformer. When a certain fault occurs, the state of the fault is associated with a plurality of pieces of information data, or one piece of information data reflects a plurality of faults, and further, different fault states of the transformer may exist simultaneously. Therefore, after knowing these characteristics, it is necessary to establish a fault diagnosis model based on multi-source information to identify the state of the equipment. If a fault occurs, the fault of the transformer can be reflected by the information state of each single component. Meanwhile, with the development of artificial intelligence of the electrical industry in China, the computerized technology development of the power transformer monitoring system is mature day by day, and data of the transformer can be acquired more conveniently. Based on the situation, a proper data index in the transformer can be selected, and a transformer fault diagnosis model overlapped by various transformer information is established, so that the fault of the transformer can be diagnosed.
The multi-source information fusion of transformer fault diagnosis is to combine the information and data of the transformer obtained by various monitoring according to a certain optimization criterion, and finally to systematically explain and describe the transformer fault and the generation reason. The method not only analyzes the relationship between the condition of the insulating oil at the main insulating part of the transformer and the fault of the transformer, but also analyzes the principle of generating characteristic products in the aging process of the insulating oil, so that a diagnostic model of the oil-paper insulating fault of the transformer is established to diagnose the fault of the transformer.
The support vector machine is a novel learning model and is mainly used for problem classification and regression. The idea of the support vector machine is to convert the inseparable problem of low-dimensional nonlinearity into the separable problem of high-dimensional linearity, so that the problem is processed more quickly and the difficulty is reduced. Therefore, it is very important to invent a transformer insulation diagnosis method based on an improved support vector machine.
Disclosure of Invention
The invention aims to provide a transformer insulation state diagnosis method based on an improved support vector machine, because the aging and the moisture of a transformer can influence the insulation performance of the transformer, and the insulation state diagnosis is difficult to make in the face of multi-source information, the invention proposes to use the improved support vector machine to perform multi-source information fusion on environmental variables of the transformer, such as micro-water content, environmental temperature, aging days and the like, and adopted return voltage method experiment characteristic quantities, such as initial slope, return voltage peak value and peak value time, to predict the polymerization degree of an unknown transformer model oil-immersed paperboard, thereby achieving the detection of the insulation state of the transformer.
Drawings
Other advantages, features and details of the present invention will be more fully understood from the following description taken in conjunction with the accompanying drawings.
FIG. 1 is a flow diagram of the operation of an improved support vector machine;
FIG. 2 is a test flow chart of a recovery voltage method;
FIG. 3 is a graph of the relative error of predicted versus actual values for the improved support vector machine.
Detailed Description
The invention is implemented as follows:
a transformer insulation state diagnosis method based on an improved support vector machine comprises the following steps:
s1: collecting related images by adopting equipment such as a direct-current power supply, an electrode box for insulation test, an electrometer 6517B and the like;
s2: extracting image related parameters by using LabVIEW software;
s3: preprocessing the experimental characteristic quantity;
s4: programming by using MATLAB software, building a support vector machine operation model, and improving an algorithm;
s5: and putting the preprocessed data into a trained model for processing, and predicting unknown parameters according to known data.
In step S1, a time domain dielectric response test method, i.e., a return voltage method, is used to collect an experimental image.
In the step S2, LabVIEW software is used to extract image-related parameters, and the main parameters include initial slope, return voltage peak, peak time, environmental variables, environmental temperature, aging days, micro-water content, and the like.
The step S3 is used for preprocessing the experimental characteristic quantity, and the main method of data preprocessing is a reduction method based on a rough set theory; a concept tree based data enrichment method; information theory ideas and knowledge discovery; an attribute selection method based on statistical analysis; and (4) genetic algorithm. Common data preprocessing methods include: data cleaning, data integration, data transformation and data reduction:
(1) data cleansing (Data cleansing): the purpose of data scrubbing is not just to eliminate errors, redundancy and data noise. The goal is to reconcile the various data sets generated according to different, incompatible rules, i.e. to remove extraneous data and noise;
(2) data Integration (Data Integration) refers to merging heterogeneous Data in a multi-file or multi-database operating environment to solve the semantic ambiguity problem. The part mainly relates to data selection, data conflict and inconsistent data processing, namely, data in a plurality of data sources are combined and stored in a consistent data storage;
(3) data Transformation (Data Transformation): a characterization of the data is sought. Using dimension transformation or transformation (including normalization, switching and projection) to reduce the number of effective variables or find invariants of data, i.e. to convert the original data into a form suitable for data mining;
(4) data Reduction (Data Reduction): based on an understanding of the discovery task and the content of the data itself, data reduction seeks to rely on useful features of the discovery target's expressive data to reduce the data model, thereby minimizing the amount of data while preserving as much of the original appearance as possible. There are two main approaches to attributes and records in databases: attribute selection and data sampling compressed data, i.e., compressed data.
In the step S4, MATLAB software is used for programming, a support vector machine operation model is built, and an algorithm is improved, so that a calculation formula can be obtained as follows:
the data of the support vector machine can be expressed as
yiEither 1 or-1, meaning xiTo which class it belongs. Each xiIs a p-dimensional vector representing all the eigenvalues (variables) of the data points. Will yiXi vector set of 1 and yiThe best separated hyperplane for the-1 vector set is:
wherein,
is the normal vector of the hyperplane, and b is the offset of the hyperplane from the origin, i.e., the intercept.If the data points are linearly separable, the hard margin (hard margin) can be expressed as:
mathematically, given a set of points xi that belong to two linearly separable classes w1 and w2, the distance of any data point to the hyperplane is equal to the ratio of | g (x) | to | ω | |, and the constructing of the hyperplane for the support vector machine is primarily aimed at finding ω, b, for the closest data point (support vector) belonging to class w1, g (x) equals 1, and the closest to w2 is-1. Therefore, there are:
this involves an optimization problem that minimizes the objective function:
and the hyperplane constraint:
s.t.yi(ωTx+b)≥1(ξ≥0,i＝1,2···n) (5)
from the lagrange function:
where ω, b is called the original variable, λiReferred to as the langerhan multiplier. These multipliers thus limit the search space of the solution to a set of feasible values, given the constraints. In the presence of inequality constraints, the Karush-Kuhn-tucker (KKT) condition is widely used, since the equation satisfies the KKT condition:
according to the constraint conditions of the formula (6) and the formula (7), the problem of finding the optimal classification surface can be changed into the optimization problem of the convex quadratic programming dual problem:
wherein λ isiThe lagrange operator is equal to or more than 0, which is the only solution for the problem of quadratic function optimization. If λi *And the optimal solution satisfies the following conditions:
where λ i is a sample other than zero, i.e., a support vector. b is a classification threshold, and the constraint conditions are as follows:
solving, and obtaining an optimal classification surface after solving:
in the equation, the summation is actually performed only for the support vectors, since the corresponding λ i value of the non-support vectors is equal to 0. The above formula is a general expression of SVM.
In practical terms, when the two classification points cannot be completely separated for the optimal classification surface, in order to obtain a balance between empirical risk and popularization performance, fault tolerance is introduced, namely a non-negative relaxation variable ζ is added to serve as a 'fine' number of wrong classification samples. The classification face at this time satisfies:
ω·x+b＝0 (12)
when 0 < ζiThen, the sample point i is correctly classified; at that time, sample point x is misclassified. Therefore, the original target function 0.5| | ω | | non-woven phosphor2Adding a penalty term, the objective function becomes:
wherein C is a constant, ζiIs a penalty factor. Approximately the same as the linear separable case, equation (15) can be implemented by following quadratic programming:
the algorithm adopts a Gaussian kernel function and an RBF kernel function:
K(x,xi)＝exp(-γ*||x-xi||2) (16)
for the nonlinear classification problem, once the classification result of the simple optimal classification surface in the initial space is not satisfactory, the initial spatial nonlinearity can be converted into the linear problem of a higher-dimensional space, and the optimal classification surface is obtained in the higher-dimensional space. And performing MATLAB programming by using the derived formula, and training the existing data, wherein a flow chart of an improved support vector machine is shown in figure 1.
And step S5, the preprocessed data is put into a trained model for processing, the unknown parameters are predicted according to the known data, the parameters of the unknown oil-immersed paper board are predicted by utilizing the known polymerization degree, and the purpose of diagnosing the insulation state of the transformer is achieved.
The invention has the beneficial effects that:
the method combines an improved support vector machine with a return voltage method for the first time, and predicts the polymerization degree of the oil-immersed paperboard by adopting an algorithm mode under the condition that the normal operation of the transformer is damaged without core suspension sampling. The peak value, the peak time and the initial slope of the recovery voltage are important parameters of insulation aging, and the diagnosis method is simple and easy. The invention can be widely applied to the oil paper insulation aging diagnosis of various transformers.
This embodiment is only illustrative of the patent and does not limit the scope of protection thereof, and those skilled in the art can make modifications to its part without departing from the spirit of the patent.
Claims (2)
1. A transformer insulation state diagnosis method based on an improved support vector machine is characterized by comprising the following steps:
s1: collecting related images by adopting equipment such as a direct-current power supply, an electrode box for insulation test, an electrometer 6517B and the like;
s2: extracting image related parameters by using LabVIEW software;
s3: preprocessing the experimental characteristic quantity;
s4: programming by using MATLAB software, building a support vector machine operation model, and improving an algorithm;
s5: and putting the preprocessed data into a trained model for processing, and predicting unknown parameters according to known data.
2. The method for diagnosing the insulation state of the transformer based on the improved support vector machine according to claim 1, wherein MATLAB software is used for programming, a support vector machine operation model is built, and an algorithm is improved, specifically, the improved support vector machine algorithm is used for predicting the polymerization degree of the oil-immersed paperboard, and the operation process is as follows:
(1) establishing and training an SVM model by using an LIBSVM tool, and calling all data obtained by an experiment;
(2) taking the first two thirds of data as a training set, and taking the last one third of data as a prediction set;
(3) normalizing the data;
(4) mapping on a high dimension by a kernel function by using a peak voltage, a slope and a main time constant;
(5) in SVM modelThe penalty factor c and the threshold g need to be in the range 10-10-1010If 10, the-10≤cg≤1010If not, the next step is directly carried out, otherwise, the optimization needs to be carried out again;
(6) searching an optimal hyperplane by using the optimized c and g models, and predicting the polymerization degree of the oil-immersed paperboard by combining with the experimental characteristic quantity;
mathematically, a set of points x is giveniThey belong to two linearly separable classes w1 and w2, the distance from any data point to the hyperplane is equal to the ratio of | g (x) | to | ω | |, the main purpose of the support vector machine is to find ω, b, for the nearest data point (support vector) belonging to the class w1, g (x) is equal to 1, and the nearest to w2 is-1, ω, b is called the original variable, λiCalled the langerhan multiplier; these multipliers thus limit the search space of the solution to a set of feasible values, given the constraints; according to the constraint conditions, the problem of finding the optimal classification surface can be changed into the optimization problem of the convex quadratic programming dual problem:
wherein λ isiThe lagrange operator is more than or equal to 0, which is a unique solution for the problem of quadratic function optimization, if lambda isi *And the optimal solution satisfies the following conditions:
solving, and obtaining an optimal classification surface after solving:
in the equation, the summation is actually performed only for the support vectors, due to the corresponding λ of the non-support vectorsiThe value is equal to 0 and the above equation is a general expression for SVMs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111322970.XA CN113899997A (en) | 2021-11-10 | 2021-11-10 | Transformer insulation state diagnosis method based on improved support vector machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111322970.XA CN113899997A (en) | 2021-11-10 | 2021-11-10 | Transformer insulation state diagnosis method based on improved support vector machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113899997A true CN113899997A (en) | 2022-01-07 |
Family
ID=79193795
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111322970.XA Pending CN113899997A (en) | 2021-11-10 | 2021-11-10 | Transformer insulation state diagnosis method based on improved support vector machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113899997A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106771755A (en) * | 2016-12-28 | 2017-05-31 | 江苏大学 | A kind of transformer oil Aging of Oil-paper Insulation in Oil state analyzing method based on return voltage |
| CN108872803A (en) * | 2018-03-29 | 2018-11-23 | 福建工程学院 | A kind of detection method of the transformer insulation state based on dielectric return voltage |
| CN112485609A (en) * | 2020-10-19 | 2021-03-12 | 重庆大学 | Raman spectrum diagnosis method for insulation aging of transformer oil paper |
| CN113468461A (en) * | 2020-03-30 | 2021-10-01 | 电子科技大学中山学院 | Oil-immersed transformer fault diagnosis method based on support vector machine and genetic algorithm |
-
2021
- 2021-11-10 CN CN202111322970.XA patent/CN113899997A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106771755A (en) * | 2016-12-28 | 2017-05-31 | 江苏大学 | A kind of transformer oil Aging of Oil-paper Insulation in Oil state analyzing method based on return voltage |
| CN108872803A (en) * | 2018-03-29 | 2018-11-23 | 福建工程学院 | A kind of detection method of the transformer insulation state based on dielectric return voltage |
| CN113468461A (en) * | 2020-03-30 | 2021-10-01 | 电子科技大学中山学院 | Oil-immersed transformer fault diagnosis method based on support vector machine and genetic algorithm |
| CN112485609A (en) * | 2020-10-19 | 2021-03-12 | 重庆大学 | Raman spectrum diagnosis method for insulation aging of transformer oil paper |
Non-Patent Citations (1)
| Title |
|---|
| 张丽平等: "采用核主成分分析和随机森林算法的变压器油纸绝缘评估方法", 四川电力技术, no. 02, 20 April 2018 (2018-04-20), pages 33 - 37 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | Fault diagnosis of power transformers using computational intelligence: A review | |
| Ma et al. | A new method for transformer fault prediction based on multifeature enhancement and refined long short-term memory | |
| CN105718958B (en) | Circuit breaker failure diagnostic method based on linear discriminant analysis and support vector machines | |
| US20240037387A1 (en) | Power transformer fault diagnosis method based on stacked time series network | |
| CN112633315A (en) | Electric power system disturbance classification method | |
| Esmaeili Nezhad et al. | A review of the applications of machine learning in the condition monitoring of transformers | |
| Singh et al. | Power system fault diagnosis using fuzzy decision tree | |
| CN113899997A (en) | Transformer insulation state diagnosis method based on improved support vector machine | |
| Tang et al. | An improved fuzzy C-means clustering algorithm for transformer fault | |
| Moreira et al. | Evaluation of pattern recognition algorithms for applications on power factor compensation | |
| Yan et al. | Research on imbalanced data fault diagnosis of on-load tap changers based on IGWO-WELM | |
| Yong et al. | Research on condition evaluation algorithm of oil-immersed transformer based on Naive Bayes | |
| Liu et al. | Recognition of dissolved gas in transformer oil by ant colony optimization support vector machine | |
| Negri et al. | Artificial Intelligence and Smart Grids: the Cable Joint Test Case | |
| Huang et al. | A Review of Transformer Fault Diagnosis Based on Information System Theory and Machine Learning | |
| Kong et al. | Dissolved gas analysis of insulating oil for power transformer state evaluation based on entropy weight and topsis method | |
| Elmabrouk et al. | An implementation of fuzzy logic technique for prediction of the power transformer faults | |
| Khaledian et al. | Event-Based Dynamic Response Modeling of Large Behind-the-Meter Solar Farms: A Data-Driven Method Based on Real-World Data | |
| Wen et al. | Charge Time Prediction Model of Power Battery based on Random Forest algorithm | |
| Hadiki et al. | Transformers Faults Prediction Using Machine Learning Approach | |
| CN113743801B (en) | Power transformer state evaluation method | |
| Zou et al. | Power Transformer Fault Diagnosis Method Based on Machine Learning | |
| Li | Construction of Transformer Fault Diagnosis and Prediction Model Based on Deep Learning | |
| Er et al. | Automated Transformer Health Classification: Feature weight Optimisation using Differential Evolution | |
| CN111814395A (en) | Static voltage stability evaluation method based on principal component analysis and confidence detection |
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 |