CN109657720B - On-line diagnosis method for turn-to-turn short circuit fault of power transformer - Google Patents
On-line diagnosis method for turn-to-turn short circuit fault of power transformer Download PDFInfo
- Publication number
- CN109657720B CN109657720B CN201811560473.1A CN201811560473A CN109657720B CN 109657720 B CN109657720 B CN 109657720B CN 201811560473 A CN201811560473 A CN 201811560473A CN 109657720 B CN109657720 B CN 109657720B
- Authority
- CN
- China
- Prior art keywords
- short circuit
- transformer
- monitoring
- fault
- turn
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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)
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
- Housings And Mounting Of Transformers (AREA)
Abstract
The invention discloses an on-line diagnosis method for turn-to-turn short circuit faults of a power transformer, which comprises the following steps: (1) collecting monitoring data before and after short circuit recorded by each monitoring index of the case transformer, checking whether each monitoring index has significant change before and after short circuit, and screening out the monitoring index with significant change before and after short circuit; (2) inputting monitoring data recorded by monitoring indexes with significant changes before and after short circuit into a random forest model, and training a short circuit fault online diagnosis model; (3) and inputting the monitoring data to be detected of the transformer to be detected into the optimized random forest model, outputting an online diagnosis result of the monitoring data to be detected, and judging whether the transformer to be detected has a fault. The method only needs to carry out remote analysis on the monitoring data to be detected of the transformer to be detected, and compared with an offline detection method, the method does not need a field test, thereby saving the test cost and improving the diagnosis efficiency.
Description
Technical Field
The invention relates to the field of fault diagnosis, in particular to an on-line diagnosis method for turn-to-turn short circuit faults of a power transformer.
Background
The power transformer is one of the important devices of the power grid system and takes charge of the work of a power grid junction. The method can avoid the hidden trouble of the power transformer in time, and has very important significance for guaranteeing the safe and stable operation of the power grid. Statistically, about 25% of transformer turn-to-turn accidents are caused by the deformation of the transformer winding after short circuit impact, and the severity of the fault is related to the magnitude of the short circuit current.
The classification of poor working conditions of short-circuit current can be divided into three categories: the I-grade poor working condition means that the short-circuit current is 50-70% of the maximum bearable short-circuit current, and the accumulated times reach 6 times and more; the class II poor working condition means that the short-circuit current is 70-100% of the maximum bearable short-circuit current; the class III poor working condition means that the short-circuit current is more than 100% of the maximum bearable short-circuit current.
At present, the diagnosis of turn-to-turn short circuit faults of domestic transformers mainly depends on observing the protection action condition of a main transformer, the analysis condition of an oil sample and the short circuit current value. If the main transformer does not trip, the oil sample is not abnormal, and the short-circuit current value is less than 50% of the bearable maximum short-circuit current value, the main transformer is considered to have no obvious fault, and follow-up oil chromatography tracking monitoring is enhanced. If the main transformer is not tripped, an oil sample is abnormal or the short-circuit current value is larger than 50% of the bearable maximum short-circuit current value, power failure tests including frequency response tests, short-circuit impedance tests, capacitance tests and the like need to be arranged as soon as possible, all test results are integrated to carry out special analysis, whether the winding is seriously deformed or not under impact is judged, and whether the transformer can continue to operate or not is determined. If the main transformer does not perform electric quantity protection action, the transformer needs to be arranged and tested, dynamic short-circuit resistance analysis is performed, and all test results are integrated to perform special analysis so as to determine whether to perform partial discharge test or to perform cover hanging and disassembly inspection.
The above short-circuit fault diagnosis methods all need to perform a power failure test or disassembly check to confirm whether the transformer winding is deformed after being impacted, belong to offline diagnosis methods, and have the defects of complex operation, difficult test, dependence on manual experience and the like.
A diagnostic method that does not require a power failure test as opposed to an off-line diagnostic method is called an on-line diagnostic method, and detects whether or not deformation is caused by a change in an ultrasonic signal or other broadband signal, measures a winding size and observes whether or not deformation is caused using a radar chart, and detects whether or not deformation is caused by a change in magnetic flux distribution.
The method is simple to operate, can directly observe the internal structure of the transformer and make a very accurate judgment result, but has the defects of high application cost of new technology and new equipment, and the methods cannot be put into practical application and large-scale popularization.
Patent specification CN 108197639 a discloses a transformer fault diagnosis method based on random forest, which includes: collecting fault gas concentration data in insulating oil in a transformer and corresponding fault types as training samples; establishing a fault decision tree according to the training samples and the generation steps of the decision tree: establishing a random forest model according to the fault decision tree; and collecting the concentration data of the fault gas of unknown fault types, inputting the concentration data into the random forest model, and obtaining the fault types by using the random forest model. The random forest model established by using the concentration data of the fault gas in the insulating oil in the transformer as the training sample has accurate diagnosis on the fault of the whole transformer, has high stability and can be applied to the technical field of transformer diagnosis.
Patent specification CN 107025514 a discloses an evaluation method for dynamically evaluating the state of transformer equipment, which comprises the following steps: counting historical data of each state quantity of the transformer equipment; calculating the importance of each state quantity by using a random forest algorithm based on the historical data of each state quantity; correcting the importance of each state quantity according to the quantized value of the correction factor, discretizing the importance of the corrected state quantity, converting the discretized importance of the state quantity into a preset parameter, and reflecting the importance level of each state quantity; performing state evaluation on the transformer equipment according to the importance level of each state quantity, and determining the real-time state of the equipment; and (4) taking the importance degree grade of each state quantity in the state evaluation method as historical data, and repeating the steps to realize the white-me optimization and dynamic evaluation of the evaluation method.
Disclosure of Invention
Aiming at the defects in the field, the invention provides an on-line diagnosis method for turn-to-turn short circuit faults of a power transformer, aiming at the transformer with deformation and deformation operation data, the data to be detected is respectively matched with the historical deformation operation data and the normal operation data of the transformer through a random forest algorithm, so that remote intelligent diagnosis is realized, the test cost is saved, the multi-source information of on-line monitoring data is integrated, and more comprehensive diagnosis is carried out.
An on-line diagnosis method for turn-to-turn short circuit fault of a power transformer comprises the following steps:
(1) collecting monitoring data before and after short circuit recorded by each monitoring index of the case transformer, checking whether each monitoring index has significant change before and after short circuit, and screening out the monitoring index with significant change before and after short circuit;
(2) monitoring data recorded by monitoring indexes with significant changes before and after short circuit are used as a training set and input into a random forest model, and a short circuit fault online diagnosis model is trained;
(3) and inputting the monitoring data to be detected of the transformer to be detected into the optimized random forest model, outputting an online diagnosis result of the monitoring data to be detected, and judging whether the transformer to be detected has a fault.
In the step (1), preferably, the monitoring indexes are voltage, current, oil temperature and power.
Preferably, the specific step of verifying whether each monitoring index has a significant change before and after the short circuit includes:
a. carrying out normality test on the distribution of the on-line monitoring data recorded by the monitoring index;
b. if the normal distribution is met, whether the arithmetic mean of the monitoring data before and after short circuit is obviously different is checked by using ANOVA; if the normal distribution is not satisfied, Kruskal-Wallis is used for testing whether the median, range and distribution of the monitoring indexes before and after short circuit are obviously different.
The normality test, ANOVA test and Kruskal-Wallis test can be performed by SPSS software.
When the significance value of the monitoring index is less than 0.05, the monitoring index is considered to have significant difference before and after short circuit.
In the step (2), the random forest is an important Bagging-based integrated learning method, can process discrete data and continuous data, and has good applicability to high-dimensional data.
Preferably, the specific steps of training the short-circuit fault online diagnosis model include:
a. randomly sampling m samples from the training set by using a bootstrapping method to select n samples, and sampling for n times to generate n sub-training sets;
b. respectively training n decision trees for the n sub-training sets;
c. splitting each decision tree according to the characteristic of the maximum information gain ratio until all training samples of the node belong to the same class;
d. and forming a random forest by the generated decision trees.
Preferably, the random forest model in step (2) may be optimally adjusted to improve the diagnosis accuracy, and the specific methods that may be adopted include:
1. and constructing a phase difference value as a new characteristic according to the three-phase unbalance principle of the transformer.
The basis for constructing the phase difference value as a new characteristic is that the amplitude-frequency response characteristics of A, B, C three-phase windings of the same voltage level of the transformer are compared, the unbalance rate of the three phases is calculated, and whether the transformer windings deform or not is judged. The principle is that the A, B, C three phases of voltage and current are normally equal in magnitude, and if one phase is deformed, the three phases will be unbalanced.
2. And setting a decision tree quantity gradient, comparing classification performance, and balancing diagnosis precision and algorithm efficiency.
The classification performance is improved with the increase of the number of the sub-trees of the decision tree, but the improvement amplitude is gradually reduced. Therefore, the number of decision trees needs to be reasonably selected to obtain a result with better comprehensive classification performance and operation duration. For higher algorithm efficiency, the running time is preferably 5-10 minutes.
The specific method for outputting the online diagnosis result of the monitoring data to be detected and judging whether the transformer to be detected has a fault or not in step (3) may be to output the classification result of each decision tree on the monitoring data to be detected, and if the number of decision trees with faults as the classification result is smaller than the number of decision trees with normal classification result, judge that the transformer to be detected is normal; and if the number of the decision trees with the fault classification result is larger than the number of the decision trees with the normal classification result, judging that the transformer to be tested has the fault.
Compared with the prior art, the invention has the main advantages that:
(1) the remote analysis is only needed to be carried out on the monitoring data to be detected of the transformer to be detected, and compared with an offline detection method, the field test is not needed, so that the test cost is saved, and the diagnosis efficiency is improved.
(2) And multi-source information of online monitoring data such as voltage, current, power, oil temperature and the like of the transformer is fused and utilized, so that more comprehensive diagnosis can be performed.
Drawings
Fig. 1 is a flowchart of an online diagnosis method for turn-to-turn short circuit fault of a power transformer in embodiment 1.
Detailed Description
The invention is further described with reference to the following drawings and specific examples. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. The experimental procedures, in which specific conditions are not noted in the following examples, are generally carried out under conventional conditions or conditions recommended by the manufacturers.
Example 1
As shown in fig. 1, the flow of the online diagnosis method for turn-to-turn short circuit fault of the power transformer includes:
and S01, collecting monitoring data before and after short circuit recorded by each monitoring index of the case transformer, wherein the monitoring indexes comprise four types of voltage, current, oil temperature and power.
And S02, checking whether the monitoring index has significant change before and after short circuit. If the monitoring data meet normal distribution, whether the arithmetic mean of the monitoring data before and after short circuit is obviously different is checked by using ANOVA; if the normal distribution is not satisfied, Kruskal-Wallis is used for testing whether the median, range and distribution of the monitoring indexes before and after short circuit are obviously different.
S03, screening out monitoring indexes with significant changes before and after short circuit, inputting the monitoring indexes into a random forest model, and training an online short circuit fault diagnosis model;
s04, optimizing and adjusting the random forest model, constructing a phase difference value as a new characteristic according to the three-phase unbalance principle of the transformer, respectively setting the number of decision trees to be 30, 100 and 200, performing classification performance comparison, and selecting a preferred parameter to balance diagnosis precision and algorithm efficiency;
s05, inputting the monitoring data to be detected of the transformer to be detected into the optimized random forest model, outputting the classification result of each decision tree on the monitoring data to be detected, and if the number of the decision trees with faults in the classification result is smaller than the number of the decision trees with normal classification result, judging that the transformer to be detected is normal; and if the number of the decision trees with the fault classification result is larger than the number of the decision trees with the normal classification result, judging that the transformer to be tested has the fault.
(1) Six transformers with deformed windings are detected and displayed off line after the six transformers are short-circuited in the Zhejiang power grid are used as case transformers. And collecting voltage, current, oil temperature and power data before and after the short circuit monitored under the three-phase three-winding of the six transformers, and dividing the data into a normal group and a fault group according to the data before and after the short circuit.
Firstly, analysis of variance is carried out to judge whether the arithmetic mean of the monitoring data of the normal group and the fault group has difference. Analysis of variance requires three assumptions to be satisfied: normality, homogeneity and independence. The results of the normality test show that the P value of each index is below 0.001, so that the original hypothesis of normal distribution is rejected, all indexes are considered to be not subject to normal distribution obviously, so that the independence T test and the common ANOVA cannot be used, and the non-parametric test method Kruskal-Wallis test which does not need to meet the normality hypothesis is considered.
The Kruskal-Wallis test, also known as the rank sum test, can be used to determine whether there is a difference in the two population distribution positions from which two independent samples are calculated. The median, range and distribution of the failed and normal cohorts were examined for differences using Kruskal-Wallis and the results are shown in Table 1. It is generally believed that when the significance of a statistic is less than 0.05, the statistic is considered to have significant differences between the two groups. Therefore, except that the difference between the A-phase current amplitude of the 110kV side, the A-phase current amplitude of the 220kV side, the reactive value of the 220kV side and the median of the reactive value of the 220kV side is not obvious between the fault group and the normal group, the differences exist in the median, the range and the distribution of other monitoring indexes of the fault group and the normal group, and the monitoring indexes of the current, the voltage, the power and the oil temperature can provide effective information for judging whether the transformer fails.
TABLE 1 nonparametric Kruskal-Wallis test
Therefore, the selected four monitoring indexes of voltage, current, power and oil temperature are used as characteristics and input into a random forest model for training, and the precision of the diagnosis model is judged by adopting ten-fold cross validation. Three precision indexes are mainly adopted: accuracy (Precision), Recall (Recall), and F1 scores. The accuracy rate is measured by the misjudgment condition of the fault, and the higher the accuracy rate is, the smaller the misjudgment rate is. The recall rate is measured by the condition of missed judgment of the fault, and the higher the recall rate is, the smaller the missed judgment rate is. Generally, the recall rate is reduced while the accuracy is improved, and the F1 score is an index for combining the two. The calculation methods of the three indexes are as follows:
as shown in table 2, by integrating the observation accuracy, the recall ratio and the F1 score, the random forest model has better diagnosis effect because the F1 scores of the other five transformers are all over 80% except that the diagnosis effect of the random forest model on the transformer 2 is slightly low.
TABLE 2 diagnosis results of random forest models on 6 transformers
| Transformer 1 | Transformer 2 | Transformer 3 | Transformer 4 | Transformer 5 | Transformer 6 | |
| Rate of accuracy | 0.9861 | 0.8792 | 0.9887 | 0.9667 | 0.9836 | 0.8595 |
| Recall rate | 0.985 | 0.6947 | 0.9741 | 0.8325 | 0.9681 | 0.7804 |
| F1 score | 0.9856 | 0.7761 | 0.9813 | 0.8946 | 0.9758 | 0.818 |
In order to further improve the diagnosis precision, the phase difference value is considered as a new characteristic, and whether the transformer winding is deformed or not can be judged by comparing the amplitude-frequency response characteristics of A, B, C three-phase windings with the same voltage level of the transformer and calculating the unbalance rate of the three phases. The principle is that the A, B, C three phases of voltage and current are normally equal in magnitude, and if one phase is deformed, the three phases will be unbalanced.
Therefore, when the features are combined, the current and voltage difference values of the AB phase and the BC phase on each side are calculated as new features. In order to retain most original information, original voltage and current data are still retained, and only 12 characteristic values are newly added.
And after the phase difference index is increased, calculating the classification performance of the random forest model on each transformer again. As shown in table 3, after the phase difference value is increased, the recall rate of the random forest model to the transformer 2 is increased by 13.71%, and the F1 score is increased by 6.4%. Meanwhile, the recall rate and the F1 score of the random forest model added with the phase difference value on the transformer 6 are both improved by about 5%.
TABLE 3 comparison of diagnostic results of random forest models before and after addition of the phasor difference indicator
In the random forest model, more decision tree subtrees can make the model have better performance, but too many subtrees can make the model run very slowly. In this embodiment, the number of records of case transformers is 2 to 10 ten thousand, and the running time exceeds 20 minutes when the number of decision trees is 500.
After comprehensive consideration, the number of the decision trees is respectively set to be 30, 100 and 200, classification performance comparison is carried out, the operation result of the random forest model is shown in table 4, the classification performance is improved along with the increase of the number of the decision trees, and the improvement range is gradually reduced. Therefore, when the number of the decision trees is 200, a result with better comprehensive classification performance and operation time can be obtained, and the operation time is 5-10 minutes.
TABLE 4 running results of random forest models of different decision tree numbers
After feature merging and parameter tuning, the online diagnosis method for turn-to-turn short circuit faults of the power transformer has a relatively considerable diagnosis result on six transformers, and the accuracy, the recall rate and the F1 score of identification of winding deformation faults after short circuit all reach more than 85%.
Furthermore, it should be understood that various changes and modifications can be made by one skilled in the art after reading the above description of the present invention, and equivalents also fall within the scope of the invention as defined by the appended claims.
Claims (4)
1. An on-line diagnosis method for turn-to-turn short circuit fault of a power transformer comprises the following steps:
(1) collecting monitoring data before and after short circuit recorded by each monitoring index of the case transformer, checking whether each monitoring index has significant change before and after short circuit, and screening out the monitoring index with significant change before and after short circuit; the monitoring indexes are voltage, current, oil temperature and power; the specific steps for checking whether the monitoring indexes are significantly changed before and after the short circuit comprise:
a. carrying out normality test on the distribution of the on-line monitoring data recorded by the monitoring index;
b. if the normal distribution is met, whether the arithmetic mean of the monitoring data before and after short circuit is obviously different is checked by using ANOVA; if the normal distribution is not satisfied, using Kruskal-Wallis to test whether the median, range and distribution of the monitoring indexes before and after short circuit are significantly different;
(2) monitoring data recorded by monitoring indexes with significant changes before and after short circuit are used as a training set and input into a random forest model, and a short circuit fault online diagnosis model is trained;
(3) inputting the monitoring data to be detected of the transformer to be detected into the optimized random forest model, outputting an online diagnosis result of the monitoring data to be detected, and judging whether the transformer to be detected has a fault;
optimizing and adjusting the random forest model in the step (2) to improve the diagnosis precision; the specific method for optimizing and adjusting the random forest model is to construct a phase difference value as a new characteristic according to the three-phase imbalance principle of the transformer.
2. The on-line diagnosis method for turn-to-turn short circuit fault of power transformer according to claim 1, wherein the specific steps of training the on-line diagnosis model for short circuit fault include:
a. randomly sampling m samples from the training set by using a bootstrapping method to select n samples, and sampling for n times to generate n sub-training sets;
b. respectively training n decision trees for the n sub-training sets;
c. splitting each decision tree according to the characteristic of the maximum information gain ratio until all training samples of the nodes belong to the same class;
d. and forming a random forest by the generated decision trees.
3. The on-line diagnosis method for turn-to-turn short circuit faults of the power transformer according to claim 1, wherein the specific method for optimizing and adjusting the random forest model is to set a decision tree quantity gradient, perform classification performance comparison, and balance diagnosis precision and algorithm efficiency.
4. The method for on-line diagnosis of turn-to-turn short circuit fault of power transformer according to claim 1, wherein the step of outputting the on-line diagnosis result of the monitored data to be tested, and the step of judging whether the transformer to be tested has fault comprises the steps of outputting the classification result of each decision tree on the monitored data to be tested, and if the classification result is that the number of the decision trees with fault is smaller than the number of the decision trees with normal classification result, judging that the transformer to be tested is normal; and if the number of the decision trees with the fault classification result is larger than the number of the decision trees with the normal classification result, judging that the transformer to be tested has the fault.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811560473.1A CN109657720B (en) | 2018-12-20 | 2018-12-20 | On-line diagnosis method for turn-to-turn short circuit fault of power transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811560473.1A CN109657720B (en) | 2018-12-20 | 2018-12-20 | On-line diagnosis method for turn-to-turn short circuit fault of power transformer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109657720A CN109657720A (en) | 2019-04-19 |
| CN109657720B true CN109657720B (en) | 2021-05-11 |
Family
ID=66115132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811560473.1A Active CN109657720B (en) | 2018-12-20 | 2018-12-20 | On-line diagnosis method for turn-to-turn short circuit fault of power transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109657720B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110514924B (en) * | 2019-08-12 | 2021-04-27 | 武汉大学 | Power transformer winding fault positioning method based on deep convolutional neural network fusion visual identification |
| CN114019298B (en) * | 2021-09-28 | 2023-12-05 | 中电华创(苏州)电力技术研究有限公司 | On-line monitoring method for turn-to-turn short circuit of generator rotor based on PCC-SVM |
| CN113872200B (en) * | 2021-10-13 | 2023-11-24 | 国网江苏省电力有限公司电力科学研究院 | Medium-voltage distribution network power failure event diagnosis and identification method, system and storage medium |
| CN116799741B (en) * | 2023-05-24 | 2024-03-26 | 华斗数字科技(上海)有限公司 | Precision equipment short-circuit monitoring protection method and system based on slope detection |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1441257A (en) * | 2003-03-27 | 2003-09-10 | 河海大学 | In-situ fault diagnosing technology for turn-to-turn short-circuit of transformer windings based on change in loss |
| CN107516108A (en) * | 2017-08-15 | 2017-12-26 | 国网四川省电力公司电力科学研究院 | Grader creation method and partial discharge of transformer method of fault pattern recognition |
| CN107782442A (en) * | 2017-10-24 | 2018-03-09 | 华北电力大学（保定） | Transformer multiple features parameter selection method based on big data and random forest |
| CN108062571A (en) * | 2017-12-27 | 2018-05-22 | 福州大学 | Diagnosing failure of photovoltaic array method based on differential evolution random forest grader |
| CN108197639A (en) * | 2017-12-14 | 2018-06-22 | 佛山科学技术学院 | A kind of Diagnosis Method of Transformer Faults based on random forest |
| CN108693478A (en) * | 2018-04-17 | 2018-10-23 | 北京理工大学 | A kind of method for detecting leakage of lithium-ion-power cell |
| CN108872778A (en) * | 2018-07-03 | 2018-11-23 | 安徽理工大学 | A kind of transformer winding fault diagnostic system based on correlation dimension and random forest |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2528044B (en) * | 2014-07-04 | 2018-08-22 | Arc Devices Ni Ltd | Non-touch optical detection of vital signs |
| CN106372747B (en) * | 2016-08-27 | 2023-07-25 | 天津大学 | Random forest-based reasonable line loss rate estimation method for transformer area |
| CN108398612A (en) * | 2018-01-12 | 2018-08-14 | 广州市扬新技术研究有限责任公司 | A kind of urban track traffic DC power-supply system short trouble localization method |
| CN108594067A (en) * | 2018-04-02 | 2018-09-28 | 湖南大学 | A kind of Multi-port direct-current distribution network line short fault distance measuring method |
| CN108959732B (en) * | 2018-06-15 | 2019-09-27 | 西安科技大学 | A kind of transmission line malfunction kind identification method based on convolutional neural networks |
-
2018
- 2018-12-20 CN CN201811560473.1A patent/CN109657720B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1441257A (en) * | 2003-03-27 | 2003-09-10 | 河海大学 | In-situ fault diagnosing technology for turn-to-turn short-circuit of transformer windings based on change in loss |
| CN107516108A (en) * | 2017-08-15 | 2017-12-26 | 国网四川省电力公司电力科学研究院 | Grader creation method and partial discharge of transformer method of fault pattern recognition |
| CN107782442A (en) * | 2017-10-24 | 2018-03-09 | 华北电力大学（保定） | Transformer multiple features parameter selection method based on big data and random forest |
| CN108197639A (en) * | 2017-12-14 | 2018-06-22 | 佛山科学技术学院 | A kind of Diagnosis Method of Transformer Faults based on random forest |
| CN108062571A (en) * | 2017-12-27 | 2018-05-22 | 福州大学 | Diagnosing failure of photovoltaic array method based on differential evolution random forest grader |
| CN108693478A (en) * | 2018-04-17 | 2018-10-23 | 北京理工大学 | A kind of method for detecting leakage of lithium-ion-power cell |
| CN108872778A (en) * | 2018-07-03 | 2018-11-23 | 安徽理工大学 | A kind of transformer winding fault diagnostic system based on correlation dimension and random forest |
Non-Patent Citations (1)
| Title |
|---|
| 配电变压器故障分析及随机森林诊断实现;单林森;《科技风》;20161231(第24期);第120页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109657720A (en) | 2019-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109657720B (en) | On-line diagnosis method for turn-to-turn short circuit fault of power transformer | |
| CN109670242B (en) | Transformer winding deformation unsupervised online monitoring method based on elliptical envelope curve | |
| CN107037279B (en) | A kind of shunt capacitor monitoring system and method | |
| CN105353256A (en) | Electric transmission and transformation device state abnormity detection method | |
| CN107992665B (en) | Online fault diagnosis and analysis method for alternating current filter of extra-high voltage converter station | |
| CN108647479B (en) | Lightning arrester fault transient waveform diagnosis method and device | |
| CN106557546A (en) | A kind of method and system extra-high voltage online monitoring data excavated and is evaluated | |
| CN110647924B (en) | GIS equipment state evaluation method based on support vector description and K-nearest neighbor algorithm | |
| CN114814501B (en) | On-line diagnosis method for capacitor breakdown fault of capacitor voltage transformer | |
| CN114966295A (en) | Super capacitor operation power frequency modulation detection system based on big data | |
| CN105911407A (en) | Transformer state fuzzy set pair assessment method based on matter-element augmentation extensive correlation | |
| CN111551887A (en) | Multidimensional identification voltage transformer metering performance online monitoring platform | |
| CN112801135B (en) | Fault line selection method and device for power plant service power system based on characteristic quantity correlation | |
| CN117310353B (en) | Method and system for testing through-flow pressurization faults of primary and secondary circuits of transformer substation | |
| CN117554885A (en) | Intelligent metering box operation abnormality detection method and system | |
| CN109784777B (en) | Power grid equipment state evaluation method based on time sequence information fragment cloud similarity measurement | |
| Hossain et al. | Distribution high impedance fault location using localized voltage magnitude measurements | |
| CN110645887A (en) | Winding deformation judgment method based on short-circuit reactance | |
| CN113487141B (en) | Bearing state evaluation method based on multisource information clustering fusion | |
| CN115561596A (en) | Lightning arrester insulation state assessment method based on random forest | |
| CN113052249B (en) | Transformer winding fault type identification method based on support vector machine and current deviation coefficient | |
| CN115308644A (en) | Transformer winding fault detection method and system based on current offset ratio difference analysis | |
| CN110943485B (en) | Index evaluation method for simulation reliability of equivalent model of doubly-fed wind power plant | |
| CN107576259A (en) | Deformation of transformer winding online test method based on very fast transient overvoltage characteristic | |
| Liu et al. | Single-phase Grounding Fault Line Selection Method Based on the Difference of Electric Energy Information Between the Distribution End and the Load End |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |