CN110726957A - Fault identification method of dry-type reactor - Google Patents
Fault identification method of dry-type reactor Download PDFInfo
- Publication number
- CN110726957A CN110726957A CN201911070723.8A CN201911070723A CN110726957A CN 110726957 A CN110726957 A CN 110726957A CN 201911070723 A CN201911070723 A CN 201911070723A CN 110726957 A CN110726957 A CN 110726957A
- Authority
- CN
- China
- Prior art keywords
- fault
- dry
- vibration signal
- som network
- network
- 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
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
Abstract
The invention relates to the technical field of dry type reactor fault identification, and particularly discloses a fault identification method of a dry type reactor, wherein the fault identification method comprises the following steps: respectively acquiring a vibration signal of the dry-type reactor in a normal state and a vibration signal of the dry-type reactor in a fault state; respectively processing the vibration signal in the normal state and the vibration signal in the fault state to obtain processing results; inputting the processing result into an SOM network, and training the SOM network to obtain a trained SOM network; inputting data to be tested to the trained SOM network to obtain an output result; and judging the fault mode of the dry type reactor according to the output result. The fault identification method of the dry-type electric reactor can well identify the fault of the dry-type electric reactor, and the SOM network is a self-organizing mapping network without a pilot, so that the fault identification method of the dry-type electric reactor has the characteristics of small network scale, short learning process time, small calculation amount, good fault tolerance of a trained network and the like.
Description
Technical Field
The invention relates to the technical field of dry-type reactor fault identification, in particular to a fault identification method of a dry-type reactor.
Background
With the increase of the running number of the dry-type reactors year by year and the increase of the running age, the quality defects of products are gradually shown due to the fact that the technology of early products is immature or the process control is not in place and the like. Meanwhile, air is adopted for heat dissipation, the operation environment is severe, the problem of insulation damage is easily caused after long-time operation, faults such as turn-to-turn short circuit are further caused, even a reactor is burnt in serious conditions, and the safe and stable operation of a power grid is influenced. According to related researches, 95% of reactor faults are caused by insulation faults and turn-to-turn short circuits, but the existing monitoring method has the problems of low sensitivity, poor reliability and the like, and the reactor can not be burnt out due to the fact that the faults cannot be found in time. Therefore, it is important to research a more effective monitoring method and find faults in time.
The processing of the vibration signal generally comprises 2 links, namely feature extraction and fault identification. The time-frequency method is commonly used for feature extraction, can give consideration to both time and frequency, can better express local features of signals, and is particularly suitable for analysis of non-stationary signals. The wavelet packet technology improves the problem of wavelet high frequency low resolution on the basis of maintaining the excellent characteristics of wavelet orthogonal base, provides a more precise analysis method for vibration signals, and has self-adaption capability to the characteristics of different signals. The signal after the wavelet packet orthogonal decomposition has the characteristics of independence of each frequency band signal and energy conservation, and is more suitable for time-frequency analysis and energy spectrum analysis of vibration signals. And the fault identification judges the system state according to the mechanical fault characteristic quantity. The fault identification method of the high-voltage circuit breaker is usually an artificial intelligence algorithm, and comprises an artificial neural network, a support vector machine, an artificial immune network and the like. The neural network has better fault-tolerant capability and generalization performance, but has the problem of local convergence. At present, most of the available network types are limited by less sample data, and a more perfect fault identification method does not exist.
Disclosure of Invention
The invention provides a fault identification method of a dry-type reactor, which solves the problem that no perfect fault identification method exists in the related technology.
As a first aspect of the present invention, there is provided a fault identification method for a dry reactor, including:
respectively acquiring a vibration signal of the dry-type reactor in a normal state and a vibration signal of the dry-type reactor in a fault state;
respectively processing the vibration signal in the normal state and the vibration signal in the fault state to obtain processing results;
inputting the processing result into an SOM network, and training the SOM network to obtain a trained SOM network;
inputting data to be tested to the trained SOM network to obtain an output result;
and judging the fault mode of the dry type reactor according to the output result.
Further, still include:
establishing a reactor turn-to-turn short circuit model;
calculating the equivalent resistance and the equivalent reactance of each turn-to-turn coil in the turn-to-turn short circuit model;
constructing a first eigenvector from the equivalent resistance and the equivalent reactance.
Further, the processing the vibration signal in the normal state and the vibration signal in the fault state to obtain the processing result respectively includes:
respectively performing wavelet decomposition on the vibration signal in the normal state and the vibration signal in the fault state to obtain a plurality of frequency band signals;
and carrying out normalization processing on the frequency band signals to construct a second feature vector.
Further, the inputting the processing result to the SOM network and training the SOM network to obtain a trained SOM network includes:
constructing a third feature vector according to the first feature vector and the second feature vector;
inputting the third feature vector to the SOM network.
Further, the performing wavelet decomposition on the vibration signal in the normal state and the vibration signal in the fault state respectively to obtain a plurality of frequency band signals includes:
and performing two-layer wavelet packet transformation on the vibration signal in the normal state and the vibration signal in the fault state by adopting a db10 wavelet to obtain four frequency band signals.
Further, the energy expression of the frequency band signal is as follows:
wherein E is2(i) Representing the energy of the ith frequency band, W (2, i) representing the ith frequency band signal of the second layer after wavelet packet two-layer decomposition, xikThe value of the kth discrete point of the ith frequency band signal W (2, i), i being 0,1, …,3, i represents four frequency bands, k being 1,2, …, N represents the number of discrete points.
Further, the second eigenvector H constructed according to the proportion of the energy of each frequency band2Expressed as:
H2=[E2(0)/S2,…,E2(3)/S2],
wherein S is2Representing the total energy of the second layer after the decomposition of the two layers of the wavelet packet,
further, still include:
and dividing the second feature vectors into two groups, wherein one group is used for inputting the SOM network for training, and the other group is used for inputting the SOM network for testing.
Further, the inputting the processing result to the SOM network and training the SOM network to obtain a trained SOM network includes:
competition, given the tth input X (t), each neuron weight vector W is calculatedi(t) relative distance D from input vectori(t):||X(t)-Wi(t)||=Di(t), if the minimum relative distance is selected as the winner, then exciting, namely: i.e. i*(t)={i:minDi(t)};
Learning, the winner and its neighbors are allowed to learn the input pattern, the adjustment of the weight factor being proportional to the difference of the weight factor from the input: Δ Wi=η(t)(X(t)-Wi(t)), η (t) represents a variable learning rate, decaying with time;
repeating the competition and learning steps until the excitatory neuron is stably corresponding to the input sample, and finishing the training.
Further, the fault condition includes winding turn-to-turn short circuit and insulation aging.
By the aid of the fault identification method of the dry-type electric reactor, vibration signals of the dry-type electric reactor are collected firstly, then the vibration signals are processed, processing results are input into the SOM network, data to be tested are input into the trained SOM network after the SOM network is trained, output results are obtained, and finally the fault mode of the dry-type electric reactor is judged according to the output results; in addition, the method for identifying the fault of the dry type reactor provided by the embodiment does not need a large database for training, so that the problems of less sample data, complicated calculation, low precision and the like in the conventional fault identification technology of the dry type reactor can be well solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
fig. 1 is a flowchart of a fault identification method for a dry reactor according to the present invention.
Fig. 2 is an equivalent model of turn-to-turn short circuit of the dry reactor provided by the invention.
Fig. 3 is a schematic diagram of a wavelet decomposition tree provided by the present invention.
Fig. 4 is a schematic diagram of an SOM network structure provided by the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
In order to make those skilled in the art better understand the technical solution of the present invention, the technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged under appropriate circumstances in order to facilitate the description of the embodiments of the invention herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In the present embodiment, a method for identifying a fault of a dry reactor is provided, and fig. 1 is a flowchart of a method for identifying a fault of a dry reactor according to an embodiment of the present invention, as shown in fig. 1, including:
s110, respectively acquiring a vibration signal of the dry-type reactor in a normal state and a vibration signal of the dry-type reactor in a fault state;
s120, processing the vibration signal in the normal state and the vibration signal in the fault state respectively to obtain processing results;
s130, inputting the processing result into an SOM network, and training the SOM network to obtain a trained SOM network;
s140, inputting data to be tested to the trained SOM network to obtain an output result;
and S150, judging the fault mode of the dry type reactor according to the output result.
By the aid of the fault identification method of the dry-type electric reactor, vibration signals of the dry-type electric reactor are collected firstly, then the vibration signals are processed, processing results are input into the SOM network, data to be tested are input into the trained SOM network after the SOM network is trained, output results are obtained, and finally the fault mode of the dry-type electric reactor is judged according to the output results; in addition, the method for identifying the fault of the dry type reactor provided by the embodiment does not need a large database for training, so that the problems of less sample data, complicated calculation, low precision and the like in the conventional fault identification technology of the dry type reactor can be well solved.
It should be noted that the fault condition includes winding turn-to-turn short circuit and insulation aging.
Specifically, the method further comprises the following steps:
establishing a reactor turn-to-turn short circuit model;
calculating the equivalent resistance and the equivalent reactance of each turn-to-turn coil in the turn-to-turn short circuit model;
constructing a first eigenvector from the equivalent resistance and the equivalent reactance.
Specifically, the dry-type air-core reactor is equivalent to a plurality of parallel inductance branches, and an equivalent circuit model is shown in fig. 2, assuming that the kth turn of the mth layer winding of the s layers of parallel windings generates turn-to-turn short circuit. In fig. 2, the short-circuit rings form a single closed loop, and the m layers of windings are divided into two sections of windings which are connected in series up and down, and the equivalent resistance of the windings is Rs+1Self-inductance of Ls+1。
Closed loop through mutual inductance Mi,s+1And the voltage of the two sections of series windings is the voltage of the end of the reactor, and the voltage of the short-circuit loop is 0. According to the principle of electromagnetic induction, although the voltage across the short-circuit loop is 0, an induced current exists in the short-circuit loop. The voltage equation of each branch after the short-circuit fault occurs is as follows.
The voltage balance equation of the coil of the ith layer is as follows:
the voltage balance equation of the short circuit loop is as follows:
calculating distributed current I of each branch of reactor1、I2、I3、…、IsAlgebraically summing the current of each branch to obtain the total current I flowing through the reactor, obtaining an equivalent resistance R and an equivalent reactance X according to the ohm law Z-U/I, and further calculating a power angle theta. Calculating equivalent impedance and power angle of n groups of reactors to construct a characteristic vector H1:
H1=θ1。
Specifically, the processing of the vibration signal in the normal state and the vibration signal in the fault state to obtain the processing result includes:
respectively performing wavelet decomposition on the vibration signal in the normal state and the vibration signal in the fault state to obtain a plurality of frequency band signals;
and carrying out normalization processing on the frequency band signals to construct a second feature vector.
Further specifically, the inputting the processing result to the SOM network and training the SOM network to obtain a trained SOM network includes:
constructing a third feature vector according to the first feature vector and the second feature vector;
inputting the third feature vector to the SOM network.
Specifically, the performing wavelet decomposition on the vibration signal in the normal state and the vibration signal in the fault state respectively to obtain a plurality of frequency band signals includes:
and performing two-layer wavelet packet transformation on the vibration signal in the normal state and the vibration signal in the fault state by adopting a db10 wavelet to obtain four frequency band signals. As shown in fig. 3, a diagram of a wavelet decomposition tree is shown.
The energy expression of the frequency band signal is as follows:
wherein E is2(i) Representing the energy of the ith frequency band, W (2, i) representing the ith frequency band signal of the second layer after wavelet packet two-layer decomposition, xikThe value of the kth discrete point of the ith frequency band signal W (2, i), i being 0,1, …,3, i represents four frequency bands, k being 1,2, …, N represents the number of discrete points.
More specifically, the second eigenvector H is constructed according to the proportion of energy in each frequency band2Expressed as:
H2=[E2(0)/S2,…,E2(3)/S2],
wherein S is2Representing the total energy of the second layer after the decomposition of the two layers of the wavelet packet,
further, a third feature vector H3Is represented by the formula:
H3=[E2(0)/S2,…,E2(3)/S2,θ]。
third eigenvector H3Inputting the data into the SOM network, and training the SOM network. And when the SOM network reaches the training error, testing the SOM network by adopting the test data so as to judge the fault mode of the reactor.
Specifically, the method further comprises the following steps:
and dividing the second feature vectors into two groups, wherein one group is used for inputting the SOM network for training, and the other group is used for inputting the SOM network for testing.
The data acquisition card acquires 20000 point vibration data per phase at the speed of 20kHz and sends the vibration data to a PC for processing when the breaker simulates action; and dividing the acquired vibration data into two groups, wherein each group comprises a normal vibration signal and a fault vibration signal, one group is used for training the SOM network, and the other group is used for testing the SOM network.
Specifically, the inputting the processing result to the SOM network and training the SOM network to obtain a trained SOM network includes:
competition, given the tth input X (t), each neuron weight vector W is calculatedi(t) relative distance D from input vectori(t):||X(t)-Wi(t)||=Di(t), if the minimum relative distance is selected as the winner, then exciting, namely: i.e. i*(t)={i:minDi(t)};
Learning, the winner and its neighbors are allowed to learn the input pattern, the adjustment of the weight factor being proportional to the difference of the weight factor from the input: Δ Wi=η(t)(X(t)-Wi(t)), η (t) represents a variable learning rate, decaying with time;
repeating the competition and learning steps until the excitatory neuron is stably corresponding to the input sample, and finishing the training.
It should be noted that the SOM network is trained 1 ten thousand times by using the training samples, and data is randomly extracted from the training samples to be trained each time.
As shown in fig. 4, the SOM network has only one "planar" layer of neurons, and each input is input to each neuron, i.e. each neuron has a connection to each input, so that for the ith neuron, the vector dimension formed by the weight coefficients between it and the input is equal to the dimension of the input vector. Each neuron has no specific output layer, and its excited state is the output. From the output state of the SOM network, not only the class to which the input mode belongs can be judged and the output node represents a certain class of mode, but also the general distribution condition of the whole data area can be obtained, i.e. the general essential characteristics of all data distribution can be captured from the sample data. Therefore, the SOM network can be used for high voltage circuit breaker failure mode identification.
In order to simplify the operation, the embodiment also performs normalization processing on the energy of 8 frequency bands of the test data. Because the SOM network adopts a competitive learning mode and has no special output layer, sample data can be input circularly, namely the SOM network is suitable for small sample data classification. In order to train the SOM network better, improve the network pattern recognition accuracy and simultaneously consider the simulation time, the invention trains the network 1 ten thousand times by adopting samples, and randomly extracts data from the training samples to train each time, thereby ensuring the reliability of the network.
Test data is input into the trained SOM network, and whether the test network can identify the fault type of the high-voltage circuit breaker or not is tested. Simulation experiments show that the dry-type electric reactor identification method based on the equivalent impedance and the SOM network has a good diagnosis effect.
Table 1 below shows the test results of the data of the SOM network 15 groups, and the test results are the same as the target values, so that the SOM network can correctly identify the failure mode, and has a certain stability, and the accuracy rate reaches 100%. When the number of training samples is increased, the recognition result is better. Therefore, the high-voltage circuit breaker fault mode identification method based on the combination of the wavelet packet energy and the SOM network has a good effect on reactor fault diagnosis.
Table 1SOM network 15 set of data test results
|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
|
1 | 1 | 1 | 1 | 1 | 2 | 2 | 2 | 2 | 2 | 3 | 3 | 3 | 3 | 3 |
|
1 | 1 | 1 | 1 | 1 | 2 | 2 | 2 | 2 | 2 | 3 | 3 | 3 | 3 | 3 |
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (10)
1. A fault identification method for a dry reactor is characterized by comprising the following steps:
respectively acquiring a vibration signal of the dry-type reactor in a normal state and a vibration signal of the dry-type reactor in a fault state;
respectively processing the vibration signal in the normal state and the vibration signal in the fault state to obtain processing results;
inputting the processing result into an SOM network, and training the SOM network to obtain a trained SOM network;
inputting data to be tested to the trained SOM network to obtain an output result;
and judging the fault mode of the dry type reactor according to the output result.
2. The method of identifying a fault in a dry reactor according to claim 1, further comprising:
establishing a reactor turn-to-turn short circuit model;
calculating the equivalent resistance and the equivalent reactance of each turn-to-turn coil in the turn-to-turn short circuit model;
constructing a first eigenvector from the equivalent resistance and the equivalent reactance.
3. The method of identifying a fault in a dry reactor according to claim 2, wherein the processing of the vibration signal in the normal state and the vibration signal in the fault state to obtain the processing results includes:
respectively performing wavelet decomposition on the vibration signal in the normal state and the vibration signal in the fault state to obtain a plurality of frequency band signals;
and carrying out normalization processing on the frequency band signals to construct a second feature vector.
4. The method for identifying a fault in a dry reactor according to claim 3, wherein the step of inputting the processing result to an SOM network and training the SOM network to obtain a trained SOM network comprises:
constructing a third feature vector according to the first feature vector and the second feature vector;
inputting the third feature vector to the SOM network.
5. The method for identifying the fault of the dry reactor according to claim 3, wherein the step of performing wavelet decomposition on the vibration signal in the normal state and the vibration signal in the fault state to obtain a plurality of frequency band signals comprises the steps of:
and performing two-layer wavelet packet transformation on the vibration signal in the normal state and the vibration signal in the fault state by adopting a db10 wavelet to obtain four frequency band signals.
6. The method for identifying the fault of the dry type reactor according to claim 5, wherein the energy expression of the frequency band signal is as follows:
wherein E is2(i) Representing the energy of the ith frequency band, W (2, i) representing the ith frequency band signal of the second layer after wavelet packet two-layer decomposition, xikThe value of the kth discrete point of the ith frequency band signal W (2, i), i being 0,1, …,3, i represents four frequency bands, k being 1,2, …, N represents the number of discrete points.
7. The method of identifying a fault in a dry reactor according to claim 6, wherein the second eigenvector H is constructed from the ratio of the energy in each frequency band2Expressed as:
H2=[E2(0)/S2,…,E2(3)/S2],
8. the method of identifying a fault in a dry reactor according to claim 2, further comprising:
and dividing the second feature vectors into two groups, wherein one group is used for inputting the SOM network for training, and the other group is used for inputting the SOM network for testing.
9. The method for identifying a fault in a dry reactor according to claim 1, wherein the step of inputting the processing result to an SOM network and training the SOM network to obtain a trained SOM network comprises:
competition, given the tth input X (t), each neuron weight vector W is calculatedi(t) relative distance D from input vectori(t):||X(t)-Wi(t)||=Di(t), if the minimum relative distance is selected as the winner, then exciting, namely: i.e. i*(t)={i:minDi(t)};
Learning, the winner and its neighbors are allowed to learn the input pattern, the adjustment of the weight factor being proportional to the difference of the weight factor from the input: Δ Wi=η(t)(X(t)-Wi(t)), η (t) represents a variable learning rate, decaying with time;
repeating the competition and learning steps until the excitatory neuron is stably corresponding to the input sample, and finishing the training.
10. A fault identification method for a dry-type reactor according to any one of claims 1 to 9, wherein the fault conditions include winding turn-to-turn short circuit and insulation aging.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911070723.8A CN110726957A (en) | 2019-11-05 | 2019-11-05 | Fault identification method of dry-type reactor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911070723.8A CN110726957A (en) | 2019-11-05 | 2019-11-05 | Fault identification method of dry-type reactor |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110726957A true CN110726957A (en) | 2020-01-24 |
Family
ID=69223763
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911070723.8A Pending CN110726957A (en) | 2019-11-05 | 2019-11-05 | Fault identification method of dry-type reactor |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110726957A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114460490A (en) * | 2022-01-17 | 2022-05-10 | 武汉朕泰智能科技有限公司 | Method for diagnosing ground short circuit fault inside high-voltage cable sheath grounding box |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1484034A (en) * | 2002-09-18 | 2004-03-24 | 新疆特变电工股份有限公司 | On-line intelligent monitoring system for transformer and intelligent analysis diagnosis method thereof |
CN105335698A (en) * | 2015-09-15 | 2016-02-17 | 河南理工大学 | Gear failure diagnosis method based on adaptive genetic algorithm and SOM (Self-Organizing Map) network |
CN105699840A (en) * | 2016-03-09 | 2016-06-22 | 珠海蓝瑞盟电气有限公司 | Method for judging inter-turn short-circuit fault of reactor |
CN106404078A (en) * | 2016-11-28 | 2017-02-15 | 山东科技大学 | Transformer online monitoring device |
CN106546869A (en) * | 2016-12-27 | 2017-03-29 | 哈尔滨理工大学 | A kind of hollow power reactor turn-to-turn insulation On-line Fault monitoring device |
CN107607303A (en) * | 2017-09-13 | 2018-01-19 | 河海大学 | Mechanical Failure of HV Circuit Breaker recognition methods based on wavelet packet Yu SOM networks |
CN109856501A (en) * | 2019-01-13 | 2019-06-07 | 广西电网有限责任公司南宁供电局 | A kind of on-load tap changers of transformers and winding deformation fault detection method |
CN109975634A (en) * | 2019-03-04 | 2019-07-05 | 河南理工大学 | A kind of fault diagnostic method for transformer winding based on atom sparse decomposition |
CN110147739A (en) * | 2019-04-29 | 2019-08-20 | 河海大学 | A kind of Reactor Fault recognition methods based on Multifractal Analysis |
CN110221139A (en) * | 2019-05-05 | 2019-09-10 | 新奥数能科技有限公司 | A kind of failure prediction method of dry-type transformer, apparatus and system |
-
2019
- 2019-11-05 CN CN201911070723.8A patent/CN110726957A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1484034A (en) * | 2002-09-18 | 2004-03-24 | 新疆特变电工股份有限公司 | On-line intelligent monitoring system for transformer and intelligent analysis diagnosis method thereof |
CN105335698A (en) * | 2015-09-15 | 2016-02-17 | 河南理工大学 | Gear failure diagnosis method based on adaptive genetic algorithm and SOM (Self-Organizing Map) network |
CN105699840A (en) * | 2016-03-09 | 2016-06-22 | 珠海蓝瑞盟电气有限公司 | Method for judging inter-turn short-circuit fault of reactor |
CN106404078A (en) * | 2016-11-28 | 2017-02-15 | 山东科技大学 | Transformer online monitoring device |
CN106546869A (en) * | 2016-12-27 | 2017-03-29 | 哈尔滨理工大学 | A kind of hollow power reactor turn-to-turn insulation On-line Fault monitoring device |
CN107607303A (en) * | 2017-09-13 | 2018-01-19 | 河海大学 | Mechanical Failure of HV Circuit Breaker recognition methods based on wavelet packet Yu SOM networks |
CN109856501A (en) * | 2019-01-13 | 2019-06-07 | 广西电网有限责任公司南宁供电局 | A kind of on-load tap changers of transformers and winding deformation fault detection method |
CN109975634A (en) * | 2019-03-04 | 2019-07-05 | 河南理工大学 | A kind of fault diagnostic method for transformer winding based on atom sparse decomposition |
CN110147739A (en) * | 2019-04-29 | 2019-08-20 | 河海大学 | A kind of Reactor Fault recognition methods based on Multifractal Analysis |
CN110221139A (en) * | 2019-05-05 | 2019-09-10 | 新奥数能科技有限公司 | A kind of failure prediction method of dry-type transformer, apparatus and system |
Non-Patent Citations (2)
Title |
---|
黄新波等: "基于功角特性的干式空心电抗器匝间绝缘在线监测技术", 《电力自动化设备》 * |
黄新波等: "干式空心电抗器匝间短路故障在线监测技术", 《电力系统自动化》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114460490A (en) * | 2022-01-17 | 2022-05-10 | 武汉朕泰智能科技有限公司 | Method for diagnosing ground short circuit fault inside high-voltage cable sheath grounding box |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109685138B (en) | XLPE power cable partial discharge type identification method | |
CN112041693B (en) | Power distribution network fault positioning system based on mixed wave recording | |
CN109612708B (en) | Power transformer on-line detection system and method based on improved convolutional neural network | |
CN109829402B (en) | GS-SVM-based bearing damage degree diagnosis method under different working conditions | |
CN109033612B (en) | Transformer fault diagnosis method based on vibration noise and BP neural network | |
CN106443310B (en) | A kind of transformer fault detection method based on SOM neural network | |
CN107505133A (en) | The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM | |
CN108646149A (en) | Fault electric arc recognition methods based on current characteristic extraction | |
CN107766816A (en) | A kind of Mechanical Failure of HV Circuit Breaker recognition methods based on LVQ neutral nets | |
CN106548230A (en) | Diagnosis Method of Transformer Faults based on Modified particle swarm optimization neutral net | |
CN112766140A (en) | Transformer fault identification method based on kernel function extreme learning machine | |
CN110174255B (en) | Transformer vibration signal separation method based on radial basis function neural network | |
CN106384302B (en) | Power distribution network reliability evaluation method and system | |
CN112130026B (en) | MMC-HVDC single-pole grounding fault positioning method based on wavelet packet energy entropy and DBN | |
CN105137354B (en) | One kind is based on neutral net electrical fault detection method | |
CN109142851A (en) | A kind of novel power distribution network internal overvoltage recognition methods | |
CN109212392A (en) | Recognition methods, system and the Caffe convolutional neural networks of direct current cables shelf depreciation defect failure | |
CN110672905A (en) | CNN-based self-supervision voltage sag source identification method | |
CN112114232A (en) | Monopole earth fault distance measuring method based on wavelet decomposition and DBN | |
CN110610212A (en) | Fault classification method and fault classification device for transformer of power distribution network | |
CN110726957A (en) | Fault identification method of dry-type reactor | |
CN112596016A (en) | Transformer fault diagnosis method based on integration of multiple one-dimensional convolutional neural networks | |
CN116089857A (en) | Transformer fault identification method based on CEEMDAN-DBN | |
CN115267428A (en) | LCC-MMC single-pole grounding fault positioning method based on VMD-ET feature selection | |
Rajamani et al. | Identification of simultaneously occurring dynamic disc-to-disc insulation failures in transformer winding under impulse excitation |
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 |
Application publication date: 20200124 |
|
RJ01 | Rejection of invention patent application after publication |