CN115144702A - Ground fault type identification method based on self-organizing competitive neural network - Google Patents
Ground fault type identification method based on self-organizing competitive neural network Download PDFInfo
- Publication number
- CN115144702A CN115144702A CN202210885492.1A CN202210885492A CN115144702A CN 115144702 A CN115144702 A CN 115144702A CN 202210885492 A CN202210885492 A CN 202210885492A CN 115144702 A CN115144702 A CN 115144702A
- Authority
- CN
- China
- Prior art keywords
- ground fault
- neural network
- self
- sequence current
- zero sequence
- 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
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Locating Faults (AREA)
Abstract
A ground fault type identification method based on a self-organizing competitive neural network belongs to the field of relay protection of power systems. The method is characterized in that: the method comprises the following steps: step a, acquiring a zero sequence current after a fault and solving a power frequency component; step b, acquiring a power frequency component amplitude variation trend function of the zero sequence current; step c, extracting characteristic parameters; d, executing the step e if the unstable grounding fault does not occur; step e, constructing a neural network based on a self-organizing competition type; and f, judging whether the tree line discharge fault or other types of stably grounded ground faults occur in the power distribution network by the self-organizing competitive neural network, and outputting. In the ground fault type identification method based on the self-organizing competitive neural network, the competitive neural network is utilized to identify the type of the ground fault in the power distribution network, particularly, the tree line discharge fault is accurately identified, and the blank in the field is filled.
Description
Technical Field
A ground fault type identification method based on a self-organizing competitive neural network belongs to the field of relay protection of power systems.
Background
The high-resistance earth faults of the medium-voltage distribution network account for 5% -10% of the total number of the single-phase earth faults, and one part of the high-resistance earth faults are tree line discharge faults of overhead lines grounded through trees. The electrical characteristic quantity of the tree line discharge fault is weak, and often cannot reach the threshold value of the starting of the protection device, and the long-term existence of the fault can cause the risk of mountain fire, thus seriously threatening the life and property safety of people.
In the present stage, the medium voltage distribution network ground fault type identification method mainly utilizes algorithms such as wavelet transformation and the like to extract time-frequency domain characteristics of ground fault zero sequence information quantity as characteristic data, or directly uses the waveform of the ground fault zero sequence information quantity as the characteristic data, and inputs the characteristic data into a support vector machine or a neural network to realize the type identification of the ground fault. The existing ground fault identification method for the medium-voltage distribution network is too complex in data preprocessing means, fuzzy boundaries and uncertainty exist in the machine learning intelligent algorithm, a reliable and accurate tree line discharge fault detection method does not exist, whether tree line discharge faults occur or not cannot be identified, and the blank of the research field exists.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method overcomes the defects of the prior art, and provides the ground fault type identification method based on the self-organizing competitive neural network, which utilizes the competitive neural network to identify the type of the ground fault in the power distribution network, particularly accurately identifies the tree line discharge fault.
The technical scheme adopted by the invention for solving the technical problem is as follows: the method for identifying the type of the ground fault based on the self-organizing competitive neural network is characterized by comprising the following steps: the method comprises the following steps:
step a, acquiring a zero sequence current signal after a fault after a ground fault occurs in a distribution line, and calculating a power frequency component of the zero sequence current signal after the fault;
b, acquiring a trend function of the amplitude change of the power frequency component of the zero sequence current signal according to the power frequency component of the zero sequence current signal after the fault;
c, extracting characteristic parameters of the zero sequence current signal amplitude variation trend function by a curve fitting means;
d, judging whether the type of the ground fault occurring in the power distribution network is an unstable grounding ground fault or not, and if the type of the ground fault occurring in the power distribution network is the unstable grounding ground fault, finishing the judgment; if the type of the ground fault occurring in the power distribution network is not the ground fault of unstable grounding, executing the step e;
step e, constructing a self-organizing competition type-based neural network;
and f, judging whether the tree line discharge fault or other types of stably grounded ground faults occur in the power distribution network by the self-organizing competitive neural network, and outputting.
Preferably, when step e is executed, the method includes the following steps:
step e-1, obtaining a symbol matrix and an Euclidean distance matrix of characteristic parameters of the amplitude variation trend function of the zero sequence current;
step e-2, calculating the state of the competitive layer element;
computing the State S of Competition layer neurons j j ：
Wherein inputsum i Is the ith element of the input layer data;
e-3, solving winning neurons which represent the neurons with the largest weighting values in the classification mode competition layer of the input samples to win, and outputting:
and e-4, calculating weight correction of the winning neuron, wherein for all input layer neurons i, the weight correction comprises the following steps:
wherein, alpha is a learning coefficient, and is generally 0 < alpha < -1; μ is the number of neurons with an output of 1, i.e
E-5, calculating Euclidean distances between the data of the input layer corresponding to the neuron with the output of 1 and all the data of the input layer, and seeking similar data;
and e-6, judging whether the convergence condition is met, if so, executing the step e-7, and if not, returning to the step e-2.
And e-7, returning to the step f.
Preferably, in step e-5, the formula for calculating the similar data is:
wherein inputsum output＝1 i is the ith element of the input layer data corresponding to the neuron whose output is 1, inputsumi is the ith element corresponding to all the input layer data, when ED (inputsum) output＝1 inputsum) is less than a predetermined threshold value.
Preferably, when the step d is executed, a mean square deviation value of the two signals is obtained according to a difference between the trend function of the zero sequence current signal power frequency component amplitude change obtained in the step b and the trend function obtained by fitting in the step c, and if the mean square deviation value of the two signals is smaller than a set threshold value, it indicates that the type of the ground fault occurring in the power distribution network is an unsteady grounded ground fault; and when the mean square deviation value of the two signals is larger than or equal to a set threshold value, indicating that the type of the ground fault occurring in the power distribution network is a stably grounded ground fault.
Preferably, in the step b, the trend function Amp _ i of the amplitude change of the power frequency component of the zero sequence current signal 0 (k) Comprises the following steps:
wherein: a. The 1 Representing the magnitude of the exponential component, A 2 Attenuation/gain factor representing an exponential component, A 3 Representing the magnitude of the constant dc component.
Preferably, the mean square error value MSE (Amp _ i) of the two signals 0 (k)，y (3) ) Comprises the following steps:
wherein n is the data length of the zero sequence current signal, W i Amp _ i, a weight of the root mean square difference 0 (k) Is a trend function of the amplitude change of the power frequency component of the zero sequence current signal, y (3) Is the difference of the trend function.
Preferably, when step c is performed, the following steps are included:
step c-1, fitting Amp _ i with a cubic polynomial 0 (k) And extracting characteristic parameters and amplitude variation trend function Amp _ i 0 (k) The fitting formula of (a) is:
y (3) ＝B 3 x 3 +B 2 x 2 +B 1 x+B 0
wherein, B 3 ～B 0 Is a fitting parameter;
step c-2, performing Taylor expansion on the trend function of the amplitude change of the power frequency component of the zero-sequence current signal, and neglecting the higher-order term of the Taylor expansion formula to obtain:
wherein: a. The 1 Representing the magnitude of the exponential component, A 2 Attenuation/gain factor representing an exponential component, A 3 Represents the magnitude of the constant dc component;
amp _ i fitting by three times 0 (k) And the characteristic parameters of the image are extracted,wherein the cubic fitting parameter is related to Amp _ i 0 (k) The corresponding relation of the Taylor expansion parameters is as follows:
wherein: a. The 1 Representing the magnitude of the exponential component, A 2 Attenuation/gain factor representing an exponential component, A 3 Represents the magnitude of the constant dc component; b is 3 ～B 0 Are fitting parameters.
Compared with the prior art, the invention has the beneficial effects that:
in the ground fault type identification method based on the self-organizing competitive neural network, the competitive neural network is utilized to identify the type of the ground fault in the power distribution network, particularly the tree line discharge fault is accurately identified, the blank in the field is filled, and the tree line discharge fault, the intermittent ground fault and other types of ground faults of the medium-voltage power distribution network are timely and accurately identified.
In the method for identifying the type of the ground fault based on the self-organizing competitive neural network, aiming at the characteristic that the amplitude variation trend of the zero-sequence current of the intermittent ground fault is approximate to a square wave signal, the frequency component is complex and is difficult to be completely fitted by using a curve fitting mode, and other fault signals can be completely fitted, whether the intermittent ground fault occurs or not is judged by using the difference between the amplitude variation trend of the zero-sequence current of the intermittent ground fault and different ground faults and the root-mean-square difference value of the fitted amplitude variation trend, and the method has the advantages of low calculated amount and easiness in setting.
The method comprises the steps of extracting characteristic parameters of a zero sequence current signal amplitude value change trend function by utilizing curve fitting means such as a cubic polynomial and the like, constructing a neural network based on a self-organizing competition structure, taking the characteristic parameters obtained by fitting as a data set of a neural network input layer and further processing the data set, determining a classification result in a mode of mutual competition and after-competition weight adjustment of neurons of the competition layer, judging whether a tree line discharge fault occurs, finally realizing identification of intermittent ground faults, tree line discharge faults and other types of ground faults of the medium-voltage distribution network, and having high classification and identification precision.
In the ground fault type identification method based on the self-organizing competitive neural network, the distribution automation device is adopted to collect the zero sequence current signals in a fixed time window after the fault occurs, the FFT is utilized to extract the power frequency component and construct the zero sequence current amplitude variation trend function, the influence of the tree line discharge fault harmonic component is reduced, the calculated amount is reduced, and the calculation speed of the distribution automation device is improved.
In the method for identifying the type of the ground fault based on the self-organizing competitive neural network, data input to an input layer of the self-organizing competitive neural network are further classified and processed, a symbol matrix and an Euclidean distance matrix are extracted, and the problems of high dispersion degree of original data, dispersed inherent characteristics and large slow calculation amount of neural network classification are solved; and by searching the corresponding relation between the input layer data corresponding to the neuron with the output of 1 and all the input layer data, the learning parameters are adaptively adjusted, and the classification speed and precision of the neural network are improved.
The zero sequence current signal can be acquired through traditional power frequency sensor acquisition or three-phase synthesis, additional primary equipment is not needed to be added, other primary equipment is not needed to be matched, and the practical application value is high.
Characteristic parameters of the zero sequence current signal amplitude change trend function are extracted by means of polynomial, fourier series fitting, prony algorithm and other curve fitting methods, a sign matrix of the characteristic parameters is constructed, and a criterion is constructed according to whether negative elements or products of elements or ratios exist in the sign matrix, so that whether tree line discharge faults and other ground faults occur is detected.
Drawings
Fig. 1 is a flowchart of a method for identifying a ground fault type based on a self-organizing competitive neural network.
Fig. 2 is a neural network construction flow chart of a ground fault type identification method based on a self-organizing competitive neural network.
Detailed Description
Fig. 1-2 show preferred embodiments of the present invention, and the present invention will be further described with reference to fig. 1-2.
the distribution automation device arranged in the distribution network judges whether an earth fault occurs in the distribution network, and executes step 1002 when the earth fault occurs.
the distribution automation device in the power distribution network collects zero sequence current signals in a fixed time window after a fault occurs according to a zero sequence transformer arranged in the power distribution network. Defining a zero sequence current signal i after the ground fault of the power distribution network 0 And (n) n is the data length of the zero sequence current signal, depends on the length of a fixed time window and a power distribution automatic sampling device, generally, the length of the fixed time window is more than or equal to 150s, and the sampling frequency is more than or equal to 6400Hz.
distribution automation device in power distribution network utilizes FFT to extract zero sequence current signal i after power distribution network ground fault occurs 0 (n) power frequency component i 0 (k) Namely:
wherein n represents the data length of the zero sequence current signal, k represents the data length of the power frequency component, and j represents the imaginary part.
power frequency component i of zero sequence current signal after ground fault occurs 0 (k) The amplitude characteristics of the contained frequency and the corresponding frequency are further obtained to obtain a trend function Amp _ i of the amplitude change of the power frequency component of the zero sequence current signal 0 (k)：
Wherein: a. The 1 Representing the magnitude of the exponential component, A 2 Attenuation/gain factor representing an exponential component, A 3 Representing the magnitude of the constant dc component.
extracting characteristic parameters of the zero sequence current signal amplitude variation trend function by utilizing polynomial fitting or Fourier series fitting or Prony algorithm and other curve fitting means;
the method specifically comprises the following steps:
step 1005-1, obtaining Amp _ i of zero sequence current signal power frequency component amplitude variation 0 (k) The fitting formula of (1);
to fit Amp _ i by a cubic polynomial 0 (k) And extracting characteristic parameters thereof, for example, amplitude variation trend function Amp _ i 0 (k) The fitting formula of (a) is:
y (3) ＝B 3 x 3 +B 2 x 2 +B 1 x+B 0
wherein, B 3 ～B 0 Are fitting parameters.
Step 1005-2, carrying out Taylor expansion on the trend function of the amplitude change of the power frequency component of the zero-sequence current signal to obtain:
in the Taylor expansion described above, the attenuation/gain factor A is given for high resistance ground faults, especially tree line discharge faults 2 Is larger, so the higher order terms in the taylor expansion can be neglected, which can be approximated as:
amp _ i fitting by three times 0 (k) Extracting characteristic parameters of the three-degree fitting parameters, wherein the three-degree fitting parameters and Amp _ i 0 (k) Taylor's teaThe corresponding relation of the expansion parameters is as follows:
calculating a zero sequence current amplitude variation trend function Amp _ i 0 (k) Trend function y obtained by curve fitting (3) To obtain the mean square error value MSE (Amp _ i) of the two signals 0 (k)，y (3) )：
Wherein n is the data length of the zero sequence current signal, W i The weight of the root mean square difference is generally set to 1.
When MSE (Amp _ i) 0 (k)，y (3) ) Less than a set threshold value
If yes, indicating that the type of the ground fault occurring in the power distribution network is an unsteady grounded ground fault, and executing step 1009; when MSE (Amp _ i) 0 (k)，y (3) ) Greater than or equal to a set threshold value
If so, it indicates that the type of the ground fault occurring in the power distribution network is a stable ground fault, and step 1007 is executed.
For presetting the threshold value, it can be set by itself, e.g.
and constructing a neural network based on a self-organizing competition type, wherein the neural network comprises an input layer and a competition layer, the number N of neurons in the input layer is consistent with the number of the characteristic parameters, and the number M of the competition layer is generally more than twice the number N of the neurons in the input layer. The weight value of the connection between the input layer and the competition layer is omega ij (i =1
As shown in fig. 2, the present step includes the following steps:
step 1007-1, obtaining a symbol matrix and an Euclidean distance matrix of characteristic parameters of a zero sequence current amplitude variation trend function;
for data (B) = [ B ] of input layer 0 B 1 B 2 B 3 ]Further processing is performed to obtain a sign matrix sgn (B) thereof, wherein:
sgn(B)＝[sgn(B 0 ) sgn(B 1 ) sgn(B 2 ) sgn(B 3 )]
calculation of B 1 、B 2 、B 3 And B 0 Euclidean distance ED (B) = [ ED (B) =between 0 B 1 ) ED(B 0 B 2 ) ED(B 0 B 3 )]Wherein:
thus, the data input to the neural network input layer is:
inputsum＝[data(B) sgn(B) ED(B)]。
step 1007-2, calculating the state of the competitive layer element;
computing state S of competition layer neuron j j ：
Wherein, inputsum i Is the ith element of the input layer data.
Step 1007-3, solving winning neurons;
and solving winning neurons which represent the classification mode of the input sample, wherein the neurons with the largest weighting values in the competition layer win, and the output is as follows:
step 1007-4, calculating weight correction of the winning neuron;
calculating weight correction of the winning neuron, and for all input layer neurons i, there are:
wherein alpha is a learning coefficient, and is generally 0-alpha < 1; μ is the number of neurons with an output of 1, i.e
Step 1007-5, self-adapting learning parameters;
calculating Euclidean distances between the input layer data corresponding to the neuron with the output of 1 and all the input layer data, and seeking similar data, wherein the calculation formula is as follows:
wherein inputsum output＝1 i is the ith element of the input layer data corresponding to the neuron with output of 1, and inputsumi is the ith element corresponding to all the input layer data. When ED (inputsum) output＝1 inputsum) is smaller than a preset threshold value, the two sets of data are considered to be similar, and the number of similar data is set as gamma. And (3) taking gamma as a reference quantity and learning coefficient alpha and gamma in direct proportion as a principle, and performing self-adaptive adjustment.
Step 1007-6, whether the convergence condition is satisfied;
judging whether the convergence condition is met, if the convergence condition is met, executing the step 1007-7, and if the convergence condition is not met, returning to the step 1007-2.
Step 1007-7, returning to the main process; returning to the main flow, step 1008 is performed.
and (4) circulating steps 1007-3 to 1007-5 in step 1007, judging whether the tree line discharge fault or other types of stably grounded ground faults occur in the power distribution network by the neural network based on the self-organizing competition type, and outputting.
The foregoing is directed to preferred embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow. However, any simple modification, equivalent change and modification of the above embodiments according to the technical essence of the present invention are within the protection scope of the technical solution of the present invention.
Claims (7)
1. A ground fault type identification method based on a self-organizing competitive neural network is characterized in that: the method comprises the following steps:
step a, after a ground fault occurs in a distribution line, acquiring a zero sequence current signal after the fault, and calculating a power frequency component of the zero sequence current signal after the fault;
b, acquiring a trend function of the amplitude change of the power frequency component of the zero sequence current signal according to the power frequency component of the zero sequence current signal after the fault;
c, extracting characteristic parameters of the zero sequence current signal amplitude variation trend function by a curve fitting means;
d, judging whether the type of the ground fault occurring in the power distribution network is an unstable grounding ground fault or not, and if the type of the ground fault occurring in the power distribution network is the unstable grounding ground fault, finishing the judgment; if the type of the ground fault occurring in the power distribution network is not the unstable grounded ground fault, executing the step e;
step e, constructing a neural network based on a self-organizing competition type;
and f, judging whether the tree line discharge fault or other types of stably grounded ground faults occur in the power distribution network by the self-organizing competition type-based neural network, and outputting.
2. The ground fault type identification method based on the self-organizing competition type neural network as claimed in claim 1, wherein: when the step e is executed, the method comprises the following steps:
step e-1, obtaining a symbol matrix and an Euclidean distance matrix of characteristic parameters of the amplitude variation trend function of the zero sequence current;
step e-2, calculating the state of the competitive layer element;
computing state S of competition layer neuron j j ：
Wherein, inputsum i Is the ith element of the input layer data;
e-3, solving winning neurons which represent the neurons with the largest weighting values in the classification mode competition layer of the input sample to win, and outputting:
and e-4, calculating weight correction of the winning neuron, wherein for all input layer neurons i, the weight correction comprises the following steps:
wherein, alpha is a learning coefficient, and is generally 0 < alpha < -1; μ is the number of neurons with an output of 1, i.e
E-5, calculating Euclidean distances between the data of the input layer corresponding to the neuron with the output of 1 and all the data of the input layer, and seeking similar data;
and e-6, judging whether the convergence condition is met, executing the step e-7 if the convergence condition is met, and returning to the step e-2 if the convergence condition is not met.
And e-7, returning to the step f.
3. The method for identifying the type of the ground fault based on the self-organizing competitive neural network as claimed in claim 2, wherein: in step e-5, the calculation formula of the similar data is:
wherein inputsum output＝1 i is the ith element of the input layer data corresponding to the neuron whose output is 1, inputsumi is the ith element corresponding to all the input layer data, when ED (inputsum) output＝1 inputsum) is less than a predetermined threshold value.
4. The ground fault type identification method based on the self-organizing competition type neural network as claimed in claim 1, wherein: when the step d is executed, acquiring a mean square deviation value of the two signals according to the difference between the trend function of the zero sequence current signal power frequency component amplitude change acquired in the step b and the trend function obtained by fitting in the step c, and if the mean square deviation value of the two signals is smaller than a set threshold value, indicating that the type of the ground fault occurring in the power distribution network is an unsteady grounded ground fault; and when the mean square deviation value of the two signals is larger than or equal to a set threshold value, indicating that the type of the ground fault occurring in the power distribution network is a stably grounded ground fault.
5. The ground fault type identification method based on the self-organizing competitive neural network as claimed in claim 1, wherein: in the step b, a trend function Amp _ i of the amplitude change of the power frequency component of the zero sequence current signal 0 (k) Comprises the following steps:
wherein: a. The 1 Representing the magnitude of the exponential component, A 2 Attenuation/gain factor representing an exponential component, A 3 Representing the magnitude of the constant dc component.
6. The method for identifying the type of the ground fault based on the self-organizing competitive neural network as claimed in claim 4, wherein: mean square error value MSE (Amp _ i) of the two signals 0 (k)，y (3) ) Comprises the following steps:
wherein n is the data length of the zero sequence current signal, W i Amp _ i, a weight of the root mean square difference 0 (k) Is a trend function of the amplitude variation of the power frequency component of the zero sequence current signal, y (3) Is the difference of the trend function.
7. The ground fault type identification method based on the self-organizing competitive neural network as claimed in claim 1, wherein: when step c is executed, the method comprises the following steps:
step c-1, fitting Amp _ i with a cubic polynomial 0 (k) And extracting characteristic parameters, amplitude variation trend function Amp _ i 0 (k) The fitting formula of (a) is:
y (3) ＝B 3 x 3 +B 2 x 2 +B 1 x+B 0
wherein, B 3 ～B 0 As fitting parameters；
Step c-2, performing Taylor expansion on the trend function of the amplitude change of the power frequency component of the zero-sequence current signal, and neglecting the higher-order terms of the Taylor expansion formula to obtain:
wherein: a. The 1 Representing the magnitude of the exponential component, A 2 Attenuation/gain factor representing an exponential component, A 3 Represents the magnitude of the constant dc component;
amp _ i fitting by three times 0 (k) Extracting characteristic parameters of the three-degree fitting parameters, wherein the three-degree fitting parameters and Amp _ i 0 (k) The corresponding relation of the Taylor expansion parameters is as follows:
wherein: a. The 1 Representing the magnitude of the exponential component, A 2 Attenuation/gain factor representing an exponential component, A 3 Represents the magnitude of the constant dc component; b is 3 ～B 0 Are fitting parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210885492.1A CN115144702A (en) | 2022-07-26 | 2022-07-26 | Ground fault type identification method based on self-organizing competitive neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210885492.1A CN115144702A (en) | 2022-07-26 | 2022-07-26 | Ground fault type identification method based on self-organizing competitive neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115144702A true CN115144702A (en) | 2022-10-04 |
Family
ID=83414688
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210885492.1A Pending CN115144702A (en) | 2022-07-26 | 2022-07-26 | Ground fault type identification method based on self-organizing competitive neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115144702A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116087692A (en) * | 2023-04-12 | 2023-05-09 | 国网四川省电力公司电力科学研究院 | Distribution network tree line discharge fault identification method, system, terminal and medium |
| CN117151198A (en) * | 2023-09-06 | 2023-12-01 | 广东海洋大学 | Underwater sound passive positioning method and device based on self-organizing competitive neural network |
-
2022
- 2022-07-26 CN CN202210885492.1A patent/CN115144702A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116087692A (en) * | 2023-04-12 | 2023-05-09 | 国网四川省电力公司电力科学研究院 | Distribution network tree line discharge fault identification method, system, terminal and medium |
| CN117151198A (en) * | 2023-09-06 | 2023-12-01 | 广东海洋大学 | Underwater sound passive positioning method and device based on self-organizing competitive neural network |
| CN117151198B (en) * | 2023-09-06 | 2024-04-09 | 广东海洋大学 | Underwater sound passive positioning method and device based on self-organizing competitive neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115144702A (en) | Ground fault type identification method based on self-organizing competitive neural network | |
| CN110082640B (en) | Distribution network single-phase earth fault identification method based on long-time memory network | |
| US11656263B2 (en) | Effective feature set-based high impedance fault detection | |
| WO2020015277A1 (en) | Arc light fault identifying device and method based on panoramic information | |
| WO2022067562A1 (en) | Method and device for diagnosing fault arc, and computer-readable storage medium | |
| CN109800660B (en) | Voltage sag source identification method and system based on big data clustering | |
| CN108139432B (en) | Method for correcting saturation effect in current transformer and intelligent electronic equipment thereof | |
| CN110609477B (en) | Electric power system transient stability discrimination system and method based on deep learning | |
| CN108535572B (en) | Metering system secondary circuit monitoring method and device based on fundamental wave zero sequence characteristics | |
| CN111537853A (en) | Intelligent detection method for partial discharge of switch cabinet based on multi-source heterogeneous data analysis | |
| CN110702966A (en) | Fault arc detection method, device and system based on probabilistic neural network | |
| CN107765065B (en) | Fundamental wave attenuation factor-based power distribution network magnetizing inrush current identification method | |
| CN111639583A (en) | Method and system for identifying power quality disturbance of power grid | |
| CN108089100A (en) | The detection method of small current neutral grounding system arc light resistance earth fault | |
| CN113592308B (en) | Monitoring data alarm threshold extraction method based on normal model | |
| Matz et al. | Automated power quality monitoring system for on-line detection and classification of disturbances | |
| CN113837895A (en) | Power distribution network abnormal event identification method and system based on power disturbance data | |
| CN107578016B (en) | Sparse representation-based automatic residual current waveform identification method | |
| CN109061354A (en) | A kind of electrical energy power quality disturbance recognition methods based on improvement PSO and SVM | |
| CN113514743A (en) | Construction method of GIS partial discharge pattern recognition system based on multi-dimensional features | |
| CN115144701A (en) | Ground fault type identification method by utilizing zero sequence current characteristic curve identification | |
| CN112269110A (en) | Arc fault judgment method | |
| JP2023502659A (en) | Machine Learning Based Method and Apparatus for Transmission Line Disturbance Classification | |
| CN115808600B (en) | DC arc detection method based on mathematical morphology and pattern recognition | |
| Lukomski et al. | Power system topology verification using artificial neural networks: maximum utilization of measurement data |
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 |