CN112240965A - Grounding line selection device and method based on deep learning algorithm - Google Patents
Grounding line selection device and method based on deep learning algorithm Download PDFInfo
- Publication number
- CN112240965A CN112240965A CN201910643661.9A CN201910643661A CN112240965A CN 112240965 A CN112240965 A CN 112240965A CN 201910643661 A CN201910643661 A CN 201910643661A CN 112240965 A CN112240965 A CN 112240965A
- Authority
- CN
- China
- Prior art keywords
- grounding
- line selection
- fault
- waveform
- image
- 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
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 21
- 238000013135 deep learning Methods 0.000 title claims abstract description 16
- 238000000034 method Methods 0.000 title claims description 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 18
- 230000001052 transient effect Effects 0.000 claims abstract description 8
- 238000010187 selection method Methods 0.000 claims abstract description 7
- 238000005070 sampling Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000011176 pooling Methods 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims description 25
- 238000013527 convolutional neural network Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 2
- 238000012360 testing method Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 1
- 238000012356 Product development Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Emergency Protection Circuit Devices (AREA)
Abstract
The invention discloses a grounding line selection device based on a deep learning algorithm, which comprises a sampling module, a starting module and a line selection module. The invention also discloses a grounding line selection method based on the deep learning algorithm, which comprises the following steps: acquiring a section of transient waveform of the grounding transient fault, displaying the zero sequence voltage and the zero sequence current waveform in the same image, marking data and a label file as input data, optimizing model parameters to an optimal value through a back propagation algorithm after convolution pooling and regression calculation, and making a classification result; and (3) leading the trained model into a grounding line selection device, judging whether a grounding fault occurs according to zero sequence voltage, processing the waveform image, sending the processed waveform image into a trained deep convolution neural network for identification, wherein the result is '0' which is a non-fault line, and the result is '1' which represents a fault line. The technical scheme can solve the problem of low line selection accuracy when the existing low-current grounding system is in single-phase grounding.
Description
Technical Field
The invention belongs to the field of relay protection of power systems, and particularly relates to a grounding line selection device and method of a low-current grounding system.
Background
In a 3kV-66kV low-current grounding system, single-phase grounding is a common fault type. When the small-current grounding system is in single-phase grounding, the fault voltage to ground is reduced, the non-fault voltage to ground is increased, the line voltage is still symmetrical, and in the state, the grounding current is very small, so that the operation can be allowed for 1-2h in order to ensure the power supply reliability. However, due to the non-fault phase arc light overvoltage, breakdown of an insulation weak part, saturation of a voltage transformer iron core, system overvoltage, cable burnout of the fault phase arc light, electric shock casualty accidents of human bodies and the like are easily caused, so that the single-phase grounding fault of the fault phase needs to be isolated in time after single-phase grounding, and safe and stable operation and power supply reliability of the system are guaranteed.
The traditional neural network has the problems of high learning sample quantity requirement, long learning time and the like due to difficult model construction, and cannot be used in practical application. The BP neural network has the advantages that the network structure is simple and easy to operate, but has two defects that the network structure is not easy to determine, and the prediction discrimination capability in large sample data set is obviously reduced. In addition, when the BP neural network is trained, the characteristic extraction must be carried out on the picture manually, some information is inevitably lost in the characteristic extraction process, and the prediction capability of the BP neural network is fundamentally reduced. Convolutional Neural Networks (CNN) are an improvement over BP Neural Networks. The method greatly reduces the number of parameters by utilizing four means of local connection, weight sharing, multi-core convolution and pooling, so that the layer number of the network can be deeper, the network structure is complex, the characteristics can be reasonably and implicitly extracted, and the method is more suitable for processing the image identification problem.
Google brain completed the development of "second generation machine learning system" TensorFlow and opened the source for code. TensorFlow has a significant improvement in performance, and architecture flexibility and portability are also enhanced over prior efforts. The Tensorflow has a multi-level structure, can be deployed in various servers, PC terminals, webpages and embedded devices, supports high-performance numerical calculation such as GPU, TPU, FPGA and the like, and is widely applied to product development inside Google and scientific research in various fields. Therefore, the neural network can be conveniently trained on the cloud, the server or the PC and then migrated to the embedded grounding route selection device by using the TensorFlow.
The invention patent CN201811393217.8 distribution network earth fault analysis method based on deep learning selects line positioning algorithm, which is a positioning method based on transient zero sequence current similarity principle and does not use deep learning algorithm to select line. The line selection positioning adopts zero sequence current synthesized by three-phase current, the single-phase grounding zero sequence current of a small current grounding system is very small and far smaller than load current, and the transformation ratio of a three-phase current transformer is far larger than that of a special zero sequence current transformer, so that the special zero sequence CT current is used for line selection in the industry.
The patents CN201810088104.0 "method for selecting single-phase ground fault line of power distribution network based on convolutional neural network", CN201710916960.6 "method for selecting single-phase ground fault line of small current ground system based on big data of power grid", CN201710138838.0 "device for selecting small current ground line based on neural network processor and its operation method", CN201610060984.1 "method for selecting single-phase short circuit line of power distribution network based on wavelet neural network", and CN201010149558.8 "method for selecting line of artificial neural network of power distribution network fault using S transformation energy sample attribute" do not adopt the function of waveform image recognition, but perform feature extraction in advance, the feature selection is unreasonable and incomplete, which inevitably results in some information loss in the process of feature extraction, and reduces the prediction capability of neural network.
Disclosure of Invention
The invention aims to provide a grounding line selection device and method based on a deep learning algorithm, which can solve the problem of low line selection accuracy when the existing low-current grounding system is grounded in a single phase.
In order to achieve the above purpose, the solution of the invention is:
a grounding line selection device based on a deep learning algorithm comprises:
the sampling module is used for collecting and sending bus zero-sequence voltage and zero-sequence current of each branch circuit to the starting module and the line selection module;
the starting module is used for judging whether a ground fault occurs or not; and the number of the first and second groups,
and the line selection module is used for identifying the bus zero sequence voltage and the zero sequence current of each branch circuit by the deep convolution neural network when the starting module judges that the ground fault occurs, and selecting a fault line.
A grounding line selection method based on a deep learning algorithm comprises the following steps:
step 3, leading the trained model into a grounding line selection device;
step 4, the grounding line selection device judges whether a grounding fault occurs according to the zero sequence voltage;
and 5, after the ground fault occurs, the ground line selection device processes the waveform image by using the method in the step 1, the processed waveform image is sent to a trained deep convolution neural network for identification, the image with the result of 0 is a non-fault line, and the image with the result of 1 represents a fault line.
In step 1, the zero sequence voltage and the zero sequence current waveforms are displayed in an overlapping manner in the same image.
In the step 2, a deep convolutional neural network is used for training and recognition.
In the step 2, the neural network is trained in the cloud server or the PC.
After the scheme is adopted, the target of the method is a 3kV-66kV low-current grounding system, a large number of single-phase grounding waveform images obtained on site are subjected to preprocessing such as graying, size compression and the like to obtain required training images, and a deep convolutional neural network model is trained. The grounding line selection device acquires three-phase voltage of a bus, zero-sequence voltage and zero-sequence current of each branch, judges the occurrence of single-phase grounding faults according to the zero-sequence voltage, performs image graying, size compression and other preprocessing on a waveform image at the moment of grounding to obtain a waveform, and uses a trained deep convolutional neural network to select lines.
Drawings
FIG. 1 is a flow chart of the line selection process using the grounding line selection device according to the present invention;
FIG. 2 is a schematic diagram of a single-phase ground fault raw waveform;
FIG. 3 is a schematic diagram of a waveform showing a combination of zero sequence current and voltage;
FIG. 4 is a schematic diagram of a waveform intercepting a fault instant;
fig. 5 is a schematic view of a waveform after graying.
Detailed Description
The technical solution and the advantages of the present invention will be described in detail with reference to the accompanying drawings.
The invention provides a grounding line selection device based on a deep learning algorithm, which comprises a sampling module, a starting module and a line selection module, wherein the sampling module comprises:
the sampling module is used for collecting and sending bus zero sequence voltage and zero sequence current of each branch circuit to the starting module and the line selection module;
the starting module is used for judging whether a ground fault occurs or not;
and the line selection module is used for identifying the waveforms of the bus zero-sequence voltage and the zero-sequence current of each branch by using a trained deep convolution neural network when the starting module judges that the ground fault occurs, and selecting a fault line.
As shown in fig. 1, the present invention further provides a grounding line selection method based on a deep learning algorithm, which includes the following steps:
1. preprocessing the grounding line selection waveform
Acquiring a field single-phase earth fault waveform and a single-phase earth fault waveform acquired in a true experiment, and dividing the waveforms into two groups: 100 training waveforms and 100 test waveforms. In order to ensure objectivity, the training waveform does not include a test waveform, and the training waveform and the test waveform respectively include about 20% of each of the metallic ground fault, the transition resistance ground fault of 500 ohms, 1000 ohms and 2000 ohms, and the arc ground fault of the system which is grounded through the arc suppression coil and is not grounded.
In order to enable the graph to be suitable for the input requirement of the convolutional neural network and improve the training and testing speed and precision, a grounding line selection waveform needs to be preprocessed before the model is trained and tested, the fault characteristics at the grounding moment mainly occur at the grounding moment, and the attenuation is fast, so that the waveform of 3ms at the grounding moment is selected, the training and recognition speed can be improved, and the line selection accuracy can also be high.
Fig. 2 is a diagram of original waveforms. U0 is bus current, I02 is fault line waveform, and I01, I03, I04 are non-fault line waveforms. To facilitate the recognition of the pattern by the convolutional network, the zero sequence voltage and the zero sequence current are displayed in the same pattern, as shown in fig. 3. In order to accelerate the learning efficiency, the transient waveform of about 3ms, in which the grounding transient fault characteristic is most obvious, is intercepted, as shown in fig. 4. To reduce the volume of the learning sample library, the image is further compressed and grayed out, as shown in fig. 5.
The original waveform contains a fault line and three non-fault lines, four training samples can be formed, the labels of the samples are respectively '0' for representing the non-fault lines, and '1' for representing the fault lines.
A large amount of training data are more favorable to the study of model, in order to increase training data sample quantity, have carried out different transient state moments in the left and right translation simulation waveform to the waveform, direct current component in the upper and lower translation simulation waveform, the opposite condition of upper and lower upset simulation waveform polarity, not only increase the quantity of training data, make the training model have robustness more moreover. The expanded training data exceeds 10000 samples.
The test data set uses the same grayed and compressed waveform as the training data.
2. Training deep convolutional neural networks
In training data, marking data and label files, storing the marked data and label files as TFRecord files supported by TensorFlow as input data, reading the files to a convolutional layer through an input layer, obtaining a basic model structure after convolution pooling and regression calculation, optimizing model parameters to an optimal value through an algorithm back propagation algorithm, and making a classification result.
3. Testing trained convolutional neural networks
And leading the trained model into a grounding line selection device, and playing back a test waveform for testing, wherein the more the training data set, the higher the accuracy. The deep convolutional neural network is adopted in the test, model parameters are continuously optimized, and test data sets are stored and continuously added to come from a learning optimization model.
10000 samples are divided into 100 times, 100 samples are trained each time, and after 60 times of training is carried out on data, the accuracy rate of line selection test using test waveforms is 81.94%. After 100 training of the data, the line selection test accuracy using the test waveform was 95.83%.
It can be seen that, when the training data is larger, the accuracy of the result is higher, and the robustness of the convolutional neural network model is higher. The method only uses 100 waveforms obtained by field and true tests for training, and can ensure that the accuracy rate of line selection exceeds 95 percent. In practical application, more waveforms are adopted for training, so that higher accuracy can be obtained.
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modifications made on the basis of the technical scheme according to the technical idea of the present invention fall within the protection scope of the present invention.
Claims (5)
1. A grounding line selection device based on a deep learning algorithm is characterized by comprising:
the sampling module is used for collecting and sending bus zero-sequence voltage and zero-sequence current of each branch circuit to the starting module and the line selection module;
the starting module is used for judging whether a ground fault occurs or not; and the number of the first and second groups,
and the line selection module is used for identifying the bus zero sequence voltage and the zero sequence current of each branch circuit by the deep convolution neural network when the starting module judges that the ground fault occurs, and selecting a fault line.
2. A grounding line selection method based on a deep learning algorithm is characterized by comprising the following steps:
step 1, acquiring a historical single-phase grounding waveform image, intercepting a section of transient waveform of a grounding instantaneous fault, displaying a zero sequence voltage waveform and a zero sequence current waveform in the same image, and further compressing and graying the image, wherein each branch circuit has an image;
step 2, training image data samples, wherein the label of each sample is '0' for representing a non-fault line, and '1' for representing a fault line, marking data and label files, storing the data and the label files as TFRecord files as input data, reading the files to a convolution layer through an input layer, obtaining a basic model structure after convolution pooling and regression calculation, optimizing model parameters to an optimal value through a back propagation algorithm, and making a classification result;
step 3, leading the trained model into a grounding line selection device;
step 4, the grounding line selection device judges whether a grounding fault occurs according to the zero sequence voltage;
and 5, after the ground fault occurs, the ground line selection device processes the waveform image by using the method in the step 1, the processed waveform image is sent to a trained deep convolution neural network for identification, the image with the result of 0 is a non-fault line, and the image with the result of 1 represents a fault line.
3. The grounding line selection method based on the deep learning algorithm as claimed in claim 2, characterized in that: in the step 1, the zero sequence voltage and the zero sequence current waveforms are displayed in an overlapped mode in the same image.
4. The grounding line selection method based on the deep learning algorithm as claimed in claim 2, characterized in that: in the step 2, a deep convolutional neural network is used for training and recognition.
5. The grounding line selection method based on the deep learning algorithm as claimed in claim 2, characterized in that: in the step 2, the neural network is trained in the cloud server or the PC.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910643661.9A CN112240965A (en) | 2019-07-17 | 2019-07-17 | Grounding line selection device and method based on deep learning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910643661.9A CN112240965A (en) | 2019-07-17 | 2019-07-17 | Grounding line selection device and method based on deep learning algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112240965A true CN112240965A (en) | 2021-01-19 |
Family
ID=74167142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910643661.9A Pending CN112240965A (en) | 2019-07-17 | 2019-07-17 | Grounding line selection device and method based on deep learning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112240965A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113377041A (en) * | 2021-05-26 | 2021-09-10 | 国网江西省电力有限公司供电服务管理中心 | Complex current waveform analog playback circuit and dynamic control method |
| CN113406439A (en) * | 2021-06-23 | 2021-09-17 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Power distribution network fault positioning method |
| CN114252739A (en) * | 2021-12-24 | 2022-03-29 | 国家电网有限公司 | Power distribution network single-phase earth fault distinguishing method, system, equipment and storage medium |
| CN115311493A (en) * | 2022-08-04 | 2022-11-08 | 国网江苏省电力有限公司电力科学研究院 | Method, system, memory and equipment for judging direct current circuit state |
| CN116008731A (en) * | 2023-02-15 | 2023-04-25 | 重庆大学 | Power distribution network high-resistance fault identification method and device and electronic equipment |
| CN117436025A (en) * | 2023-12-21 | 2024-01-23 | 青岛鼎信通讯股份有限公司 | Fault indicator-based non-fault abnormal waveform screening method |
| CN117706277A (en) * | 2024-02-02 | 2024-03-15 | 昆明理工大学 | Power distribution network fault line selection method based on graphic analysis and identification |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103163430A (en) * | 2013-03-29 | 2013-06-19 | 昆明理工大学 | Resonant grounding system fault line selection method by combining complex wavelets with ANN (artificial neural network) |
| CN104678260A (en) * | 2015-03-25 | 2015-06-03 | 南京南瑞继保电气有限公司 | System and method for single-phase grounding line selection of small current grounding system |
| CN106324432A (en) * | 2016-08-30 | 2017-01-11 | 南京南瑞继保电气有限公司 | Single-phase grounding line selection method for small-current grounding system |
| CN108154072A (en) * | 2016-12-02 | 2018-06-12 | 天津工业大学 | Insulator breakdown of taking photo by plane based on depth convolutional neural networks detects automatically |
| CN108279364A (en) * | 2018-01-30 | 2018-07-13 | 福州大学 | Wire selection method for power distribution network single phase earthing failure based on convolutional neural networks |
| CN108693439A (en) * | 2017-04-07 | 2018-10-23 | 南京南瑞继保电气有限公司 | A kind of small current neutral grounding system multi-line is the same as phase single-phase grounding selecting method |
| CN108732465A (en) * | 2018-05-30 | 2018-11-02 | 广东电网有限责任公司 | A kind of electrical power distribution network fault location method based on wavelet transformation and CNN |
| CN109324266A (en) * | 2018-11-21 | 2019-02-12 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of distribution single-phase-to-earth fault analysis method based on deep learning |
| CN110018389A (en) * | 2019-02-21 | 2019-07-16 | 国网山东省电力公司临沂供电公司 | A kind of transmission line of electricity on-line fault monitoring method and system |
-
2019
- 2019-07-17 CN CN201910643661.9A patent/CN112240965A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103163430A (en) * | 2013-03-29 | 2013-06-19 | 昆明理工大学 | Resonant grounding system fault line selection method by combining complex wavelets with ANN (artificial neural network) |
| CN104678260A (en) * | 2015-03-25 | 2015-06-03 | 南京南瑞继保电气有限公司 | System and method for single-phase grounding line selection of small current grounding system |
| CN106324432A (en) * | 2016-08-30 | 2017-01-11 | 南京南瑞继保电气有限公司 | Single-phase grounding line selection method for small-current grounding system |
| CN108154072A (en) * | 2016-12-02 | 2018-06-12 | 天津工业大学 | Insulator breakdown of taking photo by plane based on depth convolutional neural networks detects automatically |
| CN108693439A (en) * | 2017-04-07 | 2018-10-23 | 南京南瑞继保电气有限公司 | A kind of small current neutral grounding system multi-line is the same as phase single-phase grounding selecting method |
| CN108279364A (en) * | 2018-01-30 | 2018-07-13 | 福州大学 | Wire selection method for power distribution network single phase earthing failure based on convolutional neural networks |
| CN108732465A (en) * | 2018-05-30 | 2018-11-02 | 广东电网有限责任公司 | A kind of electrical power distribution network fault location method based on wavelet transformation and CNN |
| CN109324266A (en) * | 2018-11-21 | 2019-02-12 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of distribution single-phase-to-earth fault analysis method based on deep learning |
| CN110018389A (en) * | 2019-02-21 | 2019-07-16 | 国网山东省电力公司临沂供电公司 | A kind of transmission line of electricity on-line fault monitoring method and system |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113377041A (en) * | 2021-05-26 | 2021-09-10 | 国网江西省电力有限公司供电服务管理中心 | Complex current waveform analog playback circuit and dynamic control method |
| CN113406439A (en) * | 2021-06-23 | 2021-09-17 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Power distribution network fault positioning method |
| CN114252739A (en) * | 2021-12-24 | 2022-03-29 | 国家电网有限公司 | Power distribution network single-phase earth fault distinguishing method, system, equipment and storage medium |
| CN114252739B (en) * | 2021-12-24 | 2023-11-03 | 国家电网有限公司 | Power distribution network single-phase earth fault discrimination method, system, equipment and storage medium |
| CN115311493A (en) * | 2022-08-04 | 2022-11-08 | 国网江苏省电力有限公司电力科学研究院 | Method, system, memory and equipment for judging direct current circuit state |
| CN115311493B (en) * | 2022-08-04 | 2023-06-20 | 国网江苏省电力有限公司电力科学研究院 | Method, system, memory and equipment for judging direct current circuit state |
| CN116008731A (en) * | 2023-02-15 | 2023-04-25 | 重庆大学 | Power distribution network high-resistance fault identification method and device and electronic equipment |
| CN116008731B (en) * | 2023-02-15 | 2023-08-25 | 重庆大学 | Power distribution network high-resistance fault identification method and device and electronic equipment |
| CN117436025A (en) * | 2023-12-21 | 2024-01-23 | 青岛鼎信通讯股份有限公司 | Fault indicator-based non-fault abnormal waveform screening method |
| CN117706277A (en) * | 2024-02-02 | 2024-03-15 | 昆明理工大学 | Power distribution network fault line selection method based on graphic analysis and identification |
| CN117706277B (en) * | 2024-02-02 | 2024-06-07 | 昆明理工大学 | Power distribution network fault line selection method based on graphic analysis and identification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112240965A (en) | Grounding line selection device and method based on deep learning algorithm | |
| CN108279364B (en) | Power distribution network single-phase earth fault line selection method based on convolutional neural network | |
| Du et al. | Single line-to-ground faulted line detection of distribution systems with resonant grounding based on feature fusion framework | |
| CN109116203A (en) | Power equipment partial discharges fault diagnostic method based on convolutional neural networks | |
| CN109325526A (en) | A kind of distribution network failure classification method using convolution depth confidence network | |
| CN108732465A (en) | A kind of electrical power distribution network fault location method based on wavelet transformation and CNN | |
| CN110726898B (en) | Power distribution network fault type identification method | |
| CN102279358A (en) | MCSKPCA based neural network fault diagnosis method for analog circuits | |
| CN110852023B (en) | Automatic generation method and device for primary main wiring diagram of intelligent substation | |
| CN106203382A (en) | A kind of excitation surge current based on kernel function extreme learning machine and fault current recognition methods | |
| CN103677815B (en) | A kind of spacecraft low-frequency cable electrical property autotest program generates method | |
| CN114414942A (en) | Power transmission line fault identification classifier, identification method and system based on transient waveform image identification | |
| CN106443380B (en) | A kind of distribution cable local discharge signal recognition methods and device | |
| CN110346692A (en) | A kind of wire selection method for power distribution network single phase earthing failure based on time-frequency image information | |
| CN108267673B (en) | Distribution network fault line selection big data dimension reduction method and device | |
| CN110020637A (en) | A kind of analog circuit intermittent fault diagnostic method based on more granularities cascade forest | |
| CN114239703A (en) | Active power distribution system fault diagnosis method, system, equipment and storage medium | |
| CN112711865B (en) | Power transmission line inductive electric simulation refined modeling method | |
| CN112946425A (en) | Fault positioning method for mining travelling wave time-frequency domain characteristics by utilizing deep learning | |
| Wang et al. | Transmission line fault diagnosis based on wavelet packet analysis and convolutional neural network | |
| CN103558484B (en) | A kind of substation cable is by thunderbolt interference detecting apparatus and method | |
| da Silva et al. | High impedance fault location-case study using wavelet transform and artificial neural networks | |
| CN115308538A (en) | Power distribution network fault positioning method, system, terminal and storage medium based on YOLOV5 | |
| CN114252725A (en) | HHT and ResNet 18-based single-phase earth fault type comprehensive identification method | |
| CN114878973A (en) | Multi-branch distribution line lightning stroke fault positioning method and device and storage medium |
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 |