CN116863233A - Intelligent identification method for high-resistance ground faults of power distribution network based on image classification - Google Patents
Intelligent identification method for high-resistance ground faults of power distribution network based on image classification Download PDFInfo
- Publication number
- CN116863233A CN116863233A CN202310859339.6A CN202310859339A CN116863233A CN 116863233 A CN116863233 A CN 116863233A CN 202310859339 A CN202310859339 A CN 202310859339A CN 116863233 A CN116863233 A CN 116863233A
- Authority
- CN
- China
- Prior art keywords
- fault
- hilbert
- resistance ground
- ground fault
- random
- 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
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000001228 spectrum Methods 0.000 claims abstract description 48
- 238000013145 classification model Methods 0.000 claims abstract description 18
- 230000011218 segmentation Effects 0.000 claims abstract description 14
- 230000009466 transformation Effects 0.000 claims abstract description 11
- 238000005516 engineering process Methods 0.000 claims abstract description 9
- 238000005070 sampling Methods 0.000 claims abstract description 9
- 230000001052 transient effect Effects 0.000 claims description 15
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 239000010426 asphalt Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/52—Testing for short-circuits, leakage current or ground faults
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Abstract
The application provides an intelligent recognition method for high-resistance ground faults of a power distribution network based on image classification. Firstly, a sliding window technology is used for recording zero sequence voltage waveforms in real time; then classifying each sampling point in the zero sequence voltage by using a one-dimensional semantic segmentation model, and determining the starting moment of the fault; secondly, extracting a signal envelope and a Hilbert marginal spectrum by using Hilbert-Huang transformation, converting the signal envelope and the Hilbert marginal spectrum into a two-dimensional gray image as input of an image classification model, and obtaining a classification result through the image classification model; when the output combined result is 'distortion and random', a high-resistance ground fault is detected, and otherwise, the next sliding window data is continuously evaluated. The application uses Hilbert-Huang transformation to extract signal envelope and Hilbert marginal spectrum, which can effectively represent randomness and distortion of high-resistance ground fault. The application uses semantic segmentation technology in high-resistance ground fault identification for the first time, realizes pixel-level classification, and has high identification precision.
Description
Technical Field
The application relates to the technical field of power distribution networks, in particular to an intelligent recognition method for high-resistance ground faults of a power distribution network based on image classification.
Background
The power distribution network is used as a terminal link of a power system, and the power supply reliability directly influences the electricity consumption of production and life. With the increasing size of the distribution network, the structure is more and more complex, and the frequency of single-phase high-resistance ground faults of the distribution line is more and more increased. The high-resistance ground fault is a unique type of single-phase ground fault, and when an overhead line is in contact with high-resistance media such as branches, gravel, concrete, asphalt pavement, etc., an arc is easily generated, resulting in a high-resistance ground fault. The fault resistance range of the high-resistance ground fault can reach hundreds or even thousands of ohms, so that fault current is weak and nonlinear and random changes, and the traditional zero-sequence overcurrent protection device is difficult to effectively identify. However, if the faults are not recognized in time, the fault lines are cut off, the fault current exists for a long time, so that the temperature of the fault points is increased, the insulation of equipment is destroyed, the electrical equipment is damaged, and even more serious safety accidents, such as mountain forest fires, personal electric shock and the like, are caused.
Currently, the mainstream high-resistance ground fault identification methods can be divided into two types, namely a threshold-based method and an artificial intelligence-based method. The threshold-based approach has difficulty in setting an appropriate threshold to accommodate a wide range of fault conditions because the frequency domain characteristics of the nonlinear load and the high impedance load exhibit significant similarity. The method based on artificial intelligence does not need to set a threshold manually, but has high requirement on computing power. Furthermore, most of the previous studies only use short-time data and cannot characterize the main features of high-resistance ground faults over a long period of time, such as distortion and randomness.
Disclosure of Invention
In view of the above, the present application aims to provide an intelligent identification method for high-impedance ground faults of a power distribution network based on image classification, which is used for detecting zero-sequence voltages in real time, identifying transient processes of suspicious ground fault events, determining fault starting time, using hilbert-yellow transformation on longer-time zero-sequence voltage waveforms to obtain signal envelopes and hilbert marginal spectrums, and converting the signal envelopes and the hilbert marginal spectrums into two-dimensional gray images as inputs of an image classification model. And if the image classification model identification result is 'distortion and random', detecting that the high-resistance ground fault occurs.
In order to achieve the above purpose, the application adopts the following technical scheme: an intelligent recognition method for high-resistance ground faults of a power distribution network based on image classification comprises the following specific steps:
step one: collecting and recording zero sequence voltage waveform data in real time; the sliding window technology is used for recording zero sequence voltage waveforms in real time, and the length and the moving step length of the sliding window are respectively M power frequency periods and N power frequency periods;
step two: triggering suspected ground faults; processing the real-time data of each sliding window by using a one-dimensional semantic segmentation model to identify the transient process of a suspicious ground fault event and determine the fault starting time;
step three: identifying the type of the ground fault; obtaining fault zero sequence voltage data for a longer time, processing a zero sequence voltage waveform by using Hilbert-Huang transform to obtain a signal envelope and a Hilbert marginal spectrum, and then converting the signal envelope and the Hilbert marginal spectrum into a two-dimensional gray image which is used as an input of an image classification model; the output of the image classification model has four combined outputs, including "distorted and random", "distorted and non-random", "non-distorted and non-random";
step four: high-resistance ground fault identification; if the image classification model identification result is 'distortion and random', detecting that a high-resistance ground fault occurs; if the identification result is 'distortion and non-random' or 'non-distortion and random', identifying as suspicious high-resistance ground fault; if the identification result is 'non-distortion and non-random', identifying the non-fault event; and when the identification result is a suspicious high-resistance ground fault or a non-fault event, evaluating the next sliding window data.
In a preferred embodiment: step two, processing real-time data of each sliding window by a one-dimensional semantic segmentation model to identify transient processes of suspicious ground fault events and determine fault starting time;
the one-dimensional semantic segmentation model can realize pixel-level classification, and classifies each sampling point of the zero-sequence voltage waveform into two types of TP and N/A; the "TP" class is the transient course of a suspected ground fault event; the "N/A" class is the transient course of a non-suspected ground fault event; so that the starting time of the suspected ground fault event can be determined.
In a preferred embodiment: and step three, the ground fault type identification comprises two processes of feature extraction and feature classification.
In a preferred embodiment: in the characteristic extraction process, hilbert-Huang transformation is adopted to carry out empirical mode decomposition on signals, and the upper envelope and the lower envelope of the signals are calculated to be signal envelopes in the empirical mode decomposition; the integral of the hilbert spectrum on the time axis is a hilbert marginal spectrum; the method is characterized in that the Hilbert-Huang transformation is utilized to extract the signal envelope and the Hilbert marginal spectrum of the zero sequence voltage, and the randomness and the distortion of the high-resistance ground fault are respectively represented.
In a preferred embodiment: the feature extraction process specifically comprises a signal envelope and a Hilbert marginal spectrum;
the signal envelope detects the amplitude variation of the zero sequence voltage signal caused by the high-resistance ground fault; the signal envelope is calculated by the following formula:
where x (t) and y (t) are the real and imaginary parts of the signal, m, respectively x (t) and m y (t) is a moving average of the real and imaginary parts;
the signal envelope conditions of the zero sequence voltages under different fault conditions are different; for single-phase earth faults, the signal envelope remains stable after the fault occurs; for high-resistance ground faults, the signal envelope fluctuates after the fault occurs; when the signal envelope is fluctuating, then the signal is interpreted as having randomness;
the hilbert marginal spectrum is used for measuring the marginal spectrum of the signal, and the calculation formula is as follows:
where T is the duration of the signal, x (T) is the signal, and f (T) is a frequency modulation function that varies between 0 and 1;
the main frequency distribution range of the non-high-resistance single-phase earth fault and the high-resistance earth fault and the edge spectrum peak value under the same frequency show obvious difference on the Hilbert marginal spectrum; the hilbert marginal spectrum is obtained by integrating the hilbert spectrum on a time axis, and the distortion is represented by the total energy distribution of frequency.
In a preferred embodiment: in the characteristic classification process, converting a signal envelope and a Hilbert marginal spectrum into a two-dimensional gray image; then adopting an image classification model to identify signal envelopes of different samples and image features of the Hilbert marginal spectrum;
the input of the image classification model is a two-dimensional gray scale image, which has four outputs including 'distortion', 'random', 'non-distortion', 'non-random', and the combined outputs are 'distortion and random', 'distortion and non-random', 'non-distortion and non-random'; and if the combined output is 'distorted and random', judging that the high-resistance ground fault sample is obtained.
Compared with the prior art, the application has the following beneficial effects:
1. the power distribution network high-resistance ground fault identification method provided by the application firstly uses a one-dimensional semantic segmentation technology to detect zero-sequence voltage in real time, identifies the transient process of a suspicious ground fault event, and determines the fault starting time. The one-dimensional semantic segmentation technology can realize pixel-level classification, classify each sampling point of the zero-sequence voltage waveform, and further improve the accuracy of determining the fault starting time.
2. According to the high-resistance ground fault identification method for the power distribution network, disclosed by the application, the characteristic quantity is extracted from the zero-sequence voltage waveform, the distortion and randomness of the ground fault zero-sequence voltage waveform are judged by utilizing long-time data, and compared with the characteristic quantity extracted from the zero-sequence current, the high-resistance ground fault identification method for the power distribution network has the advantage of being free from the influence of the distance between a fault point and a measuring point; the Hilbert-Huang transform is used for extracting a signal envelope and a Hilbert marginal spectrum, and the signal envelope and the Hilbert marginal spectrum are respectively used for representing randomness and distortion of the high-resistance ground fault, so that the accuracy rate of high-resistance ground fault identification is improved.
Drawings
Fig. 1 is a schematic diagram of a high-resistance ground fault identification method according to a preferred embodiment of the present application.
Fig. 2 is a graph of the signal envelope and hilbert margin for a typical high-resistance fault in accordance with a preferred embodiment of the present application.
Detailed Description
The application will be further described with reference to the accompanying drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the present application; as used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
The embodiment provides an intelligent identification method for high-resistance ground faults of a power distribution network based on image classification, which is shown in fig. 1-2 and comprises the following steps:
step one: and acquiring and recording the zero sequence voltage waveform data in real time. The sliding window technology is used for recording zero sequence voltage waveform data in real time, the length and the moving step length of the sliding window are respectively 18 power frequency periods and 1 power frequency period, the sampling frequency is 5kHz, and the corresponding sliding window length and the moving step length are respectively 1800 sampling points and 100 sampling points.
Step two: the suspected ground fault triggers. Processing real-time data of each sliding window by using a 1D-UNet one-dimensional semantic segmentation model to identify a transient process of a suspicious ground fault event and determine a fault starting time; the method specifically comprises the following steps:
and dividing each sampling point in the one-dimensional zero sequence voltage waveform into two types of transient processes of suspicious ground fault events or transient processes of non-suspicious ground fault events by using a 1D-UNet semantic segmentation model so as to identify the transient processes of the suspicious ground fault events and determine the fault starting time. The 1D-UNet is a U-Net network which performs semantic segmentation on one-dimensional data and can realize pixel-level classification.
Step three: and (5) identifying the type of the ground fault. The method comprises the steps of obtaining zero sequence voltage waveform data of 50 power frequency cycles after faults, processing one-dimensional zero sequence voltage by using Hilbert-yellow transformation to obtain a signal envelope and Hilbert marginal spectrum which respectively represent randomness and distortion of high-resistance ground faults, and then converting the signal envelope and the Hilbert marginal spectrum into two-dimensional gray images serving as input of a GoogLeNet image classification model. The google net image classification model output has four combinations, including "distorted and random", "distorted and non-random", "non-distorted and non-random"; the method specifically comprises the following steps:
in a power distribution network, a high-resistance ground fault is a special single-phase ground fault. There is a significant difference between a high resistance ground fault, which is accompanied by intermittent arcing, and a single phase ground fault with a fixed high fault resistance, which is typically grounded through a fixed resistance, where arcing does not occur. Single-phase ground faults with a fixed low fault resistance of 200 ohms and a fixed high fault resistance of 3000 ohms were simulated in the distribution network, while high-resistance ground faults through gravel were simulated, with the signal envelopes and hilbert marginal spectra at the three faults shown in fig. 2.
3.1 randomness of high resistance ground faults
And performing empirical mode decomposition on the signals by using Hilbert-Huang transform, wherein the upper envelope and the lower envelope of the signals are calculated as signal envelopes in the empirical mode decomposition. The signal envelope of the zero sequence voltage is extracted by using Hilbert-Huang transformation to represent the randomness of the high-resistance grounding fault.
The signal envelope is a time domain analysis method for measuring the amplitude of the signal envelope over time. It can detect the signal amplitude variation caused by the high-resistance ground fault. The signal envelope is calculated by the following formula:
where x (t) and y (t) are the real and imaginary parts of the signal, m, respectively x (t) and m y And (t) is a moving average of the real and imaginary parts.
As shown in fig. 2, after a fault occurs, the signal envelope of a single-phase earth fault through a fixed fault resistance is regularly stabilized; the signal envelope of a ground fault through gravel is fluctuating due to non-linear and time-varying impedance. Thus, it is stated that high resistance ground faults are different from single phase ground faults through a fixed fault resistance, depending on the randomness caused by intermittent arcing. Thus, it is desirable to use the signal envelope to characterize the randomness of high resistance ground faults and distinguish them from other events.
3.2 distortion of high resistance ground faults
The integral of the hilbert spectrum over the time axis is the hilbert marginal spectrum. The Hilbert marginal spectrum of the zero sequence voltage is extracted by utilizing Hilbert-Huang transformation to represent the distortion of the high-resistance ground fault.
The hilbert marginal spectrum is a frequency domain analysis method, and is used for measuring the marginal spectrum of a signal, and the calculation formula is as follows:
where T is the duration of the signal, x (T) is the signal, and f (T) is the frequency modulation function varying between 0 and 1.
As shown in fig. 2, since the fundamental frequency component of 50Hz is the highest in proportion, the fundamental frequency component may obscure information of other frequency components in the frequency range, and thus only the hilbert marginal spectrum of 100Hz to 500Hz is considered. The main frequency distribution range and edge spectral peaks at the same frequency of a fixed-resistance grounded single-phase ground fault and a high-resistance ground fault show a significant difference between the hilbert marginal spectra. The hilbert marginal spectrum is obtained by integrating the hilbert spectrum on a time axis, and the distortion of the high-resistance ground fault is represented by the total energy distribution of the frequency.
Step four: high resistance ground fault identification. If the identification result of the GoogLeNet image classification model is 'distortion and random', identifying the GoogLeNet image classification model as a high-resistance ground fault; if the identification result is 'distortion and non-random' or 'non-distortion and random', identifying as suspicious high-resistance ground fault; if the identification result is 'non-distortion and non-random', the non-fault event is identified. And when the identification result is a suspicious high-resistance ground fault or a non-fault event, evaluating the next sliding window data.
The application provides an intelligent recognition method for high-resistance ground faults of a power distribution network based on image classification. The application firstly proposes to identify the transient process of the suspicious ground fault event and determine the fault starting time by using a one-dimensional semantic segmentation technology, and the technology can realize pixel-level classification, realize classification of each sampling point and further improve the accuracy of fault time determination. The application extracts the characteristic quantity from the zero sequence voltage, and has the advantage of being not influenced by the distance between the fault point and the measuring point compared with the characteristic quantity extracted from the zero sequence current. The application judges the distortion and randomness of the faults by using long-time data, extracts the signal envelope and the Hilbert marginal spectrum by using Hilbert-Huang transformation so as to represent the randomness and the distortion of the high-resistance ground faults, and further improves the accuracy of high-resistance ground fault identification.
Claims (6)
1. An intelligent recognition method for high-resistance ground faults of a power distribution network based on image classification is characterized by comprising the following steps of: the method comprises the following specific steps:
step one: collecting and recording zero sequence voltage waveform data in real time; the sliding window technology is used for recording zero sequence voltage waveforms in real time, and the length and the moving step length of the sliding window are respectively M power frequency periods and N power frequency periods;
step two: triggering suspected ground faults; processing the real-time data of each sliding window by using a one-dimensional semantic segmentation model to identify the transient process of a suspicious ground fault event and determine the fault starting time;
step three: identifying the type of the ground fault; obtaining fault zero sequence voltage data for a longer time, processing a zero sequence voltage waveform by using Hilbert-Huang transform to obtain a signal envelope and a Hilbert marginal spectrum, and then converting the signal envelope and the Hilbert marginal spectrum into a two-dimensional gray image which is used as an input of an image classification model; the output of the image classification model has four combined outputs, including "distorted and random", "distorted and non-random", "non-distorted and non-random";
step four: high-resistance ground fault identification; if the image classification model identification result is 'distortion and random', detecting that a high-resistance ground fault occurs; if the identification result is 'distortion and non-random' or 'non-distortion and random', identifying as suspicious high-resistance ground fault; if the identification result is 'non-distortion and non-random', identifying the non-fault event; and when the identification result is a suspicious high-resistance ground fault or a non-fault event, evaluating the next sliding window data.
2. The intelligent identification method for the high-resistance ground faults of the power distribution network based on image classification as claimed in claim 1, wherein the intelligent identification method is characterized by comprising the following steps of: step two, processing real-time data of each sliding window by a one-dimensional semantic segmentation model to identify transient processes of suspicious ground fault events and determine fault starting time;
the one-dimensional semantic segmentation model can realize pixel-level classification, and classifies each sampling point of the zero-sequence voltage waveform into two types of TP and N/A; the "TP" class is the transient course of a suspected ground fault event; the "N/A" class is the transient course of a non-suspected ground fault event; so that the starting time of the suspected ground fault event can be determined.
3. The intelligent identification method for the high-resistance ground faults of the power distribution network based on image classification as claimed in claim 1, wherein the intelligent identification method is characterized by comprising the following steps of: and step three, the ground fault type identification comprises two processes of feature extraction and feature classification.
4. The intelligent identification method for high-resistance ground faults of the power distribution network based on image classification as claimed in claim 3, wherein the intelligent identification method is characterized by comprising the following steps of: in the characteristic extraction process, hilbert-Huang transformation is adopted to carry out empirical mode decomposition on signals, and the upper envelope and the lower envelope of the signals are calculated to be signal envelopes in the empirical mode decomposition; the integral of the hilbert spectrum on the time axis is a hilbert marginal spectrum; the method is characterized in that the Hilbert-Huang transformation is utilized to extract the signal envelope and the Hilbert marginal spectrum of the zero sequence voltage, and the randomness and the distortion of the high-resistance ground fault are respectively represented.
5. The intelligent identification method for the high-resistance ground faults of the power distribution network based on image classification as claimed in claim 4, wherein the intelligent identification method is characterized by comprising the following steps of: the feature extraction process specifically comprises a signal envelope and a Hilbert marginal spectrum;
the signal envelope detects the amplitude variation of the zero sequence voltage signal caused by the high-resistance ground fault; the signal envelope is calculated by the following formula:
where x (t) and y (t) are the real and imaginary parts of the signal, m, respectively x (t) and m y (t) is a moving average of the real and imaginary parts;
the signal envelope conditions of the zero sequence voltages under different fault conditions are different; for single-phase earth faults, the signal envelope remains stable after the fault occurs; for high-resistance ground faults, the signal envelope fluctuates after the fault occurs; when the signal envelope is fluctuating, then the signal is interpreted as having randomness;
the hilbert marginal spectrum is used for measuring the marginal spectrum of the signal, and the calculation formula is as follows:
where T is the duration of the signal, x (T) is the signal, and f (T) is a frequency modulation function that varies between 0 and 1;
the main frequency distribution range of the non-high-resistance single-phase earth fault and the high-resistance earth fault and the edge spectrum peak value under the same frequency show obvious difference on the Hilbert marginal spectrum; the hilbert marginal spectrum is obtained by integrating the hilbert spectrum on a time axis, and the distortion is represented by the total energy distribution of frequency.
6. The intelligent identification method for high-resistance ground faults of the power distribution network based on image classification as claimed in claim 3, wherein the intelligent identification method is characterized by comprising the following steps of: in the characteristic classification process, converting a signal envelope and a Hilbert marginal spectrum into a two-dimensional gray image; then adopting an image classification model to identify signal envelopes of different samples and image features of the Hilbert marginal spectrum;
the input of the image classification model is a two-dimensional gray scale image, which has four outputs including 'distortion', 'random', 'non-distortion', 'non-random', and the combined outputs are 'distortion and random', 'distortion and non-random', 'non-distortion and non-random'; and if the combined output is 'distorted and random', judging that the high-resistance ground fault sample is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310859339.6A CN116863233A (en) | 2023-07-13 | 2023-07-13 | Intelligent identification method for high-resistance ground faults of power distribution network based on image classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310859339.6A CN116863233A (en) | 2023-07-13 | 2023-07-13 | Intelligent identification method for high-resistance ground faults of power distribution network based on image classification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116863233A true CN116863233A (en) | 2023-10-10 |
Family
ID=88223097
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310859339.6A Pending CN116863233A (en) | 2023-07-13 | 2023-07-13 | Intelligent identification method for high-resistance ground faults of power distribution network based on image classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116863233A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
-
2023
- 2023-07-13 CN CN202310859339.6A patent/CN116863233A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pinpart et al. | Differentiating between partial discharge sources using envelope comparison of ultra-high-frequency signals | |
| Li et al. | A generic waveform abnormality detection method for utility equipment condition monitoring | |
| CN109298287B (en) | Fault indicator data-based method for identifying fault type of low-current grounding distribution network | |
| CN109633368A (en) | The method of duration power quality disturbances containing distributed power distribution network based on VMD and DFA | |
| CN116863233A (en) | Intelligent identification method for high-resistance ground faults of power distribution network based on image classification | |
| CN110007197B (en) | Series arc fault detection system and detection method | |
| Mallick et al. | Performance characteristics of the ENTLN evaluated using rocket-triggered lightning data | |
| Cheng et al. | Series arc fault detection and implementation based on the short-time fourier transform | |
| Cheng et al. | Application of Gabor–Wigner transform to inspect high-impedance fault-generated signals | |
| CN112730964A (en) | Lightning overvoltage identification method based on overvoltage waveform characteristics | |
| CN105403816A (en) | Identification method of DC fault electric arc of photovoltaic system | |
| Gomes et al. | The effectiveness of different sampling rates in vegetation high-impedance fault classification | |
| CN112505484B (en) | Medium-voltage distribution cable latent fault identification method for small-resistance grounding system | |
| CN109581166B (en) | Power transmission line fault positioning method based on Sobel energy spectrum | |
| CN105510760A (en) | Method for detecting short circuit fault data based on wavelet analysis | |
| CN105893976B (en) | A kind of parametrization recognition methods of travelling wave signal | |
| Wooi et al. | Statistical Analysis on Preliminary Breakdown Pulses of Positive Cloud-to-Ground Lightning in Malaysia. | |
| Chang et al. | A hybrid approach for detection and classification of power quality disturbances | |
| CN113376474A (en) | Neural network fault arc identification system and method based on generalized S transformation | |
| CN104865508A (en) | Partial discharge recognition method based on data grouping quantification | |
| Hong et al. | Short-time fourier transform based analysis to characterization of series arc fault | |
| CN115453260A (en) | Power distribution network fault detection method based on mathematical morphology | |
| CN107578016A (en) | A kind of residual current waveform automatic identifying method based on rarefaction representation | |
| CN112666436B (en) | Electric arc crosstalk signal identification method based on vector analysis and calculation | |
| CN106771893B (en) | A kind of ground insulator gap discharge method for waveform identification |
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 |