CN113702739A - Electric arc detection method and device based on wavelet decomposition and neural network - Google Patents
Electric arc detection method and device based on wavelet decomposition and neural network Download PDFInfo
- Publication number
- CN113702739A CN113702739A CN202110993221.3A CN202110993221A CN113702739A CN 113702739 A CN113702739 A CN 113702739A CN 202110993221 A CN202110993221 A CN 202110993221A CN 113702739 A CN113702739 A CN 113702739A
- Authority
- CN
- China
- Prior art keywords
- current
- sequence
- wavelet
- arc
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 64
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 47
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 40
- 238000010891 electric arc Methods 0.000 title claims description 5
- 238000005070 sampling Methods 0.000 claims abstract description 47
- 230000000737 periodic effect Effects 0.000 claims abstract description 22
- 238000001914 filtration Methods 0.000 claims description 8
- 238000000034 method Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 description 33
- 238000010586 diagram Methods 0.000 description 5
- 230000009466 transformation Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000003763 carbonization Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000001228 spectrum Methods 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R29/00—Arrangements for measuring or indicating electric quantities not covered by groups G01R19/00 - G01R27/00
- G01R29/02—Measuring characteristics of individual pulses, e.g. deviation from pulse flatness, rise time or duration
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Relating To Insulation (AREA)
Abstract
本发明公开了一种基于小波分解和神经网络的电弧检测方法及装置，该方法包括：根据背景电流波形序列与当前周期电流波形采样序列作差，获取当前周期电流差分波形序列，并对所述当前周期电流差分波形序列进行离散小波变换处理，获取故障电弧检测的特征量；将所述故障电弧检测的特征量输入训练好的误差反向传播神经网络进行二分类，输出线路产生故障电弧的概率，若所述输出线路产生故障电弧的概率大于阈值，则判定线路产生故障电弧。本发明基于小波分解和神经网络的检测方法，避免了由于背景噪声带来的干扰，同时降低了算法的执行时间并提高了故障电弧检测的准确度。
The invention discloses an arc detection method and device based on wavelet decomposition and neural network. The method includes: obtaining a current periodic current differential waveform sequence according to a difference between a background current waveform sequence and a current periodic current waveform sampling sequence, and analyzing the current periodic current differential waveform sequence. The current periodic current differential waveform sequence is subjected to discrete wavelet transform processing to obtain the characteristic amount of arc fault detection; the characteristic amount of arc fault detection is input into the trained error back propagation neural network for binary classification, and the probability of arc fault generation in the output line is output. , if the probability that the output line produces an arc fault is greater than the threshold, it is determined that the line produces an arc fault. The invention is based on the detection method of wavelet decomposition and neural network, which avoids the interference caused by background noise, reduces the execution time of the algorithm and improves the accuracy of fault arc detection.
Description
Technical Field
The invention relates to the technical field of fault arc detection of low-voltage power distribution networks, in particular to an arc detection method and device based on wavelet decomposition and a neural network.
Background
The essence of the arc is a gas discharge phenomenon with concentrated energy, high temperature and high brightness, and when the switching device opens the circuit, if the voltage level between the open contacts reaches a certain value, intense white light is generated between the contacts, which is called the arc. Electric arcs in low-voltage distribution network lines are divided into good arcs and bad arcs. Good arc refers to arc caused by normal breaking, plugging, starting and the like of the load, and the arc cannot damage the load. The broken arc is an arc caused by loose connection, insulation carbonization, damage and the like, and is also called a fault arc. Fault arcs are one of the main causes of electrical fires. At present, a detection method for detecting a fault arc of a low-voltage distribution network aiming at characteristics of current and voltage of a power transmission line when the arc occurs is as follows:
the prior patent (CN101673930A) directly performs discrete wavelet transform on the original current waveform of the line to calculate wavelet coefficients. Background noise exists in the original current waveform of the line, when some nonlinear loads exist on the line and the normally running waveform of the nonlinear loads has the characteristics of fault arc waveforms such as 'zero-break region' and 'high-frequency noise', the decision maker is easy to make misjudgment by directly carrying out wavelet analysis on the original waveform of the line, and the normally running waveform of the loads is identified as the waveform when the fault arc occurs.
The existing patent (CN108562835A) extracts four main features of current shoulder leveling time, current jump value ratio, the number of times of current jump in 1s and standard deviation of current average in 1s as the input of the artificial neural network. The method does not fully mine information contained in current waveform data, has poor generalization, has large difference of current waveform characteristics when a line generates fault arcs in different application scenes, and can select the characteristics most suitable for the scenes through a large amount of experimental analysis to be used as the input of the artificial neural network.
Disclosure of Invention
The invention aims to provide an arc detection method based on wavelet decomposition and a neural network, and aims to solve the problem of low detection efficiency of fault arcs of a low-voltage distribution network in the prior art.
In order to achieve the above object, the present invention provides an arc detection method based on wavelet decomposition and neural network, comprising:
obtaining a current period current differential waveform sequence according to the difference between the background current waveform sequence and the current period current waveform sampling sequence, and performing discrete wavelet transform processing on the current period current differential waveform sequence to obtain the characteristic quantity of fault arc detection;
inputting the characteristic quantity detected by the fault arc into a trained error back propagation neural network for secondary classification, outputting the probability of the fault arc generated by a line, and if the probability of the fault arc generated by the output line is greater than a threshold value, judging that the line generates the fault arc.
Preferably, the discrete wavelet transform processing on the current period current difference waveform sequence to obtain the characteristic quantity of fault arc detection includes:
acquiring an initial scale function and an initial wavelet function according to the characteristics of a current waveform when a line generates a fault arc;
respectively constructing a high-pass filter and a low-pass filter of the initial scale function and the initial wavelet function;
and filtering the input signal by adopting the high-pass filter and the low-pass filter to obtain the characteristic quantity of the fault arc detection on the required frequency band, wherein the characteristic quantity of the fault arc detection comprises a scale coefficient and a wavelet coefficient of wavelet decomposition.
Preferably, before the difference is made from the background current waveform sequence and the current period current waveform sampling sequence, the method further comprises:
sampling a current waveform of a line in normal and steady-state operation to obtain a periodic current waveform sampling sequence in normal and steady-state operation;
and calculating the expectation of the periodic current waveform sampling sequence in normal steady-state operation to obtain the background current waveform sequence.
Preferably, before the difference is made from the background current waveform sequence and the current period current waveform sampling sequence, the method further comprises:
and sampling the current waveform of the line in the current period according to the sampling frequency of 200kHz to obtain a current waveform sampling sequence in the current period.
Preferably, the filtering the input signal by using the high-pass filter and the low-pass filter to obtain the characteristic quantity of the fault arc detection on the required frequency band includes:
and performing discrete convolution on the current period current difference waveform sequence and the low-pass filter and the high-pass filter respectively to obtain a scale coefficient and a wavelet coefficient of the wavelet decomposition respectively.
The invention also provides an arc detection device based on wavelet decomposition and a neural network, which comprises:
the acquisition module is used for carrying out difference according to the background current waveform sequence and the current period current waveform sampling sequence, acquiring a current period current differential waveform sequence, and carrying out discrete wavelet transform processing on the current period current differential waveform sequence to acquire the characteristic quantity of fault arc detection;
and the detection module is used for inputting the characteristic quantity detected by the fault arc into a trained error back propagation neural network for secondary classification, outputting the probability of generating the fault arc by the line, and if the probability of generating the fault arc by the output line is greater than a threshold value, judging that the line generates the fault arc.
Preferably, the obtaining module is further configured to obtain an initial scale function and an initial wavelet function according to characteristics of a current waveform when the line generates a fault arc;
respectively constructing a high-pass filter and a low-pass filter of the initial scale function and the initial wavelet function;
and filtering the input signal by adopting the high-pass filter and the low-pass filter to obtain the characteristic quantity of the fault arc detection on the required frequency band, wherein the characteristic quantity of the fault arc detection comprises a scale coefficient and a wavelet coefficient of wavelet decomposition.
Preferably, the method further comprises a first determining module, configured to:
sampling a current waveform of a line in normal and steady-state operation to obtain a periodic current waveform sampling sequence in normal and steady-state operation;
and calculating the expectation of the periodic current waveform sampling sequence in normal steady-state operation to obtain the background current waveform sequence.
Preferably, the first determining module is further configured to:
and sampling the current waveform of the line in the current period according to the sampling frequency of 200kHz to obtain a current waveform sampling sequence in the current period.
Preferably, the obtaining module is further configured to:
and performing discrete convolution on the current period current difference waveform sequence and the low-pass filter and the high-pass filter respectively to obtain a scale coefficient and a wavelet coefficient of the wavelet decomposition respectively.
Compared with the prior art, the invention has the beneficial effects that:
and comparing with the prior art that time domain waveforms are directly used as the input of the neural network, and the characteristic quantity of the fault arc detection is used as the input of the trained error back propagation neural network, so that the output result of the neural network is more accurate.
Further, discrete convolution is carried out on the current period current difference waveform sequence and a low-pass filter and a high-pass filter which are formed by a scale function and a wavelet function respectively, a scale coefficient and a wavelet coefficient of wavelet decomposition are obtained respectively, the influence of background noise on the waveform characteristics of the fault arc current is avoided, the execution time of an algorithm is reduced, and the accuracy of fault arc detection is improved.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of an arc detection method based on wavelet decomposition and a neural network according to an embodiment of the present invention;
FIG. 2 is a graph of a wavelet function provided by another embodiment of the present invention;
FIG. 3 is a graph of a scaling function provided by yet another embodiment of the present invention;
FIG. 4 is a waveform diagram of a low pass filter provided by another embodiment of the present invention;
FIG. 5 is a high pass filter waveform diagram provided by yet another embodiment of the present invention;
FIG. 6 is a diagram illustrating a wavelet decomposition structure provided by an embodiment of the present invention;
FIG. 7 is a schematic diagram of a neural network according to another embodiment of the present invention;
fig. 8 is a schematic structural diagram of an arc detection apparatus based on wavelet decomposition and a neural network according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be understood that the step numbers used herein are for convenience of description only and are not intended as limitations on the order in which the steps are performed.
It is to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The terms "comprises" and "comprising" indicate the presence of the described features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The term "and/or" refers to and includes any and all possible combinations of one or more of the associated listed items.
Referring to fig. 1, an embodiment of the present invention provides an arc detection method based on wavelet decomposition and a neural network, including the following steps:
s101: and obtaining a current period current differential waveform sequence according to the difference between the background current waveform sequence and the current period current waveform sampling sequence, and performing discrete wavelet transform processing on the current period current differential waveform sequence to obtain the characteristic quantity of fault arc detection.
Specifically, the current waveform during normal and steady operation of the line is sampled, a periodic current waveform sampling sequence during normal and steady operation is obtained, the expectation of the periodic current waveform sampling sequence during normal and steady operation is calculated, and the background current waveform sequence is obtained.
The characteristic frequency of the current waveform when the line is generating fault arc is usually distributed in the range of 1kHz-100 kHz. From the nyquist sampling theorem: sampling frequency fsMust be equal to or greater than the highest frequency component of interest of the signal under testNyquist frequency fN) Twice, the sampling frequency f is selectedsAt 200kHz, the frequency band of interest of the detected signal has no spectrum aliasing, and the number of sampling points of the waveform in each period is calculated according to the frequency f of the normal operation of the line
Sampling the current waveform of the line in normal steady state operation with a sampling period of NTTo obtain NTPeriodic current waveform sampling sequence for normal steady-state operation of individual line
Wherein the sequence of each cycle is
According to the calculation of NTMathematical expectation of periodic current waveform sampling sequence during normal steady-state operation of individual line
As a background current waveform sequence
Sampling the current waveform of the line in the current period according to the sampling frequency of 200kHz to obtain the current waveform sampling sequence in the current period
Sampling sequence of current waveform of current cycle
Waveform sequence with background current
Making difference to obtain current differential waveform sequence R of current periodc＝Cc-Y as input signal for wavelet decomposition, enabling filtering of background noise.
Referring to fig. 2 and 3, in particular, the coiflet1 wavelet is selected as the wavelet function Ψ (t) and the scaling function according to the characteristics of the current waveform when the line generates the fault arc
The following were used:
1) the fault arc duration of the line is usually very short, but can generate huge heat effect, the temperature can reach 3000-5000 ℃ during arc burning, and the standard regulation of UL1699 in the United states: when the arc fault circuit breaker detects 8 half-cycles of fault arcs in 500ms, the circuit breaker needs to execute tripping action to cut off a circuit, so that the arc detection algorithm has high real-time requirements, discrete wavelet transformation needs to be adopted, and a wavelet function with high vanishing moment is selected to enable the transformed signal energy to be more concentrated, reduce the complexity of the algorithm and improve the detection real-time property.
2) Because the electric arc is usually scattered in the sporadic flash waves among normal currents, the electric arc current waveform has a plurality of burrs similar to high-frequency pulses relative to the normal current waveform, a wavelet function with a short tight support set needs to be selected, the local time-frequency characteristic of the wavelet function is improved, the detection of a catastrophe point and a singular point is facilitated, and the wavelet function with good symmetry is selected as much as possible to avoid signal phase distortion after wavelet transformation.
3) In order to avoid signal phase distortion after wavelet transformation, a wavelet function with good symmetry is selected as much as possible.
Therefore, an initial scale function and an initial wavelet function are obtained according to the characteristics of a current waveform when a line generates a fault arc, a high-pass filter and a low-pass filter of the initial scale function and the initial wavelet function are respectively constructed, multiple filtering processing is carried out by adopting the high-pass filter and the low-pass filter, the characteristic quantity of fault arc detection on a required frequency band is obtained, the characteristic quantity of fault arc detection comprises a scale coefficient and a wavelet coefficient of wavelet decomposition, specifically, according to the analysis, the initial scale function and the wavelet function are selected, the high-pass filter and the low-pass filter are respectively constructed, multiple filtering is carried out on an original signal, and a coiflet1 wavelet on the required frequency band is determined as a small waveletWave function Ψ (t) and scale function
The coiflet1 wavelet has biarthogonality, compactness and approximate symmetry, filter length 6, support length 5, wavelet function vanishing moment order 2, and scale function vanishing moment order 1.
Referring to fig. 4, 5 and 6, the wavelet function Ψ (t) and the scaling function are selected according to the above
Constructing a low-pass filter and a high-pass filter, in particular defining a low-pass filter hnThe mathematical expression is
Defining a high-pass filter gnThe mathematical expression is
Using a low-pass filter hnAnd a high-pass filter gnFor input signal RcFiltering to obtain input signal RcAre respectively connected with a low-pass filter hnAnd a high-pass filter gnRespectively making discrete convolution
And
wherein c isj,n＝RcFiltered result cj-1,kAnd dj-1,kThe k scale coefficient and the k wavelet coefficient of the wavelet decomposition of the j-1 th layer, the j-1 th layer and the (j-n-1) th layer can be further calculated to form a one-dimensional vector s ═ cj-n,k,dj-n,k,...,dj-1,k). The iterative process is to pass the input signal R through a filtercIs gradually decomposed into different frequency bands, after the nth decomposition,
in order to approximate the frequency band range of the signal,
is the frequency band range of the detail signal. And (n-1) k wavelet coefficients d (j-n, k) and k scale coefficients c (j-1, k) obtained after wavelet decomposition are used as characteristic quantities for fault arc detection, and after normalization processing is carried out on the n x k coefficients, the n x k coefficients are directly used as n x k input parameters of the input layer of the neural network.
S102: inputting the characteristic quantity detected by the fault arc into a trained error back propagation neural network for secondary classification, outputting the probability of the fault arc generated by a line, and if the probability of the fault arc generated by the output line is greater than a threshold value, judging that the line generates the fault arc.
Referring to fig. 7, specifically, n × k input parameters calculated in the above steps are input into an input layer of a neural network, and n × k parameters are input into a trained error back propagation neural network, wherein the trained neural network is composed of an input layer, a hidden layer and an output layer, and a threshold p for controlling an arc fault circuit interrupting device (AFCI) to operate by a decision maker is determined according to a hazard level of a fault arc generated by the linetUnder the conventional strategy ptIf the damage degree of the fault arc generated by the line is not large and a conservative strategy is adopted, p is takent<0.5, so as to reduce the false action probability of AFCI, if the harmfulness of the fault arc generated by the circuit is larger, an aggressive strategy is adopted, and p is takent>And 0.5, reducing the probability of missed judgment of the fault arc event.
Inputting the characteristic quantity of fault arc detection into the trained error back propagation neural network for two classifications, outputting the probability of fault arc generation and the probability of normal operation of the circuit, and forming a 1 x 2 matrix [ p ] by the probability of fault arc generation and the probability of normal operation of the circuitf,pn]T，pfRepresenting the probability sum p of line fault arcingnRepresenting the probability of normal operation, if the output line is producing a fault arc, pfGreater than a threshold value ptAnd judging that the line generates fault arc.
The invention analyzes based on the differential waveform of the current waveform to be detected and the historical normal operation current waveform of the circuit, avoids the influence of background noise on the waveform characteristics of the fault arc current, improves the accuracy of fault arc detection, further balances the parameters of waveform symmetry, tight set length, filter length, vanishing moment order and the like, selects the coiflet1 wavelet as a wavelet base, ensures that the algorithm has low complexity and good real-time property, the phase distortion of the wavelet function is low, the high-frequency time-frequency characteristic is good, compared with the method of directly using the time domain waveform as the input of the neural network, the invention uses the wavelet function as the input of the neural network, the wavelet function simultaneously contains the time-frequency domain information of the waveform, improves the information entropy, ensures that the output result of the neural network is more accurate, the error back propagation neural network has the capacity of from the input to the output of complex nonlinear mapping, and is suitable for solving the problem of complex internal mechanism, and the network structure is simple, and the algorithm execution time is reduced.
Referring to fig. 8, another embodiment of the present invention provides an arc detection apparatus based on wavelet decomposition and neural network, including:
and the obtaining module 11 is configured to obtain a current period current differential waveform sequence by subtracting the background current waveform sequence from the current period current waveform sampling sequence, and perform discrete wavelet transform processing on the current period current differential waveform sequence to obtain a characteristic quantity of fault arc detection.
And the detection module 12 is configured to input the feature quantity detected by the fault arc into a trained error back propagation neural network to perform two classifications, output the probability that the line generates the fault arc, and determine that the line generates the fault arc if the probability that the output line generates the fault arc is greater than a threshold value.
For specific limitations of the arc detection apparatus based on wavelet decomposition and neural network, reference may be made to the above limitations of the arc detection method based on wavelet decomposition and neural network, and details are not repeated here. The modules in the above-mentioned arc detection device based on wavelet decomposition and neural network can be wholly or partially realized by software, hardware and combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110993221.3A CN113702739A (en) | 2021-08-26 | 2021-08-26 | Electric arc detection method and device based on wavelet decomposition and neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110993221.3A CN113702739A (en) | 2021-08-26 | 2021-08-26 | Electric arc detection method and device based on wavelet decomposition and neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113702739A true CN113702739A (en) | 2021-11-26 |
Family
ID=78655740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110993221.3A Pending CN113702739A (en) | 2021-08-26 | 2021-08-26 | Electric arc detection method and device based on wavelet decomposition and neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113702739A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114184915A (en) * | 2021-12-03 | 2022-03-15 | 国网宁夏电力有限公司电力科学研究院 | Insulation fault detection method for closing resistor of breaker for filter bank |
| CN114818798A (en) * | 2022-04-15 | 2022-07-29 | 浙江大学 | Fault arc feature extraction method and device based on self-convolution and electronic equipment |
| CN115856550A (en) * | 2022-12-19 | 2023-03-28 | 华南理工大学 | Composite insulator salt mist flashover prediction method and device and storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030038640A1 (en) * | 2001-08-24 | 2003-02-27 | Zuercher Joseph C. | Apparatus for detecting arcing and overcurrents in dc electrical systems subject to cyclic disturbances |
| US20070242399A1 (en) * | 2006-04-11 | 2007-10-18 | Crouzet Automatismes | Electric arc detection device, switchgear unit comprising one such device and method for detecting an electric arc |
| US20090171603A1 (en) * | 2007-12-28 | 2009-07-02 | Sriram Changali | Methods of detecting series arcs in electrical signals |
| CN102981088A (en) * | 2012-12-12 | 2013-03-20 | 天津市鸿远电气设备有限公司 | Detection method for fault arc |
| CN104569684A (en) * | 2015-01-14 | 2015-04-29 | 上海和伍新材料科技有限公司 | Fault electric arc detection method based on electric arc spectrum signals |
| CN104764963A (en) * | 2015-04-20 | 2015-07-08 | 河北工业大学 | Alternating-current arc fault detecting method |
| CN104811135A (en) * | 2015-04-16 | 2015-07-29 | 西安交通大学 | Photovoltaic system fault arc detection method under common-mode signal interference condition |
| CN105675966A (en) * | 2016-01-08 | 2016-06-15 | 安徽长远绿色能源有限公司 | Fault arc detection method based on difference calculation and protection device thereof |
| CN108646149A (en) * | 2018-04-28 | 2018-10-12 | 国网江苏省电力有限公司苏州供电分公司 | Fault electric arc recognition methods based on current characteristic extraction |
| CN110488161A (en) * | 2019-07-23 | 2019-11-22 | 南京航空航天大学 | A kind of detection of multi-load series arc faults and localization method |
| CN110618353A (en) * | 2019-10-22 | 2019-12-27 | 北京航空航天大学 | Direct current arc fault detection method based on wavelet transformation + CNN |
| WO2020015277A1 (en) * | 2018-07-20 | 2020-01-23 | 国电南瑞科技股份有限公司 | Arc light fault identifying device and method based on panoramic information |
| CN112305376A (en) * | 2020-10-28 | 2021-02-02 | 国网山东省电力公司青岛供电公司 | A low-voltage distribution line fault arc detection and identification method |
-
2021
- 2021-08-26 CN CN202110993221.3A patent/CN113702739A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030038640A1 (en) * | 2001-08-24 | 2003-02-27 | Zuercher Joseph C. | Apparatus for detecting arcing and overcurrents in dc electrical systems subject to cyclic disturbances |
| US20070242399A1 (en) * | 2006-04-11 | 2007-10-18 | Crouzet Automatismes | Electric arc detection device, switchgear unit comprising one such device and method for detecting an electric arc |
| US20090171603A1 (en) * | 2007-12-28 | 2009-07-02 | Sriram Changali | Methods of detecting series arcs in electrical signals |
| CN102981088A (en) * | 2012-12-12 | 2013-03-20 | 天津市鸿远电气设备有限公司 | Detection method for fault arc |
| CN104569684A (en) * | 2015-01-14 | 2015-04-29 | 上海和伍新材料科技有限公司 | Fault electric arc detection method based on electric arc spectrum signals |
| CN104811135A (en) * | 2015-04-16 | 2015-07-29 | 西安交通大学 | Photovoltaic system fault arc detection method under common-mode signal interference condition |
| CN104764963A (en) * | 2015-04-20 | 2015-07-08 | 河北工业大学 | Alternating-current arc fault detecting method |
| CN105675966A (en) * | 2016-01-08 | 2016-06-15 | 安徽长远绿色能源有限公司 | Fault arc detection method based on difference calculation and protection device thereof |
| CN108646149A (en) * | 2018-04-28 | 2018-10-12 | 国网江苏省电力有限公司苏州供电分公司 | Fault electric arc recognition methods based on current characteristic extraction |
| WO2020015277A1 (en) * | 2018-07-20 | 2020-01-23 | 国电南瑞科技股份有限公司 | Arc light fault identifying device and method based on panoramic information |
| CN110488161A (en) * | 2019-07-23 | 2019-11-22 | 南京航空航天大学 | A kind of detection of multi-load series arc faults and localization method |
| CN110618353A (en) * | 2019-10-22 | 2019-12-27 | 北京航空航天大学 | Direct current arc fault detection method based on wavelet transformation + CNN |
| CN112305376A (en) * | 2020-10-28 | 2021-02-02 | 国网山东省电力公司青岛供电公司 | A low-voltage distribution line fault arc detection and identification method |
Non-Patent Citations (4)
| Title |
|---|
| 蔡坤宝 等: "吸毒者脉象信号的小波与神经网络分析", 《重庆大学学报（自然科学版）》, vol. 30, no. 10, 31 October 2007 (2007-10-31), pages 50 - 54 * |
| 郭凤仪 等: "基于灰度-梯度共生矩阵的串联故障电弧特征", 《电工技术学报》, vol. 33, no. 1, pages 71 - 81 * |
| 马建峰 等: "模糊控制理论在铣刀磨损监控中的应用", 《北京理工大学学报》, vol. 21, no. 3, 30 June 2001 (2001-06-30), pages 296 - 299 * |
| 黄琼: "基于小波神经网络的配电网单相接地故障定位研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》, no. 3, 30 June 2001 (2001-06-30), pages 042 - 542 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114184915A (en) * | 2021-12-03 | 2022-03-15 | 国网宁夏电力有限公司电力科学研究院 | Insulation fault detection method for closing resistor of breaker for filter bank |
| CN114184915B (en) * | 2021-12-03 | 2024-07-30 | 国网宁夏电力有限公司电力科学研究院 | A method for detecting insulation fault of closing resistance of circuit breaker for filter bank |
| CN114818798A (en) * | 2022-04-15 | 2022-07-29 | 浙江大学 | Fault arc feature extraction method and device based on self-convolution and electronic equipment |
| CN115856550A (en) * | 2022-12-19 | 2023-03-28 | 华南理工大学 | Composite insulator salt mist flashover prediction method and device and storage medium |
| CN115856550B (en) * | 2022-12-19 | 2024-01-16 | 华南理工大学 | A composite insulator salt spray flashover prediction method, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113702739A (en) | Electric arc detection method and device based on wavelet decomposition and neural network | |
| CN110488161B (en) | A multi-load series arc fault detection and location method | |
| Yao et al. | Characteristic study and time-domain discrete-wavelet-transform based hybrid detection of series DC arc faults | |
| CN110059437B (en) | A method for extracting feature quantities of GIS vibration signals based on variational mode decomposition | |
| Zhang et al. | Series AC arc fault diagnosis based on data enhancement and adaptive asymmetric convolutional neural network | |
| CN105067966B (en) | The low-voltage alternating-current fault arc detection method of feature based modal components energy spectrometer | |
| CN108399147B (en) | A Transformer Inrush Current Identification Method Based on MEEMD Algorithm | |
| CN110677125B (en) | Method and device for arc fault detection | |
| CN113608023B (en) | Sampling frequency self-adaptive harmonic detection method and system | |
| CN113094983B (en) | An online simulation method for multi-dimensional time-varying characteristics of DC fault arc in photovoltaic system | |
| Yao et al. | A series DC arc fault detection method and hardware implementation | |
| Wu et al. | Smart detection technology of serial arc fault on low-voltage indoor power lines | |
| You et al. | Research of an improved wavelet threshold denoising method for transformer partial discharge signal | |
| CN108828401B (en) | Method and device for fault line selection | |
| Wang et al. | A novel series arc fault detection method based on mel-frequency cepstral coefficients and fully connected neural network | |
| CN117725393A (en) | A low-voltage arc fault identification method, device, equipment and storage medium | |
| CN106771798A (en) | A kind of fault arc detection method based on the equal difference of wavelet coefficient | |
| CN114707553B (en) | Step-by-step fault arc detection method and device, electronic equipment, and storage medium | |
| Zhang et al. | Photovoltaic dc series arc fault identification method based on precise noise reduction algorithm | |
| CN115144703A (en) | High-resistance grounding fault identification method based on zero-sequence differential current and energy moment indexes | |
| CN115327325A (en) | Charging pile series arc fault detection method and device and electronic equipment | |
| CN118897168A (en) | Ring main unit partial discharge monitoring method and device | |
| CN117169591B (en) | Power system broadband measurement methods, devices and computer equipment | |
| Liu et al. | Extended morlet wavelet-based fir phasor estimation using fake samples | |
| Pattanaik et al. | Power Quality Disturbance Detection and Monitoring of Solar Integrated Micro-Grid |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211126 |