CN104764963B - Alternating current arc fault detection method - Google Patents
Alternating current arc fault detection method Download PDFInfo
- Publication number
- CN104764963B CN104764963B CN201510188978.XA CN201510188978A CN104764963B CN 104764963 B CN104764963 B CN 104764963B CN 201510188978 A CN201510188978 A CN 201510188978A CN 104764963 B CN104764963 B CN 104764963B
- Authority
- CN
- China
- Prior art keywords
- signal
- arc fault
- frequency spectrum
- circuit
- frequency
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a kind of alternating current arc fault detection method, comprise the following steps：A, the voltage signal or current signal in loop to be detected are gathered by transformer, and voltage signal or current signal to collecting are filtered and nursed one's health, and the signal after filtering and conditioning is carried out into analog-to-digital conversion；B, the method that design treatment passes through the obtained voltage signals of step a or current signal；C, the combination of any one method or any two method or the common combination of three methods are handled the obtained voltage signals of step a or current signal in selecting step b, so as to obtain result of determination.The invention has the advantages that easy to implement, detection speed is fast, is quick on the draw, and the degree of accuracy is high.
Description
Technical field
The present invention relates to alternating current arc field of fault detection, particularly a kind of alternating current arc fault detection method.
Background technology
With the continuous social and economic development, the type and quantity of electrical equipment in being built in residential building, factory, market etc.
It is on the increase.Because electric wiring, electrical equipment internal wiring with load or overlond running, cause electric wire insulation layer gradual for a long time
Aging is even damaged.Fault electric arc is easily produced in the position of insulation damages, if can not find and take appropriate measures in time,
Fault electric arc is easy to cause fire or exploded.In addition, on aircraft in addition to the fault electric arc as caused by insulation ag(e)ing, fly
Wiring loosening caused by mechanical oscillation in machine flight course can also produce fault electric arc, serious threat the people's lives and property peace
Entirely.Traditional protection electrical equipment can only play a protective role in the case of overcurrent or short circuit to circuit, and the electricity of fault electric arc
Stream is less than the specified limit value of its action in most cases, it is likely that will not react.Therefore, also need to set up in distribution system
Arc fault detection and protection device, arc fault is prevented to cause fire or blast, protection the people's lives and property peace to reach
Full purpose.
Existing fault arc detection method is generally from following three aspects：
First, temporal signatures.
Method：Reduced including voltage or current amplitude, positive-negative half-cycle is asymmetric, zero stops the appearance in area and dashing forward for waveform slope
The features such as change.
Shortcoming：Voltage or current amplitude can also change when switching different loads.Positive-negative half-cycle is asymmetric existing
As can be influenceed by material molten degree between the conductor of trouble point two.The width that current zero stops area can be by plasma between electrode
The influences of condition such as body state, clearance distance size, impedance angle and change.The slope of voltage or current waveform is stopped area by current zero
The influence of duration length and material surface degree of oxidation.But the plasma between the conductor of arc fault point two in practice
State, gap length are unfixed, and loadtype is nor changeless in many cases.Therefore using temporal signatures very
Hardly possible detects arc fault exactly.
2nd, the amplitude of high-frequency signal
Method：The voltage or current signal collected is handled by high-pass filtering, using filtered signal as modulus
The input signal of conversion.If the amplitude of high-frequency signal exceedes a certain threshold value and its duration exceedes another threshold value, increase
The value of counter.Think to generate arc fault if default threshold value is exceeded in the value of defined time inside counting device.
Shortcoming：The amplitude of high band pulse is influenceed more obvious, high band arteries and veins by certain loads type and size of current
The threshold value of punching needs to adjust for inhomogeneous load.In the complicated distribution that load often changes, this method
It is difficult to ensure that its accuracy.
3rd, wavelet transformation is combined with BP neural network
Method：By testing the characteristic value for the wavelet transformation for obtaining multigroup a variety of loads, using these characteristic values as nerve
The training sample of network, after neutral net convergence, it can just use in the microprocessor.By two layers of details before wavelet transformation
Characteristic value is inputted neutral net, is by the output valve of neutral net and threshold value multilevel iudge by the energy of signal as characteristic value
It is no to have arc fault.
Shortcoming：BP neural network needs to be trained by substantial amounts of test data, treats just may be used after neutral net convergence
Used with being transplanted in processor.If the load type that training sample is insufficient or sample includes is not comprehensive enough, can cause
The accuracy of the detection algorithm.And early stage will be greatly increased by testing the enough train samples of acquisition
Workload.
The content of the invention
The invention aims to solve the above problems, a kind of alternating current arc fault detection method is devised.
Realize above-mentioned purpose the technical scheme is that, a kind of alternating current arc fault detection method, comprise the following steps：
A, the voltage signal and/or current signal in loop to be detected, and the voltage to collecting are gathered by transformer
Signal and/or current signal are filtered and nursed one's health, by the signal input analog-to-digital converter after filtering and conditioning, analog-digital converter
Output be discrete voltage signal and/or current signal；
B, design treatment passes through the method for the obtained voltage signals of step a and/or current signal, the method bag of the processing
Include：
Method one, voltage signal and/or current signal progress wavelet transformation are decomposed to obtain approximate signal and detail signal,
Whether there is arc fault in decision circuit by the contrast of detail signal and threshold value；
Method two, windowed DFT is done to voltage signal and/or current signal obtain the frequency spectrum of signal, calculate
The harmonic component of certain specific exponent number is relative to fundametal compoment proportion, by the contrast of the ratio and threshold value come in decision circuit
Whether arc fault is occurred；
Method three, windowed DFT is done to voltage signal and/or current signal obtain the frequency spectrum of signal, from frequency
Several points are chosen in spectrum and carry out computing according to default algorithm, are in decision circuit by the contrast of operation result and threshold value
It is no arc fault occur；
C, in selecting step b the combination of any one method or any two method or the common combination of three methods to step
The voltage signal and/or current signal that rapid a is obtained are handled, so as to obtain result of determination.
The particular content of method one in the step b is as follows：
Non-linear form filtering process is carried out to the voltage signal and/or current signal obtained in step a；
Wavelet transform is done to filtered signal, obtains the detail signal d of a certain layer；
Calculate d virtual value：Wherein Level is exponent number (d itself the resolution ratio water of this layer
It is flat), LdFor detail signal d discrete points.
By D compared with threshold value thr1.This method thinks do not have arc fault generation in circuit if D≤thr1；If D
>Then this method judges to generate arc fault in circuit thr1.
The particular content of method two in the step b is as follows：
Windowed DFT is done to signal and obtains the frequency spectrum of signal.The amplitude at frequency f (i) places is designated as y (f (i)),
The n of wherein i=1,2,3 ..., n are the discrete point number that the frequency spectrum includes.
Calculate the harmonic component of certain specific exponent number：Wherein j=1,2,
3 ..., F are the frequency of AC power, and δ span is 0~20.
Calculate the fundametal compoment in frequency spectrum：Wherein F be AC power frequency, δ span
For 0~20.
Calculate accounting of the harmonic component with respect to fundametal compoment of certain specific exponent number： J=1,2,
3……。
By ηjCompared with threshold value thr2.If ηjThen this method thinks do not have arc fault generation in circuit to≤thr2；If
ηj>Then this method judges to generate arc fault in circuit thr2.
The particular content of method three in the step b is as follows：
Windowed DFT is done to signal and obtains the frequency spectrum of signal.The amplitude at frequency f (i) places is designated as y (f (i)),
The n of wherein i=1,2,3 ....N is the discrete point number that the frequency spectrum includes.
Optional m point, wherein m≤n from frequency spectrum.Frequency corresponding to this m point is denoted as f (p successively1)、f(p2)、f
(p3)……f(pm), corresponding amplitude is followed successively by y (f (p1))、y(f(p2))、y(f(p3))……y(f(pm))。
By m selected point computing according to the following formula：Wherein j's takes
Value scope is p1、p2、p3……pm, α ＞ 0 and α ≠ 1, β ≠ 0, γ ∈ R.
Averaged:
By result compared with threshold value thr3.If result≤thr3, this method thinks do not have electric arc event in circuit
Barrier produces；If result>Thr3, then this method judge to generate arc fault in circuit.
The alternating current arc fault detection method designed using technical scheme, detection method of the invention are easy to use
Hardware is realized, and helps to reduce hard-wired cost；The detection method operand is small, can shorten operation time, detection
Speed is fast, is quick on the draw, and the degree of accuracy is high；The detection method can detect the arc fault of a variety of loadtypes well, and
The setting of threshold value need not change with the change of loadtype；The detection method it is applied widely, to 50Hz, 400Hz etc.
The power supply of multi-frequency scope is applicable；Such as the details that Fig. 6 is a certain layer obtained by wavelet transform in method one
Signal curve.The detail signal when left-half of curve is circuit normal work, right half part are that occur electric arc event in circuit
Detail signal during barrier.The amplitude for occurring this layer of detail signal after arc fault as seen from Figure 6 significantly increases, and this method can be accurate
Really detection arc fault.By taking 2~7 subharmonic as an example, table 1 lists Application way two and harmonic wave is calculated with respect to fundamental wave accounting.
From data in table, when no arc fault compared with, there is arc fault time-harmonic wave accounting and significantly increase, therefore this method
Can be with effective detection arc fault.(a) is normal and failure under the 50Hz supply frequencies that application method three is calculated in Fig. 7
Comparison diagram, (b) are normal and failure comparison diagram under 400Hz supply frequencies.When blue curve is no arc fault in figure
Result of calculation, result of calculation when red curve is arc fault.The detection method improves the standard of arc fault detection technique
Exactness, improve the reliability of relay protection system.
Brief description of the drawings
Fig. 1 is the detection algorithm flow chart of the embodiment 1 of alternating current arc fault detection method of the present invention；
Fig. 2 is the detection algorithm flow chart of the embodiment 2 of alternating current arc fault detection method of the present invention；
Fig. 3 is the detection algorithm flow chart of the embodiment 3 of alternating current arc fault detection method of the present invention；
Fig. 4 is the detection algorithm flow chart of the embodiment 4 of alternating current arc fault detection method of the present invention；
Fig. 5 is the detection algorithm flow chart of the embodiment 5 of alternating current arc fault detection method of the present invention；
Fig. 6 is the details of a certain layer that wavelet transform obtains in alternating current arc fault detection method one of the present invention
Signal curve；
Table 1 is 2~7 subharmonic the accounting for respect to fundamental wave that alternating current arc fault detection method two of the present invention is calculated
Than；
Fig. 7 be alternating current arc fault detection method three of the present invention for two kinds of supply frequencies of 50Hz, 400Hz it is normal and
The result of calculation comparison diagram of failure.
Embodiment
In order that those skilled in the art will better understand the technical solution of the present invention, below in conjunction with the accompanying drawings to the present invention
It is specifically described,
Embodiment 1
As shown in figure 1, using by the way of three kinds of methods are combined to the obtained voltages of step a or current signal at
Reason, specific processing step are as follows：
Step 1: system initialization, it is F to set supply frequency according to actual conditions, sets the threshold value n_ of cycle-index
thr。
Step 2: to counter C1, C2, C3 zero setting, to cycle-index n_cycle zero setting；
Step 3: collection voltages and/or current signal；
Step 4: signal is filtered and nursed one's health；
Step 5: by the signal input analog-to-digital converter after filtering and conditioning, the output of analog-digital converter is discrete electricity
Press signal and/or current signal；
Step 6: the signal after analog-to-digital conversion is handled signal by following two branch roads respectively：
Tie point：Method one, wavelet transformation
1st, non-linear form filtering process is carried out to signal；
2nd, wavelet transform is done to filtered signal, obtains the detail signal d of a certain layer；
3rd, d virtual value is calculated：Wherein Level is exponent number (d itself the resolution ratio water of this layer
It is flat), LdFor detail signal d discrete points；
4th, by D compared with threshold value thr1.This method thinks do not have arc fault generation in circuit if D≤thr1；Such as
Fruit D>Then this method judges to generate arc fault in circuit thr1, adds one with seasonal counter C1.
Second branch road：
Windowed DFT is done to signal first and obtains the frequency spectrum of signal.The amplitude at frequency f (i) places is designated as y (f
), (i) n of wherein i=1,2,3 ....N is the discrete point number that the frequency spectrum includes.
Then obtained frequency spectrum is handled according to following two methods respectively：
Method two, harmonic wave accounting
1st, the harmonic component of certain specific exponent number is calculated：Wherein j=1,2,
3 ..., F are the frequency of AC power, and δ span is 0~20.
2nd, the fundametal compoment in frequency spectrum is calculated：Wherein F be AC power frequency, δ value model
Enclose for 0~20.
3rd, accounting of the harmonic component with respect to fundametal compoment of certain specific exponent number is calculated： J=1,2,
3……。
4th, by ηjCompared with threshold value thr2.If ηjThen this method thinks do not have arc fault generation in circuit to≤thr2；Such as
Fruit ηj>Then this method judges to generate arc fault in circuit thr2, adds one with seasonal counter C2.
Method three, spectrum analysis
1st, optional m point, wherein m≤n from frequency spectrum.Frequency corresponding to this m point is denoted as f (p successively1)、f(p2)、f
(p3)……f(pm), corresponding amplitude is followed successively by y (f (p1))、y(f(p2))、y(f(p3))……y(f(pm))。
2nd, by m selected point computing according to the following formula：
Wherein j span is p1、p2、p3……pm, α ＞ 0
And α ≠ 1, β ≠ 0, γ ∈ R.
3rd, averaged:
4th, by result compared with threshold value thr3.If result≤thr3, this method thinks do not have electric arc in circuit
Failure produces；If result>Thr3, then this method judge to generate arc fault in circuit, make counter C3 add one.
Step 7:
If C1, C2, C3 are zero, step 3 is gone to；
If there is one to be not zero in C1, C2, C3, step 8 is gone to.
Step 8:
If w1*C1+w2*C2+w3*C3 >=n_thr*3* μ are set up, judge to generate arc fault in circuit, send report
Alert signal and break signal.
If w1*C1+w2*C2+w3*C3<N_thr*3* μ are set up, then turn to step 9.
Wherein w1, w2, w3 and μ are any real number in the range of 0~1.
Step 9:
N_cycle=n_cycle+1.
Step 10:
Judge whether n_cycle=n_thr sets up.If not, jump to step 3；If n_cycle=n_thr, jump
Go to step 2.
Embodiment 2
As shown in Fig. 2 to the obtained voltages of step a or electricity by the way of method two and the two methods of method three are combined
Stream signal is handled, and specific processing step is as follows：
Step 1: system initialization, it is F to set supply frequency according to actual conditions, sets the threshold value n_ of cycle-index
thr。
Step 2: to counter C2, C3 zero setting, to cycle-index n_cycle zero setting；
Step 3: collection voltages and/or current signal；
Step 4: signal is filtered and nursed one's health；
Step 5: by the signal input analog-to-digital converter after filtering and conditioning, the output of analog-digital converter is discrete electricity
Press signal and/or current signal；
Step 6: doing windowed DFT to signal, the frequency spectrum of signal is obtained.The amplitude at frequency f (i) places is designated as
Y (f (i)), the n of wherein i=1,2,3 ....N is the discrete point number that the frequency spectrum includes.Then by obtained frequency spectrum respectively according to
Following two methods are handled：
Method two, even-order harmonic accounting
1st, the harmonic component of certain specific exponent number is calculated：Wherein j=1,2,
3 ..., F are the frequency of AC power, and δ span is 0~20.
2nd, the fundametal compoment in frequency spectrum：Wherein F is the frequency of AC power, and δ span is
0~20.
3rd, accounting of the harmonic component with respect to fundametal compoment of certain specific exponent number is calculated： J=1,2,
3……。
4th, by ηjCompared with threshold value thr2.If ηjThen this method thinks do not have arc fault generation in circuit to≤thr2；Such as
Fruit ηj>Then this method judges to generate arc fault in circuit thr2, adds one with seasonal counter C2.
Method three, spectrum analysis
1st, optional m point, wherein m≤n from frequency spectrum.Frequency corresponding to this m point is denoted as f (p successively1)、f(p2)、f
(p3)……f(pm), corresponding amplitude is followed successively by y (f (p1))、y(f(p2))、y(f(p3))……y(f(pm))。
2nd, by m selected point computing according to the following formula：
Wherein j span is p1、p2、p3……pm, α ＞ 0
And α ≠ 1, β ≠ 0, γ ∈ R.
3rd, averaged:
4th, by result compared with threshold value thr3.If result≤thr3, this method thinks do not have electric arc in circuit
Failure produces；If result>Thr3, then this method judge to generate arc fault in circuit, make counter C3 add one.
Step 7:
If C2, C3 are zero, step 3 is gone to；
If there is one to be not zero in C2, C3, step 8 is gone to.
Step 8:
If w2*C2+w3*C3 >=n_thr*2* μ are set up, judge to generate arc fault in circuit, send alarm signal
And break signal.
If w2*C2+w3*C3<N_thr*2* μ are set up, then turn to step 9.
Wherein w2, w3 and μ are any real number in the range of 0~1.
Step 9:
N_cycle=n_cycle+1.
Step 10:
Judge whether n_cycle=n_thr sets up.If not, jump to step 3；If n_cycle=n_thr, jump
Go to step 2.
Embodiment 3
As shown in figure 3, handled using the obtained voltages of a pair of step a of method or current signal, specific processing step
It is rapid as follows：
Step 1: system initialization, it is F to set supply frequency according to actual conditions, sets the threshold value n_ of cycle-index
thr。
Step 2: to counter C1 zero setting, to cycle-index n_cycle zero setting；
Step 3: collection voltages and/or current signal；
Step 4: signal is filtered and nursed one's health；
Step 5: by the signal input analog-to-digital converter after filtering and conditioning, the output of analog-digital converter is discrete electricity
Press signal and/or current signal；
Step 6: the signal after analog-to-digital conversion is handled by a pair of signals of method：
1st, non-linear form filtering process is carried out to signal；
2nd, wavelet transform is done to filtered signal, obtains the detail signal d of a certain layer；
3rd, d virtual value is calculated：Wherein Level is exponent number (d itself the resolution ratio water of this layer
It is flat), LdFor detail signal d discrete points；
4th, by D compared with threshold value thr1.This method thinks do not have arc fault generation in circuit if D≤thr1；Such as
Fruit D>Then this method judges to generate arc fault in circuit thr1, adds one with seasonal counter C1.
Step 7:
If C1 is zero, step 3 is gone to；
If C1 is not zero, step 8 is gone to.
Step 8:
If C1 >=n_thr* μ are set up, judge to generate arc fault in circuit, send alarm signal and break signal.
If C1<N_thr* μ are set up, then turn to step 9.
Wherein μ is any real number in the range of 0~1.
Step 9:
N_cycle=n_cycle+1.
Step 10:
Judge whether n_cycle=n_thr sets up.If not, jump to step 3；If n_cycle=n_thr, jump
Go to step 2.
Embodiment 4
As shown in figure 4, the obtained voltages of step a or current signal are handled using method two, specific processing step
It is rapid as follows：
Step 1: system initialization, it is F to set supply frequency according to actual conditions, sets the threshold value n_ of cycle-index
thr。
Step 2: to counter C2 zero setting, to cycle-index n_cycle zero setting；
Step 3: collection voltages and/or current signal；
Step 4: signal is filtered and nursed one's health；
Step 5: by the signal input analog-to-digital converter after filtering and conditioning, the output of analog-digital converter is discrete electricity
Press signal and/or current signal；
Step 6: doing windowed DFT to the signal after analog-to-digital conversion obtains the frequency spectrum of signal.Frequency f
(i) amplitude at place is designated as y (f (i)), the n of wherein i=1,2,3 ....N is the discrete point number that the frequency spectrum includes.Then the side of passing through
Method two is handled：
1st, the harmonic component of certain specific exponent number is calculated：Wherein j=1,2,
3 ..., F are the frequency of AC power, and δ span is 0~20.
2nd, the fundametal compoment in frequency spectrum is calculated：Wherein F is the frequency of AC power, δ's
Span is 0~20.
3rd, accounting of the harmonic component with respect to fundametal compoment of certain specific exponent number is calculated： J=1,2,
3……。
4th, by ηjCompared with threshold value thr2.If ηjThen this method thinks do not have arc fault generation in circuit to≤thr2；Such as
Fruit ηj>Then this method judges to generate arc fault in circuit thr2, adds one with seasonal counter C2.
Step 7:
If C2 is zero, step 3 is gone to；
If C2 is not zero, step 8 is gone to.
Step 8:
If C2 >=n_thr* μ are set up, judge to generate arc fault in circuit, send alarm signal and break signal.
If C2<N_thr* μ are set up, then turn to step 9.
Wherein μ is any real number in the range of 0~1.
Step 9:
N_cycle=n_cycle+1.
Step 10:
Judge whether n_cycle=n_thr sets up.If not, jump to step 3；If n_cycle=n_thr, jump
Go to step 2.
Embodiment 5
As shown in figure 5, the obtained voltages of step a or current signal are handled using method three, specific processing step
It is rapid as follows：
Step 1: system initialization, it is F to set supply frequency according to actual conditions, sets the threshold value n_ of cycle-index
thr。
Step 2: to counter C3 zero setting, to cycle-index n_cycle zero setting；
Step 3: collection voltages and/or current signal；
Step 4: signal is filtered and nursed one's health；
Step 5: by the signal input analog-to-digital converter after filtering and conditioning, the output of analog-digital converter is discrete electricity
Press signal and/or current signal；
Step 6: doing windowed DFT to the signal after analog-to-digital conversion, the frequency spectrum of signal is obtained.Frequency
The amplitude at f (i) places is designated as y (f (i)), the n of wherein i=1,2,3 ....N is the discrete point number that the frequency spectrum includes.Then incite somebody to action
To frequency spectrum handled according to method three：
1st, optional m point, wherein m≤n from frequency spectrum.Frequency corresponding to this m point is denoted as f (p successively1)、f(p2)、f
(p3)……f(pm), corresponding amplitude is followed successively by y (f (p1))、y(f(p2))、y(f(p3))……y(f(pm))。
2nd, by m selected point computing according to the following formula：
Wherein j span is p1、p2、p3……pm, α ＞ 0
And α ≠ 1, β ≠ 0, γ ∈ R.
3rd, averaged:
4th, by result compared with threshold value thr3.If result≤thr3, this method thinks do not have electric arc in circuit
Failure produces；If result>Thr3, then this method judge to generate arc fault in circuit, make counter C3 add one.
Step 7:
If C3 is zero, step 3 is gone to；
If C3 is not zero, step 8 is gone to.
Step 8:
If C3 >=n_thr* μ are set up, judge to generate arc fault in circuit, send alarm signal and break signal.
If C3<N_thr* μ are set up, then turn to step 9.
Wherein μ is any real number in the range of 0~1.
Step 9:
N_cycle=n_cycle+1.
Step 10:
Judge whether n_cycle=n_thr sets up.If not, jump to step 3；If n_cycle=n_thr, jump
Go to step 2.
Above-mentioned technical proposal only embodies the optimal technical scheme of technical solution of the present invention, those skilled in the art
Some variations that some of which part may be made are embodied the present invention principle, belong to protection scope of the present invention it
It is interior.
Claims (2)
1. a kind of alternating current arc fault detection method, it is characterised in that comprise the following steps：
A, the voltage signal and/or current signal in loop to be detected are gathered by transformer, and the voltage to collecting is believed
Number and/or current signal be filtered and nurse one's health, by the signal input analog-to-digital converter after filtering and conditioning, analog-digital converter
Export as discrete voltage signal and/or current signal；
B, design treatment is included by the method for the obtained voltage signals of step a and/or current signal, the processing method：
Method one, voltage signal and/or current signal progress wavelet transformation are decomposed to obtain approximate signal and detail signal, passed through
The contrast of detail signal and threshold value comes in decision circuit whether arc fault occur；
Method two, windowed DFT is done to voltage signal and/or current signal obtain the frequency spectrum of signal, it is special to calculate certain
Determine the harmonic component of exponent number relative to fundametal compoment proportion, by the contrast of the ratio and threshold value come in decision circuit whether
There is arc fault；
Method three, windowed DFT is done to voltage signal and/or current signal obtain the frequency spectrum of signal, from frequency spectrum
Choose several points and carry out computing according to default algorithm, whether gone out in decision circuit by the contrast of operation result and threshold value
Arc fault is showed；
C, in selecting step b the combination of any one method or any two method or the common combination of three methods to step a
Obtained voltage signal and/or current signal is handled, and so as to obtain result of determination, the method one in the step b is specific
Content is as follows：
Non-linear form filtering process is carried out to the voltage signal and/or current signal obtained in step a；
Wavelet transform is done to filtered signal, obtains the detail signal d of a certain layer；
Calculate d virtual value：Wherein Level is the exponent number (d itself level of resolution) of this layer, Ld
For detail signal d discrete points；
By D compared with threshold value thr1；This method thinks do not have arc fault generation in circuit if D≤thr1；If D>
Then this method judges to generate arc fault in circuit thr1,
The particular content of method two in the step b is as follows：
Windowed DFT to be done to signal and obtains the frequency spectrum of signal, the amplitude at frequency f (i) places is designated as y (f (i)), wherein
The n of i=1,2,3 ...；N is the discrete point number that the frequency spectrum includes；
Calculate the harmonic component of certain specific exponent number：Wherein j=1,2,3 ..., F are
The frequency of AC power, δ span is 0~20；
Calculate the fundametal compoment in frequency spectrum：Wherein F is the frequency of AC power, and δ span is 0
~20；
Calculate accounting of the harmonic component with respect to fundametal compoment of certain specific exponent number：
By ηjCompared with threshold value thr2；If ηjThen this method thinks do not have arc fault generation in circuit to≤thr2；If ηj>
Then this method judges to generate arc fault in circuit thr2.
2. alternating current arc fault detection method according to claim 1, it is characterised in that the method three in the step b
Particular content is as follows：
Windowed DFT is done to signal and obtains the frequency spectrum of signal；The amplitude at frequency f (i) places is designated as y (f (i)), wherein
The n of i=1,2,3 ..., n are the discrete point number that the frequency spectrum includes；
Optional m point, wherein m≤n, frequency corresponding to this m point are denoted as f (p successively from frequency spectrum1)、f(p2)、f(p3)……f
(pm), corresponding amplitude is followed successively by y (f (p1))、y(f(p2))、y(f(p3))……y(f(pm))；
By m selected point computing according to the following formula：
Wherein j span is p1、p2、p3……pm, α ＞ 0 and α ≠
1, β ≠ 0, γ ∈ R,
Averaged:
By result compared with threshold value thr3, if result≤thr3, this method thinks do not have arc fault production in circuit
It is raw；If result>Thr3, then this method judge to generate arc fault in circuit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510188978.XA CN104764963B (en) | 2015-04-20 | 2015-04-20 | Alternating current arc fault detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510188978.XA CN104764963B (en) | 2015-04-20 | 2015-04-20 | Alternating current arc fault detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104764963A CN104764963A (en) | 2015-07-08 |
| CN104764963B true CN104764963B (en) | 2018-03-13 |
Family
ID=53646917
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510188978.XA Expired - Fee Related CN104764963B (en) | 2015-04-20 | 2015-04-20 | Alternating current arc fault detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104764963B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105866635B (en) * | 2015-08-05 | 2019-02-12 | 深圳市赋安安全系统有限公司 | A kind of fault electric arc detection method and device |
| CN106961248B (en) * | 2017-04-25 | 2019-03-26 | 西安交通大学 | Photovoltaic system fault arc detection method |
| US10509067B2 (en) | 2017-07-06 | 2019-12-17 | Mersen Usa Newburyport-Ma, Llc | Method for AC arc fault detection using multidimensional energy points |
| US10985695B2 (en) * | 2018-08-28 | 2021-04-20 | Analog Devices International Unlimited Company | DC arc detection and photovoltaic plant profiling system |
| CN109917241A (en) * | 2019-03-19 | 2019-06-21 | 德力西电气有限公司 | A kind of detection method and fault electric arc breaker of fault electric arc |
| CN110108972B (en) * | 2019-06-26 | 2021-10-08 | 云南电网有限责任公司电力科学研究院 | Arc detection method for single-phase high-resistance earth fault |
| CN110376489B (en) * | 2019-08-13 | 2021-07-27 | 南京东博智慧能源研究院有限公司 | Intelligent power distribution network fault arc detection system |
| CN112255515A (en) * | 2020-10-12 | 2021-01-22 | 浙江长元科技有限公司 | Fault arc detection method and device |
| CN113702739A (en) * | 2021-08-26 | 2021-11-26 | 广东电网有限责任公司 | Electric arc detection method and device based on wavelet decomposition and neural network |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5335230A (en) * | 1991-08-12 | 1994-08-02 | Ncr Corporation | Apparatus and method for automatic digitizer fault detection |
| US8159793B2 (en) * | 2008-12-22 | 2012-04-17 | General Electric Company | Arc detection using detailed and approximate coefficients from discrete wavelet transforms |
| CN101696986B (en) * | 2009-10-26 | 2012-06-13 | 吴为麟 | Fault arc detection method and protection device adopting same |
| CN202524053U (en) * | 2012-04-18 | 2012-11-07 | 天津市鸿远电气设备有限公司 | Electric arc failure protector |
| CN104251945B (en) * | 2013-06-28 | 2017-06-27 | 上海电科电器科技有限公司 | arc fault detection device |
-
2015
- 2015-04-20 CN CN201510188978.XA patent/CN104764963B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN104764963A (en) | 2015-07-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104764963B (en) | Alternating current arc fault detection method | |
| CN104614608B (en) | A kind of low pressure parallel arc fault detection means and method | |
| CN105223476A (en) | A kind of fault arc detection method based on high-frequency characteristic and device | |
| CN105067966B (en) | The low-voltage alternating-current fault arc detection method of feature based modal components energy spectrometer | |
| CN106908671A (en) | A kind of non-intrusion type household loads intelligent detecting method and system | |
| CN107069734B (en) | Active filter with lightning protection isolation and arc detection functions and method | |
| CN104569683B (en) | A kind of detection method of fault electric arc | |
| CN105223483B (en) | A kind of DC transmission engineering insulator chain defect online detection method and device | |
| CN107765108A (en) | A kind of island detection method of inverter, device and electric power system | |
| CN106324406B (en) | A kind of transformer DC magnetic bias method for diagnosing faults and device | |
| CN203299270U (en) | Device for accurately measuring CVT grid-side voltage harmonic wave | |
| CN105891606A (en) | Device for testing tower grounding resistance through disassembly-free ground lead | |
| CN205941736U (en) | Trouble arc detection device | |
| CN204575811U (en) | A kind of test macro of high-voltage switch contact | |
| CN109782122A (en) | A kind of detection method of S Injection Signal | |
| CN107092213B (en) | Passive filtering device and method with fault arc detection function | |
| CN103336205B (en) | Island effect detecting system when a kind of two harmonic currents are injected simultaneously into and method | |
| CN105676088A (en) | Device and method for testing fault arc detection apparatus | |
| CN107134777B (en) | Passive filtering device and method with lightning protection isolation and fire prevention early warning functions | |
| CN105301391B (en) | Isolated island detecting device and method based on active power filtering and harmonic compensation | |
| CN108107329A (en) | A kind of alternating current arc frequency domain detection method | |
| CN110888086A (en) | Circuit electric leakage detection positioning system based on signal injection | |
| CN204330871U (en) | A kind of low pressure parallel arc fault pick-up unit | |
| CN104062555B (en) | The discrimination method of distribution line high resistance earthing fault characteristic harmonics | |
| CN103427391B (en) | Method for identifying surge interference in microcomputer protection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180313 Termination date: 20210420 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |