CN102621377B - Fault arc detection method - Google Patents

Abstract

The invention discloses a fault arc detection method. By the fault arc detection method, the discrimination and determination accuracy of a fault arc can be effectively improved. On the basis of universal characteristics of an arc, by acquiring current data of each cycle, existence of the characteristics of zero-rest, asymmetry of positive and negative half cycles, unobvious cyclicity and abundant high-frequency harmonic waves in current waveform is analyzed, and occurrence of arc fault is determined. Compared with the prior art, the fault arc detection method has the advantages that the judgment accuracy and precision of the fault arc can be improved, and the misoperation rate of an arc-fault circuit-interrupter (AFCI) product is reduced.

Description

Fault arc detection method

Technical field

The present invention relates to electric circuit protection equipment field, especially a kind of method of the detection for fault electric arc in AFCI.

Background technology

Electric wiring in family can, loose contact aging because of short circuit, electric wire, electric equipment products failure and other reasons and produce electric arc in the line.Electric arc is a kind of air electric conduction phenomena, and produce strong and lasting electric discharge between two electrodes, concentration of energy, temperature is high, very easily causes the fire failure in family.But traditional domestic circuit failure protecting device can not produce protection to arc fault, a kind of in prevention family fire accident can be very important for the protective device of arc fault.

AFCI (Arc-Fault Circuit-Interrupter) i.e. arc-fault circuit interrupter, the basis of traditional isolating switch with the addition of the function shielded to fault electric arc, to take precautions against the fire caused due to fault electric arc.The Electrical Safety that appears as of AFCI provides reliable guarantee, is applied to aerospace field the earliest, has progressively entered now among daily life.At present, for the AFCI product of low-voltage distribution system on domestic and international market, be all develop for the load equipment of 120V, 60Hz, the AFCI product being suitable for domestic 220V/AC, 50Hz distribution system is a brand-new technology.

As the device that can shield to fault electric arc, being that it is the most basic to the accurate judgement of fault electric arc, is also most crucial technology.Break down electric arc time current waveform can be variant because of the difference of load, but some features are electric arc has, such as: electric current " current zero stop " phenomenon; Current waveform positive-negative half-cycle is asymmetric; Waveform loses periodically; Containing abundant high fdrequency component in current waveform.But due to some defects on arc-detection algorithm, make current most of AFCI product to the judgement Shortcomings part of fault electric arc electric current, easily produce identification error and misoperation.

Summary of the invention

The object of this invention is to provide a kind of fault arc detection method distinguishing and judge accuracy that effectively can improve fault electric arc.

For solving the problem, a kind of fault arc detection method of the present invention, comprises the following steps:

1) gather the current value of alternating current one-period, obtain several discrete current values;

2) calculate the number of the value in this cycle in the middle of several current values near 0, whether the electric current characterizing this cycle exists flat shoulder, obtains the Parameter N of some number near expression 0;

Judge that whether the positive-negative half-cycle of this periodic current is symmetrical by the mean value calculating this periodic current, obtain characterizing the whether symmetrical parameter I of electric current positive-negative half-cycle _avr;

Judge whether present current waveform loses periodically by the current value contrasting adjacent two cycles, obtain the parameter I characterizing current cycle _per;

FFT conversion (Fast Fourier Transform (FFT)) is carried out to the some data points collected, FFT conversion adopts hanning window (raised cosine window) windowing process, to improve measuring accuracy, obtained the content of each harmonic in current waveform by FFT conversion, finally calculate the aberration rate I of current waveform _x, judge in current waveform, whether contained higher hamonic wave exceeds standard with this;

3) successively gained four parameters are compared with corresponding threshold value, each comparing result is added to the parameter F characterizing arc fault by coefficient of correspondence _v, comprise the following steps:

A, F _vvalue resets;

B, by same for Parameter N threshold value beta ₁relatively, if N > is β ₁, then N value is multiplied by coefficient n ₁count arc fault parameter F _v; If N < is β ₁, then this judgement is not counted in F _v.

C, similar above-mentioned steps, successively by parameter I _avr, I _per, I _xwith corresponding threshold value beta ₂, β ₃, β ₄compare, if be greater than threshold value, be multiplied by corresponding coefficient n ₂, n ₃, n ₄count arc fault parameter F _v, be less than, be not counted in, obtain F _v;

4) by F _vwith threshold value beta ₅relatively, if be greater than threshold value, then this cycle is designated as inaction interval, is characterizing the parameter C of inaction interval number _fin add 1;

5) step 1 is repeated) to step 3), if F _vstill be greater than threshold value, namely continuous two cycles are inaction interval, then C _fadd 1 again; Otherwise C _freset.

6) if 8 cycles of current continuity be all judged as inaction interval, i.e. C _fcan be added to 8, then Cutoff current produces arc fault;

Above threshold value beta ₁, β ₂, β ₃, β ₄, β ₅and coefficient n ₁, n ₂, n ₃, n ₄be set-point, specifically refer to data to optimize further and determine by experiment.

Fault arc detection method of the present invention is based on the universals of electric arc, analyze current waveform whether have that zero stops, positive-negative half-cycle is asymmetric, periodicity is not obvious by gathering the current data of each cycle, and whether containing abundant these features of high-frequency harmonic, judge whether to there occurs arc fault, relative to prior art, the accuracy to fault electric arc judgement and precision can be improved, decrease the malfunction rate of AFCI product.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of fault arc detection method in the present invention.

Embodiment

In order to make those skilled in the art person understand technical solution of the present invention better, below in conjunction with drawings and embodiments, the present invention is described in further detail.

As shown in Figure 1, fault arc detection method of the present invention, comprises the following steps:

1) gather the current value of family expenses 220V/50Hz alternating current one-period with the sample frequency of 1600Hz, obtain 32 discrete current values, be designated as I ₀, I ₁..., I ₃₁, form the current waveform data set in this cycle;

2) calculate the number of the value in the middle of 32 current values in one-period near 0, whether the electric current characterizing this cycle exists flat shoulder, and concrete calculating formula is:

$N=\underset{k=0}{\overset{31}{\mathrm{\&Sigma;}}}{Z}_k$

Obtain the Parameter N of some number near expression 0

The implication of above-mentioned function is the total number of sample points meeting flat shoulder characteristic in the sample magnitude in calculating 220V/50Hz ac period, according to the current value gathering family expenses 220V/50Hz alternating current one-period with the sample frequency of 1600Hz, obtain 32 discrete current values, be designated as I ₀, I ₁..., I ₃₁, form the current waveform data set in this cycle; Calculate the number of the value in the middle of 32 current values in one-period near 0, whether the electric current characterizing this cycle exists flat shoulder, the Z in function _kbe the function of flat shoulder characteristic, the expression formula of function is:

Z _k=1, | I _k| <0.5 and | I _k– I _k+ 1|<0.05;

Z _k=0, other;

In function, k is the integer from 0 to 31, I _krepresent each corresponding current value, | I _k| <0.5, represents that the current amplitude of current collection point is less than 0.5, | I _k– I _k+1| <0.05 represents that the current amplitude of current collection point and the difference of more lower current amplitude are less than 0.05, I _k+1represent the current amplitude of next collection point, when meeting | I _k| <0.5 and | I _k– I _k+1| <0.05 two sufficient and necessary conditions, Z _kvalue be 1, otherwise Z _kvalue be 0;

Judge that whether the positive-negative half-cycle of this periodic current is symmetrical by the mean value calculating one-period electric current, specific formula for calculation is as follows:

$I_{\mathrm{avr}}=\frac{1}{32}\underset{k=0}{\overset{31}{\mathrm{\&Sigma;}}}{I}_k$

Obtain characterizing the whether symmetrical parameter I of electric current positive-negative half-cycle _avr.

Judge whether present current waveform loses periodically, specifically calculates by the current value contrasting adjacent two cycles

$I_{\mathrm{per}}=\underset{k=0}{\overset{31}{\mathrm{\&Sigma;}}}|I_k^{\&prime;}-I_k|$

Obtain the parameter I characterizing current cycle _per.

Carry out FFT conversion to 32 data points collected, FFT conversion adopts hanning windowing process, to improve measuring accuracy, is obtained the content of each harmonic in current waveform, finally calculate the aberration rate I of current waveform by FFT conversion _x, judge in current waveform, whether contained higher hamonic wave exceeds standard with this.

3) successively gained four parameters are compared with corresponding threshold value, each comparing result is added to the parameter F characterizing arc fault by certain coefficient _v, concrete implementation step is as follows:

F before a, each judgement _vvalue resets;

B, by same for Parameter N threshold value beta ₁relatively, if N > is β ₁, then N value is multiplied by coefficient n ₁count arc fault parameter F _v; If N < is β ₁, then this judgement is not counted in F _v.Wherein β ₁for given threshold value, coefficient n ₁for given coefficient.

C, similar above-mentioned steps, successively by parameter I _avr, I _per, I _xwith corresponding threshold value beta ₂, β ₃, β ₄compare, if be greater than threshold value, be multiplied by corresponding coefficient n ₂, n ₃, n ₄count arc fault parameter F _v, be less than, be not counted in, wherein β ₂, β ₃, β ₄for given threshold value, coefficient n ₂, n ₃, n ₄be given coefficient.Such as, if above-mentioned comparative result is be greater than, then obtain:

Fv＝n ₁*N+n ₂*I _avr+n ₃*I _per+n ₄*I _x

4) by F _vwith threshold value beta ₅relatively, if be greater than threshold value, then this cycle is designated as inaction interval, is characterizing the parameter C of inaction interval number _fin add 1, wherein β ₅for given threshold value.

5) gather the current value in next cycle, and repeat step 1) to step 3), if F _vstill be greater than threshold value, namely continuous two cycles are inaction interval, then C _fadd 1 again; Otherwise C _freset.

6) if 8 cycles of current continuity be all judged as inaction interval, i.e. C _fcan be added to 8, then Cutoff current produces arc fault;

Above threshold value beta ₁, β ₂, β ₃, β ₄, β ₅and coefficient n ₁, n ₂, n ₃, n ₄be set-point, specifically refer to data to optimize further and determine by experiment.

The present invention is based on the universals of electric arc, analyze current waveform whether have that zero stops, positive-negative half-cycle is asymmetric, periodicity is not obvious by gathering the current data of each cycle, and whether containing abundant these features of high-frequency harmonic, judge whether to there occurs arc fault, relative to prior art, the accuracy to fault electric arc judgement and precision can be improved, decrease the malfunction rate of AFCI product.

Claims (1)

1. a fault arc detection method, is characterized in that, comprises the following steps:

1) gather the current value of alternating current one-period, obtain several discrete current values;

2) calculate the number of the value in this cycle in the middle of several current values near 0, whether the electric current characterizing this cycle exists flat shoulder, and obtain the Parameter N of some number near expression 0, described N is obtained by following formula:

The implication of above-mentioned function is the total number of sample points meeting flat shoulder characteristic in the sample magnitude in calculating 220V/50Hz ac period, according to the current value gathering family expenses 220V/50Hz alternating current one-period with the sample frequency of 1600Hz, obtain 32 discrete current values, be designated as I ₀, I ₁..., I ₃₁, form the current waveform data set in this cycle; Calculate the number of the value in the middle of 32 current values in one-period near 0, whether the electric current characterizing this cycle exists flat shoulder, the Z in function _kbe the function of flat shoulder characteristic, the expression formula of function is:

Z _k=1, | I _k| <0.5 and | I _k– I _k+ 1|<0.05;

Z _k=0, other;

In function, k is the integer from 0 to 31, I _krepresent each corresponding current value, | I _k| <0.5, represents that the current amplitude of current collection point is less than 0.5, | I _k– I _k+1| <0.05 represents that the current amplitude of current collection point and the difference of more lower current amplitude are less than 0.05, I _k+1represent the current amplitude of next collection point, when meeting | I _k| <0.5 and | I _k– I _k+1| <0.05 two sufficient and necessary conditions, Z _kvalue be 1, otherwise Z _kvalue be 0;

Judge that whether the positive-negative half-cycle of this periodic current is symmetrical by the mean value calculating this periodic current, obtain characterizing the whether symmetrical parameter I of electric current positive-negative half-cycle _avr, described I _avrobtained by following formula:

；

Judge whether present current waveform loses periodically by the current value contrasting adjacent two cycles, obtain the parameter I characterizing current cycle _per, described I _perobtained by following formula:

；

FFT conversion (Fast Fourier Transform (FFT)) is carried out to the some data points collected, FFT conversion adopts hanning window (raised cosine window) windowing process, to improve measuring accuracy, obtained the content of each harmonic in current waveform by FFT conversion, finally calculate the aberration rate I of current waveform _x, judge in current waveform, whether contained higher hamonic wave exceeds standard with this;

3) successively gained four parameters are compared with corresponding threshold value, each comparing result is added to the parameter F characterizing arc fault by coefficient of correspondence _v, comprise the following steps:

A, F _vvalue resets;

B, by same for Parameter N threshold value beta ₁relatively, if N > is β ₁, then N value is multiplied by coefficient n ₁count arc fault parameter F _v; If N < is β ₁, then this judgement is not counted in F _v;

C, similar above-mentioned steps, successively by parameter I _avr, I _per, I _xwith corresponding threshold value beta ₂, β ₃, β ₄compare, if be greater than threshold value, be multiplied by corresponding coefficient n ₂, n ₃, n ₄count arc fault parameter F _v, be less than, be not counted in, obtain F _v;

4) by F _vwith threshold value beta ₅relatively, if be greater than threshold value, then this cycle is designated as inaction interval, is characterizing the parameter C of inaction interval number _fin add 1;

5) step 1 is repeated) to step 3), if F _vstill be greater than threshold value, namely continuous two cycles are inaction interval, then C _fadd 1 again; Otherwise C _freset;

6) if 8 cycles of current continuity be all judged as inaction interval, i.e. C _fcan be added to 8, then Cutoff current produces arc fault;

Above threshold value beta ₁, β ₂, β ₃, β ₄, β ₅and coefficient n ₁, n ₂, n ₃, n ₄be set-point.