CN102981088B - Fault arc detection method - Google Patents

Abstract

The invention discloses a kind of fault arc detection method distinguishing and judge accuracy that effectively can improve fault electric arc, based on the universals of electric arc, by gather the current data of each cycle analyze current waveform whether exist lose periodically, asymmetric, the flat shoulder of positive-negative half-cycle and rate of change excessive, 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.

Description

Fault arc detection method

Technical Field

The invention relates to the field of circuit protection equipment, in particular to a method for detecting fault arcs in AFCI.

Background

Electrical lines in a home can generate arcs in the lines due to short circuits, wire degradation, poor contact, electrical product failures, and the like. The electric arc is an air conduction phenomenon, strong and durable discharge is generated between two electrodes, energy is concentrated, temperature is high, and a household fire accident is easily caused. However, the conventional household circuit fault protection device cannot protect against the arc fault, and a protection device against the arc fault is very important in preventing a household fire accident.

AFCI (Arc-fault circuit-Interrupter), an Arc fault circuit Interrupter, adds a function of protecting a fault Arc on the basis of a conventional circuit breaker to prevent a fire caused by the fault Arc. The emergence of AFCI provides reliable guarantee for power utilization safety, is applied to the field of aerospace at first, and gradually enters the daily life of people at present. At present, AFCI products aiming at low-voltage distribution systems in domestic and foreign markets are developed aiming at 120V and 60Hz load equipment, and the AFCI products suitable for domestic 220V/AC and 50Hz distribution systems are a brand-new technology.

As a device capable of protecting the fault arc, accurate determination of the fault arc is the most basic and most core technology. The current waveform at which a fault arc occurs will vary from load to load, but some characteristics are common to arcs, such as: the phenomenon of current zero break; the positive and negative half cycles of the current waveform are asymmetric; the waveform loses periodicity; the current waveform contains a large amount of high-frequency components. However, due to defects of an arc detection algorithm, most of AFCI products have defects in judging fault arc current, and identification errors and misoperation are easy to generate.

Disclosure of Invention

The invention aims to provide a fault arc detection method which can effectively improve the accuracy of distinguishing and judging fault arcs.

In order to solve the above problems, the method for detecting a fault arc according to the present invention includes the following steps:

1) collecting current value of 220V/50Hz alternating current in one period at 3200Hz sampling frequency to obtain discrete 64 current values marked as I₀、I₁、…,I_k,…、I₆₃Forming a current waveform data set of the period;

2) by comparing the current value I of the current cycle₀、I₁、…,I_k,…、I₆₃And a normal period current value I0₀、I0₁、…,I0_k,…、I0₆₃To determine whether the current waveform is different from the waveform of the normal periodic currentThe specific calculation formula is as follows:

obtaining a parameter I representing the difference between the current period and the normal current period_per. Wherein I0₀、I0₁、…,I0_k,…、I0₆₃Representing 64 current values of a normal period as a reference for judging a fault period;

3) will I_perAnd threshold value β₁By comparison, if I_per＜β₁Then use I₀、I₁、…,I_k,…、I₆₃Median update I0₀、I0₁、…,I0_k,…、I0₆₃And proceeding to step 1); if I_per＞β₁Entering step 4);

4) calculating the average value of the current in the current period to judge whether the positive and negative half cycles of the current waveform in the period are symmetrical, wherein the specific calculation formula is as follows:

obtaining a parameter I for representing whether the positive and negative half cycles of the current are symmetrical_aver；

5) Calculating the maximum absolute value | I! of 64 current data in the current period_maxAnd maximum value | I0! of the absolute values of 64 current values in normal period_maxTaking C ═ I0 as non-woven fabrics_max-|I|_maxObtaining a parameter C representing whether the current peak value of the current period is reduced or not;

6) counting the number of values near 0 in 64 current values in the current period to represent whether the current in the period has a flat shoulder, wherein the specific calculation formula is as follows:

wherein,α is a coefficient less than 1

Obtaining a parameter N representing the number of values near 0 in the current value of the current period_zero；

7) Calculating the difference value of two adjacent current values in the current period, and finding out the maximum value to represent the maximum change rate of the current value in the period, wherein the specific calculation formula is as follows:

D＝max(di₁，di₂，…,di_k,…，di₆₃)

wherein di_k＝I_k-I_k-1

Obtaining a parameter D representing the maximum value of the current change rate

Calculating the difference value of two adjacent current values in the normal period, and finding out the maximum value to represent the maximum change rate of the current value in the normal period, wherein the specific expression is as follows:

D0＝max(d0i₁，d0i₂，…,d0i_k,…，d0i₆₃)

wherein, d0i_k＝I0_k-I0_k-1

Calculating D0-D to obtain a parameter Delta D representing whether the current cycle current waveform slope is suddenly changed;

8) will be parameter I_aver、C、N_zeroΔ D corresponds to the felling value β₂、β₃、β₄、β₅Comparing and judging I_aver、N_zeroWhether both parameters are greater than the corresponding thresholds β₂、β₄Such asIf yes, entering step 9);

if not, judging the period as a normal period and using I₀、I₁、…,I_k,…、I₆₃Median update I0₀、I0₁、…,I0_k,…、I0₆₃Then, entering the step 1) to judge the next period;

9) judging whether at least one parameter of C and delta D is larger than corresponding value β₃、β₅If yes, judging the period as a fault period, and entering the step 10);

if not, judging the period as a normal period, and then entering the step 1) to judge the next period;

10) repeating the step 1) to the step 9), and calculating the number Sum of fault cycles in 25 adjacent cycles;

11) judging whether the number Sum of fault cycles in the adjacent 25 cycles is more than or equal to 8, and if the Sum is more than or equal to 8, judging that the arc fault occurs; if Sum is less than 8, entering the step 1);

note that the upper cut value is β₁、β₂、β₃、β₄、β₅For a given value, the specific value can be further optimized and determined by experimental data.

The method for detecting the fault arc is based on the common characteristics of the arc, and judges whether the arc fault occurs by analyzing whether the current waveform has lost periodicity, asymmetrical positive and negative half cycles, flat shoulder and overlarge change rate or not by collecting current data of each cycle.

Drawings

FIG. 1 is a flow chart of a fault arc detection method of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the present invention will be further described in detail with reference to the accompanying drawings and embodiments.

As shown in fig. 1, the fault arc detection method of the present invention includes the following steps:

1. a method of fault arc detection comprising the steps of:

1) collecting current value of 220V/50Hz alternating current in one period at 3200Hz sampling frequency to obtain discrete 64 current values marked as I₀、I₁、…,I_k,…、I₆₃Forming a current waveform data set of the period;

2) by comparing the current value I of the current cycle₀、I₁、…,I_k,…、I₆₃And a normal period current value I0₀、I0₁、…,I0_k,…、I0₆₃Whether the current waveform is different from the waveform of the current in the normal period is judged, and the specific calculation formula is as follows:

obtaining a parameter I representing the difference between the current period and the normal current period_per. Wherein I0₀、I0₁、…,I0_k,…、I0₆₃Representing 64 current values of a normal period as a reference for judging a fault period;

3) will I_perAnd threshold value β₁By comparison, if I_per＜β₁Then use I₀、I₁、…,I_k,…、I₆₃Median update I0₀、I0₁、…,I0_k,…、I0₆₃And enter intoStep 1); if I_per＞β₁Entering step 4);

4) calculating the average value of the current in the current period to judge whether the positive and negative half cycles of the current waveform in the period are symmetrical, wherein the specific calculation formula is as follows:

obtaining a parameter I for representing whether the positive and negative half cycles of the current are symmetrical_aver；

5) Calculating the maximum absolute value | I! of 64 current data in the current period_maxAnd maximum value | I0! of the absolute values of 64 current values in normal period_maxTaking C ═ I0 as non-woven fabrics_max-|I|_maxObtaining a parameter C representing whether the current peak value of the current period is reduced or not;

6) counting the number of values near 0 in 64 current values in the current period to represent whether the current in the period has a flat shoulder, wherein the specific calculation formula is as follows:

wherein,α is a coefficient less than 1

Obtaining a parameter N representing the number of values near 0 in the current value of the current period_zero；

7) Calculating the difference value of two adjacent current values in the current period, and finding out the maximum value to represent the maximum change rate of the current value in the period, wherein the specific calculation formula is as follows:

D＝max(di₁，di₂，…,di_k,…，di₆₃)

wherein di_k＝I_k-I_k-1

Obtaining a parameter D representing the maximum value of the current change rate

Calculating the difference value of two adjacent current values in the normal period, and finding out the maximum value to represent the maximum change rate of the current value in the normal period, wherein the specific expression is as follows:

D0＝max(d0i₁，d0i₂，…,d0i_k,…，d0i₆₃)

wherein, d0i_k＝I0_k-I0_k-1

Calculating D0-D to obtain a parameter Delta D representing whether the current cycle current waveform slope is suddenly changed;

8) will be parameter I_aver、C、N_zeroΔ D corresponds to the felling value β₂、β₃、β₄、β₅Comparing and judging I_aver、N_zeroWhether both parameters are greater than the corresponding thresholds β₂、β₄If yes, entering step 9);

if not, judging the period as a normal period and using I₀、I₁、…,I_k,…、I₆₃Median update I0₀、I0₁、…,I0_k,…、I0₆₃Then, entering the step 1) to judge the next period;

9) judging whether at least one parameter of C and delta D is larger than corresponding value β₃、β₅If yes, judging the period as a fault period, and entering the step 10);

if not, judging the period as a normal period, and then entering the step 1) to judge the next period;

10) repeating the step 1) to the step 9), and calculating the number Sum of fault cycles in 25 adjacent cycles;

11) judging whether the number Sum of fault cycles in the adjacent 25 cycles is more than or equal to 8, and if the Sum is more than or equal to 8, judging that the arc fault occurs; if Sum is less than 8, entering the step 1);

note that the upper cut value is β₁、β₂、β₃、β₄、β₅For a given value, the specific value can be further optimized and determined by experimental data.

The method for detecting the fault arc is based on the common characteristics of the arc, and judges whether the arc fault occurs by analyzing whether the current waveform has lost periodicity, asymmetrical positive and negative half cycles, flat shoulder and overlarge change rate or not by collecting current data of each cycle.

Claims (1)

1. A method of fault arc detection comprising the steps of:

1) collecting current value of 220V/50Hz alternating current in one period at 3200Hz sampling frequency to obtain discrete 64 current values marked as I₀、I₁、…,I_k,…、I₆₃Forming a current waveform data set of the period;

2) by comparing the current value I of the current cycle₀、I₁、…,I_k,…、I₆₃And a normal period current value I0₀、I0₁、…,I0_k,…、I0₆₃Whether the current waveform is different from the waveform of the current in the normal period is judged, and the specific calculation formula is as follows:

obtaining a parameter I representing the difference between the current period and the normal current period_perWherein I0₀、I0₁、…,I0_k,…、I0₆₃Representing 64 current values of a normal period as a reference for judging a fault period;

3) will I_perAnd threshold value β₁By comparison, if I_per＜β₁Then use I₀、I₁、…,I_k,…、I₆₃Median update I0₀、I0₁、…,I0_k,…、I0₆₃And proceeding to step 1); if I_per＞β₁Entering step 4);

4) calculating the average value of the current in the current period to judge whether the positive and negative half cycles of the current waveform in the period are symmetrical, wherein the specific calculation formula is as follows:

obtaining a parameter I for representing whether the positive and negative half cycles of the current are symmetrical_aver；

5) Calculating the maximum absolute value | I! of 64 current data in the current period_maxAnd maximum value | I0! of the absolute values of 64 current values in normal period_maxTaking C ═ I0 as non-woven fabrics_max-|I|_maxObtaining a parameter C representing whether the current peak value of the current period is reduced or not;

6) counting the number of values near 0 in 64 current values in the current period to represent whether the current in the period has a flat shoulder, wherein the specific calculation formula is as follows:

wherein,α is a coefficient of less than 1

Obtaining a parameter N representing the number of values near 0 in the current value of the current period_zero；

7) Calculating the difference value of two adjacent current values in the current period, and finding out the maximum value to represent the maximum change rate of the current value in the period, wherein the specific calculation formula is as follows:

D＝max(di₁，di₂，…,di_k,…，di₆₃)

wherein di_k＝I_k-I_k-1

Obtaining a parameter D representing the maximum value of the current change rate

Calculating the difference value of two adjacent current values in the normal period, and finding out the maximum value to represent the maximum change rate of the current value in the normal period, wherein the specific expression is as follows:

D0＝max(d0i₁，d0i₂，…,d0i_k,…，d0i₆₃)

wherein, d0i_k＝I0_k-I0_k-1

Calculating D0-D to obtain a parameter Delta D representing whether the current cycle current waveform slope is suddenly changed;

8) will be parameter I_aver、C、N_zeroΔ D corresponds to the felling value β₂、β₃、β₄、β₅Comparing and judging I_aver、N_zeroWhether both parameters are greater than the corresponding thresholds β₂、β₄If yes, entering step 9);

if not, judging the period as a normal period and using I₀、I₁、…,I_k,…、I₆₃Median update I0₀、I0₁、…,I0_k,…、I0₆₃Then, entering the step 1) to judge the next period;

9) judging whether at least one parameter of C and delta D is larger than corresponding value β₃、β₅If yes, judging the period as a fault period, and entering the step 10);

if not, judging the period as a normal period, and then entering the step 1) to judge the next period;

10) repeating the step 1) to the step 9), and calculating the number Sum of fault cycles in 25 adjacent cycles;

11) judging whether the number Sum of fault cycles in the adjacent 25 cycles is more than or equal to 8, and if the Sum is more than or equal to 8, judging that the arc fault occurs; if Sum is less than 8, entering the step 1);

note that the upper cut value is β₁、β₂、β₃、β₄、β₅For a given value, the specific value can be further optimized and determined by experimental data.