HK40032724A - Arc protection device and operating method thereof - Google Patents

Description

Arc protection device and method of operating the same

The invention relates to the field of fire safety and electrical engineering, namely to a method and a device for detecting arcing in electrical networks and electrical installations, in order to prevent fires, said method and device complying with the standard according to IEC 62606: 2013 standard: general requirements for arc fault detection devices and the GOST IEC 62606-: arc Fault Detection Devices for house and Similar applications, general Requirements or other technical Requirements complying with the Requirements of Devices suitable for protecting electrical networks and electrical installations from the effects of arcs (Arc protection Devices, hereinafter also referred to as AFDD).

An arc protection device (AFDD) is a device comprising an arc fault detection unit (also arc action detection unit, spark detection unit) and a circuit interruption unit (also disabling means, circuit breaker). AFDD is intended to operate in difficult electromagnetic environments and should comply with most of the conflicting requirements regarding sensitivity, noise immunity, and absence of false activations that take the device out of normal operation. Therefore, in order for an AFDD to operate successfully, it is desirable to collect, if possible, general information about the processing that takes place in the protected circuit.

Thus, U.S. patent application No.20160187409 describes an AFDD that detects and analyzes current pulses at a low frequency band, voltage pulses at a low frequency band, and the intensity of high frequency components of the current. When it is detected that the current in the low frequency band rises above a predetermined threshold level, the analysis process is initiated. In the final phase of the analysis, the high-frequency current oscillations generated by the arc-specific broadband fluctuations are examined in several frequency subbands. However, these fluctuations are rather weak as characteristic features of arc faults, which poses certain problems for developing AFDD with large rated current values. A large rated current value corresponds on average to a wider service area covering a larger number of connected electrical devices and to a protected line having a longer length and more branches, which may result in a significant reduction in the transmission of high frequency components of the current from the location where the arcing event occurred to the AFDD. As a result, the signal separated from these fluctuations by the band-pass filter may be reduced to the level of background noise present in the protected line at that time. Therefore, the above fluctuations were used in Siemens 5SM6 AFDD to detect arcing (see, e.g., https:// support.industry.siemens.com/cs/document/109744645/brochure). At high background noise levels, the AFDD simply changes to a functionally limited mode, thereby notifying the user of the corresponding indication.

Thus, in large service areas, the detection of wide-band current fluctuations used in the above-mentioned and other known AFDDs can be difficult due to the small amplitude of the analyzed signal.

The current pulses selected by the high frequency filter at the time of an arc fault have a significantly larger amplitude under otherwise identical conditions. For example, U.S. patent No.5280404 or russian patent No.2528137 disclose the use of these pulses to detect arcing.

The AFDD and method for detecting arcing according to russian patent No.2528137 may be considered the closest analog of the subject matter of the present invention.

Solutions based on this principle, including the closest analogs, react and trigger even under typical low power high frequency events, such as when lighting a light emitting tube unit with an electromagnetic start control device.

It is an object of the present invention to obviate the above limitations and disadvantages of known arc protection devices and methods of operating the same.

The invention has the technical effects that: increasing the accuracy of arcing event detection; reducing the number of false activations of the arc protection device; and increasing the working range, the number of connected electrical devices, the length of the protected circuit and the number of branches it contains.

The above objects are achieved and the technical result claimed is achieved according to the proposed method of operating an arc protection device, wherein a voltage signal and a current signal of a protected circuit are measured in each voltage half-cycle; determining the presence of a separated arc in each voltage half-cycle based on the measured deviations of the voltage and current; information about the separated arc over a set period of time is collected and stored as arc action parameters. When the arcing event parameter reaches a preset value, an arc fault is detected and a signal for disconnecting the protected circuit is generated. Meanwhile, the process of measuring the voltage and the current includes: measuring a current at a high frequency band; measuring a current at an intermediate frequency band; measuring the voltage at the low frequency band and measuring the voltage at the intermediate frequency band.

The introduction of current measurements in the medium frequency band and voltage measurements in the medium frequency band into the conventional method of operating AFDD does not only simply increase the number of criteria for determining arcing in the protected circuit, but also enables monitoring of the occurrence of arcing in situations where the arcing signal is fully representative on the noise background and signal from normally operating equipment, not only normally operating equipment in the protected circuit but also in neighbouring circuits of the power grid. This makes it possible to reduce the number of false activations of the arc protection device. Thus, the claimed method ensures an extension of the service area covering a larger number of connected electrical apparatuses and increases the length of the protected circuit and the number of branches it contains, compared to the conventional operating method of AFDD.

In a preferred embodiment of the invention, the current in the intermediate frequency band is measured, the voltage in the intermediate frequency band is measured, and the voltage in the low frequency band is measured only in the case where the value of the current measured in the high frequency band, and preferably the current in the high frequency band measured in the sub-band from 1MHz to 10MHz at the part where the rise of the electrical network voltage module is detected, is greater than a predetermined level. This enables a reduction in the amount of "excess" information processed by the arc protection device, and thus reduces the requirements on the amount and speed at which information is processed in the device.

In determining the separated arc, the arcing current may be determined by measuring the voltage at a low frequency band and measuring the current at a medium frequency band. If the arcing current does not correspond to the preset value, the measurement of the current event is stopped, which also makes it possible to reduce the amount of excess information processed by the arc protection device and, in addition, to exclude a false activation of the arc protection device.

While determining the arc-action current, the polarity of the voltage signal in the intermediate frequency band is preferably determined by a voltage measurement of the intermediate frequency band, and the presence of a separated arc in the protected circuit is confirmed by the polarity of the voltage signal in the intermediate frequency band and the polarity of the current signal in the intermediate frequency band. This enables the arcing signal to be detected with high accuracy against a noise background in the protected circuit and adjacent circuits.

Moreover, the above task is solved and the technical effects claimed are achieved by the proposed arc protection device comprising: a power supply unit; at least one current sensor; at least one voltage reading unit; a high-frequency current signal detection unit; an intermediate frequency current signal detection unit; a microcontroller and a disconnecting member. The microcontroller is configured to: interacting with a high frequency current signal detection unit and an intermediate frequency current signal detection unit to measure a current in a high frequency band and a current in an intermediate frequency band, respectively; interacting with a voltage reading unit to measure a voltage at a low frequency band and a voltage at a middle frequency band; determining a detached arc fault event; collecting information about the arc over a set period of time and storing the information in the form of arcing parameters; and generating a disconnect signal when the arcing parameter reaches a preset value.

The claimed arc protection device is an embodiment of such a device implementing the claimed method of operating an arc protection device; and all the advantages of the claimed method are realized in the claimed apparatus.

Some possible embodiments of the invention are disclosed in detail below, without however limiting the invention, with reference to the attached drawings, in which:

fig. 1 shows an embodiment of an arc protection device and its connection to a circuit to be protected.

Fig. 2 shows a general schematic of operating an arc protection device according to the claimed method.

Fig. 3 shows a schematic diagram of the high-frequency start-up (HF start-up) of the measurement.

Fig. 4 shows a typical histogram of the voltage value distribution at the time of an arc fault.

Fig. 5 illustrates typical noise generated by the power supply unit of the portable computer and the light emitting diode lamp.

Fig. 6 shows a modulation curve of the high frequency voltage signal identification threshold.

Fig. 7 shows a schematic diagram of the detection of a detached arc at HF start-up.

Fig. 8 shows a waveform diagram of the current rise during an arc fault.

Fig. 9 shows an example of a signal from the power supply monitoring circuit.

Fig. 10 shows an example of a signal from an intermediate frequency voltage sensor.

Fig. 11 shows a schematic diagram of the conditions for detecting a disturbance by a sensor of the arc protection device.

Fig. 12 shows a polarity table for the intermediate frequency current signal and the intermediate frequency voltage sensor signal when switching the load.

Fig. 1 shows an embodiment of an arc protection device (or AFDD) (1) according to the invention and its connection to a circuit to be protected.

The AFDD (1) comprises a voltage reading unit (2), a current sensor (3), a high-frequency current signal detection unit (4), an intermediate-frequency current signal detection unit (5), a microcontroller (6), a power supply unit (7) and a disconnection member (8).

These components of the AFDD (1), except for the disconnecting member (8), constitute an arc fault detection unit.

The AFDD (1) may be connected to a line between, on the one hand, a power entry board (9) or another power supply unit or power supply network and, on the other hand, an electrical installation (10) of the circuit to be protected.

The voltage reading unit (2) comprises two voltage sensors: a low frequency voltage sensor and a medium frequency voltage sensor. The low-frequency voltage sensor is used together with a microcontroller (6) to detect and subsequently analyze the current value of the electrical network voltage at a considerable sampling rate, in particular 10kHz to 40kHz, preferably 20kHz to 30 kHz. The intermediate frequency voltage sensor is used together with a microcontroller (6) for detecting and subsequently analyzing voltage pulses in an intermediate frequency band substantially from 1kHz to 50kHz, preferably from 5kHz to 50 kHz.

In general, the low frequency voltage sensor and the medium frequency voltage sensor may be separate devices. The number of low-frequency voltage sensors and the number of intermediate-frequency voltage sensors and the number of voltage reading units (2) may vary depending on the task to be solved and may be defined, for example, by the number of circuits and/or phases to be protected.

The voltage reading unit (2) may have any conventional configuration and in the simplest case is a voltage divider for measurements in the low frequency band and a differentiating circuit for measurements in the mid frequency band.

The current sensor (3) is used to acquire a current signal from which the intermediate frequency current signal (the current measured in a band substantially from 0.1kHz to 30kHz, preferably from 0.3kHz to 15 kHz) and the high frequency current signal (the current measured in a band substantially from 1MHz to 10 MHz) are subsequently separated and analyzed by a high frequency current signal detection unit (4), an intermediate frequency current signal detection unit (5) and a microcontroller (6).

Depending on the task to be solved, several current sensors (3) can be used. For example, two current sensors (3) may be used, one of which is operated or combined in conjunction with the high frequency current signal detection unit (4) and the second of which is operated or combined in conjunction with the intermediate frequency current signal detection unit (5).

The current sensor (3) may be of any conventional construction and in the simplest case is a current transducer.

The microcontroller (6) is used to process the signals received from the voltage reading unit (2), the current sensor (3), the high-frequency current signal detection unit (4) and the medium-frequency current signal detection unit (5) and to determine the event of arcing occurring in the protected line and to generate a control signal for disconnecting the connecting member (8).

The power supply unit (7) supplies the microcontroller (6) with power and, if necessary, the disconnecting means (8).

The disconnecting means (8) disconnect the circuit supplying power to the electrical installation (10), i.e. disconnect the protected circuit from the electrical supply network, upon receiving a control signal from the microcontroller (6). Depending on the design of the device, the circuit can be opened not only in the path L of the phase current (as shown in the example in fig. 1) but also in the path N of the neutral current.

The arc protection device (1) operates as follows.

The microcontroller (6) analyses the signals received from the voltage reading unit (2) and the signals received from the current sensor (3) by the high-frequency current signal detection unit (4) and the medium-frequency current signal detection unit (5). The parameters required according to the standards of the GOST IEC 62606-.

In a first step, the microcontroller (6) determines the presence of a separated arc (IA) in the current half-cycle of the electric network voltage and estimates the parameters of the separated arc by analyzing and comparing the following signals:

1) a signal received from a low frequency voltage sensor (this signal is denoted LF-V in fig. 2) to determine the current electrical network voltage;

2) a signal received from the current sensor (2) in a band from approximately 1MHz to 10MHz, i.e. in a high frequency band (this signal is denoted HF-I in fig. 2);

3) a signal received from the current sensor (2) in a band, i.e. in the middle frequency band, of approximately from 0.1kHz to 20kHz, preferably from 0.3kHz to 15kHz (this signal is denoted MF-I in fig. 2);

4) the signal received from the intermediate frequency voltage sensor at a frequency of approximately from 1kHz to 50kHz, preferably from 5kHz to 50kHz (this signal is denoted MF-V in fig. 2).

After receiving and analysing these signals, the microcontroller (6) determines whether a detached arc (IA) is present and if the answer is that a detached arc is present, then in a second step of AFDD operation, the sequence of confirmed IAs is analysed for the purpose of determining arcing effects. If arcing is confirmed, the microcontroller (6) generates a signal for the disconnecting means (8) to disconnect the protected circuit from the electrical network.

Each step is described in more detail below starting with the first step.

The phase of the zero crossing of the electrical network voltage is determined from the signal from the low frequency voltage sensor, i.e. the signal LF-V.

Then, the time interval when a subsequent measurement is to be made is determined. These time intervals correspond to the rise region of the electrical network voltage, i.e. in the angular sectors of the voltage sinusoid of 0-90 ° and 180-. Repeated arc faults are highly unlikely to occur in the fall region of the electrical network voltage module, and therefore it is not reasonable to measure any type of accidental or regular noise in this region.

The current signal from the current sensor (2) is then subjected to frequency filtering by a high frequency current signal detection unit (4) to remove frequency components in the band up to approximately 1MHz, thereby forming an HF-I signal, which is then amplitude discriminated by a comparator (not shown in the figure) which may for example form part of the microcontroller (6) or be manufactured as a separate unit.

The purpose of this frequency filtering is to substantially separate the arcing signal, which has a large intensity high frequency component, from the noise signal generated by most electrical equipment. If the current signal obtained after said frequency filtering, i.e. the HF-I signal, exceeds a preset value Ipv, it will initiate a procedure for making subsequent measurements, in particular the measurement MF-I, MF-V, LF-V, which throughout this application is called "high frequency start or HF-start". Furthermore, as already discussed, HF start-up is only allowed in the rise region of the electrical network voltage module.

The HF start-up is schematically shown in fig. 3.

The amplitude discrimination threshold for the high frequency signal, which is determined by the preset value Ipv, is set based on a compromise between the detection of the arcing signal and the stability of AFDD (1) with respect to noise signals generated by a normally operating electrical installation (10). However, the amplitude of the arcing signal and the amplitude of the noise signal depend on the current voltage value and therefore on the phase of the current voltage value relative to the zero-crossing electrical network voltage. Under otherwise identical conditions, the magnitude of the arcing current signal at the high and mid frequency bands generated by arcing is directly proportional to the voltage at which an arc fault occurs within a given half-cycle of the network.

For example, fig. 4 shows that when using a cable according to IEC standard 62606: 2013, item 9.9.2.7, wherein the arc current presence value is equal to approximately 7A. Here, the voltage range in volts is represented on the X-axis and the number of counts is represented on the Y-axis. It can be seen in fig. 2 that approximately half of arc faults occur at voltages below 100V, and that this distribution behavior depends on factors that are difficult to predict. A low voltage arc fault may form a highly dominant part while having a small signal amplitude, which makes its detection difficult. Meanwhile, the level of most of the noise sharply drops in a range of voltage values lower than 100V. For example, fig. 5 shows the noise phase at the output of the current sensor (3), which is generated by a number of typical sources with respect to the network voltage, in particular by a free-running power supply unit (fig. 5a) and by a power supply unit running under load (fig. 5b) and by an LED lamp (fig. 5 c).

For this reason, in order to optimize the ratio of arcing signal/noise signal in each half-cycle of the network, it is reasonable to modulate the discrimination level of the high-frequency current signal (i.e. the HF-start threshold) to ensure that it is proportional to the network voltage within the range of network voltage values of 30V and above. At lower values of the network voltage, this level is kept constant to avoid excessive increase in the sensitivity of AFDD (1) to noise. The modulation behavior in the positive half-cycle of the network is shown in fig. 6, which shows the modulation curve for the threshold Uthr of the detected high frequency current signal (HF-I). This modulation can be performed as a modulation synchronized with the network voltage of the reference voltage potential of a comparator identifying the high frequency signal at the output of the high frequency current signal detection unit (4), which signal is proportional to the pulses of the current sensor (3).

Thus, HF-actuation is used to initiate measurement of the MF-I, MF-V, LF-V signal to determine and confirm the occurrence of IA, which is schematically illustrated in FIG. 7.

The purpose of processing the signal in the middle frequency band (signal MF-I in fig. 7) from the current sensor (3) and the signal from the low frequency voltage sensor (signal LF-V in fig. 7) after HF-start is to estimate the arc current value based on the conductivity of the arcing zone of the protected circuit. The amplitude of the high frequency current signal HF-I cannot reliably determine the arc current because it depends strongly on the a priori unknown configuration of the circuit. Meanwhile, for example, IEC standard 62606: 2013 clearly define a low level sensitivity of AFDD to arcing current, i.e. 2.5A for an electrical network voltage of 230V, as studies show that the probability of ignition at lower arcing current values is small. However, reaction to low current disturbances generated by multiple electrical facilities (10) will result in false enablement of the AFDD (1), thus introducing conductivity identification of the circuit area where the network is shut down at a given time into the arc fault detection algorithm.

This task can be solved as follows. In parallel with the processing of information in a high frequency band received from the current sensor (3), the AFDD (1) performs processing on a current signal at a significantly lower frequency by a circuit implemented using an intermediate frequency current signal detection unit (5). Hereinafter, this circuit is referred to as a Power Monitoring Circuit (PMC), and the PMC output signal is referred to as a PMC signal. The conditions for using PMC in AFDD (1) lie in the following facts: when the transient processing period determined by the reactive parameters of the circuit expires, the intermediate frequency current signal MF-I is horizontally stable with respect to its rise in value before the event occurs and is equal to the quotient obtained by dividing the current network voltage by the effective resistance of the arcing zone. Fig. 8 shows an example of a current waveform diagram at the time of an arc fault in the case of the same present network voltage for loads with different resistances in the circuit, the reactive components of the impedances being 540Ohm (fig. 8a), 147Ohm (fig. 8b) and 104Ohm (fig. 8c), wherein the upper curve shows the rise of the intermediate frequency current signal MF-I and the lower curve shows the HF-start signal. At the same time, it appears clearly that the rise of the intermediate frequency current signal (in its stable region) is directly proportional to the electrical conductance of the arcing circuit region.

The result is that the electrical conductivity of the arcing zone can be determined by dividing the rising value of the intermediate frequency current signal MF-I by the present network voltage value LF-V, which is read continuously by the microcontroller (6) from the voltage reading unit (2) at a comparatively high rate, in particular from 10kHz to 40kHz, more preferably from 20kHz to 30 kHz. In order to select the frequency band, time interval, and algorithm for analyzing the PMC signal, the following requirements should be considered:

1) the PMC signal is formed in an intermediate frequency current signal detection unit (5) by frequency filtering the signal coming from the same current sensor (3) from which the high frequency current signal is received, or from any type of additional current sensor;

2) the process of analyzing the PMC signal can only be invoked by HF-start;

3) the frequency band used for analyzing the PMC signal should be rather low (and correspondingly, the analysis time should be rather long) so that the amplitude of the PMC signal does not depend on the reactive component of the protected circuit and is defined only by the active component of the impedance of the arc-action region;

4) the analysis time should be fairly small compared to the network voltage half-cycle (and correspondingly the frequency band used for analysis should be fairly wide) so that the analysis of the PMC signal after a possible low current high frequency pulse does not take too much time, since otherwise the reaction of AFDD (1) to other events within a given half-cycle may be prevented.

The conflict between the above-mentioned requirement 4 for the analysis time and the requirement 3 for the frequency band for analyzing the PMC signal necessitates additional experimental studies to be performed. As a result, an optimal frequency band for the PMC signal is determined that ranges from approximately 0.1kHz to approximately 20kHz, most preferably from 0.3kHz to 15kHz, and the time interval for analyzing the response to a single high frequency disturbance is 5 μ s to 50-100 μ s post-disturbance.

The circuit for forming the PMC signal may be, for example, an oscillation circuit in the intermediate frequency current signal detection unit (5). Alternatively, all necessary frequency filtering may be performed by the microcontroller (6). Other embodiments of PMC are also possible.

For example, the following can be considered: wherein the circuit for forming the PMC signal is the above-described oscillating circuit having a resonant frequency of 10kHz and a Q factor of slightly less than 0.5. The PMC signal at the capacitor of such a circuit has the form shown in fig. 9; and the quality characteristic of the current rise, which is practically insensitive to the value and arrangement of the reactive components of the protected circuit, is the average value of this pulse in the interval from 5 to approximately 70 mus from the HF-start.

Dividing the current rise value calculated at the time of receiving the PMC signal by the present network voltage value at the time of high frequency interference enables the value of the active conductivity component of the possible arcing region to be determined. The current consumption of the possible arcing region is determined from the obtained values; and in this way compares the determined current consumption with a threshold value Ithr set at a certain reserve below the specification of AFDD (1) (for standard IEC 62606: 2013, this value is 2.5A at a network voltage of 230V). If the value Iarc of the rise in arc current corresponding to the arcing event to the resistance exceeds the threshold value Ithr, i.e., in the case of Iarc < Ithr, the event is ignored; and the microcontroller (6) returns to the ready state to analyse the next event in a given half cycle of the network (see figure 7).

After HF-start-up, and simultaneously with the analysis of the PMC signal described above, the MF-V signal from the intermediate frequency voltage sensor is processed (see fig. 7). When the voltage pulse signal is formed, for example, by double differentiation of the network voltage at the AFDD (1) terminal with a time constant of approximately 40 μ s for both differential sections, the voltage pulse signal in any configuration of the circuit has a clearly defined polarity within a time interval of 5 μ s to 15 μ s from the time of HF-start. For example, fig. 10a and 10b show the MF-V signal from the if sensor at two different points of the extension and branch circuits at similar current rises. It can be seen that these signals have a clearly defined same polarity during the above-mentioned time interval.

The purpose of processing the MF-V signal from the intermediate frequency voltage sensor is to determine its polarity to identify the location of the arcing source and the sign of the interference within the protected circuit. This is necessary to avoid the influence of processing occurring in circuits having other phases or the same phase outside the protected circuit. The strength of the induced high-frequency signal from the outside can be quite high so as to invoke HF start-up, which is particularly dangerous near the zero-crossing phase of the protected circuit, because the HF start-up threshold at that time is low and the coefficient for converting the PMC signal into the conductivity of the arc region is high, while the processing of adjacent phases in the circuit can take place at the maximum voltage value and for this reason can have a great strength at that time.

This problem can be solved by comparing the polarity of the PMC signal, i.e. the intermediate frequency current signal MF-I, with the polarity of the signal MF-V from the intermediate frequency voltage sensor.

Let us consider a polarity diagram for the MF-I and MF-V signals from the MF-V voltage sensors when connecting and disconnecting the load in the protected circuit and the adjacent circuit. The conditional scheme for interference detection may be presented in the form shown in fig. 11. Here, V and i are the voltage and current sensors of the AFDD, respectively; z is the switched load in the protected circuit; z1 is the load in the adjacent circuit connected to the same phase L1 as the protected circuit; z2 is the load in the other phase L2 which has the same polarity as phase L1 at the time of interference; z3 is the polar phase of the phase L1 and the polarity of the phase in interferenceLoad of the third, opposite phase L3; z_NIs the impedance of the neutral line N. The local connection point N of the load neutral line and the initial neutral voltage generation point N are connected in the circuit₀Non-zero impedance Z between_NThe intermediate frequency components of the switching process, including the arc fault, conductively affect the voltage sensor V and the current sensor i. The switching of loads Z1, Z2, Z3 occurs at point N relative to N₀And thus the potential with respect to the phase potential rises, thereby generating voltage and current rises in the voltage sensor V and the current sensor i.

Fig. 12 shows a polarity table of the pulses of the intermediate frequency current signal MF-I (or PMC signal) and the MF-V signal from the intermediate frequency voltage sensor in connecting and disconnecting the load in the protected circuit and the adjacent circuit in the positive network half-cycle according to the scheme shown in fig. 11. When the current in the protected circuit increases, the pulse signal polarity of the intermediate frequency current signal is taken to be positive; and all polarities will simply change sign within a negative half cycle.

As can be seen from fig. 12, the combination of the polarity of the intermediate frequency current signal and the polarity of the signal from the intermediate frequency voltage sensor, which is not repeated during another switching, corresponds to the connection of the load Z which occurs in particular in the event of an arc fault in the protected circuit. Thus, simultaneously with the analysis of the PMC response after HF-start, the corresponding signal from the intermediate frequency voltage sensor is directed to confirm its presence and polarity. In the simplest variant, this can be done by a separate comparator or alternatively by a microcontroller (6). If there is no such confirmation within the next 5 μ s to 15 μ s, the PMC signal analysis is stopped and the microcontroller (6) returns to a ready state to analyze the next event.

If all the above criteria are met, it is deemed that a separated arc (IA) is identified in the protected circuit in this half-cycle (see fig. 7) and AFDD (1) is transferred to the second analysis step.

In a second step, AFDD is transferred from the identification of IA to the identification of the arcing itself. This step can be achieved by various methods known from the above-mentioned patent documents, for example. In the following, a further variant of using the confirmed IA to determine arcing events is presented, which variant can be used with known methods to implement the present invention.

When the microcontroller (6) detects a separated arc in a given half-cycle, a certain number of fractions are added to a digital counter, which is for example an integral part of the microcontroller (6). If there is a non-zero sum of the fractions in the digital counter, then the preset sum of the fractions is subtracted therefrom in each half cycle. The values for adding and subtracting are chosen such that at a sufficient rate of recurrence of the half-cycle with respect to IA, the sum of the fractions reaches a preset limit at which the microcontroller (6) issues a command to the disconnecting means (8) to disconnect the protected circuit. Here, a sufficient recurrence rate of the half-cycles with respect to IA is understood to be a rate combined with the number of load arcing connections in the protected circuit, which is highly unlikely under practical operating conditions of a sparkless generating circuit. Also, the sum added to the digital counter may be made dependent on the value of the arcing current calculated during the half cycle, for example may increase as it increases for the purpose of shortening the device activation time at large arcing currents.

If an arcing event is identified, the microcontroller (6) issues a control signal for disconnecting the connecting means (8), and the disconnecting means (8) disconnects the protected electrical circuit of the electrical installation (10) from the electrical network.

The invention thus enables a more accurate and unambiguous determination of arcing events in a protected circuit, significantly reducing the number of false activations. Thus, the area of monitoring possible arcing in the protected circuit is extended, the number of electrical installations connected to the protected circuit is increased, the length of the protected circuit and the number of branches it contains are expanded.

Claims (7)

1. A method of operating an arc protection device, wherein a voltage signal and a current signal of a protected circuit are measured in each voltage half-cycle; determining the presence of a separated arc in each voltage half-cycle based on the measured voltage signal and the measured current signal; collecting information about the separated arc over a set period of time and storing the information in the form of parameters for detecting arcing events; when the parameter for detecting the arc action reaches a preset value, detecting the arc action; and generating a signal to disconnect the protected circuit from a network; wherein measuring the voltage signal and the current signal comprises: measuring a current signal at a high frequency band from 1MHz to 10 MHz; measuring a current signal at an intermediate frequency band from 0.1kHz to 20 kHz; measuring a voltage signal at a low frequency band having a frequency lower than a frequency of the intermediate frequency band; and measuring a voltage signal at a mid-band from 1kHz to 50 kHz.

2. The method of operating an arc protection device according to claim 1, wherein in a case where the measured current signal in the high frequency band is higher than a preset value, the current signal in the middle frequency band is measured, the voltage signal in the middle frequency band is measured, and the voltage signal in the low frequency band is measured.

3. Method of operating an arc protection device according to claim 2, wherein the current at the high frequency band is measured at a rise region of the network voltage module, which corresponds to the time interval when the subsequent measurement is to be made, and which lies in the angular sectors of the voltage sinusoid of 0-90 ° and 180-270 ° after the zero crossing of the electrical network voltage.

4. The method of operating an arc protection device of claim 1, wherein the arc current is determined by: measuring the voltage signal at a low frequency band, at a measurement rate of 10kHz to 40 kHz; and measuring the current signal at the intermediate frequency band.

5. The method of operating an arc protection device according to claim 4, wherein the polarity of the voltage signal at the intermediate frequency band is determined by measuring the voltage at the intermediate frequency band substantially simultaneously with determining the arc current.

6. An arc protection device comprising a power supply unit, at least one voltage reading unit, at least one current sensor, a high frequency current signal detection unit, an intermediate frequency current signal detection unit, a disconnection means configured to be disconnected from a network upon receiving a signal for disconnection, and a microcontroller configured to: interacting with the high-frequency current signal detection unit and the low-frequency current signal detection unit to measure and analyze a current signal at a high frequency band and a current signal at a medium frequency band, respectively; interacting with the voltage reading unit to measure voltages at a middle frequency band and voltages at a low frequency band; determining a separated arc; collecting information about the separated arc over a set period of time and storing said information in the form of parameters for detecting arcing events; and generating a disconnection signal when said parameter for detecting arcing reaches a preset value.

7. The arc protection device of claim 6 wherein the voltage reading unit comprises at least one low frequency voltage sensor and at least one intermediate frequency voltage sensor.