Method and circuit device for detecting electric arc and photovoltaic inverter having the circuit device

Abstract

The present application describes a method for detecting an arc in a DC, wherein the circuit has a DC load, a DC source for supplying power to the DC load, and a circuit device arranged between the DC source and the DC load. The power flow between the input terminal and the output terminal of the circuit device is suppressed by means of a switching unit through a cyclic interruption, so that the power flow is enabled in an active time window with a first duration and suppressed in a disabled time window with a second duration. By detecting an input current flowing at the input terminal and / or an input voltage applied at the input terminal and comparing the value of the input current detected in the disabled time window and / or the value of the input voltage with a current threshold or a voltage threshold, if the input current detected in the disabled time window is lower than the current threshold and / or the input voltage detected in the disabled time window does not exceed the voltage threshold, a signal about the presence criterion of an arc is issued. A circuit device for detecting an arc and a photovoltaic inverter having the circuit device are also described.

Description

[0001] Technical field of the invention

[0002] The invention relates to a method and a circuit arrangement for detecting an electric arc, and a photovoltaic (PV) inverter having such a circuit arrangement. Prior art

[0003] Electric arcs can cause non-negligible damage in electrical systems, which can, for example, lead to fires in the electrical system and in the building in which it is located. Injuries to people can also occur. Identifying electric arcs in electrical systems is therefore an important requirement. This is especially true when the electrical system has direct current (DC) circuits, because in DC circuits, unlike alternating current (AC) circuits, there are usually no voltage zero crossings that could extinguish (or at least facilitate extinguishing) a burning arc.

[0004] Two different types of arcs are usually distinguished. A series arc occurs when there is a contact fault, for example when a cable or plug is loose but the contacts are very close together, so that there is a local high electric field strength in the gap between the contacts. An example of a series arc is also a switching arc that occurs between the contacts of a relay during a switching process. Parallel arcs are less common and occur, for example, when the positive and negative wires of an electrical system are close to each other and their insulation is damaged. Here, there may also be local high field strengths between the damaged conductors. The power losses converted in a series arc are usually significantly smaller than those converted in a parallel arc. Therefore, a series arc between a power source and a load supplied by this power source causes a relatively small change in the transmitted power. In contrast, a parallel arc can lead to a large change in the electrical power provided by the power source and ultimately consumed by the load.

[0005] Conventional methods for detecting electric arcs use the noise spectrum that is usually emitted by the electric arc. Here, when there is an electric arc in the electrical system, a high-frequency interference voltage or a high-frequency interference current is superimposed on the useful voltage or the useful current. The high-frequency interference voltage and / or the high-frequency interference current can be electrically filtered and detected on an electrical conductor that simultaneously carries the useful current or the useful voltage. When a certain interference voltage and / or a certain interference current is exceeded, a signal of an electric arc is emitted. This method is described, for example, in WO 2012 / 116 722 A1. In order to increase the sensitivity of the detection, multiple different frequencies within the emitted noise spectrum can also be detected simultaneously.

[0006] Another known method for detecting an electric arc is to use the sudden changes in voltage and / or current that accompany the formation of an electric arc. In this case, the current and / or voltage are usually monitored with a high time resolution in order to detect the sudden changes. In this method, it is also possible to check, if necessary, to what extent the observed changes in current and / or voltage differ from normal fluctuations in the power consumption of the electrical load. If the observed changes in current and / or voltage differ significantly from normal consumption fluctuations of the system, an electric arc is signaled. This method is disclosed, for example, in document WO 2011 / 017721 A1.

[0007] Finally, document WO 2013 / 171329 A1 discloses a method for arc detection which combines the detection of noise signals with the detection of sudden changes in the current. In this case, the temporal occurrence of the detected noise signal and the temporal occurrence of the detected current change can each be provided with a time stamp. By comparing the two time stamps, an additional plausibility test for the presence of an arc can be determined.

[0008] All methods generally use relatively complex and therefore expensive measuring systems, which become increasingly complex the more reliably a possible arc is detected or the more reliably false signals of an arc and associated downtime of the electrical system are avoided.

[0009] From document DE 102 25 259 B3 an electrical plug connector is known which has a main contact which leads during the unplugging process and an auxiliary contact which lags in parallel with it. When unplugging the plug connector, the main contact inevitably separates first from its mating connector and then the auxiliary contact only lastly from its mating connector. In order to avoid arcs associated with the unplugging process, a semiconductor switch which is connected in series with the auxiliary contact is periodically switched on and off between the contact separations of the main contact and the auxiliary contact.

[0010] Invention Task

[0011] The object of the present invention is to provide a method for detecting arcs, in particular series arcs, in DC circuits, which method can be implemented relatively cost-effectively. In particular, complex detection systems should be avoided as much as possible, and relatively simple and cost-effective measuring techniques should be used. Furthermore, the object of the present invention is to provide a circuit arrangement suitable for implementing the method and a photovoltaic (PV) inverter having such a circuit arrangement.

[0012] Description of the invention

[0013] The method according to the present invention is intended to detect an arc in a direct current (DC) circuit, the DC circuit having a DC load, a DC source for powering the DC load, and a circuit device arranged between the DC source and the DC load. Here, the circuit device includes: an input end, which has two input terminals for connecting the DC source; an output end, which has two output terminals for connecting the DC load; and a switching unit arranged between the input end and the output end, and the circuit device has an input capacitance between the input terminals at the input end. The switching unit has at least two operating modes and can be operated in these two operating modes. Specifically, the switching unit is designed to enable a power flow P between the DC source and the output end of the circuit device in a first operating mode, and to inhibit a power flow between the DC source and the output end of the circuit device in a second operating mode. The method comprises the following steps:

[0014] - cyclically interrupting a power flow P between an input and an output of the circuit arrangement by means of the switching unit, such that the power flow P is enabled in an active time window having a first duration Δt1, i.e., where P≠0, and the power flow P is inhibited in a deactivated time window having a second duration Δt2, i.e., where P=0,

[0015] - Detect the input current I flowing at the input terminal ein and / or the input voltage U applied to the input ein ,

[0016] - Input current I to be detected during the disable time window ein and / or input voltage U ein The value of the current threshold I TH and / or voltage threshold U TH To compare, and

[0017] - If the input current I detected during the disable time window ein Below the current threshold I TH and / or the input voltage U detected in the disable time window ein The voltage threshold U is not exceeded TH , then the arc presence standard is issued signal.

[0018] The input current I flowing at the input terminal ein It is the current that flows from the DC source into the circuit device via one of the input terminals, or flows out of the circuit device again via another of the input terminals. During the active time window, the input current I einrepresents a current that characterizes the power flow P between the input and output of the circuit arrangement. However, during the disable time window, the power flow P between the DC source and the output of the circuit arrangement is interrupted by means of the switching unit. In this case, the input current I ein Flows only between the DC source and the input capacitor.

[0019] The input voltage U applied to the input terminal ein Located between the two input terminals. Since the input capacitor is also arranged between the input terminals, the input voltage U ein It also corresponds to the voltage at the input capacitor.

[0020] Signaling the presence criterion of an arc indicates that a criterion has been identified from which the presence of an arc can be inferred.

[0021] In particular, the arc to be detected may be a series arc. The circuit arrangement may be completely or partially a component of an electrical device, such as an energy conversion unit, such as an inverter. The DC load may be constructed as a DC load in the conventional sense and for this purpose only comprises a DC consumer. However, the DC load may also include a DC / AC converter, which is connected to an alternating voltage (AC) load on the output side. The AC load may be an electrical device powered by an alternating voltage. However, alternatively, the AC load may also be a power supply network, into which the DC / AC converter feeds the AC power converted from the DC power. The cyclic interruption of the power flow P refers to the repeated interruption of the power flow P in a manner that satisfies the rules, which interruption may be performed periodically, but need not be performed periodically. In the interruption that is performed periodically, the sum of the first duration Δt1 and the second duration Δt2 is constant, i.e. Δt1+Δt2=constant, while in the interruption that is performed non-periodically, the sum is not constant. If the power fluctuations consumed by the DC load can be classified as being small enough to be negligible, the power flow P may be interrupted periodically. If this cannot be assumed, the cyclic interruption of the power flow is usually performed aperiodically.

[0022] The present invention utilizes the following effect: the power flow P that is usually continuous between the DC source and the DC load in the DC circuit is converted into a discontinuous power flow P by the switching unit. This applies at least to the first region of the DC circuit, which is arranged between the DC source and the output of the circuit device. Here, in the first region of the DC circuit, that is, on the entire section between the DC source and the output of the circuit device, the power flow P can (but is not absolutely necessary) be interrupted. More precisely, it is sufficient to interrupt the cycle of the power flow only on the partial section between the DC source and the output of the circuit device. Specifically, for example, when the power flow is interrupted only on the partial section between the DC source and the output, the charging process of the capacitance that is still decaying there can be carried out in the remaining partial sections of the first region of the DC circuit. However, by suppressing the power flow on the partial section of the first region of the DC circuit, in response to this, the power flow on the remaining partial sections in the first region of the DC circuit is at least significantly reduced. In the second region of the DC circuit arranged between the output of the circuit device and the DC load, the power flow can be discontinuous or continuous.

[0023] Since there is no power flow from the DC source to the output of the circuit arrangement in the second disabled time window, a zero crossing of the current I is intentionally caused in the DC circuit, at least in the first region of the DC circuit, and at least on a section of the DC circuit extending from the DC source to the first region of the output of the circuit arrangement. In summary, this applies to the entire first region of the DC circuit between the DC source and the output of the circuit arrangement, if the power flow P is not completely interrupted there, then at least it is greatly reduced everywhere. In this case, energy is also simultaneously drawn from a possible series arc in the first region of the DC circuit. By drawing away the energy, the arc can no longer burn, but is extinguished (at least temporarily within the disabled time window). Now, when a power flow P from the DC source to the output of the circuit arrangement is generated again in the active time window after the disabled time window, the risk of the power flow P being accompanied by a burning arc is greatly reduced compared to the previous active time window in which the arc is still burning. In other words: if the arc burns in the active time window that is earlier in the two active time windows, the arc is most likely no longer present in the active time window that is later in the two active time windows. In this case, the probability increases with increasing second duration of the deactivated time window or of a deactivated time window between two activated time windows.

[0024] In the disable time window, the DC load can be supplied by an energy buffer (e.g., a capacitor) already present at the input of the DC load or by an output capacitor which is additionally provided and connected in parallel with the output of the circuit arrangement. In addition, the second duration Δt2 of the disable time window can be selected to be very small, so that the capacitor or the output capacitor is only negligibly discharged, so that insufficient supply of the DC load and possible associated malfunctioning of the DC load can be avoided to a large extent.

[0025] In another embodiment, the method according to the present invention further comprises the following steps:

[0026] - detecting a current I characteristic of the power flow P and / or a voltage U characteristic of the power flow P in two consecutive active time windows,

[0027] - comparing the value of the current I and / or the voltage U of the detected active time window with the corresponding value of the detected previous active time window, and

[0028] If the value of the current I and / or the voltage U of an active time window differs from the corresponding value of the preceding active time window by more than a threshold value, a signal is issued regarding the presence of a further arc.

[0029] The current I to be detected, which characterizes the power flow P, may be the current flowing between the DC source and the input of the circuit arrangement. Alternatively, it may be the current flowing between the input of the circuit arrangement and the switching unit or the current flowing between the switching unit and the output of the circuit arrangement. The detected voltage U may be the voltage applied to the input of the circuit arrangement (i.e., the input voltage U ein ) and / or a voltage present at the output of the circuit arrangement. Within the scope of the invention, only one current characterizing the power flow P between the DC source and the output is detected, but alternatively also multiple currents may be detected. Alternatively or additionally to this, only one voltage characterizing the power flow P between the DC source and the output of the circuit arrangement may be detected, but optionally multiple voltages may also be detected.

[0030] In another embodiment of the method, the evaluation of the voltage and the current is performed in two consecutive active time windows, and in each of the two temporally offset active time windows, the current I and / or the voltage U associated with the power flow P in the corresponding active time window are respectively detected. Since the power loss due to the arc is also associated with the arc, the electrical variables of the current I and / or the voltage U change according to the presence of the arc in the DC circuit. Therefore, it can be inferred from the comparison of the values ​​of the detected current I and / or the voltage U of the two temporally offset active time windows whether one of the two active time windows has an arc compared to the corresponding other of the two active time windows. This is particularly applicable to the case where the consumption of the DC load over time is known for other reasons and / or does not change in time, at least not significantly. In particular, if the value of the current I and / or the voltage U of the temporally subsequent active time window differs from the corresponding value of the current I and / or the voltage of the previous active time window by more than a predetermined threshold value, a signal about the presence of an arc can be issued.

[0031] In the method according to the invention, no special requirements are imposed on the measuring unit for detecting electrical variables from the current I and / or the voltage U. The measuring unit does not have to have high accuracy nor high time resolution, i.e. high measurement frequency. This is because it is not important to detect sudden changes in the electrical variables. More precisely, for example, in another embodiment of the method, it is sufficient to detect only the value of the electrical variable representing the corresponding active time window, such as the average value of the corresponding active time window. For this reason, the measuring unit for detecting the electrical variable can be constructed very low-cost. Usually, the measuring unit already exists in the DC circuit, which is why no additional costs are incurred in this case. Because according to the invention, even in the normal operation of the DC load, the power flow P between the DC source and the output of the circuit device in the DC circuit also occurs discontinuously, basically in each energy packet (Energiepaket), so there is no need for a complex detection unit to trigger a one-time interruption of the continuously occurring power flow. More precisely, the method according to the invention operates as if in each active time window, at least in the first region of the DC circuit there is an arc, in particular a series arc, which must be extinguished in the disabled time window by producing an interruption of the power flow P from the DC source by means of the switching unit. The information about whether an arc was actually present in the preceding active time window is derived only from the detection of electrical variables of the current I and / or voltage U in the disabled time window or in a temporally subsequent active time window and the comparison thereof with a threshold value or with a correspondingly detected variable of the current I and / or voltage U of the temporally preceding active time window.

[0032] In an advantageous embodiment of the method, the switching unit can, in response to the signaling of the arc presence criterion and / or the signaling of the further arc presence criterion, permanently interrupt the power flow P between the DC source and the output of the circuit arrangement, at least for a period of time that is greater than the second duration Δt2 of the disable time window. For example, such interruption in response to the signaling of the arc presence criterion and / or the signaling of the further arc presence criterion can last for several minutes or until manually confirmed. In this way, a specialist has the opportunity to identify and repair possible damage in the DC circuit caused by a detected arc.

[0033] Thus, the arc presence criteria may alternatively or additionally be considered in deciding whether an arc is present (in which case the arc itself has been extinguished in the disable window). In any case, the accuracy of arc detection is improved when multiple arc presence criteria are considered compared to the accuracy of arc detection when only a single arc presence criterion is considered. Depending on the choice of alternative or additional considerations, improvements are achieved in avoiding false positive or false negative identifications. The arc presence criteria determined by evaluation in the disable time window may also be used to improve identification in the active time window, for example by adjusting a threshold value used for evaluation in the active time window based on the presence of the arc presence criterion.

[0034] The active time window and the previous active time window, whose values ​​of the current I and / or the voltage U are compared with one another, do not necessarily necessarily follow one another directly. Rather, they can be separated from one another by a plurality of disabled time windows and at least one further active time window. However, in an advantageous embodiment of the method, the active time window and the previous active time window can also be separated from one another by exactly one disabled time window. In this way, changes in the consumption of the DC load that occur during this period are less important when comparing the electrical variables.

[0035] In an advantageous embodiment of the method, the value of the second duration Δt2 of the disable time window can be selected so that it is not only high enough to extinguish an arc that may be present in the DC circuit, but more precisely high enough to suppress the previously extinguished arc from reigniting in an active time window that may follow the disable time window. In particular, the value of the second duration can depend on a thermal time constant associated with the arc and should in particular be greater than the corresponding thermal time constant. The thermal time constant describes the decrease in the conductivity of the arc column, which is constructed as a plasma, over time when no more electrical energy is supplied to the arc, for example because the power flow in the first region of the DC circuit is interrupted by means of a switching unit. In particular, if the time for which the voltage U is applied to the arc column again after the arc has extinguished is greater than the thermal time constant, reignition of the arc does not occur. In addition, the value of the minimum necessary interruption time of the current flow can also depend on the power converted in the burning arc. Therefore, the value of the second duration Δt2 of the disable time window can also be selected as a function of the values ​​of the current I characterizing the power flow P and / or the voltage U characterizing the power flow P detected in the immediately preceding active time window. In particular, the value of the second time period Δt2 can increase with increasing values ​​of the detected current I and / or detected voltage U of the preceding time window. Experiments have shown that it is sufficient if the value of the second time period Δt2 is between 0.1 ms and 10.0 ms, preferably between 0.3 ms and 4.0 ms.

[0036] The input current I detected during the disable time window can be ein The value of and the input voltage U detected in the disable time window ein The second duration Δt2 of the disabled time window is determined by the value of . Specifically, for example, the second duration Δt2 of the disabled time window is defined in the following manner: the detected input current I ein Below another current threshold I TH2 and / or detected input voltage U ein Exceeding another voltage threshold U TH2 Here, the other current threshold I TH2 is selected so, in particular so low, that if a series arc is present between the DC source and the input of the circuit arrangement, it cannot continue to burn. TH2 This ensures that any series arc that may exist between the DC source and the input terminal is reliably extinguished.

[0037] Of course, the input current I einThis can be detected by a corresponding current sensor. However, this requires an additional current sensor in addition to the voltage sensor that is usually present. However, in the disabled time window, the input current I flowing between the DC source and the input terminal (especially the input capacitor) ein It can also be applied to the input capacitor by the input voltage U ein Specifically, the input voltage U dropped across the input capacitor is measured over time. ein The time derivative of provides a measure of the input current flowing between the DC source and the circuit input. In this way, the input voltage U can be measured using only one voltage sensor. ein and input current.

[0038] In another variant of the method, the switching unit of the circuit arrangement is formed by a DC / DC converter, in particular a step-up converter. In this embodiment, the circuit arrangement also has an output capacitor connected in parallel with its output. Advantageously, in the active time window, the output voltage U applied to the output capacitor can be amplified by the switching unit near the end of the first duration Δt1. aus For example, when at least 50%, preferably at least 75%, particularly preferably at least 90% of the first duration Δt1 of the active time window is reached or exceeded, the output voltage U applied to the output capacitor aus Here, at the beginning of the active time window, the output voltage U applied to the output capacitor aus In any case, in a DC / DC converter with a freewheeling diode, the output voltage U applied to the output capacitor aus It should be at least or exceed the value of the no-load voltage of the DC source. Only in this way can it be ensured that the switching unit together with the output voltage U applied on the output side aus The power flow between the input and output of the circuit device is suppressed together. Additionally, the energy cached in the output capacitor and in particular also in the adjacent disabled time window for supplying power to the DC load can be increased in this way. The upper limit of the second duration Δt2 of the adjacent disabled time window can be increased, and the risk of incorrectly supplying power to the DC load, in particular during the disabled time window, can be avoided or at least reduced. On the other hand, the loss of the DC / DC converter for caching energy only occurs at the end of the active time window, that is, only during a small part of the first duration Δt1, rather than during the entire first time window. Additionally, the component providing the output capacitor is loaded with a high voltage value only in a small part of the first duration.

[0039] In one embodiment of the method, the value of the first duration Δt1 in the active time window can be selected such that the arc energy resulting from the power loss of the assumed arc in the active time window does not exceed a predetermined maximum permissible energy value E max Therefore, in this case, in the current active time window, it is assumed that there is a fictitious (but not necessarily real) arc. However, based on the assumption that the fictitious arc burns in the active time window, the current I in the active time window and the assumed arc voltage drop ΔU can be calculated according to the following formula: LB To estimate the first duration Δt1 of the current active time window:

[0040]

[0041] Depending on the time resolution of the current measuring unit, it is also possible to use a plurality of detected values ​​of the current I. In this case, the average current I can be used in the denominator of the equation given above for the current I Mittelwert For the arc voltage drop ΔU LB There are experimentally determined or empirically based values. For example, in common PV systems, a value between 30 V and 40 V can be used as a good approximation for the arc voltage drop.

[0042] In detecting arcs in DC circuits of photovoltaic (PV) systems, if the maximum possible energy value E max With E also listed in the relevant standards max = 200 J. In particular, when the energy transmitted in the active time window does not exceed the value 200 J, a restart of the PV system can be performed in a simplified form and in particular without the use of electrical specialists in the case that an arc was actually detected in the previous active time window.

[0043] In one embodiment of the method, a plurality of DC sources can be connected to a DC load in parallel with each other via a circuit arrangement in order to jointly supply the DC load with their respective power flows P. In this case, the method for detecting, in particular, a series arc can be performed simultaneously for each DC source, wherein, when the method is performed for different DC sources, the deactivation time windows associated with the respective cyclic interruptions of the power flows P are as far as possible temporally staggered from one another and do not overlap as far as possible. Advantageously, the cyclic interruptions of the power flows P are performed in such a way that the temporal overlap between the deactivation time windows of the different DC sources is at least minimized. In this way, a robust operation of the DC load is ensured by minimizing the fluctuations in the total power flow resulting from the sum of the individual power flows P of the plurality of DC sources.

[0044] A circuit arrangement according to the invention for detecting an arc in a DC circuit comprises:

[0045] - an input terminal having two input terminals for connecting a DC source and an output terminal having two output terminals for connecting a DC load,

[0046] - Input capacitance between the input terminals,

[0047] a switching unit arranged between the input and the output, wherein the switching unit is designed and arranged to enable a power flow P between the input and the output in a first operating mode and to inhibit a power flow P between the input and the output in a second operating mode,

[0048] A measuring unit for determining the input voltage U applied to the input ein and / or an input current I flowing at one of the input terminals ein ,as well as

[0049] A control unit for controlling the switching unit and, if applicable, the measuring unit.

[0050] The circuit arrangement is characterized in that the control unit is designed and provided for carrying out the method according to the invention together with the switching unit and the measuring unit.

[0051] In order to enable the switching unit to suppress the power flow P between the input and the output, the switching unit can have one or more controllable switches, in particular semiconductor switches. Alternatively or additionally, the switching unit for suppressing the power flow P can also have one or more uncontrollable semiconductor switches, such as diodes. The control unit can be designed as a separate control unit. Alternatively, however, the control unit can also be designed as a component of a control unit that is already present in the DC circuit. The advantages already explained in conjunction with the method are obtained.

[0052] In an advantageous embodiment of the circuit arrangement, the switching unit may have a semiconductor switch which is arranged in one of the connecting lines from one of the input terminals to one of the output terminals. The semiconductor switch may not have an intrinsically constructed or separately constructed freewheeling diode connected in parallel with the current-carrying terminals of the semiconductor switch. However, as an alternative to this, the semiconductor switch may also have a freewheeling diode connected in parallel with the current-carrying terminals of the semiconductor switch. Here, the freewheeling diode may be constructed as an intrinsic freewheeling diode or as a separately constructed freewheeling diode. If the semiconductor switch has a freewheeling diode, the freewheeling diode may be arranged in particular such that it is polarized in the cut-off direction relative to the normal current direction of the DC circuit.

[0053] The circuit device may have an energy buffer connected in parallel with the output, in particular an output capacitor. With the output capacitor, the circuit device is able to buffer the limited energy of the DC source in the active time window, with the help of which the DC load is powered in the subsequent disabled time window. In this way, even if the DC load does not include the possibility of buffer storage, it is possible to power the DC load during the disabled time window. In this case, the power flow P is discontinuous only in the first region of the DC circuit, while the power flow P is continuous in the second region of the DC circuit. According to an embodiment, the circuit device or the switching unit of the circuit device according to the present invention can be designed as a DC / DC converter, in particular a boost converter. Here, the power flow P from the DC source to the output of the circuit device can be interrupted by disabling the circuit device, in particular the switching unit of the circuit device, and the voltage applied at the output of the circuit device is greater than the voltage applied at the input of the circuit device. Experiments have shown that it is sufficient to disable within a few (advantageously, up to 8) switching cycles of the DC / DC converter.

[0054] According to the invention, a photovoltaic (PV) inverter comprises: at least one input having two input terminals for connecting a PV string as a DC source; an output for connecting an alternating voltage (AC) grid; a DC / AC converter for converting a direct voltage into an alternating voltage; and a circuit arrangement according to the invention. Here, the input of the circuit arrangement is connected to at least one input of the PV inverter, and the output of the circuit arrangement is connected to the input of the DC / AC converter. The inverter can be a single-phase PV inverter having at most one phase conductor terminal and one neutral conductor terminal on the output side. Alternatively, the inverter can also be a multi-phase PV inverter having a plurality of (in particular three) phase terminals on the output side and one neutral conductor terminal on the output side. The control unit of the circuit arrangement can be integrated into the control unit of the PV inverter or into a superordinate control unit of a PV system including the PV inverter. As a result, the advantages already mentioned in conjunction with the method and the circuit arrangement are obtained.

[0055] When the method according to the invention is used to detect an arc in a PV system, in particular a series arc, wherein the DC source comprises a PV string and the DC load comprises a single-phase DC / AC converter connected to an alternating voltage (AC) grid on the output side, it is advantageous to cyclically interrupt the power flow P so that the AC current I flowing through the output terminals of the DC / AC converter AC(t) crosses zero during the second duration Δt2 of the disable time window. That is, in a single-phase DC / AC converter, power is extracted from the buffer of the PV inverter (e.g., the intermediate circuit capacitor of the DC intermediate circuit) in a sinusoidal pulse manner having a frequency corresponding to twice the frequency of the AC grid connected to the PV inverter, wherein the AC current I generated by the DC / AC converter AC Now, if the disable time window is aligned in time with the AC current I AC The zero crossing overlap of (t) can minimize the voltage ripple of the DC intermediate circuit.

[0056] In one embodiment of the invention, the PV inverter may have only a single input, to which a PV string as a DC source is connected. However, in another embodiment of the invention, the PV inverter may also have a plurality of inputs, which are designed to be connected to PV strings as DC sources, respectively. Here, the inputs may be connected in parallel to one another via a DC / DC converter to a common DC intermediate circuit. The DC intermediate circuit, which may include an intermediate circuit capacitor, may in turn be connected to the input of a DC / AC converter. Here, a circuit arrangement according to the invention may be associated with each input of a PV inverter, wherein each input of the PV inverter is respectively connected to the input of a circuit arrangement associated therewith, and the output of each circuit arrangement is respectively connected to the input of a DC / AC converter. Within the scope of the invention, each circuit arrangement is at least partially, and if necessary also completely, formed by a DC / DC converter.

[0057] In one embodiment of the PV inverter, the control unit of the circuit arrangement can be contained in a superordinate control unit of the PV inverter. Here, the superordinate control unit is simultaneously configured to control the DC / AC converter and the DC / DC converter of the PV inverter so that during operation of the PV inverter, the voltage applied to the DC intermediate circuit remains above the maximum value of the no-load voltage of the individual PV strings.

[0058] Advantageous embodiments of the invention are indicated in the following description of the drawings and in the dependent claims, the features of which may be used individually and in any desired combination with one another. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The present invention is described below with the aid of the accompanying drawings, in which:

[0061] Figure 1 shows a DC circuit having a DC source, a DC load and a circuit arrangement according to a first embodiment of the invention;

[0062] Figure 2shows a circuit arrangement according to a second embodiment of the invention;

[0063] Figure 3 shows the time curves of the input current and the input voltage of the circuit arrangement according to the invention during the method according to the invention;

[0064] Figure 4 shows an operating mode of a circuit arrangement according to the invention, a time profile of a current and a voltage during a further specific embodiment of a method according to the invention; and

[0065] Figure 5 A PV system having a PV inverter according to an embodiment of the present invention is shown.

[0066] Description of the drawings

[0067] exist Figure 1 In the figure, a direct current (DC) circuit 6 is shown having a DC source 2, a DC load 3 and a circuit arrangement 1 according to a first embodiment of the present invention, the DC source 2 having an internal resistance 2a. The circuit arrangement 1 has an input terminal 7, which has a first input terminal 7.1 and a second input terminal 7.2, to which the DC source 2 is connected. The circuit arrangement 1 also has an output terminal 8, which has a first output terminal 8.1 and a second output terminal 8.2, which is connected to the DC load 3. Each input terminal 7.1, 7.2 is connected to the corresponding output terminal 8.1, 8.2 via a connecting line 23, 24, respectively. A switching unit 10 is arranged between the input terminal 7 and the output terminal 8. The switching unit 10 is designed to enable a power flow P between the DC source 2 and the output terminal 8 during a first operating mode BM1, and to inhibit a power flow P between the DC source 2 and the output terminal 8 during a second operating mode BM2. For example, a power flow P between the input terminal 7 and the output terminal 8 can be enabled during a first operating mode BM1 and a power flow P between the input terminal 7 and the output terminal 8 can be blocked during a second operating mode BM2 by means of the switching unit 10. For this purpose, the switching unit 10 has at least one controllable switch and, if necessary, also a plurality of controllable switches. The switch can be an electromechanical switch and, advantageously, also a semiconductor switch. Furthermore, the switching unit 10 can also have one or more uncontrollable semiconductor switches, in particular diodes. The switching unit 10 is controlled by a control unit 5. In addition, the circuit arrangement 1 includes a measuring unit 4, which is designed to measure an input voltage U applied to the input terminal 7 of the circuit arrangement 1. ein and / or the input current I flowing through the input terminal 7 ein Additionally, the measuring unit 4 can also be designed to measure the output voltage U applied to the output terminal 8. aus and / or the output current I flowing through the output terminal 8 aus ( Figure 1 The measuring unit 4 is connected to the control unit 5 for communication and control. The circuit arrangement 1 also has an input capacitor 12, which is connected in parallel with the input terminal 7 between the input terminals 7.1 and 7.2. In addition, the circuit arrangement 1 may optionally include an output capacitor 11 ( Figure 1 The output capacitor is connected in parallel with the output terminal 8 between the output terminals 8.1 and 8.2.

[0068] Now, during operation of the circuit arrangement 1, the power flow P between the DC source 2 and the output 8 of the circuit arrangement 1 is interrupted cyclically, i.e. repeatedly in a regular manner, by means of the switching unit 10. Thus, the power flow P between the DC source 2 and the output 8 occurs discontinuously, wherein as described below in conjunction with Figure 4 As explained, active time windows 31 with a first duration Δt1 alternate with deactivated time windows 32 with a second duration Δt2. Thus, during the active time windows 31, a non-zero power flow P occurs between the DC source 2 and the output terminal 8, wherein in the deactivated time windows 32, the power flow P takes the value 0. The control unit 5 is configured to emit a signal regarding an arc presence criterion, which indicates the presence of an arc 9, in particular the presence of a series arc.

[0069] exist Figure 2 , a circuit device 1 according to a second embodiment of the present invention is shown. The circuit device 1 is similar to Figure 1 This is similar to the first embodiment already described in Figure 1 The following mainly describes the differences between the second embodiment and the first embodiment.

[0070] According to a second embodiment, the switching unit 10 is designed as a DC / DC converter 21, in particular a step-up converter, and is configured to convert the input voltage U applied to the input terminal 7 to ein Converted into an output voltage U applied to the output terminal 8 aus For this purpose, the switching unit 10 of the circuit arrangement 1 has an inductor 22 and a first semiconductor switch S1 connected in series therewith in a first connecting line 23 between the first input terminal 7.1 and the first output terminal 8.1. A first terminal of a second semiconductor switch S2 is connected to a connecting point 25 between the inductor 22 and the first semiconductor switch S1, and a second terminal of the second semiconductor switch S2 is connected to a second connecting line 24. The first semiconductor switch S1 as well as the second semiconductor switch S2 can each have an inherent or separately constructed freewheeling diode D1 or D2 (in Figure 2 denoted by dashed lines).

[0071] In other words, during the first operation mode BM1, in the following Figure 3 or Figure 4 In the active time window 31 with a first duration Δt1 shown in FIG. 1 , the circuit arrangement 1 is operated as a step-up converter and, by means of suitable clocking of the first semiconductor switch S1 and, if necessary, the second semiconductor switch S2, the input voltage U ein Converted to ratio input voltage U ein Large output voltage U aus Here, advantageously, the output voltage U aus is selected so that it is greater than the no-load voltage U0 of the DC source 2 connected to the input terminal 7 (at Figure 2 with the output voltage U aus The associated energy is temporarily stored in the output capacitor 11 arranged in parallel with the output terminal 8. In the second operating mode BM2, i.e. in the following Figure 3 or Figure 4 In the deactivation time window 32 with the second duration Δt2 shown in FIG. 3 , the clock of the semiconductor switches S1, S2 is interrupted by the control unit 5. During the second operating mode BM2, the semiconductor switches S1, S2 are continuously disconnected. The second duration Δt2 may last for several clock cycles of the DC / DC converter 21. Now, due to the voltage U applied at the output 8 aus is chosen so that the voltage U aus is greater than the no-load voltage U0 of the DC source 2 connected at the input, so that the freewheeling diode D1 (if present) associated with the first semiconductor switch S1 blocks the current flow from the first input terminal 7.1 to the first output terminal 8.1 and thereby blocks the power flow P from the DC source 2 to the output terminal 8 of the circuit arrangement 1.

[0072] By means of the control unit 5 , the circuit arrangement 1 is set in an alternating manner to the first operating mode BM1 during a first duration Δt1 and to the second operating mode BM2 during a second duration Δt2 , thereby generating a discontinuous power flow P from the DC source 2 to the output 8 of the circuit arrangement 1 .

[0073] exist Figure 3 , input current curves 33, 34 and input voltage curves 36, 37 of a circuit arrangement 1 according to the invention are shown over time during the method according to the invention. According to the invention, the power flow P between the input 7 and the output 8 is cyclically interrupted by means of the switching unit 10 of the circuit arrangement 1 according to the invention, and thus cyclically changes between an active time window 31 and a deactivated time window 32.

[0074] For the case where there is no arc 9, the input current curve 33 and the input voltage curve 36 are shown. Firstly, during the first active time window 31, a power flow P is generated between the input terminal 7 and the output terminal 8, and a constant input current I associated therewith is generated. ein and constant input voltage U ein Here, the input voltage U ein The value of is obtained by subtracting the voltage drop of the internal resistance 2a from the value of the no-load voltage U0 of the DC source 2.

[0075] During the deactivation time window 32 , the power flow P between the input terminal 7 and the output terminal 8 is interrupted by means of the switching unit 10 , so that the input current I ein It can only flow from the DC source 2 to the input capacitor 12 to charge the input capacitor 12. Since the current value decreases according to the e-function during this charging process, the voltage drop across the internal resistance 2a of the DC source 2 also decreases accordingly, so the input voltage U ein If the duration Δt2 of the disable time window 32 is correspondingly longer, the input capacitor will be charged to the value of the no-load voltage U0 of the DC source 2, wherein the input current will take the value 0A, so that there is no longer a voltage drop across the internal resistance 2a of the DC source 2, and accordingly, the input voltage U ein Set to the level of the no-load voltage U0 of the DC source 2.

[0076] However, in the method for detecting an arc according to the present invention Figure 3 In the variant shown in FIG. , the length of the duration Δt2 of the deactivation time window 32 is selected only so that the input current I ein does not completely become zero, but only drops to such an extent that any existing arc 9 can be reliably extinguished. ein It also does not rise to the value of the no-load voltage U0 of the DC source 2, but rather rises to a lower value.

[0077] In particular, the selection of the shortest possible duration Δt2 of the deactivation time window 32 can also take into account that the duration of the interruption of the power flow P between the input 7 and the output 8 occurring during the deactivation time window 32 can be kept as short as possible.

[0078] After the duration Δt2 of the deactivation time window 32, in the subsequent next active time window 31, the power flow P between the input terminal 7 and the output terminal 8 is enabled again by means of the switching unit 10, so that Figure 3 As identified in the ein rises again, and the input voltage U ein It decreases again so that after the transient process has ended, the input voltage and the input current adopt constant values ​​related to the power flow P in the first active time window 31 .

[0079] The input current curve 34 for the case where the arc 9 exists also shows that in the disabled time window 32, a decreasing trend occurs during the charging of the input capacitor 12, and then the arc 9 is extinguished when the current value is small enough, so the input current I ein abruptly becomes zero, and in the subsequent active time window 31, after the power flow P between the input terminal 7 and the output terminal 8 is enabled again by means of the switching unit 10, the input current I ein The value of remains zero. Thus, as mentioned above, in the absence of the arc 9, in the method according to the invention, the input current I is preferably not set ein Reduced to zero, it can be compared with the input current I ein The value of the current threshold I TH To detect the presence of an arc 9 and to issue a signal regarding the arc presence criterion.

[0080] Although the input voltage curve 37 for the case where the arc 9 exists also shows a rise in the disabled time window 32, the rise is smaller than the rise of the input voltage curve 36 for the case where the arc 9 does not exist. In particular, after the arc 9 is extinguished, due to the generated input current I ein The value of is 0A, so the input capacitor 12 is no longer charged. ein The increase is smaller than when there is no arc 9, so in the disable time window 32, the input voltage U ein The value and voltage threshold U TH After the power flow P to the output 8 can be enabled again by the switching unit 10 in the subsequent active time window 31, the input capacitor 12 is discharged through the DC load 3 via the switching unit 10, so that the input voltage U ein The value of decreases to zero in the input voltage curve 37 as time increases.

[0081] Current threshold I TH and voltage threshold U TH A fixed value determined empirically can be used. However, the voltage threshold U TH In particular, it is selected relative to the no-load voltage U0 of the DC source 2. Since the no-load voltage U0 of the DC source 2 may also be subject to large fluctuations (e.g. in a PV generator due to changes in external influences such as irradiation or temperature), a variable (if necessary adaptively adjustable, e.g., depending on the time-varying no-load voltage U0(t)) voltage threshold value U TH .

[0082] Figure 3The input current curves 33 and 34 in the first active time window 31 are the input current I ein The deviations in the values ​​are due to the different power flows P caused by the power losses generated by the arc 9.

[0083] As described above, the length of the duration Δt2 of the disable time window 32 may be selected such that the input current I ein is not completely reduced to zero, but only reduced to such an extent that any existing arc 9 can be reliably extinguished. However, in a variant of the method according to the invention, in which only the input voltage U ein The duration Δt2 of the disable time window 32 can also be selected so that the input current I ein In this case, due to the higher values ​​in the input voltage curve 36 and the resulting greater difference from the input voltage curve 36 , the possible presence of an arc can be more clearly inferred than when a shorter duration Δt2 is selected.

[0084] The duration Δt2 of the deactivation time window 32 can be set based on empirically determined values. In order to ensure reliable arc detection even in the case of possible fluctuations in the power flow P, and in particular also to keep the duration Δt2 of the deactivation time window 32 as short as possible, the duration Δt2 is preferably set adaptively and is selected, for example, as a function of the detected value of the current I representing the power flow P and / or the detected value of the voltage U representing the power flow P in the immediately preceding active time window 31. Alternatively, the duration Δt2 can also be selected as a function of the input current I detected in the current or one of the previous deactivation time windows 32. ein and / or the input voltage U detected in the current or one of the previous deactivation time windows 32 ein For example, in one embodiment, the duration Δt2 of the disable time window 32 is determined by the following method: ein The value is lower than another current threshold I TH2 Or the input voltage U ein The value exceeds another voltage threshold U TH2 In the case of an adaptive adjustment of the duration Δt2 of the deactivation time window 32 , in particular a maximum and / or a minimum value of the duration Δt2 can also be specified.

[0085] Another current threshold I TH2 and another voltage threshold U TH2 Empirically determined fixed values ​​can be used, wherein in particular the further voltage threshold U TH2It is preferably selected relative to the no-load voltage U0 of the DC source 2. It is also possible to use a further voltage threshold U t which is variable (if necessary, adaptively adjustable, for example, as a function of the no-load voltage U0 (t) which varies with time). TH2 Typically, another current threshold I TH2 and another voltage threshold U TH2 are respectively selected to be greater than the current threshold I TH and voltage threshold U TH .

[0086] exist Figure 4 In the Figure 2 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 92 ein (t), U ein (t). The time curve here reflects the behavior of the circuit arrangement 1 close in time to the ignition of the series arc 9 at time t0. Figure 3 different, Figure 4 The input current curve 35 and the input voltage curve 38 are shown only schematically, since only the input current I 35 which occurs after the transient process has ended is shown in each case. ein Or input voltage U ein The value of .

[0087] As shown in the figure above, the circuit device 1 is in the first operating mode BM1, the active time window 31 of duration Δt1 alternates with the circuit device in the second operating mode BM2, the disabled time window 32 of duration Δt2. In the first active time window 31, the current I ein (t) takes the value of I1 and voltage U ein (t) takes the value U1. The values ​​of the current I1 and the voltage U1 are derived from the power consumption of the DC load 3 connected to the output terminal 8 of the circuit arrangement 1. In the adjacent disable time window 32, the power flow P between the DC source 2 and the output terminal 8 of the circuit arrangement 1 is interrupted for a second duration Δt2. As previously described in connection with Figure 3 As described above, the current I flowing through the input terminal 7 ein (t) drops to a value close to 0A, while the voltage U applied to the input terminal 7 ein(t) increases to a value close to the no-load voltage U0 of the DC source 2. Since the actual DC source 2 (e.g., PV generator) has an internal resistance 2a, the no-load voltage U0 is usually higher than the voltage U1 of the loaded DC source 2. In the next (here, second) active time window 31, the current I ein (t) first corresponds to the value of the previous active time window 31 until the time point t0, at which the arc 9 ignites in the DC circuit 6, in particular between the DC source 2 and the input terminal 8 of the circuit arrangement 1. Starting from the time point t0, the current value, and possibly also the voltage value, suddenly drops slightly from the original value I1, U1 to the value I2, U2. The sudden drop in current is based on the power loss and the associated voltage drop of the series arc 9 starting from t0. For example, the series arc 9 burns across a microgap in the broken conductor. In the adjacent disabled time window 32, the series arc 9 is extinguished by interrupting the power flow P between the DC source 2 and the output terminal 8 of the circuit arrangement 1 again. The current I flowing through the input terminal 7 ein (t) first drops to a value close to 0 A again, whereby the arc is extinguished, and as a result, the current I ein (t) takes a value of 0 A. Compared with the voltage U2 in the immediately preceding second active time window 31, the voltage U applied to the input terminal 7 in the second disabled time window 32 is ein (t) continues to rise, but as above combined Figure 3 As explained, the voltage U ein (t) reaches a lower value than in the previous first disabled time window 32. In the subsequent third active time window 31, enabling the power flow P through the switching unit 10 is in principle possible, but now it is blocked by the microgap, which is why Figure 3 shows the voltage U ein (t) and current I ein (t) has negligibly small values.

[0088] The control unit 5 now converts the current I of the active time window 31 detected by the measuring unit 4 into ein (t) and / or voltage U ein The value of (t) is compared with the value of the corresponding previous active time window 31. Figure 4 When the value between the second active time window 31 and the third active time window 31 in the control unit 5 exceeds the threshold value, the control unit 5 determines that the difference between these values ​​exceeds the threshold value and issues a signal regarding the arc presence criterion in response. In response to issuing the signal regarding the arc presence criterion, if necessary, in combination with (e.g., in combination with Figure 3 If the other arc presence criterion is not met, the switching unit 10 remains in the second operating mode BM2 until manual confirmation by a skilled person, in which the power flow P between the DC source 2 and the output 8 is suppressed.

[0089] exist Figure 5 , a photovoltaic (PV) system 47 is shown, which has a PV inverter 40 according to an embodiment of the present invention. Here, the PV inverter 40 is constructed as a so-called multi-string inverter, which has multiple (here exemplarily two) DC-side input terminals 43, which have input terminals 43.1 and 43.2 for respectively connecting PV strings 45 as DC sources 2. Each input terminal 43 is connected in parallel with an intermediate circuit capacitor of a common DC intermediate circuit 42 via a DC / DC converter 21 (for example, via a boost converter). The intermediate circuit capacitor is connected to the input of the DC / AC converter 41. The output of the DC / AC converter 41 is connected via corresponding AC separation elements (in Figure 4 The PV inverter 40 is connected to an alternating voltage (AC) grid 46 via an output terminal 44 of the PV inverter 40 (not shown).

[0090] The PV inverter 40 comprises a circuit arrangement 1, one for each DC-side input 43. The circuit arrangement 1, in particular its switching unit 10, is designed as a DC / DC converter and is formed at least partially by an already existing DC / DC converter 21 of the PV inverter 40. The input 7 of the circuit arrangement 1 corresponds to the input 43 of the PV inverter 40. The output 8 of the circuit arrangement 1 corresponds to the output of the DC / DC converter 21. The circuit arrangement 1 has a common control unit 5, which is a component of the central control unit of the PV inverter 40 and is designed and provided for controlling the DC / DC converter 21 and the DC / AC converter 41. The measuring unit 4 of the circuit arrangement is likewise a component of the DC / DC converter 21 and is connected to the measuring unit 4 of the circuit arrangement. Figure 5 Not explicitly shown.

[0091] The method according to the invention is carried out by the control unit 5 simultaneously for each DC source 2 configured as a PV string 45. Thus, in normal operation of the PV inverter 40, each PV string 45 has a discontinuous power flow P in the respective DC line between its PV modules and the respective DC-side input 43 of the PV inverter 40. In this case, an active time window 31 with a first duration Δt1 alternates with a deactivation time window 32 with a second duration Δt2. Advantageously, a discontinuous power flow P in two PV strings 45 can be achieved in this case, so that the deactivation time window 32 of one PV string 45 overlaps as little as possible with the deactivation time window of the other PV string 45, i.e. the deactivation time windows are as far as possible offset from one another. If an arc 9 (in particular a series arc) is now detected in one PV string 45 by means of the control unit 5, only the switching unit 10 associated with the PV string in which the series arc was detected, i.e. the DC / DC converter 21 associated with the PV string in which the series arc was detected, is placed in the second operating mode BM2 for a longer period of time and, if necessary, until manual confirmation by an expert. In contrast, the respective other PV string 45 can continue to operate with a discontinuous power flow P in the direction of its associated DC-side input terminal 43 .

[0092] Reference Mark List

[0093] 1 Circuit Device

[0094] 2 DC Source

[0095] 3 DC load

[0096] 4 Measurement units

[0097] 5 Control unit

[0098] 6 Direct current (DC) circuit

[0099] 7 Input

[0100] 7.1, 7.2 Input terminals

[0101] 8 Output

[0102] 8.1, 8.2 Output terminals

[0103] 9 Arc (series)

[0104] 10 Switching unit

[0105] 11 Output Capacitor

[0106] 12 Input Capacitance

[0107] 21 DC / DC Converter

[0108] 22 Inductor

[0109] 23, 24 Connection lines

[0110] 25 Connection points

[0111] 31,32 Time Window

[0112] 33,34,35 Input current curve

[0113] 36,37,38 Input voltage curve

[0114] 40 Photovoltaic (PV) Inverter

[0115] 41 DC / AC Converter

[0116] 42 DC intermediate circuit

[0117] 43 Input

[0118] 43.1, 43.2 Input terminals

[0119] 44 Output

[0120] 45 PV string

[0121] 46 AC voltage (AC) power grid

[0122] 47 Photovoltaic (PV) systems

[0123] Δt1, Δt2 duration

[0124] S1, S2 semiconductor switch

[0125] BM1, BM2 operation mode

[0126] D1, D2 diodes

[0127] P Power Flow

[0128] U ein Input voltage

[0129] U aus Output voltage

[0130] I ein Input Current

[0131] I aus Output current.

Claims

1. A method for detecting an arc in a direct current (DC) circuit, the DC circuit having a DC load, a DC source for supplying the DC load, and a circuit arrangement arranged between the DC source and the DC load, - in, The circuit device comprises: an input end having two input terminals for connecting the DC source; an output end having two output terminals for connecting the DC load; and a switching unit arranged between the input end and the output end, and the circuit device has an input capacitor between the input terminals at the input end, - wherein the switching unit is designed to enable a power flow P between the DC source and the output in a first operating mode (BM1) and to inhibit the power flow between the DC source and the output in a second operating mode (BM2), The method comprises the following method steps: - cyclically interrupting the power flow P between the input and the output by means of the switching unit, such that the power flow P is enabled in an active time window having a first duration Δt1 and is inhibited in a deactivated time window having a second duration Δt2, - Detecting the input current I flowing at the input terminal ein and / or an input voltage U applied to the input terminal ein , - The input current I to be detected during the disable time window ein and / or input voltage U ein The value of the current threshold I TH Or voltage threshold U TH To compare, and - If the input current I detected during the disable time window ein Below the current threshold I TH and / or the input voltage U detected in the disabled time window ein does not exceed the voltage threshold U TH , a signal is issued regarding the presence of an arc.

2. The method according to claim 1, wherein: In response to the issuance of a signal regarding the arc presence criterion, the power flow P between the input and the output is permanently interrupted by means of the switching unit.

3. The method according to claim 1, further comprising the following method steps: - detecting a current I characteristic of the power flow P and / or a voltage U characteristic of the power flow P in two consecutive active time windows, - comparing the detected values ​​of the current I and / or voltage U of the active time window with the corresponding values ​​detected of a previous active time window, and If the value of the current I and / or the voltage U of the active time window differs from the corresponding value of the previous active time window by more than a threshold value, a signal is issued regarding a further arc presence criterion.

4. The method according to claim 3, wherein: In response to the issuing of a signal regarding the arc presence criterion and / or in response to the issuing of a signal regarding the further arc presence criterion, the power flow P between the input and the output is permanently interrupted by means of the switching unit.

5. The method according to claim 3, wherein: The active time window and the previous active time window are separated from each other by exactly one disabled time window.

6. The method according to claim 4, wherein: The active time window and the previous active time window are separated from each other by exactly one disabled time window.

7. The method according to any one of claims 1 to 6, wherein: The value of the second duration Δt2 of the disabled time window is sufficient to extinguish an arc that may exist in the DC circuit and to inhibit the arc extinguished in the disabled time window from reigniting in an active time window, which may immediately follow the disabled time window in time.

8. The method according to claim 7, wherein: The value of the second time duration Δt2 is between 0.1 ms and 10.0 ms.

9. The method according to any one of claims 1 to 6 and 8, wherein: The value of the second time period Δt2 is selected as a function of a value of the current I characteristic of the power flow P and / or a value of the voltage U characteristic of the power flow P detected in an immediately preceding active time window.

10. The method according to any one of claims 1 to 6 and 8, wherein: According to the input current I detected in the disable time window ein The value of the input voltage U detected in the disabled time window ein The second duration Δt2 of the disabled time window is determined by the value of 11. The method according to claim 10, wherein: The second duration Δt2 of the disable time window is defined in the following manner: the detected input current I ein Below another current threshold I TH2 and / or the detected input voltage U ein Exceeding another voltage threshold U TH2 .

12. The method according to any one of claims 1 to 6, 8 and 11, wherein: In the disable time window, the input voltage U applied to the input capacitor ein The input current I is measured over time. ein .

13. The method according to any one of claims 1 to 6, 8 and 11, wherein: The switching unit is formed by a DC / DC converter, and the circuit arrangement has an output capacitor connected in parallel with the output, and wherein, in the active time window, the voltage U applied to the output capacitor is amplified by the switching unit near the end of the first time period Δt1. aus .

14. The method according to any one of claims 1 to 6, 8 and 11, wherein: The value of the first duration Δt1 of the active time window is chosen so that the arc energy resulting from the power loss of the arc assumed in the active time window does not exceed a predetermined maximum permissible energy value E max .

15. The method according to any one of claims 1 to 6, 8 and 11, wherein: The DC source has a PV string and the DC load has a single-phase DC / AC converter connected to an AC voltage grid at the output side, and wherein the power flow P is interrupted so that the AC current I flowing through the output of the DC / AC converter AC (t) crosses zero during a second duration Δt2 of the disable time window.

16. The method according to any one of claims 1 to 6, 8 and 11, wherein: A plurality of DC sources are connected to the DC load in parallel via switch units, respectively, so as to jointly power the DC load with their respective power flows P, wherein the method is performed for each DC source simultaneously so that the time overlap of the disabled time windows of the plurality of DC sources is minimized.

17. The method according to claim 2, wherein: In response to the signalling of the arc presence criterion, the power flow P between the input and the output is interrupted for a period of several minutes by means of the switching unit.

18. The method according to claim 2, wherein: In response to signaling the arc presence criterion, the power flow P between the input and the output is interrupted by means of the switching unit until manual confirmation.

19. The method according to claim 4, wherein: In response to the signaling of the arc presence criterion and / or the signaling of the further arc presence criterion, the power flow P between the input and the output is interrupted for a period of several minutes by means of the switching unit.

20. The method according to claim 4, wherein: In response to signaling the arc presence criterion and / or signaling the further arc presence criterion, the power flow P between the input and the output is interrupted by means of the switching unit until manual confirmation.

21. The method according to claim 8, wherein: The value of the second duration Δt2 is between 0.3 ms and 4.0 ms.

22. The method according to claim 9, wherein: The second time duration Δt2 increases as the value of the current I and / or the voltage U increases.

23. The method according to claim 13, wherein: The switching unit is formed by a boost converter.

24. The method according to claim 14, wherein: The predetermined maximum allowable energy value is E max = 200 J.

25. A circuit arrangement for detecting an arc in a DC circuit, comprising: - an input terminal having two input terminals for connecting a DC source and an output terminal having two output terminals for connecting a DC load, - the input capacitance between the input terminals, - a switching unit arranged between the input terminal and the output terminal, wherein: The switching unit is designed to release a power flow P between the input terminal and the output terminal in a first operating mode (BM1) and to inhibit a power flow P between the input terminal and the output terminal in a second operating mode (BM2), - a measuring unit for determining an input voltage U applied to the input terminal ein and / or an input current I flowing at one of the input terminals ein ,as well as - a control unit for controlling the switching unit, It is characterized in that The control unit is designed and arranged to carry out the method according to any one of claims 1 to 24 together with the switching unit and the measuring unit.

26. The circuit arrangement according to claim 25, characterized in that The switching unit has a semiconductor switch, which is arranged in one of the connecting lines from one of the input terminals to one of the output terminals, wherein the semiconductor switch does not have a freewheeling diode, or wherein the semiconductor switch has an inherent freewheeling diode, which is polarized in a cut-off direction relative to a normal current direction of the DC circuit.

27. The circuit arrangement according to claim 25 or 26, characterized in that The circuit arrangement or the switching unit of the circuit arrangement is designed as a DC / DC converter.

28. The circuit arrangement according to claim 25, characterized in that The measuring unit is additionally used to determine the voltage U present at the output aus and / or a current I flowing at one of the output terminals aus .

29. The circuit arrangement according to claim 25, characterized in that The control unit is further used to control the measuring unit.

30. The circuit arrangement according to claim 27, characterized in that The circuit arrangement or the switching unit of the circuit arrangement is designed as a step-up converter.

31. A photovoltaic (PV) inverter, comprising: - at least one input having two input terminals for connecting a PV string as a DC source; and an output terminal for connection to an AC voltage grid, - a DC / AC converter for converting a DC voltage into an AC voltage, and - A circuit arrangement according to any one of claims 25 to 30, the input of the circuit arrangement being connected to the at least one input of the PV inverter and the output of the circuit arrangement being connected to the input of the DC / AC converter.

32. The PV inverter according to claim 31, comprising: a plurality of input terminals for respectively connecting PV strings as DC sources, wherein: The input ends are respectively connected in parallel to a common DC intermediate circuit via DC / DC converters, and wherein the DC intermediate circuit is connected to the input end of the DC / AC converter; and a plurality of circuit devices according to any one of claims 25 to 30, wherein each input end of the PV inverter is respectively connected to the input end of the circuit device associated therewith, and wherein the output end of each circuit device is respectively connected to the input end of the DC / AC converter.

Images