MONITORING DEVICE FOR EMERGENCY STANDBY OPERATION

Abstract

To determine an AC insulation resistance between the AC connections of an inverter and a ground potential, the AC connections are connected to an AC energy sink in an emergency standby operation in order to transmit energy from a DC energy source to an AC energy sink and an insulation test is performed in emergency standby operation. An AC fault current flowing between an AC connection and the ground potential is determined, and the AC insulation resistance is calculated from the AC fault current.

Claims

1. A method for determining an AC insulation resistance between the AC connections of an inverter and a ground potential, wherein a DC voltage is applied between DC connections of the inverter, and an AC voltage is applied between AC connections of the inverter, wherein the DC connections are connected to a DC power source, and wherein, in emergency standby operation, the AC connections are connected to an AC energy sink in order to transmit energy from the DC energy source to the AC energy sink, wherein, in emergency standby operation, an insulation test is performed, wherein an AC fault current flowing between an AC connection and the ground potential is determined from an insulation voltage, preferably the AC voltage component of the insulation voltage, present between one of the intermediate circuit connections and the ground potential or an intermediate circuit midpoint and the ground potential, and from an insulation impedance between the intermediate circuit connection and the ground potential, or the intermediate circuit midpoint and the ground potential, wherein a phase shift between an AC phase angle of the AC voltage and a fault current phase angle of the AC fault current is calculated, and wherein the AC insulation resistance is calculated from the phase shift.

2. The method according to claim 1, wherein the insulation test is repeated several times, preferably cyclically, with a test cycle rate, particularly preferably with a test cycle rate in the range of seconds.

3. The method according to claim 1, wherein the insulation impedance is approximated by a parasitic DC capacitance or by a parasitic DC capacitance and a parallel DC insulation resistance.

4. The method according to claim 1, wherein, during the insulation test, a test signal with a test frequency greater than an AC frequency of the AC voltage is modulated onto the AC voltage.

5. The method according to claim 1, wherein, when a fault current threshold is exceeded by the AC fault current and/or when an insulation resistance threshold of the AC insulation resistance is undershot, the inverter is switched to a fault current operation and/or at least partially switched off.

6. The method according to claim 1, wherein a mains operation of the inverter is provided, wherein, while switching from emergency standby operation to mains operation, the AC connections of the inverter are separated from the AC energy sink and/or connected to an energy supply network in order to transmit energy from the DC energy source to the energy supply network, and wherein the insulation test is not run during mains operation.

7. The method according to claim 1, wherein a DC insulation fault is detected by connecting a high-ohmic resistor between an intermediate circuit connection and the ground potential, wherein a DC voltage component of the insulation voltage between one of the intermediate circuit connections and the ground potential is determined before and after connecting the high-ohmic resistor, and wherein a ratio of the insulation voltages is determined, and a DC insulation fault is inferred from the ratio.

8. A monitoring device for determining an AC insulation resistance between AC connections of an inverter and a ground potential, wherein DC connections of the inverter can be connected to a DC power source, and, in an emergency standby operation, AC connections of the inverter can be connected to an AC energy sink, wherein the monitoring device comprises a fault current detection unit which is configured to determine an AC fault current flowing between an AC connection and a ground potential, from an insulation voltage, preferably the AC voltage component of the insulation voltage, present between one of the intermediate circuit connections and the ground potential or an intermediate circuit midpoint and the ground potential, and from an insulation impedance between the intermediate circuit connection and the ground potential or the intermediate circuit midpoint and the ground potential, and wherein the monitoring device comprises a calculation unit which is configured to calculate a phase shift between an AC phase angle of the AC voltage and a fault current phase angle of the AC fault current, and to calculate the AC insulation resistance from the phase shift.

9. The monitoring device according to claim 8, wherein a voltage measuring unit is provided which is configured to measure an insulation voltage, preferably the AC voltage component of the insulation voltage, present between one of the intermediate circuit connections and the ground potential, and wherein the fault current detection unit is configured to calculate the AC fault current from the insulation voltage, preferably the AC voltage component of the insulation voltage, and from an insulation impedance between the intermediate circuit connections and the ground potential.

10. An arrangement of an inverter and a monitoring unit according to claim 8.

Description

[0026] The present invention is described in greater detail below with reference to FIGS. 1 through 5, which show schematic and non-limiting advantageous embodiments of the invention by way of example. The following are shown:

[0027] FIG. 1 an arrangement consisting of an inverter, an AC energy sink, a DC energy source, and a monitoring device,

[0028] FIG. 2 the arrangement with a fault current detection unit as a fault current measuring unit,

[0029] FIG. 3 the arrangement with a voltage measuring unit for measuring an insulation voltage at an insulation impedance,

[0030] FIG. 4 an alternative arrangement for measuring an insulation impedance,

[0031] FIG. 5 waveforms of an AC frequency, a test frequency, and associated fault currents, phase angles, AC voltages, fault current phase angles, and insulation resistances.

[0032] FIG. 1 shows an arrangement consisting of an inverter 2, an AC energy sink 4, and a DC energy source 3. AC connections AC1, AC2 of the inverter 2 are connected to the AC energy sink 4, and DC connections DC+, DC of the inverter 2 are connected to the DC power source 3. The AC connections AC1, AC2 can represent, for example, two phases of the inverter 2. The inverter 2 further comprises an optional intermediate circuit with intermediate circuit connections ZK+, ZK, between which a DC intermediate circuit voltage Uzk is present.

[0033] It is also possible to connect an inverter 2, which is fundamentally multi-phase, e.g., three-phase, to a two-phase AC energy sink 4, wherein two of the three phases of the inverter 2 are connected to the AC energy sink 4. A multi-phase, e.g., three-phase, AC energy sink 4 can also be connected to the AC connections of the inverter 2, wherein several, e.g., three, AC connections are each formed by the phases of the inverter 2.

[0034] The inverter 2 converts a DC input voltage Ue, provided by the DC power source 3, via the DC intermediate circuit voltage ZK+, ZK into an AC voltage ua, which in turn is made available to the AC energy sink 4. In this way, energy is transmitted from the DC energy source 3 to the AC energy sink 4. The DC voltage Ue is present between the DC connections DC+, DC, and the AC voltage ua is present between the AC connections AC1, AC2. An electric machine or another electrical consumer, for example, can be provided as an AC energy sink 4. For example, batteries/storage batteries and/or photovoltaic cells can be provided as a DC energy source 3.

[0035] Filter capacitors and/or filter inductors can be provided in each case at the DC connections DC+, DC, and also at the AC connections AC1, AC2 (not shown in the figures).

[0036] The inverter 4 can also be designed to be bidirectional, which means that an AC voltage ua applied to the AC connections AC1, AC2 can also be converted to the DC intermediate circuit voltage ZK+, ZK and further to the DC input voltage Ue applied to the DC connections DC+, DC. In this way, energy can also be transported from the AC connections AC1, AC2 to the intermediate circuit, which is connected, for example, to a current storage/battery which directly consumes PV current. A bidirectional inverter 2 is also referred to as a hybrid inverter. Therefore, consumers can also be supplied with separate AC connections, i.e., for example, from the current storage/battery. The AC connections can be opened by means of an AC isolator so that the inverter 2 is disconnected from the energy supply network.

[0037] In the figures, the inverter 2 is therefore in an emergency standby operation NE, since the AC isolator is open, which is why, for simplification, the energy supply network and the AC isolator are not shown. If the inverter 2 is switched to mains operation (not shown), then the AC connections AC1, AC2 are connected to the mains phases of an energy supply network (closed AC isolator), and the AC energy sink 4 can be disconnected. If the inverter 2 is switched back into emergency standby operation NE, the AC connections AC1, AC2 are disconnected from the mains phases of the energy supply network and connected to the AC energy sink 4. The energy sink 4 can, for example, be at least one socket, integrated into the inverter 2, at which at least one consumer can be connected and supplied.

[0038] In such an embodiment, the inverter 2 can therefore supply, for example, those consumers in emergency standby operation NE which are connected to the integrated socket. All other consumers can be separated by an AC isolator.

[0039] In a further embodiment, it is also possible for the integrated socket to be supplied by the energy supply network even during mains operation. In this case, the AC energy sink can nevertheless be disconnected in the circuitry and can be bridged, so to speak, with a connection to the power supply network.

[0040] In emergency standby operation NE, network protection devices do not affect the energy supply network, since the inverter 2 is separated from the energy supply network and therefore also from the mains protection devices. Instead, an IT system, in conjunction with the AC energy sink 4, is present, which is not connected to a ground potential GND by definition. An AC insulation resistance R.sub.isoAC is present between the AC connections AC1, AC2 of the inverter 2 and the ground potential GND, wherein, according to the invention, a monitoring device 1 for determining the ohmic AC insulation resistance R.sub.isoAC is provided. The monitoring device 1 comprises a fault current detection unit 10 which is designed, for performing an insulation test I, to determine an AC fault current i.sub.fAC flowing between an AC connection AC1. AC2 and a ground potential GND. A calculation unit 11 calculates, using the AC fault current i.sub.fAC, the AC insulation resistance R.sub.isoAC, The AC in the AC insulation resistance R.sub.isoAC relates to the AC side of the inverter 2, at which the AC insulation resistance R.sub.isoAC occurs.

[0041] If no insulation fault occurs in the inverter 2, the AC insulation resistance R.sub.isoAC is high. However, if an insulation fault occurs in the inverter 2, the AC insulation resistance R.sub.isoAC decreases. If an insulation fault occurs, the AC insulation resistance R.sub.isoAC is low.

[0042] If the fault current detection unit 10 is designed as a fault current measuring unit, as shown in FIG. 2, the AC fault current i.sub.fAC can be measured directly to perform the insulation test I. The fault current measuring unit 10 can be a residual current measuring unit (RCMU), which encloses the AC connections AC1, AC2 and measures a sum current.

[0043] Accordingly, the fault current detection unit 10 and the calculation unit 11 can be connected to a control unit (not shown) of the inverter, which is designed to control the fault current detection unit 10 and the calculation unit 11.

[0044] Alternatively or additionally, the fault current detection unit 10 can be configured to calculate the AC fault current i.sub.fAC for performing the insulation test I. For this purpose, an AC voltage component of an insulation voltage U.sub.iso can be determined between one of the intermediate circuit connections ZK+, ZK and a ground potential GND, as shown in FIG. 3. In conjunction with a known insulation impedance Z.sub.iso which is present between one of the intermediate circuit connections ZK+, ZK and the ground potential GND, the AC fault current i.sub.fAC can be calculated in the fault current detection unit 10. The AC insulation resistance R.sub.isoAC can then be correspondingly calculated from the AC fault current i.sub.fAC. Particularly in multi-phase inverters 2, an AC voltage component of the insulation voltage U.sub.iso can also be determined between an intermediate circuit midpoint ZKm and a ground potential GND, as shown in FIG. 4. The AC fault current i.sub.fAC can be calculated as described above with respect to FIG. 3. The insulation voltage U.sub.iso can be determined using a voltage measuring unit 12, as shown in FIGS. 3 and 4. The voltage measuring unit 12 preferably measures both a DC voltage component and an AC voltage component of the insulation voltage U.sub.iso.

[0045] Preferably, during the, preferably entire, insulation test I, a test signal s with a test frequency f greater than an AC frequency f.sub.AC of the AC voltage ua is modulated onto the AC voltage ua, which is indicated in FIGS. 1 through 4 by ua (+s). To generate the test signal, a test signal generator can be provided; to modulate the test signal onto the AC voltage ua, a test signal coupling unit can be provided.

[0046] The insulation impedance Z.sub.iso is preferably approximated by a known parasitic DC capacitance C.sub.isoDCfor example, a DC-side capacitance of a DC voltage source (e.g., a PV generator). The DC capacitance C.sub.isoDC can be required for other routines in the inverter 2 and, accordingly, preferably be determined continuously. In this case, the already determined DC capacitance C.sub.isoDC can be used for the calculation of the AC fault current i.sub.fAC. It can be assumed that the insulation impedance Z.sub.iso approximately corresponds to the reactance X.sub.C which is formed by the parasitic DC capacitance C.sub.isoDC at the AC frequency f.sub.AC of the AC voltage, or when using a test signal s at a test frequency f of the test signal s. It is also possible to approach the insulation impedance Z.sub.iso by a known parasitic DC capacitance C.sub.isoDC and a parallel ohmic DC insulation resistance R.sub.isoDC, as indicated in FIG. 3. The DC in the DC insulation resistance R.sub.isoDC and in DC capacitance C.sub.isoDC relates to the DC side of the inverter 2 at which the DC insulation resistance R.sub.isoDC and DC capacitance C.sub.isoDC occurs. Especially with a small ohmic DC insulation R.sub.isoDC of the insulation impedance Z.sub.iso, the insulation impedance Z.sub.iso can again be approximated by the reactance X.sub.C, because, if the DC insulation resistance R.sub.isoDC is low, it can already be assumed that a DC insulation fault exists, which means that a calculation of the AC insulation resistance R.sub.isoAC can be omitted.

[0047] In both cases, the magnitude and the phase angle of the AC fault current i.sub.fAC can be calculated directly from the quotient i.sub.fAC=U.sub.iso/Z.sub.iso (at the AC frequency f.sub.AC or, if a test signal s is used, at the test frequency f of the test signal s). To calculate the AC fault current i.sub.fAC, an alternating voltage component of the insulation voltage U.sub.iso is preferably used at the AC frequency f.sub.AC or, if a test signal s is used, at the test frequency f of a test signal s.

[0048] Regardless of whether the AC fault current is measured according to FIG. 2 or is calculated according to FIG. 1, FIG. 3, and/or FIG. 4, the AC fault current i.sub.fAC serves as the input for the calculation unit 11, which calculates the AC insulation resistance R.sub.isoAC. For this purpose, the following steps can be provided:

[0049] In a next step, the phase shift between the AC phase angle .sub.AC of the AC voltage ua (or of the test signal s) and the fault current phase angle .sub.ifAC of the AC fault current i.sub.fAC is calculated: =.sub.AC.sub.ifAC. The phase shift is therefore the distance of the zero crossings between the AC voltage ua (or of the test signal s) and the AC fault current i.sub.fAC. If a test signal s is superimposed on the AC voltage ua, then the component with the test frequency f of the test signal s is relevant from the resulting voltage ua+s.

[0050] From the known relationship tan([Z])=Im[Z]/Re[Z] (Z corresponds to Z.sub.iso), as mentioned above, when the DC insulation resistance R.sub.isoDC is high, the approximate equation tan()=R.sub.isoAC/X.sub.C can be formed, which then can be solved for R.sub.isoAC, which allows the AC insulation resistance R.sub.isoAC to be calculated.

[0051] The AC insulation resistance R.sub.isoAC is preferably calculated cyclically, for example, is in the range of seconds.

[0052] To calculate the AC insulation resistance R.sub.isoAC, a test signal s with a test frequency f can also be used. The test frequency f, which is superimposed/modulated onto the AC frequency f.sub.AC, can correspond to the AC frequency f.sub.AC of the AC voltage ua, or preferably a higher frequency. The test signal s is preferably continuously present during the insulation test I.

[0053] FIG. 5 contrasts the measured values for an insulation test using the AC frequency f.sub.AC of 50 Hz (right) and using a test frequency f of 150 Hz (left).

[0054] The top graph shows an AC voltage component of the insulation voltage U.sub.iso with the AC frequency f.sub.AC and the test frequency; in the graph underneath, the AC fault current i.sub.fAC with the AC frequency f.sub.AC and the test frequency f can be seen. It is apparent that the fault current i.sub.fAC (or its amplitude), when using a test signal s with a test frequency f greater than the AC frequency f.sub.AC (i.e., in this case, 150 Hz>50 Hz), is higher than the fault current i.sub.fAC when using the AC frequency f.sub.AC.

[0055] Furthermore, FIG. 5 shows on the left the AC phase angle .sub.AC of the test signal s and the associated fault current phase angle .sub.ifAC (dashed), and FIG. 5 on the right shows the AC phase angle .sub.ifAC of the AC voltage and the associated fault current phase angle (dashed) .sub.ifAC. It is therefore apparent that the fault current phase angle .sub.ifAC at the AC frequency f differs from the fault current phase angle .sub.ifAC at the test frequency f. During an insulation test I, an AC insulation resistance R.sub.isoAC is calculated, which is preferably repeated cyclically with a test cycle rate. The test cycle rate may be in the range of seconds, and may, for example, be 1 to 5, preferably 3, seconds, which means that the insulation test I is repeated every 1 to 5, preferably every 3, seconds. An AC insulation fault exists if the AC insulation resistance R.sub.isoAC, calculated from the phase shift between the AC phase angle .sub.AC of the AC voltage ua (or the test signal s) and the fault current phase angle .sub.ifAC, is under a predetermined insulation resistance threshold.

[0056] Since the reactance X.sub.C of the DC capacitance C.sub.isoDC rises with the frequency, the AC fault current i.sub.fAC also rises with rising frequency (see FIG. 5, i.sub.fAC), which allows an influence by a possibly fluctuating parasitic DC capacitance C.sub.isoDC to be compensated for. The measurement of an AC voltage component of the insulation voltage U.sub.iso is not affected at all or only slightly by an increased frequency (which is shown here with the test frequency f=150 Hz compared to the AC frequency f) since, although the current rises, the reactance X.sub.C falls to the same extent.

[0057] Since the inverter 4 does not have any galvanic isolation between the AC connections and the DC connections, a DC insulation fault can also be detected. This can be done by the fault current detection unit 10 (for example, a residual current measuring unit (RCMU) according to FIG. 2), which measures both an AC voltage component and the DC voltage component of the fault current i.sub.fAC. A residual current measuring unit is particularly suitable for detecting suddenly occurring DC insulation faults.

[0058] In general, however, a DC insulation fault can also be detected by connecting a high-ohmic resistor R between one of the intermediate circuit connections ZK+, ZK and the ground potential GND by means of a test switch S, as shown in dashed lines in FIG. 3 and also valid for the variant according to FIG. 4. The test switch S can be controlled by the monitoring device 1.

[0059] A DC insulation fault can be detected by connecting a high-ohmic resistor R between one of the intermediate circuit connections ZK, ZK+ and the ground potential GND, and determining a DC voltage component of the insulation voltage U.sub.iso between one of the intermediate circuit connections ZK+, ZK and the ground potential GND before and after connecting the high-ohmic resistor R. The ratio of these two direct current components to one another allows a conclusion to be drawn regarding the level of insulation resistance R.sub.isoDC. The higher the insulation resistance R.sub.isoDC, the more strongly the potential at the intermediate circuit connection ZK+, ZK is shifted against the ground potential GND by connecting the high-ohmic resistor R. The greater the ratio of the insulation voltages U.sub.iso to each other, the greater the DC insulation resistance R.sub.isoDC.

[0060] In order to achieve a sufficient speed and current strength in the variant according to FIG. 2, it is advantageous if a switch provided in the residual current fault current detector is replaced or supplemented by a semiconductor switch to realize the test switch S. In all variants, a DC voltage component of the insulation voltage U.sub.iso is determined between one of the intermediate circuit connections ZK+, ZK and the ground potential GND before and after connecting the high-ohmic resistor R.

[0061] For combined insulation monitoring of DC insulation faults and AC insulation faults, the following sequence of a routine can now preferably be used: [0062] 1. Check whether there is a DC insulation fault. [0063] 2. If there is a DC insulation fault, repeat the check. [0064] 3. If there is no DC insulation fault, switch the inverter 2 to mains operation or emergency standby operation. [0065] 4. Check whether there is a DC or AC insulation fault. This is done in a cyclical/serial manner with the test cycle rate, according to the previously described operations.

[0066] Step 4 can be provided if, according to a specification, AC and DC insulation faults must be checked continuously during emergency standby operation in the IT network. When an AC or DC insulation fault is detected, i.e., an excessively low AC insulation resistance and/or DC insulation resistance R.sub.isoDC, the inverter 2 can accordingly be switched off.