DE102015101870A1 - Method for compensating a portion of a differential current flowing via a leakage capacitance and suitable inverters for carrying out the method - Google Patents

Abstract

In order to compensate for a portion of a differential current flowing via a leakage capacitance, which is detected as current sum via lines (4, 5) which carry the current of an alternating current generator (8), a signal (13) is generated starting from a measuring leakage current (itest), at least the ohmic resistance (Radj) of an adjustable measuring resistor (15), which together with a further capacitor (16) is a low - pass filter (14) for at least one measuring capacitor (10) to ground cause the voltages at the alternator (8) the signal (13) is formed, tracked in such a way that an effective value of the differential current after subtraction of the signal (13) at the current value of the ohmic resistance (Radj) has a minimum.

Description

The invention relates to a method for compensating a portion of a differential current flowing through a leakage capacitance with the features of the preamble of independent claim 1 and to an inverter, which is suitable for carrying out this method and has the features of the preamble of independent claim 13.

In particular, the AC generator is one in which an inverter is provided which converts a DC current provided from a DC power source into an AC power. The DC power source may in particular be a photovoltaic generator.

The monitoring of a differential current is a common procedure to detect the occurrence of a critical fault current, which must lead due to safety requirements for switching off electrical systems. An occurring fault current increases the residual current. In the case of the effective value of the differential current, however, an increasing fault current makes itself noticeable to varying degrees depending on how large the other components of the differential current are. In particular, these leakage rates include a leakage current that occurs when voltages to ground change such that discharge capacitances that are effective with respect to ground recharge. Such a high current leakage current occurs, for example, in an inverter that feeds electrical energy from an ungrounded photovoltaic generator into an AC grid. Here, large-area photovoltaic panels of the photovoltaic generator have large discharge capacities towards earth; and to the DC voltage inputs of the inverter and the connected leads of the photovoltaic generator are variable voltages to earth. The large capacities of the photovoltaic panels with respect to earth are not constant, but for example dependent on the weather.

STATE OF THE ART

From the EP 2 372 857 A1 For example, in order to determine the fault current portion of a differential current sensed as a current sum across the current of an alternator, it is known to generate an electrical signal that is dependent on voltages on the alternator to ground and in phase with a capacitive leakage component of the differential current and to subtract from the difference current. Practically, this electrical signal is generated by a leakage current, which cause the voltages on the alternator by measuring capacitors to ground. Before subtracting the differential current, the electrical signal is multiplied by a scaling factor. This scaling factor is continuously updated according to a tracking method in such a way that the effective value of the differential current after subtraction of the scaled electrical signal has a minimum at the current value of the scaling factor. The fault current component of the differential current is retained when subtracting the scaled signal. The minimum of the RMS value of the differential current after subtraction of the scaled electrical signal is reached precisely if the scaling factor optimally takes into account the current discharge capacitances. Then the leakage current multiplied by the scaling factor through the measuring capacitors to earth should be exactly the capacitive leakage component of the differential current, and the tracking scaling factor should reflect all real changes in the leakage capacitance of the alternator. It is thus possible to draw conclusions about the current value of the discharge capacitances from the value of the measuring capacitors and the current value of the scaling factor.

In the application of the from the EP 2 372 857 A1 However, it is apparent from the known method and the apparatus described there for carrying out the method that the minimum of the effective value of the differential current achieved by tracking the scaling factor after subtraction of the scaled electrical signal is often significantly greater than the effective value of the fault current component of the differential current. This means that, even after subtraction of the optimally scaled electrical signal, the differential current still comprises an uncompensated component in addition to the fault current of interest, which disturbs the monitoring for the occurrence of a critical fault current.

OBJECT OF THE INVENTION

It is the object of the invention to provide a method for compensating a current flowing through a leakage capacitor portion of a differential current having the features of the preamble of claim 1 and a suitable for implementing this method inverter with the features of the preamble of independent claim 13, with which also the occurrence of a small compared to the differential current critical leakage current is reliably detected.

SOLUTION

The object of the invention is achieved by a method having the features of independent claim 1 and an inverter having the features of independent claim 13. Advantageous embodiments of the The method according to the invention and the inverter according to the invention are defined in the dependent claims.

DESCRIPTION OF THE INVENTION

In the method according to the invention for compensating a fraction of a differential current flowing via a leakage capacitance, the differential current is detected as a current sum via lines which carry the current of an alternating-current generator. The considered lines may carry a DC input current or an AC output current of an AC generator inverter. Importantly, account is taken of a complete set of lines through which the AC generator current flows.

Further, in the inventive method, starting from a Meßableitstrom, the voltages on the alternator cause at least one measuring capacitor to ground, a signal generated, typically at least at first as an electrical signal. The Meßableitstrom through the at least one measuring capacitor flows parallel to the capacitive leakage current from the alternator to ground. In other words, the measuring capacitor is connected in parallel with the leakage capacitance of the alternator, and accordingly the leakage current to the leakage current flowing through the leakage capacitance behaves according to the ratio of the capacitance of the at least one measuring capacitor to the leakage capacitance of the alternator.

Furthermore, in the method according to the invention, an ohmic resistance influencing the signal is predetermined in order to minimize an effective value of the differential current after subtraction of the signal. The predetermined ohmic resistance is either (i) that of a real, in a Meßableitstrom leading line with the at least one measuring capacitor in series resistance or (ii) that of a real resistor, which forms an analog low-pass filter for the signal together with another capacitor , or (iii) that of a fictitious resistor connected in series with the at least one measuring capacitor in a line carrying the measuring leakage current and whose influence on the spectrum of the measuring leakage current is reproduced by analog or digital filtering of the signal.

The fictitious resistance is actually not connected in series in the line leading the measuring leakage current to the at least one measuring capacitor. Instead, by analog or digital filtering of the signal, the influence is simulated that a resistor connected in series with the at least one measuring capacitor would have on the spectrum of the measured leakage current and thus also on the signal.

The present invention is based on the finding that the leakage current flowing in the operation of an AC generator via its leakage capacitance is not a purely capacitive leakage current. In other words, this fraction of the differential current does not have a phase angle of π / 2 to a resistive fault current. The reason is that the leakage current flowing via the leakage capacitance is resistively damped in practice, as a result of which its phase angle is smaller than π / 2 in relation to a resistive fault current. As a consequence, the fraction of the differential current flowing across the leakage capacitance with an electrical signal which is in phase with a purely capacitive measuring leakage current and which correspondingly has a phase angle of π / 2 to a resistive fault current, may not be complete, and often not even substantially be compensated. However, if the leakage current path of an alternator is modeled by a series connection of its leakage capacitance and an effective ohmic resistance, such a compensation is achievable. In the method according to the invention, the effective ohmic resistance is modeled by the predetermined ohmic resistance of the resistor, which is real or fictitious connected in series with the measuring capacitor or real forms together with another capacitor the low-pass filter for the signal. As well as the effective ohmic resistance, which is connected in series with the leakage capacitance of the alternator, such a resistor forms together with the measuring capacitor or the further capacitor an RC element, the phase and amplitude of the signal relative to those applied to the alternator Suspensions of earth shifts. If the RMS value of the differential current after subtraction of the signal at the given ohmic resistance is minimized, this RC element is optimally adapted for imaging the RC element from the leakage capacitance and the series-connected effective resistance.

That a real or fictitious connected in series with the at least one measuring capacitor resistance basically has the same effect on phase and amplitude of the signal relative to the voltages on the alternator as the series connected to the leakage capacitance of the alternator effective ohmic resistance, can be immediately understood. But also the low-pass filter, which is formed by the resistor together with the other capacitor for the signal, has the same influence on the signal. However, in the first case, there is a minimum of the RMS value of the differential current from which the signal is subtracted when the product of the leakage capacitance and the series connected one effective ohmic resistance is equal to the product of the measuring capacitor with the ohmic resistance of the adjustable resistor connected in series, while in the second case the product of the leakage capacitance with the series connected effective ohmic resistance in the minimum of the RMS value of the compensated differential current equal to the product of ohmic Resistor of the variable resistor and the capacity of the other capacitor is.

In principle, the ohmic resistance in the method according to the invention can be preset with a constant value. It turns out, however, that, for example, in a photovoltaic generator as a DC power source of an AC generator, the resistive damping of the discharge current flowing over its leakage capacity is substantially determined by the currently prevailing weather. The resulting variations in the leakage current component of the differential current may even be greater than those due to changes in leakage capacity with the weather. It therefore proves useful to model the effects of changing weather on the phase and amplitude distribution of the leakage current flowing via the leakage capacitance by changing the effective ohmic resistance connected in series with the leakage capacitance. In the method according to the invention, the ohmic resistance, which influences the signal used for compensation, can therefore be set variably. Specifically, it can be tracked in such a way that an effective value of the differential current after subtraction of the signal at the current value of the ohmic resistance has a minimum.

In addition, in the method according to the invention, if a gain of the measurement leakage current or of the signal is tracked as another signal influencing the signal in such a way that the effective value of the differential current after subtraction of the signal at the current value of the amplification factor has a minimum Leakage capacity of the alternator compensated. By simultaneously tracking the ohmic resistance of the variable resistor and the gain of the gain, both components of the RC element which determines the leakage current flowing through the leakage capacitance from the alternator are replicated in the signal. Thus, regardless of the current dominance of the variation of the leakage capacitance or the resistive attenuation of the leakage current flowing through it, optimized compensation of this portion of the differential current is achieved.

In the method according to the invention, the signal can compensate for all components of the differential current which flow via the leakage capacitance. In error-free operation of the alternator, these components make up the differential current at least almost completely, because occurring pure resistive leakage currents remain very small due to high insulation resistance of the alternator. In the case of continuous tracking of the ohmic resistance and of the amplification factor in order to reduce the effective value of the differential current after the subtraction of the signal, the method according to the invention also at least partially compensates for an occurring fault current from the signal. However, since critical fault currents are characterized by much faster increases in their current strength, as possible weather-related changes in the ohmic resistance and the gain factor, it can easily be ensured that the critical fault currents in the inventive method are not compensated before their detection.

A typical critical fault current is characterized by a sudden increase to a current of 30 mA, 60 mA or 150 mA and must be switched off within a period of 300 ms, 150 ms or 40 ms. If every critical increase of the fault current is already detected within 40 ms and the tracking of all tracking variables influencing the signal takes place with a time constant which is 40 ms or greater, such an increase in the inventive method at best during a Nachführschritts and thus only one small part compensates, which does not question its recognition as a critical fault current.

In general, the effective value of the differential current in the method according to the invention can be compared after subtraction of the signal with a limit value for sudden changes and, when the limit value is exceeded, output an error signal, which can result in particular in switching off the alternator. In this case, the tracking of all tracking signals influencing the signal takes place with a time constant which is at least as great as a time constant of the limit value for sudden changes. The time constant of the tracking may also be at least 5 times or 10 times greater than a time constant of the threshold value for sudden changes.

If, in addition, the signal is compared with an absolute limit value taking into account all the tracking quantities influencing the signal and an error signal is output when the limit value is exceeded, the method according to the invention also displays critical leakage currents which regardless of their composition are cause to turn off the alternator.

Specifically, in the method according to the invention, the signal can be subtracted directly from the differential current as an analog electrical signal by means of a differential current transformer detecting the differential current. In this case, the uncompensated residual current itself is not measured at all. However, it can be reconstructed from the measured compensated differential current and the signal.

Alternatively, in the method according to the invention, the signal can be subtracted from a voltage signal of a differential current-detecting summation current transformer, the voltage signal indicating the uncompensated differential current. This subtraction can also be based on digitized values. For this purpose, the signal - before or after any amplification and before or after any filtering - digitized and then subtracted in digital form from the also digitized voltage signal of the summation current transformer.

The ohmic resistance of an adjustable resistor can be tracked in the inventive method according to a basically known tracking method. In parallel, the amplification factor of the gain of the Meßableitstroms or the signal can be tracked by a tracking method. So that these two tracking methods do not interfere with each other, they should have different time constants. In this case, the time constant of the tracking method for the ohmic resistance of the variable resistor can be greater than that for the gain factor, because fluctuations in the leakage capacitance, for example in a photovoltaic panel, can occur at a higher frequency than fluctuations in the series-connected effective ohmic resistance.

In the method according to the invention, it may be preferable to measure the voltages on the alternator with respect to ground in front of the inverter, i. H. at its DC input side, and the residual current after the inverter, d. H. on the AC output side, to be considered. Then, the differential current also includes fault currents occurring in the inverter, and the occurrence of such fault currents is detected when monitoring the differential current. In addition, particular consideration is given to the voltages on the alternator to ground which are applied to the DC input side of the inverter to a DC generator, for example to a photovoltaic generator, with high leakage capacitance to ground.

Due to the continuous adaptation of the ohmic resistance and the amplification factor, with the method according to the invention not only a very good compensation of the portion of the differential current flowing via the leakage capacitance of the alternator to earth can be achieved; It is also possible to deduce from the capacitance of the at least one measuring capacitor and the current values of the ohmic resistance and the amplification factor - optionally taking into account the value of the further capacitor - the leakage capacitance and the effective ohmic resistance connected in series therewith. These values, as well as the immediate values of the ohmic resistance and the amplification factor, can additionally be monitored for irregularities or even output for informational purposes only.

In practice, it turns out that with the method according to the invention the compensation of the components flowing through the leakage capacitance on the differential current, which is measured at an inverter, with the electric current from a photovoltaic generator is fed into an AC network, to a much greater extent as if a purely capacitive Meßableitstrom is adjusted only by tracking its gain to the current dissipation capacity of the photovoltaic generator.

An inverter according to the invention, which is suitable for carrying out the method according to the invention, has a differential current detection device for detecting a differential current as current sum via lines which carry a total input DC current or a total DC output current of the inverter. Furthermore, a compensation device for compensating a conditional by the operation of the inverter portion of the differential current is present. This compensation device has at least one measuring capacitor in a current path from the DC voltage side of the inverter to ground, a signal generator, which generates a signal based on a Meßableitstrom through the at least one measuring capacitor, a subtractor, which subtracts the signal from the detected differential current, and a rms value measuring device which measures an RMS value of the differential current after subtraction of the signal by the subtractor.

According to the invention, an ohmic resistance influencing the signal can be predetermined in order to minimize the rms value measured by the rms value measuring device. The predeterminable ohmic resistance is, as in the method according to the invention, either (i) that of a real resistor, which in one the Meßableitstrom or (ii) that of a real resistor which together with another capacitor forms a low pass filter for the signal, or (iii) that of a fictitious resistor in series with the at least one measuring capacitor, whose influence on the spectrum of the measuring leakage current is simulated by a filter for the signal. In other words, the filter simulates an adjustable resistor connected in series with the measurement capacitor with respect to its effect on the phase and amplitude distribution of the signal. This filter can in principle be realized by the analog low-pass filter, which is formed by the real resistor together with the further capacitor for the signal. But it can also be an adjustable digital filter, for example.

The specifiable ohmic resistance may in particular be that of an adjustable resistor. Each real adjustable resistor can be realized by a potentiometer.

In addition, the inverter according to the invention can have an amplifier for the measured leakage current or the signal, wherein a gain factor of the amplifier can be specified as a further variable influencing the signal in order to minimize the effective value measured by the rms value measuring device.

A changing device of the inverter according to the invention can be provided to automatically track each variable variable influencing the signal in such a way that the effective value measured by the rms value measuring devices has a minimum. In particular, this alteration means may implement a tracking method for each of these quantities.

In addition, an evaluation device can be provided which compares the rms value measured by the rms value measuring device with a limit value for sudden changes and / or the signal with an absolute limit value and outputs an error signal when one of the limit values is exceeded. This error signal can lead to the immediate shutdown of the inverter.

The subtractor of the inverter according to the invention can subtract the signal from the differential current by means of a summation current transformer detecting the differential current. Alternatively, the subtractor may subtract the signal from a voltage signal of a differential current sensing sum current transformer. In this case, A / D converters can be provided which digitize the signal and the voltage signal of the summation current transformer before the subtractor. The A / D converter for the signal may be arranged in front of the amplifier for the signal, the amplifier then being to be implemented digitally. The digitizer may be arranged in front of the filter for the signal, which simulates the influence of the fictitious variable resistor in the line carrying the measuring leakage current, if this is carried out digitally.

Advantageous developments of the invention will become apparent from the claims, the description and the drawings. The advantages of features and of combinations of several features mentioned in the description are merely exemplary and can take effect alternatively or cumulatively, without the advantages having to be achieved by embodiments according to the invention. Without thereby altering the subject matter of the appended claims, the following applies with regard to the disclosure content of the original application documents and the patent: Further features can be found in the drawings - in particular the illustrated arrangement and operative connection of several components. The combination of features of different embodiments of the invention or of features of different claims is also possible deviating from the chosen relationships of the claims and is hereby stimulated. This also applies to those features which are shown in separate drawings or are mentioned in their description. These features can also be combined with features of different claims. Likewise, in the claims listed features for further embodiments of the invention can be omitted.

The features mentioned in the patent claims and the description are to be understood in terms of their number that exactly this number or a greater number than the said number is present, without requiring an explicit use of the adverb "at least". For example, when talking about an element, it should be understood that there is exactly one element, two elements or more elements. These features may be supplemented by other features or be the only characteristics that make up the product in question.

The reference numerals contained in the claims do not limit the scope of the objects protected by the claims. They are for the sole purpose of making the claims easier to understand.

BRIEF DESCRIPTION OF THE FIGURES

The invention is explained in more detail below with reference to embodiments with reference to the accompanying drawings and described.

1 shows a common mode model of a photovoltaic generator.

2 shows a compensation according to the invention of a current flowing through a leakage capacitance portion of a differential current, taking into account the common mode model according to 1 ,

3 shows a further compensation of the current flowing through a leakage capacitance portion of the differential current with further consideration of the common mode model according to 1 ,

4 shows a further embodiment of the further compensation of the current flowing through a leakage capacitance portion of the differential current.

5 shows yet another embodiment of the further compensation of the current flowing via a leakage capacitance portion of the differential current; and

6 FIG. 12 shows a combination of two tracking methods for parallel adaptation of two quantities, which in the embodiments of the further compensation of the portion across the leakage current flowing through a leakage capacitance according to the 3 to 5 can be used.

DESCRIPTION OF THE FIGURES

1 shows a common mode model of a photovoltaic generator 1 , The common mode model illustrates the causes of leakage currents coming from the photovoltaic generator 1 due to voltages on the photovoltaic generator 1 flow from earth to earth. The common mode model is divided into a resistive branch 2 and a branch 3 in which a leakage capacitance C _gene between the photovoltaic generator 1 and earth is effective, so over the capacitive branch 3 only one leakage current i _{gene can} flow, caused by changes in the voltages to earth at the photovoltaic generator 1 is driven. Although in the branch 3 is connected in series with the leakage capacitance C _Gen an effective resistance R _gene , which indicates a resistive attenuation of the leakage current flowing through the leakage capacitance C _Gen , becomes the branch 3 here also as a capacitive branch 3 designated. The effective ohmic resistance R _gene together with the leakage capacitance C _Gen an RC element, the frequency-dependent phase and amplitude of the capacitive branch 3 flowing leakage current i _gene compared to those at the photovoltaic generator 1 applied voltages determined. The resistive branch 2 of the common mode model is determined by the insulation resistance R _{iso of} the photovoltaic generator 1 determined, in the proper state of the photovoltaic generator 1 much larger than the resistance R _gene is. Therefore, it is usually through the resistive branch 2 flowing resistive leakage much smaller than that through the capacitive branch 3 flowing leakage current i _gene . In the event of an insulation fault, however, the insulation resistance R _iso breaks down and a significant resistive leakage current through the resistive branch occurs 2 on.

If a current sum over connecting lines 4 and 5 of the photovoltaic generator 1 is formed, this is not zero, but there is a differential current having the Ableitstromanteile which, via the capacitive branch 3 and the resistive branch 2 flow. It is also a large differential current, on a large leakage current i _gene by the capacitive branch 3 is not yet an indication of a fault of the photovoltaic generator 1 because, depending on the type and size of its photovoltaic panels and the weather, the leakage capacitance C _{gene can be} very large and the resistance R _{gene can be} very small and because when connecting the ungrounded photovoltaic generator 1 To an inverter large voltage changes to ground on the connecting cables 4 and 5 may occur due to the operation of the inverter. On the other hand, that is through the resistive branch 2 flowing leakage current in the proper state of the photovoltaic generator 1 only very small. This means that every increase in the resistive leakage current must be interpreted as a fault current. For very large leakage current i _gene through the capacitive branch 3 However, it is difficult to change the resistive leakage current, ie occurring fault currents, by monitoring the differential current across the connecting lines 4 and 5 to recognize. This is not only due to the different sizes of these two leakage currents, but also because they add up vectorially. However, the monitoring of the differential current to occurring fault currents can be greatly facilitated by the proportion of the leakage current i _gene through the capacitive branch 3 is compensated at the detected differential current. For this purpose, a signal is to be generated, which is the leakage current i _gene through the capacitive branch 3 as accurately as possible. When this signal is subtracted from the differential current, only the resistive leakage current of interest remains through the resistive branch 2 and thus a possible fault current.

An embodiment of such a compensation of the portion of the differential current flowing via the leakage capacitance C _Gen is in 2 illustrated. Here is shown that the connecting cables 4 and 5 from the photovoltaic generator 1 to a inverter 6 lead, that of the photovoltaic generator 1 generated electricity converts into an alternating current and converts this alternating current into an alternating current network 7 feeds. The photovoltaic generator 1 and the inverter 6 together form an alternator 8th , In 2 is from the common mode model according to 1 only the capacitive branch 3 reproduced because of the implementation of the compensation of a faultless alternator 8th is assumed, in which over the resistive branch 2 according to 1 no relevant leakage current flows. The operation of the inverter 6 conditionally changes in the voltages on the connecting cables 4 and 5 towards earth, which has a leakage current i _gene through the capacitive branch 3 float. This leakage current i _gene occurs as part of a differential current, which with a summation current transformer 9 over the connection lines 4 and 5 is detected. For compensation of the leakage current i _gene at the voltage signal U _{DI of} the summation current transformer 9 are measuring capacitors 10 with its one side to the connection lines 4 and 5 connected and grounded on their other side. About the measuring capacitors 10 a measuring leakage current i _{test flows} to earth, the same operating voltages on the connecting cables 4 and 5 is driven as the leakage current i _gene . With a current measuring device 11 the measured leakage current i _{test is} measured, and the current measuring device 11 outputs a voltage signal u _mess as a measure of the measured leakage current i _test . In that regard, the current measuring device is used 11 here as a signal generator 12 for a signal 13 , which starts from the measuring leakage current i _test through the measuring capacitors 10 is generated. The signal 13 is then using a low pass filter 14 filtered, which has an adjustable resistor 15 and a capacitor leading to ground 16 having. The lowpass filtered signal 13 is calculated as compensation _current i _komp by N compensation windings 17 around the core 18 of the summation current transformer 9 guided until it flows over a protective resistor R _b to earth. From a logic 19 becomes an effective value of the output signal U _{DI of} the summation current transformer 9 continuously by tracking the ohmic resistance of the adjustable resistor 15 minimized. In this way, the replica of the capacitive branch 3 the common mode model of the photovoltaic generator 1 through the measuring capacitors 10 , the signal generator 12 , the low-pass filter 14 , the N compensation windings 17 and the protective resistor R _b adapted to the current value of the effective ohmic resistance R _Gen connected in series with the leakage capacitance C _Gen. It is assumed here that the leakage capacitance C _gene remains at least substantially constant and that (i) the measuring capacitors 10 or their capacity C _PE , (ii) one of the current measuring device 11 predetermined ratio between the measured Testableitstrom i _test and the output from her voltage signal u _mess , (iii) the number N of the compensation windings and (iv) the protective resistor R _b are well tuned to this substantially constant leakage capacitance C _gene . Any change in the ohmic resistance R _Gen , however, is due to a change in the ohmic resistance of the variable resistor 15 followed until the minimum of the RMS value of U _DI is reached again. Ideally, the rms value of U _{DI is} compensated to zero. Thus, an errant fault current makes the effective value of U _DI alone and can therefore be very easily compared with a critical fault current limit. In addition, the signal can 13 be monitored as a measure of the absolute leakage current to the observance of a limit value.

3 shows an alternator 8th which, except for a further adaptation to changing values of the leakage capacitance C _Gen and the ohmic resistance R _Gen in the capacitive branch 3 of the common mode model according to 1 in the compensation of not by a ground fault during operation of the inverter 6 conditional leakage current i _{gene of} the embodiment of the alternator 8th according to 2 equivalent. In concrete terms, the logic works 19 here as a function of the effective value U _{DI of} the output signal of the summation current transformer 9 on an amplifier 20 and a filter 21 for the signal 13 one. The filter can do this 21 the low pass filter 14 according to 2 be, or it may otherwise be one with the two measuring capacitors 10 in a line leading the measurement leakage current i _test 22 series connected fictitious adjustable resistor in terms of its influence on the phase and amplitude of the signal 13 relative to the voltages on the connecting cables 4 and 5 simulate. Here is the ohmic resistance of this adjustable resistor of the logic 19 so trackable that the rms value of the output signal U _{DI of} the summation current transformer 19 reached its minimum. At the same time, the logic varies 19 a gain G of the amplifier 20 also with the aim of minimizing the effective value of the voltage signal U _DI . Hereby additionally occurring variations of the leakage capacity C _{Gen are} simulated. With optimum tuning of the amplifier 20 and the filter 21 is the voltage signal U _DI in the error-free operation of the alternator 8th zero, thus showing very sensitively everyone on the alternator 8th occurring fault current. So that such a fault current does not occur with the signal 13 is compensated, all tracked quantities are the signal 13 influence, ie the ohmic resistance R _adj and the gain G, nachzuführen with a time constant, which is preferably not smaller than a time constant, which is relevant for a sudden increase in critical leakage current. If any critical fault current does not need to be turned off faster than within 40ms and accordingly the rms value of the voltage signal U _{DI is} monitored for increases within 40 ms, the time constant of the tracking can also be set to 40 ms.

4 illustrates the following possible variations in the alternator 8th across from 3 : The summation current transformer 9 does not detect the residual current across the connecting cables here 4 and 5 of the photovoltaic generator 1 but over from the inverter 6 to the AC mains 7 leading output lines 23 . 24 . 25 , Thus, fault currents are detected, which are in the range of the inverter 6 occur. Next is instead of the filter 21 the low pass filter 14 for the signal 13 intended.

u

die treibende Spannung gegenüber Erde ist,

C_Gen

die Ableitkapazität ist,

R_Gen

der mit der Ableitkapazität C_Gen in Reihe geschaltete ohmsche Widerstand ist,

C_PE

die Kapazität der Messkondensatoren 10 ist,

S

das Verhältnis des Spannungssignals u_mess zu dem Testableitstrom i_test ist,

G

der Verstärkungsfaktor des Verstärkers 20 ist,

R_adj

der ohmsche Widerstand des verstellbaren Widerstands 15 ist,

C_adj

die Kapazität des Kondensators 16 des Tiefpassfilters 14 ist,

N

die Anzahl der Kompensationswicklungen 17 ist und

R_b

der ohmsche Widerstand des Schutzwiderstands 26 ist.

Taking into account the in 4 Registered symbols: With faultless alternator 8th the minimum of the rms value of the voltage signal U _DI is zero and the capacitive leakage current i _{Gen is} equal to the compensation current i _comp . From this follows the equation: u / (R _Gen + 1 / jωC _gene ) = u × 2jωC _PE × S × G × 1 / (1 + jωR _adj C _adj ) × N / R _b , in which

u

is the driving tension towards earth,

C _gene

the leakage capacity is,

R _gene

the series resistor connected in series with the leakage capacitance C _{Gen is} ohmic resistance,

C _PE

the capacity of the measuring capacitors 10 is

S

the ratio of the voltage signal u _mess to the _test leakage current i _test is,

G

the amplification factor of the amplifier 20 is

R _adj

the ohmic resistance of the adjustable resistor 15 is

C _adj

the capacity of the capacitor 16 the low-pass filter 14 is

N

the number of compensation windings 17 is and

R _b

the ohmic resistance of the protective resistor 26 is.

Furthermore, in the minimum of the effective value of the voltage signal U _DI : R _Gen × C _Gen = R _adj × C _adj .

This results in the leakage capacitance C _Gen in the capacitive Ableitzweig 3 to: C _Gen = (2 * C _PE ) * S * G * N / R _b

And connected with the discharge capacitance C _gene in series ohmic resistance R _gene to: R _gene = R _adj * C _adj / C _gene .

So can with the alternator 8th not only flowing component i _gene are compensated to the differential current, but it can also be determined that this leakage current i _{Gen-determining} parameters R _gene and C _gene on the leakage capacitance C _gene.

In the embodiment of the alternator 8th according to 5 is the picture of the capacitive branch 3 the common mode model of the photovoltaic generator 1 compared to the embodiment according to FIG 4 modified in the following details: The measuring leakage current i _test flows only through a single measuring capacitor 10 that with the one connecting line 5 of the photovoltaic generator 1 connected is. But it also flows through an adjustable resistor 27 who is directly in the line 22 from the measuring capacitor 10 is arranged to earth. The ohmic resistance R _{adj of} this adjustable resistor 27 is from the logic 19 changed to the RMS value of a difference signal from a subtractor 29 to minimize. The subtractor 29 subtracted from the voltage signal U _{DI of} the summation current transformer 9 that with an A / D converter 30 digitized with another A / D converter 31 digitized and with this digitally trained amplifier 20 amplified signal 13 , By changing the ohmic resistance R _{adj of} the adjustable resistor 27 fits the logic 19 here the replica of the capacitive branch 3 the common mode model of the photovoltaic generator 1 so that the product of the capacitance C _{PE of} the measuring capacitor 10 and the ohmic resistance R _{gene of} the adjustable resistor 27 is equal to the product of the leakage capacitance C _gene with the series connected effective resistance R _gene . The simultaneous change of the gain G of the amplifier 20 scales the signal 13 to calculate the difference between the capacitance C _{PE of} the measuring capacitor 10 and the leakage capacitance C _gene . Accordingly applies here: C _gene = C _PE × S × G and R _gene = R _adj × C _PE / C _gene .

By not here in the range of the summation current transformer 9 Compensation of the capacitive leakage current must be effected, for example, by the A / D converter 30 be designed for the full differential current and still be sensitive enough to fully detect the fault current component of the differential current and its changes.

6 illustrates a combination of two tracking methods, according to which the logic 19 in the design of the alternator 8th according to the 3 to 5 on the one hand the ohmic resistance R _{adj of} the adjustable resistor 27 or the adjustable resistor in the low-pass filter 14 or of the filter 21 simulated adjustable resistance and on the other the gain G of the amplifier 20 as the signal 13 can optimize influencing variables. Between the signal thus influenced 13 , ie, the compensation _current i _comp , and the differential _current i _Diff is supplied to the subtractor 29 in the form of the summation current transformer 9 formed the compensated difference current reproducing voltage signal U _DI . An RMS measuring device 28 forms an effective value of U _DI . The signal then branches to the two individual tracking methods for optimizing the ohmic resistance of the variable resistor (below) and the gain of the amplifier 20 (above). The signal is each a sample & hold stage 32 subsequently supplied with a subtraction node the current value of the respective previous value of the signal 13 deducted. If the output signal of the subtraction node is positive, it means that the current value of the signal 13 closer to its desired minimum than the previous value. A previously chosen direction of changing the value of each of the signal 13 influencing size is therefore to be maintained. This is done via an operational amplifier 34 a JK flip-flop 35 controlled, whose output signal with an integrator 36 is integrated. The JK flip-flop 35 will be with the same clock signal 37 clocked like the Sample & Hold level 32 , Here is the time constant of the clock signal 37 for tracking the ohmic resistance R _{adj of} the adjustable resistor 27 here larger than that for the tracking of the gain G in order to avoid mutual interference of the tracking of the two sizes. Typically, the time constant for the tracking of the ohmic resistance is ten times greater than that for tracking the gain G. In addition, both time constants are significantly greater than the time constant valid for a critical fault current, ie the time during which a critical fault current reaches a certain current level , As an alternative to tracking both the signal 13 influencing variables with different time constants, an alternating tracking of the two sizes is possible. That is, while the value of one size is optimized, the value of the other size is recorded and vice versa. With a suitable choice of the integration constants of the rms value measuring device 28 and the integrator 36 and the time constant of the clock signals 37 The tracking methods very quickly oscillate to values at which the rms value of the voltage signal U _{DI is kept} at its minimum. This minimum is zero in the error-free case. The rms value of a zero voltage rising voltage signal U _DI then corresponds to the rms value of a fault current flowing to the ac generator 8th occurs.

LIST OF REFERENCE NUMBERS

1

photovoltaic generator

2

resistive branch

3

capacitive branch

4

connecting cable

5

connecting cable

6

inverter

7

AC power

8th

Alternator

9

Summation current transformer

10

measuring capacitor

11

Current measurement device

12

signal generator

13

signal

14

Low Pass Filter

15

adjustable resistance

16

capacitor

17

compensation winding

18

Core of the summation current transformer 9

19

logic

20

amplifier

21

filter

22

management

23

output line

24

output line

25

output line

26

protection resistor

27

adjustable resistance

28

RMS measuring device

29

subtractor

30

A / D converter

31

A / D converter

32

Sample & hold stage

33

subtracting node

34

operational amplifiers

35

JK flip-flop

36

integrator

37

clock signal

C _gene

leakage capacitance

R _gene

ohmic resistance in series with the leakage capacitance C _gene

i _gene

capacitive leakage current

C _PE

Capacitance of a measuring capacitor 10

i _test

Messableitstrom

u _mess

Voltage signal of the current measuring device 11

S

Ratio of u _mess to i _test

G

Amplification factor of the amplifier 20

R _adj

ohmic resistance of the adjustable resistor 15 . 27

C _adj

Capacitance of the capacitor 16

R _b

ohmic resistance of the protective resistor 26

I _Diff

differential current

i _comp

compensating current

U _DI

Voltage signal of the summation current transformer 9

N

Number of compensation turns 17

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• EP 2372857 A1 [0004, 0005]

Claims (20)

A method for compensating for a via a leakage capacitance (C _gene) flowing proportion of a difference current (I _DIFF), - wherein the differential current (I _DIFF) (as current sum through lines 4 . 5 or 23 to 25 ) detecting the current of an alternator ( 8th ), where a signal ( 13 ) is generated from a Meßableitstrom (i _test ), the voltages at the alternator ( 8th ) by at least one measuring capacitor ( 10 ) cause earth, - at least one signal ( 13 ) is given to influence an effective value of the differential _current (i _Diff ) after subtraction of the signal ( 13 ), characterized in that the at least one predefined variable is an ohmic resistance (R _adj ) of a resistance ( 15 . 27 ), which - in a line carrying the measuring leakage current (i _test ) ( 22 ) with the at least one measuring capacitor ( 10 ) is connected in series or - together with another capacitor ( 16 ) a low-pass filter ( 14 ) for the signal ( 13 ) or - fictitiously in a line carrying the measuring leakage current (i _test ) ( 22 ) with the at least one measuring capacitor ( 10 ) is connected in series and its influence on the spectrum of the Meßableitstroms (i _test ) by filtering the signal ( 13 ) is modeled. Method according to Claim 1, characterized in that the ohmic resistance (R _adj ) of the resistor ( 15 . 27 ) is tracked in such a way that an effective value of the differential _current (i _Diff ) after subtraction of the signal ( 13 ) has a minimum at the current value of the ohmic resistance (R _adj ). Method according to one of the preceding claims, characterized in that as further the signal ( 13 ) an amplification factor (G) of a gain of the Meßableitstroms (i _test ) or the signal ( 13 ) is tracked in such a way that the effective value of the differential _current (i _Diff ) after subtraction of the signal ( 13 ) has a minimum at the current value of the gain (G). Method according to claim 2 or claim 3, characterized in that the ohmic resistance (R _adj ) of the resistance ( 15 . 27 ) and / or the amplification factor (G) is tracked continuously according to a tracking method. Method according to claim 2 and claim 3, characterized in that the ohmic resistance (R _adj ) of the resistance ( 15 . 27 ) and the amplification factor (G) are tracked consecutively according to a tracking method, wherein a time constant of the one tracking method is at least 5 times greater than a time constant of the other tracking method. Method according to Claim 2 and Claim 3 or one of Claims 4 and 5 dependent thereon, characterized in that from the capacitance (C _PE ) of the at least one measuring capacitor ( 10 ) and the current values of the ohmic resistance (R _adj ) and the amplification factor (G) to a leakage capacitance (C _gene ) and a series connected with this effective ohmic resistance (R _gene ) is closed. Method according to one of the preceding claims, characterized in that the effective value of the differential _current (i _Diff ) after subtraction of the signal ( 13 ) is compared with a limit value for sudden changes and an error signal is output when the limit value is exceeded. Method according to claim 7 and claim 4 or 5, characterized in that the tracking of all the signal ( 13 ) tracking variables having a time constant which is at least as large as a time constant of the limit value for sudden changes. Method according to one of the preceding claims, characterized in that the signal ( 13 ) with an absolute limit value and an error signal is output when the limit value is exceeded. Method according to one of the preceding claims, characterized in that the signal ( 13 ) by means of a differential current (i _Diff ) detecting summation current transformer ( 9 ) is subtracted from the difference _current (i _Diff ). Method according to one of the preceding claims 1 to 9, characterized in that the signal ( 13 ) of a voltage signal (U _DI ) of a differential current (i _Diff ) detecting the summation current transformer ( 9 ) is subtracted. Method according to claim 11, characterized in that the signal ( 13 ) and the voltage signal (U _DI ) of the summation current transformer ( 9 ) and the digitized signal ( 13 ) is subtracted from the digitized voltage signal (U _DI ). Inverter ( 6 ) with, - a differential current detection device for detecting a differential current (i _Diff ) as a current sum via lines ( 4 . 5 or 23 to 25 ), which has a total input DC current or a whole AC output current of the inverter ( 6 ), and - compensation means for compensating a portion of a differential current (i _Diff ) flowing via a leakage capacitance (C _Gen ), the compensation unit comprising: - at least one measuring capacitor ( 10 ) in a current path from the DC side of the inverter ( 6 ) to earth, - a signal generator ( 12 ), which starting from a Meßableitstrom (i _test ) through the at least one measuring capacitor ( 10 ) a signal ( 13 ), - a subtractor ( 29 ), the signal ( 13 ) is subtracted from the difference _current (i _Diff ), and - an effective value measuring device ( 28 ), which has an effective value of the differential _current (i _Diff ) after subtraction of the signal ( 13 ) by the subtractor ( 29 ), at least one signal ( 13 ) influencing size can be predetermined to the of the RMS ( 28 ) to minimize the measured RMS value, characterized in that - the at least one signal ( 13 ) an influencing predefinable variable is an ohmic resistance (R _adj ) of a resistance ( 15 . 27 ), which - in a line carrying the measuring leakage current (i _test ) ( 22 ) with the at least one measuring capacitor ( 10 ) is connected in series or - together with another capacitor ( 16 ) a low-pass filter ( 14 ) for the signal ( 13 ) or - fictitiously in a line carrying the measuring leakage current (i _test ) ( 22 ) with the at least one measuring capacitor ( 10 ) is connected in series and its influence on the spectrum of the Meßableitstroms (i _test ) by a filter ( 21 ) for the signal ( 13 ) is modeled. Inverter ( 6 ) according to claim 13, characterized in that the predetermined ohmic resistance (R _adj ) that of an adjustable resistor ( 15 . 27 ). Inverter ( 6 ) according to claim 14, characterized in that an amplifier ( 20 ) for the measuring leakage current (i _test ) or the signal ( 13 ) and that a gain (G) of the amplifier (G) 20 ) as further the signal ( 13 ) influencing size can be predetermined to the of the RMS ( 28 ) to minimize the measured RMS value. Inverter ( 6 ) according to claim 14 or 15, characterized in that a logic ( 19 ), each of which is the signal ( 13 ) so that the predeterminable variable can be corrected in an automatic manner such that the value determined by the rms value measuring device ( 28 ) measured RMS value has a minimum. Inverter ( 6 ) according to one of the preceding claims 13 to 16, characterized in that an evaluation device is provided, which by the RMS ( 28 ) measured RMS value with a threshold for sudden changes and / or the signal ( 13 ) compares with an absolute limit value and outputs an error signal when one of the limit values is exceeded. Inverter ( 6 ) according to one of the preceding claims 13 to 17, characterized in that the subtractor ( 29 ) the signal ( 13 ) by means of a differential current (i _Diff ) detecting summation current transformer ( 9 ) is subtracted from the difference _current (i _Diff ). Inverter ( 6 ) according to one of the preceding claims 13 to 17, characterized in that the subtractor ( 29 ) the signal ( 13 ) of a voltage signal (U _DI ) of a differential current (i _Diff ) detecting the summation current transformer ( 9 subtracted. Inverter ( 6 ) according to claim 18, characterized in that A / D converter ( 30 . 31 ) are provided, the signal ( 13 ) and the voltage signal (U _DI ) of the summation current transformer ( 9 ) before the subtractor ( 29 ) digitize.

Images