WO2014086783A2

WO2014086783A2 - Method for preventing excess current in a power electronics device and power electronics device

Info

Publication number: WO2014086783A2
Application number: PCT/EP2013/075396
Authority: WO
Inventors: Peter Schade; Bjoern Ellermeyer
Original assignee: Sma Solar Technology Ag
Priority date: 2012-12-05
Filing date: 2013-12-03
Publication date: 2014-06-12
Also published as: DE102012111855B4; WO2014086783A3; DE102012111855A1

Abstract

To prevent excess current in a power electronics device (1), which has a signal processing unit (8) for generating a variable for the strength of a current (15) occurring in the power electronics device (1), wherein a change to an input variable (12) of the signal processing unit (8) affects the current (15) in a manner so as to be delayed by a first time period, the input variable (12) of the signal processing unit (8) is evaluated for fulfilment of a criterion, on the basis of which the risk of excess current can be detected, and if the criterion is fulfilled, the operating control (10) of the power electronics device (1) is manipulated in such a manner that excess current is prevented in the power electronics device (1) before a critical current rise occurs, wherein the detection of the input variable (12), the evaluation of said input variable with regard to the criterion and manipulation of the operating control (10), including an effect of such manipulation on the current (15), is carried out within a second time period which is shorter than the first time period.

Description

Method for preventing overcurrents in a power electronic device and power electronic device

description

The present invention relates to a method for avoiding overcurrents in a power electronic device. Furthermore, the invention relates to a power electronic device with a device for preventing overcurrents.

Power electronic devices are often used to convert electrical energy, in particular between two current forms, that is, from alternating current to direct current, from direct current to alternating current, or from an alternating current with a certain frequency to an alternating current having a different frequency. For this purpose, currents flow in so-called Leistungsstrompfaden the power electronic device, the strength of which often by controlling components of the power electronic device, such as

Power semiconductors, is set so as to control a power provided by the power electronic device at the output thereof or a power transmitted by the power electronic device.

To generate manipulated variables for setting the currents in the power current paths, power electronic devices usually have a signal processing in which control algorithms are implemented in many cases. In particular, in the case of regulations, operating states can occur which lead via the manipulated variables at the output of the signal processing to such high currents in the power current paths that the components contained therein are excessively stressed and thereby, in particular during prolonged exposure to such currents, destroyed. Such currents exceeding a permissible level are referred to as overcurrents, the occurrence of which in power electronic devices must be avoided.

To avoid overcurrents in power electronic devices, it is therefore customary to arrange current measuring devices in such current paths in which overcurrents may occur and to compare the currents measured thereby with a threshold value, which is the current limit for an uncritical operating state, and then when the threshold value is exceeded to carry out by the measured current measures that lead to an interruption of the current flow in the current path. In the prior art approach described above, since the current has often already exceeded a critical value when it exceeds the threshold, fast, mostly dedicated, hardware for detecting the overcurrent and initiating the interception measures is required to prevent components from being destroyed , An inexpensive realization using in

Power electronic devices for purposes of operation or signal processing anyway existing processors is not regularly possible. At least it requires a correspondingly low interpretation of the threshold in order to detect this quickly enough in case of increasing overcurrent and to be able to interrupt the flow of current. However, a low interpretation of the threshold value can lead to an already uncritical current, which rises only insignificantly above the low threshold, erroneously recognized as dangerous overcurrent and causes an unnecessary interruption of the current flow in the current path, which then usually with a least Temporary failure of the power electronic device is connected.

A problem in setting a suitable threshold in the described procedure according to the prior art is in particular that one in the

Leistungsstrompfaden occurring current rate of increase is not necessarily known from the outset and may possibly vary, especially if a control is included in the signal processing. This existing ignorance of the rate of current rise, even with the use of fast, dedicated hardware, requires a correspondingly low threshold design to safely avoid overcurrents, which, as described, may result in erroneous overcurrent detection.

In order to avoid erroneous overcurrent detection, JP2012125092A proposes to increase the threshold value for detecting overcurrents for lower DC input voltages of the inverter according to a predetermined characteristic using the prior art method for an inverter as described above. When the threshold value is exceeded, the current flow is interrupted by switching off the control signals for the power semiconductors (IGBTs). Although the adjustment of the threshold value described in the publication JP2012125092A reduces the problems described above, it does not solve them. When using processors or slow hardware as well as due to the not necessarily known current slew rate still has the threshold off Safety reasons are chosen to be correspondingly low, so that an erroneous

Detection of overcurrents can not be excluded.

The document US 2012/0212871 A1 discloses an overcurrent detection circuit for a battery. The battery is part of a "Battery Pack", which includes a control unit, is set by the dependent on a voltage value of the battery, a current flow direction at terminals of the "Battery Pack" either for charging or discharging the battery. Parallel to the control unit, a circuit for overcurrent protection in the battery is provided, which instead of a conventional measurement of the

Current flow into the battery compares their current voltage value with a time-delayed voltage value of the battery and when exceeding a predetermined

Threshold present by the difference of these at different times

Voltage values of the battery engages in the control unit such that the flow of current in the battery is interrupted. According to the description recognizes the circuit for overcurrent protection in the document US 2012/0212871 A1, the overcurrents in the battery, however, only when an overcurrent already flows or has flowed.

US Pat. No. 7,324,341 B2 describes a network-feeding inverter with a DC / DC converter for current shaping and a subsequent pole reverser. In addition to measures to protect against too high or too low voltages and too high temperatures, this inverter also provides measures to protect against excessive currents. However, due to the measurement of the current at the output of the DC / DC converter used for current shaping, overcurrents are also only recognized if they have already occurred.

It is therefore an object of the invention to provide a method for avoiding overcurrents in a power electronic device, which in its design does not require consideration of a potentially further increasing current after its detection and which can preferably be implemented using a processor or even slower hardware.

This object is achieved by a method having the features of the independent claim 1 or by a power electronic device for carrying out such a method according to claim 11. The dependent claims relate to others

Embodiments of the invention. The method according to the invention for avoiding overcurrents is used in a power electronic device which has signal processing for generating a manipulated variable for the strength of a current occurring in the power electronic device, wherein an effect of changing an input of the signal processing to the current delays by a first period of time entry. In the signal processing may be both an analog and a digital signal processing, as well as a mixed analog and digital signal processing, which then outputs a control variable for the power directly to, for example, a power semiconductor. Any existing driver circuits or a signal processing for generating, for example, a pulse width modulation (PWM) to provide the current, are thus to be regarded as components of the signal processing. Also, the adjustment of the strength of the current in a power current path itself can be analog or digital.

In the method according to the invention, as in the case of known methods, a current value in the power current path of the power electronic device is not evaluated for the presence of an overcurrent, but the input variable of the signal processing is evaluated with regard to the fulfillment of a criterion based on which the risk of an overcurrent occurs is recognizable. Upon fulfillment of the criterion, intervention in the operational management of the power electronic device is then made in such a way that the occurrence of overcurrents in the power electronic device is avoided. Characterized in that in the inventive method, the detection of the input variable, the evaluation of this input variable in terms of the criterion and the intervention in the management

including an effect of this intervention on the current within a second period of time shorter than the first period of time by which an effect of changing an input of the signal processing on the current occurs delayed, overcurrents in power current paths of the power electronic device can be avoided before it even comes to a critical current increase.

The intervention in the operational management of the power electronic device to prevent the occurrence of overcurrents can mean, for example, that a current flow in the power current path is interrupted directly, for example, by a current path arranged in the relay, a power semiconductor or a fuse element. The intervention may also mean, for example, that the manipulated variable for setting the current, d. H. For example, a drive signal for a power semiconductor, is suppressed or that is intervened, for example, by the operation management in the signal processing, that no more manipulated variables are generated or only those that one

Counteract the formation of an overcurrent. It can also be the example Signal processing completely shut down. The duration of the intervention in the operational management, which is a component of the second time period, is always the time duration from the beginning of the intervention to the effect on the current in the power current path in the form of a shutdown or reduction of the stream. Accordingly, depending on the nature of the intervention, durations may be included as a component in the second time period, which also constitute a component of the first time duration.

With regard to the criterion by means of which the danger of occurrence of overcurrents can be recognized, depending on the type of input variable of the signal processing and depending on the type of signal processing, different embodiments of the invention result

inventive method. For example, the criterion may include a change in the input of the signal processing within a predetermined first time interval by more than a predetermined value. For example, such a criterion can be used as an input variable if a continuous voltage characteristic is present

Detect voltage dip, which may have an overcurrent in the power electronic device depending on the nature of the signal processing. For sinusoidal mains voltage signals with an effective value of the voltage of 230 V which are frequently present in power electronic devices, a voltage change of more than 10 V to more than 100 V, for example of more than 40 V, can then be taken, for example, from the mains voltage voltage of 2 ^■ 2307 = 3257 Voltage drop can be classified. For example, in the case of digital signal processing, the first time interval in which this change occurs may be the length of the sampling interval in which the

For example, at a sampling frequency of 16 kHz, a first time interval of 62.5μ8 or multiples thereof may be detected.

The criterion by means of which the danger of an occurrence of overcurrents can be recognized can, in a further embodiment, be a change in the input quantity of the

Include signal processing to a value of zero. A value of zero also includes a value close to zero, which, for example, only due to

Noise influences from a value of exactly zero. In a practical

Implementation of the criterion can be ensured, for example, by ensuring that all values smaller than, for example, 1% of a maximum possible value of the

Signal processing input quantity or 1% of a last occurring value of the signal processing input value, to be treated as a value of zero. Even with the present criterion can be, for example, in the presence of a steady

Voltage curve as an input variable to detect a voltage dip, depending on the type of Signal processing may cause an overcurrent in the power electronic device.

In a further embodiment of the method according to the invention that comprises

A criterion by which the risk of an overflow phenomenon is recognizable, the presence of a value of the signal processing input value of zero for a longer period of time than a predetermined second time interval, again including a value close to zero in accordance with the previously explained explanations. Even with this criterion can be, for example, in the presence of a continuous voltage waveform as input a voltage dip detect, which can have an overcurrent in the power electronic device depending on the nature of the signal processing. Again, in the case of digital signal processing, the choice of the size of the second time interval may correspond to the length of the sampling interval or multiples thereof, i.e. at a sampling frequency of 16 kHz, for example a second time interval of 62.5μ8 or multiples thereof.

Likewise, the criterion by means of which the danger of occurrence of overcurrents can be recognized, in a further embodiment, may comprise exceeding a predetermined threshold value by the value of the input variable of the signal processing. With such a criterion, many situations can be detected which, depending on the input quantity and the type of signal processing, can result in an overcurrent in the power electronic device.

Likewise, in a further embodiment, the criterion by means of which the risk of an occurrence of overcurrents can be detected, may include the undershooting of a predetermined threshold value by the value of the input variable of the signal processing. Depending on the input quantity and the type of signal processing, this criterion can be used to detect other situations that may result in an overcurrent in the power electronic device.

Of course, combinations in the form of logical combinations of the criteria, some of which have been exemplified above, are possible and in many cases even useful to increase the security in detecting the risk of overflows, and in particular the residual risk of unnecessary

Shutdown of the power electronic device to minimize. In the case of a combination of several criteria, these then each constitute subcriteria of the criterion, by means of which the danger of an occurrence of overflows can be recognized. The method according to the invention is particularly advantageous if the signal processing comprises a control-technical algorithm, since large control deviations, which can already be recognized on the basis of the input variable of the signal processing, are frequently too

possibly very fast and strong rising overcurrents in the

power electronic device can lead.

Since the detection of the input of the signal processing, the evaluation of this input with respect to the criterion and the intervention in the operation in the method according to the invention are carried out within the second period shorter than the first period to the effect of changing an input of the signal processing delayed on the stream, the evaluation of the

Input variable in terms of the criterion preferably include a processor-assisted processing of a program flow, especially if the signal processing, as is often the case in power electronic devices, is done by a processor-assisted processing of a program flow. An evaluation of the input variable in terms of the criterion by means of dedicated, possibly also slow hardware is just as possible.

In a later closer considered embodiment feeds that

Power electronic device, the power in an AC network. In such network-feeding inverters or converters, which are often decentralized

Energy generation are used, is an application of the invention

Method particularly advantageous because they usually have a current control, are to be avoided in the overcurrents in the device and during feeding into the grid. A current control often uses a voltage as input

AC mains, with a voltage drop at the supplied from the mains

AC voltage can then lead to fast rising, strong overcurrents. A further preferred embodiment of the method according to the invention therefore comprises the case in which the input variable of the signal processing is a voltage of a

AC network is.

An inventive power electronic device comprises a device for

Implementation of the method according to the invention. This may, for example, be considered in more detail below in one embodiment

Netzeinspeis- inverting inverter, which energy from a DC or DC voltage source, for example, from a photovoltaic generator, in AC converts it into a public or private electrical grid.

With regard to a number mentioned in connection with features in the patent claims and the description, this always has to be understood that at least this number is present.

Furthermore, the reference numerals contained in the claims are not limited to a particular embodiment, but merely serve a better purpose

Understanding the reading of the claims.

The invention will be described in more detail below with reference to embodiments using three figures.

1 shows a power electronic device with a device for carrying out a method according to the invention,

FIG. 2 shows the sequence of a method according to the invention in a time comparison to a signal processing process of a power electronic device, and

FIG. 3 shows an exemplary evaluation of an input variable with regard to a

Criteria by which the danger of an overcurrent occurrence can be identified, using the example of a voltage of an alternating current network

In Fig. 1, a power electronic device 1 with a device (OP) 2 for carrying out a method according to the invention is shown schematically. It is in the power electronic device 1 shown here to an inverter 3 with a

Inverter 4, which converts energy from a DC or DC voltage source 5 into an alternating current and feeds it into a public or private electrical supply network 7.

The inverter device 4 itself may consist of one or more stages, for example of an inverter bridge and possibly additionally one or more converter stages, for example DC / DC converter stages. It can also contain current or voltage intermediate circuits. The inverter 3 of Fig. 1 feeds the

AC single-phase in the supply network 7 a. Of course, a power electronic device 1 according to the invention can also be a polyphase, for example be three-phase in a supply grid feeding inverter. In the present example, the energy from a photovoltaic generator 6 is provided. The Gleichstrombzw. DC voltage source 5 can also be other energy suppliers, such

For example, include a battery or a fuel cell. The DC or DC voltage source 5 may just as well be a current or voltage intermediate circuit, the further converter stages are connected upstream.

The power electronic device 1 has a signal processing (SP) 8, which processes in the present example measured with a voltage measuring device 9 value of the AC mains voltage as input and a manipulated variable for the strength of one in the power electronic device 1, in this case in a power current path within the Inverter 4, occurring current generated.

Furthermore, the electronic power device 1 has an operational management (OC) 10, which controls the operation of the power electronic device 1 and for this purpose transmits values and commands to the inverter 4 and the signal processor 8. Optionally, an additional reading of values and operating states from the

Inverter 4 and the signal processing 8 may be provided by the operation management 10.

The device 2 for carrying out the method according to the invention evaluates the

Input of the signal processing, in the present case with the

Voltage measuring device 9 measured value of the mains AC voltage of

Supply network 7, in terms of meeting a criterion by which the risk of occurrence of overcurrents can be seen. If the criterion is fulfilled, an intervention in the operational management will cause the occurrence of overcurrents in the

power electronic device 1 prevented. For this purpose, the operational management transmits values or commands to the inverter 4 and / or the signal processor 8, which cause a shutdown or reduction of the current in the power current path before a dangerous overcurrent occurs.

Optionally, additional transmission of values and commands to the device 2 for carrying out the method according to the invention by the operational management 10 and / or read-in of values and operating states from the device 2 for carrying out the method according to the invention can be provided by the operational management 10. The signal processing 8, the operation management 10 and the device 2 for carrying out the method according to the invention can be realized both analog and digital, as well as mixed analog and digital. In addition to an implementation by dedicated hardware, especially in a digital implementation, a processing of a

Program sequence on a processor done. The signal processing 8, the

Operation management 10 and the device 2 for carrying out the inventive

Method can be spatially or functionally separated. However, they may as well be realized in whole or in part both spatially and functionally in a common unit. That is, for example, the operations management 10, in particular, if it comprises only a few actions, be integrated into the signal processing 8. Furthermore, the signal processing 8, the operational management 10 and the device 2 for carrying out the method according to the invention can be arranged completely or partially spatially or, in particular, when implemented by hardware, also functionally in the inverter device 4. A separate arrangement in individual separate functional blocks, as in FIG. 1, serves only for a better overview and a better understanding of the function of the invention.

2 shows the sequence of a method according to the invention in a time comparison to a signal processing process of a power electronic device. The numbering of the steps shown in the flowchart is carried out below in the order of the first appearance in the description and does not specify an order for the execution of the steps due to the existing returns and the parallel execution of individual steps the signal processing in a first step 13 a

Manipulated variable generated by the strength of a in the second step 14

power electronic device occurring current 15 is set. In this case, the effect of a change in the input variable 12 to the current 15 within a first time period t-ι. In the method according to the invention, a value of the input variable 12 is first detected in a third step 16. In a fourth step 17, the input variable is evaluated with regard to the fulfillment of a criterion by means of which the danger of an occurrence of overcurrents can be detected. If the criterion is not satisfied, then after a return to the third step 16, the next value of the

Input 12. If the criterion is satisfied, in a fifth step 18 an intervention is made in the operational management of the power electronic device which avoids the occurrence of overcurrents in the power electronic device, and then after a return to the third step 16 the detection of a next value of the input variable 12 performed. The avoidance of overflow through the Operation management 18 takes place, as indicated in Fig. 2 by the dashed arrows, optionally by influencing the signal processing in the first step 13, by influencing the adjustment of the current by means of the manipulated variable in the second step 14 or by a direct influence of the current 15, or by combinations thereof. The detection of the input variable 12 in the third step 16, the evaluation of this input variable with respect to the criterion in the fourth step 17 and the intervention in the operation management in the fifth step 18 including an effect of this intervention on the current 15 according to the invention, as shown in Fig. 2 , within a second time period t ₂ , which is shorter than the first time period t | , whereby a training of a

Overcurrent can be avoided before it occurs. The second time duration t ₂ is set, for example, by an appropriate selection of components for the signal processing 8, the operation management 10 and the device 2 for performing the method according to the invention, and when processing a program flow on a processor by a corresponding structuring of the program flow in that it is shorter than the first time t-1. It is also possible that the processing of the

The method according to the invention with another, for example higher, sampling or clock rate takes place as the signal processing.

FIG. 3 illustrates an exemplary evaluation of an input variable of a signal processing of a power electronic device with regard to a criterion by means of which the danger of an occurrence of overcurrents can be detected. As an example of the input variable, a voltage U _A c of an AC network is shown, which, for example, the AC line voltage measured by the voltage measuring device 9

Supply network 7 may be from Fig. 1. In public or private electrical supply networks, the voltage of the AC network has a sinusoidal waveform with a frequency of often 50 Hz or 60 Hz and has an effective value of, for example, 230 V, corresponding to a net coupling voltage of 2 ^■ 2307 = 3257th In the present case, the values the input quantity as samples 21 of the voltage of the AC mains, with power electronics devices a sampling in the kHz range is common, for example, a sampling at a frequency of 8 kHz, 16 kHz or 48 kHz. A sampling interval At then has a size of 125μ8, 62.5μ8 or 20.8 MS.

In the voltage curve in FIG. 3, a voltage dip occurs at the time t _E. Such a breakdown of the voltage can, for example, in netzeinspeisenden

Power electronic devices that have a current control as part of their signal processing, leading to a rapidly increasing, strong overcurrent. As a criterion for Detection of such a voltage drop, and thus as a criterion by means of which the risk of an occurrence of overcurrents can be recognized, serves here as the difference AU between a sample 22 before the voltage dip and a next sample 23 after the voltage dip, ie over a predetermined time interval, which here corresponds to the length of the sampling interval At, is greater than a predetermined value U _diff . A choice of a suitable value for U _Diff is made depending on the application, in particular depending on the characteristics of the input variable and the signal processing. In the present case of a mains coupling voltage of 325 V, a value for U _diff of 10 V to 100 V, for example 20 V at a sampling interval of 20.8 μβ, 40 V at a

Sampling interval of 62.5μ8 or 80V at a sampling interval of 125μ8, used to classify a voltage dip.

It can also be seen in FIG. 3 that in the case of the voltage dip illustrated there, the next sample value 23 has a value of zero after the voltage dip and that after this next sample value 23 after the voltage dip further samples 24 follow, which have a value of zero. This results in further criteria by which the risk of an overcurrent occurring is recognizable, that a change in the input signal processing signal to a value of zero, or that a value of the input signal processing of zero for a longer period than a

given second time interval, wherein the length of the second time interval can be selected, for example as a multiple of a length of a sampling interval At, then a maximum length of the second time interval is determined by the fact that according to the illustration in Fig. 2 in a method according to the invention, the second time period t ₂ shorter than the first time t-ι must be measured. The further criteria can be used both individually and in combination, as well as in combination with the aforementioned criterion, in which the change AU is evaluated over a predetermined time interval Δt to the exceeding of a predetermined value U _D i _ff Criteria then each sub-criteria of the criterion, by means of which the risk of occurrence of overcurrents can be seen represent.

For a further criterion by means of which the danger of an occurrence of overcurrents can be recognized, threshold values + THR and -THR to be optionally used are shown in FIG. 3. Exceeding the threshold value + THR by the sample value 22 before the voltage dip, or correspondingly below the threshold value -THR by a corresponding sample value before the voltage dip, may for example preferably be combined with the criterion in which the change AU is greater than or equal to

predetermined time interval At to the exceeding of a predetermined value U _D i _ff is used to prevent a shutdown of the power electronic device in cases where the current threatens to assume no critical values despite a voltage drop in the voltage of an AC network. An interpretation of the threshold + THR, or according to -THR takes place depending on the application. In the present example, values in the range of 90% to 10% of the net dome voltage, in the present case a mains dome voltage of 325 V, ie values of 292.5 V to 32.5 V, for example 162.5 V, are meaningful.

The criterion described above, in which the value of the input variable is evaluated with respect to the exceeding or undershooting of the threshold value + THR or -THR, as well as the criterion in which the change AU over a predetermined time interval At to the exceeding of a predetermined value U _{Diff is} evaluated individually or in combination for the detection of other forms of

Voltage dips, which are not shown in the present example in Fig. 3. For example, a voltage may break to a non-zero value and maintain it for a longer period of time than a predetermined third time interval. Likewise, the voltage may collapse to a nonzero value and then continue to sinusoidally with reduced line cup voltage. Even with such voltage dips exists depending on the type of

Signal processing the risk of an occurrence of overcurrents, which are then avoided by the inventive method.

LIST OF REFERENCE NUMBERS

1 power electronic device

2 device

3 inverters

4 inverter device

5 DC or DC voltage source

6 photovoltaic generator

7 Electricity supply network

8 signal processing

9 tension measuring device

10 management

12 input size

13 step

14 step

15 electricity

16 step

17 step

18 step

21 sample

22 sample

23 sample

24 sample

Claims

WO 2014/086783 "1 °" PCT / EP2013 / 075396

1 . A method for preventing overcurrents in a power electronic device (1), which has a signal processing (8) for generating a manipulated variable for the strength of a current (15) occurring in the power electronic device (1), wherein an effect of a change of an input variable (12 ) of the

Signal processing (8) to the current (15) delayed by a first time period occurs, characterized in that the input variable (12) of the signal processing (8) is evaluated with respect to the fulfillment of a criterion by which the risk of occurrence of overcurrents can be seen and that upon fulfillment of the criterion in such a way in the operational management (10) of the power electronic device (1) is intervened such that the occurrence of overcurrents in the

power electronic device (1) is avoided before it comes to a critical current increase, wherein the detection of the input variable (12), the evaluation of this input variable with respect to the criterion and the intervention in the

Operation (10) including an effect of this intervention is performed on the current (15) within a second period of time which is shorter than the first

Period of time.

2. The method according to claim 1, characterized in that the criterion by which the risk of an occurrence of overcurrents is recognizable, a change of the input variable (12) of the signal processing (8) within a predetermined first time interval by more than a predetermined value.

3. The method of claim 1 or 2, characterized in that the criterion by which the risk of occurrence of overcurrents is recognizable, a change of the input variable (12) of the signal processing (8) to a value of zero includes.

4. The method according to any one of the preceding claims, characterized in that the criterion by which the risk of an occurrence of overcurrents is recognizable, the presence of a value of the input variable (12) of

Signal processing (8) of zero for a longer period of time than a predetermined second time interval.

5. The method according to any one of the preceding claims, characterized in that the criterion by which the risk of an occurrence of overcurrents is recognizable, the exceeding of a predetermined threshold value by the value of the input variable (12) of the signal processing (8).

6. The method according to any one of the preceding claims, characterized in that the criterion by which the risk of an occurrence of overcurrents can be seen, the falling below a predetermined threshold value by the value of the input variable (12) of the signal processing (8).

7. The method according to any one of the preceding claims, characterized in that the signal processing (8) comprises a control algorithm.

8. The method according to any one of the preceding claims, characterized in that the evaluation of the input variable (12) with respect to the fulfillment of the criterion includes a processor-assisted processing of a program flow.

9. The method according to any one of the preceding claims, characterized in that the power electronic device (1) feeds the current (15) into an AC network.

10. The method according to any one of the preceding claims, characterized in that the input variable (12) of the signal processing (8) is a voltage of an AC network.

1 1. Power electronic device (1) with a device (2) for carrying out a method according to one of Claims 1 to 10.

12. Power electronic device (1) according to claim 1 1, characterized in that the power electronic device (1) energy from a DC or

DC power source (5), in particular from a photovoltaic generator (6) converts into alternating current and feeds it into a public or private electrical supply network (7).